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
|
|---|---|---|---|---|---|
Parasitic infections , which are among the most common infections worldwide , disproportionately affect children; however , little is known about the impact of parasitic disease on growth in very early childhood . Our objective was to document the prevalence of parasitic infections and examine their association with growth during the first three years of life among children in coastal Kenya . Children enrolled in a maternal-child cohort were tested for soil transmitted helminths ( STHs: Ascaris , Trichuris , hookworm , Strongyloides ) , protozoa ( malaria , Entamoeba histolytica and Giardia lamblia ) , filaria , and Schistosoma infection every six months from birth until age three years . Anthropometrics were measured at each visit . We used generalized estimating equation ( GEE ) models to examine the relationship between parasitic infections experienced in the first three years of life and growth outcomes ( weight , length and head circumference ) . Of 545 children , STHs were the most common infection with 106 infections ( 19% ) by age three years . Malaria followed in period prevalence with 68 infections ( 12% ) by three years of age . Filaria and Schistosoma infection occurred in 26 ( 4 . 8% ) and 16 ( 2 . 9% ) children , respectively . Seven percent were infected with multiple parasites by three years of age . Each infection type ( when all STHs were combined ) was documented by six months of age . Decreases in growth of weight , length and head circumference during the first 36 months of life were associated with hookworm , Ascaris , E . histolytica , malaria and Schistosoma infection . In a subset analysis of 180 children who followed up at every visit through 24 months , infection with any parasite was associated with decelerations in weight , length and head circumference growth velocity . Multiple infections were associated with greater impairment of linear growth . Our results demonstrate an under-recognized burden of parasitism in the first three years of childhood in rural Kenya . Parasitic infection and polyparasitism were common , and were associated with a range of significant growth impairment in terms of weight , length and/or head circumference .
Parasitic infections , which are among the most common infections worldwide , disproportionately affect children [1] . It is increasingly recognized that both protozoan and helminthic diseases are common among children under the age of five years . Of particular concern , their associated disease burden is experienced during the period of life most critical for physical and cognitive development [1] . The association between parasitic infection and growth impairment has long been established [2] . However , few studies have evaluated the impact of parasitic disease on growth in very early childhood [3–5] . The mechanism by which parasitic disease impairs child growth is not fully understood , but is thought to be related to host systemic responses to infection ( fever , decreased appetite ) , to intestinal disease disruption of host gut processes , and to anemia [2 , 4] . Schistosoma japonicum and S . mansoni infections have been associated with growth inhibition in school age children , even among those with low parasitic burden [2 , 6] . It is also evident that parasite-related growth deficits can be overcome by specific treatment of these infections . In a pediatric population in Kenya , weight gain deficits were ameliorated with intermittent albendazole treatment for soil transmitted helminth infections [7] . Other studies have shown improvement in growth and physical activity in children actively treated for STH infections [1 , 8] . Similar findings were seen in coastal Kenyan pre-school aged children infected with helminths [3] . In Peruvian pre-school aged children , Gyorkos and colleagues found a greater degree of stunting and reduced length-for-age Z-score in children infected with moderate to heavy helminth infections as compared to children with no or light infections [4] . However , such studies have not provided long-term longitudinal data on infants followed from birth . The goal of the present prospective study was to monitor the incidence of childhood parasitic infections ( intestinal parasites , malaria , filaria and Schistosoma haematobium infection ) from birth to the age of 36 months in an area known to be endemic for multiple parasite species . This , in turn , provided better definition of the association between incident parasitic infection and growth impairment experienced during early childhood .
Healthy pregnant women and their offspring born at the Msambweni District Hospital on the south coast of Kenya were enrolled in this mother-child cohort study . Approval for the study was obtained from the Kenya Medical Research Institute National Ethical Review Committee and from the Institutional Review Board for Human Studies at University Hospitals of Cleveland Case Medical Center . Mothers provided written informed consent for their own participation and that of their infants . This observational study was performed at Msambweni District Hospital on the southern coast of Kenya from 2007–2010 . Healthy pregnant women and their offspring were enrolled in a three-year longitudinal maternal-child cohort study . Pregnant mothers were included if they were permanent residents of Msambweni and received pre-natal and post-natal care at the District Hospital’s antenatal clinic ( ANC ) . Mother-infant pairs were excluded if: a ) they experienced a complicated infant delivery resulting in significant infant morbidity at birth; b ) the infant was born at less than 36 weeks gestation; c ) the mother had known chronic illness; d ) the mother had severe anemia ( Hgb < 6 g/dl ) requiring hospitalization and urgent treatment; e ) the mother had permanent disability that impeded study participation and/or comprehension; or f ) the family had plans to relocate after delivery . Upon enrollment , mothers completed a detailed questionnaire that queried their education level , spouse’s occupation , and household income . At later time points , they and their offspring were screened for malaria , Schistosoma haematobium , filariasis , and intestinal helminthes and protozoa . Parasitic infections were tested via blood , urine and stool . Sera and plasma were frozen at -20°C for subsequent determination of filarial Og4C3 antigen testing , filarial BMA specific ( Brugia malayi antigen ) IgG4 antibody , and Schistosoma haematobium specific ( soluble worm antigen of S . haematobium ) IgG4 levels . Red blood cell ( RBC ) pellet was used for DNA extraction for detection of malaria parasites via RTQ-PCR [5] . A drop of blood was used to make a malaria blood smear and measure hemoglobin by Hemocue method [9 , 10] . Malaria positive was defined as either positive blood smear or positive PCR or both . W . bancrofti infection was detected by assay for circulating antigen in plasma samples with the use of a commercial Og4C3 antigen detection assay ( TropBioMed , Townsville , Australia ) and also assessed by ELISA detection of BMA-specific IgG4 antibodies [11–13] . Positive filaria infection was defined as either positive Og4C3 antigen or presence of BMA-specific IgG4 antibodies or both . A single sample of fresh stool was examined in duplicate to quantify ova of intestinal parasites using the Ritchie Method [14] . A single fresh urine sample was filtered and examined for S . haematobium eggs [11 , 13] . Infection status with S . haematobium was also assessed by ELISA detection of SWAP-specific ( soluble worm antigen of S . haematobium ) IgG4 antibodies in plasma samples [13] . Positive S . haematobium infection was defined as positive for either eggs identified in urine or the presence of SWAP-specific IgG4 antibodies or both . Infants were followed from birth every six months until three years of age . Twins were included and analyzed as individuals since they comprised <2% of the cohort . In addition to follow up visits , children were evaluated during any episodes of acute illness that occurred between scheduled visits . A select group of specially-trained clinical staff performed physical examinations , including anthropometrics ( weight , length and head circumference ) on all infants at delivery and at each scheduled follow up visit . Following Kenya Ministry of Health guidelines , infants received amodiaquine ( if <5 kg ) or artemether/lumefantrine ( Coartem ) ( if >5 kg ) for malaria treatment . Infants were treated with mebendazole for soil transmitted helminth infections . For purposes of analysis , we first transformed the growth outcomes into age and gender-specific Z-scores , using the WHO Anthro software ( WHO , Geneva , Switzerland ) [15] . The Z-score is a standardized score indicating how many standard deviations a child’s growth parameter differs from an established mean value for children of the same age among the world’s population . For our analysis , the key dependent variables of interest were weight , height and head circumference Z-scores over the first three years of life . The main exposure variables of interest were the number and types of infection at any time point ( cumulative infections in mothers and their offspring ) . Basic descriptive analyses of all the measurements were conducted on each time point in the study . The statistical analysis was performed using SAS version 9 . 3 ( SAS Institute Inc . , Cary NC ) . In initial exploratory analysis for bivariate comparisons of continuous data , the Student’s t-test was used , and for categorical data , chi-square tests were used . Next , logistic regression models were created to assess the association of maternal infection during pregnancy with infant infection at each follow up visit . The primary study outcome , the association of infant infection at any point prior to the measured time point ( cumulative infections ) with the rate of growth and growth at each time point , was evaluated using general estimating equations models ( GEE ) for longitudinal data . The use of GEE models allowed us to account for the heteroscedasticity of repeated outcome measurements on the same subjects over time . For soil transmitted helminth infections , infection by each species was analyzed alone ( Ascaris , Trichuris , hookworm , or Strongyloides ) and then as an aggregate class ( “STH” ) . Infections were treated as categorical , either present or absent for all types . The final models examined the relationship between the infant’s parasitic infections and their weight , length , and head circumference Z-scores over the first three years of life , while controlling for the infant’s sex , birth weight , birth length , birth head circumference , and maternal education . Because maternal education was highly correlated with household expenditures , it was used as a proxy for socioeconomic status . To control for secular changes during the study , infant’s infection , time , and the interaction of infection and time were included in the models . Tukey’s method of adjustment was used to assess significance when comparing across multiple time points . A significance level of 0 . 05 was used for all statistical tests . To better refine the estimates of the effects of incident infection on growth trajectory , a subset analysis was performed on the cohort of 180 children who did not have any missed scheduled visits from birth to 24 months . This 24 month endpoint was chosen for this analysis because of a significant limitation in the number of children who were able to complete every visit through 36 months ( only 91 children ) .
During the study period , 545 infants were followed and tested for parasitic infections over the first three years of life . Of the 545 , 13 pairs of twins ( 26 infants ) were included . Maternal and infant characteristics are described in Table 1 . During the infant observation period , soil transmitted helminths were the most common infection , with 106 of 545 or 19% ( CI95: 16–23% ) , of children experiencing infections by age three ( Fig 1 ) . Malaria followed in period prevalence with 68 ( 12% , CI95: 10–16% ) infected by three years of age . Filaria and S . haematobium infection occurred in 26 ( 4 . 8% , CI95: 3–7% ) and 16 ( 2 . 9% , CI95: 2–5% ) children , respectively . Thirty-two percent ( 176 of 545 ) of children experienced at least one parasitic infection and 7% ( 37 of 545 ) were infected with more than one parasite in the first 36 months of life ( Fig 2 ) . All infection categories were documented as early as six months of age . For Schistosoma infection , the infants less than 18 months were only positive by IgG4 assays . Of the three positive infants positive for filarial infection under 18 months of age , one was only positive via BMA IgG4 assay . Prenatal maternal infections increased the odds of infant infections ( see S1 Table ) . Infant parasite infection in the first three years of life was associated with decreases in several growth parameters ( Table 2 ) , although prevalence of marked stunting and wasting were low overall ( <2% of cohort for each ) . The average length , weight and head circumference Z-scores for all children at each follow up time point are shown in Fig 3 . STH infection by 24 months was significantly associated with a lower growth attainment in terms of length and head circumference relative to WHO standards . STH infection by 30 months was also associated with a lower growth in head circumference by 30 months . When evaluating soil-transmitted helminths individually , hookworm infection by 24 months was associated with below average growth in length and head circumference at 24 months and relative deficit in overall weight gain over the first 36 months of life . Hookworm infection by 30 months was associated with a decreased Z-score for head circumference at 30 months , while infection by 36 months of age was associated with a relative decrease in weight gained at 36 months . Ascaris infection by either 12 or 18 months was associated with a decrease in weight Z-scores at each age milestone , respectively . Ascaris infection by 24 months was associated with a decrease in length achieved at 24 months . Strongyloides infection by 30 months was associated with a relatively lower head circumference at that age . Of intestinal protozoa , Entaomeba histolytica infection by 24 months was associated with a lower weight Z-score at 24 months , and infection by 30 months was associated with a lower head circumference score at 30 months . Schistosoma infection by 36 months was associated with a reduced Z-score for head circumference at 36 months . Malaria infection was associated with a lower linear growth ( length ) attainment at 6 , 12 and 18 months and lower Z-scores for head circumference at 24 and 36 months . Of note , in the study cohort , no significant reductions were observed in terms of weight , length , or head circumference scores following filarial infection . In the subset of 180 infants who attended all follow up visits through two years of age , as in the overall group ( N = 545 ) , soil transmitted helminths proved to be the most common infection ( 47 infants , 26% , CI95: 19–31% ) by 24 months of age . Among this more closely followed group of children , a number of growth deficits were significantly associated with infection . Hookworm infection by 24 months was associated with a decrease in weight and head circumference Z-score at 24 months ( Table 2 ) . Ascaris infection by 24 months was associated with a decrease in linear growth ( length Z-score ) at 24 months . E . histolytica infection by 24 months was also associated with a length deficit by 24 months . Malaria infection was associated with a statistically significant reduction in weight and length attained at 18 months , and in head circumference at 24 months . By contrast , filarial infection was associated with a statistically significant increase in weight attained at 24 months . Among this full follow-up group , the relative rate of growth attained in weight , length or head circumference by 24 months was decreased among children with any parasitic infection as compared to the uninfected group ( Table 3 ) . Between the group with complete follow up and the cohort with incomplete follow up ( i . e . N = 545 cohort minus 180 ) , there were significant differences between maternal education and number of infections . In the complete follow up group ( N = 545 ) , mothers had higher education ( p = 0 . 005 ) and children experienced more infections ( p = 0 . 004 ) over the first 24 months of life . Overall , polyparasitism ( defined as children experiencing more than one type ( any soil transmitted helminth , filaria , malaria and/or Schistosoma haematobium ) of parasitic infection at any time point ) was associated with decreases in length ( p = 0 . 004 ) by 36 months of life as compared to children who did not experience any parasitic infections in the first 36 months of life . To demonstrate the relative timing of polyparasitism effects on infant growth , WHO growth charts are shown for two infants who experienced multiple parasitic infections in the first 24 months of life ( Fig 4 ) .
Our results confirm that parasitic infections are common among the youngest age groups of children ( under three years old ) who live in endemic areas such as rural coastal Kenya . This study found that parasitic infections during the first three years of life were associated with significant differences in growth and that children can be infected within the first six months of life . By 24 and 36 months of age , many children were multiply infected ( polyparasitized ) , with 7% of the children experiencing more than two different parasitic infections . STHs were the most common infections encountered in our cohort . This is not surprising as young children have frequent contact with contaminated soil and explore their environment by touching and tasting early in life . An increase in the absolute number of STH infections was observed between six to 18 months , at the time when children are becoming mobile . STHs infect the gut , a compartment with relatively limited effects of host immunity , free from the robust response of the immune system found in host tissues and circulation , and have a constant supply of nutrients . They can successfully inhabit this environment for long periods of time , resulting in prolonged chronic infection [16] . Intestinal worms may disrupt nutrition by 1 ) feeding on host tissue causing blood , iron , and protein loss; 2 ) impairing digestion and absorption; 3 ) causing generalized inflammatory responses that lead to decreased appetite and increased metabolic demand , thus , diverting nutrients and energy of the host [1 , 2 , 16] . Hookworms are one of the most important species in terms of disease causing profound anemia even at low egg counts [1] . Ascaris and hookworm were associated with decreased growth in terms of weight , body length , and head circumference , likely via the mechanisms mentioned earlier . In addition to intestinal parasitic diseases , other parasites such as malaria and Schistosoma infection often lead to anemia and contingent host inflammatory responses , resulting in decreased energy intake and compromised host nutrition [6 , 16] . In particular , malaria was associated with multiple growth deficits in our cohort . Overall , our data indicate that growth effects were more often significant at 24 and 36 months of age , likely because the deleterious effects of infection take time to significantly retard growth . Lower height-for-age Z-scores ( i . e . stunting ) indicates long term insult ( chronicity ) , whereas , lower weight-for-length Z-score represents acute insult . In a resource-limited setting , these deficits likely represent both the effects of parasitic infection and the effects of comorbidities such as undernutrition . It is important to note how early in life these effects can be detected . Schistosoma infection was associated with a decrease in head circumference Z-score at 36 months; a finding that was statistically significant , even though only a relatively small number of children were infected . However , it is worth noting that these young children were only positive for Schistosoma infection via antibody testing , which could be the result of persistent maternal IgG4 antibody in the child’s circulation . Whatever the source of the positive serology , it is worth noting that this group of children experienced alteration in growth either from exposure to Schistosoma haematobium in utero , or from early life infection . Malaria , hookworm , and other soil transmitted helminth infections were also associated with decreases in head circumference . The rate of head growth during the first three years of life is significantly correlated with cognitive development and brain growth [2 , 17] , and it is notable that , significant cognitive developmental delays have been associated with other types of infection experienced in early life [5 , 16 , 18] . Future directions for study should include evaluation of the associations between parasitic infections , decreased growth in head circumference , and loss of cognitive potential , especially in the first two years of life , as this is the period known to be most important for brain development . Over the three year period of the study , follow up of the 545 children was incomplete , such that inclusion of different children in different age milestone data groups likely led to variability in the strength of associations between infections and growth outcomes . To overcome this limitation , we performed a subgroup analysis of children who were followed at every visit until the age of two . Among these children , cumulative infection rates were higher , likely due to more frequent testing and better ascertainment of early infection status . Furthermore , among this group , larger growth rate deficits were documented in weight , length and head circumference in infected children . The relative rate of change in Z-scores for weight , length and head circumference over the first 24 months of life ( i . e . growth velocity ) was decreased in children with any parasitic infection versus children without infection . Unexpectedly , we observed some associations between parasitic infections and an increase in growth parameters . Filaria infection by 24 months was associated with an increase in weight Z-score . Studies in experimental animal models have shown an association between filarial infection and obesity in mouse models [19] . Our cohort children , although they had no clinical signs of lymphedema , may have had excess inflammation with interstitial fluid and fat deposition , which might explain this observation . Ascaris infection by 36 months of life was associated with a significant increase in head circumference Z-score . Again , we cannot currently explain this finding based on individual child , family , or SES factors . Polyparasitism , which was defined as infection with more than one parasite at any given time during the first 36 months of life , was associated with growth faltering in a dose dependent relationship . The number of child infections over the three years was found to be negatively associated with length changes over time . While most previous studies have correlated poor growth outcomes with intensity of infection , the finding of multiple infections associated with linear growth in early life is novel [4] . An interesting finding with respect to estimated socio-economic standing ( SES ) was the better rates of follow up among the lower SES families . This may have been due to a greater number of working women in the higher SES groups , making adherence with regular follow up visits more challenging . There were strengths and limitations for this study . We recognize that a cohort bias exists , as we were comparing infants within one district over a limited period of time , and did not evaluate growth patterns or parasitic prevalence outside this community . We acknowledge that there are many other factors that may have contributed to poor growth and nutrition in this resource limited setting , including lack of access to protein-rich foods , and intercurrent illness with diarrhea or other conditions . Although we have data on the children’s care while ill between visits , we did not have data on other potential co-infections or comorbidities that might have impacted growth . Unlike some previous studies , we did not attempt to tease out details of nutritional status , factors that could have had confounding effects on growth parameters related to chronic anemia , iron and micronutrient deficiencies , and lack of calorie-rich foods . Use of comprehensive 24 hour food recall questionnaires and iron or micronutrient serum markers would have bolstered our anthropometric assessments . However , weight , length and head circumference for age and sex are , in and of themselves , very reliable tools for growth assessment when performed by trained teams , as in our study [15 , 20 , 21] . As for any tool that requires human measurement , both random and systematic error can be introduced into the measured data . Inter-rater variability was reduced in this study by using the same staff , who were fully trained to perform standardized measurements , throughout the study period . Because we had precise knowledge of each child’s birth date , and , hence , their age in days , our Z-scores were likely to be more accurate than those reported in past cross-sectional surveys , where children’s ages have based on recall , which is much less reliable . In our study , food availability was not specifically measured , however , maternal education , used as a proxy for socioeconomic standing , was used to estimate access to nutrition . In terms of other environmental factors , we could not measure the complex interaction of other risk factors for acquiring infection , i . e . , lack of shoes increasing risk of hookworm infection and poor sanitation , which cannot be simply derived from SES parameters and infection status . The longitudinal analysis of the association of infection with growth parameters required the use of cumulative infection effects instead of those of current infections at each time point . This was due to the small number of infected children at any given time point . Further , although the children were treated , we did not test children for cure , such that infections could have persisted . As a result , the cumulative effect of detected infections on growth may have been a more accurate reflection of the effects of infection in our cohort . We did not analyze the correlation between infection intensity and growth . Those infections with intensity data collected ( schisto and STH ) would have all been classified as “light” infections; therefore , all infections were categorized as either present or absent . Previous studies have found higher intensity STH infection to correlate with decreased body length in young children and likely an important contributor to overall poor nutrition and growth [4] . We recognize that many standard methods for testing for STH parasitic infections are not sensitive overall , and thus , infection rates we measured could have been falsely low , thereby leading to a Type II error [22 , 23] . By using multiple testing methods , sensitivity for detecting malaria , filarial and Schistosoma infections was increased in our study . However , testing with antibody assays for S . haematobium infection and filariasis in children less than 18 months may have been biased by persisting maternal antibody leading to misclassification bias ( false positives ) . Despite these limitations , we feel that this longitudinal cohort study offers a more accurate documentation of the early onset parasitic infections among young children in the tropics and their effects on early growth . Currently , most deworming campaigns are geared toward school-age children , among whom the impact of helminth infections has been well established . Operationally , this is the more “cost efficient” age group on which to focus therapy . However , this current ‘preventive chemotherapy’ policy is partly based on the notion that parasitic infection in the first five years of life is insignificant . Additionally , it is believed that deworming more heavily infected school-age children will yield a greater impact on parasite transmission community-wide . It is thus assumed that school age deworming will provide “trickle down” benefits in terms of growth and cognitive development to pre-school children . However , the findings of the present study reveal that helminthic parasite infection can occur very early in life and is associated with decreased physical growth , despite the low overall community prevalence for some of these parasites . It will be worth pursuing a better understanding of prevalence and effects of infection in these vulnerable pre-school age groups to most effectively target therapeutic interventions . Finally , if parasite transmission is to be fully disrupted , control programs must also logically target the younger , usually non-symptomatic age groups to prevent environmental contamination via egg dissemination . New prospective cohort studies are now underway in our study area to evaluate the associations between parasitic infections and growth , development , physical fitness , quality of life and vaccine response in both preschool and school aged children . These studies will address many of the limitations discussed here . | Parasitic infections are extremely common worldwide and children are especially vulnerable to these infections during critical periods of growth and development . There is a paucity of information about how frequently very young children ( under the age of five years ) are infected with parasites and the effects of parasitic infections on their growth and development . The findings from this study reveal that not only does infection occur early in life; it is associated with decreases in physical growth , even with low overall prevalence for some parasites . Decreases in growth of weight , length and head circumference during the first 36 months of life were associated with hookworm , Ascaris , E . histolytica , malaria and Schistosoma infection . In a subset analysis of 180 children who followed up at every visit through 24 months , infection with any parasite was associated with decelerations in weight , length and head circumference growth velocity . Multiple infections were associated with greater impairment of linear growth . It seems worthwhile to pursue a better understanding of prevalence and effects of parasitic infection in this vulnerable group to effectively target therapeutic interventions . And finally , if parasite transmission is to be significantly disrupted , control programs targeting these young , usually asymptomatic , age groups may be critical . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [] | 2015 | Parasitism in Children Aged Three Years and Under: Relationship between Infection and Growth in Rural Coastal Kenya |
Microarrays enable comparative analyses of gene expression on a genomic scale , however these experiments frequently identify an abundance of differentially expressed genes such that it may be difficult to identify discrete functional networks that are hidden within large microarray datasets . Microarray analyses in which mutant organisms are compared to nonmutant siblings can be especially problematic when the gene of interest is expressed in relatively few cells . Here , we describe the use of laser microdissection microarray to perform transcriptional profiling of the maize shoot apical meristem ( SAM ) , a ~100-μm pillar of organogenic cells that is required for leaf initiation . Microarray analyses compared differential gene expression within the SAM and incipient leaf primordium of nonmutant and narrow sheath mutant plants , which harbored mutations in the duplicate genes narrow sheath1 ( ns1 ) and narrow sheath2 ( ns2 ) . Expressed in eight to ten cells within the SAM , ns1 and ns2 encode paralogous WUSCHEL1-like homeobox ( WOX ) transcription factors required for recruitment of leaf initials that give rise to a large lateral domain within maize leaves . The data illustrate the utility of laser microdissection-microarray analyses to identify a relatively small number of genes that are differentially expressed within the SAM . Moreover , these analyses reveal potentially conserved WOX gene functions and implicate specific hormonal and signaling pathways during early events in maize leaf development .
The paralogous WUSCHEL1-like homeobox ( WOX ) genes narrow sheath1 ( ns1 ) and narrow sheath2 ( ns2 ) function from two , lateral foci within the maize shoot apical meristem ( SAM ) ( Figure 1 ) [1 , 2] , which comprises a pool of over 1 , 200 pluripotent cells that ultimately generates all lateral organs of the vegetative shoot . Maize leaf development begins with the initialization of approximately 200 leaf founder cells in the SAM , a recruitment process whereby cells occupying the periphery of the SAM are signaled to become founder cells of the incipient leaf [3 , 4] . The ns1 and ns2 duplicate genes encode redundant functions during maize leaf development . Single mutations in either ns gene are nonphenotypic [1 , 2 , 5–7] . Plants harboring recessive mutations in both of the ns genes fail to initialize founder cells within a specific , lateral domain of the SAM; failure to transduce this founder-cell recruitment signal results in the preprimordial deletion of an extensive lateral domain from the mutant maize leaf ( Figure 1 ) [6 , 7] . Although redundantly expressed in two foci comprising only approximately eight to ten total cells [2] , NS1 and NS2 propagate a founder-cell recruitment signal throughout the lateral domain of the SAM from where much of the lower portion of the maize leaf is derived . The molecular nature of this recruitment signal and the mechanism of its transduction within the SAM are unknown . The technique of laser microdissection ( LM ) permits the facile isolation of specific cells and tissues from plants [8] . Nanogram quantities of total RNA extracted from laser-microdissected tissues are linearly amplified by T7 RNA polymerase and used in microarray analyses . The combined use of LM and microarray technologies enables comparative analyses of discrete developmental fields while eliminating the transcriptional noise contributed by multiple tissues and downstream developmental events [9] . LM has been applied to global expression analyses of maize vascular and epidermal tissues , maize roots , as well as Arabidopsis embryos and floral organs [10–14] . The relatively large size of the maize SAM , approximately 200 founder cells are recruited into the incipient maize leaf versus 25–30 in Arabidopsis [3 , 15] , renders the maize plant especially tractable to LM strategies owing to the unique utility of this new technology to microdissect localized gene expression patterns within plant tissues . Here we describe analyses of differential gene expression in whole maize SAMs derived from ns mutants and nonmutant siblings , genetically nearly identical tissues , whose differences stem solely from the loss of NS1 homeobox gene expression in approximately eight to ten cells in the lateral SAM domain . In microarray analyses of more than 37 , 000 cDNAs representing approximately 21 , 721 maize genes , 66 genes are identified as differentially expressed in the ns mutant SAM , which demonstrates the power of LM-microarray to focus global analyses of gene expression to discrete , developmental domains . Quantitative real-time-PCR ( qRT-PCR ) corroborated the differential expression of 18 implicated genes and identified transcripts that are enriched in maize shoot meristematic tissues . In situ hybridization analyses revealed previously undescribed expression patterns for ten genes , six of which exhibit differential expression within the NS lateral domain of the SAM . Genes predicted to be involved in hormonal transport and signaling , signal transduction , and growth are especially implicated during NS-mediated leaf induction , and potentially conserved WOX gene functions during the regulation of two-component response pathways and of jasmonate-induced gene expression are identified .
Whole SAMs , comprised of the meristem proper as well as the founder cells of the incipient leaf , were laser microdissected ( Figure 2 ) from serial sections of ns mutant ( genotype ns1-R and ns2-R ) and nonmutant ( genotype Ns1/ns1-R and ns2-R ) seedlings grown under controlled conditions ( see Materials and Methods ) . Although previous analyses revealed that NS1 and NS2 perform redundant functions during maize leaf development [1 , 2 , 5–7] , the microarray experiments described herein measured differential gene expression conferred by NS1 function , given that the ns mutant and nonmutant samples analyzed in this study were all homozygous for the ns2-R mutation . Technical barriers ( described in Materials and Methods ) precluded the use of our laser-microdissected RNA together with maize oligonucleotide microarrays . Therefore , following extraction and linear amplification of SAM RNA [11] , expression profiles were generated for a combined total of 37 , 662 maize cDNA sequences ( including approximately 21 , 721 maize genes ) spotted onto three different microarray chips ( SAM 1 . 1 , SAM 2 . 0 , and SAM 3 . 0 ) , which were generated specifically for use in microarray analyses of the maize SAM ( see Materials and Methods for descriptions of SAM chip contents; further details are provided at http://www . plantgenomics . iastate . edu/maizechip ) . SAM 3 . 0 in particular contains 10 , 816 sequences obtained from maize shoot apices ( SAM plus four leaf primordia ) as part of a SAM expressed sequence tag ( EST ) discovery program performed during this project [16] . We performed six biological replicates , each comprised of ten laser-microdissected SAMs from mutant and nonmutant seedlings . For each array platform , three of the six pairs were measured with Cy3 from mutant SAMs and Cy5 dye for nonmutant . Dye assignments were reversed for the other three replications . Cy5 minus Cy3 differences were computed for each slide following normalization . The differences were used to test for evidence of differential expression between mutant and nonmutant SAMs using a linear model analysis for each gene ( see Materials and Methods ) [17] . With the intent of focusing on genes most likely to be differentially expressed , 56 genes with p-values <0 . 001were selected; ten additional genes were selected with p-values between 0 . 001 and 0 . 01 and fold changes greater than 1 . 5 or less than 0 . 67 ( Table 1 ) . Bioinformatic predictions of function were performed for all genes differentially expressed in ns mutant apices and are presented at Gene Expression and Visualization Application ( GENEVA , http://sam . truman . edu/geneva/geneva . cgi ) , a SAM gene-expression database created during this project [18] . A total of 11 functional categories are identified ( Figure 3 ) , including genes predicted to be involved in two-component signaling , auxin transport and signaling , jasmonate-induction/sugar signaling , intercellular transport , RNA processing , chromatin remodeling , transcription regulation , growth/cell division , ribosome structure , and general metabolism . Of the 66 genes identified in our microarray analyses , just one ( Zmhp1 ) has been characterized biochemically [19] , and none have been subjected to genetic analysis in maize . In addition , nine genes of unknown function are differentially expressed in ns mutant SAMs ( Table 1 ) . Excluding unknowns and genes predicted to be involved in general metabolism or “housekeeping” function , 28 out of the remaining 40 genes are predicted to be involved in some aspect of signal transduction or cell differentiation/growth . To ensure that the cells comprising the SAM lateral domains were not injured or degraded due to proximity to the ultraviolet laser during microdissection , internal controls included comparisons of ns2 transcript abundance in laser-microdissected SAMs versus whole seedlings . qRT-PCR and microarray data confirmed that ns2 expression is indeed enriched in the laser-microdissected SAM samples ( unpublished data ) , suggesting that the SAM lateral domains are not compromised by the ultraviolet laser-microdissection procedure . The identification of additional , previously uncharacterized gene expression within the SAM lateral domain ( Figures 4A , 5 , and 6; as described below ) provides further proof that these SAM domains were not destroyed during microdissection . qRT-PCR of cDNA prepared from RNA microdissected from ns and nonmutant SAMs corroborated the differential expression of 18 out of 22 genes tested , whose microarray fold changes were large enough to detect by qRT-PCR methodology ( i . e . , fold changes ≤0 . 67 and ≥1 . 5 ) ( Table 2 ) . Overall , the qRT-PCR and microarray data exhibited remarkable agreement in quantitative fold change between ns mutant and nonmutant SAMs ( Pearson correlation coefficient was 0 . 856 , r2 = 0 . 733 , p = 0 . 000006 ) . A total of four genes originally detected as differentially expressed in ns apices by microarray analyses could not be verified by qRT-PCR and were removed from further consideration . One such false positive gene encoding a predicted F-Box protein ( BM078718 ) may have been detected by cross-hybridization to a TIR1-like F-box gene ( CD001847 ) that was later verified as differentially expressed via in situ hybridization ( described below ) ( Figure 5I–5L ) . This example illustrates one drawback to the use of long cDNA arrays in expression profiling and underscores the importance of secondary verification of microarray data by qRT-PCR and in situ hybridization . As part of a larger effort ( http://maize-meristems . plantgenomics . iastate . edu ) to identify maize genes whose expression is enriched in shoot meristems , qRT-PCR analyses of transcript accumulation in a variety of maize tissues were performed for thirteen genes contained in Table 1 ( Figure 4; unpublished data ) . Maize tissues examined in these analyses included the vegetative SAM , immature tassel inflorescence , immature ear inflorescence , whole seedlings , expanded mature leaves , expanded seedling leaves , and seedling root ( see Materials and Methods ) . In addition , in situ hybridization was used to characterize the expression patterns of 14 differentially expressed maize genes , four of which exhibited weak or indiscernible staining and could not be interpreted . A total of ten genes examined via in situ hybridization of maize apices yielded interesting mRNA accumulation patterns ( Figures 5 and 6 ) ; six of which exhibited lateral SAM domain-specific differential expression and are thereby predicted candidate genes functioning within or nearby the NS expression foci ( Figure 1C ) during maize leaf initiation . For example , a predicted DEAD box helicase gene ( DV621960 ) is expressed in the lateral founder cell regions of the initiating nonmutant leaf primordium ( red arrows in Figure 5A ) , whereas no expression is noted in ns mutant sibling SAMs ( Figure 5A–5D ) . These in situ hybridization data are in agreement with our microarray data ( Table 1 ) , which indicated that expression of DV621960 is significantly down-regulated in ns apices ( p = 1 × 10−5 ) . Characterized by their shared ability to unwind RNA helices , the 32 DEAD box HELICASE proteins of Arabidopsis are implicated in a variety of RNA metabolic processes ( including transcription and pre-mRNA splicing , ribosome biogenesis and translation , gene expression , and meristematic cell division ) , although their functions as yet described are specific and noninterchangeable [20 , 21] . Indeed , qRT-PCR analyses of DV621960 transcript accumulation in various maize tissues reveal that expression of this maize DEAD box helicase-like gene is enriched in the SAM ( Figure 4A ) . Moreover , the lateral SAM domain-specific differential accumulation of DV621960 suggests that its expression is activated by NS function and is related to founder-cell recruitment . As described below , the lateral domain-specific differential expression of five additional maize genes ( CD001847 , DN210415 , Zmhp1 , DN221438 , and DN232668 ) in the ns mutant SAM implies that domain-specific hormonal and signaling mechanisms are important during maize leaf initiation . Two-component response regulators comprise an evolutionarily conserved signal transduction pathway involving the transfer of phosphate from a sensor HISTIDINE KINASE to a RESPONSE REGULATOR ( ARR ) effector molecule [22] . HISTIDINE PHOSPHOTRANSFER proteins mediate these exchanges , whereupon the phosphorylated ARR regulates the activation of specific cellular responses . In plants the two-component system is implicated in numerous developmental functions including responses to ethylene and cytokinin [23 , 24] , whereas pseudo-response regulators function to control circadian rhythms [25] . Previously , microarray analyses of WUS1 induction revealed that four Arabidopsis arr genes are direct targets of WUS1 transcriptional repression [26] . Both the predicted maize response regulator gene BG840771 and the maize phosphotransfer gene Zmhp1 are up-regulated in the ns mutant SAM ( Table 1 ) . qRT-PCR reveals that transcripts from the predicted response regulator gene BG840771 accumulate to 4 . 05-fold higher in the ns SAM ( Table 2 ) , which is consistent with our microarray data ( Table 1 ) . In addition , tissue-specific qRT-PCR revealed that transcripts of BG840771 are especially abundant in the vegetative SAM as compared to other maize tissues ( Figure 4B ) . These data further illustrate the utility of LM-microarray analyses to identify new genes whose expression and implicated function are likely to be enriched in the SAM . ZmHP1 has been shown to function in vitro as a phospho-donor for the maize response regulators ( ZmRRs ) ZmRR1 , ZmRR4 , ZmRR8 , and ZmRR9 , whereas accumulation of ZmHP1 in both the cytoplasm and nucleus is consistent with its predicted function during two-component signaling [19] . In situ hybridization of Zmhp1 reveals transcript accumulation at two lateral foci within the nonmutant SAM ( red arrows in Figure 5E and 5G ) , which mimics the ns expression domain ( Figure 2C ) [2] . In contrast , Zmhp1 transcripts accumulate throughout the ns mutant SAM ( Figure 5F and 5H ) ; this up-regulated mutant expression is consistent with our microarray data ( Table 1 ) . Polar auxin transport is required for leaf initiation and lateral margin development in plants [27 , 28] . Auxin is transported to sites of leaf initiation via the PINFORMED1 ( PIN ) family of efflux proteins [29–32] , a process that requires ADP-ribosylation factor ( ARF ) -GAP-mediated vesicular cycling of PIN proteins [33 , 34] . Auxin signaling involves targeted proteolysis of transcriptional regulators wherein the F-box protein TIR1 functions as an auxin receptor [35 , 36] . The two genes predicted to be involved in auxin biology are down-regulated in our analyses of ns mutant SAMs ( Table 1 ) , including a putative ARF-GAP encoding gene ( DY402633 ) with predicted orthology to the Arabidopsis van3/scf1 gene required for vesicle trafficking and PIN recycling [34 , 37] and the predicted maize orthologue of the tir1 auxin receptor ( CD001847 ) . Whereas in situ hybridizations reveal that Zm*tir1 is expressed throughout the nonmutant SAM as well as in the margins and vasculature of leaf primordia ( Figure 5I and 5K ) , Zm*tir1 transcript abundance is diminished specifically in the lateral domain of the ns mutant SAM ( red arrows in Figure 5J and 5L ) . These data suggest that auxin transport and auxin signaling are correlated with NS-mediated recruitment of leaf founder cells within this lateral SAM domain . A total of six genes predicted to be induced by jasmonate , a phytohormone functioning during plant development and defense [38] , are down-regulated in the ns mutant SAM ( Table 1 ) . Auxin signaling induces jasmonate responses and both hormones share common downstream signaling pathways [39–41] , suggesting that down-regulation of jasmonate responses in ns apices might be related to defects in auxin signaling . Notably , all the jasmonate-induced genes identified herein encode putative lectins , carbohydrate-binding receptor proteins that are implicated during sugar transport in plants [42] . qRT-PCR corroborated the differential expression of four jasmonate-induced genes ( Table 2 ) , and in situ hybridization of a putative jacalin-related LECTIN encoding gene ( DN210415 ) reveals a butterfly-shaped expression pattern at the insertion of the leaf primordium into the apex ( Figure 5M and 5O ) . The “wings” of this expression pattern ( red arrows in Figure 5M ) are diminished or absent in ns mutants ( Figure 5N and 5P ) , revealing down-regulated lectin mRNA accumulation in lateral domains of the mutant apex . Although transcript accumulation in leaf primordia was not measured in our microarray analyses , DN21415 transcripts are also detected in the vasculature of nonmutant leaves and are not detected in ns mutant leaf primordia . Notably , the apparent Arabidopsis orthologue of this particular jacalin-related lectin gene is known to be a direct target of the NS-related WOX protein WUS1 [26] . Furthermore , a putative sugar transporter-encoding gene ( DY400928 ) was qRT-PCR-verified as down-regulated ( 0 . 11-fold compared to nonmutant ) in the ns mutant SAM ( Table 2 ) , implying further that sugar transport and/or sugar signaling is required for NS function . In situ hybridization reveals accumulation of this putative sugar-transporter transcript in initiating leaf primordia and in leaf vascular traces within the nonmutant apex ( Figure 6E–6H ) . Transcriptome analyses of ns mutant SAMs suggest that numerous GTP-binding proteins are involved during NS1 signaling within the initiating maize leaf . A total of six genes predicted to encode GTP-binding proteins are significantly misexpressed in the ns mutant ( Table 1 ) , including genes predicted to be involved in vesicle trafficking ( three Rab GTPases and an ARF GTPase ) , signal transduction ( a heterotrimeric G protein ) , and cell growth and division ( a GTP1/OBG family GTPase ) [43] . In the nonmutant SAM , transcripts encoding a putative Rab-class ARF-GTPase ( DN221438 ) are detected in the meristem center , as well as in lateral stripes overlapping the NS functional domain ( Figure 5Q and 5R ) . In the ns mutant SAM , the lateral expression of DN221438 is wider and more pronounced ( red arrows in Figure 5R and 5T ) , consistent with the up-regulation observed in microarray hybridizations and implicating NS during negative regulation of this putative ARF-GTPase encoding gene . Likewise , a second predicted Rab-class GTPase encoding gene ( DN232668 ) homologous to the Arabidopsis GTP-binding protein gene ara3 [44] is also up-regulated in the ns mutant SAM ( Table 1 ) . In situ hybridization reveals that the ara3-like GTPase ( DN232668 ) gene is expressed in the midrib/central founder-cell domain of the nonmutant SAM ( red arrows in Figure 6A and 6C ) , whereas in mutant apices loss of NS function results in the expansion of this expression domain into the lateral SAM domain and thereby encompasses the entire founder-cell ribbon ( Figure 6B and 6D ) . We found two maize genes with predicted functions in chromatin remodeling ( AI820200 and DN224375 ) that are significantly differentially expressed in ns mutant apices ( Table 2 ) . In situ hybridizations reveal that one such gene , encoding a predicted amine oxidase implicated in histone modification [45] , is unexpectedly up-regulated in the midrib/central domain as well as in the lateral domain of ns mutant founder cells ( Figure 6I–6L ) . A maize ftsZ-related gene ( CB381550 ) is significantly down-regulated in the ns mutant SAM ( Table 1 ) and shares homology with Arabidopsis genes encoding tubulin-like , structural proteins required for cell division in chloroplasts [46] . In situ hybridizations of nonchlorophyllic maize apices reveal ftsZ-like transcript accumulation in actively dividing tissues including young leaf primordia and leaf founder cells , as well as in the SAM apical tip ( Figure 6M and 6N ) . Although statistical parameters suggest robust support for differential expression of this gene in each of our six biological replicate samples ( p = 3 . 8 × 10−5 ) , no domain-specific differences in CB381550 expression are observed in situ hybridizations of ns samples . It is likely that in the absence of domain-specific changes in mRNA localization , our in situ hybridization protocols are unable to discriminate a 0 . 72-fold quantitative change in transcript accumulation . Lastly , a previously undescribed maize yabby-like gene ( CD650947 ) , the putative orthologue of the drooping leaf gene of rice [47] , is also down-regulated in our analyses of ns mutants . In nonmutant apices yabby-like transcripts accumulate in newly initiating leaf primordia ( Figure 6O and 6P ) , an expression pattern that is consistent with the predicted role of YABBY proteins during initiation and expansion of lateral organ primordia [47–49] . No domain-specific changes in mRNA accumulation are noted in ns apices , however qRT-PCR corroborates the down-regulation of this maize yabby-like gene in the ns mutant SAM that was observed in microarray analyses ( Table 2 ) . Furthermore , tissue specific qRT-PCR reveals that transcripts of the yabby-like gene CD650947 accumulate in the SAM as well as in shoot meristem-enriched tissues such as the young tassel and young ear , each of which bears numerous spikelet and spikelet pair meristems ( Figure 4C ) .
The power of LM to focus microarray comparisons to small developmental fields is demonstrated in these analyses of ns mutant and nonmutant SAMs . Among the genes represented in this relatively small dataset are a number of likely candidates implicated during NS-mediated leaf initiation and predicted to function in hormonal signaling , signal transduction , or growth ( discussed below ) . Although the primary developmental function of NS is shown to be localized in the shoot apex during recruitment of leaf founder cells in a lateral domain of the SAM [1 , 6 , 7] , loss of NS function is predicted to cause widespread changes in gene expression in the growing seedling owing to enormous differences in the differentiation and expansion of marginal/lateral leaf tissues that happens during normal leaf development downstream of NS function ( Figure 1A and 1B ) . The differential gene expression that ensues in mutant leaves following the ns-induced leaf domain deletion event is unlikely to address our experimental question , namely , the mechanisms of founder cell recruitment in the SAM . Microarray analyses of the ns SAM provide a resource for new gene discovery; of the 66 differentially expressed genes identified in these analyses , all but one are previously undescribed maize genes , and at least three exhibit enriched transcription in maize shoot meristematic tissues ( Figure 4 ) . In addition , six candidate genes chosen for in situ hybridization analyses exhibited differential expression within or overlapping the SAM lateral domain of NS function and expression ( Figures 1 , 5 , and 6 ) [1 , 2] . Our data illustrate the combined utility of in situ hybridization and LM technology to focus microarray comparisons to a discrete developmental field , thereby limiting the number of identified differentially expressed genes to those transcribed in close vicinity to the domain of NS function . Although the use of six biological replicates enabled the selection of ns differentially expressed genes , it is noted that the majority of the candidate genes represented in Table 1 exhibit relatively modest fold changes in expression level . This result was expected , considering that the NS micro-domain of expression and function comprises a relatively small number of cells in the lower lateral region of the maize shoot apex ( Figure 1C ) [1 , 2] . During SAM LM ( Figure 2 ) the RNA contribution of cells comprising the NS lateral domain is diluted by RNA collected from the rest of maize SAM , which may explain the relatively low fold changes observed in our microarray analyses . Unfortunately , without the use of technically prohibitive NS expression markers during LM of the SAM , reliable microdissection of the NS lateral micro-domain away from the rest of the SAM proper is not feasible . Excluding genes of unpredicted function , the majority of transcripts differentially expressed in the ns SAM ( 28/40 ) are putatively involved in some aspect of developmental signaling or growth regulation ( Table 1 ) . These include genes involved in two-component response pathways , auxin signaling , jasmonate-induced pathways , as well as GTP-binding proteins implicated during signal transduction or cellular trafficking . These results are consistent with previous models for NS function during transduction of a cell-autonomous , founder-cell recruitment signal required for maize leaf initiation . Especially intriguing are those genes whose expression domains mirror or overlap the NS lateral foci ( Figure 5; Figure 6A–6D ) . We hypothesize that these coexpressed genes may function closely downstream of NS . Subsequent reverse genetic and molecular/biochemical analyses of these implicated genes will test this hypothesis and help to elucidate the NS signaling pathway . Our microarray data and in situ hybridization analyses of the maize tir1-like gene ( CD001847; Figure 5I–5L ) suggest that auxin activity is involved in NS-mediated recruitment of leaf founder cells in the lateral domain of the maize SAM . Previous analyses among multiple laboratories implicate PIN-mediated auxin transport during leaf initiation , and leaf initiation correlates with knox gene down-regulation within the SAM [50–52] . More recent studies established a mechanistic link between these two correlated phenomena , illustrating that knox down regulation requires auxin [28 , 53] . Initial analyses of NS function revealed that ns mutants fail to down-regulate KNOX accumulation in the lateral domain of the SAM , which correlates with the failure to recruit founder cells from this meristem domain [6 , 7] . In light of the recently established links between knox down-regulation , leaf initiation , and auxin activity , our microarray analyses and Zm*tir1 expression data suggest that NS-mediated founder-cell recruitment and KNOX down-regulation require TIR-mediated auxin signaling within the SAM lateral domain . In addition , the expression of six genes predicted to encode jasmonate-induced lectins is consistently down-regulated in the ns mutant SAM ( Table 1 ) . Lectins are carbohydrate-binding proteins that function to facilitate the intercellular transport of sugars [42] . In situ hybridization of a jacalin-related lectin gene ( DN210415 ) reveals a pattern of transcript accumulation that spreads laterally at the insertion site of the newly initiated leaf primordium into the shoot ( Figure 5M–5P ) . The decreased lateral accumulation of DN210415 transcripts in ns mutant shoots suggests that NS promotes the novel expression pattern of this jasmonate-induced gene . Likewise , a predicted sugar-transporter gene ( DY400928 ) is under expressed in ns 1-R mutants ( Figure 6E–6H; Tables 1 and 2 ) , which further implicates a role for carbohydrate transport during NS function . Numerous studies reveal a hormone-like role for sugar signaling in plants , in which carbohydrate transport regulates gene transcription [54] . Subsequent analyses are required to determine if this putative sugar transporter and the jasmonate-induced lectins perform metabolic functions or signaling functions during NS-mediated founder-cell recruitment . At present , ns1 and its duplicated paralogue ns2 are the only maize wox genes for which genetic analyses of function are described . Arabidopsis includes 15 WOX family members [55] , seven of which have been subjected to genetic analyses ( WUS1 [56]; PRESSED FLOWER/WOX3 [57]; WOX2 [55]; PRETTY FEW SEEDS2/WOX6 [58]; STIMPY/WOX9 [59]; WOX5 [60]; WOX4 , J . Ji and M . J . Scanlon , unpublished data ) . Although the phenotypes of individual WOX mutants are varied ( affecting shoot and root meristems , embryogenesis , and organogenesis of lateral organs and the vascular procambium ) , the combined genetic and molecular expression data suggest that an evolutionarily conserved general function of WOX proteins is to promote the organization of embryonic/meristematic cells or lateral organ initials . For example , WUS1 organizes proliferation in the central zone of Arabidopsis shoot meristems via repressing the transcription of several two-component response regulator genes ( ARR5 , ARR6 , ARR7 , and ARR15 ) , which function to reduce SAM size [26] . Meanwhile , our analyses demonstrate that NS is required to repress the expression of two maize genes predicted to function in two-component signaling pathways that operate within the maize SAM ( Figures 4A and 5E–5H; Table 1 ) . These include a SAM-enriched response regulator-like gene ( BG840771 ) as well as the maize histidine phosphotransfer gene Zmhp1 , whose nonmutant RNA accumulation pattern ( Figure 5E and 5G ) mirrors that of the ns duplicate genes ( Figure 1C ) . Taken together , these data suggest that transcriptional repression of specific , two-component signaling pathways may be conserved functions of WUS1 and NS . As likewise observed in our studies of ns meristems ( Figure 5M–5P; Table 1 ) , microarray analyses of WUS-induced gene expression also revealed the up-regulation of numerous jasmonate-induced lectin genes [26] , including an apparent homolog of the maize jacalin-related lectin gene ( DN210415 ) , which is down-regulated in the ns mutant SAM . Although preliminary , these analyses suggest that NS and WUS1 may share a conserved WOX function to activate the expression of specific lectin genes during plant development . The LM-microarray analyses described herein provide a starting point toward reverse genetic and biochemical analyses of the mechanisms of NS-mediated founder cell recruitment during maize leaf initiation .
The ns 1:1 line was propagated by crossing ns mutant plants ( genotype ns1-R and ns2-R ) onto nonmutant siblings ( genotype Ns1/ns1-R and ns2-R ) for over twenty successive generations as described [5 , 6] . Seedlings of this near-isogenic ns 1:1 line were grown in an environmentally controlled chamber with light intensity 220–250 μES−1M−2; 25 °C for 15 h of light; 20 °C for 9 h of dark; 50% for humidity , and harvested for LM at 2 wk after germination . Seedlings were fixed in acetone and paraffin-embedded as described [16] . SAM cells were laser microdissected from 10-μm sections ( ten to 12 sections per SAM ) using the P . A . L . M . Laser Microbeam ( http://www . palm-microlaser . com ) . Expression of the lateral SAM domain control gene ns2 was highly enriched in our laser-dissected SAM samples compared to that of the whole seedlings , as monitored by RT-PCR and microarray ( unpublished data ) . We used six biological replicates in these experiments , each comprised of ten to 12 whole ns or nonmutant laser microdissected SAMs ( ranging from 2 . 4 mm2 to 4 . 2 mm2 of tissue ) ( Table S1 ) . RNA was isolated using the PicoPure RNA extraction kit ( Arcturus Molecular Devices , http://www . moleculardevices . com ) , and two rounds of RNA amplification were performed using T7-RNA polymerase as described [11] with changes described in [61] . Yields of amplified SAM RNA ranged from 16 . 7 μg to 57 . 6 μg quantities . We reverse transcribed two μg of amplified SAM RNA with Superscript II ( Invitrogen , http://www . invitrogen . com ) and 0 . 5 μg of random primers ( Roche Diagnostics , http://www . roche . com ) . The resultant cDNA were indirectly labeled with Cy dyes assisted by amino allyl incorporation as described [11]; dye bias was removed by swapping Cy dyes between the RNA samples . Microarray hybridizations were performed as described [11] . Two technical problems prevented our use of laser microdissected , amplified RNA in hybridizations with the maize oligo arrays that are currently in production at the University of Arizona ( http://www . maizearray . org ) . First , amplified RNA prepared as described above is antisense orientated and therefore unusable with sense-directed oligo arrays . Although , linear RNA amplification protocols that generate sense-oriented RNA are available , these truncate the 3′ ends of the amplified RNA product . Because the majority of the maize oligo array sequences are generated from 3′ ends of maize ESTs , sense-amplified RNA generated by these protocols is also suboptimal . A second caveat is that the maize EST sequences that were used to design the maize oligo microarrays are underrepresented for sequences derived from the vegetative maize SAM; SAM-specific cDNA libraries were not deeply sequenced in previous maize EST projects . Therefore , we initiated a SAM EST discovery project in which cDNA was prepared from hand-dissected maize apices ( SAM plus P1–P4 ) and 31 , 036 apex ESTs were generated [16] and submitted to GenBank ( http://www . ncbi . nlm . nih . gov/Genbank ) . These include 3 , 503 SAM ESTs that are not found among EST libraries from any other maize tissues [16]; these SAM EST sequences are also not represented on the current maize oligo arrays . In light of the obstacles preventing our use of the maize oligo arrays , three microarrays ( SAM 1 . 1 , SAM 2 . 0 , and SAM 3 . 0 ) containing a combined total of 37 , 662 informative cDNAs including approximately 21 , 721 maize genes were constructed . SAM 1 . 1 contains 14 , 401 informative spots corresponding to ~9 , 423 maize cDNAs; SAM 2 . 0 contains 8 , 991 informative spots representing ~7 , 599 maize genes . Whereas SAM 1 . 1 and SAM 2 . 0 contain maize UniGenes [62] as well as cDNAs derived from maize inflorescences , the SAM 3 . 0 chip is particularly enriched for cDNAs derived from the maize SAM . Gene chip SAM 3 . 0 contains 14 , 270 informative spots and approximately 12 , 257 new genes , including more than 10 , 800 cDNAs derived from dissected maize apices and identified during this project ( described above and in [16] ) . All three SAM microarrays contain 45 control genes that are known to be expressed in the maize SAM and/or young leaf primordia . SAM chips may be ordered online ( http://www . plantgenomics . iastate . edu/maizechip ) , and their gene content may be searched via the online tool MADI ( MicroArray Data Interface , http://schnablelab . plantgenomics . iastate . edu:8080/madi ) . Hybridized arrays were scanned using a ScanArray 5000 ( GSI Lumonics , http://www . gsilumonics . com ) at 10-μm resolution . Image processing utilized Digital Genome System software ( MolecularWare , http://www . calbatech . com ) . Signals were background corrected and LOWESS normalized within each slide to remove intensity-dependent dye bias [63] . Normalization across slides was accomplished by median centering data from each channel [64] . Our model for the normalized log-scale signal intensities for any given gene is as follows: where yijk denotes the normalized log-scale signal intensity from SAM type i ( i = 1 , 2 for mutant and nonmutant SAMs , respectively ) , dye j ( j = 3 , 5 for Cy3 and Cy5 dyes , respectively ) , and slide k ( k = 1 , 2 , 3 , 4 , 5 , 6 ) ; μ is an intercept term; τi denotes the effect of SAM type i; δj denotes the effect of dye j; sk denotes the random effect of slide k; and eijk denotes a random residual term . The slide and residual random effects are assumed to be independent and normally distributed with a single variance for slides and a single variance for residuals . All parameters are allowed to vary from gene to gene though we have suppressed a gene-specific subscript to simplify notation . To obtain tests of SAM type effects ( H0 : τ1 = τ2 ) , the difference between normalized signals ( Cy5 minus Cy3 ) was computed for each spot . Based on our model ( 1 ) above , the six differences ( denoted d1 , . . . , d6 ) can be modeled as dk = β0 + β1 xk + ɛk , where β0 = δ5 − δ3 , β1 = τ1 − τ2 , ɛk is a difference of the form eijk − ei'j'k , and x1 , . . . , x6 are −1 , −1 , −1 , 1 , 1 , and 1 to correspond to our design in which three of the slides have Cy3 and Cy5 assigned to mutant and nonmutant SAMs , respectively , and three have the opposite assignment . In this simple linear regression model , the intercept term accounts for gene-specific dye effects not removed in normalization ( δ5 − δ3 ) , and the slope term accounts for the SAM type effect of interest ( τ1 − τ2 ) , which corresponds to the mean difference in normalized log-scale expression between mutant and nonmutant SAMs . The resulting p-values from the tests for SAM type effects were converted to q-values using the method of Storey and Tibshirani [65] to estimate the false discovery rate ( FDR ) associated with any p-value threshold for significance . Functional annotation of differentially expressed genes was performed as described [18] . MIAME guidelines utilized in these experiments are described in Text S1; all microarray data are available at Gene Expression Omnibus ( GEO; http://www . ncbi . nlm . nih . gov/geo ) . qRT-PCR analyses were performed on cDNA synthesized from the identical SAM-amplified RNA samples used in our analyses , using either Taq-Man or SYBR-Green probes ( Table S2 ) [66] . We used three biological replicates , upon which three technical replicates were performed . Reactions were normalized to control ubiquitin expression as described [67] . Tissue-specific qRT-PCR analyses were performed as above using the SYBR-Green methodology and gene-specific probes ( Table S1 ) ; three technical replicates were performed . Expression of each transcript was normalized to the level of ubiquitin controls; relative gene expression values were graphed using the iQ5 Optical System Software version 1 . 0 ( Bio-Rad , http://www . bio-rad . com ) , wherein no expression was assigned a value of zero . All tissues were derived from the maize inbred B73 . Tissues included the laser microdissected SAM from 14-d-old seedlings; fully expanded mature leaf ( leaf 10 ) ; fully expanded leaf from 14-d-old seedlings ( leaf 4 ) ; seedling roots; immature ears ( 6-mm long ) containing multiple spikelet meristems and spikelet pair meristems with glume primordia; and immature tassels ( 8 mm ) containing branch primordia , multiple spikelet meristems , and spikelet pair meristems with glume primordia . Except for LM-derived SAM tissues ( RNA extracted and amplified as above ) , all RNA extractions of maize tissues were performed using the Trizol method , as described [2] . Maize 14-d-old seedlings were grown in controlled conditions ( above ) and processed for in situ hybridization as described [49] . For each gene-specific probe analyzed , at least six replicate samples each of ns1-R mutant and nonmutant sibling were analyzed . Cartoons of transcript accumulation patterns modeled in Figures 5 and 6 depict SAM expression only . Expression in leaf primordia , which was not measured in our microarray analyses , is not depicted in cartoons .
The Gene Expession Omnibus ( GEO ) database ( http://www . ncbi . nlm . nih . gov/geo ) accession numbers for the genes discussed in this paper are ns , GSE7248; knotted1 , AY260164; and Zmhp1 , DN233962 . | Unlike animals , plants exhibit a prolonged period of organogenesis , generating new leaves throughout their life cycle . This ability to maintain an embryo-like state is dependent upon the activity of shoot meristems , whose dual functions are to supply an inner core of pluripotent cells that sustain the shoot meristem while simultaneously generating new leaves derived from cells at the meristem periphery . Deciphering the complex combinations of molecular signals that transform meristematic cells into leaf primordia is a central question in plant developmental biology . In this study , we used the power of focused laser light to microdissect shoot meristems from neighboring leaf and stem tissue in the maize plant . Once isolated , we compared patterns of gene expression in normal shoot meristems to those of genetically mutant shoot meristems that form abnormal , narrow leaves . Out of more than 21 , 000 maize genes analyzed , 66 genes were identified as misexpressed in the mutant shoot meristems . All but one of the differentially expressed genes are previously unstudied in maize , and the majority are predicted to function during cell division , growth , or developmental signaling . Many of these novel genes are expressed in specific domains of the shoot meristem , consistent with their predicted function during maize leaf initiation . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods",
"Supporting",
"Information"
] | [
"none",
"plant",
"biology",
"zea",
"eukaryotes",
"plants"
] | 2007 | Laser Microdissection of Narrow Sheath Mutant Maize Uncovers Novel Gene Expression in the Shoot Apical Meristem |
Host innate immune defences play a critical role in restricting the intracellular propagation and pathogenesis of invading viral pathogens . Here we show that the histone H3 . 3 chaperone HIRA ( histone cell cycle regulator ) associates with promyelocytic leukaemia nuclear bodies ( PML-NBs ) to stimulate the induction of innate immune defences against herpes simplex virus 1 ( HSV-1 ) infection . Following the activation of innate immune signalling , HIRA localized at PML-NBs in a Janus-Associated Kinase ( JAK ) , Cyclin Dependent Kinase ( CDK ) , and Sp100-dependent manner . RNA-seq analysis revealed that HIRA promoted the transcriptional upregulation of a broad repertoire of host genes that regulate innate immunity to HSV-1 infection , including those involved in MHC-I antigen presentation , cytokine signalling , and interferon stimulated gene ( ISG ) expression . ChIP-seq analysis revealed that PML , the principle scaffolding protein of PML-NBs , was required for the enrichment of HIRA onto ISGs , identifying a role for PML in the HIRA-dependent regulation of innate immunity to virus infection . Our data identifies independent roles for HIRA in the intrinsic silencing of viral gene expression and the induction of innate immune defences to restrict the initiation and propagation of HSV-1 infection , respectively . These intracellular host defences are antagonized by the HSV-1 ubiquitin ligase ICP0 , which disrupts the stable recruitment of HIRA to infecting viral genomes and PML-NBs at spatiotemporally distinct phases of infection . Our study highlights the importance of histone chaperones to regulate multiple phases of intracellular immunity to virus infection , findings that are likely to be highly pertinent in the cellular restriction of many clinically important viral pathogens .
The intrinsic , innate , and adaptive arms of host immunity cooperatively supress the replication and spread of invading viral pathogens . Conferred by constitutively expressed host factors , intrinsic immunity ( also known as intrinsic antiviral resistance and cell autonomous immunity ) restricts virus replication from the outset of infection [1–4] . By way of contrast , activation of Pattern Recognition Receptors ( PRRs ) by microbial specific Pathogen-Associated Molecular Patterns ( PAMPs ) leads to the induction of innate immune defences and the coordinated upregulation of a broad repertoire of host antiviral genes , principally cytokines ( including interferons; IFNs ) and interferon stimulated gene ( ISG ) products [3–6] . This induced immune response confers an enhanced antiviral state to limit virus spread and prime adaptive immune responses . Accordingly , many viruses have evolved counter measures to antagonize intrinsic and innate immune defences to promote the efficient onset of replication , propagation , and transmission to new hosts . A component step in the activation of intrinsic and innate immune defences during herpesvirus infection is the rapid recruitment of host factors to infecting viral genomes , which enter the nucleus as naked linear dsDNA molecules [1 , 3 , 7 , 8] . Recent microscopy and biochemical evidence has shown that the sequential recruitment of host factors to nuclear infecting viral genomes plays an important role in the spatial and temporal ( spatiotemporal ) regulation of intrinsic and innate immune defences , through the assembly of viral genomes into chromatin that can be epigenetically silenced and the activation of innate PRRs , respectively [9–19] . Notably , many of these host factors have been shown to either reside or transiently associate with promyelocytic leukaemia nuclear bodies ( PML-NBs , also known as ND10; [9 , 11 , 17 , 20–26] ) , which rapidly associate with and entrap infecting viral DNA ( vDNA ) as a component of the intrinsic antiviral immune response [9 , 12 , 17 , 19 , 27–31] . Saturation of intrinsic host defences under high genome loads leads to the induction of innate immune defences in a PML- and IFI16 ( interferon gamma inducible protein 16 ) -dependent manner , which establishes an antiviral state to restrict virus propagation and spread [9 , 13 , 32–36] . Importantly , intrinsic host defences can inhibit the onset of HSV-1 lytic replication in restrictive cell types independently of the induction of innate immune defences [9 , 37] . The importance of PML-NBs in the regulation of intracellular immunity is highlighted by the fact that many DNA and RNA viruses have evolved independent strategies to disrupt these dynamic nuclear sub-domains during infection [1 , 3 , 38] . One of the first proteins expressed during Herpes Simplex Virus 1 ( HSV-1 ) infection is ICP0 , a viral RING-finger ubiquitin ligase with SUMO-Targeting Ubiquitin Ligase ( STUbL ) properties [23 , 39 , 40] . ICP0 mediates the degradation and dispersal of host factors , including PML and IFI16 [9 , 23 , 36 , 41–45] , away from infecting viral genomes to disable intrinsic and innate immune defences activated in response to infection [1 , 3 , 46] . Viral mutants that do not express functionally active ICP0 are highly susceptible to intrinsic silencing under low multiplicities of infection ( MOI ) and are hypersensitive to interferon ( IFN ) treatment [47–52] . The use of such mutants has been critical for the characterization of many aspects relating to the antiviral roles of PML-NBs in the regulation of intracellular immunity during virus infection . Recent studies have shown that the HIRA ( histone cell cycle regulator ) histone H3 . 3 chaperone complex , composed of HIRA , Ubinuclein 1 ( UBN1 ) , and Calcineurin-binding protein 1 ( CABIN1 ) , relocalizes to PML-NBs in response to replication-defective herpesvirus infection [17 , 18] . This host response has been linked to the intrinsic epigenetic silencing of viral gene expression through the HIRA-dependent deposition of variant histone H3 . 3 onto viral genomes at PML-NBs in restrictive cell types [17 , 18 , 53] . This finding is consistent with the replication-independent assembly of HSV-1 genomes into chromatin containing H3 . 3 shortly after nuclear entry ( < 1 hour post-infection ( hpi ) ; [16 , 18 , 19 , 53 , 54] ) . Importantly , infections with replication defective herpesviruses prematurely activate PPRs to trigger innate immune defences leading to the secretion of cytokines . This host defence is naturally antagonized by ICP0 during wild-type ( WT ) HSV-1 infection under physiological infection conditions [9 , 33 , 36 , 55–59] . Notably , exogenous cytokine stimulation ( IFN-β and IFN-γ ) can induce HIRA localization to PML-NBs independently of infection [17] , supporting a role for HIRA at PML-NBs as a component of the innate immune response . Exogenous IFN stimulation has been shown to enrich HIRA at gene bodies across the human genome , including a subset of ISGs [17 , 60 , 61] . However , the extent to which HIRA influences the activation or regulation of innate immunity during HSV-1 infection remains to be established . We therefore sought to investigate the spatiotemporal kinetics of HIRA localization to infecting HSV-1 genomes and PML-NBs in the sequential regulation of intrinsic and innate immune defences under productive lytic infection conditions pertinent to a clinical setting . We show that the recruitment of HIRA to infecting HSV-1 genomes , as a component of the intrinsic antiviral immune response , occurs with kinetics that are temporally distinct to those of PML , the principle scaffolding protein of PML-NBs [62 , 63] . Under productive infection conditions that promote the activation of innate immune defences , HIRA localized to PML-NBs in cells proximal to a developing infection in a Janus-Associated Kinase ( JAK ) dependent manner , identifying a role for virus-induced cytokine signalling in the recruitment of HIRA to PML-NBs . RNA-seq analysis revealed that HIRA was required for the efficient induction of cellular genes involved in the regulation of innate immunity , including MHC-I antigen presentation , cytokine signalling , and the expression of a broad repertoire of ISGs . ChIP-seq analysis revealed that PML promoted HIRA enrichment at ISG bodies , identifying a role for PML in the HIRA-dependent regulation of innate immune defences . Depletion of HIRA inhibited ISG expression , which led to a relief in the cellular restriction of ΔICP0 HSV-1 propagation . Our research uncovers distinct kinetics in the spatiotemporal recruitment of HIRA to infecting viral genomes and PML-NBs during different phases of HSV-1 infection and identifies dual roles for HIRA in the sequential regulation of intrinsic and innate immunity to herpesvirus infection . Our data demonstrates the importance of histone chaperones to regulate multiple phases of intracellular host immunity , findings likely to be highly pertinent in the cellular restriction of many clinically important viral pathogens .
Recent studies have reported that HIRA accumulates at PML-NBs in response to infection with replication-defective herpesviruses ( HSV-1 and HCMV ) [17 , 18] . This host response was linked to the intrinsic epigenetic restriction of viral gene expression at PML-NBs [17 , 18] . However , the stable localization of HIRA at PML-NBs was only observed at comparatively late times post-infection ( 2–6 days; [17 , 18] ) , which is atypical of intrinsic PML-NB host factors that silence viral gene expression from the outset of nuclear infection [9 , 19 , 27–31] . We hypothesized that this delayed recruitment may be indicative of innate immune defences activated in response to replication-defective herpesvirus infection . We therefore examined the kinetics of HIRA recruitment to infecting WT or ICP0-null mutant ( ΔICP0 ) HSV-1 genomes pre-labelled with EdC ( 5-Ethynyl-2’-deoxycytidine; HSV-1EdC and ΔICP0EdC , respectively ) and PML-NBs ( Fig 1 ) . Antibody validation experiments confirmed the specificity of the HIRA monoclonal antibody used for indirect immunofluorescence and western blotting in this study ( S1 Fig ) . Consistent with recent studies [9 , 10 , 64] , infecting HSV-1EdC genomes were readily detectible by click chemistry following virion release into the nucleus and to be stably associated with PML-NBs by 90 minutes post-infection ( mpi ) , prior to their disruption and release by ICP0 by 180 mpi ( Fig 1A and 1B; [9 , 10] ) . At these times , HIRA did not stably localize to either vDNA or PML-NBs ( Fig 1A–1C ) . It is possible that low levels of ICP0 expression were sufficient to disperse HIRA away from vDNA prior to the disruption of PML-NBs [9] . To test this , HIRA localization was evaluated during ΔICP0EdC HSV-1 infection . While HIRA was observed to localize with vDNA and PML-NBs on rare occasions ( Fig 1A , ΔICP0EdC ) , quantitation revealed that the frequency of this colocalization was below coincident threshold levels ( weighted colocalization frequency < 0 . 2 , dotted line; Mock ) at both 90 and 180 mpi ( Fig 1B and 1C ) . In contrast , PML remained stably colocalized with ΔICP0EdC vDNA ( median weighted colocalization frequency > 0 . 8 , Fig 1B; [9] ) . Genome counts demonstrated that HSV-1EdC and ΔICP0EdC infected nuclei contained equivalent input genome loads ( median average 2–3 genomes per nuclei , Fig 1D ) . As bio-orthogonal nucleic acid labelling is not compatible with live-cell kinetic studies , we cannot rule out the possibility of highly transient or low antigen abundance HIRA interactions with either vDNA or PML-NBs . We conclude that HIRA does not stably colocalize with input vDNA or PML-NBs under infection conditions in which key intrinsic PML-NB restriction factors ( PML , Sp100 , Daxx , and ATRX; [9] ) rapidly entrap and silence vDNA following its nuclear entry . As ICP0 induces the proteasome-dependent degradation of host restriction factors [12 , 45 , 46] , we next examined the stability of HIRA over the course of WT or ΔICP0 HSV-1 infection . In contrast to PML , which is a well-characterized substrate of ICP0 [23 , 39–44 , 65] , western blot analysis of infected whole cell lysates revealed that HIRA protein levels remained stable throughout infection ( Fig 1E ) . We conclude that HIRA is not a substrate for ICP0 mediated degradation . As nuclear infection in and of itself was not sufficient to induce the stable recruitment of HIRA to vDNA or PML-NBs by 180 mpi , we next evaluated HIRA localization at PML-NBs under infection conditions that enabled the induction of innate immune defences [9] . To this end , cell monolayers were infected at a low-multiplicity of infection to enable the formation of viral plaques by 24 hpi ( Fig 2 , S2 and S3 Figs ) . Under these infection conditions , cells on the periphery of developing plaques would be exposed to innate immune signals derived from the infected cells within the body of the adjacent plaque . Cells were infected with recombinant viruses that express eYFP . ICP4 ( the major immediate early viral transcription factor ) to enable visual identification of productively infected cells , where eYFP-ICP4 accumulates in viral replication compartments prior to cytoplasmic export ( Fig 2A , S2 and S3 Figs; red arrows ) . Within mock- or WT HSV-1 infected cell monolayers , HIRA remained dispersed throughout the nucleus with little to no localization at PML-NBs , either within uninfected ( no detectable ICP4 expression ) or infected ( detectable ICP4 expression ) cells ( Fig 2B and 2C , S2 Fig ) . In contrast , in cell monolayers infected with ΔICP0 HSV-1 HIRA stably localized at PML-NBs in both ICP4 positive and negative cells ( Fig 2A , S3 Fig; red and white arrows , respectively ) . Up to 80% of ICP4 negative cells within ΔICP0 HSV-1 infected monolayers had stable HIRA localization at PML-NBs at 24 hpi ( Fig 2B and 2C ) , with HIRA detectable at the majority of individual PML-NBs within any given nucleus ( Fig 2B ) . These data are consistent with the cytokine-mediated induction of signal-transduction cascades stimulating the recruitment of HIRA to PML-NBs in cells proximal to a developing ΔICP0 HSV-1 infection; a host response antagonized by ICP0 during WT HSV-1 infection through the degradation and dispersal of innate immune regulators , including the vDNA PRR IFI16 , from infecting viral genomes which suppresses the induction of IFN expression [9 , 36 , 57 , 58] . To test the role of JAK-STAT signalling in the relocalization of HIRA to PML-NBs we examined HIRA localization in ΔICP0 HSV-1 infected cell monolayers treated with the JAK inhibitor Ruxolitinib ( Ruxo ) [9 , 66 , 67] . Ruxolitinib treatment effectively abolished the stable localization of HIRA at PML-NBs in ΔICP0 HSV-1 infected cell monolayers ( Fig 2D , S4 Fig ) . We conclude that virus-induced cytokine signalling stimulates the recruitment of HIRA to PML-NBs in response to an emerging HSV-1 infection in a JAK-dependent manner . These data are consistent with findings reported by Rai et al . , which show exogenous IFN stimulation to promote the stable localization of HIRA at PML-NBs independently of virus infection [17] . Taken together , these data suggest a role for HIRA at PML-NBs as a component of the host innate immune response to prime cells for imminent infection [32 , 33 , 68] . To investigate the potential roles of HIRA localization at PML-NBs as a component of the innate immune response , we first corroborated and extended HIRA localization studies to PML-NBs following exogenous IFN stimulation ( Fig 3; [17] ) . Stimulation of hTERT immortalized diploid foreskin fibroblast ( HFt ) cells with type-I ( IFN-β ) or type-II ( IFN-γ ) IFN efficiently induced the recruitment of HIRA to PML-NBs in 80% of cells by 24 h post-treatment ( Fig 3A and 3C–3E ) . These data confirm that HIRA recruitment to PML-NBs can occur independently of cellular senescence and virus infection [17 , 69–71] . The recruitment of HIRA to PML-NBs did not alter the sub-cellular localization of the resident PML-NB histone H3 . 3 chaperone Daxx ( death domain associated protein 6 , [62 , 72 , 73]; Fig 3B and 3E ) . The accumulation of HIRA at PML-NBs following IFN treatment also occurred independently of the induction of HIRA transcription ( Fig 3F ) , confirming that HIRA is not an ISG [17] . HIRA recruitment to PML-NBs following IFN-β stimulation occurred in a range of primary ( IMR-90 , MRC5 , and HFs ) and immortalized ( MRC5t , HFt , RPE , and HaCat ) fibroblast and epithelial cell types ( Fig 3G ) . In contrast , IFN stimulation of carcinoma cells ( U2OS , SAOS , HeLa , and A549 ) failed to induce the recruitment of HIRA to PML-NBs ( Fig 3G , S5 Fig; [17] ) , even though these cell types expressed similar levels of HIRA mRNA and protein within the nucleus ( Fig 3H , S5 Fig . ) . Collectively , these data corroborate and extend findings by Rai et al . [17] , showing that HIRA localization at PML-NBs in response to IFN stimulation occurs in a cell-type dependent manner and is independent of the induction of cellular senescence . Phosphoproteomic studies have shown HIRA to be extensively phosphorylated ( https://www . phosphosite . org/proteinAction ? id=1279&showAllSites=true ) , with in vitro studies supporting a role for cyclin-dependent kinase 2 ( CDK2 ) and glycogen synthase kinase 3β ( GSK-3β ) phosphorylation in the regulation of HIRA H3 . 3 chaperone activity and sub-cellular localization [70 , 74] . HIRA phosphorylation by GSK-3β has been linked to its localization at PML-NBs during cellular senescence [70] . We therefore examined the potential role of phosphorylation in the recruitment of HIRA to PML-NBs following IFN-β stimulation . Inhibition of GSK-3β only had a modest effect on HIRA localization at PML-NBs in response to cytokine stimulation ( 250 nM GSKi , CHIR-99021; IC50 ≤ 10 nM; Fig 3I and 3J ) , at doses sufficient to inhibit GSK-3β dependent cell cycle arrest ( S6 Fig; [75] ) . In contrast , HIRA remained nuclear diffuse following IFN-β stimulation in cells treated with Flavopiridol ( ≥ 50 nM; IC50 ≤ 200 nM; Fig 3I and 3K ) , a broad-spectrum CDK inhibitor ( CDKi ) . Notably , Flavopiridol is also known to inhibit RNA polymerase II transcript elongation through the inhibition of P-TEFb ( CDK9/cyclin T1 ) [76–78] . We conclude that CDKs , either directly or indirectly , influence HIRA recruitment to PML-NBs following IFN-β stimulation . Further studies are warranted to determine the role of CDK phosphorylation in the regulation of HIRA recruitment to PML-NBs in response to cytokine signalling . We next evaluated HIRA localization at PML-NBs in cells depleted of individual PML-NB constituent proteins to test their respective roles in the recruitment of HIRA following IFN stimulation ( Fig 4 , S7 Fig ) . HFt cells were stably transduced with lentiviral vectors expressing non-targeting control ( shCtrl ) or PML- , Sp100- , Ubc9- , Daxx- , or ATRX-targeting shRNAs ( shPML , shSp100 , shUbc9 , shDaxx , and shATRX , respectively; as described [23 , 27–29 , 79] ) . Depletion of core PML-NB constituent proteins did not reduce HIRA mRNA expression or alter its nuclear distribution relative to that of control cells ( Fig 4A and 4B ) . Following IFN-β stimulation , HIRA localized at PML-NBs in cells depleted of Daxx and ATRX ( α-thalassemia mental retardation X-linked protein; Fig 4C and 4D ) , demonstrating that HIRA recruitment to PML-NBs can occur independently of this resident histone H3 . 3 chaperone complex [62 , 72 , 73] . In contrast , HIRA localization at PML-NBs was abrogated in the absence of PML , the main scaffolding protein of PML-NBs [62 , 63] , or Sp100 ( Fig 4C and 4D , S7A Fig ) . In the absence of either PML or Sp100 , HIRA did not substantially accumulate into nuclear foci following IFN stimulation . As Sp100 is not required for PML-NB formation [62] , these data demonstrate that HIRA accumulates at pre-existing PML-NBs , as opposed to nucleating the accumulation of PML-NB proteins from the nucleoplasm at HIRA foci . As covalent and non-covalent interactions with SUMO play an integral role in mediating many protein-protein interactions associated with PML-NBs [23 , 63 , 80] , we examined the role of SUMOylation in the localization of HIRA at PML-NBs . Disruption of the SUMO pathway through the depletion of Ubc9 , the sole E2 SUMO conjugating enzyme , is known to inhibit PML and Sp100 SUMO-modification leading to the formation of enlarged PML nuclear aggregates ( one to two foci per nucleus ) that contain Sp100 [23] . Following IFN stimulation , HIRA was readily observed to localize at PML nuclear aggregates in Ubc9 depleted cells ( Fig 4C and 4D ) . As Sp100 has been shown to interact with PML in a SUMO-modification independent manner and not to influence PML SUMO-modification directly [28 , 81] , these data suggest that HIRA recruitment to PML-NBs occurs in a manner that is dependent on Sp100 but independent of de novo covalent SUMO-modification . Correspondingly , the molecular mass of endogenous HIRA remained unaltered in HFt cells treated with IFN-β ( S7C Fig ) , indicating that HIRA itself is not extensively SUMOylated to promote its localization at PML-NBs in response to cytokine stimulation . Collectively , these data suggest that Sp100 plays an important role in mediating the recruitment of HIRA to PML-NBs in response to cytokine signalling . However , in contrast to immunoprecipitation studies that have shown HIRA to interact with PML [82] , we were unable to detect an interaction between HIRA and Sp100 following IFN stimulation ( S7D Fig ) . Thus , the localization and retention of HIRA at PML-NBs in response to cytokine stimulation is likely to be multifactorial , indicative of the complex network of interactions associated with PML-NBs which rapidly alter in number , size , and distribution is response to cytokine stimulation [38 , 83] . Notably , a minor population of IFN stimulated cells could be observed to contain enlarged ‘donut shaped’ foci that contained both PML and Sp100 but not HIRA ( Fig 4C , S7B Fig; inserts ) . While the significance of this minor population of enlarged foci remains to be determined , these data highlight that additional factors are likely to influence HIRA recruitment to PML-NBs in response to IFN stimulation . We conclude that PML and Sp100 are both required to mediate the recruitment and retention of HIRA at PML-NBs in the majority of foci . However , further studies are warranted to identify the precise network of protein-protein interactions required to facilitate the stable localization of HIRA at PML-NBs following IFN stimulation . In order to investigate the role of HIRA in the regulation of intracellular immunity during virus infection , we conducted RNA-seq analysis in HSV-1 infected cells depleted of HIRA ( Fig 5 ) . HFt cells were stably transduced with lentiviral vectors expressing non-targeting control or HIRA-targeting shRNAs ( shCtrl and shHIRA , respectively ) . qPCR and western blotting confirmed HIRA depletion without influencing PML expression levels ( Fig 5A and 5B , S1 Fig ) . Cells were mock treated ( no treatment , nt ) , or either treated ( t ) with IFN-β ( 100 IU/ml ) or infected with WT or ΔICP0 HSV-1 ( MOI 1 PFU/cell ) for 17 h prior to RNA extraction for next-generation sequencing ( NGS ) . High confidence reads ( FDR Q-value ≤ 0 . 0001 , ≥ log2 fold change ) were used for gene expression and gene ontology ( GO ) analysis ( Fig 5C–5G ) . As expected , treatment with IFN-β or infection with HSV-1 ( WT or ΔICP0 ) induced a strong transcriptional response in comparison to non-treated control cells , with an enrichment in upregulated genes involved in immune system regulation ( Fig 5C and 5F [shCtrl ( + ) log2 fold change]; S1 Table ) . Depletion of HIRA alone led to a significant shift in gene expression ( Fig 5D ) , with an enrichment in upregulated genes involved in developmental biology , muscle contraction , and neuronal systems ( Fig 5E ( + ) log2 fold change; S2 Table ) , and a downregulation of genes involved in extracellular matrix organization ( Fig 5E ( - ) log2 fold change ) . Relative to control cells , HIRA depleted cells did not significantly upregulate the expression of genes involved in immune system regulation in response to IFN treatment or infection with HSV-1 ( WT or ΔICP0; Fig 5F shCtrl to shHIRA ( + ) log2 fold change , dotted box; S1 and S3 Tables ) . These data suggest that HIRA directly or indirectly promotes the transcriptional upregulation of genes involved in the immune response to HSV-1 infection or IFN-β stimulation . An expanded analysis of the immune system node revealed that infection ( WT or ΔICP0 HSV-1 ) of HIRA depleted cells resulted in a significant downregulation in the expression of genes involved in MHC class I antigen presentation , neutrophil degranulation , and interferon signalling relative to infected control cells ( Fig 5G [ ( - ) log2 fold change]; S4 and S5 Tables ) . These data indicate that HIRA plays a critical role in the transcriptional regulation of a wide variety of genes involved in the immune response to virus infection . Correspondingly , transcriptome profiling of ISG expression revealed that this group of immuno-regulatory genes was downregulated in HIRA depleted cells treated with IFN or infected with HSV-1 ( WT or ΔICP0 ) relative to treated or infected control cells ( Fig 5H , S6 Table ) [84] . Notably , HSV-1 infection of HIRA-depleted cells resulted in a higher degree of ISG downregulation relative to that observed in HIRA-depleted cells treated with IFN-β ( Fig 5H; red dots ) , highlighting that the nature of immune stimulus can differentially influence the transcriptional induction of ISGs . Taken together , we conclude that HIRA plays important roles in the transcriptional upregulation of a broad repertoire of genes involved in the innate immune response to HSV-1 infection or IFN stimulation . It is becoming evident that PML has independent roles in the regulation of intrinsic and innate immune defences during herpesvirus infection [3 , 9 , 32–34 , 68] . To test whether PML facilitates the HIRA dependent induction of innate immune defences , we performed ChIP-seq analysis to evaluate the distribution of HIRA across the human genome in cells in which PML was depleted or not ( Fig 6 ) . Primary fibroblast cells were stably transduced with lentiviral vectors expressing PML-targeting or non-targeting control shRNAs ( shPML or shCtrl , respectively; [27] ) . Western blot analysis confirmed PML depletion ( Fig 6A ) . Cells were either mock treated ( nt ) or treated ( t ) with IFN-β for 24 h prior to native ChIP and NGS , as described [69] . Peak calling revealed a genome wide increase in HIRA binding to host genes following IFN stimulation in both control and PML-depleted cells ( Fig 6B , S7 Table; [17] ) . Notably , the pattern of HIRA binding to genes was altered in IFN treated PML depleted cells in comparison to IFN treated control cells ( Fig 6C , condition ii vs iii; P < 0 . 001 ) . These data suggest a role for PML in the enrichment of HIRA binding to a specific subset of genes following IFN stimulation . In order to investigate this , we examined the levels of HIRA enrichment over a panel of equivalent sized interferon or non-interferon stimulated coding gene bodies ( 49 genes per condition; [84] ) . Following IFN-β treatment , the level of HIRA enrichment on ISGs was significantly reduced in PML depleted cells in comparison to IFN treated control cells ( Fig 6D–6F , S7 Table; a representative example for Mx1 is shown in Fig 6D ) . While a modest increase in HIRA enrichment was observed for non-interferon stimulated genes , the level of enrichment was not significant nor was it dependent on PML ( Fig 6G and 6H ) . Collectively , these data demonstrate that PML plays a role in the enrichment of HIRA onto ISGs following IFN stimulation . Taken together with our transcriptomic analysis ( Fig 5 ) , these data demonstrate that IFN treatment stimulates HIRA binding to the gene bodies of a variety of ISGs in a PML-dependent manner to facilitate their transcriptional upregulation . As HSV-1 infection can inhibit the termination of host gene transcription , which could influence the interpretation of our transcriptomic analysis ( Fig 5; [85] ) , we performed validation studies to analyse the expression of a subset of ISGs ( Mx1 , ISG54 , ISG15 , and OAS1 ) during WT or ΔICP0 HSV-1 infection in control and HIRA depleted HFt cells . Consistent with our transcriptomic analysis , ΔICP0 HSV-1 infection of control cells efficiently induced ISG transcription to levels equivalent to those observed for IFN-β treatment alone ( Fig 7A and 7B; dotted line ) . As expected , this host response was impaired during WT HSV-1 infection at both 9 and 17 hpi ( Fig 7A and 7B; dotted line ) . HIRA depleted cells infected with either WT or ΔICP0 HSV-1 had a significant decrease in ISG transcript levels relative to infected control cells ( Fig 7A and 7B ) . Western blot analysis of infected whole cell lysates demonstrated that expression of these ISGs was significantly reduced in HIRA depleted cells infected with HSV-1 ( WT or ΔICP0; Fig 7C and 7D ) . In agreement with our qPCR analysis , the levels of ISG expression during WT HSV-1 infection of control cells was lower than that observed during ΔICP0 HSV-1 infection , consistent with ICP0 disruption of host innate immune defences [9 , 36 , 57 , 58] . While the induction of ISG transcription was also significantly reduced in IFN-β-treated cells depleted for HIRA relative to control cells ( S8A Fig; dotted line ) , only a minor reduction in ISG protein expression was observed ( S8B and S8C Fig ) . These data corroborate our transcriptomic analysis ( Fig 5H ) and demonstrate that HIRA plays a key role in the regulation of ISG expression in response to virus infection , a host response likely to be saturated under elevated levels of exogenous cytokine stimulation . We conclude that HIRA plays a critical role in the regulation of host innate immune defences that are activated in response to HSV-1 infection under low MOI conditions ( ≤ 1 PFU/cell ) pertinent to a clinical setting . These data identify for the first time a role for HIRA in the induction of innate defences in response to infection with replication-competent herpesvirus . HIRA has recently been reported to act as an intrinsic antiviral regulator to HSV-1 infection through the deposition of histone H3 . 3 onto vDNA in the epigenetic silencing of viral gene expression [17 , 18] . In light of our observations demonstrating a role for HIRA in the induction of innate immune defences to HSV-1 infection ( Figs 5–7 ) , we investigated if HIRA performs independent roles in the intrinsic and innate immune responses to HSV-1 infection , similar to those reported for PML ( Fig 8; [9] ) . We have previously shown that pharmacological inhibition of JAK-STAT signalling by Ruxolitinib impairs the induction of ISG expression during HSV-1 infection [9] . Inactivation of innate immune defences in Ruxolitinib treated cells did not influence the relative plaque-formation efficiency ( PFE ) of WT or ΔICP0 HSV-1 in comparison to equivalently-infected DMSO-treated control cells ( Fig 8A; [9] ) . Thus , inhibition of innate immune defences does not influence the probability of either WT or ΔICP0 HSV-1 to initiate lytic replication and plaque formation by 24 hpi [9] . However , plaque measurements revealed that Ruxolitinib treatment led to an increase in WT and ΔICP0 HSV-1 plaque diameter ( Fig 8B; a median increase of 8 and 40% , respectively ) . Titration of CRV ( cell-released virus ) from Ruxolitinib treated cell monolayers revealed a significant increase in ΔICP0 , but not WT , HSV-1 titres relative to infected DMSO-treated control cells ( Fig 8C ) . Thus , the induction of innate immune defences restricts the rate of ΔICP0 HSV-1 propagation and spread , but not the initiation of plaque formation [9 , 37] . In contrast , PFE assays in HIRA depleted cells demonstrated an 8 to 10-fold increase in ΔICP0 HSV-1 PFE relative to equivalently-infected control cells , with only a modest ( < 2-fold ) inhibitory effect on WT HSV-1 PFE ( Fig 8D and 8E ) . Correspondingly , infection of HIRA depleted cells resulted in a 10-fold increase in VP5 ( major capsid protein and HSV-1 late gene product ) positive cells following ΔICP0 , but not WT , HSV-1 infection at 6 hpi in a JAK-independent manner ( Fig 8F and 8G ) . As the onset of vDNA replication is required to stimulate HSV-1 late gene expression , we analyzed HIRA localization to input genomes ( HSV-1EdC or ΔICP0EdC ) or de novo synthesized vDNA pulse-labelled with EdC following the initiation of HSV-1 DNA replication ( S9 Fig ) . HIRA was readily observed to stably colocalize with ΔICP0 vDNA at 6 hpi under both labelling conditions and to restrict the levels of ΔICP0 HSV-1 gene expression; host responses antagonized by ICP0 that correlate with the degradation of PML and dispersal of PML-NB proteins during WT HSV-1 infection ( S9A–S9E Fig ) . These data corroborate and extend findings reported by Rai et al . [17] , and demonstrate that HIRA contributes to the intrinsic restriction of ΔICP0 HSV-1 independently of the induction of innate immune defences . These data also demonstrate that the stable recruitment of HIRA to infecting viral genomes occurs with temporally distinct kinetics to that observed for PML and PML-NB associated restriction factors ( Fig 1; [9] ) . As expected , viral release assays demonstrated significantly elevated titres of CRV during ΔICP0 , but not WT , HSV-1 infection of HIRA depleted cells relative to equivalently-infected control cells at 24 and 48 hpi ( Fig 8H; [17] ) . Importantly , while Ruxolitinib treatment of infected control cells significantly enhanced ΔICP0 HSV-1 CRV titres relative to DMSO treatment , Ruxolitinib treatment of infected HIRA depleted cells failed to equivalently increase ΔICP0 HSV-1 titres ( Fig 8I ) . These data demonstrate that HIRA plays a key role in the cytokine-mediated restriction of ΔICP0 HSV-1 propagation following the saturation of intrinsic host defences and the onset of productive virus infection . We conclude that HIRA plays independent roles in the regulation of intrinsic and innate immunity to HSV-1 by restricting the initiating cycle of viral gene expression and the induction of host innate immune defences that limit virus propagation , respectively . These combined host defences are antagonized by ICP0 during WT HSV-1 infection through the degradation of PML , which disperses HIRA from infecting viral genomes and reduces HIRA-mediated transcriptional upregulation of ISGs . We conclude that HIRA plays a crucial role in regulating multiple phases of host immunity in response to HSV-1 infection .
The replication-independent deposition of histone H3 . 3 into host chromatin has been predominantly linked to two histone chaperone complexes , namely Daxx/ATRX and HIRA/UBN1/CABIN1 , which play important roles in the regulation of host chromatin structure , transcription , and DNA repair [86 , 87] . The Daxx/ATRX complex typically mediates H3 . 3 deposition into cellular nucleosomes associated with telomeric and pericentric heterochromatin [88–92]; while the HIRA/UBN1/CABIN1 complex has been linked to the deposition of H3 . 3 into more transcriptionally accessible euchromatin and at sites of DNA damage [89 , 93–96] . Recent reports have shown the HIRA H3 . 3 chaperone complex localizes at PML-NBs in response to infection with replication-defective herpesviruses to promote the assembly of viral genomes into chromatin for subsequent epigenetic silencing [17 , 18] . However , the kinetics of HIRA recruitment to infecting viral genomes and PML-NBs under lytic replication conditions had yet to be examined . Here we report that HIRA displays spatiotemporally distinct kinetics of recruitment to infecting HSV-1 genomes and PML-NBs , which independently contribute to the regulation of intrinsic and innate immune defences in response to herpesvirus infection . We recently reported that core PML-NB associated proteins ( PML , Sp100 , Daxx , and ATRX ) rapidly ( < 90 mpi , post-addition of virus ) associate with and entrap infecting herpesvirus genomes as a component of the intrinsic antiviral immune response [9] . With respect to Daxx/ATRX , these findings are consistent with their reported roles as intrinsic host factors that contribute to the entrapment , chromatin assembly , and silencing of viral gene expression at PML-NBs independently of the induction of innate immune defences [9 , 18 , 19 , 29–31] . In contrast to the stable recruitment of PML to infecting viral genomes ( Fig 1A and 1B; [9] ) , we did not observe the stable enrichment of HIRA to either vDNA or PML-NBs at 90 or 180 mpi ( Fig 1A–1C ) . As bio-orthogonal nucleic acid labelling is not compatible with live-cell kinetic studies , we cannot rule out the possibility of highly transient or low antigen abundance HIRA interactions with vDNA or PML-NBs ( Fig 1A ) . Nevertheless , our observations are in contrast with the stable recruitment phenotypes reported for HIRA at PML-NBs in response to infection with replication-defective herpesviruses at 2–6 days post-infection ( [17 , 18]; discussed further below ) or IFN stimulation 8–24 hours post-treatment ( Fig 3D; [17] ) . Under productive infection conditions , HIRA recruitment to infecting viral genomes could only be observed following the saturation of PML-NB host defences and the onset of ΔICP0 HSV-1 DNA replication by 6 hpi ( S9 Fig; [9] ) . These data demonstrate that stable recruitment of HIRA to vDNA occurs with temporally distinct kinetics to that of core PML-NB host factors ( Fig 1; [9] ) , and provides supporting evidence that ΔICP0 HSV-1 genomes can be restricted at multiple phases of infection to influence the progression of viral gene expression ( Fig 8D–8G , S9D and S9E Fig; [9 , 19 , 31] ) . This HIRA-mediated host response is antagonised by ICP0 , in a manner analogous to , but temporally distinct from , that of Daxx/ATRX [9 , 29] . ICP0 disrupts the stable localization of HIRA at vDNA without inducing its proteasome-dependent degradation ( Fig 1E , S9A–S9C Fig ) . While the mechanism ( s ) of this disruption remains to be fully established , the degradation of SUMO-conjugated proteins by the STUbL-like properties of ICP0 is known to inhibit the recruitment of Daxx/ATRX to infecting viral genomes and early-stage vDNA replication complexes [9 , 22 , 23 , 97] . Thus , ICP0 is likely to inhibit the localization of HIRA at vDNA indirectly through the degradation of other substrates that mediate HIRA recruitment or retention at vDNA independently of the degradation of HIRA itself . While this correlates with the ICP0-dependent degradation of PML ( S9 Fig ) , a known interaction partner of HIRA [82] , further investigation is warranted to test the underlying mechanism as many chromatin-associated host factors are asynchronously recruited to vDNA during the initial stages of HSV-1 replication [10 , 98–100] . Our data supports and corroborates conclusions drawn by Rai et al . , which identified HIRA to play an important role in the intrinsic antiviral restriction of ΔICP0 HSV-1 [17] , and emphasizes the importance of multiple histone chaperone complexes ( Daxx/ATRX and HIRA/UBN1/CABIN1 ) in the spatiotemporal regulation of intrinsic immune defences during different phases of herpesvirus infection [29 , 31 , 79] . While HIRA depletion did not significantly influence the levels of WT HSV-1 replication under low MOI conditions ( 0 . 0005 PFU/cell; Fig 8D and 8H ) , we could observe diminished levels of WT HSV-1 gene expression at early time points post-infection ( 2–6 hpi , S9D Fig ) . We therefore cannot discount a transient role for HIRA in the formation of viral chromatin that may promote the onset of viral gene expression [8 , 53] . Nor can we discount a repressive role for HIRA at PML-NBs in other infection contexts , for example during the establishment of viral latency in sensory neurones , where the presence or absence of specific viral or cellular factors may influence the mechanism of viral chromatin assembly and silencing [18] . It is likely , therefore , that the role ( s ) of HIRA in the regulation of viral chromatin assembly that can lead to viral genome silencing will be dependent on the context of infection ( presence or absence of ICP0 ) , permissiveness of cell-type to restrict ΔICP0 HSV-1 replication , and stage of host immune response to infection ( see below ) . Following the productive onset of lytic replication , HIRA stably colocalized with PML-NBs in ICP4 negative cells at the periphery of developing ΔICP0 HSV-1 plaques . HIRA recruitment to PML-NBs under these infection conditions occurred in a JAK-dependent manner and was significantly impaired during WT HSV-1 infection ( Fig 2 ) . These data are consistent with an established role for ICP0 to disrupt PRR activation and the induction of innate immune signalling cascades that lead to IFN-β production and secretion [9 , 12 , 36 , 45 , 101] . We therefore identify that virus-induced cytokine signalling cascades to play an important role in the spatiotemporal localization of HIRA at PML-NBs during infection of normal diploid cells . Replication-defective herpesviruses promote the activation of PRRs that induce innate immune defences and cytokine secretion [33 , 55 , 56 , 59 , 102] , which likely accounts for the stable localization of HIRA at PML-NBs observed under replication-defective infection conditions [17 , 18] . These data are consistent with findings that IFN stimulation alone is sufficient to induce the re-localization of HIRA to PML-NBs in a range of primary and immortalized cell-types ( Fig 3G; [17] ) in a PML- and Sp100-dependent manner ( Fig 4 ) . Further investigation is warranted to determine why this recruitment is impaired in carcinoma cell-types that are responsive to IFN signalling ( Fig 3G , S5 Fig ) , but may reflect changes in PML-NB composition , CDK expression levels , or host chromatin epigenetic status , which are frequently altered upon carcinoma transformation . Collectively , our data demonstrate that HIRA recruitment to PML-NBs in the context of lytic replication in diploid cells represents a sequential step in the paracrine activation of innate immune defences to prime cells for imminent infection . These data have important implications with respect to the interpretation of previous host factor recruitment studies that have utilized ΔICP0 HSV-1 ( or similar ) plaque-edge recruitment assays [21 , 103 , 104] , as these recruitment phenotypes should now be considered in the context of an activated innate immune response . Correspondingly , further experiments are warranted in order to establish if the localization of HIRA at PML-NBs following replication-defective herpesvirus infection is a direct consequence of vDNA entry into the nucleus alone or an indirect consequence of PRR-induced cytokine-mediated autocrine signalling . Under replication-competent infection conditions , our transcriptomic and infection biology analysis clearly demonstrate that HIRA plays important roles in the induction of innate immune defences and ISG expression during HSV-1 infection that correlates with its enrichment of localization at PML-NBs ( Figs 2 , 5 and 7 ) . HIRA depletion abrogated the cytokine-mediated restriction of ΔICP0 HSV-1 ( Fig 8I ) , a viral mutant known to be hypersensitive to IFN [50–52] . Thus , we identify for the first time a requirement for HIRA in the induction of innate immune defences that contribute to the cellular restriction of virus propagation , host defences impaired by ICP0 during the initiating cycle of WT HSV-1 infection ( see above ) . While HIRA has been reported to be enriched across the human genome in response to cytokine stimulation [17] , including a subset of ISGs [17 , 60 , 61] , here we demonstrate that PML promotes the enrichment of HIRA onto ISG bodies ( Fig 6 , S7 Table ) . These data are consistent with a growing body of evidence to support a role for PML and PML-NBs in the regulation of innate immune defences in response to virus infection [9 , 32–34 , 68 , 83] . HIRA has been reported to promote histone H3 . 3 deposition into nucleosomes associated with ISGs to facilitate RNA pol . II transcript elongation [60 , 61] . As Daxx maintains a soluble non-nucleosomal pool of histone H3 . 3 at PML-NBs [17 , 18 , 82 , 88 , 105] , the HIRA chaperone complex may transit through PML-NBs in response to IFN stimulation to load histone H3 . 3 for ISG nucleosome deposition . Thus , PML depletion may impair HIRA enrichment onto ISG bodies by dispersing the concentrated free pool of non-nucleosomal histone H3 . 3 at PML-NBs [82 , 88] . Notably , PML-NBs are also known to localize in close proximity to transcriptionally-active gene-dense chromosome regions and to influence chromatin looping [106–109] , including regions associated with the MHC locus [107] . The localization of HIRA at PML-NBs may therefore concentrate HIRA to regions of host chromatin that are transcriptionally activated in response to IFN-stimulation to enhance histone H3 . 3 deposition to promote their rapid expression , a hypothesis consistent with the HIRA-dependent upregulation of MHC class I genes observed in our study ( Fig 5G; [107] ) . Alternatively , PML may enhance HIRA binding to ISGs by stimulating the onset of transcription directly through the loading of STAT-1/-2 and HDAC-1/-2 onto ISG promoters [33 , 110] . Importantly , none of these hypothesises are mutually exclusive and further studies are required to identify the role ( s ) of PML and PML-NBs in the HIRA-dependent induction of ISG expression during herpesvirus infection . Surprisingly , while both transcriptomic and qPCR studies demonstrate a role for HIRA in the transcriptional upregulation of ISGs in response to exogenous IFN-β stimulation ( Fig 5H , S8A Fig ) , depletion of HIRA only had a minor impact on ISG protein expression ( S8B and S8C Fig ) . These data imply that HIRA mediated induction of ISG expression is saturable under elevated levels of cytokine stimulation ( ≥ 100 IU/ml ) . In the context of virus infection , these data suggest that HIRA is likely to play a critical role in the regulation of innate immune defences under physiological infection conditions that result in comparatively low levels of cytokine secretion , for example in the context of WT herpesvirus infections that express a full complement of immune antagonists ( Figs 5H and 7 ) . This observation is likely to have an important role in vivo , where tissue physiology will further limit cytokine diffusion prior to immune cell migration and the priming of adaptive immune responses [111 , 112] . In summary , we show that the histone H3 . 3 chaperone HIRA plays independent roles in the regulation of intrinsic and innate immune defences to HSV-1 infection that restrict the initiation and propagation of herpesviruses , respectively . We identify these immune defences to occur at spatiotemporally distinct phases of infection , highlighting a critical role for HIRA in the regulation of multiple phases of host immunity in response to pathogen invasion .
Primary human foreskin fibroblast ( HFs ) cells were obtained from Thomas Stamminger ( University of Erlangen; [27] ) . Primary human lung fibroblast ( MRC5 and IMR-90 ) cells were obtained from the ATCC ( CCL-171 and CCL-186 , respectively ) . HFs and MRC5 cells were immortalized by retrovirus transduction to express the catalytic subunit of human telomerase ( HFt and MRC5t , respectively ) , as previously described [25] . Primary and hTERT immortalized human fibroblasts , human retinal pigmented epithelial ( RPE-1; ATCC , CRL-4000 ) , human keratinocyte ( HaCat; AddexBio , T0020001 ) , human osteosarcoma ( U2OS and SAOS; ECACC , 92022711 and 89050205 ) , human cervical carcinoma ( HeLa; a gift from R . Everett , MRC-UoG CVR ) , human lung adenocarcinoma ( A549; a gift from B . Hale , University of Zurich ) , and human embryonic kidney ( HEK 293T; a gift from R . Everett , MRC-UoG CVR ) cells were grown in Dulbecco’s Modified Eagle Medium ( DMEM; Life Technologies , 41966 ) . HFt and MRC5t cells were cultured in the presence of 5 μg/ml of Hygromycin ( Invitrogen , 10687–010 ) to maintain hTERT expression . Transduced HFt and IMR-90 cells expressing shRNAs were cultured in the presence of Puromycin ( Sigma-Aldrich , P8833; 1 μg/ml or 0 . 5 μg/ml for selection or maintenance , respectively ) . Medium for all cell lines was supplemented with 10% foetal bovine serum ( FBS; Life Technologies , 10270 ) , with the exception of IMR-90 cells that were supplemented with 20% FBS , plus 100 units/ml penicillin and 100 μg/ml streptomycin ( Life Technologies , 15140–122 ) . Cell lines were maintained at 37°C in 5% CO2 . 5-Ethynyl-2’-deoxycytidine ( EdC; Sigma-Aldrich , T511307 ) , MG132 ( Calbiochem; 474790 ) , Flavopiridol ( CDK inhibitor; Selleckchem , S1230 ) , CHIR-99021 HCl ( GSK-3α/β inhibitor; Selleckchem , S2924 ) , and Ruxolitinib ( JAK inhibitor; Selleckchem , S1378 ) were prepared in DMSO and used at the indicated concentrations . Interferon beta ( IFN-β; Calbiochem , 407318 ) was prepared in Milli-Q H2O and used at the indicated concentrations . Wild-type ( WT ) HSV-1 strain 17syn+ ( HSV-1 ) , its ICP0-null mutant derivative dl1403 ( ΔICP0; [47] ) , and their respective variants that express eYFP . ICP4 [113] were propagated in RPE cells and titrated in U2OS cells , as described [49] . The purification of EdC labelled HSV-1 virions was performed as essentially described [9] . Briefly , RPE cells were infected with WT or ΔICP0 HSV-1 ( MOI 0 . 001 or 0 . 5 PFU/cell , respectively ) . At 24 h post-infection ( hpi ) , EdC was added at a final concentration of 1 μM every 24 h until extensive cytopathic effect was observed . Supernatants containing labelled cell-released virus ( CRV ) were clarified by centrifugation ( 423 xg for 10 min ) , filtered through a 0 . 45 μm sterile filter , and purified through NAP-25 Sephadex column ( GE Healthcare; 17-0852-01 ) prior to titration in U2OS cells [49] . Plasmids encoding short hairpin ( sh ) RNAs against a non-targeted control sequence ( shCtrl; 5’-TTATCGCGCATATCACGCG-3’ ) , PML ( shPML; 5’-AGATGCAGCTGTATCCAAG-3’ ) , ATRX ( shATRX; 5’- CGACAGAAACTAACCCTGTAA-3’ ) , Daxx ( shDaxx; 5’-GGAGTTGGATCTCTCAGAA-3’ ) , Ubc9 ( shUbc9; 5’- GAAGTTTGCGCCCTCATAA-3’ ) , Sp100 ( shSp100; 5’-GTGAGCCTGTGATCAATAAT-3’ ) , or HIRA ( shHIRA clone F2; 5'-TGAATACCGACTTCGAGAAAT-3' , clone F3; 5'-TCAGGACCGTTAGCCATAATC-3' , clone F4; 5’-TGAATACCGACTTCGAGAAAT-3’ ) were used to generate lentiviral supernatant stocks for transduction of HFt or IMR-90 cells as described [23 , 27–29 , 79] . Pooled stably transduced cells were used for experimentation . The following antibodies were used for immunofluorescence or western blotting: Primary rabbit polyclonal: anti-actin ( Sigma-Aldrich , A5060 ) , anti-Daxx ( Upstate , 07–471 ) , anti-ATRX ( Santa Cruz , H300 ) , anti-PML ( Bethyl Laboratories , A301-167A; Jena Biosciences , ABD-030 ) , anti-Sp100 ( GeneTex , GTX131569 ) , anti-Mx1 ( Santa Cruz , sc-50509; ProteinTech , 13750-1-AP ) , anti-ISG15 ( ProteinTech , 15981-1-AP ) , anti-ISG54 ( IFIT2 , proteinTech , 12604-1-AP ) , and anti-histone H3 ( abcam , ab1791 ) . Primary mouse monoclonal: anti-HIRA ( Millipore , 04–1488 ) , anti-ICP0 ( 11060 , [114] ) , anti-ICP4 ( 58s , [115] ) , anti-VP5 ( DM165 , [116] ) anti-UL42 ( Z1F11; [117] ) , and anti-PML ( abcam , ab96051 ) . Primary antibodies were detected using the following secondary antibodies: DyLight-680 or -800 conjugated goat anti-rabbit or -mouse ( Thermo; 35568 and SA5-35571 ) , Alexa -488 , -555 , or -647 conjugated donkey anti-rabbit or -mouse ( Invitrogen; A21206 , A21202 , A31572 , A31570 , A31573 , A31571 ) , or HRP conjugated goat anti-mouse ( Sigma-Aldrich , A4416 ) . Unless otherwise stated , cells were infected with serial dilutions of HSV-1 or ΔICP0 and rocked every 10 min for 1 h prior to overlay with medium supplemented with 2% Human Serum ( HS; MP Biomedicals , 2931149 ) . 24 to 36 hpi , cells were washed twice in PBS ( Sigma-Aldrich , D1408 ) , simultaneously fixed and permeabilized in 1 . 8% formaldehyde ( Sigma-Aldrich , F8775 ) and 0 . 5% NP40 ( BDH , 56009 ) in PBS for 10 min , then washed twice in 0 . 1% Tween in PBS ( PBST ) . Cells were blocked with 5% skimmed milk powder ( SMP; Marvel ) in PBST ( blocking buffer ) for 30 min before incubation with an anti-VP5 monoclonal antibody diluted in blocking buffer for 90 min . Cells were washed three times with PBST , incubated with HRP conjugated anti-mouse IgG diluted in blocking buffer for 60 min , then washed with PBST three times . Plaques were visualized with True Blue peroxidase developing solution ( Insight , 50-78-02 ) according to the manufacturer’s instructions , and washed with Milli-Q H2O prior to plaque counting or imaging using an Axio Observer Z . 1 microscope ( Zeiss ) with differential interference contrast . For plaque formation efficiency ( PFE ) assays , plaque counts are expressed relative to the number of plaques on infected control cell monolayers at the equivalent dilution of input virus . Results are presented as relative fold change ( number of plaques sample/number of plaques control ) . Plaque diameters were measured using Zen blue ( Zeiss ) imaging software . Cells were pretreated with 5 μM Ruxolitinib or DMSO as a carrier control for 1 h prior to infection with HSV-1 ( MOI 0 . 001 or 0 . 0005 PFU/cell , as indicated ) , or ΔICP0 ( MOI 1 or 0 . 5 PFU/cell , as indicated ) . Following absorption , cell monolayers were washed twice with DMEM prior to overlay with medium containing 5 μM Ruxolitinib or DMSO . Supernatants containing CRV were collected at the indicated times post-infection . Virus titres were calculated by titration on U2OS cells , as described [49] . 3x106 cells were treated with IFN-β ( 100 IU/ml ) for 24 h prior to harvesting by trypsinization and sedimentation by low speed centrifugation ( 1500 rpm for 5 minutes at RT ) . Cell pellets were resuspended in 2 ml of ice-cold IP buffer ( 50 mM HEPES pH 7 . 4 , 1% NP40 , 150 mM NaCl , 10% glycerol , 1mM EDTA , 2 mM Dithiothreitol ) containing a cocktail of protease inhibitors ( Roche , 11873 580 001 ) . Cell suspensions were cup horn sonicated at 4°C ( Branson digital sonifier 450; 2x 30 second pulses at 20% amplitude ) prior to clarification by ultra-centrifugation at 25 , 000 rpm for 20 mins ( Beckman OptimaMax-XP ultra ) . Cell lysates were pre-cleared using 30 μl of equilibrated Protein G beads ( Millipore , 16–201 ) at 4°C for 30 mins with rotation . Beads were collected by centrifugation ( 1 min at 13 , 000 rpm ) and the clarified lysate transferred to a fresh tube . 1 μg of purified rabbit IgG ( Sigma-Aldrich , I5006 ) or rabbit anti-Sp100 ( GeneTex , GTX131569 ) was incubated with 1 ml of clarified lysate at 4°C for 2 h with rotation . Immune complexes were captured by adding 30 μl of equilibrated Protein G beads at 4°C for 90 mins . Beads were collected by centrifugation ( 1 min at 13 , 000 rpm ) and washed 3x in 500 μl of IP buffer prior to final resuspension in 1x SDS-PAGE loading buffer ( see below ) . Samples were boiled for 10 mins prior to SDS-PAGE and western blot analysis . Treated or infected cells were washed twice with PBS . Whole cell lysates were collected in 1x SDS-PAGE loading buffer containing 2 . 5 M Urea ( Sigma-Aldrich , U0631 ) and 150 mM Dithiothreitol ( DTT; Sigma-Aldrich , D0632 ) . Proteins were resolved on NuPAGE 4–12% Bis-Tris Protein gels ( Invitrogen , NP0322BOX ) in MES ( Invitrogen; NP0002 ) or MOPS buffer ( Invitrogen , NP0001 ) and transferred onto 0 . 2 μm nitrocellulose membrane ( Amersham , 15249794 ) for 90 min at 30 volts in Novex transfer buffer ( Invitrogen , NP0006-1 ) according to the manufacturer’s instructions . Membranes were blocked in PBS with 5% FBS ( Block ) for a minimum of 1 h at room temperature . Membranes were incubated in primary antibody diluted in Block for a minimum of 1 h , washed three times with PBST for 5 min each , then incubated in secondary antibody diluted in Block for 1 h . Following three 5 min washes in PBST , one 5 min wash in PBS , and one rinse in Milli-Q H2O , membranes were imaged on an Odyssey Infrared Imager ( LiCor ) . The intensity of protein bands was quantified with Odyssey Image Studio Software . Cells were seeded overnight on to 13 mm coverslips prior to treatment or infection at the indicated MOI and times at 37°C . For click chemistry assays , cells were washed in serum free DMEM prior to overlay in complete medium or fixation . At indicated times , cells were washed twice in CSK buffer ( 10 mM HEPES , 100 mM NaCl , 300 mM Sucrose , 3 mM MgCl2 , 5 mM EGTA ) , simultaneously fixed and permeabilized in 1 . 8% formaldehyde and 0 . 5% Triton-X100 ( Sigma-Aldrich , T-9284 ) in CSK buffer for 10 min , and washed twice in CSK . Coverslips were then blocked with 2% HS in PBS for 30 min prior to click chemistry followed by immunostaining . Where applicable , EdC-labelled vDNA was detected using the Click-iT Plus EdU Alexa Fluor 555 Imaging Kit ( ThermoFisher scientific , C10638 ) according to the manufacturer’s instructions . For host and viral protein labelling , cells were incubated with primary antibodies diluted in 2% HS in PBS for 60 min , then washed in 2% FBS in PBS three times , before incubation with secondary antibodies and DAPI ( Sigma-Aldrich , D9542 ) in 2% HS in PBS for 60 min . Coverslips were then washed in 2% FBS in PBS three times , and twice in Milli-Q H2O prior to mounting on Citiflour AF1 ( Agar Scientific , R1320 ) . Coverslips were examined using a Zeiss LSM 880 confocal microscope using the 63x Plan-Apochromat oil immersion lens ( numerical aperture 1 . 4 ) using 405 nm , 488 nm , 543 nm , and 633 nm laser lines . Zen black software ( Zeiss ) was used for image capture , generating cut mask channels , and calculating weighted colocalization coefficients . Exported images were processed with minimal adjustment using Adobe Photoshop and assembled for presentation using Adobe Illustrator . HFt cells were infected with WT ( MOI 0 . 001 PFU/cell ) or ΔICP0 ( MOI 2 PFU/cell ) HSV-1 expressing EYFP . ICP4 to enable the initiation of viral replication and plaque formation to occur , as previously described [49] . At 24 hpi , infected cell monolayers were fixed and immunostained , as described above . HFt cells were seeded into 48-well plates at a density of 6x104 cells/well . 24 h post-seeding , the cells were washed 2x in DMEM medium containing low serum ( 1% FBS ) prior to incubation in low serum medium for 24 h . Cells were released from cell cycle arrest by the addition of DMEM medium ( 10% FBS ) containing 1 μM EdC and either DMSO ( carrier control ) or GSK-3α/β inhibitor ( CHIR-99021 HCl ) at the indicated concentrations . Cells were fixed and permeabilized at the indicated time points prior to click chemistry ( as described above ) . The number of EdC positive nuclei ( proliferating cells containing de novo replicated DNA ) were quantified using an automated Celigo imaging cytometer ( Nexcelom biosciences ) , as per the manufacture’s instructions . Cells were mock , IFN-β stimulated ( 100 IU/ml ) , or infected with HSV-1 or ΔICP0 at the indicated MOI . Total RNA was isolated at the indicated times post-treatment using the RNAeasy Plus Kit ( Qiagen , 74134 ) , according to the manufacturer’s instructions . Reverse transcription ( RT ) was performed using the TaqMan Reverse Transcription Reagents kit ( Life Technologies , N8080234 ) with oligo ( dT ) primers . cDNA samples were analyzed in triplicate using TaqMan Fast Universal PCR Master Mix ( Life Technologies , 4352042 ) with the following TaqMan gene specific primer- ( FAM/MGB ) probe mixes ( Life Technologies ) : assay ID PML ( Hs00231241_m1 ) , Ubc9 ( Hs00163336_m1 ) , Daxx ( Hs00985566_g1 ) , ATRX ( Hs00997529_m1 ) , Sp100 ( Hs00162109_m1 ) , HIRA ( Hs00231498_m1 ) , Mx1 ( HS00895608_m1 ) , ISG15 ( Hs01921425_s1 ) , ISG54 ( Hs01922738_s1 ) , OAS1 ( Hs00973635_m1 ) , or GAPDH ( 4333764F ) on a 7500 Fast Real time PCR system ( Applied Biosystems ) . Relative mRNA levels were determined using the ΔΔCt method ( normalized to GAPDH ) and expressed relative to indicated treatments . Mean ( RQ ) and standard deviations ( RQmin/max ) are presented . 2x105 cells/well were seeded into 12-well dishes and incubated at 37°C for 24 h prior to experimentation . Cell monolayers were mock treated , stimulated with IFN-β ( 100 IU/ml ) or infected with WT or ΔICP0 HSV-1 ( MOI 1 PFU/cell ) for 17 h . Total RNA from three technical replicate experiments was isolated using an RNAeasy Plus kit ( Qiagen , 74134; as described above ) . RNA concentration and integrity was determined using Qubit Fluorimeter ( Life Technologies , Q32855 and Q32854 ) and Agilent 4200 TapeStation ( Agilent , 5067–5579 and 5067–5584 ) reagents and instruments , respectively . All samples had a RIN score of ≥ 8 . 5 . 500 ng of total RNA was used to prepare libraries for sequencing using an Illumina TruSeq Stranded mRNA HT kit ( Illumina , 20020594 ) and SuperScript2 Reverse Transcriptase ( Invitrogen , 18064014 ) according to the manufacturer's instructions . Libraries were pooled in equimolar concentrations and sequenced using an Illumina NextSeq 500 sequencer ( Illumina , FC-404-2005 ) . At least 95% of the reads generated presented a Q score of ≥ 30 . Reads were quality assessed using FastQC software ( https://www . bioinformatics . babraham . ac . uk/projects/fastqc/ ) , sequence adaptors removed using TrimGalore ( https://www . bioinformatics . babraham . ac . uk/projects/trim_galore/ ) , and sequences aligned to the human genome ( GRCh38 . 86 ) via Ensembl ( Ensembl genome browser 86 ) using HISAT2 [118] . FeatureCount [119] was used to count reads mapped to individual annotated gene files . EdgeR was used to calculate relative gene expression levels [120] . False discovery rate ( FDR ) values were calculated using the Benjamini–Hochberg method in EdgeR for each pairwise comparison ( Q value ) . Sequences have been deposited in EMBL-EBI ( https://www . ebi . ac . uk/ena ) under accession number PRJEB27501 . Pathway analysis was conducted using Reactome ( https://reactome . org ) [121 , 122] . High-confidence Ensembl gene IDs ( Q < 0 . 0001 , ≥ log2 fold change ) were converted to uniprot IDs using the Reactome mapping tool and used to identify pathways enriched ( over-represented ) for mapped entities ( S1–S5 Tables ) . Reactome FDR values < 0 . 05 were considered significant for pathway enrichment , ranked , and plotted as heat maps in GraphPad Prism . Native ChIP was performed for HIRA as described [69] using an equimolar mixture of mouse mAbs to HIRA ( WC15 , WC19 , WC117 , WC 119; [74] , a gift from Professor Peter Adams ( Beatson Institute for Cancer Research , University of Glasgow ) or mouse mAb to HA tag ( Covance , MMS-101R ) as a species and class-matched negative control . Analysis of the ChIP-seq data was performed by assessing read quality using FASTQC ( https://www . bioinformatics . babraham . ac . uk/projects/fastqc/ ) and trimming reads using Trim-Galore ( https://www . bioinformatics . babraham . ac . uk/projects/trim_galore/ ) . Single-end ChIP-seq reads were aligned to the human genome ( hg19 ) using Bowtie2 [123] . Additional quality and filtering steps were undertaken; demarking duplicate reads with Picard Tools ( http://picard . sourceforge . net ) as well as by retaining reads that map uniquely to single loci ( Alignment Statistics: S7 Table ) . Peak calling was performed using USeq ( v8 . 6 . 0 ) [124] , with window and extension sizes of 200 bp and 150 bp , respectively , utilizing non-specific input DNA as a control . Venn diagrams were drawn using the package Eulerr [125] and the significance and fold change of each overlap was determined with a bespoke script ( https://github . com/neilrobertson/BICRCode ) using a permutation based approach over 1000 iterations . HIRA enrichment on ISG bodies was compared against a random , equally sized subset of genes from the Ensembl gene set with status ‘known’ and biotype ‘coding’ ( 22 , 008 total genes; version 73 of the archive ) . The HIRA signal for any given window was calculated as the total number fractional reads within a window , with the product divided by the total number of reads in the dataset divided by one million . Composite profiles were generated by dividing gene length into 40 bins , each corresponding to 2 . 5% of the total gene length , with ten additional 500 bp windows extending from the transcriptional start and end sites of each gene ( ± 5 kb ) . The average normalised ChIP-seq signal was then calculated for each window and normalised to input DNA . Sequences have been deposited in GEO ( https://www . ncbi . nlm . nih . gov/geo/ ) under accession number GSE128173 . | Host innate immune defences play critical roles in the cellular restriction of invading viral pathogens and the stimulation of adaptive immune responses . A key component in the regulation of this arm of host immunity is the rapid induction of cytokine signalling and the expression of interferon stimulated gene products ( ISGs ) , which confer a refractory antiviral state to limit virus propagation and pathogenesis . While the signal transduction cascades that activate innate immune defences are well established , little is known about the cellular host factors that expedite the expression of this broad repertoire of antiviral host genes in response to pathogen invasion . Here we show that HIRA , a histone H3 . 3 chaperone , associates with PML-NBs to stimulate the induction of innate immune defences in response to HSV-1 infection . Our study highlights the importance of histone chaperones in the coordinated regulation of multiple phases of host immunity in response to pathogen invasion and identifies a key role for HIRA in the induction of innate immunity to virus infection . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [] | 2019 | The histone chaperone HIRA promotes the induction of host innate immune defences in response to HSV-1 infection |
Myriad infectious and noninfectious causes of encephalomyelitis ( EM ) have similar clinical manifestations , presenting serious challenges to diagnosis and treatment . Metabolomics of cerebrospinal fluid ( CSF ) was explored as a method of differentiating among neurological diseases causing EM using a single CSF sample . 1H NMR metabolomics was applied to CSF samples from 27 patients with a laboratory-confirmed disease , including Lyme disease or West Nile Virus meningoencephalitis , multiple sclerosis , rabies , or Histoplasma meningitis , and 25 controls . Cluster analyses distinguished samples by infection status and moderately by pathogen , with shared and differentiating metabolite patterns observed among diseases . CART analysis predicted infection status with 100% sensitivity and 93% specificity . These preliminary results suggest the potential utility of CSF metabolomics as a rapid screening test to enhance diagnostic accuracies and improve patient outcomes .
Encephalomyelitis ( EM ) is a condition characterized by inflammation of the brain ( encephalitis ) and spinal cord ( myelitis ) that frequently causes permanent disability . There are myriad causes of EM syndromes , which are in aggregate relatively common [1–4] and include viral , bacterial , fungal , protozoal and prion infections , autoimmune encephalitis , intoxications , and metabolic encephalopathies , while other EM cases have unknown causes [5] . Clinicians face significant challenges to the rapid and accurate diagnosis and treatment of EM . Due to the rarity of a definitive diagnosis , many arbovirus and other viral causes of EM , including rabies , have limited evidence-based therapies; this may change with newer broad-spectrum antivirals currently in clinical trials [6] . Treatment of autoimmune EM relies on corticosteroids , immunoglobulin , plasmapheresis , cytotoxic agents and biologicals [7 , 8] , which are typically contra-indicated until infections can be excluded . Physicians are often forced to treat empirically for infections and delay appropriate therapy for autoimmune EM , thereby worsening patient outcomes . Moreover , for many causes of EM , no rapid diagnostic testing exists , and long delays pending laboratory test results commonly occur before definitive treatment may be initiated; however , superior outcomes depend on early intervention . Because there are numerous causes of EM , including multiple infectious agents that overlap or coincide in geographic distribution , diagnosis reliant on single-target testing is unsatisfactory as it requires quantities of tests that are not only prohibitive in cost but also involve collecting unsafe volumes of blood or cerebrospinal fluid ( CSF ) from patients . Improved diagnostics and proxy markers of therapeutic efficacy are sorely needed , especially as new treatment regimens develop . In recent years , the development and expansion of omics technologies have presented opportunities for discovering disease mechanisms and biomarkers of clinical significance [9–11] . Metabolomics , the comprehensive study of small-molecule metabolites in a biofluid or tissue , offers a set of clues to the biochemical workings of a body system , organ , or compartment in a given physiological state , and has diverse applications in improving clinical diagnosis and treatment of central nervous system ( CNS ) diseases and intoxications [9 , 12–17] . Metabolomics panels may also provide information about a broad spectrum of metabolic processes involved in a disease presentation compared to traditional single-molecule assays . Metabolites present in CSF may originate from brain metabolic processes , including intermediate and end products of energy metabolism , neurotransmission , inflammation and oxidative stress responses; thus , their analysis provides insights into metabolic disturbances occurring in CNS diseases . Among the methodological approaches taken in metabolomics studies of CSF , 1H-NMR spectroscopy carries advantages for exploratory studies both in the scope of metabolite detection and its quantitative ability [18] . An additional advantage of this method is the lack of sample consumption , given practical limitations on the volume of CSF usually available . Further , many CNS diseases and intoxications are prevalent in countries where advanced imaging facilities , reference laboratories and therapeutics are in short supply . Recent studies have applied 1H NMR-based metabolomics of CSF to identify single-molecule biomarkers and panels of metabolites associated with a range of neurological diseases such as infectious meningitis [14] , multiple sclerosis ( MS ) [13 , 19 , 20] , Alzheimer’s [21 , 22] Parkinson’s [23] and Huntington’s diseases [24] . Further , this method has detected metabolic changes characterizing different stages of disease progression in rabies and MS [12 , 25] . Proxy markers of disease progression or response to therapy may also accelerate therapeutic trials while lowering their cost . Despite significant advances in the application of NMR metabolomics in the investigation of certain CNS diseases , such as multiple sclerosis , its potential to describe metabolic changes occurring in many infectious neurological diseases has been less studied . Lyme disease and West Nile Virus ( WNV ) are vastly under-studied in this sense , despite being the most common causes of vector-borne bacterial and viral disease , respectively , in the United States [26 , 27] . Rabies is an important global zoonosis but may be underdiagnosed in some contexts due to challenges in distinguishing it clinically from other CNS infections , such as cerebral malaria , in areas where these are endemic [28] . Infectious diseases that invade the CNS have distinct molecular mechanisms driving their respective pathologies [29 , 30] . Further , pathogen strategies to replicate while evading host immune responses can involve the disruption of a range of endogenous metabolic processes [31] , many of which have yet to be illuminated for specific diseases; thus , explorative studies of the CSF metabolome in different disease states can provide an important window for examining potential pathogen effects on metabolism within the CNS to lay the groundwork for future targeted diagnostics or therapeutic interventions . In the present study CSF samples from patients representing diverse infectious and non-infectious diseases of the CNS were analyzed by 1H NMR-spectroscopy to determine if metabolomics profiling could distinguish diseases . We find preliminary evidence of the existence of discriminating metabolic features .
Twenty-seven patients were diagnosed with CNS Lyme disease ( n = 5 , all ages , at the New York State Department of Health ) , WNV meningoencephalitis ( n = 5 , all ages , New York State Department of Health ) , Clinically Isolated Syndrome ( CIS ) of multiple sclerosis ( MS , n = 4 , adults , Intermountain Healthcare ) , rabies ( n = 10 , all ages , at Canadian Food Inspection Agency , Centers for Disease Control and Prevention , Kimron Veterinary Institute , National Institutes of Health-Colombia , and New York State Department of Health ) , or Histoplasma meningitis ( n = 3 , anonymous , at Indiana University School of Medicine ) . Due to ethical concerns surrounding the collection of CSF from healthy individuals , healthy controls were not available for this study . Specimens obtained as discard material from 25 anonymous children aged 5–20 years at the Children’s Hospital of Wisconsin with no concurrent microbiological testing and no known encephalopathy or encephalitis served as a control group . This population includes mostly children with cancer in remission or children being treated for pseudotumor cerebri , a common non-inflammatory condition . Given patient samples were anonymous discard material , the study was ruled to not be human research requiring informed consent by the Children’s Hospital of Wisconsin IRB ( protocol CHW 10/24 ) . For rabies patients , for whom multiple specimens were available , the specimen taken closest to the fourth day of hospital admission was selected to minimize the influence of hypoglycemia , ketosis or renal insufficiency on presentation to the CSF metabolome . While the CSF was collected for diagnostic purposes , precise timing is uncertain other than for rabies patients . Initially , four specimens from patients with Histoplasma meningitis were analyzed , but one specimen had a metabolite profile inconsistent with CSF and was excluded on the basis of containing implausible values . Three Histoplasma specimens remained after this exclusion . After collection , specimens were stored refrigerated and/or frozen until transport on dry ice to the site of analysis , where they were stored at -80°C until sample preparation . Once defrosted , samples were filtered using washed Amicon Ultra-0 . 5 mL centrifugal filters with a cut-off of 3000 MW ( Millipore , Billerica , MA ) to remove lipids and proteins . When needed , filtrate volume was adjusted to 207 μL when preparing for 3mm NMR tubes or 585 μL when preparing for 5mm NMR tubes with Type I ultrapure water from Millipore Synergy UV system ( Millipore , Billerica , MI ) . Samples were prepared for analysis by the addition of 23 μL or 65 μL of internal standard containing approximately 5 mmol/L of DSS-d6 [3- ( trimethylsilyl ) -1-propanesulfonic acid-d6] , 0 . 2% NaN3 , in 99 . 8% D2O to 207 μL or 585 μL of CSF filtrate , respectively . The pH of each sample was adjusted to 6 . 8 ± 0 . 1 by adding small amounts of NaOH or HCl . A 180 or 600 μL aliquot was subsequently transferred to 3 mm or 5mm Bruker NMR tubes , respectively , and stored at 4 oC until NMR acquisition ( within 24 hours of sample preparation ) . NMR spectra were acquired as previously described [12] on a Bruker Avance 600-MHz NMR equipped with a SampleJet autosampler using a NOESY-presaturation pulse sequence ( noesypr ) at 25°C . NMR spectra were manually phased and baseline-corrected using NMR Suite v6 . 1 Processor ( Chenomx Inc . , Edmonton , Canada ) , and Chenomx NMR Suite v . 8 . 1 Profiler ( Chenomx Inc . , Edmonton , Canada ) was used for quantification of metabolites . Selected NMR spectral data from a previous rabies study in this lab [12] were compared to additional samples acquired from Lyme , WNV , histoplasmosis and MS patients . After correcting metabolite concentrations for dilution , data were cluster-analyzed 2 ways for comparison using RStudio software ( RStudio Version 1 . 0 . 136 , Boston , MA , USA ) or Stata software ( SE 14 , College Station , TX , USA ) . First taking a data-driven approach , concentrations were log10-transformed before principal component analysis ( PCA ) was carried out on the covariance matrix of the centered data as an unsupervised search for trends . Alternatively , to provide clinical context , data were normalized to z-scores using published reference ranges in CSF ( www . hmdb . ca ) . In instances when published norms were discrepant , those that encompassed the range of our control population were selected . In rare instances when normal ranges were unavailable , means and standard deviations were constructed using our 25 controls . Normalization by z-scores constructed from population norms generated more skewed data than log10-transformation across the entire spectrum of diseases and controls . Factor analysis better tolerates skewed data than PCA and was applied to the z-scores . Based on the separation found by PCA and factor analysis , differences in metabolite concentrations by infection status and by individual disease diagnoses were assessed on the untransformed data using Mann-Whitney U tests and Kruskal-Wallis tests , respectively . P-values were adjusted for multiple comparisons using false discovery rates . Homogeneity of variance between groups was tested using the Levene test to inform interpretation of the rank sum test results . For metabolites with significant differences by Kruskal-Wallis testing , Dunn’s multiple comparisons tests were performed between each pair of groups to determine which diseases were different from each other . For these tests , p-values were Bonferroni-adjusted within the 15 multiple comparisons carried out for each metabolite . After adjustment , p-values of less than 0 . 05 were considered significant . Cliff’s Delta statistics [32] were calculated to assess the degree of overlap in metabolite concentrations by infection status and between diseases that were found to have significant differences by the Dunn’s test . Untransformed data were also analyzed by predictive analysis [33 , 34] . Classification and regression trees ( CART ) and Random Forests were performed using Salford Predictive Modeler software suite CART and suite Random Forests ( Salford Systems , San Diego , CA , USA ) . For CART , parent node and terminal node were 10 and 5 , respectively . 10% leave-out samples were used for cross-validation . Random Forests are collections of decision trees , and each tree was grown on a random ( ~2/3 ) subsample of the data . The remaining data were used to determine the performance of the trees . The number of trees to build was 1000 . The number of predictors considered for each node was the square root of the number of potential predictors , and the parent node minimum cases was 2 . The variable importance was assessed using the GINI method . Target variable and predictors were the same as for CART .
A major clinical challenge is determining whether infection exists as a contraindication to immunosuppression . Unsupervised PCA was performed on metabolite data from patients diagnosed with a neurological disease and controls . Six compounds ( acetaminophen , ethanol , ethylene glycol , glycerol , propylene glycol , and valproate ) of likely exogenous origin were excluded from cluster analysis models . The first two principal components ( PC ) in this model accounted for 37 . 8 percent of the variation in metabolite concentrations . Prominent overlap was apparent between controls and MS , which separated distinctly from infectious diseases along PC 1 ( Fig 1 ) . In a scores plot of the first two components , PC 2 identified an apparent outlier in the WNV group , which upon closer examination was observed to have extremely low levels of citrate , lactate , and amino acids coupled with markedly high glutamate , pyruvate , acetate and 2-oxoglutarate compared to the rest of the samples . Since the general patterns generated by PCA did not change when this individual was removed from the dataset , the results shown in Fig 1 reflect this exclusion in order to better visualize clusters in the data . When overlaid with loadings vectors , the scores plot of the first two PCs revealed two patterns of metabolites among infectious diseases , one characterized by higher levels of ketone bodies and the other by higher levels of pyruvate , glutamate , 2-oxoglutarate , carnitine , and glycine ( Fig 1 ) . Pearson correlation coefficients reflect moderate to high correlation among the metabolites in each pattern , with correlation coefficients ranging from 0 . 77 to 0 . 93 among ketone bodies and from 0 . 23 to 0 . 58 among metabolites in the second pattern . In contrast , metabolites including acetate , isobutyrate , myo-inositol , threonine , and glutamine appeared to characterize controls and MS using loadings vectors . The contribution of ketone bodies to the PCA analysis prompted a second , clinically applicable analysis using z-scores of normal human values for each metabolite while excluding the potentially non-specific markers of dehydration and starvation , which yielded similar results . Unsupervised factor analysis discriminated CNS disease from controls , with 2 factors accounting for 35 . 6 percent of the variation . The WNV sample that appeared as an outlier by PCA was not influential in this analysis . Factor analysis excluding ketones and creatinine did not discriminate infections from normal as well as did the PCA analysis . Given the graphical separation by infection status shown by PCA and factor analysis , Mann-Whitney U tests were performed to test for differences in metabolite concentrations between patients with an infectious CNS disease and those with no CNS infection ( MS and controls ) . All metabolites were included . These results are summarized in Table 2 . After correcting for multiple comparisons , significant univariate differences were detected in the concentrations of 29 compounds; these included several metabolites that appeared to drive separation in the PCA ( ketones , pyruvate , carnitine , and glycine ) . Median concentrations of glutamate and 2-oxoglutarate were significantly higher in infectious diseases than patients with no infectious disease , and there was a trend towards higher citrate concentrations in the infectious disease group ( p = 0 . 07 ) . Also , in agreement with the PCA results , median concentrations of isobutyrate , fructose , N-acetylneuraminate , and serine were higher in the noninfectious disease group , and acetate exhibited different distributions between the groups . In a similar univariate analysis on z-scores for 43 variables , nine metabolites were identified ( Table 2 , among bolded metabolites ) , all of which were also identified using the previous method . While CNS infections overlap as a syndrome , they are caused by viruses , bacteria , fungi , protozoa and prions that require different therapies . We therefore evaluated PCA discrimination within CNS diseases without the influence of controls . In the resulting model , PC1 and PC2 cumulatively accounted for 38 . 9 percent of the variation , and when loadings vectors were overlaid with PC scores , the resulting Gabriel’s biplot revealed the most important metabolites to be ketone bodies , glutamine , glutamate , and threonine . In a scores plot of the first two PCs , moderate separation by disease diagnosis pointed to differential as well as overlapping metabolic patterns among diseases ( Fig 2 ) , which were further dissected in additional analyses and are summarized in Tables 3 and 4 . After removing ketones and creatinine , factor analysis of z-scores did not separate cleanly between disease groups . After correcting for multiple comparisons , Kruskal-Wallis tests on untransformed data detected significant differences among diseases and controls in the concentrations of 31 metabolites . Metabolites and diseases for which concentrations were significantly different from control samples according to Dunn’s multiple comparisons tests are shown in Table 4 . In particular , the CSF of WNV patients had markedly higher concentrations of pyruvate ( p = 0 . 0008 ) and formate ( p = 0 . 0005 ) , and Lyme disease and WNV patients shared higher levels of formate and glycine compared to controls . Rabies patients had significantly different concentrations of energy-related metabolites including ketone bodies , lactate and 2-hydroxybutyrate , some of which were also elevated in WNV but not in histoplasmosis or Lyme disease . CART analysis differentiated infection status with 100% sensitivity and 93% specificity ( Table 5 ) . High pyroglutamate alone discriminated WNV , Lyme and histoplasmosis from controls . MS or rabies could be identified from controls with 100% sensitivity and 76% specificity by high 2-hydroxybutyrate or low 2-hydroxybutyrate and high carnitine . Random Forest analyses confirmed the importance of the majority of metabolites identified by CART .
NMR metabolomics distinguished infectious and inflammatory disorders using laboratory-confirmed samples of 5 disorders using 2 approaches to normalization of the data , and 2 unsupervised cluster analytical approaches . CART decision analysis easily differentiated bacterial ( Lyme ) , fungal ( Histoplasma ) and viral ( WNV ) causes of encephalomyelitis from controls . Decision analysis also differentiated rabies and the prodromal form of MS from controls , while separation by cluster analyses was incomplete between MS and controls . Notably , the greatest source of variation in metabolomics data found by PCA was the presence or absence of an infectious pathogen . If replicated , this finding is of paramount clinical impact because treatments for infections require almost polar opposite therapeutics than those for autoimmune diseases . There was also substantial agreement in the identification of influential metabolites between different approaches to data normalization and reduction and predictive approaches , including CART and random forest analysis . Metabolites driving separation in PCA ( pyruvate , glutamate , quinolinate , 2-oxoglutarate , carnitine , and glycine ) potentially suggest alterations in energy metabolism , excitotoxicity and antioxidant response . Patterns of these metabolites were not uniform . Rather , overlapping as well as distinguishing metabolic features were seen , highlighting the potential utility of measuring a suite of metabolites rather than searching for individual metabolic biomarkers for diseases , which may not exist . Overlap of profiles makes strong clinical sense given that EM syndromes overlap in signs and symptoms . The overlap also supports a clinical rationale for syndromic metabolic therapies across a range of infectious or autoimmune causes of EM . Distinguishing features provide promise of rapid , relatively specific diagnoses that enable prompt pathogen or process-directed therapies . Significant differences by disease group were found in the CSF concentrations of several metabolites known to be involved in the synthesis of the antioxidant glutathione ( GSH ) and related pathways , including glycine , formate , pyroglutamate , and 2-hydroxybutyrate . The transsulfuration pathway links the methylation cycle of one carbon metabolism to GSH synthesis and produces 2-hydroxybutyrate as a secondary byproduct during the conversion of cystathionine to cysteine [35 , 36] . Formate , an endogenous and bacterial metabolite that along with glycine was found at significantly higher levels in WNV and Lyme disease patients compared to controls in this study , is formed as a byproduct in several pathways including the tryptophan kynurenine pathway [37] , pterin metabolism [38] and protein demethylation ( following hypermethylation by S-adenosyl-L-methionine [39] ) , while it is also consumed in the folate cycle during the conversion of tetrahydrofolate ( THF ) to 10-formyl-THF [40] . An end product of purine catabolism , neopterin , has been found to be elevated in patients with rabies [41] , Lyme disease , and other neuroinfections , while remaining low in MS and other neuroinflammatory conditions [42] . Pyroglutamate , which converts to glutamate before being incorporated into GSH and also activates amino acid transport systems at the blood brain barrier [43] , was higher in histoplasmosis , Lyme disease and WNV and was an important predictor distinguishing these conditions from control samples . Given individual metabolites can participate in a number of biochemical pathways , further studies are required to parse out the mechanisms at play in the diseases studied here . A likely interpretation is that infection or inflammation in the CNS is associated with redox imbalances including glutathione metabolism and NADH/NAD+ ratios . It is of particular interest that these metabolites may profile mechanisms leading to insulin resistance and vascular disease [36] , given that low dose insulin therapy was added to the Milwaukee protocol , version 4 , with statistical improvements in survival [44] . Our analytical design sought to minimize the effects of starvation/ketosis and dehydration/uremia on the metabolic profile of rabies by prioritizing rabies samples taken four days after admission . Nevertheless , PCA analysis identified the importance of ketone bodies in identifying rabies . Factor analysis that deliberately excluded primary ketones , urea and creatinine from analysis still identified isopropanol and methanol ( Table 3 ) , both downstream metabolites of ketones , as discriminators of rabies . RF and CART analyses also identified ketones and carnitine ( fatty acid oxidation ) as predictors of rabies but not other infections ( Table 5 ) . Despite our experimental design , CNS ketosis may be a valid indicator of rabies encephalitis . This study was originally intended to further explore the specificity of NMR metabolomics for the diagnosis of rabies , which is often confused with Guillain-Barre syndrome , acute psychosis and N-methyl-D-aspartate receptor ( NMDAR ) encephalitis and currently requires multiple tests for diagnosis at remote reference laboratories . Our findings suggest that the utility of the approach may instead lie in excluding competing diagnoses , many of which are more treatable . NMR metabolomics performed on a par with current rabies diagnostics ( 100% sensitivity , 76% specificity ) and is likely complementary ( particularly after 5 days ) . When restricted to the first week of hospitalization with rabies ( when most patients die ) , NMR metabolomics did not perform as well as for other infections; gene expression studies of rabies CSF and detection of rabies-specific antibodies also performed poorly in the first week . Rabies can clearly be delineated from controls by NMR at later time points , and NMR of CSF also measures recovery [12] . The promise of an NMR metabolomics profile as a proxy marker for therapeutic response would be welcome for rabies , WNV , NMDAR encephalitis or acute disseminated encephalomyelitis for which efficacious treatments remain undefined . This study is exploratory and is limited by the number of samples available for CNS diseases of rare incidence . The possibility of confounding effects of age , sex , disease stage , or other acute variations in metabolic processes should be considered in interpreting these results . Our control group was aged 5–20 years , while ages in the disease group ranged from 4 to 83 years . However , we confirmed that the distribution of metabolites of our controls overlapped with adult norms reported by the international Human Metabolomics Database ( www . hmdb . ca ) . Further , clear inter-disease differences within groups of adult diseases ( MS , WNV ) were evident in PCA ( Fig 2 ) , suggesting disease was much more influential in driving variation than was age . Sensitivity analyses in rabies in a larger dataset [12] did not identify meaningful age differences , although we cannot exclude the possibility that this might occur for other inflammatory diseases of the CNS . Another potential source of confounding is the timing of sample collection , which was not precisely known for samples other than rabies . All forms of encephalitis are treated empirically upon hospitalization , so early diagnostic samples such as those analyzed here may reflect early empirical therapies that often overlap ( e . g . , rehydration , provision of glucose , use of antibacterials , sedation ) but may also differ between diseases . Our choice of rabies samples centered on the fourth day of hospitalization was intended to minimize effects of dehydration and malnutrition , but may have biased rabies samples toward normality . Finally , differences in some metabolites should be interpreted with caution , since low concentrations in some specimens precluded exact quantification ( carnitine and glycine ) , which may have artificially led to statistical differences . Other metabolites ( glutamine and pyroglutamate ) are potentially affected by protein removal [45] , although this has not been shown in CSF . This study provides justification for further analysis of samples from these and other causes of encephalomyelitis . Several prominent and as of yet unidentified peaks observed in the spectra of some patients may indicate the presence of important metabolites involved in disease pathogenesis that have not yet been elucidated . While further studies with larger sample sizes will be needed to determine the clinical utility of NMR in the diagnosis of EM , NMR or other ‘omics technologies may in the future serve as a rapid initial screening test that would allow medical practitioners to initiate treatment with antivirals or biological immune modifiers , while patient samples can then be triaged to appropriate reference laboratories for confirmation without delaying treatment . Rabies and many arbovirus reference laboratories require specialized containment facilities , immunization of laboratory workers , and highly trained personnel who perform subjective assays such as immunofluorescence . Reference laboratories for rabies , arboviruses , bacteria and fungi are often dispersed geographically , leading to substantial requirements in volume , delay , and cost for diagnosis of encephalomyelitis when all are considered . NMR and MS instruments , on the other hand , exist at most research universities , i . e . at a state or provincial rather than national level . NMR analytical procedures are easily standardized and permit detection of multiple diseases using a single experiment , as illustrated here . NMR spectra can be transmitted electronically for analysis , which can be automated [46] . Decision analytical approaches such as CART and RF offer diagnostic flow charts that are easily implemented once validated , with quantifiable diagnostic probabilities . Considering current challenges , its relative ease of use makes NMR metabolomics of CSF a potentially important tool for emergent diseases and distinguishing between autoimmune and infectious EM . | Inflammation of the brain and spinal cord , known as encephalomyelitis , is a dangerous condition that can be caused by a wide range of pathogens , such as viruses and bacteria , and other medical conditions including autoimmunity or drug intoxications . Given the many possible causes , it is often difficult for clinicians treating patients with encephalomyelitis to identify the underlying cause , which in turn determines the appropriate treatment . Infections and other diseases causing neurological inflammation work by distinct biological mechanisms and , consequently , can cause unique biochemical changes that can be observed in cerebrospinal fluid of affected individuals . The researchers used a metabolomics technique to measure a range of small molecules in cerebrospinal fluid and examine biochemical differences in patients with encephalomyelitis caused by Lyme disease , West Nile Virus , multiple sclerosis , rabies , or fungal infection . The researchers found distinct differences in the biochemical profiles of patients whose encephalomyelitis was caused by infections versus patients with no infection , and also identified different patterns among the individual diseases . This study showed that metabolomics may be useful in improving diagnosis and treatment of diseases affecting the central nervous system by enhancing understanding of their unique effects on metabolism . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"cell",
"physiology",
"medicine",
"and",
"health",
"sciences",
"body",
"fluids",
"pathology",
"and",
"laboratory",
"medicine",
"multiple",
"sclerosis",
"nervous",
"system",
"neurodegenerative",
"diseases",
"pathogens",
"immunology",
"tropical",
"diseases",
"microbiology",
"cell",
"metabolism",
"viruses",
"rabies",
"demyelinating",
"disorders",
"clinical",
"medicine",
"metabolites",
"rna",
"viruses",
"neglected",
"tropical",
"diseases",
"pharmacology",
"drug",
"metabolism",
"infectious",
"diseases",
"zoonoses",
"medical",
"microbiology",
"metabolic",
"pathways",
"microbial",
"pathogens",
"pharmacokinetics",
"biochemistry",
"west",
"nile",
"virus",
"cell",
"biology",
"anatomy",
"flaviviruses",
"clinical",
"immunology",
"physiology",
"viral",
"pathogens",
"autoimmune",
"diseases",
"neurology",
"biology",
"and",
"life",
"sciences",
"viral",
"diseases",
"metabolism",
"cerebrospinal",
"fluid",
"organisms"
] | 2018 | NMR metabolomics of cerebrospinal fluid differentiates inflammatory diseases of the central nervous system |
The pathogenesis of severe acute respiratory syndrome coronavirus ( SARS-CoV ) is likely mediated by disproportional immune responses and the ability of the virus to circumvent innate immunity . Using functional genomics , we analyzed early host responses to SARS-CoV infection in the lungs of adolescent cynomolgus macaques ( Macaca fascicularis ) that show lung pathology similar to that observed in human adults with SARS . Analysis of gene signatures revealed induction of a strong innate immune response characterized by the stimulation of various cytokine and chemokine genes , including interleukin ( IL ) -6 , IL-8 , and IP-10 , which corresponds to the host response seen in acute respiratory distress syndrome . As opposed to many in vitro experiments , SARS-CoV induced a wide range of type I interferons ( IFNs ) and nuclear translocation of phosphorylated signal transducer and activator of transcription 1 in the lungs of macaques . Using immunohistochemistry , we revealed that these antiviral signaling pathways were differentially regulated in distinctive subsets of cells . Our studies emphasize that the induction of early IFN signaling may be critical to confer protection against SARS-CoV infection and highlight the strength of combining functional genomics with immunohistochemistry to further unravel the pathogenesis of SARS .
Infection with SARS-CoV causes lower respiratory tract disease with clinical symptoms that include fever , malaise , and lymphopenia [1] . Approximately 20%–30% of SARS patients require management in intensive care units , and the overall fatality rate has approached 10% . Interestingly , children seem to be relatively resistant to SARS , but the reason for this restriction is not known [2–4] . The clinical course of SARS follows three phases [5 , 6] . In the first phase , there is active viral replication and patients experience systemic symptoms . In the second phase , virus levels start to decrease while antibodies , which are effective in controlling infection , increase . However , pneumonia and immunopathological injury also develop in this phase . Ultimately , in the third phase , fatal cases of SARS progress to severe pneumonia and acute respiratory distress syndrome ( ARDS ) , characterized by the presence of diffuse alveolar damage ( DAD ) [1 , 7] . It has been hypothesized that the pathological changes are caused by a disproportional immune response , illustrated by elevated levels of inflammatory cytokines and chemokines , such as CXCL10 ( IP-10 ) , CCL2 ( MCP-1 ) , interleukin ( IL ) -6 , IL-8 , IL-12 , IL-1β , and interferon ( IFN ) -γ [8–13] . These in vivo data have been confirmed with in vitro experiments , demonstrating that SARS-CoV infection induces a range of cytokines and chemokines in diverse cell types [14–19] . In contrast , production of type I IFNs seems to be inhibited or delayed by SARS-CoV in vitro [14–18 , 20–22] . Moreover , no IFN-α or IFN-β has been detected in the sera of SARS patients or in lungs of SARS-CoV–infected mice [23–25] . Recent in vitro studies demonstrated that type I IFN inhibition or delay may be orchestrated by SARS-CoV proteins ORF 3B , ORF 6 , and N [26] . The inhibition of IFN production would benefit SARS-CoV replication , since pretreatment of cells with IFN before SARS-CoV infection efficiently prevents replication in these cells [21 , 27–30] . Furthermore , prophylactic treatment of macaques with pegylated IFN-α reduces SARS-CoV replication in the lungs [31] . Although IFN production was absent in clinical samples , gene and protein expression profiles in these patients were likely impacted by clinical treatments and concurrent preexisting disease . In addition , most if not all virus–host response information is from clinical blood/sera samples that were taken relatively late during infection—little is known about what happens early during infection . Animal studies are of great value to decipher the host's initial innate immune response , without confounding clinical treatment ( steroid and mechanical ventilation ) or underlying co-morbidity . In order to elucidate early host responses during the acute phase of SARS-CoV infection , we infected cynomolgus macaques with SARS-CoV and used macaque-specific microarrays and real-time ( RT ) -PCR techniques to study host gene expression profiles . Adolescent cynomolgus macaques infected with SARS-CoV develop DAD similar to SARS patients , but clear most of the virus in the lungs by day 6 [7] . Because SARS-CoV replicates predominantly in the lower respiratory tract of macaques , the virus infects a range of cells , including type 1 and type 2 pneumocytes , that are different from those analyzed in vitro . The ability to simultaneously examine virus replication and host response gene expression profiles in the lungs of these animals during the acute phase of SARS offers the opportunity to further unravel the pathogenesis of SARS .
Six cynomolgus macaques were inoculated with SARS-CoV strain HKU-39849 and lung tissues were collected at day 1 ( n = 2 , 1A and 1B ) or day 4 ( n = 4 , 4A–4D ) . No lesions or clinical symptoms were detected on day 1 after SARS-CoV infection , whereas on day 4 , three out of four monkeys were lethargic , with one of these animals showing mildly labored breathing . Pathological changes at day 4 post infection included DAD , characterized by flooding of the alveoli with edema fluid , infiltration of neutrophils , damage to the alveolar and bronchial epithelia , and occasional type 2 pneumocyte hyperplasia , as described earlier [31] . Four mock-infected animals were included in the study to serve as a reference for host response without viral challenge and to examine outbred inter-animal variation . Our previous experience with A/Texas/36/91 influenza virus demonstrated that viral mRNA was detected in representative samples of the lung rather than throughout the whole lung [32] . Based on this experience , the level of infection in separate lung samples was evaluated using RT-PCR . SARS-CoV mRNA was detected in all animals , and 13 pieces out of the total of 16 lung pieces from infected animals contained high levels of virus , while the three remaining pieces of lung contained very low levels of virus ( ∼3–4 logs lower , Figure 1A ) . No viral RNA could be detected in the samples from the mock-infected animals . For gene expression experiments , lung samples from SARS-CoV–infected animals were compared to a reference lung sample from mock-infected animals . The three samples with lower virus levels ( 1A-low , 4A-low , and 4D-low ) were analyzed individually so as not to dilute the gene expression of pooled pulmonary samples with higher SARS-CoV levels and also to potentially further define pulmonary infection . Samples from animals with high viral mRNA levels showed greater gene expression changes ( ∼2 , 000 genes day 1 , ∼800 genes day 4 ) compared to samples from animals with low levels of viral mRNA ( ∼400 genes ) , indicating a response of lung tissue to the virus ( Figure 1B ) . Additionally , the two day 1 animals showed higher numbers of differentially expressed genes than the day 4 animals . In contrast , gene expression analysis of the separate mock samples revealed limited differentially expressed genes . In order to examine how gene expression would be influenced by presence of virus , timing after inoculation , and individual animal variation , global expression profiling was performed . Hierarchical clustering methods were used to order rows ( genes ) and columns ( samples ) to identify groups of genes or samples with similar expression patterns [33 , 34] . These data were plotted as a heat map in which each matrix entry represents a gene expression value ( Figure 2A ) . Red corresponds to higher gene expression than that of the controls; green corresponds to lower gene expression . This analysis yielded 2 , 050 genes with day 1 samples on one side of the heat map and day 4 samples on the other side of the heat map , indicating an influence of timing after inoculation . There are two major roots to the hierarchical dendrogram , with the larger root composed of all the day 1 samples and the three day 4 samples with the highest virus levels . The smaller root is composed of the remaining day 4 samples with the lowest SARS-CoV levels . Although transcriptional profiling shows some variation when comparing samples from the same animal , the underlying gene expression is similar with a reduction in fold change in the “low” samples . These comparisons suggest that both individual animal variation and the “asynchronous” nature of the infection in the animals' lungs are factors involved in determining transcription of cellular genes . To validate that the host response from infected animals comprises a stronger transcriptional profile than individual variation from mock-infected animals , differential gene expression patterns in the separate mock samples were investigated , but only 38 genes were differentially expressed ( Figure 2B ) . These results suggest that underlying basal levels of gene transcription do not confound expression levels after infection . Even in a basal state , some low-level lung-to-lung variations were identified within the same animal but not enough to disrupt segregation of lung pieces based on mock-infected animals . In order to elucidate common responses to SARS-CoV throughout the infection as well as unique responses at different time points after inoculation , a Venn diagram was generated with each set ( circle ) holding to the parameters of an absolute fold change > 2 and p < 0 . 0001 in at least two animals ( Figure 3A ) . The day 1 set contained 1 , 278 genes and the day 4 set contained 950 genes . When examining host responses that were similar throughout the course of the infection , the intersection of the day 1 and day 4 sets indicates that 597 genes show shared responses . The heat map of these 597 genes is shown in Figure 3B . If more stringent criteria were used to find common responses in all six animals , using the 1 , 278 genes from the day 1 set and the 129 genes that are differentially expressed in all day 4 animals , a subset of 97 genes was identified . This subset included IFN-stimulated genes ( ISGs ) , like IFITs , MX1 , GBP1 , and G1P2 , and also various chemokines and cytokines , such as CXCL10 ( IP-10 ) , CCL2 ( MCP-1 ) , IL-6 , and IL-8 ( Figure S1 ) . These same cytokines and chemokines have been reported to be up-regulated in human SARS cases [9–12] . This set also included cathepsin L ( CTSL ) , which has been shown to be required for SARS-CoV entry into a cell [35] . Even though only 97 genes were commonly regulated in all animals , indicated with blue bars in Figure 3B , the heat map highlights that the other 500 genes show similar expression trends . Both sets of common-response genes showed similar functionality: cellular growth and proliferation , cell death , cellular movement , immune response , and cell-to-cell signaling . Next , we analyzed genes that were differentially expressed exclusively on either day 1 or day 4 in order to find signature gene expression patterns for each day . Genes identified as unique responses at day 1 ( 681 genes ) and at day 4 ( 353 genes ) in the Venn diagram showed unique functionality ( Figure 3C ) . The gene expression profile at day 1 shows a prominent innate host response to viral infection; top functional categories on day 1 are the immune response , the hematological system , and the immune and lymphatic system . Genes like IFN-γ , CCL4 ( MIP-1-β ) , CSF3 , IL1A , and TNF are included in these categories . At day 4 , a smaller panel of unique differentially expressed genes that play a role in cell cycle , cellular assembly , and DNA repair were identified like CCNB2 , CCNE1 , CDCA5 , CENPA , CHAF1A , and PRC1 . In order to investigate genes that are most strongly regulated after SARS-CoV infection , genes included in the Venn diagram ( Figure 3A ) that also held to an absolute fold change > 5 were queried ( Figure S2 ) . From this set , genes that were involved in the immune response and lung repair processes were used to generate a heat map ( Figure 4 ) . A number of genes that have been reported to be up-regulated in SARS patient sera , such as CCL2 ( MCP-1 ) , CXCL10 ( IP-10 ) , IL-6 , and IL-8 , were strongly ( ∼20-fold ) induced in all animals . Many cell cycle and matrix genes indicative of tissue repair processes were also highly differentially expressed at day 4 ( e . g . , ANLN , AREG , CDC2 , CDKN3 , CKS2 , FOSL1 , and KIF2C ) . Likewise , tissue factor pathway inhibitor 2 ( TFPI2 ) , an anticoagulant , was strongly up-regulated during infection in all animals ( averaging ∼20-fold ) , as well as PLSCR1 , SERPINE1 ( PAI1 ) , and THBS1 , all genes involved in pro-coagulation and platelet activation , were induced . Concomitant expression of TFPI2 with these pro-coagulation genes might function as an inhibitory response to restrain the activation of the coagulation pathway during acute inflammation . Surprisingly , expression of diverse IFN-α genes and expression of IFN-β was up-regulated ∼10- to 20-fold in the day 1 samples . Furthermore , IFN-γ , a type II IFN , was efficiently transcribed on day 1 after SARS-CoV infection ( ∼5-fold ) . Other genes associated with the induction of IFNs like DDX58 ( Rig-I ) , IRF-7 , and signal transducer and activator of transcription 1 ( STAT1 ) , were also highly induced ( ∼8-fold ) . Up-regulation of type I IFNs in these SARS-CoV infected macaques is remarkable , since SARS-CoV inhibits IFN production in many in vitro studies . We did not detect induced IFN-β mRNA expression using Ma104 cells or Caco2 cells and the SARS-CoV-HKU virus ( unpublished data ) . Not only IFNs , but also several IFN-responsive genes ( e . g . , G1P2 , GBP1/2 , IFI/IFITs , MX1/2 , ISG20 , and OAS1/2/L ) were highly transcribed , showing a persistent activation of the innate immune response . Furthermore , suppressor of cytokine signaling 1 ( SOCS1 ) is induced at the onset of infection , presumably to establish negative feedback to attenuate cytokine signaling . Of note , IFIT1 ( ISG56/IFI56 ) , often used to gauge IFN induction , was up-regulated an average of ∼13-fold . To further explore some of the pathogenic and antiviral pathways that are induced after SARS-CoV infection , we investigated the transcription of various cytokines , chemokines , IFNs , ISGs , and transcription factors involved in the JAK/STAT pathway . As can be seen in Figure 5A , a wide range of chemokines and cytokines are differentially expressed after SARS-CoV infection in macaque lungs , especially on day 1 after infection . Besides previously mentioned chemokines , we detected monocyte chemotactic protein genes like CCL8 ( MCP-2 ) and CCL7 ( MCP-3 ) , but also CCL11 ( eotaxin ) , a chemotactic protein for eosinophils . In the samples with low SARS-CoV mRNA levels , the induction of chemokines is less evident , suggesting that the presence of these molecules is restricted to areas in the lung where virus is present . Furthermore , SARS-CoV–infected macaques showed a stronger induction of IFNs ( 14 unique genes ) and ISGs ( 20 unique genes ) on day 1 than day 4 and when virus was present at high levels . Note that besides IFN-α , IFN-β , and IFN-γ , the IFN-λs ( IL-29 , IL-28A , IL-28B ) , which are type I IFNs , were induced in samples with high SARS-CoV levels . In the absence of viral RNA , no IFNs , but interestingly , a number of ISGs ( 17 unique genes ) were detected , suggesting paracrine stimulation ( Figure 5B ) . Differential expression of a selection of strongly up-regulated genes , CXCL10 ( IP-10 ) , IL-6 , IL-8 , and IFN-β , was confirmed using RT-PCR ( Figure 6 ) . In accordance with the microarray data , the RT-PCR data showed that CXCL10 ( IP-10 ) , IL-6 , IL-8 , and IFN-β were all expressed at levels that were approximately 100 times higher in the SARS-CoV–infected animals at day 1 than in the uninfected control animals and were still elevated on day 4 after infection . As can be seen in Figure 6 , the induction of IFN-β was strongly correlated to the presence of virus ( rspearman= 0 . 88 , p < 0 . 0001 ) . For CXCL10 ( IP-10 ) , IL-6 , and IL-8 the correlation is less evident , which is not surprising since these cytokines can be induced by other factors than the virus itself . In order to visualize the host response in the lungs of SARS-CoV–infected macaques , IFN-β production and translocation of phosphorylated STAT1 was studied using immunohistochemistry . In the lungs of the SARS-CoV–infected macaques , a modest number of cells stained positive for IFN-β at day 1 post infection , whereas no IFN-β–positive cells could be detected in mock-infected macaques ( Figure 7A–7C ) . Notably , most of the cells that stained positive for IFN-β were located very close to blood vessels , but not in the alveoli where most SARS-CoV antigen-positive cells ( mainly type 2 pneumocytes at 1 day post infection ) are located . To examine whether the IFNs that are produced in the lungs of these SARS-CoV–infected macaques are biologically active and able to induce STAT1 phosphorylation and translocation , lung sections of the infected macaques were stained with antibodies against phosphorylated STAT1 . As shown in Figure 7D and 7E , no phosphorylated STAT1 could be detected in the lungs of PBS-infected macaques , while in the lungs of SARS-CoV–infected macaques , cells with phosphorylated STAT1 in their nucleus were abundantly present . Subsequently , the same pieces of lung from SARS-CoV–infected macaques at day 1 were double stained for phosphorlylated STAT1 and SARS-CoV ( Figure 7 F ) . Notably , phosphorylated STAT1 was not detected in the nucleus of SARS-CoV–infected cells ( type 2 pneumocytes ) , while cells directly adjacent to these SARS-CoV–infected cells stained for phosphorylated STAT1 in many , but not all , foci containing SARS-CoV–positive cells . Thus , type I IFNs are produced in the lungs of SARS-CoV–infected macaques , and are able to activate the JAK/STAT pathway . However , translocation of STAT1 does not occur in SARS-CoV–infected pneumocytes . Although recent studies indicate that the SARS-CoV ORF6 protein is able to inhibit nuclear translocation of STAT1 in vitro , this was not demonstrated in experiments using infectious SARS-CoV [26] . In order to assess whether SARS-CoV inhibits phosphorylation and translocation of STAT1 , MA104 cells were infected with SARS-CoV for 24 h and then either fixed directly or treated with type I IFN . Cells infected with SARS-CoV , but not treated with IFN , stained positive for SARS-CoV ( unpublished data ) , but lacked staining for phosphorylated STAT1 , indicating that SARS-CoV or other soluble mediators are not able to induce STAT1 phosphorylation ( Figure 8 ) . After treatment of the MA104 cells with IFN , phosphorylated STAT1 could be detected in the nucleus of most cells , but not in the nucleus of SARS-CoV–infected cells ( Figure 8 ) . This demonstrates that SARS-CoV inhibits the translocation of phosphorylated STAT1 to the nucleus , confirming our in vivo data . Besides inhibiting translocation of phosphorylated STAT1 , SARS-CoV also seems to reduce STAT1 phosphorylation , as the majority of SARS-CoV–infected cells contained low levels of phosphorylated STAT1 in their cytoplasm .
Pathogenic viruses escape the antiviral action of the IFN system by inhibiting both IFN production and signaling pathways . Here , we report that even though production and signaling of type I IFNs is inhibited by SARS-CoV in vitro as well as in SARS-CoV–infected cells in vivo , high levels of type I IFNs are induced in the lungs of SARS-CoV–infected macaques . These IFNs are able to activate STAT1 , followed by the transcription of numerous ISGs . Using immunohistochemistry , we revealed that these antiviral signaling pathways were differentially regulated in distinctive subsets of cells . Our results emphasize the strength of combining functional genomics with immunohistochemistry to further unravel the pathogenesis of SARS-CoV infection in cynomolgus macaques . To our knowledge , this study represents the first functional genomics investigation of SARS-CoV infection in cynomolgus macaques . All experimental animals showed signs of infection because viral mRNA could be detected in random samples from the lung , indicating that the virus had spread throughout the whole lung at the time of necropsy . Furthermore , pathological examination of SARS-CoV–infected macaques at day 4 post infection revealed multifocal DAD [31] . Unlike 10% of humans with SARS , which are mainly restricted to the elderly , adult macaques used in this study do not succumb to SARS-CoV infection . However , the SARS-CoV–induced pathology in these macaques likely resembles the pathological changes seen in the majority of human SARS patients that recover from the disease . Although none of the current animal models has fully reproduced all features of SARS , the most important aspects of this disease are observed in experimentally infected macaques , providing valuable insights into the initial innate immune response after infection without confounding clinical treatment or underlying co-morbidity . Using macaque-specific microarrays , we were able to observe that with early infection , high levels of viral mRNA corresponded to a strong cellular host response . This strong host response is dominated by genes involved in the immune response and includes a wide range of genes corresponding with what is seen in human ARDS . During the acute phase of human ARDS , activated neutrophils and macrophages enter the alveoli and produce a number of cytokines and chemokines such as IL-6 , IL-8 , and CXCL10 ( IP-10 ) [36] , as were found in the lungs of our SARS-CoV–infected macaques . Researchers have postulated that these genes also predict adverse SARS patient outcome [37] . During the chronic phase of human ARDS , type 2 pneumocytes start to proliferate and differentiate in order to repair the damaged lung . At day 4 , the macaque lung shows similar evidence of lung repair , and numerous genes are up-regulated that are involved in cellular growth and proliferation , cell cycle regulation , and DNA replication and repair . Genes involved in cell cycle regulation and proliferation have been previously reported in coronavirus infections other than SARS-CoV and have been characterized by an accumulation of infected cells in the G0/G1 phase [14 , 38] . We also detected a strong presence of genes involved in the coagulation pathway , including TFPI2 , SERPINE1 , and TIMP1 . The idea of a pro-coagulation profile mimics the clinical-pathological observations of SARS patients that showed unusually disseminated small vessel thromboses in the lungs [5 , 39] . Additionally , cathepsin L was up-regulated in all SARS-CoV–infected macaques . Induction of this gene after SARS-CoV infection is quite interesting because cathepsin L is an endosomal protease that is necessary for SARS-CoV to infect a cell [35] . Remarkably , SARS-CoV infection in macaques leads to a strong transcription of IFNs . Not only IFN-α , IFN-β , and IFN-λ ( all type I IFNs ) , but also IFN-γ , a type II IFN , were all highly up-regulated , especially on day 1 after infection . The expression of IFN-β , which strongly correlated to the amount of virus present , continued throughout day 4 and was confirmed using immunohistochemistry; IFN-β–positive cells could be detected in the lungs of the SARS-CoV–infected macaques . The induction of IFN-β in these SARS-CoV–infected macaques is surprising , because several reports have shown that SARS-CoV inhibits or delays type I IFN production in a number of cell types [14–18 , 20 , 22] . For example , SARS-CoV blocks a step in the activation of IRF-3 , a transcription factor that is required for IFN-β induction [21] . In addition , the SARS-CoV proteins ORF3B , ORF6 , and nucleocapsid have been shown to function as IFN antagonists , as has the SARS-CoV nsp1 gene that prevents the production of Sendai virus–induced IFN-β in 293 cells [26 , 40] . Interestingly , it was recently shown that plasmacytoid dendritic cells ( pDCs ) are able to produce IFN-α and IFN-β after SARS-CoV infection , while conventional DCs did not produce these type I IFNs [41] . pDCs are known for their ability to produce very high amounts of IFN-α and IFN-β and are considered first-line sentinels in immune surveillance in the lung [42–46] . We speculate that the IFN-β–producing cells detected in the lungs of SARS-CoV–infected macaques are pDCs . Future studies may address the nature of these IFN-producing cells once technical difficulties in detecting pDCs in macaque tissues have been tackled . These studies may also shed light on whether decreasing numbers of pDCs observed in clinical blood samples from human SARS patients are caused by sequestering of pDCs by the lungs , destruction of pDCs by SARS-CoV , or destruction or suppression of pDCs by steroid treatment [47] . When IFNs are produced , they bind to their receptors on the cell membrane , after which STAT1 , a key member of the JAK/STAT pathway , is phosphorylated and subsequently translocated to the nucleus , followed by the production of a wide range of IFN-stimulated genes . In vitro , SARS-CoV inhibited translocation of STAT1 to the nucleus , and phosphorylation of STAT1 was strongly reduced . However , the inhibition of STAT1 phosphorylation was not absolute because cells with low levels of phosphorylated STAT1 in their cytoplasm were also detected . In accordance with our data , Kopecky-Bromberg et al . recently showed that the SARS-CoV protein ORF6 is able to inhibit STAT1 translocation [26] . This strategy is not unique to SARS . Other viruses have been shown to be able to block signaling of IFNs by affecting phosphorylation and/or translocation of the STAT proteins . For example , measles virus V protein inhibits translocation of STAT1 , but does not affect phosphorylation , whereas measles virus P protein blocks both of these processes [48] . Other paramyxoviruses , like Rinderpest virus , Nipah virus , Hendra virus , and mumps virus , as well as flaviviruses like West Nile virus and Japanese encephalitis virus , are able to block activation of STAT1 and STAT2 [49–52] . Inhibition of STAT1 phosphorylation is not always complete . For example , Sendai virus suppresses tyrosine phosphorylation of STAT1 during the early stages of infection , but this block becomes leaky after a couple of hours with phosphorylated STAT1 accumulating in the cytoplasm [53] . In contrast to these in vitro data , we observed phosphorylated STAT1 in the nuclei of numerous cells in the lungs of SARS-CoV–infected macaques , indicating that these cells had been activated by the IFNs produced in the lung . However , phosphorylated STAT1 was not detected in SARS-CoV–infected cells . The observations made in this study indicate that SARS-CoV–infected macaques produce IFNs in response to virus infection and are further capable of activating the STAT1 pathway in cells surrounding the SARS-CoV–infected cells . The importance of IFNs in controlling SARS-CoV infection has been suggested in several animal studies . Mice clear SARS-CoV in the absence of NK cells , T cells , or B cells , suggesting that innate immune responses are sufficient to limit SARS-CoV infection in these animals [23] . Indeed , STAT1 knock out mice , which are resistant to the effects of IFNs , to some extent show a worsening of pulmonary disease and an increase in viral replication in the lungs compared to normal mice after infection with SARS-CoV [54] . Although IFN treatment was not conducted in SARS-CoV infection mouse studies , prophylactic treatment of macaques with pegylated IFN-α protects type 1 pneumocytes from infection with SARS-CoV [31] . In addition , potent antiviral activity is observed in vitro when cells are treated with IFNs before they are infected with SARS-CoV [27 , 29 , 30] . Although we cannot determine the effect of neutralizing IFN-β in SARS-CoV–infected animals , based on the experiments utilizing recombinant IFNs in these animals , we postulate that type I IFNs are partly responsible for the relatively mild clinical symptoms that are seen in SARS-CoV–infected macaques . In addition , a recent study again demonstrated the importance of IFNs in viral infections , as macaques infected with the highly pathogenic and fatal 1918 influenza virus showed limited induction of type I IFNs ( only IFNA4 reached fold changes > 5 ) and delayed induction of ISGs , while macaques infected with the low-pathogenic K173 influenza virus showed a strong induction of these antiviral molecules early during infection [55] . Notably , IFN-β was not up-regulated ( absolute fold change < 2 ) in any of the influenza virus–infected animals , even in those animals that recovered , unlike SARS-CoV–infected macaques that showed a very strong presence of IFN-β . In conclusion , our study demonstrates that cynomolgus macaques can be infected with SARS-CoV , as indicated by presence of viral mRNA at different locations throughout the lung at day 1 and day 4 , with gross pathology becoming noticeable at day 4 . Furthermore , we show that infection of cynomolgus macaques with SARS-CoV leads to a strong immune response , including the induction of various cytokines and chemokines , resembling the host response seen in human SARS patients . Strikingly , despite the fact that SARS-CoV infection blocks the production of IFNs in vitro , type I IFNs are strongly induced in the lungs of SARS-CoV–infected macaques . The production of IFN early during infection leads to widespread activation of STAT1 and the production of ISGs . This suggests that , although SARS-CoV blocks IFN signaling in infected cells , locally produced IFNs are capable of activating non-infected cells and possibly can prevent infection of these cells . Thus , SARS-CoV infection in macaques leads to the differential activation of both pathogenic and antiviral signaling pathways in vivo , and the outcome may be determined by the relative contribution of these signaling pathways .
Six cynomolgus macaques ( Macaca fascicularis ) were infected intratracheally with 1 × 106 TCID50 SARS-CoV ( HKU-39849 ) as described earlier [31] . Virus stocks were generated in Vero E6 cells that were defective in IFN production . Two animals were euthanized on day 1 after infection and four animals were euthanized on day 4 . In addition , four animals were mock ( PBS ) infected and euthanized on day 4 , serving as a negative control group . One lung from each monkey was fixed in 10% formalin for histopathology and immunohistochemistry while the other was used for real-time PCR and microarrays . Lung samples were randomly excised from three different lung areas ( cranial , medial , caudal ) and stored in RNAlater ( Ambion , http://www . ambion . com/ ) . Sixteen pieces of lung were taken from the SARS-CoV–infected animals , two to three pieces of lung per animal . Twelve pieces of lung were taken from the mock-infected animals , three pieces of lung per animal . Individual lung samples in RNAlater were transferred to TRIZOL Reagent ( Invitrogen , http://www . invitrogen . com/ ) , homogenized using Polytron PT2100 tissue grinders ( Kinematica , http://www . kinematica . ch ) , and then processed to extract RNA . All experiments were executed under a biosafety level 3 , and approval for animal experiments was obtained from the Institutional Animal Welfare Committee . Infected macaque lung samples were co-hybridized with a reference mock-infected macaque lung sample on macaque oligonucleotide arrays containing 131 viral probes , corresponding to 26 viruses , and 22 , 559 rhesus probes , corresponding to ∼18 , 000 rhesus genes . The reference mock-infected sample was created by pooling equal mass quantities of total RNA extracted from the 12 individual lung pieces from mock-infected animals . An Agilent 2100 bioanalyzer was used to check the purity of the total RNA prior to cRNA probe production with the Agilent Low RNA Input Fluorescent Linear Amplification kit ( Agilent Technologies , http://www . agilent . com/ ) . Arrays were scanned with an Agilent DNA microarray scanner , and image analysis was performed using Agilent Feature Extractor Software ( Agilent Technologies ) . Each microarray experiment was done with two technical replicates using dye reversal [56] . All data were entered into a custom-designed database ( Expression Array Manager ) and analyzed with Resolver 4 . 0 ( Rosetta Biosoftware , http://www . rosettabio . com/ ) and Spotfire DecisionSite for Functional Genomics ( Spotfire , http://www . spotfire . com/ ) . In our data analysis , genes were selected to be included for transcriptional profile based on two criteria: a greater than 99 . 99% probability of being differentially expressed ( p ≤ 0 . 0001 ) and an expression level change of 2-fold or greater . Ingenuity Pathway Analysis ( Ingenuity Systems , http://www . ingenuity . com/ ) was used to functionally annotate genes according to biological processes and canonical pathways . In accordance with proposed MIAME standards , primary data are available in the public domain through Expression Array Manager at http://expression . microslu . washington . edu/expression/index . html [57] . RT-PCR was performed to detect SARS-CoV mRNA and to validate cellular gene expression changes as detected with microarrays . Each reaction was run in triplicate using Taqman 2x PCR Universal Master Mix ( Applied Biosystems , http://www . appliedbiosystems . com/ ) with primers and probe specific for the SARS-CoV nucleoprotein gene [7] , or for macaque cellular genes ( sequences shown in Table 1 ) . Differences in gene expression are represented as the fold change in gene expression relative to a calibrator and normalized to a reference , using the 2−ΔΔCt method [58] . GAPDH ( glyceraldehydes-3-phosphate dehydrogenase ) or 18S rRNA were used as endogenous controls to normalize quantification of the target gene . The samples from the mock-infected macaques were used as a calibrator . Formalin-fixed , paraffin-embedded lung samples from SARS-CoV–infected and mock-infected macaques were stained for SARS-CoV , phosphorylated STAT1 , and IFN-β using mouse-anti-SARS-nucleocapsid ( Clone Ncap4 , mouse IgG2b; Imgenex , http://www . imgenex . com/ ) , mouse-anti-phospho-STAT1 ( Clone ST1P-11A5 , mouse IgG2a-κ; Zymed Laboratories , http://www . invitrogen . com/ ) , and rabbit-anti -IFN-β ( Chemicon , http://www . chemicon . com/ ) , respectively . After deparaffinization , antigen retrieval was performed using a citrate buffer for the SARS-CoV and STAT1 staining . No antigen retrieval was performed when staining for IFN-β . Goat-anti-mouse IgG2a HRP , goat-anti-mouse IgG2b AP ( Southern Biotech , http://www . southernbiotech . com/ ) , and anti-rabbit IgG-HRP ( DAKO , http://www . dako . com/ ) were used as secondary antibodies . Signals were developed with Fast Red and DAB ( Sigma , http://www . sigmaaldrich . com/ ) and counterstained with Mayer's hematoxylin . MA104 cells ( African green monkey foetal kidney cells , ECACC ) were cultured in Eagle's Minimal Essential Medium ( EMEM; Cambrex , http://www . cambrex . com/ ) supplemented with 2 mM glutamine , 1% non-essential amino acids and 10% foetal bovine serum . Cells were seeded in 96-well plates and infected with SARS-CoV ( MOI 0 . 5 ) , and 24 h after infection , selected wells were treated with universal type I IFN ( 5 , 000 U/ml , Sigma ) for 30 min at 37 °C . Subsequently , cells were fixed with 10% neutral-buffered formalin and treated with 70% ethanol . SARS-CoV–infected cells were visualized using purified human IgG from a convalescent SARS patient ( CSL ) , followed by staining with an antibody to human IgG , linked to Alexa Fluor 594 ( Invitrogen ) . Phosphorylated STAT1 was visualized using mouse-anti-phospho-STAT1 ( Zymed ) , followed by staining with a FITC-linked antibody to mouse IgG . | Severe acute respiratory syndrome coronavirus ( SARS-CoV ) infection causes a progressive atypical pneumonia . In typical cases , largely confined to adult and elderly individuals , acute respiratory distress syndrome develops , and admission to an intensive care unit is required . Although these complications can be fatal , most SARS patients recover , suggesting that protective immune responses are operational . In this study , we simultaneously examined virus replication and host–response gene expression profiles in macaque lungs during the acute phase of SARS to gain more insight into the early events that take place after SARS-CoV infection . We show that a strong host response is induced in the lungs of SARS-CoV–infected macaques , illustrated by the induction of several pathogenic cytokines and chemokines . Interestingly , antiviral pathways are activated as well , demonstrated by the presence of phosphorylated signal transducer and activator of transcription 1 ( STAT1 ) transcription factors throughout the lung , but not in SARS-CoV–infected cells . A subset of cells was shown to produce interferon-β , a cytokine involved in the resistance to many viral infections and able to activate STAT1 . Activation of this antiviral pathway upon SARS-CoV infection may be an important escape route of the host to withstand the devastating effects of SARS-CoV . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"primates",
"viruses",
"virology",
"vertebrates",
"animals",
"mammals"
] | 2007 | Functional Genomics Highlights Differential Induction of Antiviral Pathways in the Lungs of SARS-CoV–Infected Macaques |
Middle East respiratory syndrome coronavirus ( MERS-CoV ) is a novel human coronavirus that emerged in 2012 , causing severe pneumonia and acute respiratory distress syndrome ( ARDS ) , with a case fatality rate of ~36% . When expressed in isolation , CoV accessory proteins have been shown to interfere with innate antiviral signaling pathways . However , there is limited information on the specific contribution of MERS-CoV accessory protein 4b to the repression of the innate antiviral response in the context of infection . We found that MERS-CoV 4b was required to prevent a robust NF-κB dependent response during infection . In wild-type virus infected cells , 4b localized to the nucleus , while NF-κB was retained in the cytoplasm . In contrast , in the absence of 4b or in the presence of cytoplasmic 4b mutants lacking a nuclear localization signal ( NLS ) , NF-κB was translocated to the nucleus leading to the expression of pro-inflammatory cytokines . This indicates that NF-κB repression required the nuclear import of 4b mediated by a specific NLS . Interestingly , we also found that both in isolation and during infection , 4b interacted with α-karyopherin proteins in an NLS-dependent manner . In particular , 4b had a strong preference for binding karyopherin-α4 ( KPNA4 ) , which is known to translocate the NF-κB protein complex into the nucleus . Binding of 4b to KPNA4 during infection inhibited its interaction with NF-κB-p65 subunit . Thereby we propose a model where 4b outcompetes NF-κB for KPNA4 binding and translocation into the nucleus as a mechanism of interference with the NF-κB-mediated innate immune response .
Middle East respiratory syndrome coronavirus ( MERS-CoV ) is a human coronavirus that emerged in the Middle East in 2012 . Since then , 2100 confirmed cases and 730 deaths have been reported , with an estimated mortality rate of ~36% ( http://www . who . int/csr/don/10-november-2017-mers-oman/en/ ) . Currently , MERS-CoV remains a public health threat , as new infections continue to occur and no approved vaccines or antivirals are available . Furthermore , MERS-CoV is endemic in camels , making it likely that the virus will be an ongoing concern . Evidence from a number of studies has demonstrated that MERS-CoV strongly inhibits IFN-α/β production [1–3] , although our knowledge on how the virus interferes with the host antiviral response is still limited . The contribution of viral proteins to the inhibition of the innate antiviral and pro-inflammatory responses has been mainly addressed in cells overexpressing individual viral proteins , not in the context of virus infection . The nuclear factor-κB ( NF-κB ) family of proteins are critical regulators of both innate and adaptive immunity [4] . In naïve cells , NF-κB proteins are present in the cytoplasm in association with inhibitory proteins ( IκBs ) . Cellular stimuli , including pathogens and stress , induce IκB phosphorylation , ubiquitination and degradation by the proteasome , releasing NF-κB proteins and allowing their nuclear translocation . In the nucleus , NF-κB proteins promote the transcription of pro-inflammatory cytokines and chemokines , stress-response proteins , and anti-apoptotic proteins . NF-κB activity is essential for lymphocyte survival and activation , and for initiating and propagating optimal immune responses . By contrast , the constitutive activation of the NF-κB pathway is often associated with inflammatory diseases , such as rheumatoid arthritis and asthma . Exacerbated NF-κB activity has been demonstrated to be important for the inflammatory immunopathology induced by respiratory viruses such as severe acute respiratory syndrome ( SARS ) -CoV [5 , 6] and influenza A virus [7] . Here , we found that nuclear MERS-CoV 4b protein contributes to the inhibition of the host NF-κB-mediated pro-inflammatory response by preventing the nuclear translocation of NF-κB . In addition , we identified karyopherin-α4 ( importin-α3 ) , which is responsible for the nuclear import of NF-κB , as a 4b binding protein . We found that 4b binding to karyopherin-α4 inhibited its interaction with NF-κB-p65 subunit , thus preventing NF-κB nuclear translocation and , as a consequence , expression of NF-κB-dependent pro-inflammatory cytokines during infection . The ability of 4b to block NF-κB is likely to be relevant for MERS-CoV-induced in vivo pathology and virulence .
To address whether MERS-CoV accessory genes 3 , 4a , 4b or 5 interfere with the innate immune response during infection , we infected Huh-7 cells with previously engineered MERS-CoV deletion mutants ( Δ3 , Δ4ab , Δ5 ) [8] and analyzed the expression of interferon β ( IFN-β ) and the pro-inflammatory cytokines IL-6 , IL-8 , and TNF-α . Viruses deleted in the accessory proteins , like the parental virus [3] , were unable to induce IFN ( S1 Fig ) in contrast to the effective induction of IFN-β by poly ( I:C ) transfection , implying that additional proteins expressed by the virus act as IFN antagonists [9] . Interestingly , dual deletion of genes 4a and 4b led to a significant increase in the expression levels of IL-6 , IL-8 , and TNF-α as compared to the wild-type infection ( S1 Fig ) . Since the transcription of these genes depends on NF-κB [4] , these results suggested that one or both proteins repressed NF-κB activation during MERS-CoV infection . In order to determine the specific contribution of 4a and 4b proteins to the inhibition of the NF-κB response during infection , individual deletion mutants MERS-CoV-Δ4a and MERS-CoV-Δ4b were engineered ( Fig 1A ) . The 4a and 4b genes have overlapping coding sequences and are expressed from the same subgenomic mRNA , and thus 4b is likely expressed by either internal ribosomal entry or “leaky scanning” of ribosomes [10] . The 4a deletion preserved the transcription-regulating sequences ( TRS ) preceding 4a gene , which comprise the conserved sequence ( CS ) , and 26 nt downstream of the CS , including the first 22 nt of ORF 4a . In addition , the 3’ 101 nt of the 4a gene preceding ORF 4b were maintained in order to preserve translation of 4b protein , as suggested for other CoV ORFs that are translated by similar mechanisms [11] . In the case of MERS-CoV-Δ4b the 4b gene was almost entirely deleted excluding the 4a/4b overlapping sequences and the 3’ most 53 nt , in order to maintain the TRS for transcription of gene 5 ( Fig 1A ) . MERS-CoV-Δ4a and MERS-CoV-Δ4b were recovered from plasmids pBAC-MERSFL-Δ4a and pBAC-MERSFL-Δ4b , respectively , in Huh-7 cells with titers similar to those of wild-type virus . After two passages in cell culture , the recombinant viruses were sequenced to confirm their stability . The replication kinetics of these singly mutated viruses in Huh-7 cells was not significantly different from that of the wild-type virus ( Fig 1B ) , with only a slight decrease in titers at some time points . These results confirm that proteins 4a and 4b were not required for virus replication in cell culture [8] . The presence or absence of 4b and 4a proteins in Huh-7 cells infected with the recombinant viruses MERS-CoV-Δ4a , MERS-CoV-Δ4b and MERS-CoV-Δ4ab was confirmed by Western-blot and immunofluorescence assays . The expression levels of 4a and 4b proteins ( A in S2 Fig ) and their subcellular distribution ( B and C in S2 Fig ) were similar to those of the wild-type virus , indicating that 4a and 4b gene expression was not affected in the deletion mutants . While 4a protein showed a cytoplasmic distribution [9 , 12 , 13] partially colocalizing with viral dsRNA , 4b protein localized to the nucleus of infected cells , as previously described [10] . The expression of mRNAs coding for NF-κB-dependent pro-inflammatory cytokines was analyzed in Huh-7 cells infected with the deletion mutants Δ4a and Δ4b at a multiplicity of infection ( MOI ) of 1 PFU/cell ( Fig 1C ) . While deletion of 4a did not have any effect on the expression of IL-6 , IL-8 and TNF-α as compared to the wild-type virus , deletion of 4b significantly increased their expression to similar or even higher levels than those observed in MERS-CoV-Δ4ab , indicating that the 4b protein contributed to the repression of NF-κB-dependent gene expression . To determine whether import of 4b protein into the nucleus was required for interference with the NF-κB pathway , MERS-CoV mutants defective in 4b nuclear localization were engineered ( Fig 2A ) . MERS-CoV 4b protein includes a bipartite NLS consisting of two triplets of positively charged amino acids [10] , named site 1 ( NLS-S1 ) and site 2 ( NLS-S2 ) ( Fig 2A ) . Mutation of NLS-S1 in the context of 4b overexpression completely prevented nuclear import , while mutation of NLS-S2 only partially hindered nuclear localization of 4b [10] . Two consecutive lysine residues located between NLS-S1 and NLS-S2 might also contribute to the nuclear localization of 4b , and were included in our mutagenic analysis . Since the 4b NLS-S1 is located within the 4a/4b overlapping sequence , its mutagenesis required the separation of 4a and 4b genes by duplicating the common region . A recombinant MERS-CoV-DUP cDNA with a 189 nt duplication including the 4a/4b overlapping sequences ( 88 bp ) and 4a preceding sequences ( 101 bp ) , likely required for 4b translation , was generated as a control ( DUP , Fig 2A ) . Two mutant MERS-CoV cDNAs including the 4a/4b duplication were then engineered either containing alanine mutations of NLS-S1 ( RKR→AAA , MERS-CoV-DUP-mNLS-S1 ) , or both the intermediate Lys residues and NLS-S2 ( KK→AA and KRR→AAA , MERS-CoV-DUP-mNLS-S2 ) . Recombinant viruses were rescued from infectious cDNAs in Huh-7 cells and the expression of 4a and 4b proteins was characterized . No significant differences in the expression levels of 4a and 4b proteins in Huh-7 cells were observed by Western-blot analysis , as compared to the wild-type virus ( A in S3 Fig ) , confirming that genome engineering of 4b NLS mutants did not affect virus gene expression . Accordingly , the cytoplasmic distribution of 4a protein observed in MERS-CoV-wild-type-infected cells was not affected in the NLS mutants ( B in S3 Fig ) . Analysis by confocal microscopy of 4b subcellular localization ( Fig 2B ) confirmed that 4b protein was mostly detected in the nucleus in MERS-CoV-DUP-infected cells , as observed in cells infected with wild-type virus . In contrast , mutation of 4b NLS-S1 in MERS-CoV-DUP-mNLS-S1 prevented the nuclear localization of 4b , which was mainly restricted to the cytoplasm . Mutation of NLS-S2 in MERS-CoV-DUP-mNLS-S2 only partially prevented the nuclear import of 4b , with protein detected in both the nucleus and the cytoplasm . These results demonstrated that in the context of infection , 4b NLS-S1 was essential for nuclear localization , while NLS-S2 had only a partial effect on 4b nuclear import , as described in 4b overexpression experiments [10] . The growth kinetics of MERS-CoV NLS mutants in Huh-7 and Calu-3 ( a bronchial epithelial cell line ) cells ( Fig 2C and C in S3 Fig ) confirmed that neither the duplication of 4a-4b sequences nor the mutation of 4b NLS-S1 or NLS-S2 affected viral replication as compared to the wild-type virus . This indicates that the nuclear localization of 4b protein during infection did not have a significant impact on virus replication in cell culture . To evaluate the relevance of 4b nuclear localization in the antagonism of the innate immune response during infection , Huh-7 and Calu-3 cells were infected with 4b-NLS mutants at a MOI of 1 PFU/cell . The expression of NF-κB-mediated pro-inflammatory cytokines IL-6 , IL-8 and TNF-α was analyzed by RT-qPCR at 24 hpi ( Fig 3A–3C ) . Mutation of NLS-S1 led to significantly elevated expression levels of NF-κB dependent pro-inflammatory cytokines , similar to those induced by MERS-CoV-Δ4b . Mutation of NLS-S2 , leading to a partially nuclear 4b protein , increased the expression of pro-inflammatory cytokines to intermediate levels as compared to the NLS-S1 mutation . These results indicated that nuclear translocation of 4b was required to inhibit the expression of NF-κB dependent pro-inflammatory cytokines both in Huh-7 and Calu-3 cells , suggesting that this effect might be relevant in the natural infection . Thus , MERS-CoV infection induced an NF-κB-dependent response either in the absence of 4b or in the presence of 4b proteins deficient in nuclear localization . MERS-CoV infection of Calu-3 cells in the absence of 4b protein only led to a two-fold increase in IFN-β levels , while no difference in IFN-β levels was observed in Huh-7 cells . This is in contrast to the high expression levels induced by the positive control poly ( I:C ) ( Fig 3C ) . 4b NLS mutants did not induce IFN-β production in Calu-3 cells , suggesting a limited contribution of 4b protein to the suppression of the IFN- β response and confirming the idea that multiple viral proteins participate in the antagonism of the innate immune response . To confirm that the pro-inflammatory response induced by MERS-CoV in the absence of nuclear 4b protein was mediated by NF-κB , Huh-7 cells were treated with the NF-κB inhibitor parthenolide [14] at the same time cells were infected . Parthenolide treatment inhibited the expression of NF-κB dependent pro-inflammatory cytokines induced by 4b mutant viruses without affecting virus production ( Fig 3A and 3B ) . This confirmed that NF-κB mediated the pro-inflammatory response observed in 4b-mutant virus infected cells . To determine the step in the NF-κB signaling pathway targeted by 4b protein , the amount of NF-κB p65 subunit was first analyzed by Western-blot in cells infected with 4b-mutant MERS-CoVs . No differences in p65 levels were observed among MERS-CoV mutants regardless of the NF-κB-mediated response ( Fig 3D ) . Since activation of NF-κB requires the degradation of the inhibitor IκBα , levels of IκBα were analyzed by Western-blot in cells infected with 4b-mutant MERS-CoVs ( A in S4 Fig ) . It was difficult to detect IκBα degradation in the context of viral infection , most likely indicating that activation of the pathway was not complete . Alternatively , activated NF-κB has been reported to rapidly induce expression of IκBα to restore NF-κB inhibition [15] . To examine the impact of viral infection on IκBα degradation , we further treated infected cells with TNF-α for different time periods at 14 hours post-infection , which is expected to induce a sustained reduction in IκBα levels . At this time point , significant levels of 4b had accumulated ( B in S4 Fig ) , providing a scenario where we could evaluate potential inhibition of IκBα degradation by 4b . Degradation of IκBα in infected cells treated with TNF-α was associated with induction of pro-inflammatory mRNAs , thus confirming that TNF-α activated the NF-κB pathway ( C in S4 Fig ) . Furthermore , 4b protein significantly reduced the expression levels of pro-inflammatory cytokines induced by TNF-α in infected cells as compared to infection with Δ4b virus , reinforcing the potential of 4b protein to repress the NF-κB response ( C in S4 Fig ) . However , no differences in IκBα levels were observed in infections with Δ4b and 4b NLS-mutant MERS-CoVs after TNF-α stimulation , suggesting that the 4b protein did not inhibit the degradation of IκBα ( Fig 3E and A in S4 Fig ) . Together , these results indicated that inhibition of the NF-κB response by the 4b protein did not occur at steps prior to the nuclear translocation of NF-κB . In a co-immunopecipitation ( Co-IP ) mass spectrometry screen for 4b interacting partners , we identified several peptides corresponding to karyopherin ( KPNA ) -α4 and KPNA-α3 ( also known as importin-α3 and importin-α4 , respectively ) that were immunoprecipitated with overexpressed 4b-FLAG protein but not the SARS-CoV control protein nsp15-FLAG ( S5 Fig ) . These interactions were confirmed by reciprocal co-immunoprecipitation experiments followed by Western blotting ( Fig 4A and 4B ) . Specifically , pull-down of 4b-FLAG from Huh-7 cells with a FLAG antibody isolated endogenous karyopherin-α4 and karyopherin-α3 , but not the control protein actin ( Fig 4A ) . Conversely , pull-down of overexpressed KPNA4-FLAG in Huh-7 cells with FLAG antibody isolated the co-expressed 4b-HA protein but not the 4a-HA protein ( Fig 4B ) . To confirm that 4b interacts with KPNA4 in the context of infection , KPNA4-FLAG or control GFP expressing plasmids were transfected into Huh-7 cells , prior to infection with wild-type virus . At 20 hpi , the cell lysates were immunoprecipitated with FLAG antibody followed by Western-blot analysis for either endogenous 4b or 4a proteins ( Fig 4C ) . Significant amounts of 4b but not 4a protein were isolated with KPNA4-FLAG but not GFP expressing cells , indicating a strong and specific interaction . The nuclear translocation of most cellular proteins is mediated by α-karyopherins ( or importin-αs ) , which directly bind the NLS within cargo proteins , and karyopherin-β ( or importin-β ) . The latter interacts with nucleoporins to promote nuclear import [16] . Since MERS-CoV 4b protein includes a bipartite NLS and localizes to the nucleus during infection , we tested whether the bipartite NLS sequences of 4b were required for KPNA4 binding . The basic amino acids in NLS-S1 or NLS-S2 were replaced with alanine in plasmids ( 4b-mNLS-S1–FLAG or 4b-mNLS-S2–FLAG ) and subsequently transfected into Huh-7 cells . Cell lysates were analyzed by co-immunoprecipitation with FLAG antibodies followed by Western-blot ( Fig 4D ) . Endogenous KPNA4 did not co-immunoprecipitate with 4b proteins mutated either in NLS-S1 or NLS-S2 , in contrast to wild-type 4b protein . This indicates that MERS-CoV 4b protein requires both NLS-S1 and NLS-S2 for efficient interaction with KPNA4 . It also strongly suggests that 4b interacts with KPNA4 as a mechanism to enter the nucleus . The α-karyopherins are divided into 3 sub-families . The importin-α1 subfamily , consisting of KPNA2 and KPNA7 , are general importers of cargo that bear a classic NLS; the importin-α2 subfamily , consisting of KPNA4 and KPNA3 , are known for their specificity for NF-κB and the regulator of chromosome condensation ( RCC ) -1; and finally the importin-α3 subfamily , consisting of KPNA1 , KPNA5 , and KPNA6 , is best known for binding phosphorylated STATs as well as several Influenza A virus proteins and Ebola virus VP24 [17] . In our screen , 4b protein only pulled down importin-α2 subfamily members ( S5 Fig ) , which are known to bind NF-κB . To directly test whether 4b preferentially interacted with importin-α2 subfamily proteins , we analyzed the ability of 4b protein to interact with α-karyopherins of each subfamily . KPNA1 ( importin-α3 subfamily ) , KPNA2 ( importin-α1 subfamily ) , and KPNA3 and KPNA4 ( importin-α2 subfamily ) were expressed as KPNA-FLAG fusion proteins in Huh-7 cells along with HA-tagged 4b protein . Cells were collected 48 hours post-transfection and cell lysates were immunoprecipitated with an anti-FLAG monoclonal antibody . The presence of KPNA and 4b protein was detected by Western-blot with anti-FLAG and anti-HA antibodies , respectively ( Fig 5A ) . Quantification of relative binding to each karyopherin ( Fig 5B ) showed that 4b protein preferentially bound to KPNA4 , and to a lesser extent , KPNA3 . Binding of 4b protein to karyopherin-α family members during infection was assayed in a similar fashion . Huh-7 cells were transfected with plasmids expressing KPNA-FLAG fusion proteins and infected 48 hours later with wild-type MERS-CoV at an MOI of 0 . 1 PFU/cell . Cells were collected at 20 hpi and cell lysates were immunoprecipitated with an anti-FLAG antibody and analyzed with anti-4b and anti-FLAG antibodies ( Fig 5C ) . In the context of infection , MERS-CoV 4b protein also preferentially bound to KPNA4 over all other karyopherins , including KPNA3 ( Fig 5C and 5D ) . However , 4b could at least minimally bind to other KPNAs , possibly explaining the ability of 4b-mNLS-S2 to partially access the nucleus . The favorable interaction between 4b and KPNA4 is consistent with the idea that 4b prevented the nuclear import of NF-κB by binding to KPNA4 [18] . Together , the above-mentioned results indicated that 4b protein specifically bound to KPNA4 both in the absence and in the presence of infection . Additionally , during MERS-CoV infection , the expression of pro-inflammatory cytokines was inhibited in the presence of wild-type nuclear 4b protein , in contrast to 4b cytoplasmic mutants . Therefore , we hypothesized that in infected cells , wild type 4b protein would interfere with NF-κB nuclear translocation . To test this possibility , we analyzed the nuclear-cytoplasmic distribution of 4b and the NF-κB p65 subunit at 24 hpi in Huh-7 and Calu-3 cells infected at an MOI of 0 . 1 PFU/cell either with wild-type MERS-CoV or with MERS-CoV recombinant viruses expressing cytoplasmic 4b-NLS mutant proteins . Western blot analysis of nuclear and cytoplasmic fractions and confocal microscopy analysis showed that during infection with the control viruses , MERS-CoV-wild-type or DUP , 4b protein mainly localized to the nucleus while most of the p65 protein was detected in the cytoplasm ( Fig 6A–6C and S6 Fig ) . In contrast , in infections either with MERS-CoV-Δ4b or MERS-CoV-DUP-mNLS-S1 , p65 was detected both in the nucleus and the cytoplasm . In MERS-CoV-DUP-mNLS-S2 infection , the dual distribution of 4b protein between the nucleus and the cytoplasm was associated with the presence of p65 in both cellular compartments . This observation suggested that partial access to the nucleus of 4b-mNLS-S2 was enough to allow a significant increase in p65 nuclear import as compared to that observed in the presence of nuclear wild-type 4b protein . Together , these results indicated that 4b protein with an intact NLS prevented nuclear import of NF-κB in Huh-7 cells and in the physiologically relevant Calu-3 cells , suggesting that this mechanism might also be critical in a natural lung infection . Since KPNA4 binds both p65 [18] and 4b ( Fig 4D ) proteins through their NLS sequences , it is conceivable that 4b and p65 compete for binding to the same site in KPNA4 . To address this question , co-IP experiments with FLAG antibodies in Huh-7 cells transfected with a KPNA4-FLAG plasmid and infected with MERS-CoVs expressing 4b-NLS mutant proteins were performed . These results confirmed that binding of KPNA4 to 4b protein was also NLS-dependent in the context of infection ( Fig 7 ) . Interestingly , in wild-type MERS-CoV-infected cells , immunoprecipitation of KPNA4 did not pull-down p65 while it did it in Δ4b and 4b NLS mutant virus infected cells . Together , these results strongly suggest that 4b interacts with KPNA4 in a NLS-dependent manner as a mechanism to gain entry into the nucleus , leading to the inhibition of p65 nuclear import by KPNA4 .
MERS-CoV strongly represses the innate immune response [1] , although the specific factors that contribute to this repression during infection are unknown . To date , all studies involving MERS-CoV proteins and their interactions with the innate immune response have been done with viral proteins expressed in isolation [10 , 12 , 13 , 19 , 20] . These studies can give a glimpse into the mechanisms used by the virus to block interferon and other innate immune factors . However , proteins expressed in isolation may not function as they do in the context of an infection , due to excessive protein production , improper localization , or the presence/absence of specific cellular or viral proteins [21] . Previous reports indicated that when expressed in isolation , 4b blocked IFN expression induced by Sendai virus ( SeV ) or by overexpression of cellular factors that function in the IFN-induction pathways [20] . Inhibition of IRF-3 or IRF-7-induced IFN production required the NLS sequences of 4b , whereas deletion of 4b NLS did not affect IFN expression induced by other factors such as MDA5 , MAVS , IKKε and TBK-1 . These results suggested that 4b could inhibit the induction of IFN-β in both the nucleus and the cytoplasm . Here , we show that MERS-CoV-Δ4b infection did not induce IFN in Huh-7 cells , while a limited IFN response was observed in Calu-3 cells , suggesting that additional mechanisms independent of 4b expression are used by MERS-CoV to inhibit IFN production [3 , 22 , 23] . CoV accessory proteins are not required for viral replication in cell culture although potentially contribute to pathogenesis in vivo , possibly by interfering with the innate immune response . In contrast to essential viral proteins , which are conserved across the CoV family , the accessory proteins show high diversity among CoV genera . In particular , the 4b protein is specific for MERS-CoV and does not show significant homology to other CoV accessory proteins , with the exception of two related bat CoVs HKU4 and HKU5 ( homology at the amino acid level lower than 30% ) and hedgehog CoVs included within the same lineage C of the genus Betacoronavirus [24–26] . No significant homology to other viral or mammalian proteins was found in protein databases . Here we demonstrate the role of the MERS-CoV accessory protein 4b during infection , using recombinant viruses that either completely delete 4b or alter its ability to traffic to the nucleus . We found that 4b was required to prevent a robust NF-κB dependent response during infection of both Huh-7 and Calu-3 cells ( S1 Fig and Fig 1C ) . This function required the NLS sequences of 4b , as opposed to a previous report [10] , in which overexpressed 4b proteins lacking the NLS were still able to inhibit IFN-β induction and to a lesser extent NF-κB signaling . Importantly , we found that the nuclear localization of 4b was associated with the cytoplasmic retention of NF-κB ( Fig 6 ) and consequently , with the repression of NF-κB-mediated expression of pro-inflammatory cytokines ( Fig 3A and 3C ) . We also found that both in isolation and during infection , 4b interacted with α-karyopherin proteins in an NLS-dependent manner . Interestingly , 4b had a strong preference for binding karyopherin-α4 , which is also known to translocate the NF-κB protein into the nucleus [18] . Interestingly , 4b binding to KPNA4 during infection inhibited its interaction with NF-κB-p65 subunit ( Fig 7 ) . Thereby we propose a model where 4b outcompetes NF-κB for KPNA4 binding and translocation into the nucleus ( Fig 8 ) . Alternatively , other functions of 4b or additional MERS-CoV proteins might also contribute to the NF-κB antagonism , as described for other coronaviruses , which include multiple proteins that antagonize the host IFN and NF-κB responses [22 , 27] . It has been reported that MERS-CoV 4b has phosphodiesterase ( PDE ) activity and antagonizes OAS-RNase L pathway by enzymatically degrading 2’ , 5’-oligoadenylate ( 2-5A ) , the activators of RNase L [28] . RNase L is an effector of the IFN-induced antiviral response and cleaves viral and host single-stranded RNA , leading to cell death and preventing viral replication and spread . Nuclear localization of 4b is exceptional among the other known viral PDEs , found in the cytoplasm . However , in the context of MHV infection , minor expression levels of MERS-CoV 4b in the cytoplasm are sufficient to prevent RNase L activation and , in fact , removal of the NLS did not improve RNase L antagonism . The relevance of 4b PDE activity to promote MERS-CoV replication requires further confirmation in appropriate cell types . Our results here and this previous study [28] suggest that 4b protein would provide MERS-CoV with multiple mechanisms for evasion of innate immunity , such as the repression of the NF-κB-mediated response and the inhibition of RNase L activation . Although CoVs are cytoplasmic viruses , they do interact with the nucleus in diverse ways including relocation to the nucleus of specific viral proteins and the disruption of host nuclear transport [29] . This interaction can be beneficial either for viral replication or for the antagonism of host antiviral response . Some CoV genus-specific proteins , such as MERS-CoV 4b and SARS-CoV 3b , localize to the nucleus [10 , 12 , 20 , 30] , although their precise effect on host nuclear processes is still unknown . The nucleocapsid ( N ) and nsp1 proteins have also been detected in the nucleus , although this is not a conserved property in all CoVs . N protein nuclear localization has been related to cell cycle arrest and recruitment to the cytoplasm of nuclear proteins involved in viral RNA synthesis [29 , 31] . The presence in the nucleus of MERS-CoV nsp1 has been involved in the selective inhibition of nuclear-transcribed mRNA expression [32] , while porcine epidemic diarrhea virus ( PEDV ) nsp1 mediates the degradation of CREB-binding protein in the nucleus , thus suppressing IFN I production [33] . CoV nsp1 and SARS-CoV ORF 3b proteins are sufficiently small to diffuse passively through the nuclear pore complex [30 , 32] . In contrast , nuclear import of larger proteins such as N and MERS-CoV 4b require host transporters and is mediated by NLS sequences [34] . Interference with the nuclear import machinery is a common viral strategy to evade the innate immune response . Viruses have evolved different mechanisms to subvert nucleo-cytoplasmic transport by targeting soluble transport receptors such as karyopherins as well as components of the nuclear pore complexes ( NPC ) . Viruses with a nuclear stage , such as DNA viruses and some RNA viruses , promote nuclear import of viral genome and proteins by altering the NPC [35] , which leads to a general inhibition of host nucleo-cytoplasmic trafficking . Herpes simplex virus ICP27 protein interacts with the nucleoporin Nup62 , an structural component of the NPC , leading to the inhibition of nucleo-cytoplasmic transport pathways mediated by transportin or importins [36] . Similarly , the Epstein-Barr virus ( EBV ) BGLF4 kinase interacts with the nucleoporins Nup62 and Nup153 inhibiting canonical NLS-mediated nuclear import while promoting the nuclear import of non-NLS-containing EBV lytic proteins [37] . Cytoplasmic viruses also share similar strategies to inhibit nuclear import of signals required for host immune defenses . SARS-CoV ORF6 retains KPNA2 and karyopherin-β in the cytoplasm , thus preventing STAT1 nuclear import and antagonizing IFN signaling pathways [38–40] . A similar mechanism of inhibition has also been described for Ebola virus VP24 [41] . Enterovirus 71 ( EV71 ) infection suppresses IFN responses by inhibiting STAT1/2 nuclear transport through the induction of caspase-3-dependent degradation of KPNA1 [42] . Poliovirus and rhinovirus infection induce proteolysis of the NPC components Nup153 and p62 , leading to a general inhibition of nuclear import [43 , 44] . The association of viral proteins with the host nuclear import pathway confirms its relevance in virus-host interactions , regulating innate immunity and cellular responses to infection . NF-κB is a critical factor for the up regulation of innate immunity and inflammation in response to viral infections [4] . Therefore , viruses in diverse families have evolved a variety of mechanisms to target nuclear transport of the NF-κB complex and thereby inhibit its activation . The nucleocapsid protein of Hantaan virus inhibits NF-κB activation by limiting p65 nuclear translocation , as shown in overexpression experiments [45] , presumably by its interaction with importin-α through a putative NLS . Japanese encephalitis virus NS5 competitively inhibits NF-κB nuclear translocation by binding to KPNA2 , 3 and 4 , presumably through two NLSs within NS5 , which surprisingly does not accumulate in the nucleus [46] . Alternatively , viral proteins such as Herpes simplex virus 1 UL24 inhibit the NF-κB pathway by binding p65 and p50 subunits and preventing their nuclear translocation [47] . In CoVs , there is also evidence supporting the relevance of the NF-κB pathway in pathogenesis . Increased NF-κB activation is a determinant of the exacerbated lung inflammatory pathology caused by SARS-CoV infection [6] . Consequently , inhibition of NF-κB signaling significantly reduced pulmonary inflammation and increased survival of SARS-CoV-infected mice [5] . Dual regulation of NF-κB signaling during HCoV-229E infection has been described , consisting in the simultaneous induction of the NF-κB pathway , which has a pro-viral effect , and the restriction of the NF-κB response to limit transcription of pro-inflammatory and other potentially antiviral host cell genes [48] . Here , our results with MERS-CoV 4b protein are compatible with this concept of repression of the NF-κB-mediated pro-inflammatory response during CoV infection . Viral antagonists of the innate immune response contribute to the pathogenic potential of coronaviruses in vivo [27 , 49] . In particular , inhibition of nuclear import by SARS-CoV ORF6 has been shown to mediate pleiotropic alterations in host gene expression during infection [50] that enhance viral pathogenesis in vivo [51] . In addition , MERS-CoV 4b protein expressed in mice by chimeric MHV viruses acted as an RNase L antagonist contributing to pathogenesis in vivo [28] . From this study and from the results reported here , it might be speculated that deletion of MERS-CoV 4b protein would enhance the host antiviral response leading to virus attenuation in vivo . However , the possibility that the increased NF-κB-response induced by 4b deletion could result in lung inflammatory pathology [5] cannot be excluded until in vivo studies are performed . Humanized mice ( hDPP4-KI ) infected with mouse-adapted MERS-CoV develop lethal disease and will be useful for determining the role of the 4b protein in pathogenesis . Human and mouse KPNA4 proteins share 99% amino acid identity , including binding sites of NLSs present in NF-κB p65 and p50 subunits [18] . Consequently , the interaction between MERS-CoV 4b and human KPNA4 reported here is expected to function similarly in mouse cells , inhibiting NF-κB nuclear translocation in the MERS-CoV mouse model . In one study , no differences in lethality were observed in human DPP4 knockin mice infected with mouse adapted MERS-CoV isolates differing in the presence or absence of ORF3-4a-4b accessory genes in addition to other non-synonymous mutations [52] . In a second study , two adapted viral clones with large deletions in ORF4b , in addition to other missense mutations throughout the genome , caused increased lung inflammation and mouse weight losses [53] . However , none of these studies used congenic strains of virus solely differing in 4b expression . Infection with such strains will provide insight into the role of 4b in the context of severe infection .
Baby hamster kidney cells ( BHK-21 ) were obtained from American Type Culture Collection ( ATCC CCL-10 ) . Human liver-derived Huh-7 cells were kindly provided by R . Bartenschlager ( University of Heidelberg , Germany ) . Cells were grown in BioWhittaker ( Lonza ) or Dulbecco’s modified Eagle medium ( DMEM ) with 2% glutamine , 1% non-essential amino acids ( Sigma ) and 5% ( BHK-21 ) or 8% ( Huh-7 ) fetal bovine serum ( FBS ) ( HyClone , ThermoFisher ) . The bronchial epithelial cell line Calu-3 2B4 [54] was kindly provided by CT Tseng ( University of Texas Medical Branch , USA ) . Calu-3 cells were grown in BioWhittaker ( Lonza ) 1 g/L glucose medium with 2% glutamine , 1% non-essential amino acids ( Sigma ) and 15% fetal bovine serum ( FBS ) ( HyClone , ThermoFisher ) . Virus titers were determined on Huh-7 cells following standard procedures and using closed flasks or plates sealed in plastic bags . For plaque assays , infected cells were overlaid with DMEM containing 0 . 6% low-melting agarose and 2% FBS , and at 72 hpi , cells were fixed with 10% formaldehyde and stained with 0 . 1% crystal violet . All work with MERS-CoV infectious viruses was performed in biosafety level 3 facilities at CNB-CSIC or the University of Iowa according to the guidelines set forth by each institution . E . coli DH10B ( Gibco/BRL ) cells were transformed by electroporation using a MicroPulser unit ( Bio-Rad ) according to the manufacturer’s instructions and grown on antibiotic-selective LB medium . MERS-CoV ORF4a-HA and ORF4b-HA nucleotide sequences were synthesized and cloned into pUC57 ( GenScript ) . They were then PCR amplified , restriction digested and ligated into the pLKO plasmid . A pLKO ORF4b-3XFLAG plasmid was made by creating a PCR product replacing the HA tag with a 3XFLAG tag . This PCR product was restriction digested and ligated back into the pLKO plasmid . The resulting constructs were confirmed by restriction digest , PCR , and direct sequencing . The sMacro and GFP pcDNA3 plasmids were previously described [55] . The NSP15-3XFLAG plasmid was a generous gift of Michael Buchmeier ( University of California-Irvine ) , and the KPNA-FLAG plasmids were a generous gift of Megan Shaw ( Mount Sinai Medical Center ) . Cells were transfected using Polyjet ( Amgen ) or Lipofectamine 2000 ( Fisher Scientific ) as per the manufacturer’s instructions . Genes 4a and 4b were deleted from the MERS-CoV infectious cDNA clone by PCR-directed mutagenesis . The plasmid pBAC-MERSFL was used as the template for amplification of overlapping PCR fragments with oligonucleotides MERS-del4a-26022-VS ( 5′-CTATGGATTACGTGTCTCTGCTTACAGCTATCCTTTGCTGGTTATACTGAATCT-3′ ) and MERS-del4a-26022-RS ( 5’-AGATTCAGTATAACCAGCAAAGGATAGCTGTAAGCAGAGACACGTAATCCATAG -3’ ) for 4a deletion or MERS-del4b-26811-VS ( 5’-GCGAGGAAGAGGAGCCATTCTCCAACTAAGTAATAAACAAATTGTTCATTCTTATCCC-3’ ) and MERS-del4b-26811-RS ( 5’-GGGATAAGAATGAACAATTTGTTTATTACTTAGTTGGAGAATGGCTCCTCTTCCTCGC -3’ ) for 4b deletion . The final PCR product was amplified with external oligonucleotides SA25412-VS ( 5’-CTGCACTGGTTGTGGCAC-3’ ) and MERS-26970-Nhe RS ( 5′ GCTAAAGCAGCTACATAGCCGCTAGCAGGAATG 3′; NheI is underlined ) , digested with PacI and NheI , and cloned into the same restriction sites of pBAC-MERS-3′ [8] . The PacI-RsrII digestion products from plasmids pBAC-MERS-3′-Δ4a and pBAC-MERS-3′-Δ4b were cloned into the same sites of pBAC-MERSFL ( Fig 2A ) . pBAC-MERSFL-Δ4a included a deletion from nucleotides 25 , 874 to 25 , 991 of the MERS-CoV genome , while the deletion in pBAC-MERSFL-Δ4b extended from nucleotides 26 , 183 to 26 , 781 . For all constructs , sequences were checked after cloning by sequencing . Recombinant BACs were isolated and purified using a large-construct kit ( Qiagen ) , following the manufacturer’s specifications . The complete sequence of 4a and 4b genes in the MERS-CoV infectious cDNA [8] was replaced with chemically synthesized fragments ( GeneArt , Invitrogen ) introduced into PacI-NheI sites ( nt 25841 to 26945 ) , including either the duplication of 4a/4b region alone or in combination with NLS mutations . The resulting pBAC-MERS-CoV-DUP contained a duplication of the last 189 nt of 4a gene , including the 4a/4b overlapping sequence and the preceding 4a 101 nt . Furthermore , three silent point mutations were introduced into 4a gene ( T26094C , T26109C and A26112C ) to prevent premature translation initiations of 4b from the sgmRNA 4ab . pBAC-MERS-DUP-mNLS-S1 included the 4b NLS-S1 ( aa 22–24 ) mutated to alanine ( RKR→ AAA ) in a sequence context identical to the control pBAC-MERS-CoV-DUP , while pBAC-MERS-DUP-mNLS-S2 included alanine mutations in 4b NLS-S2 ( aa 36–38 ) and two intermediate Lys residues , ( aa 30–31 ) ( KRR → AAA and KK → AA , respectively ) . The genetic integrity of the cDNAs was verified by restriction analysis and sequencing . To recover infectious viruses , BHK-21 cells were transfected with the infectious cDNA clones using Lipofectamine 2000 ( Invitrogen ) according to the manufacturer’s specifications . At 6 h post-transfection , cells were trypsinized , plated over a confluent monolayer of Huh-7 cells , and incubated at 37°C for 72 h . The cell supernatants were harvested and passaged once on fresh Huh-7 cells and then virus titers were determined . The genetic stability of rescued viruses was verified by sequencing the genomic region comprising the 4a and 4b genes using RT-PCR . Confluent monolayers of Huh-7 or Calu-3 cells were infected at a MOI of 0 . 1 and 0 . 001 PFU/cell . Cell supernatants were collected at 24 , 48 and 72 hpi and virus titers were determined as described above . Total intracellular RNA was extracted from transfected or infected cells with the RNeasy miniprep kit ( Qiagen ) according to the manufacturer’s specifications . Total cDNA was synthesized with random hexamers from 100 ng of total RNA using a High-Capacity cDNA reverse transcription kit ( Invitrogen ) . For qPCR assays , specific TaqMan assays ( Apply Biosystems ) for TNF-α ( Hs99999043_m1 ) ; IL-6 ( Hs00985641_m1 ) ; IL-8 ( Hs99999034_m1 ) ; IFN-β ( Hs02621180_s1 ) and hydroxymethylbilane synthase ( HMBS , Hs00609297_m1 , as an endogenous control ) were used . MERS-CoV genomic RNA ( gRNA ) synthesis was analyzed using a custom TaqMan assay as described [8] . Data acquisition , analysis and normalization were performed as described [8] . Two micrograms of polyinosinic-polycytidylic acid [Poly ( I:C ) ] were used for reverse transfection of cells with 12 μg of Lipofectamine 2000 ( Invitrogen ) , as recommended by the manufacturer . The relative quantifications were performed using the 2-ΔΔCt method [56] . Rabbit polyclonal antisera ( pAb ) specific for MERS-CoV 4a and 4b proteins were ordered from BioGenes ( Germany ) . In brief , rabbits were immunized with synthetic peptides PDGCSLYLRHSSLFAQSEEEEPFSN ( 25 C-terminal residues of protein 4a ) and SIRSSNQGNKQIVHSYPILHHPGF ( 24 C-terminal residues of protein 4b , as described ) [10] , according to standard protocols . The specificity of the antisera was assessed by Western-blot and immunofluorescence microscopy of MERS-CoV-infected Huh-7 cells , with pre-immune sera and mock-infected cells as negative controls . At either 48 hours post-transfection or 20 hours post-infection , Huh-7 cells were collected and pelleted by low speed centrifugation prior to lysis . Cell pellets were lysed with IP buffer ( 0 . 5–2% NP-40 , 300 mM NaCl , 5% glycerol , and 50 mM Tris pH 8 . 0 ) , with protease/phosphatase inhibitor cocktails ( Roche ) , PMSF , and a universal nuclease ( ThermoFisher Scientific ) for 1 hour at 4ºC . Next , nuclei were pelleted by centrifugation ( 16 , 000 x g for 15 min at 4ºC ) . One aliquot of cell lysate was saved as the input control and boiled in SDS sample buffer in the presence of sample reducing agent . Supernatants were then applied to FLAG antibody conjugated magnetic beads ( Sigma Aldrich #M8823 ) and were mixed for 2 hours-overnight at 4ºC . Beads were washed with TBS-Tween before elution with FLAG peptide ( 1 μg/ml ) for 1 hour and then boiled in SDS sample buffer in the presence of sample reducing agent . For mass spectrometry analysis , proteins precipitated by anti-FLAG antibody were resolved on a NuPAGE 4–12% gradient gel ( Novex ) and subsequently stained using a Silver Stain kit ( Pierce ) according to the manufacturer’s instructions . Protein bands unique to the 4b-3XFLAG transfected sample were extracted . In addition , gel bands from the NSP15-3XFLAG sample with migrating positions corresponding to those of 4b-3XFLAG specific bands were also extracted as negative controls . Extracted gel samples were submitted to the Keck Mass Spectrometry and Proteomics Facility ( School of Medicine , Yale University ) for liquid chromatography ( LC ) -mass spectrometry analysis for protein identification . Whole protein extracts were prepared from Huh-7 or Calu-3 cells in loading buffer [0 . 1 M Tris-HCl , pH 6 . 8; 20% glycerol; 4% w/v sodium dodecyl-sulfate ( SDS ) ; 0 . 2% bromophenol blue and 0 . 05% β-mercaptoethanol] . Nuclear and cytoplasmic fractions of Huh-7 or Calu-3 cells were prepared using a protocol adapted from [57] . Briefly , cell monolayers were washed three times with PBS , collected and resuspended in lysis buffer [0 . 1% IGEPAL CA-630 ( Sigma-Aldrich ) in PBS] . Extracts were centrifuged at 1000 x g for 1 min to recover the cytoplasmic fraction in the supernatant and the nuclei in the pellet . The cytosolic fraction was removed and mixed with Laemmli sample buffer 1:1 , while the pellet was resuspended in PBS with 0 . 1% IGEPAL and centrifuged at 1000 x g for 1 min . The supernatant was discarded and the pellet , designated as nuclear fraction , was treated with DNase ( Roche ) at 37ºC for 1 hour and then mixed 1:1 with Laemmli sample buffer . Cell extracts or IP elutions were lysed in sample buffer containing SDS , protease and phosphatase inhibitors ( Roche ) , β-mercaptoethanol , and a universal nuclease ( Fisher Scientific ) . Proteins were resolved on an SDS polyacrylamide gel , transferred to a polyvinylidene difluoride ( PVDF ) or nitrocellulose membrane , hybridized with a primary antibody , reacted with either HRP or infrared ( IR ) dye-conjugated secondary antibody , visualized using chemiluminescent substrate ( Bio-Rad ) or a Li-COR Odyssey Imager ( Li-COR ) , and analyzed using Image Studio software . Primary antibodies used for immunoblotting included anti-FLAG monoclonal antibody ( Sigma-Aldrich #F1804 ) ; anti-HA monoclonal antibody ( Biolegend #901513 ) ; anti-MERS-CoV 4b polyclonal antibody and anti-MERS-CoV 4a polyclonal antibodies ( this study ) ; anti-KPNA3 and anti-KPNA4 polyclonal antibodies ( Thermo Scientific #PA5-18238/PA5-21033 ) ; anti-MERS-CoV N protein [8]; anti-histone H3 ( Cell Signaling #9715 ) ; anti-p65 ( Cell Signaling #6956 ) ; anti-GAPDH mAb ( Cell Signaling #5174 ) . Secondary antibodies used included horseradish peroxidase-conjugated anti-rabbit or anti-mouse ( Sigma #A0545/A0168 ) antibodies; or IR conjugated anti-rabbit or anti-mouse ( Li-COR , #926-68071/926-32210 ) antibodies . Huh-7 or Calu-3 cells were grown to 95% confluence on sterile glass coverslips and infected with MERS-CoV at a MOI of 0 . 1 PFU/cell . At 24 hpi , cells were fixed and the virus was fully inactivated by incubating with 4% paraformaldehyde in PBS at RT for 45 min . Cells were permeabilized with methanol at −20°C for 10 min and blocked with 10% FBS in PBS at RT for 1 h . MERS-CoV 4a and 4b proteins were detected with the specific pAbs described above . dsRNA was detected with J2 mAb ( SCICONS ) and NF-κB was detected with anti-p65 ( Cell Signaling #6956 ) diluted in PBS 5% FBS . Coverslips were washed 4 times with PBS and incubated with secondary antibodies conjugated to Alexa Fluor 488 or 594 ( Invitrogen ) diluted 1:500 in 5% FBS in PBS at RT for 45 min . Nuclei were stained using DAPI ( 4′ , 6-diamidino-2-phenylindole ) diluted 1:200 in PBS . Finally , coverslips were mounted in ProLong Gold antifade reagent ( Invitrogen ) and analyzed on a Leica SP8 confocal microscope . Images were acquired with the same instrument settings and analyzed with Leica software . | Middle East respiratory syndrome coronavirus ( MERS-CoV ) is a highly pathogenic human CoV that continues to cause lethal human infections , primarily in the Middle East . Virus accessory genes are potential contributors to pathology , possibly by interfering with the innate immune response . However , understanding their interactions with host proteins in the context of infection is rudimentary . Here , we provide evidence that the MERS-CoV accessory protein 4b functioned , at least in part , to prevent a robust NF-κB dependent response during infection . This effect depended on the nuclear localization of 4b , which was associated with the cytoplasmic retention of NF-κB . We show that 4b interacted with α-karyopherin proteins ( importins ) involved in the nuclear import of NF-κB , inhibiting the binding of α-karyopherin to NF-κB-p65 subunit . We propose that 4b contributes to the evasion of the innate immune response by binding α-karyopherin proteins , leading to the inhibition of NF-κB nuclear transport . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"nuclear",
"import",
"innate",
"immune",
"system",
"medicine",
"and",
"health",
"sciences",
"coronaviruses",
"immune",
"physiology",
"cytokines",
"pathology",
"and",
"laboratory",
"medicine",
"respiratory",
"infections",
"pathogens",
"cell",
"processes",
"immunology",
"microbiology",
"pulmonology",
"viruses",
"developmental",
"biology",
"rna",
"viruses",
"immunoprecipitation",
"molecular",
"development",
"antibodies",
"cellular",
"structures",
"and",
"organelles",
"research",
"and",
"analysis",
"methods",
"immune",
"system",
"proteins",
"proteins",
"medical",
"microbiology",
"microbial",
"pathogens",
"immune",
"response",
"precipitation",
"techniques",
"immune",
"system",
"cytoplasm",
"biochemistry",
"cell",
"biology",
"viral",
"pathogens",
"physiology",
"biology",
"and",
"life",
"sciences",
"organisms"
] | 2018 | MERS-CoV 4b protein interferes with the NF-κB-dependent innate immune response during infection |
Non-additive interactions between genomes have important implications , not only for practical applications such as breeding , but also for understanding evolution . In extreme cases , genes from different genomic backgrounds may be incompatible and compromise normal development or physiology . Of particular interest are non-additive interactions of alleles at the same locus . For example , overdominant behavior of alleles , with respect to plant fitness , has been proposed as an important component of hybrid vigor , while underdominance may lead to reproductive isolation . Despite their importance , only a few cases of genetic over- or underdominance affecting plant growth or fitness are understood at the level of individual genes . Moreover , the relationship between biochemical and fitness effects may be complex: genetic overdominance , that is , increased or novel activity of a gene may lead to evolutionary underdominance expressed as hybrid weakness . Here , we describe a non-additive interaction between alleles at the Arabidopsis thaliana OAK ( OUTGROWTH-ASSOCIATED PROTEIN KINASE ) gene . OAK alleles from two different accessions interact in F1 hybrids to cause a variety of aberrant growth phenotypes that depend on a recently acquired promoter with a novel expression pattern . The OAK gene , which is located in a highly variable tandem array encoding closely related receptor-like kinases , is found in one third of A . thaliana accessions , but not in the reference accession Col-0 . Besides recruitment of exons from nearby genes as promoter sequences , key events in OAK evolution include gene duplication and divergence of a potential ligand-binding domain . OAK kinase activity is required for the aberrant phenotypes , indicating it is not recognition of an aberrant protein , but rather a true gain of function , or overdominance for gene activity , that leads to this underdominance for fitness . Our work provides insights into how tandem arrays , which are particularly prone to frequent , complex rearrangements , can produce genetic novelty .
Both evolutionary biologists and breeders have long been interested in non-additive interactions among alleles at the same locus . For example , explanations for heterosis or hybrid vigor , a staple of modern agriculture , share many conceptual formalities with models proposed by Bateson , Dobzhansky and Muller to explain how negative heterosis could result from two or more genes that accumulate different changes in separate lineages . The associated phenotypes of hybrid weakness , sterility or lethality in turn may ultimately lead to reproductive isolation and hence speciation ( [1]–[3] , reviewed in [4] , [5] ) . Hybrid incompatibilities form a continuum from the grey zone of developmental abnormalities through the clearer phenotype of F1 sterility to the severest form , lethality , and it is important to understand the genetic and molecular causes for the entire spectrum of incompatibilities . F1 incompatibilities have been found in as many as 2% of Arabidopsis thaliana intra-specific hybrids [6] . Several similar cases in A . thaliana and other species involve interactions between alleles of disease resistance genes with other loci in the genome , which cause an autoimmune syndrome known as hybrid necrosis [6]–[8] . That hybrid necrosis is such a relatively common phenomenon is easily explained , since genes involved in plant defense are highly variable between different individuals of the same species [9] , [10] , and thus make a perfect substrate for causing problems when different genomes are combined . Moreover , several important classes of defense genes , including those encoding nucleotide binding-leucine rich repeat ( NB-LRR ) proteins and receptor-like kinases ( RLKs ) , commonly occur in tandem arrays , and new alleles are easily created through gene duplication , illegitimate recombination and gene conversion [11]–[19] . In addition to inappropriate activation of the immune system or sterility , aberrant development is often observed in incompatible plant hybrids [20] , [21] . Both Triticum and Nicotiana interspecific hybrids frequently suffer from tumor-like tissue proliferation [22] , [23] . In Nicotiana hybrids , wounding and physiological stresses enhance tumor formation , and tumors may differentiate into recognizable tissues [24] . Genetically-induced tumors have also been described in hybrids of Brassica , Datura , Solanum and Lilium [24] . Developmental abnormalities in intra-specific moss hybrids have recently been linked to putatively structurally divergent regions [25] , similar to the association of hybrid necrosis with structurally diverse disease resistance loci . While the known cases of F1 hybrid incompatibility are mostly caused by interaction between alleles at unlinked loci , of particular interest are situations of heterozygous advantage ( overdominance ) or disadvantage ( underdominance ) due to interaction of divergent alleles at the same locus . Overdominance has been advanced as an important contributor to hybrid vigor , or heterosis [26]–[28] . Conversely , underdominance may underlie hybrid weakness , sterility or lethality , and thus contribute to speciation [20] , [21] , [29] . It should be noted that cases of heterozygous disadvantage are underdominant with respect to fitness but can be overdominant in the genetic sense: a plant may become less fit due to increased activity of the gene ( s ) involved . Although evidence for both single-gene over- and underdominance is easily found in whole-genome expression studies ( e . g . [28] ) , few cases with phenotypic consequences are understood at the molecular level . Schwartz and Laughner [30] reported four decades ago an example in maize , where two partially compromised forms of alcohol dehydrogenase can form a fully active homodimer; a similar case has been described for complementing alleles at the ARF GTPase-encoding GNOM locus of Arabidopsis thaliana [31] . In tomato , a heterozygote for a loss-of-function allele of the SFT gene has increased yield [32] . Finally , a particularly revealing study comes from rice , where sterility ensues when two divergent alleles at the S5 locus are combined [33] . Since this is not observed when either allele is heterozygous with a third , presumably non-functional allele , one can infer that the combination of the two S5 alleles results in gain-of-function activity of the encoded aspartate protease . The S5 interactions thus provide an example of the complex relationship between biochemical and fitness effects , as the underdominant fitness effects are not simply a consequence of reduced gene activity . It also provides a counterpoint to the SFT case , where reduced gene activity has overdominant fitness effects [32] . Here , we report on an intraspecific A . thaliana F1 hybrid , where heterozygosity at a single locus causes a pleiotropic syndrome that includes smaller stature and reduced seed set as well as ectopic outgrowths on leaf petioles . The causal receptor-like kinase ( RLK ) gene , OUTGROWTH-ASSOCIATED PROTEIN KINASE ( OAK ) , is found in a structurally hypervariable tandem cluster of related RLK genes . During duplication of the ancestral RLK gene , coding sequences were recruited to form a promoter with a new expression domain . Divergence in the extracellular domain of the protein led to evolution of alleles that now interact in the Bla-1/Sha hybrid to produce phenotypes not seen in the parents , making this a case of underdominance for fitness caused by overdominance for gene expression .
The aberrant phenotype of Blanes-1 ( Bla-1 ) /Shahdara ( Sha ) F1 hybrids was identified in a survey of more than 1 , 300 crosses among over 300 A . thaliana accessions from the world-wide range of the species [6] . Bla-1/Sha F1 plants had a range of phenotypes that were not normally seen in inbred accessions , including the Bla-1 and Sha parents , or in other F1 hybrids: outgrowths on the adaxial surface of the petioles , leaf twisting , leaf lesions , and loss of apical dominance reflected by precocious and increased release of side shoots ( Figure 1a–1c ) . These phenotypes were observed regardless of the direction of the cross . Raising plants in long days at 23°C instead of 16°C restored apical dominance and largely suppressed leaf twisting and lesioning . This partial suppression of the hybrid phenotype at higher temperatures is similar to the suppression of necrosis seen in the Uk-1/Uk-3 and other hybrids with autoimmune defects [6] . Because the ectopic outgrowth phenotype was particularly striking and reliably observed in all F1 plants , we decided to investigate it in detail . The same phenotype with little variation was seen in approximately 50% of all F2 progeny , compatible with a single-gene , heterozygous genetic basis . The outgrowth phenotype segregated independently of the lesioning in the F2 and subsequent generations . Outgrowths were occasionally noted in the Bla-1 parent , but with incomplete penetrance that varied greatly between experiments ( Table S1 ) . Onset of outgrowth formation in Bla-1 , when it occurred , was much later than in the F1 hybrids . Crosses of each parental line to the reference accession Col-0 did not produce any progeny with outgrowths , but they were , as expected , seen in about one quarter of progeny after Col-0/Bla-1 and Sha/Col-0 F1 hybrids were crossed to each other . Analysis of transverse sections revealed that outgrowths originated from proliferating parenchyma and/or epidermal cells on the adaxial surface of the petiole ( Figure 1d–1f ) . The vascular system of the petioles appeared normal . Because of their determinate nature , we concluded that the outgrowths did not constitute undifferentiated callus . We also asked whether the gene ( s ) causing the hybrid phenotypes of outgrowth and lesioning might affect overall plant performance . In a segregating F2 population of five-week old plants , we found that outgrowths alone were correlated with a 29% reduction in rosette weight , while lesioning or lesioning plus outgrowths reduced growth by over 50% ( Table S2; 2-way ANOVA outgrowths p = 0 . 0003 , lesioning p<0 . 0001 ) . In addition , we assessed seed set as a proxy for lifetime fitness . Due to confounding factors such as differential flowering times in Sha and Bla-1 , we measured seed set after the incompatibility was reconstituted in the Col-0 reference background ( see below for further details ) . Seed set was reduced by 90% in F1 hybrids that were phenotypically comparable to the natural hybrids ( two-tailed , unequal variance t-test: p<<0 . 001; Figure S1 ) . In two other independent crosses that resulted in a more severe incompatibility phenotype , all the hybrids died within two months , and thus did not produce any seeds at all . This indicates that the Bla-1/Sha OAK incompatibility greatly reduces lifetime fitness . Because wounding and physiological stresses enhance the formation of tumors in Nicotiana , where these may differentiate into recognizable tissues [24] , we examined the effects of wounding , by pricking the petioles of Bla-1/Sha F1 plants with a fine needle . Outgrowth formation was not enhanced , but we found that increased humidity suppressed outgrowth formation ( Figure S2 ) . This is reminiscent of the suppression of constitutive activation of disease resistance in the ssi4 mutant by high humidity [34] . Compared to normal tissue , induction of callus from Nicotiana hybrid tumors requires less auxin [35] . Some A . thaliana tumor forming lines also produce callus tissue that can continue to proliferate on hormone-free media [36] . To test auxin response in our system , transverse sections of leaf and petiole tissue were induced to form callus . Although the Bla-1 parent had a relatively higher auxin requirement for callus formation , there was no difference between the Sha parent and the Bla-1/Sha hybrids ( Figure S3 ) . Thus , the outgrowths are probably genetically distinct from the Nicotiana tumors . Microarray analysis with triplicate Affymetrix ATH1 arrays using RNA extracted from three-week-old aerial tissue identified 356 genes differentially expressed in the hybrids compared to the parents . There was no significant up- or down-regulation of any particular known pathways or reactions based on the SkyPainter tool [37] , but several , often overlapping , Gene Ontology ( GO ) categories were enriched among the differentially expressed genes , most notably several related to pathogen response ( Table S3; [38] ) . Whether this reflects a link to disease resistance remains unclear , since some well-known markers for pathogen response , such as PR1 or the defensin gene PDF1 . 2 ( b ) , were down-regulated in the hybrids ( Tables S4 and S5 ) . In any case , as with the morphological phenotype , there was no overwhelming connection to the hybrid necrosis syndrome as seen in many other incompatible A . thaliana F1 hybrids [21] . Using F2 and F3 progeny , we mapped the outgrowth phenotype to a single genomic region on chromosome 5 containing 17 genes in the reference accession Col-0 ( At5g59560 to At5g59700; Figure S4 ) . A tandem array of four genes that encode a distinct clade of closely related receptor-like kinases ( RLKs; At5g59650 to At5g59680 ) [17] were of particular interest , because RLKs are one of the most variable gene families in the A . thaliana genome [9] . We recovered the genomic regions from At5g59616 ( encoding a protein kinase-related protein ) to At5g59690 ( histone H4 ) by long-range PCR from Bla-1 and Sha , and found the RLK cluster to be highly variable ( Figure 2a ) . In Col-0 only , there are two transposons and a pseudogene upstream of the RLK genes . In Sha , the first RLK gene in the cluster , At5g59650 , is missing and the upstream gene At5g59616 is only partially present . In both Bla-1 and Sha , a 150 bp remnant of the second RLK gene , At5g59660 , indicates that a deletion likely occurred in the Bla-1/Sha lineage . Also in both Bla-1 and Sha , the third RLK gene of the cluster , At5g59670 , has been duplicated to give rise to At5g59670a and At5g59670b ( Table S6 ) . In addition to Bla-1 and Sha , the At5g59670 duplication was detected by PCR analysis of the OAK promoter in 36 of 87 diverse A . thaliana accessions ( Table S7 ) , while a Col-0 like promoter was found in 45 accessions . Assays for both promoter types were positive in two accessions , indicating either illegitimate recombination or a different duplication event . The PCR assays failed in the remaining four accessions . Reconstruction of the ancestral state of the tandem array , by comparison with the close relative A . lyrata [39] , suggested the presence of three tandem RLK genes in the last common ancestor of A . thaliana and A . lyrata . The central gene was duplicated in the A . thaliana lineage to produce At5g59660 and At5g59670 , whereas in A . lyrata , there have been subsequent duplications of the two flanking RLK genes , resulting in a cluster with six genes . Given the presence of a remnant of At5g59660 in Bla-1 and Sha and that the Col-0-like At5g59670 is found in over half the accessions tested , the ancestral state of this cluster in A . thaliana is likely to have been a cluster of four RLK genes as found in Col-0 ( Figure 2b ) . To determine whether any of the RLK genes contribute to the outgrowth phenotype , a genomic copy of each gene from Bla-1 and Sha was individually introduced into the Bla-1 , Sha and Col-0 backgrounds . Only plants transformed with At5g59670b from Bla-1 or Sha developed outgrowths ( Figure 3a ) . Unexpectedly , while At5g59670b from Bla-1 induced outgrowths most effectively in Sha , and At5g59670b from Sha in Bla-1 , outgrowths were also seen , albeit at lower frequency , upon transformation of either gene into the recurrent parent or into Col-0 . This suggests a dosage effect , perhaps due to elimination of negative regulatory elements or epigenetic marks in the transgene that normally suppress expression of the endogenous locus , such that the transgenic proteins are present at an elevated level compared to native OAKSha or OAKBla-1 . This is supported by some transgenic lines in which we saw a 3∶1 ratio of normal to affected plants in the T2 generation , such that a hemizygous state gives a wild-type phenotype while homozygosity for the transgene leads to a Bla-1/Sha-like phenotype . A similar increase in incompatibility severity after transgenic reconstitution was also observed for DM1 in the case of Uk-1/Uk-3 [6] . To determine whether the RLKs were not only sufficient , but also necessary for the outgrowths , artificial miRNAs ( amiRNAs ) were designed against individual RLKs [40] . Only Bla-1/Sha plants with an amiRNA directed against At5g59670b showed a suppression of the hybrid phenotype ( outgrowths , leaf twisting and apical dominance; Figure 2b and Figure S5 ) . We therefore refer to At5g59670b as OUTGROWTH-ASSOCIATED PROTEIN KINASE ( OAK ) . The Bla-1 and Sha OAK primary transcripts are each 3 . 9 kb long , with 13 exons , and a 5′ untranslated region of 92 nt ( expressed in Bla-1 and Sha petioles ) or up to 123 nt ( expressed in Sha pedicels and peduncles ) , as determined by 5′ RACE-PCR . Both OAK alleles encode proteins of 873 amino acids , with 9% of residues being different . The majority of polymorphisms are located in a 152 amino acid region , between positions 180 and 331 , where 55 residues differ ( Figure 3c ) . Among the remaining 721 residues , there are only 19 replacements . Like many other plant RLKs , the OAK proteins include a signal peptide , potential leucine-rich repeats ( LRRs; in OAK , four to five ) , a transmembrane domain , and a cytoplasmic kinase domain ( Michael Hothorn , personal communication; Figure 3d and Figure S6 ) . In addition , two related regions with similarity to a carbohydrate-binding domain in ER-localized malectin proteins from animals [41] are found between the signal peptide and the LRRs ( http://toolkit . tuebingen . mpg . de/hhpred/; Michael Hothorn , personal communication ) . Interestingly , the region that is very different between the Bla-1 and Sha proteins , from residue 180 to 331 , coincides almost perfectly with the second predicted malectin-like domain , from residue 169 to 331 . An analysis of OAK and its homologs ( OAKSha , OAKBla-1 , At5g59670aSha , At5g59670aBla-1 and At5g59670Col-0 ) , using the Codeml program of PAML to assess dN/dS ratios , did not provide evidence for directional or diversifying selection across the entire protein [42] , [43] . However , an Bayesian Posterior Probability analysis of positive selection at individual residues , using At5g59670Col-0 as a reference , suggested that several codons in the second malectin-like domain are under positive selection [44] . A broader analysis of 34 accessions from which OAK sequences could be recovered supported these conclusions ( Figure 3d ) . To determine if the second malectin-like region in OAK homologs is generally hypervariable , we performed a sliding window analysis of all eleven RLKs in the Col-0 , Bla-1 and Sha clusters ( Figure S7 ) . Most highly conserved are the LRR and kinase domains . We also examined in detail the duplicated genes encoding the At5g59670 proteins . At5g59670aSha and OAKSha stood out , because they are identical across the first 598 amino acids of the protein . At the nucleotide level , the two genes include an identical 2 . 7 kb fragment , which most likely reflects a recent gene conversion event that extends from 13 bp upstream of the translational start site to the first 60 bp of the kinase encoding sequences . In conclusion , the divergence between the second malectin-like domain of OAKBla-1 and OAKSha is not representative of the variation between RLKs encoded by orthologs and paralogs in this cluster . To determine the contribution of non-coding and coding sequences of OAK to the outgrowth phenotype , we performed a series of domain swaps between OAKBla-1 , OAKSha , and At5g59670Col-0 ( Figure 4a ) . Similar to plants transformed with the non-chimeric fragments , T1 transformants frequently showed more severe phenotypes than were observed in the F1 hybrids . This indicated that divergent OAK alleles have the potential to cause even stronger incompatibilities than seen between the accessions Bla-1 and Sha . The first major conclusion from the experiments with the chimeric transgenes was that the promoter region contributed to the outgrowth phenotype , because outgrowths were only observed when a particular recombinant protein was expressed from either the OAKBla-1 or OAKSha promoter , but never with the At5g59670Col-0 promoter ( Figure 4b ) . GUS reporter experiments demonstrated that the OAK promoters from Bla-1 and Sha were active in the vascular system of the petioles , in a pattern consistent with the location of the outgrowths ( Figure 5 ) . In contrast , the At5g59670Col-0 promoter drove expression in the leaf lamina , explaining why it could not cause petiole outgrowths . The activity domain of the At5g59670aBla-1 promoter was similar to that of the At5g59670Col-0 promoter , but with additional expression in the lamina of the cotyledons . Finally , the At5g59670aSha promoter was active in all seedling tissues , but in isolated patches that differed from plant to plant . Thus , despite the encoded proteins being closely related , the promoters conditioned a surprisingly wide spectrum of expression patterns , with differences both between duplicates within an accession and among orthologs from different accessions . The OAKBla-1 and OAKSha promoters are more similar to each other than are the coding regions , being 97% identical in the 1 , 238 bp upstream of the start codon . OAK promoter sequences could be recovered from a further 32 accessions . Pairwise identity for all 34 accessions including Bla-1 and Sha was between 97 and 100% . Given the high similarity of the promoter region , the duplication of At5g59670 to form OAK is unlikely to have occurred more than once . Therefore while the change in expression domain has determined how the incompatibility is expressed , the causative changes for the incompatibility are not within the promoter region . In comparison , over the first 1 , 077 bp of the coding region , the pairwise identity for the 34 accessions ranged from 87 to 100% , with a mean of 94% . One accession that was identical to Sha throughout both the promoter and coding region was Kondara , which we found to be incompatible with Bla-1 as well . Across the entire RLK cluster , there were only two nucleotide differences in 17 . 5 kb , and both were in non-coding sequences . Kondara was therefore not considered separately in any of the sequence analyses . Further crosses of Bla-1 and Sha to other accessions with the OAK gene revealed that while most accessions are compatible , a similar incompatibility phenotype is seen in Sha x Bak-2 , Sha x Leo-1 , Mer-6 x Bla-1 and Leb-3 x Bla-1 hybrids ( all incompatibilities between Bla-1-like and Sha-like haplotype groups based on the second malectin domain; Figure S8 ) . Less severe incompatibilities with a late onset of outgrowth formation were found in crosses of Bla-1 to a number of accessions with a second malectin domain that fell into a different haplotype group ( ICE91 , ICE92 , ICE152 , ICE153 , Vash-1 and Valsi-1 ) . Using NeighborNet implemented in SplitsTree [45] , we examined the relationship between the RLKs from the 34 accessions based on the promoter sequences and the extracellular domains ( amino acids 1 to 360; Figure 6a , 6b ) . Similarity in the coding region was not always reflected in promoter similarity , and vice versa , suggesting a history of recombination or gene conversion events . The SplitsTree analysis suggested four major haplotypes at the OAK locus . Analysis with STRUCTURE [46] , where we treated polymorphisms in the OAK locus as linked markers on a chromosome , confirmed that there are four major haplotype groups , with half of the accessions studied showing contributions from more than one haplotype ( Figure 6c ) . Within-locus switching between haplotype groups was confirmed by visual inspection of sequence alignments between individual accessions . This likely reflects high levels of gene conversion or recombination within the OAK gene . A search of the Col-0 reference genome for the possible origin of the OAK promoter revealed that most of it probably arose from the coding region of one of the RLK genes , spanning intron 2 to exon 7 ( encoding amino acids 207 to 383 of At5g59670 ) . Although these regions are only 60 to 70% identical to the OAK promoter ( BLASTN v2 . 2 . 25 , E-value 1×10−61 ) , they present the best matches in the Col-0 genome ( second best hit is to LRR-RLK gene At3g46330 , E-value 3×10−13 ) indicating that this is the most likely origin of the OAK promoter . While the promoter includes potential coding sequences , there are several in-frame stop codons upstream of the predicted OAK translation start . The OAKBla-1 and OAKSha promoters show similar levels of identity with RLK coding sequences across the cluster , but it seems most likely that the duplication of the At5g59670 gene involved an additional duplication that led to conversion of the region coding largely for the second malectin-like domain into a promoter . Interestingly , this is also the portion of the coding sequence that is most different between Bla-1 and Sha . The 260 bp promoter region immediately upstream of the start codon of OAK is most similar to sequences found in triplicate in the At5g59670Col-0 promoter ( Figure S9 ) . A second conclusion of the chimeric transgene experiments was that in addition to the promoter , the protein , and the extracellular domain in particular , contributed to the outgrowth phenotype ( Figure 4a , 4b ) . The At5g59670Col-0 protein did not cause an incompatibility phenotype even when expressed under the OAKBla-1 or OAKSha promoters . Swapping the extracellular and cytoplasmic domains between the OAKBla-1 and OAKSha proteins showed that the cytoplasmic domains were broadly equivalent . However , introduction of the extracellular domain of OAKBla-1 into the Sha genotype , or vice versa , greatly increased the proportion of affected T1 plants . This result is supported by the incompatibility between Leo-1 and Sha , where Leo-1 has an extracellular domain identical to Bla-1 , but only two amino acid differences in the cytoplasmic domain compared to Sha ( Figure S10 ) . Further attempts to narrow down the causal region within the extracellular domain with additional chimeras were not successful . We tested the hypothesis that the outgrowth phenotype resulted from ectopic activation of a kinase-dependent signaling pathway by mutating key residues in the kinase catalytic domain [47] . Double mutants of D693N and K695R should lack all kinase activity . In the Sha background , over 80% of T1 plants carrying the Bla-1 kinase-active construct had a moderate or severe phenotype , while only one third of T1 plants transformed with the Bla-1 kinase-dead construct had any phenotype , and this was always mild . When the Sha kinase-dead construct was transformed back into the Sha accession , all T1 transformants were wild type in appearance , which contrasts with 30% of T1 plants expressing the Sha kinase-active construct having a mild to severe phenotype ( Figure 7a ) . Results were comparable with Bla-1 transformants , although in this case some plants with a moderate phenotype were observed after transformation with the Sha kinase-dead construct . Because RLKs can form homo- and heterodimers [48] , we tested the effects of combining Bla-1 and Sha kinase-dead versions in the neutral Col-0 reference background . We transformed both kinase-active and -dead versions individually into Col-0 and then generated the four possible combinations by crossing ( Figure 7b , 7c ) . The F1 hybrids in which only one of the transgenes expressed a kinase-active version had a less severe phenotype than those carrying both Bla-1 and Sha kinase-active versions . All F1 progeny from five crosses using OAK kinase-dead forms of both Bla-1 and Sha were wild type in appearance . This finding not only confirmed that kinase activity of OAK is required for its function , but also suggested that OAK can act as a heteroallelic dimer or multimer , because a kinase active version of one OAK allele can at least partially complement a kinase-dead version of the other OAK allele . In addition , these data indicated that other RLKs present at the OAK cluster in Col-0 are unlikely to be involved in the outgrowth phenotype . Further circumstantial evidence suggesting that OAK proteins form dimers or multimers was obtained by expressing only the extracellular domain of OAKBla-1 or OAKSha in hybrid plants . Expression under the native promoter in particular suppressed the outgrowth phenotype in many OAKBla-1/OAKSha heterozygous plants ( Figure S11 ) . We propose that by binding to OAK proteins , the extracellular domains reduce the number of active OAKBla-1 or OAKSha heterodimers . Curiosity led us to examine the consequences of mis-expressing the incompatible OAK alleles from the Col-0 promoter in the putative ancestral domain of the leaf lamina . We introduced ProAt5g59670-Col∶OAKBla and ProAt5g59670-Col∶OAKSha chimeric transgenes into the Col-0 reference background , and crossed the transformants , which were wild type in appearance , to each other . As described above , performing this experiment with the OAK wild-type alleles from Bla-1 and Sha reproduced the Bla-1/Sha hybrid phenotype with petiole outgrowths . Co-expressing the Bla-1 and Sha OAK proteins from the Col-0 promoter resulted in a new incompatibility phenotype , ranging from patches of cell death visible to the naked eye on the leaf lamina and abbreviated inflorescences , to severely stunted plants ( Figure 7d–7f ) . It is striking that the altered expression domain leads essentially to a diametrically opposite phenotype , ectopic cell death instead of ectopic cell proliferation . Tissue necrosis and ectopic cell death are typical responses to pathogen infection that rely on salicylic acid signaling [49] . To determine whether the cell death we observed was associated with increased activity of this pathway , we used a transgene that drives constitutive expression of a bacterial salicylate hydroxylase , nahG , which converts salicylic acid to catechol [50] . The Pro35S∶nahG transgene suppressed the cell death phenotype caused by co-expression of OAKBla-1 and OAKSha proteins from the Col-0 promoter , but had no effect on the ectopic outgrowths and other phenotypes seen when the proteins were expressed from their own promoters in Col-0 ( Figure S12 ) . This not only indicated that OAK proteins can couple to alternative downstream signaling pathways ( as is known for the BAK1 RLK [51] ) , but also that the ancestral function might have involved detection of microbes , a known function of different RLKs [52]–[54] . Mutation of other key genes in disease resistance pathways ( PAD4 , EDS1 , and NDR1 ) [49] had no effect on the aberrant phenotypes caused by co-expression of the OAK alleles under either the OAK or the Col-0 At5g59670 promoter .
The A . thaliana genome encodes over 600 RLKs . Approximately two thirds of A . thaliana RLKs are predicted to contain structurally diverse extracellular domains [15] , which often include LRRs [56] . These extracellular domains are involved in perceiving a wide range of ligands , including small proteins , steroids , and carbohydrates . The function and ligands of most plant RLKs are unknown , but known activities of LRR-RLKs include both control of plant development ( e . g . , BRI1 in brassinosteroid response [57] , CLV1 in meristem maintenance [58] and ERECTA in pleiotropic patterning processes [59] ) and microbe detection ( e . g . , Xa21 , FLS2 and GmNARK [52]–[54] ) . The RLK genes constitute one of the most variable gene families in A . thaliana , which has been interpreted as many RLKs evolving in response to pathogen pressure [9] . Local and genome-wide duplications , along with gene conversion , have contributed to the expansion and diversification of RLKs in plants [12] , and RLK genes are overrepresented in tandem arrays [15] , [60] , although those with known roles in plant development are generally not located in tandem arrays [17] . Circumstantial evidence that might point to an interaction of OAK-like RLKs with microbes include the microarray results and the high variability of the OAK gene cluster . OAK does not appear to be required for normal development , since amiRNA-mediated knockdown of OAK activity has no obvious adverse effects . However , it is also possible that OAK acts redundantly in plant development given that the incompatibility phenotype manifests itself primarily as morphological abnormalities . In addition , the mis-expression experiments using the Col-0 promoter revealed that OAKs can trigger typical salicylic-acid dependent cell death as is often seen in response to pathogen attack , although OAK coupling to downstream signaling pathways may be dependent on the expression pattern of alternative interactors . Following the BAK1 paradigm [51] , it is conceivable that the availability of OAK interaction partners determine its activity in plant development versus microbe-interactions . The similarity of the OAK extracellular domains to the carbohydrate-binding protein malectin [41] might indicate that OAK-like RLKs interact with carbohydrates found on the surface of microbes . Alternatively , their function might be detection of damaged self , according to the concept of indirect recognition of pathogens through damage-associated molecular patterns ( DAMPs ) [61] . A role for OAK in plant immunity through perception of self damage would be reminiscent of previously reported cases of hybrid incompatibility that involve disease resistance genes [6]–[8] , [62] . Some RLKs function as hetero- or homodimers , with auto- and trans-phosphorylation required for function of the complex . For example , BAK1 and BRI1 form heteromultimers , and a multi-step pathway involving auto- and trans-phosphorylation events activates downstream signaling [63] . Our experiments with kinase-dead versions demonstrated that kinase activity is important for OAK function . The limited effects of the kinase-dead Sha allele in the Bla-1 background , and vice versa , indicate partial complementation by the opposite kinase-active allele , which is suggestive of heteroallelic dimer or multimer formation . In addition , the suppression of the hybrid phenotypes by expression of the Bla-1 or Sha OAK extracellular domain alone provides further support for this scenario . We do not know whether the change in expression pattern associated with the acquisition of a new promoter by the Bla-1 and Sha OAK alleles subsequently became subject to positive selection , or whether these alleles lack a beneficial function all together . However , the fact that the unusually high divergence in sequence between the two alleles is largely restricted to the second malectin-like domains suggests positive selection or a gene conversion event . We speculate that these sequence changes also altered the affinity for potential ligands . The fact that the Bla-1 and Sha proteins on their own can cause a hybrid-like phenotype , albeit less effectively than when they are combined , suggests that each protein on its own can interact with this potential , unknown ligand . We speculate that OAK heterodimers have increased affinity for such a ligand , leading to ectopic activation of the downstream signaling pathway and aberrant development . Several incompatibilities in F1 and F2 hybrids have recently been linked to disease resistance ( R ) genes . At least one of the A . thaliana factors , and likely another in A . thaliana and rice each , appears to be encoded in a highly polymorphic cluster of NB-LRR genes , the most common class of R genes , and at the same time the most polymorphic gene family in plants [6] , [8] , [9] , [62] , [64] , [65] . Indeed , more broadly , copy number variation is a recurring factor in reproductive isolation [66] . It has been proposed that the occurrence of disease resistance genes in clusters is critical for generating diversity of resistance specificities , because the tandem arrays support high rates of gene conversion and illegitimate recombination [67] . Indeed , complex histories of transposon insertions , translocations , and gene duplications and rearrangements have also contributed to the formation of NB-LRR gene clusters [11] , [13] , [16] , [18] , [19] . RLK genes share with NB-LRR genes the frequent occurrence in tandem arrays and extreme diversity [9] , [12] , [15] . The complex evolutionary history of the OAK cluster is thus not atypical for this gene family . Most hybrid incompatibilities described so far involve multiple loci and as such are classical examples of the Bateson , Dobzhansky and Muller model where derived alleles of two or more genes interact to produce underdominant fitness outcomes ( e . g . [8] , [21] , [62] , [68] ) . In contrast , the incompatibility we describe here is due to interaction of two different alleles at a single locus . Due to the high level of polymorphisms , it is difficult to know what the ancestral allele at the OAK locus looked like immediately after duplication . The incompatible OAK alleles may have evolved through mutations within both the Sha and Bla-1 lineages , with the current alleles remaining compatible with the ancestral allele . Alternatively , all important mutation and gene conversion events may have occurred in only one lineage , through multiple intermediate allelic forms that were never incompatible with the immediately ancestral allele [69] . Either way , evolution of the current situation would not require that plants passed through a fitness valley with heterozygosity for the two incompatible OAK alleles . Not many cases of single-gene hybrid incompatibility have been described in plants: in rice , incompatible alleles of the S5 locus cause most hybrids between the japonica and indica varieties to be female sterile [33] . It is not inconceivable that heterodimers are involved , similar to what appears to be the case for OAK , and dimer formation may be an important pre-condition for evolution of single-gene incompatibilities . We note that passage through a fitness valley is not required so long as the genetic changes causing incompatibility evolve in multiple steps within separate genetic backgrounds . In this way , two alleles could cause underdominance for fitness and reduce or abolish gene flow , but only upon crossing of lines that have diverged independently from a common ancestor . If there were strong positive selection for two different alleles that caused underdominance or sterility in hybrids , then they could eventually contribute to a speciation event . In animals single-gene single-generation speciation occurs in snails , where shell chirality is maternally determined , with opposite chirality forming a strong pre-mating barrier [70] , [71] . Extenuating factors that could allow rapid speciation based on a single locus , even after one generation , include transient silencing of genes , for example , by parental imprinting , or incomplete sterility of the hybrid . If an incompatible allele arises , but is silenced for one generation , this would allow for the production of multiple offspring that are pre-or post-zygotically incompatible with individuals carrying the ancestral allele . Offspring with the new allele can self or interbreed to establish a subpopulation before this allele is lost again by genetic drift . Similarly , if the heteroallelic combination is sublethal , then F2 offspring homozygous for the new allele can be produced . If , in turn , the homozygous form is subject to positive selection , the allele may become established in the population [70] . Such as scenario is particularly applicable to self-fertilizing species such as Arabidopsis thaliana . Whether the sort of developmental abnormalities we have observed in Bla-1/Sha F1 hybrids can contribute to cumulative reproductive isolation is of course not known . Nevertheless , that OAK has the potential to greatly reduce reproductive success can be inferred from the severe phenotypes in some plants transformed with active OAK constructs , the necrosis seen when incompatible OAKs are co-expressed from the Col-0 promoter , and the decrease in lifetime fitness as measured via seed set . All together , we propose that the occurrence of genes in variable tandem repeats , such as NB-LRR genes in several hybrid necrosis cases [6] , [8] , [62] , or RLKs as in the present case , predisposes them to being sources for the creation of novel hybrid phenotypes . Whether , as with other mutations , these are normally disadvantageous or not , will require further systematic analyses of hybrid incompatibilities in a broad range of taxa .
Bla-1 ( N28079 ) and Sha ( N28735 ) were obtained from the European Arabidopsis Stock Centre . Plants were grown at 16°C with 16 hours light , or 23°C with 8 or 16 hours of light , as indicated . Transgenic seedlings were selected on soil by BASTA resistance , and at least 90 T1 plants phenotyped , unless otherwise indicated . Genomic constructs spanned sequences from immediately downstream of the translational stop codon of the preceding gene to 200 bp downstream of the predicted translational stop . AmiRNAs were designed using WMD3 ( http://wmd3 . weigelworld . org/ ) . Constructs were transformed into plants by the Agrobacterium tumefaciens floral-dip method [72] using strain GV3101 pMP90RK or ASE . For reporter gene analysis , the promoter region between the stop codon of the previous gene and the translational start codon of the OAK homolog was inserted into pGWB433 using Gateway LR clonase ( Invitrogen , Darmstadt , Germany ) . Independent ProOAK∶OAKBla-1 and ProOAK∶OAKSha T1 plants in Col-0 that did not show any morphological defects were crossed to each other to create F1 populations , which were raised in randomly distributed individual pots without selection for the transgenes . Plants were genotyped , and seeds collected from each plant after three months of growth and weighed . The weight of individual seeds was determined by weighing 500 seeds for each of three plants per genotype , and total and individual seed weight were used to calculate total seed number per plant . Plants were grown in 23°C ( long days ) at 65% ambient humidity; or under mild drought-stress with minimal watering ( but equal ambient humidity ) ; or in saturated humidity with water surrounding the pots and the tray covered . Bla-1 and Bla-1/Sha petioles were fixed in 3 . 7% formaldehyde , 5% acetic acid , 50% ethanol , embedded in an ASP300 ( Leica , Nussloch , Germany ) tissue processor in paraffin . Transverse sections of 8 µm thickness , stained with 0 . 02% Toluidine Blue after dewaxing , were examined with a Zeiss Axioplan 2 microscope . Seeds were stratified for one week on ½ strength MS plates . Seedlings were grown in Percival LE Intellus chambers ( Perry , IA , USA ) under 23°C long days until the 4-6 leaf stage . At least 40 transverse sections per genotype of leaves ( 1 mm thick ) and petioles ( 2 mm thick ) were placed on callus induction medium ( 3 . 1 g/L Gamborg's B5 salts , 2% glucose , 2 . 6 mM MES , pH 5 . 7 , 0 . 8% agar ) with 2 . 2 µM to 22 nM 2 , 4-dichlorophenoxyacetic acid ( 2 , 4-D ) and 200 nM to 200 pM kinetin . Callus formation was assessed after 12 days . RNA was extracted from leaves of individual plants using the Qiagen ( Hilden , Germany ) Plant RNeasy Mini kit . One µg of RNA was DNaseI treated , and cDNA synthesized with hexamer primers ( Fermentas RevertAid kit , St . Leon-Rot , Germany ) . qRT-PCR was performed with Invitrogen ( St . Louis , MO , USA ) SYBR Green PCR Mastermix and the MJR Opticon Continuous Fluorescence Detection System ( Bio-Rad , Hercules , CA , USA ) . Two technical replicates were performed per sample . Expression was normalized to β-TUBULIN-2 ( At5g62690 ) and an amplification efficiency of 2 . 0 per cycle was used in the calculations . The average across three biological replicates is shown with standard deviation . The 5′ untranslated regions of OAK were identified by 5′ RACE ( GeneRacer , Invitrogen , Darmstadt , Germany ) on RNA from petioles ( Bla-1 and Sha ) or pedicels and peduncles ( Sha ) . Twelve-day old seedlings grown on ½ strength MS plates with kanamycin selection were fixed in 90% acetone on ice for 20 minutes . X-gluc stained tissue [72] was examined with a Leica MZFLIII microscope . Affymetrix ( Santa Clara , CA , USA ) ATH1 microarrays were probed as described [73] . Coarse mapping was performed with the Sequenom ( San Diego , CA , USA ) MassARRAY platform . For high-resolution mapping , approximately 750 F2 and F3 plants were genotyped with microsatellite and CAPS markers [72] . For the sliding window analysis of divergence , amino acid sequences were aligned with MUSCLE ( http://www . ebi . ac . uk/Tools/muscle/ ) and nucleotide sequences with BlastX ( http://blast . ncbi . nlm . nih . gov/Blast . cgi ) . For analysis of population structure , nucleotide sequences were aligned with Lasergene SeqMan . Networks were calculated with SplitsTree [45] using the default parameter settings for NeighborNet . For analysis of haplotypes and recombination , STRUCTURE ( version 2 . 3 . 2 . 1 ) [46] was used with 200 , 000 iterations for the burnin and 800 , 000 iterations for the final analysis . A k value of 4 was used based on the SplitsTree results , with all other parameters as default . Analyses of potential positive selection was performed with the Codeml programme implemented in PAML ( version 3 . 15 ) , using default settings [74] . A likelihood ratio test was used to identify residues under positive selection with Bayesian posterior probability calculated through the Bayes Empirical Bayes ( BEB ) tool [44] . Sites with dN/dS>1 and a high probability ( >95% ) are likely to be under positive selection . A 2-way ANOVA analysis for interaction of lesioning , outgrowth formation and biomass was performed using a web service ( http://faculty . vassar . edu/lowry/anova2×2 . html ) . | While intraspecific hybrids are vitally important in modern agriculture because they often perform better than their inbred parents , certain hybrid combinations fail to develop normally and are inferior to their parents . We have identified an Arabidopsis thaliana hybrid with several aberrant growth phenotypes that are caused by divergence at a single locus encoding the receptor-like kinase OUTGROWTH-ASSOCIATED PROTEIN KINASE ( OAK ) . OAK belongs to a group of similar genes arranged in a tandem cluster that varies substantially between A . thaliana strains . OAK originated through duplication within the cluster with concurrent recruitment of coding sequences from nearby genes to form a new promoter with a novel expression pattern . Kinase activity of OAK is required for its effects , indicating that it is not recognition of an aberrant protein but rather a true gain of function that leads to the incompatibility . Most of the incompatibility seems to come from divergence within the extracellular ligand-binding domain of the OAK protein , indicating that heterodimers of OAK may have higher affinity for a natural substrate compared to either homodimer . Finally , mis-expression of the incompatible OAK alleles from the promoter present in the reference strain of A . thaliana also leads to genetic incompatibility , but with different phenotypic outcomes . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"plant",
"growth",
"and",
"development",
"plant",
"biology",
"population",
"genetics",
"gene",
"function",
"plant",
"science",
"model",
"organisms",
"molecular",
"genetics",
"arabidopsis",
"thaliana",
"genetic",
"polymorphism",
"plant",
"genetics",
"biology",
"evolutionary",
"genetics",
"plant",
"and",
"algal",
"models",
"plant",
"evolution",
"genetics",
"gene",
"duplication",
"evolutionary",
"biology",
"genetics",
"and",
"genomics"
] | 2011 | Complex Evolutionary Events at a Tandem Cluster of Arabidopsis thaliana Genes Resulting in a Single-Locus Genetic Incompatibility |
Transmission-blocking ( TB ) vaccines are considered an important tool for malaria control and elimination . Among all the antigens characterized as TB vaccines against Plasmodium vivax , the ookinete surface proteins Pvs28 and Pvs25 are leading candidates . These proteins likely originated by a gene duplication event that took place before the radiation of the known Plasmodium species to primates . We report an evolutionary genetic analysis of a worldwide sample of pvs28 and pvs25 alleles . Our results show that both genes display low levels of genetic polymorphism when compared to the merozoite surface antigens AMA-1 and MSP-1; however , both ookinete antigens can be as polymorphic as other merozoite antigens such as MSP-8 and MSP-10 . We found that parasite populations in Asia and the Americas are geographically differentiated with comparable levels of genetic diversity and specific amino acid replacements found only in the Americas . Furthermore , the observed variation was mainly accumulated in the EGF2- and EGF3-like domains for P . vivax in both proteins . This pattern was shared by other closely related non-human primate parasites such as Plasmodium cynomolgi , suggesting that it could be functionally important . In addition , examination with a suite of evolutionary genetic analyses indicated that the observed patterns are consistent with positive natural selection acting on Pvs28 and Pvs25 polymorphisms . The geographic pattern of genetic differentiation and the evidence for positive selection strongly suggest that the functional consequences of the observed polymorphism should be evaluated during development of TBVs that include Pvs25 and Pvs28 .
Transmission-blocking ( TB ) vaccines are considered an important tool for malaria control and elimination [1] . TB vaccines aim to disrupt malaria transmission by eliciting antibody mediated responses against antigens expressed during sexual or sporogonic stages of the parasite thereby inhibiting its development inside Anopheles mosquitoes . Thus far , the search of suitable targets for TB vaccines has yielded promising results . In particular , antibodies against some of the multiple parasite proteins have shown excellent TB activities [1] . Among those antigens , the ookinete surface proteins Pvs28 and Pvs25 have been considered candidates to be incorporated in TB vaccines against Plasmodium vivax . These proteins may have originated as result of a duplication event and their orthologous genes ( referred to as p28 and p25 ) have been described in many Plasmodium species [2 , 3] . P28 and P25 are the two most abundant membrane proteins expressed on the zygote and ookinete surfaces; indeed , they might represent as much as 25% of the total ookinete surface proteins [4] . Their structure has been characterized as a triangular prism of EGF-like domains tethered on the cell by a glycosylphosphatidylinositol ( GPI ) anchor at the C-terminus [5 , 6] . Although their specific functions are still not clear , it is known that they are essential for the survival of ookinetes in the mosquito midgut [4] . In particular , studies in P . berghei strongly suggest that P28/P25 proteins have multiple , and partially redundant functions during ookinete and oocyst development [4] . Although they share a common origin and their functions appear to overlap , preliminary studies indicate that P25 proteins are expressed earlier than P28 . Specifically , P25 is expressed prior to fertilization , achieving peak synthesis in the initial hours soon after , and then most abundantly expressed on the surface of the developing zygotes and ookinetes [7] . In contrast , P28 proteins are expressed slightly later on the ookinete surface until the young oocyst stage [7] . In the context of developing TB vaccines , antibodies against these proteins interfere both with ookinete maturation and oocyst formation [8] . In particular , mice antisera against recombinant Pvs28 and Pvs25 recognized both antigens in short term cultures of parasite sexual-stages derived from patients with P . vivax malaria , and significantly suppressed oocyst development in four Anopheles species fed with blood infected with P . vivax Salvador I strain [8] . In addition , in a preclinical trial conducted in Aotus monkeys , animal immunization with recombinant Pvs25 elicited specific antibodies able to fully block parasite infection in membrane feeding assays ( MFAs ) [9] . Furthermore , in a phase I clinical trial conducted with Pvs25 sera obtained from the vaccinated volunteers induced significant inhibition of P . vivax transmission in Anopheles dirus mosquitoes using an ex-vivo MFA [1 , 10] . Moreover , TB immunity elicited with orthologous proteins in P . falciparum and other malaria parasites has been shown as well [8 , 11 , 12] . Unlike the extensive polymorphism commonly observed in several Plasmodium blood stage surface antigens , these proteins are considered to be conserved [13–17] . Therefore , the immunogenicity , TB potential , and limited polymorphism support the use of Pvs28 and Pvs25 as suitable targets for TB vaccines . Here , we study a worldwide sample of pvs28 and pvs25 coding alleles . We detected strong geographic differentiation between populations in Asia and the Americas with replacements at specific amino acid residues novel in the Americas . We also found that these genes can be as polymorphic as some merozoite antigens such as MSP-8 and MSP-10 , with most of their variation accumulating in the EGF2 and EGF3 like domains of both proteins . Finally , our analysis indicates that positive selection may be acting on the accumulation of pvs28 and pvs25 polymorphisms .
We report pvs28 and pvs25 complete CDS sequences from geographically and temporally diverse laboratory strains provided by William Collins at the Centers for Disease Control and Prevention ( CDC ) . We obtained pvs28 gene sequences from the following laboratory strains grouped by their geographic origin: Africa ( Mauritania I ) , Central America ( El Salvador II , Honduras III , Nicaragua , and Panama ) , South America ( Río Meta from Colombia ) , Asia ( Vietnam II , India VII , Thailand and Malaysia ) , and Oceania ( Sumatra from Indonesia , Indonesia XIX , Chesson , and Harris from Papua New Guinea ) . We also obtained 11 pvs28 and 15 pvs25 sequences from Venezuelan archived samples [18] . In addition , we included 259 pvs28 ( total of 284 in the global alignment ) and 310 pvs25 ( total of 325 in the final alignment ) sequences available at the GenBank ( release 208 , June 2015 ) . Those included data of pvs25 from Venezuelan and laboratory strains [19] and sequences from pvs28 and pvs25 from China ( Yunnan Province ) [13] , India ( Delhi , Chennai , Kamrup , Nadiad and Panna ) [16] , Iran [17] , Korea ( ROK ) [15] , Southern Mexico [14] , Thailand ( Tak Province ) [12] and , Bangladesh [20] . Additionally , we report 58 sequences for p28 and p25 orthologous genes from the following species of nonhuman primate parasites ( NHPPs ) : Plasmodium cynomolgi ( p28 from: Berok , B , BX-20 , Cambodian , ceylonensis , PT1 , PT2 , RO , Smithsonian , Gombak , and Mulligan strains; p25 from: Berok , B , BX-20 , Cambodian , ceylonensis , PT1 , PT2 , RO , and Smithsonian strains ) , Plasmodium inui ( p28 from: Celebes I and II , Leaf Monkey II , N34 , OS , Philippine , Leucosphyrus , Perak , Taiwan I and II , and Perlis; p25 from Celebes II , Hawking , Leaf Monkey I , Leucosphyrus , Mulligan , and Perlis ) , Plasmodium knowlesi ( p28 from: H , Hackeri , Malayan from Malaysia , Nuri from India , and the Philippine strain; p25 from: Philippine , Hackeri , and Malayan ) , Plasmodium coatneyi , Plasmodium fieldi ( p28: Hackeri and N-3; p25: ABI and N-3 from Malaysia ) , Plasmodium hylobati , Plasmodium simiovale ( Sri Lanka ) , and a parasite from African primates , Plasmodium gonderi . Information about these species and strains can be found in Coatney et al . [21] . In order to estimate the phylogenetic relationships for the genes encoding pvs28 and pvs25 and their NHPPs orthologs , we also included the published sequences in PlasmoDB , version 24 [22] and NCBI for the Asian species P . cynomolgi ( PCYB_062530 , PCYB_062520 ) and P . inui ( San Antonio 1 , GCA_000524495 . 1 and; AY639974 ) ; the Laverania group that includes P . falciparum ( 3D7_1030900 and 3D7_10310003 ) , Plasmodium gaboni ( Pgk strain , GCA_000576715 . 1 ) , and Plasmodium reichenowi ( PRCDC_1030200 and PRCDC_1030300 ) ; the human parasite Plasmodium ovale ( AB051632 and AB051631 ) ; and the rodent parasites Plasmodium bergei ( strain Anka , AF232051 and XM_670232 ) , Plasmodium chabaudi ( AF232048 and XM_739934 ) , and Plasmodium yoelii ( AF232055 and XM_720005 ) . The phylogenies were rooted with the avian parasite Plasmodium gallinaceum ( M96886 and J04008 ) . In the specific case of P . cynomolgi ( strain B ) , in addition to the p28 orthologous PCYB-062530 , the two other paralogous genes were also retrieved ( PCYB-062510/PCYB-007100 ) from PlasmoDB . To the best of our knowledge , only P . ovale [23] and P . cynomolgi [24] have one and two paralogs to p28 respectively . In the specific case of P . ovale , we only included the pos28-1 ( AB051632 ) sequence in our phylogenetic analyses . For all the samples processed in this study , DNA was extracted from whole blood by using QIAamp DNA Blood Mini Kit ( Qiagen GmbH , Hilden , Germany ) . All the p28 and p25 genes reported in this study were amplified by polymerase chain reaction ( PCR ) . PCR reactions were carried out in 50 μl volume that included 1 . 5 mM MgCl2 , 1 X PCR buffer , 1 . 25 mM of each deoxynucleosidetriphosphate , 0 . 4 mM of each primer , and 0 . 03 U/μM of AmpliTaqDNA polymerase ( Applied Biosystems , Roche-USA ) . For the pvs28 gene , we used the primers 5’-TTTGTTCATTTTTGACATACTCACTT-3’ and 5’-ATGCGCGGTGTGTTATTTGGAG-3’ with an annealing temperature ( Ta ) of 50°C . For P . cynomolgi , P . fieldi , P . fragile , P . inui and P . simiovale , we used the primers 5’-CCAACTGCATTATACAAAAAC-3’ and 5’-ATCTTCTTCGGCGAAAAAA-3’ ( Ta: 47°C ) . For P . knowlesi and P . hylobati , we used 5’-TGCCACCCCTTGTTCAAAATG-3’ and 5’-GWACTGACTCTGYGADACC-3’ ( Ta: 54°C ) . In some cases , a nested PCR was required by using the primer sets 5’-ACTTGCTCACTCGACTTAACC-3’ and 5’-CGTTTTTCTTGTCCCTTTGTCAC-3’ ( Ta: 53°C ) for P . vivax and 5’-ATACAAAAACGACTCCCCCTTT-3’ and 5’-CGTATGACTTGAACTGACTC-3’ ( Ta: 47°C ) for NHPPs . The pvs25 gene and its orthologs were amplified with the primers 5’-CTGACTTTCGTTTCACAGCA-3’ and 5’-ACATCACAAGTCCGTAAGTT-3’ ( Ta: 53°C ) . In the case of nested PCRs , we combined the external primers 5’-CTGACTTTCGTTTCACAGCA-3’ and 5’-CATCACAAGTCGGTAAGT-3’ ( Ta: 53°C ) with the internal 5’-TTCGACCGCTCAATTCGCC-3’ and 5’-CAAGTCGGTAAGTTCAGTAAAG-3’ ( Ta: 55°C ) . The amplification conditions for both genes were as follow: 5 min at 95°C , followed by 35 cycles with 1 min of denaturation at 94°C , 1 min at the specific Ta and elongation at 72°C for 2 min . After 35 cycles , a final elongation step at 72°C for 10 min was carried out . Amplified products from the two independents PCRs were either directly purified or gel extracted and cloned in pGEM-T easy Vector Systems I following the manufactory protocol ( Promega , USA ) . For at least two clones , both strands were sequenced using an Applied Biosystems 3730 capillary sequencer . All the sequences reported in this investigation are deposited in the GenBank under the accession numbers KU285229 to KU285332 . For both genes , p28 and p25 , independent alignments of their nucleotide sequences for P . vivax and their close NHPs malaria species were performed by using the MUSCLE algorithm [25] implemented in SeaView4 [26] on translated sequences followed by visual inspection and manual editing . The protein domains ( signal peptide , EGF-like domains and GPI anchor ) were assigned in the alignments following the description used by Saxena et al . 2006 [5] . In the case of pvs28 , the low complexity regions ( LCRs ) were not included in the polymorphism and phylogenetic analyses; however , those were studied separately as defined by Rich et al . 1997 [27] . We estimated the polymorphism by gene and by domain within each Plasmodium species by using the population statistics π ( the average number of substitutions between any two sequences ) , number of segregating polymorphic sites ( S ) , and haplotype diversity ( Hd ) . The polymorphism was also explored by computing Tajima’s D statistic [28] . The distribution of the genetic diversity across the p28 and p25 gene-sequences was described by calculating π on a sliding-window of 50 base pairs ( bp ) with a step size of 10 sites . The statistic was calculated in each window , assigned to the nucleotide at the midpoint of the window and plotted against the nucleotide position . All these calculations were performed using DnaSP v5 . 10 . 01 [29] . Evidence of natural selection was explored by estimating the average number of synonymous substitution per synonymous site ( dS ) and non-synonymous substitutions per non-synonymous site ( dN ) between a pair of sequences under the Nei Gojobori method [30] , with the Jukes and Cantor corrections as implemented in the MEGA6 [31] . The difference between dS and dN and its standard error was estimated by using bootstrap with 1 , 000 pseudo-replications , as well as a two tailed codon based Z-test on the difference between dS and dN as described in Nei and Kumar 2000 [32] . Under the neutral model , synonymous substitutions accumulate faster than non-synonymous because they do not affect the parasite fitness and/or purifying selection is expected to act against nonsynonymous substitutions ( dS≥dN ) . Conversely , if positive selection is maintaining polymorphism , a higher incidence of nonsynonymous substitutions is expected ( dS<dN ) . We assumed as a null hypothesis that the observed polymorphism was not under selection ( dS = dN ) . We also used the random effects likelihood ( REL ) method as implemented in HyPhy , which uses flexible , but not overly parameter-rich rate distributions [33] and allows both dS and dN to vary across sites independently . REL allows for tests of selection at a single codon site while taking into consideration rate variation across synonymous sites . It is often considered the only method for inferring selection from low divergence alignments such as pvs28 and pvs25 . Evidence for natural selection was also explored in P . vivax by using the McDonald & Kreitman ( MK ) test which compares intra and inter-specific number of synonymous and non-synonymous changes [34] . In this analysis we compared P . vivax with their close NHPPs P . cynomolgi , P . inui and P . knowlesi for both p28 and p25 genes . Significance was assessed using a Fisher’s exact test for the 2 x 2 contingency table as implemented in the DnaSP . In order to study the genetic relationships among worldwide haplotypes , a median joining network was estimated for a set of 284 cosmopolitan sequences of pvs28 and 325 of pvs25 genes by using Network v4 . 6 . 1 . 0 ( Fluxus Technologies 2011 ) . Transversions were set equal to transitions and the epsilon parameter set equal to 0 with only one round of star contraction , which collapses star-like structures in the network into single nodes . The total number of sites included in these analyses excluding gaps or missing data were 547 out of 744 for pvs28 and 558 out of 660 for the pvs25 genes . In addition , we also used DnaSP to estimate the fixation index ( FST ) based on haplotype-frequencies among these geographical regions . In order to investigate whether intragenic recombination generates allelic diversity in the P . vivax ookinete genes , the genetic algorithms for recombination detection ( GARD ) were used to screen for the recombination breakpoints in both alignments , as implemented in Datamonkey ( http://www . datamonkey . org/ ) [35 , 36] . Default parameters for the detection of recombination breakpoints and donor-recipient pairs were used with a significance cut-off of 0 . 05 . The evolutionary relationships among the p28 and p25 genes in Plasmodium spp . were investigated using Bayesian methods implemented in MrBayes v3 . 2 with the default priors [37] . A General Time-Reversible model ( GTR+I+Γ ) was used because it had the lowest likelihood value and possessed the fewest number of parameter that best fit the data ( p28 and p25 ) as was estimated by MEGA6 . For both phylogenies ( p28 and p25 ) , two independent chains were sampled every 200 generations in runs lasting 6 × 106 Markov Chain Monte Carlo steps , and after convergence was reached , we discarded 50% of the sample as ‘burn-in’ period . Convergence is reached when the value of the potential scale reduction factor is between 1 . 00 and 1 . 02 and the average standard deviation of the posterior probability is below 0 . 01 [37] . Additionally , the adaptive branch-site random effects likelihood ( aBSREL ) approach [38] , implemented in Datamonkey , was run to detect evidence of episodic positive selection on all branches using both phylogenies ( p28 and p25 ) . It allows for different Ka/Ks ratios among sites and branches . We performed a likelihood ratio test ( LRT ) comparing the null model ( ω = 1 ) against the alternative , where the branch was undergoing some form of selection ( ω ≠ 1 ) . In addition , we used BUSTED , implemented also in Datamonkey , which is an approach to identify gene-wide evidence of episodic positive selection , where the non-synonymous substitution rate is transiently greater than the synonymous rate [39] . In these analyses we selected both human malarias P . vivax and P . falciparum branches because BUSTED requires pre-specified subset of lineages .
A total of 284 and 325 sequences were studied for the pvs28 and pvs25 genes respectively . Tables 1 and 2 describe the polymorphism found in the complete gene-sequences and their subsequent domains for both genes . The overall genetic diversity , as estimated by π , revealed that both genes are relatively conserved when compared to other vaccine candidates as has been reported by previous studies analyzing a smaller sample size [3 , 12–17] . The pvs28 gene showed a slightly higher , but not significant polymorphism level ( π = 0 . 0037 ± 0 . 0011 ) than pvs25 ( π = 0 . 0023 ± 0 . 0010 ) . When we compared the polymorphism observed between isolates from Asia and the Americas , we found that pvs28 samples from Asia were slightly more polymorphic ( π = 0 . 0035 ± 0 . 0012 ) than the ones from the Americas ( π = 0 . 0024 ± 0 . 0011 ) ( Table 1 ) . In contrast , we observed a similar genetic diversity in the pvs25 sequences from both Asia ( π = 0 . 0017 ± 0 . 0008 ) and the Americas ( π = 0 . 0018 ± 0 . 0010 ) ( Table 2 ) . Nevertheless , in both genes the standard errors for π overlapped when Asia and the Americas were compared . Because of P28 and P25 EGF-like domains carry critical epitopes recognized by P . vivax TB antibodies [5 , 8] , we also estimated π by gene-domain in addition to the geographic regions . In both genes , the EGF2 and EGF3 were the most polymorphic domains of the proteins ( Tables 1 and 2 ) . Fig 1 shows the distribution of the polymorphism across the pvs28 and pvs25 genes using a sliding window approach of the nucleotide diversity π . The polymorphism for both genes was distributed unevenly; the most conserved areas were located at the secretory signal sequence at the N-terminus , the EGF1 and EGF4 like domains , and the GPI anchor at the C-terminus . In contrast , central regions like the EGF2 and EGF3 domains accumulated higher variability . We also studied the polymorphism found in pvs28 ( Table 3 ) and pvs25 ( Table 4 ) genes by country . Regardless of sampling differences , the haplotype diversity Hd was similar for both genes ( pvs28-Hd = 0 . 765 , pvs25-Hd = 0 . 724 ) . Yet , the number of haplotypes found in pvs28 ( H = 53 ) seems to be slightly higher than the ones found in the pvs25 gene ( H = 35 ) . We report estimates by country with at least 10 sequences or more in our alignment . A high haplotype diversity was observed for pvs28 in Bangladesh ( Hd = 0 . 895 ) , Thailand ( Hd = 0 . 993 ) for Asia , and Venezuela ( Hd = 0 . 873 ) from the Americas ( Table 3 ) . For pvs25 , high haplotype diversity was observed in samples from China ( Hd = 0 . 6478 ) , Venezuela ( Hd = 0 . 8000 ) , and Mexico ( Hd = 0 . 6513 ) ( Table 4 ) . In order to explore how natural selection was involved in the maintenance of the observed polymorphism , we estimated the average number of synonymous ( dS ) and non-synonymous ( dN ) changes between two sequences ( Tables 1–4 ) . Overall , we found a significant excess of non-synonymous over synonymous polymorphic changes in pvs28 sequences ( p = 0 . 0271 , Table 1 ) . Nevertheless , when we estimated the average dS and dN by region , this pattern was maintained in Asia ( p = 0 . 0318 ) but not in the Americas ( p = 0 . 4926 ) , specifically in Thailand ( p = 0 . 0280 ) , India ( p = 0 . 0090 ) , and Bangladesh ( p = 0 . 0238 ) . Although there was an excess of non-synonymous ( dN = 0 . 0027 ) over synonymous substitutions ( dS = 0 . 0007 ) in pvs25 polymorphism ( Table 2 and Table 4 ) , the differences were not significant ( p = 0 . 1040 ) ( Table 2 ) . We examined the amino acid replacements observed in the P28 and P25 proteins by using P . vivax Salvador I strain as a reference; our observations are summarized in S1 and S2 Tables respectively . In the Pvs28 protein , we observed 44 amino acid changes , most of them in low frequency ( <0 . 1% on 284 sequences ) . The EGF1 domain had the lowest number of amino acid replacements with only the replacement 52 ( M/L ) found in 25% ( S1 Table ) . In contrast , the EGF2 domain showed the highest number of changes ( 12 replacements ) but most of them observed in low frequency ( <1% ) . The most frequent replacements were at 65 ( T/K ) found in 21 . 8% of the sequences , followed by 87 ( D/N ) and 98 ( L/I ) with 8 . 5% and 4 . 2% frequency respectively . The EGF3 domain displayed a total of 13 amino acid changes , the most common were in positions 110 ( N/Y ) found in 5 . 6% , 116 ( L/V ) in 10 . 9% and 140 ( T/S ) in 16 . 2% of the total samples . The EGF4 domain ( no including LCR of tandem repeats ) and the GPI anchor region showed together only 14 changes in low frequency ( <3 . 0% ) . It is important to emphasize that the polymorphisms in positions 87 ( D/N ) and 110 ( N/Y ) were observed only in the 40 sequences from the Americas with a frequency of 60% and 40% respectively . Here , we report these polymorphisms for the first time in Colombia , El Salvador ( Sal II , [40] ) , Honduras , Nicaragua , Panama and Venezuela . Moreover , some of the frequent amino acid changes observed in Pvs28 ( 52M/L , 65T/K and 116L/V ) were also present in P . cynomolgi and P . inui ( S1 Table ) . A similar analysis for the Pvs25 antigen ( n = 325 ) showed a total of 34 low frequency ( <1% ) amino acid substitutions ( S2 Table ) . The most frequent changes for the EGF2 were 87 ( Q/K ) found in 12 . 7% and 97 ( E/Q ) in 50 . 3% of the worldwide sequences . Glutamine ( Q ) at position 87 is one of the contacting residues involved in the binding of the transmission blocking antibody 2A8 ( Fab VH domain ) with the ß loop ( EGF2 ) of the Pvs25 protein [5] . This amino acid was found to be mutated to lysine ( K ) in several field isolates from Iran , Mauritania , Brazil , Colombia , Mexico and Venezuela . In the EGF3 domain , changes at positions 130 ( I/T ) in 89 . 1% of the sequences in addition to 131 ( Q/K ) in 8 . 2% were the most common . In the case of p25 orthologous genes in close NHP malarias , different amino acid changes were also identified in all these positions ( S2 Table ) . To show the location of the observed mutations on the Pvs28 and Pvs25 structures , the three dimensional structure for Salvador I Pvs28 was modelled by using Phyre2 [41] on the Pvs25 structure as template [5]; 65% of the Pvs28 structure was modelled with 99 . 4% confidence . Positions of mutations for both Pvs25 and Pvs28 were visualized using Visual Molecular Dynamics ( VMD [42] ) . Mutations with a frequency >1% were mapped by residue location and colored according to domain ( Fig 2 , S1 and S2 Tables ) . Residues putatively under positive selection were indicated with arrows ( see results from REL method explained later in the text ) . The haplotype networks based on the pvs28 and pvs25 sequences are shown in Figs 3 and 4 respectively . We identified 63 distinct haplotypes among 284 pvs28 sequences from 18 regions/countries . Although the sampling effort per country did not allow us to reliably estimate and compare the haplotypes’ relative frequencies , there were some emerging patterns . In particular , we focused on the number of countries/areas where a given haplotype had been found since it is informative of its geographic range . The pvs28 network presented two distinctive features referred here to as A and B . The feature A suggests a star-like shape consistent with an expansion of the P . vivax population for part of the network [18] while the feature B refers to the reticulated structure observed in Asia . Only 12 haplotypes ( 19 . 1% ) were shared by two countries or more whereas 51 haplotypes ( 81 . 0% ) were restricted to single countries . Importantly , only three haplotypes were found with a relatively broad distribution ( H1 , H2 , H35; see Fig 3 ) . The haplotype denominated as H1 was the most frequent ( 40 . 1% , 114/284 ) and showed a worldwide distribution ( Fig 3 ) . The haplotype H35 with 59 sequences ( 20 . 8% ) was the second most predominant and the most common in the Indian samples included in this study ( 77 . 3% ) . It was not only found in five distant Indian regions [17] but also in Bangladesh , China , Iran , and Thailand ( Fig 3 ) . The third haplotype in terms of its frequency was H2 ( 4 . 6% , 13/284 ) and belongs to a more divergent cluster which includes only samples from the Americas; specifically , Mexico , Colombia , El Salvador ( Sal II strain ) , Nicaragua , Panama , and Venezuela . The network results suggest that the haplotype H2 could have originated from the most frequent H1 haplotype . Because of the study performed in Korea , which is geographically smaller , involved a large sample collected between 1996 and 2007 [15] , we can speculate that haplotype H1 might be the most common in that region . Feature B of the pvs28 network ( Fig 3 ) showed a group of haplotypes from Bangladesh , China , India , Malaysia , Thailand , and Vietnam forming reticulations . This pattern corresponds to several divergent haplotypes found in very low frequency in this set of samples . The pvs25 haplotype network depicts 35 distinct haplotypes among 325 sequences from 15 regions/countries . We found five haplotypes in high frequency ( H1 , H4 , H20 , H23 , and H25; see Fig 4 ) . The haplotype denominated as H4 corresponded to 152 ( 46 . 8% ) sequences from Bangladesh , China , India , Indonesia , Iran , Korea , and Thailand . This haplotype is related to H1 , the second most predominant with 70 ( 21 . 5% ) sequences distributed in China , India , Korea , Mexico , and Thailand . The other three haplotypes ( H20 , H23 and H25 ) were linked to the most frequent H1 and H4 by long branches . Finally , haplotype 25 was only found in Mexico and Venezuela . To further determine genetic differentiation among populations , the FST fixation index was estimated . Supporting our median joining network results , FST values estimated for both genes suggest high genetic differentiation among P . vivax ookinete genes in different regions ( FST > 0 . 15 , Tables 5 and 6 ) . Pairwise comparisons between Venezuela and Asia regions ( China , Korea , India , Thailand , and Bangladesh ) , produced high FST values for both genes ranging from 0 . 426 to 0 . 542 in pvs28 ( Table 5 ) and from 0 . 457 to 0 . 748 for the pvs25 gene ( Table 6 ) . As expected , a similar pattern was observed in pairwise comparisons between Mexico and Asia regions in both genes , suggesting some degree of differentiation between Asia and the Americas populations . However , when Mexico and Venezuela populations were compared , high FST values were also observed in the pvs28 ( 0 . 251 ) and pvs25 ( 0 . 457 ) coding genes . In contrast , P . vivax populations from Bangladesh compared to China and Thailand , are consistent with a minimal genetic divergence , suggesting no genetic population structure among these regions for pvs28 ( FST <0 . 05 ) . Tajima’s D produced consistent negative values for both pvs28 ( Table 1 ) and pvs25 ( Table 2 ) genes for all populations . In most cases , the results of the test were statistically significant with the exception of the American populations . In order to investigate if recombination generated allelic diversity , the genetic algorithm recombination detection ( GARD [35] ) was performed . No evidence of intragenic recombination was detected in these ookinete genes . In accordance with previous reports , the amino acid sequence alignments of the P28 and P25 proteins suggest that both are conserved among Plasmodium spp . ( S1 Fig ) [2 , 5 , 6 , 23] . All the p28 and p25 sequences included in this study have a conserved hydrophobic signal sequence at the N-terminus ( residues 1–23 , SignalP 4 . 1 Server ) [43] , followed by four cysteine-rich epidermal growth factor EGF-like domains and a short GPI anchor region at the C-terminus ( S1 Fig ) . An invariable number of 20 ( ~9 . 7% ) and 22 ( ~10% ) cysteine residues were found in all NHPPs P28 and P25 proteins respectively . The EGF4-like domain in P28 proteins contains four rather than six cysteines lacking of the 5–6 disulfide bridge . P28 orthologs showed a high average content of Lys ( ~7 . 50% ) , Leu ( ~7 . 23% ) , Asn ( ~8 . 92% ) , Thr ( ~7 . 79% ) and Val ( ~7 . 47% ) ( S2A Fig ) . Likewise , for P25 proteins we found an average content of Glu ( ~9 . 25 ) , Gly ( ~6 . 86 ) , Lys ( ~9 . 67 ) , Leu ( ~7 . 54 ) and Val ( ~8 . 32 ) ( S2B Fig ) . We compared the polymorphism of pvs28 and pvs25 with their orthologs in P . cynomolgi , P . inui and P . knowlesi . S3 and S4 Tables show the genetic variation found in the coding sequence ( CDS ) and the different domains of p28 and p25 genes respectively . We found that P . cynomolgi ( P28-π = 0 . 0340 ± 0 . 0049 , P25-π = 0 . 0284 ± 0 . 0041 ) and P . inui ( P28-π = 0 . 0400 ± 0 . 0045 , P25-π = 0 . 0133 ± 0 . 0026 ) had higher genetic polymorphism than their orthologs in P . vivax . In contrast , the p28 and p25 polymorphisms observed in P . knowlesi ( P28-π = 0 . 0023 ± 0 . 0011 , P25-π = 0 . 0038 ± 0 . 0015 ) were similar to pvs28 and pvs25 ( Tables 1 and 2 ) . The P . knowlesi orthologs also have shown no polymorphism in most of the gene domains ( S3 and S4 Tables ) . Although , the P . cynomolgi p28 paralog PCYB_007100 had similar genetic diversity ( π = 0 . 0340 ± 0 . 0044 ) to the one considered the P . cynomolgi ortholog to pvs28 , PCYB_062530 ( π = 0 . 0340 ± 0 . 0049 ) , the nucleotide diversity was different across both genes ( PCYB_062530 and PCYB_007100 , S3B Fig ) . We estimated the average pairwise dS and dN for NHPPs orthologs to p28 and p25 . In the case of p28 , especially for P . cynomolgi , the diversity found in both paralogous genes was biased toward synonymous sites , a pattern consistent with purifying selection ( S3 Table ) . This contrasts with the pattern of positive selection found in pvs28 . Nevertheless , a similar pattern was found in P . inui . Although there was an excess of synonymous over non-synonymous polymorphisms in P . cynomolgi and P . knowlesi p25 CDS , the differences were not significant using the codon based Z-test ( S4 Table ) . Interesting , the dS-dN estimated by domains suggests different patterns of selection acting in the EGF1 ( negative selection ) and EGF3 ( positive selection ) like domains in P . cynomolgi ( p < 0 . 05 , S4 Table ) . Again , a similar number of synonymous ( dS ) and non-synonymous ( dN ) substitutions was found for p25 in P . inui ( S4 Table ) . The assumption of neutrality was further examined in pvs28 and pvs25 against their orthologs in P . cynomolgi and P . inui by using the MK test . This test showed an excess of nonsynonymous over synonymous polymorphisms in the pvs25 gene when divergence was compared in P . cynomolgi and P . inui ( p < 0 . 05 , S5 Table ) . In both cases , the neutrality indexes ( NI ) were bigger than 1 and the significance of the test was explained by an excess of replacement polymorphisms in pvs25 . These results suggest a possible pattern of balancing selection since a preponderance of non-synonymous intra-species polymorphisms was observed . Similar trends but no significant departures from neutrality were found for pvs28 ( S5 Table ) . The results of the REL method are depicted in Fig 5 ( see S1 Table for reference ) and the estimated Bayes Factors ( BF ) summarize the evidence provided by the data in favor of positive selection . The method detected three codons in pvs28 where the data provided strong evidence for positive selection with BFs bigger than 50 ( codon 52 with a BF of 97 . 69 , codon 113 with a BF of 101 . 51 , and codon 116 with a BF of 367 . 92; see S1 Table for reference ) and five with some evidence for positive selection with BFs bigger than 10 but less than 50 ( codons 53 , 65 , 95 , 98 , and 123; S1 Table ) . In the case of pvs25 , we found only five codons ( 35 , 97 , 130 , 132 , and 135; see S2 Table for reference ) where the data provided some evidence of those being under positive selection with BFs bigger than 10 but less than 50 ( Fig 5 ) . Residues with mutations in high frequency ( >1% ) were mapped on the pvs25 and pvs28 protein structures depicted in Fig 2 . The Bayesian phylogenies of p28 and p25 are depicted in Fig 6 . The avian malarial parasite P . gallinaceum was included as outgroup . Overall , they are comparable with previous published phylogenies using nuclear and mitochondrial genes [19 , 44–47] . Briefly , in both phylogenies , three major clades were identified: 1 ) the Laverania subgenus , 2 ) the clade of rodent malarias , and 3 ) the P . vivax clade . Plasmodium vivax is part of a monophyletic group with closely related NHPPs found in Southeast Asia . The African primate parasite P . gonderi was consistently placed at the base of this monophyletic group in both phylogenies . The difference between p28 and p25 phylogenies was the relative position of P . ovale ( Fig 6 ) . The p28 phylogeny resembled the phylogenetic tree obtained previously based on the nuclear genes ß-tubulin , CDC-2 and the plastid gene tufA [19] . Two additional phylogenies containing all p28 and p25 strains obtained in this study for P . cynomolgi , P . inui , and P . knowlesi were estimated using P . gonderi as outgroup ( S4A and S4B Fig respectively ) . The three p28 paralogs found in P . cynomolgi genome ( B strain , PlasmoDB ) were included ( S4A Fig ) . The p28 phylogeny was slightly different to one that included all species; however , the major relationships were maintained ( e . g . P . inui-P . hylobati , P . fieldi-P . simiovale and P . knowlesi-P . coatneyi ) . We could amplify the three different copies only for the P . cynomolgi strain Berok ( S4A Fig ) , thus we could not confirm that all the P . cynomolgi strains have the two recent paralogs . Noteworthy , all p28 P . cynomolgi paralogs formed a monophyletic group that included the ortholog ( PCYB-062530 ) to pvs25 . This suggests duplication events in P . cynomolgi that took place after divergence from the common ancestor shared with P . vivax . To the best of our knowledge , only P . ovale and P . cynomolgi have reported paralogs to the p28 gene . However , we cannot rule out that such duplication events have occurred in others Plasmodium spp . In the case of p25 ( S4B Fig ) , the relationship among NHP malarias from Southeast Asia were the same as those obtained in the phylogeny containing all the species ( Fig 6 ) . Phylogenetic-based methods were used to explore the role that positive selection may have played in the divergence of these two loci . In the case of p28 , no evidence of episodic diversifying selection was found in any of the 31 total branches using aBSREL ( p ≤ 0 . 05 , corrected for multiple testing ) . However , BUSTED revealed evidence for positive selection acting only on the P . falciparum lineage ( p = 0 . 002 , S6 Table ) . In contrast , no evidence of episodic diversifying selection was found in the p25 gene using aBSREL and BUSTED ( S6 Table ) . In order to fully characterize p28 polymorphism , we examined the LCR of short tandem repeats located at the big C-loop of the EGF4 domain [6] . The description of P28 motifs and amino acid variation is summarized in S7 and S8 Tables . This LCR is also present in all NHPPs that form part of the monophyletic group with P . vivax including P . gonderi from Africa and the Pos28-1 of P . ovale ( S7 Table , S1 Fig ) . In contrast , such LCR is almost absent in species belonging to the Laverania clade ( P . falciparum and related species ) and rodent malarias with the exception of P . yoelii ( S7 Table , S1 Fig ) . In the case of P . vivax , the consensus tandem repeat unit consists of five amino acid motif ( GSGGE ) . The pattern from all P . vivax sequences can be summarized as [ ( G/E/S ) S ( G/R/D ) GE]2–6 , where the first and third positions are polymorphic ( S7 and S8 Tables ) . Interestingly , all the amino acid changes were observed in Asia . The last tandem unit was not a repetitive motif , showing aspartic acid ( D ) in high frequency at the fifth position , but lower for glutamic acid ( E ) and glycine ( G ) : [ ( G/S ) S ( G/D ) G ( D/E/G ) ] . This terminal unit was not included in our polymorphism estimations . Noteworthy , glycine was also the most abundant amino acid in the LCR in all NHP malarias included here ( S7 Table ) . Since proteins domains containing LCRs might be natural immunogenic carrying possible targets for antigenic epitopes [48] , we explored the role of natural selection acting on the observed polymorphism . When the repetitive motifs were aligned among them , a significant ( p < 0 . 05 ) excess of synonymous over non-synonymous substitutions was observed in P . vivax and NHP malarias ( S8 Table ) , suggesting that the motif is conserved and its sequence might be under purifying selection .
Although pvs28 shows slightly higher polymorphism than pvs25 , those differences appear not to be significant . The genetic diversity found in sequences from Asia and the Americas for both genes was similar . This pattern was observed even when there were fewer sequences from the Americas than Asia . This is consistent with studies based on mitochondrial genome sequences ( potentially neutral loci ) and complete genomes showing that the diversity of P . vivax population in the Americas is comparable to Asia [18 , 49] . Pvs28 and pvs25 showed higher genetic variability compared to other sexual stage TB antigens reported in P . vivax as pvs48/45 ( π = 0 , 00053 ) , the Willebrand factor A domain-related protein ( WARP ) ( π = 0 . 00010 ) and also previous estimations of pvs25 ( π = 0 . 00065 ) and pvs28 ( π = 0 . 00000 ) in Korea [50] . The Korean study likely differs from ours because of its limited geographic scope . It is worth noticing that whereas the observed polymorphism is lower than those reported in many merozoite surface antigens such as AMA-1 [51] , there are merozoite stage antigens such as MSP-8 and MSP-10 with similar levels of polymorphism to those reported here for the pvs28 and pvs25 genes [45] . The neighbor haplotype-network for both pvs28 and pvs25 genes formed a star-like shape consistent with the suggested underlying demographic history of a population expansion for P . vivax [18] . This could also explain the significant and negative Tajima’s D estimated for the gene ( Tables 1 and 2 ) . The low global proportion of haplotypes shared between countries for both genes suggests substantial genetic differentiation among P . vivax populations , as confirmed by high FST values ( Tables 5 and 6 ) . We also observed some degree of geographic clustering for haplotypes from the Americas; specifically , a divergent clade for the pvs28 gene characterized by the replacements located at the positions 87 ( D/N ) and 110 ( N/Y ) that were only found in the Americas ( S1 Table ) . Both networks suggest that some of the haplotypes from the Americas could be derived from Asian populations [52 , 53]; however , the pattern is consistent with previous finding indicating that there was not a recent or single introduction of P . vivax into the continent [18] . We performed a comparative polymorphism analysis between pvs25 and pvs28 and their orthologous genes in the Asian Old World monkey parasites that are closely related: P . cynomolgi , P . inui and P . knowlesi . In contrast to the relatively low genetic diversity observed in P . vivax and P . knowlesi , the orthologs in P . cynomolgi and P . inui exhibited significantly higher variability . Similar observations have been also reported for genes expressed in asexual Plasmodium stages [19 , 45] . This pattern could be the result of the different demographic histories of these two parasites when compared to P . vivax and P . knowlesi . Consistently with the effect of demographic differences , the same pattern has been found in the mtDNA and other genes that are considered neutral [19 , 54] . Interestingly , both Pvs28 and Pvs25 proteins showed higher variation at the EGF2 and EGF3 like domains where epitope recognition sites have been identified for blocking antibodies in Pvs25 [5] , and predicted for Pvs28 [6] and Pb28 in P . berghei [55] . Noteworthy , EGF-like domains in the orthologous protein Pfs25 have shown differential immune blocking activity after being separately expressed as a yeast-secreted recombinant protein . In particular , antibodies against the EGF2 domain elicited the strongest blocking activity indicating that this domain might be a good target for TBVs [56] . The EGF-like domains in Plasmodium spp . are relatively conserved among genes and closely related species ( S1 Fig ) . A similar pattern has been described in other EGF-like domains expressed in surface proteins from the merozoite , including MSP-4 [57] , MSP-5 [58] , MSP-8 and MSP-10 [45] . When we estimated the genetic diversity in P . cynomolgi , P . inui , and P . vivax , we observed regions with relatively high polymorphism in EGF2 and EGF3 in both genes ( S3 Fig ) . How this variation affects protein folding and functionality is a matter that remains elusive . However , it has been proposed that EGFs domains can accommodate genetic changes such as gene polymorphism , mutations , insertions and deletions [55] . Consequently , structural folds in the P28 and P25 proteins may not be significantly affected by the observed amino acid changes in natural populations thereby preserving functionality . Previous investigations suggested that the p28 and p25 coding genes were originated as result of a gene duplication event that was prior to the origin of the species included in this investigation [4] . When a duplicated gene neither adapts to a more specialized function nor is silenced by deleterious mutations and continues producing a functional protein , purifying selection could act on both paralogs keeping some level of functional redundancy [59] . Consistently , gene knockouts of either p25 ( P25Sko ) or p28 ( P28Sko ) alone in P . berghei have non-significant effects on oocyst production in infected Anopheles stephensi mosquitoes . However , concomitant disruption of both genes ( Dko ) strongly inhibited oocyst production up to 99% [4] . It is worth noting that duplication events have been reported for p28 in P . ovale [23] and P . cynomolgi [24] ( confirmed here in the Berok strain , S4 Fig ) . Furthermore , in the case of P . cynomolgi , we found evidence of an excess of synonymous over nonsynonymous substitutions in the p28 paralogous gene PCYB-007100 and PCYB-062530 suggesting purifying selection ( S3 Table , S4 Fig ) . Thus , without evidence indicating pseudogenization and patterns consistent with purifying selection , we can speculate that both p28 paralogous remain functional in P . cynomolgi . We searched for evidence of episodic selection as a consequence of changes in ecology/vectors during the evolution of the species include in this study; however , we did not find it . Only the branch leading to P . falciparum indicated positive selection in p28 , a finding that is worth exploring whenever additional Laverania species become available . We also explored the effect of selection on the pvs28 and pvs25 polymorphisms by performing the MK test and applying codon models such as REL . Their caveat is that these tests usually underperform even when adaptive evolution is present so they are regarded as conservative [60] . The MK test detected evidence for balancing selection in pvs25 . A similar pattern of synonymous/non-synonymous sites within P . vivax and its divergence to P . cynomolgi and P . inui was found for pvs28 ( see S5 Table ) , but not significant ( p > 0 . 05 ) . The Bayesian base method ( REL ) , however , detected codons under selection in pvs28 . In particular , the data provided very strong evidence for selection on three pvs28 codons with two of those codons , 113 and 116 ( Fig 2 ) , yielding BF factors above 100 . These two codons are located in the EGF3 domain . We also found other codons in pvs28 and pvs25 where the data provided some evidence of those being under positive selection but their BF did not exceed our 50 threshold defined a priori . Those residues are indicated with yellow arrows in Fig 2 . The patterns consistent with positive selection acting on the pvs28 and pvs25 polymorphism deserve special attention . Whereas the genetic polymorphism observed on surface antigens from the asexual stage has been commonly associated to the selective pressure exerted by the vertebrate immune system [61 , 62] , proteins expressed in the sexual stage may adapt to diverse microenvironments inside Anopheles mosquitoes where parasites have to go through in order to complete their life cycle [63] . The fact that anti-Pvs28 and anti-Pvs25 polyclonal antibodies completely block parasite transmission ( Pv-Sal I ) in four species of Anopheles mosquitoes [8] indicates that these proteins are essential during this phase of the parasite life-cycle . Whether pvs25 and pvs28 facilitate the Plasmodium transit through Anopheles physical barriers and by so doing , increase the parasite ( and may be the vector ) fitness is a matter that needs to be investigated [64 , 65] . The evolutionary and functional implications of LCR in P28 proteins are still elusive . In the case of asexual Plasmodium stages , they may have a role interacting with the host adaptive immune system [66 , 67] . However , such adaptive immune responses are absent in Anopheles vectors with the exception of some components from the vertebrate immune system contained in the blood bolus . The P28 LCR has been predicted to be part of a big C-loop , a fast evolving region forming a sheet over the ookinete surface that may affect the binding properties of the protein [6] . Furthermore , other studies have found that terminal LCR , like the one observed in P28 , may confer more protein binding capacity [68] . This evidence suggests that the P28 LCR is functionally important . This possibility finds also support on the significant excess of synonymous over nonsynonymous changes on the motifs of most of the NHPPs P28 studied ( p >0 . 05 ) ( S7 Table ) , which indicate evolutionary constrains and not simply conservation from continued homogenization due to gene conversion . Nevertheless , assessing the importance of the LCR on P28 requires experimental evidence that is not currently available . In summary , we explored the genetic polymorphism of pvs28 and pvs25 , and investigated the long term evolution of the genes encoding these antigens within the genus Plasmodium . Although they were less diverse than many pre-erythrocytic and erythrocytic stage expressed antigens; their polymorphisms were comparable to others such as MSP-8 and MSP-10 . We also found that these genes exhibit comparable diversities in the Americas and in Asia indicating that the use of TBVs against Pvs28 and Pvs25 will likely face similar challenges in both regions . Furthermore , we found two amino acid replacements in Pvs28 ( positions 87 ( D/N ) and 110 ( N/Y ) ) that appear to be specific for the Americas . Finally , there are polymorphisms that could be maintained by positive selection in both genes and the importance of such observation deserves to be explored . The observation that anti-Pvs28 and anti-Pvs25 polyclonal antibodies can block parasite transmission in some species of Anopheles mosquitoes [8] indicates that polymorphism in these proteins could indeed affect the parasite fitness . In particular , pvs25 and pvs28 polymorphisms could be the result of differences in vectors acting as selective pressure in some ecological contexts . Consequently , it is possible that a vaccine elicited transmission blocking immune response may not be equally effective across all vector-parasite associations in all epidemiological settings . In this context , exploring the diversity of local alleles and their interactions with specific Anopheles species could provide useful information on how to assess TBV efficacy , as well as , how to better deploy these vaccines , even partially effective ones , in the context of malaria control and elimination . | Plasmodium vivax is the most prevalent human malarial parasite outside Africa . The fact that patients can relapse due to the parasite dormant liver stages , among other biologic and epidemiologic characteristics of vivax malaria , facilitates the persistence of the disease in many endemic areas . These challenges have fueled the search for new control tools , including transmission blocking ( TB ) vaccines targeting the parasite sexual stages . Here we study the genetic diversity of two major TB vaccine antigens , Pvs25 and Pvs28 . We show that these genes are relatively conserved worldwide but still harbor diversity that is not evenly distributed across the genes . These patterns are shared by the same proteins in closely related parasite species suggesting their functional importance . We also identify strong geographic differentiation between the circulating variants found in Asia and the Americas . Finally , evolutionary genetic analyses indicate that the observed variation in both genes could be maintained by natural selection . Thus , these polymorphisms may confer an adaptive advantage to the parasite . These results indicate that the genetic variation found in these genes and their geographic distribution should be considered by vaccine developers . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"sequencing",
"techniques",
"invertebrates",
"parasite",
"groups",
"medicine",
"and",
"health",
"sciences",
"plasmodium",
"population",
"genetics",
"tropical",
"diseases",
"geographical",
"locations",
"parasitic",
"diseases",
"animals",
"parasitic",
"protozoans",
"parasitology",
"apicomplexa",
"protozoans",
"molecular",
"biology",
"techniques",
"population",
"biology",
"insect",
"vectors",
"research",
"and",
"analysis",
"methods",
"sequence",
"analysis",
"malarial",
"parasites",
"sequence",
"alignment",
"epidemiology",
"molecular",
"biology",
"disease",
"vectors",
"insects",
"arthropoda",
"people",
"and",
"places",
"mosquitoes",
"haplotypes",
"asia",
"natural",
"selection",
"genetics",
"biology",
"and",
"life",
"sciences",
"malaria",
"evolutionary",
"biology",
"evolutionary",
"processes",
"organisms"
] | 2016 | Evolution of the Transmission-Blocking Vaccine Candidates Pvs28 and Pvs25 in Plasmodium vivax: Geographic Differentiation and Evidence of Positive Selection |
Streptococcus pneumoniae ( the pneumococcus ) colonizes the human nasopharynx and is a significant pathogen worldwide . Pneumolysin ( Ply ) is a multi-functional , extracellular virulence factor produced by this organism that is critical for pathogenesis . Despite the absence of any apparent secretion or cell surface attachment motifs , Ply localizes to the cell envelope of actively growing cells . We sought to characterize the consequences of this surface localization . Through functional assays with whole cells and subcellular fractions , we determined that Ply activity and its release into the extracellular environment are inhibited by peptidoglycan ( PG ) structure . The ability of PG to inhibit Ply release was dependent on the stem peptide composition of this macromolecule , which was manipulated by mutation of the murMN operon that encodes proteins responsible for branched stem peptide synthesis . Additionally , removal of choline-binding proteins from the cell surface significantly reduced Ply release to levels observed in a mutant with a high proportion of branched stem peptides suggesting a link between this structural feature and surface-associated choline-binding proteins involved in PG metabolism . Of clinical relevance , we also demonstrate that a hyperactive , mosaic murMN allele associated with penicillin resistance causes decreased Ply release with concomitant increases in the amount of branched stem peptides . Finally , using a murMN deletion mutant , we observed that increased Ply release is detrimental to virulence during a murine model of pneumonia . Taken together , our results reveal a novel role for branched stem peptides in pneumococcal pathogenesis and demonstrate the importance of controlled Ply release during infection . These results highlight the importance of PG composition in pathogenesis and may have broad implications for the diverse PG structures observed in other bacterial pathogens .
Streptococcus pneumoniae ( the pneumococcus ) is a Gram-positive commensal of the human nasopharynx . Though asymptomatic , nasal carriage is considered a prerequisite for the establishment of invasive pneumococcal disease [1] . The pneumococcus is an extracellular pathogen that elaborates a multitude of virulence determinants that contribute to the pathogenesis of invasive pneumococcal diseases such as otitis media , pneumonia , meningitis and bacteremia , depending on the site of infection . Pneumolysin ( Ply ) is one such conserved , multi-functional virulence factor [2] . As a member of the cholesterol-dependent cytolysin family of pore-forming toxins , Ply is cytotoxic to a variety of eukaryotic host cells [3–5] . Additional activities attributed to Ply include complement activation , induction of host cell signaling cascades , and stimulation of a diverse array of cytokines [6–9] . Ply must be extracellular to carry out the aforementioned functions , however , unlike all other Gram-positive cholesterol-dependent cytolysins , Ply lacks a canonical N-terminal signal peptide commonly associated with Sec-mediated secretion . Additionally , Ply does not encode any of the currently known motifs necessary for cell envelope attachment [10] . Despite these observations , previous studies have demonstrated that Ply is present in culture supernatants [11] and the cell wall compartment during growth [12] . Furthermore , these studies ruled out a role for the major pneumococcal autolysin LytA in this process , suggesting that autolysis alone could not account for extracellular Ply . Therefore , Ply export from the cytoplasm to the cell envelope has been proposed to occur via an active process that remains to be discovered [13] . A defining characteristic of the Gram-positive cell envelope is a thick layer of peptidoglycan ( PG ) that encompasses the cell . PG is a rigid , yet dynamic , macromolecule that confers the characteristic shape of bacterial cells and provides protection against lysis from turgor pressure . The basic structure of PG is conserved and consists of a mesh-like network of glycan strands situated circumferentially around the cell composed of alternating N-acetylglucosamine and N-acetylmuramic acid residues crosslinked through short peptides emanating from the latter sugar moiety [14] . Variation among PG types largely exists at the level of stem peptide composition and the penicillin-binding protein ( PBP ) -catalyzed transpeptidation reactions that serve to generate crosslinks between them [15] . Pneumococcal PG is characterized by a combination of linear and branched stem peptides [16] . Formation of branched stem peptides occurs through addition of a dipeptide branch on the third position lysine residue of a nascent PG precursor molecule during the membrane-associated steps of PG biosynthesis . This activity is catalyzed by two gene products encoded within the murMN operon [17–19] . Crosslink formation between the dipeptide branch and an adjacent stem peptide results in formation of a crossbridge , thus incorporating branched stem peptides into the existing PG network . Therefore , murMN not only affects the peptide composition of PG but also is directly involved in the structural integrity of the mature molecule . Although the exact role of branched stem peptides in pneumococcal biology is ill-defined , it has been shown that murMN expression is necessary for penicillin resistance ( PenR ) [20] . Furthermore , PG isolated from PenR strains displays a marked shift towards a high proportion of branched stem peptides [21] . This shift is attributed to significant divergence within the coding region of murM [22 , 23] with some mutations conferring increased catalytic activity to MurM [18] . Thus , murMN is necessary for PenR and mosaic murM alleles are commonly found in PenR isolates; however , expression of murMN is not sufficient for PenR [24] . The exact association between PenR and murMN remains unclear , but it has been hypothesized that expression of low-affinity PBPs , which confer PenR , demonstrate altered substrate specificity [21] , which may drive the selective pressure for mosaic murM alleles capable of generating higher quantities of the preferred branched stem peptide substrate . In addition to its protective role , PG serves as a scaffold to which numerous secreted molecules are anchored including , but not limited to , a diverse array of proteins that serve a variety of functions for pneumococcal physiology and pathogenesis . Attachment of these proteins can be direct , as in the case of sortase-mediated covalent linkage to PG , or indirect , such as the non-covalent interaction between choline-binding proteins and the PG-linked wall teichoic acids [25] . Despite an extensive knowledge of protein export and the mechanisms responsible for their physical tethering to the cell surface , little is known about the traversal of secreted proteins not destined for attachment to the cell surface during PG maturation . Given its mesh-like structure , it has been postulated that PG acts as a barrier to the release of secreted proteins [26] . In support of this model , early observations in Bacillus amyloliquefaciens demonstrated that washed cells continue to release the secreted protein α-amylase even after inhibition of protein synthesis and Sec-mediated secretion , suggesting the existence of a surface-associated reservoir of this protein [27] . The functional consequences of Ply localization to the cell envelope remain unexplored . In this study , we tested the hypothesis that surface-associated Ply is active and contributes to pneumococcal pathogenesis . Our results indicate that Ply activity and release into the extracellular milieu is inhibited by PG structure . Ply release from the cell appears to be dependent on both the incorporation of branched stem peptides in the PG layer and the action of surface-bound choline-binding proteins . Finally , we demonstrate the importance of appropriate Ply release during infection and the role of branched stem peptides in this process .
To assess the amount of functional , surface-accessible Ply compared to the amount present in the cell wall compartment and cytoplasm , hemolysis assays were performed with washed pneumococci . While washed bacteria accounted for only two percent of the total Ply-dependent hemolytic activity , the isolated cell wall fraction harbored ~30% of the total activity ( Fig 1A ) . This discrepancy suggests that the native cell wall structure is capable of masking Ply exposure on the cell surface , and its liberation is dependent on enzymatic digestion of the PG layer . Protoplasts exhibited the highest hemolytic activity ( Fig 1A ) , indicating that the majority of Ply is retained in the cytoplasm and/or membrane fraction . The activities observed for both cell wall and protoplast fractions correlate well with the amount of Ply detected in each fraction by Western blot analysis [12] , which we confirm here ( S1 Fig , wt ) . Given that the washed cell sample demonstrated hemolytic activity , we sought to determine if the Ply responsible for this hemolysis remains surface-associated or is released from the cell surface . To address this question , paired whole cell samples were incubated with sheep red blood cells or buffer alone for a fixed amount of time . After this incubation , the bacterial cells were removed from the buffer alone sample by centrifugation and the cell-free supernatant was tested for hemolytic activity . The cell-free supernatant harbored the same activity as the whole cell sample indicating that all of the hemolysis observed with washed cells is due to Ply that has dissociated from the cell surface ( Fig 1B ) . Given this , we wondered whether the low amount of Ply release relative to the total Ply present in the cell wall compartment could be explained by binding of Ply to the PG layer . Consistent with the apparent absence of any PG-binding motif , we were unable to detect binding of Ply to purified PG over a range of protein and PG concentrations using a pull-down assay ( S2 Fig ) . Collectively , these data suggest that cell envelope-associated Ply is either not surface-exposed or is somehow inhibited from functioning while still cell-associated . Stem peptide composition within pneumococcal PG displays a high degree of heterogeneity through the cell cycle and this diversity is further extended between different pneumococcal strains [23 , 28] . One feature contributing to this variation is the presence of both linear and branched stem peptides , the latter of which are formed by products of the murMN operon ( Fig 2A ) [17] . MurM and MurN act sequentially to catalyze the tRNA-dependent addition of a dipeptide onto the lysine residue of a PG precursor molecule; MurM acts first to add either a serine or alanine residue , which provides the substrate for the MurN-dependent addition of an alanine [17–19] . Deletion of murMN has no effect on growth in vitro or the apparent amount of crosslinking within PG , yet manipulation of the murMN operon causes drastic changes in the composition of stem peptides and the type of crosslinks that connect them [17 , 24] . Therefore , we reasoned that studies of murMN would allow us to determine the effects of PG composition and structure on Ply release . Deletion of the murMN operon caused a two-fold increase in Ply release as measured by hemolytic activity of whole cells when compared to the wildtype ( wt ) ( Fig 2B , ΔmurMN ) , suggesting a role for the products of this operon in controlling Ply release . Overexpression of murM has previously been shown to favor production of branched stem peptides at the expense of linear stem peptides [29] . To test if increasing the proportion of branched stem peptides would yield the opposite phenotype of ΔmurMN , we created a merodiploid strain carrying a second copy of murMN downstream of a maltose-inducible promoter [30] . In the absence of inducer , murMN overexpression caused a decrease in hemolytic activity of washed cells to the same magnitude as the murMN deletion ( Fig 2B , malM-murMN ) suggesting that basal expression from the maltose promoter is sufficient to increase murMN transcript levels , which was supported by quantitative reverse transcription-PCR ( qRT-PCR ) ( S3 Fig ) . Supplementation of the growth medium with inducer did not further augment this decrease ( Fig 2B ) despite increased expression of the entire operon compared to growth without inducer as measured by qRT-PCR ( S3 Fig ) . Thus , changes in murMN expression are associated with differential Ply release from the cell . To rule out the possibility that genetic manipulation of the murMN operon caused alterations in Ply production or stability , which could account for the phenotypes observed we also measured the hemolytic activity of cell lysates . Mutants lacking or overexpressing the murMN operon all harbored the same total hemolytic activity as wt ( Fig 2C ) , supporting the notion that all strains tested contain the same amount of Ply during the course of the experiment . Furthermore , we determined that Ply localization to the cell wall compartment was unaffected by deletion or overexpression of murMN ( S1 Fig ) , indicating that the phenotype observed for washed , whole cells is not due to defects in Ply production or trafficking to the cell surface . Experiments described later will demonstrate that modest changes in the specific localization of Ply can have profound consequences . In order to verify that ΔmurMN and malM-murMN exhibited distinct stem peptide profiles , we purified PG from each strain and analyzed its peptide composition by reversed phase-high performance liquid chromatography . The wt strain was included as a control . As depicted in Fig 2D , wt PG contained both linear and branched stem peptides . Peptide structures of assigned peaks can be found in S4 Fig . Within the monomeric species , the linear tripeptide ( peak 1 ) represented 20 . 8% of the total peptide material analyzed compared to 3 . 2% for the branched counterparts ( peaks 3 and I ) ( Table 1 ) . However , dimers containing at least one branched structure ( peaks 5 , 6 , 7 , IV , V , VI ) were modestly increased compared to the directly crosslinked linear dimer ( peak 4 ) ( Table 1 ) . These data demonstrate that branched stem peptides can be found throughout wt PG in a manner similar to that observed in PG from other penicillin-sensitive ( PenS ) laboratory strains [20] . In contrast to wt , PG from ΔmurMN was characterized by a virtual loss of branched peptides and a concomitant overrepresentation of linear peptides ( Fig 2D and Table 1 ) . In particular , linear peptides accounted for 90 . 9% of the total material analyzed from this strain , with the directly crosslinked dimer being the most abundant at 57 . 2% ( Table 1 ) . Strikingly , overexpression of murMN caused a drastic shift in the PG stem peptide profile compared to both wt and ΔmurMN ( Fig 2D ) . The abundance of monomers containing a branched structure ( peaks 3 and I ) increased to 13 . 6% , approximately four-fold higher than in the wt ( Table 1 ) . Furthermore , the enrichment in branched peptides was particularly noticeable in the crosslinked material of this strain . Dimers containing a branched structure ( peaks 5 , 6 , 7 , IV , V , VI ) represented 55 . 3% of the total peptide material at the expense of the linear dimer ( peak 4 ) , which decreased five-fold compared to the wt ( Table 1 ) . Additionally , there was a near complete loss in the linear trimer ( peak 10 ) in malM-murMN ( Table 1 ) . Given that the stem peptide profile from malM-murMN was prepared from this strain grown without inducer , these data indicate that a modest ~1 . 5-fold overexpression of murMN is sufficient to cause profound changes to the PG layer ( S3 Fig ) . The peptide profiles depicted in Fig 2D are wholly consistent with previously published results from murMN deletion and overexpression mutants in diverse strain backgrounds [20 , 29] . Taken together , these data strongly support a role for branched stem peptides in limiting Ply release into the extracellular environment . PG is a dynamic molecule that is continuously remodeled during growth and division through the activity of numerous enzymes collectively referred to as PG hydrolases . These factors catalyze PG degradation by cleaving distinct bonds within this structure and , consequently , if not properly regulated can result in cell lysis [31] . Given the association between PG and Ply observed thus far , we hypothesized that Ply release could be due to cleavage of the cell wall by PG hydrolases . To address this possibility , we took advantage of the fact that the major pneumococcal PG hydrolases contain choline-binding domains and are therefore displayed on the cell surface by virtue of binding to the choline residues of teichoic acids [2] . This interaction is non-covalent and can be disrupted by the addition of exogenous choline , causing release of choline-binding proteins ( CBPs ) from the cell surface [32] . Thus , choline treatment would simultaneously remove multiple PG hydrolases ( e . g . LytA , LytB , LytC , CbpD ) from the cell surface as well as other CBPs that harbor distinct functions , allowing us to assess the contribution of this entire subset of proteins to Ply release . Prior incubation with 2% choline decreased the hemolytic activity of supernatants prepared from whole cells of wt , ΔmurMN , and malM-murMN compared to the no choline wash control ( Fig 3A ) . By contrast , choline treatment had no effect on the total hemolytic activity from cell lysates of either strain tested ( Fig 3B ) . This suggests that the ability to release Ply is dependent on the presence of CBPs on the cell surface . Strikingly , the magnitude by which Ply release decreased was dependent on the strain background tested ( Fig 3C ) . The most pronounced change was observed in ΔmurMN , which decreased approximately four-fold after the choline wash relative to the control sample ( Fig 3C ) . By comparison , wt experienced a two-fold drop in Ply release while malM-murMN was modestly , yet significantly , reduced by 1 . 5-fold ( Fig 3C ) . These results suggest that CBPs contribute to Ply release but this effect is sensitive to the proportion of branched stem peptides within PG . Given that Ply is present in both the cell wall compartment and the cytoplasm , we wanted to address the origin of the released Ply observed in washed , whole cells . It is formally possible that the hemolytic activity of whole cells could be explained by specific Ply release from the cell wall fraction due to PG cleavage . Alternatively , the activity could be explained by lysis of a sub-population of cells , which would release Ply from both the cell wall and the cytoplasm . To distinguish between these possible explanations , we reasoned that we could test for the presence of a strictly cytoplasmic marker in addition to Ply . If Ply release is the result of lysis , then we should also detect the cytoplasmic marker; if lysis is not the primary mechanism , there should be enrichment in Ply over the cytoplasmic marker . A commonly used , robust and easily detectable cytoplasmic marker is β-galactosidase , encoded by E . coli lacZ [33] . Therefore , we replaced the coding region of the endogenous β-galactosidase , bgaA , with that of lacZ under the control of a constitutive promoter in the wt , ΔmurMN , and malM-murMN strains and tested for the presence of LacZ in whole cells and sonicated lysates with and without choline treatment . Miller assays to detect β-galactosidase activity were performed on the same samples used to measure hemolytic activity depicted in Fig 3A and 3B . Supernatants from whole cells of ΔmurMN expressing lacZ contained approximately twice as much β-galactosidase activity as wt or malM-murMN ( Fig 3D , no choline wash ) . Overexpression of murMN resulted in a modest decrease in LacZ release from whole cells that was not significantly different from wt ( Fig 3D , no choline wash ) . Interestingly , choline treatment abolished the two-fold increase in β-galactosidase activity observed in supernatants from whole cells of ΔmurMN , reducing it to levels comparable to that of the untreated wt sample ( Fig 3D ) . Importantly , neither expression of murMN nor choline treatment affected the total β-galactosidase activity of cell lysates ( Fig 3E ) , indicating a similar amount of LacZ was present in each strain and condition tested . Therefore , LacZ release increased upon deletion of murMN in a manner dependent on CBPs , whereas wt and malM-murMN had similar levels of LacZ release that are unaffected by CBPs ( Fig 3F ) . To determine whether there was any relationship between the amount of Ply and LacZ released from washed cells of each strain , we calculated the percentage of each protein released from whole cells and fit the data to a straight-line model . By this metric , a slope of 1 is indicative of an equal proportion of Ply and LacZ in the supernatants , which would suggest lytic release of each protein from the cytoplasm . As shown in Fig 3G , there was enrichment in the amount of Ply present in each sample , particularly for ΔmurMN lacZ , as determined by skew towards the y-axis , and the resulting slope was significantly different than 1 . However , a similar analysis performed with the choline-treated samples revealed a slope that was not significantly different than 1 ( Fig 3H ) . Taken together , these data suggest that CBPs contribute to Ply release primarily from the cell wall compartment , but this effect is dependent on the proportion of branched stem peptides in the PG layer . Additionally , some but not all of the Ply release observed in each strain can be attributed to cell lysis . This is particularly apparent for ΔmurMN , which releases half as much LacZ as Ply in a CBP-dependent manner ( Fig 3G ) . However , in the absence of CBPs , Ply release can be solely attributed to lysis , presumably due to the actions of other PG hydrolases within the cell ( see Discussion ) . Given the natural diversity of murM alleles among clinical isolates [22 , 23] , we hypothesized that different murMN alleles would have different effects on Ply release . To test this , we replaced the ΔmurMN deletion locus with the murMN coding regions from different pneumococcal strains . Thus , each murMN allele is under transcriptional control of the native wt murMN promoter on the chromosome . Importantly , introduction of the wt murMN allele back into ΔmurMN restored wt levels of Ply release ( Fig 4A , murMNTIGR4 ) , indicating that the two-fold increase observed upon deletion of murMN ( Fig 2B ) can be attributed specifically to loss of this operon and not due to a second-site mutation that may have occurred elsewhere in the genome during strain construction . Next , we amplified the murMN coding regions from a PenS ( R36A ) or PenR ( Pen6 ) strain [20] and used these to repair ΔmurMN . Purified MurM derived from a PenR strain was previously shown to harbor increased catalytic activity in vitro compared to a PenS counterpart [18] . Expression of the murMNPen6 allele caused a two-fold decrease in Ply release as compared to the wt ( Fig 4A ) , representing a four-fold decrease compared to the parent ΔmurMN strain ( compare Fig 2B , ΔmurMN to Fig 4A , murMNPen6 ) . This enhanced inhibition of Ply release was specific to murMNPen6 , as expression of murMNR36A phenocopied murMNTIGR4 ( Fig 4A ) . Thus , restoration of wt levels of Ply release was achieved with either murMNTIGR4 or murMNR36A , whereas introduction of the highly active murMNPen6 inhibited Ply release to a level comparable to that observed upon murMN overexpression in the malM-murMN strain . To address whether these Ply release phenotypes were accompanied by changes in PG stem peptide composition , we purified PG from each repaired strain and analyzed the stem peptide profile as described above . Strikingly , the profiles from each strain expressing a given murMN allele were noticeably different than that of the ΔmurMN parent strain ( compare Fig 4C to ΔmurMN in Fig 2D ) . Strains expressing murMNTIGR4 and murMNR36A exhibited comparable stem peptide profiles to the wt strain with respect to the presence of both linear and branched stem peptides ( Fig 4C ) . While ΔmurMN lacked any branched monomers ( peaks 3 and I ) , these peptides could be detected in murMNTIGR4 and murMNR36A at 3 . 9% and 4 . 3% , respectively ( Table 1 ) . Additionally , expression of each murMN allele caused an approximate two-fold decrease in the directly crosslinked linear dimer ( peak 4 ) compared with ΔmurMN , accompanied by an increase in the abundance of branched dimers ( peaks 5 , 6 , 7 , IV , V , VI ) to levels similar to that observed in the wt ( Table 1 ) . Thus , the stem peptide profile and Ply release phenotype of ΔmurMN could be restored by expression of wt murMN or the PenS-associated allele from R36A . Expression of the murMNPen6 allele also lead to the formation of branched stem peptides , albeit to a much greater extent than observed with either of the other two murMN alleles tested ( Fig 2D ) . There was significant enrichment in monomers with a branched structure ( peaks 3 and I ) , increasing from zero in ΔmurMN to 13 . 2% ( Table 1 ) . Perhaps more striking was the complete reversal in the dimer structures; more than half of the total peptide material of the ΔmurMN parent was represented by the linear dimer , whereas there was complete replacement of this with the branched forms in murMNPen6 ( Table 1 ) . Thus , expression of the PenR-associated murMNPen6 allele caused a shift towards a highly branched PG reminiscent of the murMN overexpressing strain . Of note , the stem peptide profile of murMNPen6 is more similar to that of Pen6 itself than the wt strain described herein [20] . Thus , the pneumococcal stem peptide profile is largely dictated by murMN expression and activity of the resulting gene products , which is accompanied by differences in Ply release from the cell . Stem peptide analysis of the wt and various murMN mutants described revealed several differences in discrete peptide species between each strain . We were interested in determining whether there were any global trends from this analysis that could best explain the observed differences in Ply release of each strain . As anticipated , the ratio of branched to linear stem peptides in the total material analyzed was highly dependent on the expression of murMN . Deletion of murMN caused enrichment in linear stem peptides , whereas murMN overexpression led to a three-fold enrichment in branched stem peptides compared to wt ( Fig 5A ) . Repair of ΔmurMN with the murMNPen6 allele caused a more pronounced enrichment in branched peptides , representing a four-fold increase over the wt ( Fig 5A ) . The incorporation of branched stem peptides into the crosslinked material was also highly dependent on murMN and mimicked the trends highlighted in the total peptide material . The percentage of oligomeric species ( dimers and trimers ) containing a crossbridge , which is indicative of a crosslink that connects the branch structure to an adjacent stem peptide , increased to approximately 80% upon murMN overexpression or expression of murMNPen6 , which is up from 50% in the wt and the other strains expressing PenS-associated murMN alleles ( Fig 5B ) . Given these gross differences in PG stem peptide profiles we sought to determine whether any specific relationships existed between the PG composition of each strain and the amount of Ply released . We calculated the amount of Ply released as a percentage of the total for each of the strains described in Figs 2 and 4 , and plotted it against the total amount of branched stem peptides within each strain . Intriguingly , this analysis revealed a statistically significant , strong negative correlation ( Fig 5C ) . We extended this analysis further by determining whether particular subsets of branched peptide species are more strongly associated with Ply release . There was no significant correlation between Ply release and the percentage of monomers containing a branched structure ( Fig 5D ) . However , the proportion of branched oligomers showed a significant negative correlation with the amount of Ply release observed ( Fig 5E ) . These results suggest that it is not just the presence of branched stem peptides , but also their incorporation into the mature , crosslinked PG that inhibits Ply release . To assess the contribution of branched stem peptides and Ply release to pneumococcal virulence , we competed ΔmurMN or murMNTIGR4 against wt in a murine model of pneumonia . Neither mutant demonstrated a fitness defect as determined by the competitive index ( Fig 6A ) . However , infection with ΔmurMN caused a 125-fold decrease in the median number of recovered wt bacteria as compared to the murMNTIGR4 competition ( Fig 6B ) . As a control , we also performed single-strain lung infections with wt and found that the titers achieved by the wt alone were 23-fold higher than those during co-infection with ΔmurMN ( Fig 6B ) . However , wt reached similar titers either alone or during competition with the repaired murMNTIGR4 strain ( Fig 6B , compare wt alone to murMNTIGR4 , medians not significantly different ) , suggesting that the absence of branched stem peptides in ΔmurMN negatively impacts virulence . We hypothesized that the increased Ply release in the ΔmurMN strain ( Fig 2B ) was detrimental to the virulence of the co-infecting wt strain . To address whether the inhibitory effect of co-infection with ΔmurMN was Ply-dependent , we deleted the ply structural gene in the ΔmurMN background and competed this double mutant against wt in vivo . In support of our hypothesis , wt titers increased 53-fold compared to the ΔmurMN co-infection and reached levels that were statistically indistinguishable from those observed during competition with murMNTIGR4 ( Fig 6B , compare ΔmurMN Δply to murMNTIGR4 , medians not significantly different ) . As a control , we also competed the single Δply mutant against wt and were able to recover wt bacteria to approximately the same level as in co-infection with murMNTIGR4 or ΔmurMN Δply ( Fig 6B ) . Similar to the other strains tested , neither ΔmurMN Δply nor Δply showed a fitness defect during competition with the wt ( Fig 6A ) . Taken together , co-infection with ΔmurMN limits the ability of wt to achieve high titers in vivo and this inhibitory effect can be relieved by restoration of the murMN operon or deletion of ply . This strongly suggests that appropriate Ply release in vivo is dependent on PG stem peptide composition and maintenance of this ability is necessary for pneumococcal virulence .
Despite the lack of a signal peptide and a cell surface-localization motif , Ply appears to be exported through the pneumococcal membrane and associates with the cell envelope by an unresolved mechanism [12 , 13] . In this study , we demonstrate that exported Ply is not surface-accessible and its activity and release into the surrounding medium is dependent on enzymatic digestion of the PG layer . This may suggest that surface-localized Ply is inhibited due to an inability to fold into a pore-forming competent state while associated with the cell envelope . Alternatively , but not mutually exclusive , the accumulation of Ply at this site may be due to restricted mobility of the exported protein through the PG matrix . Our results cannot distinguish between these two possibilities but it is clear from our experiments that Ply activity and release is dependent on PG hydrolysis . Theoretical and experimental analysis has suggested that purified PG can accommodate globular proteins ranging from ~25–50 kilodaltons ( kDa ) [34] . However , this likely represents a static portrait of PG pore size , as it does not account for the dynamic remodeling of this layer that occurs during cell growth [14 , 35] . Ply is a 53-kDa protein , putting it on the upper boundary of the experimentally defined range noted above . The existence of a “periplasm-like” space delineated by the plasma membrane and cell wall matrix within the Gram-positive cell envelope has been proposed by several groups and experimentally investigated in different species [36–39] . It is tempting to speculate that , upon export through the membrane , Ply is transiently restricted within this “periplasm-like” space due to its limited ability to diffuse through the cell wall matrix , which may be modulated by altering the proportion of branched stem peptides in PG . Interestingly , Ply was released from washed pneumococci in the absence of host cells ( i . e . red blood cells ) , suggesting that this process may not strictly require a host-derived signal or direct contact . This is in contrast to the pH-dependent release of phospholipase C ( PC-PLC ) from the cell wall of the intracellular pathogen Listeria monocytogenes [40] . PC-PLC , like Ply , lacks a sorting motif yet localizes to the cell wall compartment of L . monocytogenes and , under native conditions , exists in a state that cannot be detected with antibodies; though digestion of the PG layer allows for complete release and exposure [40 , 41] . Entry into the host cell cytosol causes a rapid release of surface-associated PC-PLC only upon acidification of the intracellular environment , which is required for optimal virulence of L . monocytogenes [40] . Attempts to detect surface-localized Ply by immunofluorescence have been unsuccessful , perhaps due to the inability of antibodies to reach the niche that Ply occupies within the cell envelope . It is important to note that we cannot rule out the possibility that a specific trigger exists in vivo to stimulate Ply release , however , it appears that this stimulus is not required per se and that release is at least partly dependent on PG structure and the presence of CBPs on the cell surface . Of note , a recent report demonstrated that phagocytosis of the pneumococcus by macrophages results in Ply-dependent death of both host cell and bacterium . This phenomenon was potentiated in a mutant exhibiting hypersensitivity to lysozyme digestion , thus implicating this host factor in contributing to Ply release in vivo [42] . Pneumococcal PG contains both linear and branched stem peptides , the proportions of which vary depending on the expression level of murMN as well as the activity of the encoded proteins ( Figs 2D , 4C and Table 1 ) [17 , 29] . Branched stem peptide formation was inversely related to the amount of Ply released as determined through deletion and overexpression of the murMN operon . MurM and MurN are non-essential and the pneumococcus grows similarly in vitro both with and without branched stem peptides [20] . Therefore , the precise role of branched stem peptides in pneumococcal biology remains an open question . Previous observations demonstrated that murMN expression was associated with an altered propensity to autolyse upon treatment with a variety of cell wall-targeting compounds suggesting that differences in stem peptide composition confer altered susceptibility to autolysis [24] . However , the underlying mechanisms of this phenomenon are unclear . One clue that may help illuminate this issue comes from our studies of Ply and LacZ release from the murMN deletion and overexpression mutants . Our observations suggest that PG lacking branched stem peptides ( exemplified by ΔmurMN ) may be more sensitive to the action of a subset of CBPs capable of mediating PG hydrolysis even in the absence of any overt perturbation to the cell wall and/or membrane . This sensitivity can be suppressed by increasing the abundance of branched stem peptides within the PG layer . Whether variation in branched stem peptide abundance is accompanied by altered susceptibility of the PG layer to the action of PG hydrolases , including those that are CBPs , remains an interesting and unexplored possibility . Furthermore , these results suggest that Ply can be released specifically from the cell surface in the absence of lysis , which is likely the result of controlled PG remodeling by CBPs . Intriguingly , a recent report revealed a role for PG hydrolases ( including pneumococcal LytA , a CBP ) in limiting recognition by peptidoglycan responsive proteins of the innate immune system through controlled trimming of the PG layer [43] . This supports the hypothesis that regulated PG cleavage is critical during infection for additional reasons other than promoting cell growth and division . Removal of CBPs from the cell surface caused an overall decrease in both Ply and LacZ release , especially in the ΔmurMN mutant ( Fig 3H ) . However , both proteins could be detected in equal proportions , suggesting cytoplasmic leakage due to autolysis that may be the result of incomplete removal of surface-bound CBPs or the activity of other , non-choline-binding PG hydrolases . Of note , the pneumococcus encodes numerous known and putative PG hydrolases that are likely unaffected by choline treatment [44] . Intriguingly , CBP-mediated autolysis was also recently shown to be necessary for the release of LytA amidase from the cytoplasm during logarithmic growth [45] . LytA , like Ply , lacks a N-terminal signal peptide suggesting that basal autolysis during growth can allow for the release of other leaderless extracellular proteins that may then reassociate with the cell surface . However , it is important to point out that Ply released upon autolysis was not able to bind to the surface of actively growing cells [13] . The association between PenR and murMN led us to test whether Ply release would be affected by the expression of altered murMN alleles that appear to naturally arise in response to antibiotic pressure . Repair of the ΔmurMN mutant with a PenR-associated murMN allele , but not a PenS-associated allele , caused a four-fold reduction in the amount of Ply release compared to the ΔmurMN parent strain accompanied by significant changes to the stem peptide profile . Given that each allele was under control of the endogenous murMN promoter on the chromosome , these data support previous observations that variant PenR-associated murM alleles encode proteins with higher catalytic activity compared to a PenS MurM [18] . Also , this lends support to the model that PG containing more branched stem peptides represents a more restrictive environment for Ply release . The biological consequences of PenR can lead to pleiotropic effects that can cause fitness defects in different pathogens , including the pneumococcus [46–48] . The precise mechanism ( s ) that select for murMN mutations in the context of PenR low-affinity PBPs are not well understood . There is some evidence suggesting that certain PBPs display different substrate specificities for transpeptidation [49 , 50] . Given that low-affinity PBPs demonstrate decreased affinity for penicillin , which is structurally similar to their natural substrate , it seems plausible to speculate that selection for low-affinity PBPs could create a PG structure that may phenocopy the loss of murMN with respect to both Ply release and decreased virulence . Therefore , this fitness defect may aid in selection for murMN mutations that compensate for altered PG architecture and Ply release . Whether Ply release plays a role in the selection for compensatory mutations in the context of low-affinity PBPs is an intriguing hypothesis that remains to be explored . Infection with the murMN deletion mutant caused a significant Ply-dependent decrease in the ability of wt to achieve high titers in vivo . Perhaps more surprising was that , despite the gross changes in stem peptide composition , this mutant was as fit as the wt during competitive lung infection . Importantly , this result indicates that the altered stem peptide composition in the ΔmurMN mutant does not result in a fitness defect in this strain compared to the wt . Ply is a potent inflammatory agonist that contributes to cytokine production and can cause influx of phagocytic cells [9 , 51] . Therefore , we propose that the increased Ply release caused by infection with ΔmurMN causes a robust , perhaps premature , immune response that results in enhanced bacterial clearance . Neither the wt nor the mutant is equipped to handle this heightened response , which results in a reduction of both strains during this co-infection . It is unclear whether the pneumococcus modulates the proportion of branched stem peptides in the PG layer during infection . Of note , murM expression appears to be specifically upregulated in the presence of epithelial cells in vitro [52] , supporting the hypothesis that the stem peptide composition of PG may be actively modulated under certain conditions mimicking those found in vivo . The finding that repair of ΔmurMN to a murMN+ state restored both Ply release in vitro and caused wt titers to rebound in vivo strongly suggests that it is the specific localization of Ply , not the amount produced , that dictates successful infection . Both of these strains are ply+ and produce the same amount of total Ply in vitro . Thus , the discrepancy in bacterial loads in vivo observed between the groups appears to be independent of Ply production . Indeed , a Δply mutant did not exhibit a fitness defect during co-infection in this model ( Fig 6A ) . At first glance , this appears to be in opposition to the dogma that Ply is a critically important virulence factor of the pneumococcus [53–55] . However , these classical studies were all conducted using single-strain infections with either wt or a ply mutant and not a competition between the two . Indeed , our results are consistent with the observations of Benton et al , who demonstrated that growth of a ply mutant in a blood model of infection could be rescued upon co-infection with a wildtype strain [56] . The lack of attenuation observed for Δply during competition against the wt actually provides further support for our argument that Ply is released from the cell wall compartment during infection as it appears the Δply strain can be complemented in trans by Ply released from the wt . Based on this notion , we propose that the dynamics of Ply release in wt cells is carefully balanced so as to promote virulence without causing immune-mediated clearance . In summary , we have demonstrated that Ply release from the pneumococcal cell wall compartment and cytoplasm can occur in the absence of a host signal but is limited by the native PG network . Based on these observations , we propose that PG acts as a barrier that causes an accumulation of exported Ply at the cell surface that may be released upon remodeling by PG hydrolases , particularly those that are CBPs . Additionally , we identified a novel role for branched stem peptides in restricting Ply release from the cell presumably by affecting the type of crosslinks that create the PG structure . The acquisition of PenR , and other determinants necessary for its expression , may differentially affect Ply release and , as a result , the outcome of an infection . Finally , we demonstrate that branched stem peptides play a critical role in maintaining the PG barrier , which allows for precise control over Ply release . We propose that this is necessary to establish the balance between virulence and immune activation , thus suggesting that PG architecture may play an important regulatory role in the pathogenesis of diverse microbial pathogens . This is the first indication that the cell wall-associated Ply may contribute to pneumococcal virulence .
All experiments were performed with the serotype 4 strain , TIGR4 , or its isogenic mutant derivatives , which are described in Table 2 . Strains were routinely grown from frozen glycerol stocks on tryptic soy agar plates containing 5% sheep blood ( Northeast Laboratory ) overnight at 37°C in a 5% CO2 environment . The growth from overnight plates was suspended in Todd Hewitt broth ( BD Biosciences ) supplemented with 0 . 5% ( wt/vol ) yeast extract ( Fisher Scientific ) ( THY medium ) , diluted to an OD600 of 0 . 02 , and grown to a final OD600 of 0 . 8 in a 37°C incubator with 5% CO2 for all experiments . THY medium was routinely supplemented with 0 . 5% Oxyrase ( Oxyrase , Inc . ) . When necessary , the following antibiotics at the indicated concentrations were used: chloramphenicol ( Cm ) ( 4 μg/mL ) and spectinomycin ( Spec ) ( 200 μg/mL ) . Mutations were generated using allelic exchange with linear PCR amplicons . For deletion constructs , linear amplicons were created via splicing by overlap extension ( SOE ) PCR . 1 kb of flanking sequence immediately upstream and downstream of the target gene was amplified from TIGR4 gDNA and fused to either a CmR or SpecR cassette , which were amplified from pAC1000 and pAC578 ( Table 2 ) , respectively . To repair ΔmurMN with different murMN alleles , the coding regions and intervening sequence of murMN were amplified from gDNA prepared from the appropriate strains ( Table 2 ) and fused to a SpecR cassette and the flanking arms of homology surrounding the native murMN locus . Construction of the murMN overexpression strain was done essentially as described [30] . Generation of lacZ-expressing strains was performed by replacing the coding region of bgaA ( SP_0648 ) on the pneumococcal chromosome with a transcriptional fusion of E . coli lacZ to the SpecR cassette . To do so , the coding region of lacZ amplified from E . coli MG1655 gDNA ( Table 2 ) was placed downstream of the SpecR cassette and was flanked on either side by 1 kb of sequence immediately upstream and downstream of the bgaA coding region by SOE PCR . Transformation of the pneumococcus was carried out as described previously [57] . All mutant constructs were verified using PCR and DNA sequencing . Mid-exponential growth phase cells were normalized to OD600 = 0 . 8 ( approx . 108 CFU/mL ) , collected by centrifugation , washed once in assay buffer [AB; phosphate-buffered saline ( PBS ) , 0 . 1% bovine serum albumin , 10 mM dithiothreitol] and resuspended in AB . An aliquot of each sample was sonicated at 4°C at maximum amplitude for 2 minutes in a water bath sonicator ( Branson , Inc . ) using a 10 second on , 5 second off duty cycle . Whole cell , sonicated , or subcellular fractions were serially 2-fold diluted in AB in 96-well V-bottom plates ( Greiner Bio-One ) . For each experiment , an 8% solution of triple-washed sheep red blood cells ( SRBC ) was freshly prepared and 50 μL of this was added to each well containing 100 μL of sample or control wells containing either AB alone or distilled water . To test whether the cell-free supernatant harbored hemolytic activity , AB alone was substituted for the SRBC solution in the first hour incubation and then 100 μL of the supernatant was transferred to a new plate to which SRBC was added as described above . Plates were incubated at 37°C for 1 hour after which the bacterial cells and any unlysed SRBC were pelleted by centrifugation at 4000 x g for 10 min . One hundred microliters of each supernatant was removed to a 96-well flat bottom plate and the absorbance at 570 nm was measured . After subtracting the AB only value from each sample , the percent activity from each well was determined relative to the distilled water control , which was set to 100% hemolysis . Plotting the OD570 versus the OD600 of each sample yielded a sigmoidal curve , from which the linear portion of the curve was used to extrapolate the OD600 at which 50% hemolysis occurred . The reciprocal of this value is defined as the number of hemolytic units and each whole cell sample or subcellular fractionation was divided by its paired sonicated sample to determine percent hemolytic activity . For removal of choline-binding proteins , strains were grown and processed for hemolysis assays as described above with the following modifications . After washing away media , cells were washed once in PBS and split evenly in two separate tubes . After centrifugation , pellets were concentrated five-fold in either PBS or PBS containing 2% choline chloride ( Sigma ) and incubated at room temperature with agitation for 20 minutes . After centrifugation and washing once in AB , an aliquot was removed for sonication to determine total hemolytic activity and the remainder of each sample was incubated in AB at 37°C for 1 hour . Cells were collected by centrifugation and the supernatant , containing released proteins , was assayed for hemolytic activity as described above . Released and sonicated fractions were prepared as described in the section above and the presence of LacZ in each sample was determined by β-galactosidase assays using the colorimetric substrate ortho-nitrophenyl-β-D-galactopyranoside ( ONPG ) ( Sigma ) as described in [58] . A control reaction containing ONPG alone was included in all experiments and served as the blank sample . Preliminary experiments demonstrated that whole cell lysates from a strain lacking bgaA ( endogenous β-galactosidase ) produced background levels of ONPG hydrolysis . Serial five-fold dilutions of ΔbgaA::spec-lacZ ( AC5073 ) whole cell lysates in triplicate revealed that β-galactosidase activity was linear down to 0 . 89±0 . 014% of the starting material ( OD600 = 0 . 8 ) ( S5 Fig ) . Cell wall material was prepared using a previously published protocol [20] . Peptidoglycan was further purified from approximately 5–20 mg of cell wall material by treatment with 48% hydrofluoric acid for 48 hours at 4°C with agitation . Hydrofluoric acid was removed through extensive washing with 100 mM Tris-HCl pH 7 . 0 and its absence confirmed by measuring the pH . Purified peptidoglycan was then collected by lyophilization and subjected to enzymatic digestion with purified pneumococcal LytA amidase to liberate stem peptides as described previously [20] . Stem peptides were separated and analyzed by reversed phase-high performance liquid chromatography ( RP-HPLC ) as described elsewhere [20] . Cell wall digestion was performed as previously described [13] , with some modifications . Briefly , cells from exponentially growing cultures were collected by centrifugation , washed once with 50 mM Tris-Cl , pH 7 . 5 and resuspended in 100 μL cell wall digestion buffer [50 mM Tris-Cl , pH 7 . 5 , 30% ( w/v ) sucrose , 1 mg/mL lysozyme , 300 U/μL mutanolysin , 1x protease inhibitor cocktail ( Roche ) ] . Cell wall digestion was allowed to proceed for 2 hours at 37°C on a roller drum after which protoplasts were separated from cell wall material via centrifugation at 13 , 000 x g for 10 min . The supernatant was collected as the cell wall fraction . Protoplasts were resuspended in 100 μL 50 mM Tris-Cl , pH 7 . 5 . An equivalent amount of each subcellular fraction was mixed with 2x Laemmli sample buffer and heated in a boiling water bath for 10 minutes prior to loading cell equivalents on a 10% SDS-PAGE gel . Gels were run at 125 V until the dye front reached the bottom of the gel and then proteins were transferred to a nitrocellulose membrane at 25 V for 1 . 5 hours . Membranes were blocked with NAP-blocker ( G-Biosciences ) diluted 1:2 with Tris-buffered saline containing 0 . 1% Tween-20 ( TBST ) and then cut to allow for simultaneous probing with each antibody . Primary antibodies against Ply at 1:1000 ( Statens Serum Institut ) and CodY at 1:1500 ( a gift from A . L . Sonenshein ) were diluted accordingly in NAP-blocker mixed 1:4 with TBST and applied to each membrane at room temperature for 1 hour with rocking . Membranes were washed three times for 5 minutes each with TBST . Appropriate Cy5-conjugated secondary antibody at 1:1000 ( Invitrogen , Inc . ) was applied to each membrane as described above for the primary and then each blot was washed as described above . Membranes were scanned with a Fuji FLA-9000 instrument and the amount of fluorescence was quantitated using MultiGauge analysis software ( Fujifilm , Corp . ) Wildtype cells from a mid-log culture were concentrated to an OD600 of 3 . 5 in PBS and sonicated as described above . After centrifugation to remove cellular debris , the supernatant was transferred to a new tube and used as the source of Ply . Pull-down assays were performed to determine if Ply binds PG . To this end , purified wt PG was mixed with wt cell lysate in a final reaction volume of 30 μL and allowed to incubate on a roller drum at 37°C for 2 hours . Control samples lacking either PG or lysate were performed in parallel . Insoluble PG was pelleted by centrifugation at 20 , 000 x g for 30 minutes at room temperature and the supernatant ( unbound fraction ) was removed to a separate tube . The pellet ( bound fraction ) was washed once with 90 μL PBS , collected by centrifugation , and then resuspended in 30 μL PBS . The presence of Ply within each supernatant and pellet fraction was detected by Western blot analysis as described above with the exception that samples were resolved by SDS-PAGE using NuPAGE 4–12% Bis-Tris gels ( Life Technologies ) run at 200 V for 35 minutes . RNA extraction and qRT-PCR was performed exactly as described [59] to analyze murM and murN transcript levels in exponentially growing cultures of wt and malM-murMN grown in THY medium with and without added 0 . 8% maltose . Lung infections were carried out as described in [59] except that mice were euthanized 30–36 hours post-inoculation . 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 “B2014-37” that was used for this study . All animal experiments were housed in a centralized and AAALAC-accredited research animal facility that is fully staffed with trained husbandry , technical and veterinary personnel . | Pneumolysin ( Ply ) is a protein toxin produced by Streptococcus pneumoniae that contributes to the ability of this organism to cause invasive disease . Release of this protein from the bacterial cell is necessary for many of its functions but the underlying mechanisms driving this process are not well characterized . Previous research demonstrated that Ply localizes to the cell wall compartment . Here , we address the consequences of this localization and reveal a role for the major cell wall structural component , peptidoglycan , in inhibiting Ply activity and release into the extracellular environment . Peptidoglycan is an essential , mesh-like sac that encases the cell , and alterations affecting its composition lead to differences in the amount of Ply released . How molecules interact with and traverse through the restrictive matrix of the cell wall and its associated structures is incompletely understood , particularly with respect to protein secretion and surface attachment . Our results argue that proper maintenance of cell wall-associated Ply is dependent on surface architecture and may be critical for S . pneumoniae pathogenesis . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2015 | Peptidoglycan Branched Stem Peptides Contribute to Streptococcus pneumoniae Virulence by Inhibiting Pneumolysin Release |
Trypanosoma cruzi , the causal agent of Chagas disease , is monophyletic but genetically heterogeneous . It is currently represented by six genetic lineages ( Discrete Typing Units , DTUs ) designated TcI-TcVI . TcI is the most geographically widespread and genetically heterogeneous lineage , this as is evidenced by a wide range of genetic markers applied to isolates spanning a vast geographic range in Latin America . In total , 78 TcI isolated from hosts and vectors distributed in 5 different biomes of Brazil , were analyzed using 6 nuclear housekeeping genes , 25 microsatellite loci and one mitochondrial marker . Nuclear markers reveal substantial genetic diversity , significant gene flow between biomes , incongruence in phylogenies , and haplotypic analysis indicative of intra-DTU genetic exchange . Phylogenetic reconstructions based on mitochondrial and nuclear loci were incongruent , and consistent with introgression . Structure analysis of microsatellite data reveals that , amongst biomes , the Amazon is the most genetically diverse and experiences the lowest level of gene flow . Investigation of population structure based on the host species/genus , indicated that Didelphis marsupialis might play a role as the main disperser of TcI . The present work considers a large TcI sample from different hosts and vectors spanning multiple ecologically diverse biomes in Brazil . Importantly , we combine fast and slow evolving markers to contribute to the epizootiological understanding of TcI in five distinct Brazilian biomes . This constitutes the first instance in which MLST analysis was combined with the use of MLMT and maxicircle markers to evaluate the genetic diversity of TcI isolates in Brazil . Our results demonstrate the existence of substantial genetic diversity and the occurrence of introgression events . We provide evidence of genetic exchange in TcI isolates from Brazil and of the relative isolation of TcI in the Amazon biome . We observe the absence of strict associations with TcI genotypes to geographic areas and/or host species .
Trypanosoma cruzi , a protozoan parasite ( Kinetoplastidea: Trypanosomatidae ) , is known to possess a complex epidemiology and is widely distributed from the southern states of the United States of America to the Argentinian Patagonia . T . cruzi is a pervasive zoonosis capable of affecting more than 150 domestic and wild mammal species , distributed across 8 orders . T . cruzi infection in humans , may result in Chagas disease [1–3] . Transmission to humans is mainly vectorial in endemic areas and over 100 species of hematophagous triatomine insects can harbor the parasites [4 , 5] . Moreover , migration of individuals from highly endemic regions to the United States and Europe has resulted in significant public health concerns in recipient countries [6] . Domestic transmission of Chagas disease ( CD ) in Brazil by Triatoma infestans has been successfully interrupted [7] . However , human infection by T . cruzi is re-emerging as a food-borne disease in previously non-endemic areas [8–10] . Annual outbreaks have occurred , particulary in the northern Brazilian Amazon region during the past decade . Here , some local products derived from fruit juice have been contaminated with infected feces of triatomine bugs of different genera [9 , 11 , 12] . T . cruzi is characterized by a remarkable genetic heterogeneity [13 , 14] and is currently comprised of six lineages or discrete typing units ( DTUs ) , designated TcI to TcVI [15 , 16] . In addition , recent evidence also supports the existence of a seventh lineage ( TcBat ) associated to bats [17] . The most genetically distant DTUs are TcI and TcII [18] . The evolutionary origins of TcIII and TcIV was initially proposed to be the result of an old hybridization between TcI and TcII [19] , however more recent evidence shows that TcIII and TcIV have no hybrid origin , but rather are a monophyletic group with TcI that diverged from TcII [20 , 21] . TcV and TcVI are known hybrid lineages which share haplotypes with TcII and TcIII [22 , 23] . Whether given subpopulations of the parasite are associated with particular vector or host species or with distinct human disease characteristics is still unresolved . TcI is the most frequently isolated DTU in the sylvatic environment , infecting diverse host and vector species across the Americas with an ancestral parental origin estimated at ~0 . 5–0 . 9 MYA [24 , 23] . In Brazil , it is also the most widely distributed DTU , in terms of geography and diversity of host and vector species . Furthermore , it is , by far , the most genetically diverse DTU [25–30] . Llewellyn et al . [31] applied Multilocus Microsatellite Typing ( MLMT ) to the study of TcI population substructure in samples that originated from eight countries , isolated from 18 host and vector species , across 48 tandem repeats [32] . Results revealed extensive intra DTU diversity and spatial structuring of specific genotypes associated with acute oral outbreaks or vectorial infections in Venezuela . In addition , remarkable genetic diversity , through multiclonality , was observed when a single Didelphis reservoir host of TcI was studied [33] . Attempts to subdivide TcI strains into epidemiologically relevant groups are ongoing [34] . Herrera et al . [28] and Cura et al . [35] described five haplotypes associated with transmission cycles in Colombia , Chile and Bolivia . Ramirez et al . [36] , used MLST to identify TcI genotypes specifically associated with human infection ( TcIDOM ) and others associated with peridomestic/sylvatic areas . MLST exploits nucleotide diversity present in four to ten single-copy housekeeping genes and has previously been applied to the study of T . cruzi using different marker combinations for lineage assignment and intraspecific characterization [18 , 37] . Evidence for genetic exchange in TcI has been reported , for example , in strains isolated from Didelphis marsupialis and Rhodnius prolixus in the Amazon Basin [38] and in a domestic/peridomestic TcI population in Ecuador [39] . Experimental generation of intra-lineage hybrids suggest that TcI also displays a potential for genetic exchange [40] . Mitochondrial DNA in T . cruzi has a unique structure and function consisting of approximately 20–50 maxicircles ( ~20kb ) and thousands of smaller minicircles ( ~1 . 4kb ) [38] . Maxicircle DNA is uniparentally inherited and represents a useful taxonomic marker as it is highly mutable in comparison to nuclear DNA . Messenger et al . [41] developed a high resolution maxicircle multilocus sequence typing ( mtMLST ) scheme to describe intra-DTU diversity in TcI , revealing multiple mitochondrial introgression events and heteroplasmy within South American TcI . Introgression had already been detected in North America [21 , 22] and in Brazil [42] and also Bolivia [43] . Together these studies illustrate several remarkable characteristics of TcI , namely the immense geographic distribution , diversity of host and vector species , extensive genetic diversity , and the capacity for genetic exchange . However , little is known about TcI in Brazil and extraordinarily there is only one relevant Brazil centric publication specifically addressing diversity of TcI [42] . Unlike Colombia and Venezuela , in Brazil there is no evidence of population substructure in the context of geographical distribution of intra DTU genotypes , distribution of host/vector species , or genotypes associated with acute outbreaks of CD in Brazil . In the present work , we comprehensively analyze a large cohort of Brazilian TcI isolates from five ecologically disparate biomes . Through the use of high resolution nuclear markers ( MLST and MLMT ) and a maxicircle region ( COII ) , we investigate the phyloepizootiology of TcI from different Brazilian biomes . The study described herein was conducted with the following major hypothesis: DTU I of T . cruzi in Brazilian isolates displays extensive heterogeneity with no particular association of subpopulations to geographic areas , or host/vector species .
A total of 78 TcI isolates were supplied by Coleção de Trypanosoma sp de Mamíferos Silvestres , Domésticos e Vetores COLTRYP/FIOCRUZ . deposited by several researchers and maintained in liquid nitrogen . DNA was extracted immediately following initial isolation in NNN medium and one round of expansion in LIT . The isolates had previously been confirmed as TcI using Mini-Exon PCR [44] . In this work , TcI isolates were characterized using three high resolution methods comprising MLST , MLMT and maxicircle sequencing ( COII ) using appropriate reference isolates . Full isolate details are shown in S1 Table and include characterization methods applied to each sample , isolate localities and collection dates . To increase the robustness of the results , microsatellite information from 50 additional isolates , published by Lima et al . [42] , was included in our MLMT analyses . Isolates originated from vector and mammalian reservoir hosts across five Brazilian biomes; namely Atlantic Forest , Amazon , Caatinga , Cerrado , and Pantanal ( Fig 1 and S1 Table ) . The Cerrado biome is primarily open scrubland ( savannah ) covering approximately 2 million km2 of Central Brazil , comprising 23% of the total land surface area [45] . Scrubland is interspersed with gallery forests and is seasonally dry but with permanent swamplands dominated by Mauritia flexuosa palm trees [46] . The Pantanal biome is a large seasonal floodplain covering approximately 140000 km2 at the core of South America [47] . It is a biodiversity hotspot and freshwater ecosystem of global significance containing diverse mammal species and habitats . Climate instability results in periodic floods and droughts , affecting the population number and behavior of some species [48] . The Amazon biome is in the largest hydrographic basin of the world , comprising 44% of the South American subcontinent . The biome is a complex mosaic of very diverse ecosystems , dominated by tropical rain forests , with semi-arid regions , and a variety of man-made landscapes . The Amazon biome contains the greatest biological diversity ( in absolute terms ) on the planet [49] . The Atlantic Forest biome extends from the south of Pernambuco to the south of Rio Grande do Sul , and it is characterized by humid tropical forest . This biome is extensively impacted by human activities . It originally encompassed 12 percent of the national territory but only 1 to 5 percent ( less than 100 , 000 km2 ) is intact today [50] . Containing more than 8 , 000 endemic species , the Atlantic Forest is recognized as one of the world’s most significant biodiversity hotspots [51] . The Caatinga biome in northeast Brazil , comprises a semi-arid ecological landscape with only 1% of its territory currently conserved , it is threatened by agriculture and cattle ranching [52] . This biome is characterized by clay and sandy soils with open plains supporting flora that is typical of semi-arid regions [53] . The TcI COII locus was amplified and sequenced according to Messenger et al . [41] . Nucleotide sequences per gene fragment are available at GenBank under accession numbers: MF781085-MF781124 . Phylogenies were constructed implementing the substitution model based on the Akaike Information Criterion ( AIC ) in MEGA 6 [68] . To compare nuclear and mitochondrial topologies , Maximum-Likelihood ( ML ) phylogenies were constructed ( T92+I model , Tamura 3-parameter ) which assumes that a fraction of sites is evolutionarily invariable [68] . TcIII ( CM17 ) and TcIV ( Saimiri3 cl1 , X10/610 cl5 , ERA cl2 and 10R26 ) strains were included as outgroups ( accession numbers: JQ581330 . 1 , JQ581331 . 1 , JQ581329 . 1 , JQ581328 . 1 and JQ581327 . 1 , respectively ) [41] . Twenty-five microsatellite loci were amplified as previously described by Llewellyn et al . [31] with some modifications ( S3 Table ) . Markers were distributed across 11 chromosomes , including six groups of physically linked loci [69] . The following reaction conditions were implemented across all loci: a denaturation step of 4 mins at 95°C , 30 amplification cycles 95°C ( 20 s ) , 57°C ( 20 s ) , 72°C ( 20 s ) with a final 20 mins elongation step at 72°C . Reactions were performed in a final volume of 10 μL containing , 1X ThermoPol Reaction Buffer ( New England Biolabs ( NEB , UK ) , 4 mM MgCl2 , 34 μM dNTPs; 0 . 75 mM of each primer , 1 unit of Taq polymerase ( NEB , UK ) and 1 ng of genomic DNA . Five fluorescent dyes were used to label forward primers , 6-FAM & TET ( Proligo , Germany ) , NED , PET & VIC ( Applied Biosystems , UK ) . Allele sizes were determined using an automated capillary sequencer ( ABI 3730 , Applied Biosystems , UK ) , in conjunction with a fluorescently tagged size standard ( GeneScan– 500 LIZ , Applied Biosystems , UK ) , and manually checked for errors in GeneMapper software v3 . 7 ( Applied Biosystems , UK ) . Microsatellite data were assessed in accordance with Lewellyn et al . [31] . Individual-level clustering defined by Neighbour-Joining ( NJ ) phylogenies ( DAS: 1 –proportion of shared alleles at all loci/n ) between microsatellite genotypes was calculated in MICROSAT v . 1 . 5 [70] under the infinite-alleles model ( IAM ) . To accommodate multi-allelic genotypes ( ≥3 alleles per locus ) , a script was developed in Microsoft Visual Basic to generate random multiple diploid re-samplings of each Multilocus profile . A final pair-wise distance matrix was derived from the mean across multiple re-sampled datasets and used to construct a NJ phylogenetic tree in PHYLIP v3 . 67 [71] . Majority rule consensus analysis of 10 , 000 bootstrap trees was performed in PHYLIP v 3 . 6 by combining 100 bootstraps generated in MICROSAT v . 1 . 5 [70] , each drawn from 100 randomly re-sampled datasets . Population assignment with a prior assumption of subdivision by collection sites was estimated with the Bayesian clustering program Structure v . 2 . 3 [72] . We assumed the admixture model due to the lack of information regarding ancestry , with correlated allele frequencies ( i . e . frequencies in different populations are similar as a consequence of migration or shared ancestry ) [72] . Simulations were set at 106 Markov Chain Monte Carlo ( MCMC ) interactions , with 2 . 5 x 105 iterations as burn-in . Ten independent runs were performed for each value of K ( that correspond to the number of groups , 2–10 ) , as suggested by Pritchard et al . [72] . The most likely K value was estimated with the ΔK method [73] . An alternative approach to summarize genetic polymorphism was performed using a non-parametric approach , free from Hardy-Weinberg assumptions . Briefly , a K-means clustering algorithm , executed in ADEGENET [74] was used to identify the optimal number of ‘true’ populations , with reference to the BIC , which reaches a minimum when approaching the best support for assignment of isolates to the appropriate number of clusters . The relationship between clusters and the strains contained within them was evaluated using a discriminant analysis of principal components ( DAPC ) , as described in Jombert et al . [74] . A single randomly sampled diploid dataset , generated using a custom Microsoft Visual Basic script to re-sample random multiple diploid combinations of each Multilocus profile , was used for all subsequent analyses , as described in Jombert et al [75] . Population genetic statistics were calculated considering strains assigned to their DAPC-derived population clusters . DTU-level genetic diversity was evaluated using sample size corrected allelic richness ( Ar ) in FSTAT v 2 . 9 [76] . Intra-population sub-clustering was calculated as mean pairwise DAS values and associated standard deviations in MICROSAT v1 . 5 [70] . Sample size corrected private ( population-specific ) allele frequency per locus ( PA/L ) was calculated in HP-Rare [77] . Mean FIS , a measure of the distribution of heterozygosity within and between individuals , was calculated per population in FSTAT 2 . 9 . FIS varies between -1 ( all loci are heterozygous for the same alleles ) and +1 ( all loci are homozygous for different alleles ) . DTU-level heterozygosity indices were calculated in ARLEQUIN v3 . 11 [78] and associated significance levels for p-values derived after performing a sequential Bonferroni correction to minimize the likelihood of Type 1 errors [79] . Population subdivision was estimated using pairwise FST , linearized with Slatkin’s correction , in ARLEQUIN v 3 . 11 . Statistical significance was assessed via 10 . 000 random permutations of alleles between populations . Three different strategies were performed to group the samples and calculate pairwise FST values: i ) using isolate collection locations to investigate local diversity , ii ) to assess levels of gene flow between the five ecologically disparate biomes and , iii ) investigate the role of host/vector specificity in the context of host movement and the distribution of TcI genotypes . Within-population subdivision was evaluated in ARLEQUIN v 3 . 11 [74] using a hierarchal analysis of molecular variance ( AMOVA ) . A Mantel test for the effect of isolation by distance within populations ( pairwise genetic vs . geographic distance ) was implemented in GENAIEX 6 . 5 using 10 , 000 random permutations [80] . The association between host/vectors and genotypic clusters based on DAPC were calculated using contingency tables along with a Chi-squared test . Nucleotide sequences per gene fragment are available from GenBank under the accession numbers: MF615620-MF615679; MF615680-MF615739; MF615740-MF615799; MF615800-MF615859; MF615860-MF615919; MF615920-MF615979 .
Individual gene fragment trees revealed multiples polytomies in all six phylogenetic trees ( S1–S9 Figs ) . Substantial congruences between the phylogenetic trees generated with SNP duplication ( with Bayes ) and Average State ( with NJ ) were observed ( S1–S6 Figs ) , The two fragments with the most pronounced inconsistencies between Bayes and NJ were PDH and RHO1 ( S4 and S6 Figs ) . S7–S9 Figs , show the comparison between the six gene trees . RB19 and RHO1 each produced a cluster corresponding to isolates from Atlantic Forest , Cerrado and Pantanal , which are mostly congruent . However , no two gene fragments showed completely identical topologies . The remaining loci ( CoAR , LAP , PDH and GTP ) , which had lower TE and DP values , generally yielded trees that were less congruent . None of the gene fragments showed 100% congruence between their clusters . Topological incongruence analyses revealed a mean of 2 . 86 incongruences per branch and 25% of branches with at least n-1 incongruent fragments . These correspond to moderate levels of incongruence ( S11 Fig ) , where moderate incongruence was defined as being between 20 and 40% [64] . The ILD tests of discrepancies were no higher than expected , indicating that the combination of six gene fragments produces reliable branches ( ILD = 0 . 05 ) . To evaluate intra-DTU diversity of TcI , phylogenies were inferred from the concatenated alignment of six gene fragments ( Fig 2 ) . Both , NJ and Bayesian methods produced similar results , although NJ analysis showed lower bootstrap values . Clusters with >50% support in both analyses are indicated . The presence of several sub-clusters was observed , revealing considerable intraspecific diversity within TcI and also similar genotypes circulating sympatrically over large geographical areas . Specific phyloepidemiological observations are as follows . Cluster A grouped isolates originating from very distant localities including the Atlantic Forest , Cerrado and Pantanal biomes ( BPP = 81% and bootstrap <50% ) . Of particular note , cluster A ( Fig 2 ) lacked genotypes present in the Amazon , in congruence with results from maxicircle phylogenies ( below ) . Also , similar genotypes were isolated from different species . For example , isolates from Didelphis spp , primates , chiroptera , one rodent , and triatomine bugs grouped within cluster A ( Fig 2 ) . Amazonian isolates were genetically diverse and were mostly contained within a single clade ( Fig 2 ) . Cluster B ( Fig 2 ) contained genotypes from the Atlantic Forest and Cerrado biomes , comprising an enormous geographical distance ( ~1 . 130 km ) . Interestingly , cluster C comprised isolates from distant biomes , Amazon and Atlantic Forest . Likewise , isolates from Abaetetuba ( 11609 ) and Cachoeira do Arari ( 10272 ) , separated by vast geographical distances ( ~78 . 38 km ) , were grouped in the same cluster . Of note , a single isolate FRN26 , Oecomys mamorae from the Pantanal , was genetically dissimilar from all other TcI strains and placed in different topological positions in the context of MLST and maxicircle phylogenies . Although isolates were collected in different years and localities ( S1 Table ) , no clear clustering by collection date or biome was apparent; however , statistical tests were not able to rule out the existence of some association . Details of these tests are described below . Haplotype analysis , applied to nuclear loci , was used to generate phylogenies and investigate the allelic origins of heterozygous isolates from homozygous putative donor genotypes ( Fig 3 and S12–S16 Figs ) . Here isolates with haplotypes present in two different genetically clusters that also contained respective homozygous donor isolate genotypes were considered potential hybrids . Three genetic loci ( GTP , PDH and RB19 ) revealed heterozygous isolates and allelic profiles that could be derived from homozygous genotypes ( Fig 3 , S12 and S13 Figs , respectively ) . In more detail , Fig 3 shows the GTP locus and alleles from homozygous donor isolates: 10285 , haplotypes 1 and 2 , in one clade; and 14943 , haplotypes 1 and 2 , in another . Within GTP , two isolates contain heterozygous allelic profiles , 12630 and 12624 , corresponding to one allele from each homozygous donor . For PDH , five isolates contain heterozygous allelic profiles: 2892 , 2896 , 10285 , 17645 and G41 ( S12 Fig ) . The potential parental alleles for PDH were: 2879 and 2869 , in one clade , and 2880 , in another clade ( S12 Fig ) . Similarly , for RB19 , eight isolates showing potential genetic exchange were identified . The most plausible parental alleles for each putative hybrid are shown in S13 Fig , while the SNP profile for putative homozygous donors and the corresponding heterozygous profiles are shown in the S4–S6 Tables . Putative recombinants were different in PDH , GTP , and RB19; possibly indicating that there have been multiple genetic exchange events over time . Although we detect the signature of genetic exchange through heterozygous genotypes and their associated homozygous “donor” isolates , we observe no evidence of genetic exchange at the level of individual alleles , since allelic mosaics were not detected using RDP3 software . Sixty two COII sequences produced a 449 bp alignment , 10 unique haplotypes and 64 polymorphic sites . Maximum-Likelihood trees ( Fig 4 ) revealed two major clades and almost complete congruence with cluster A derived from concatenated MLST ( S17 Fig ) . This cluster contains strains from the Atlantic Forest , Cerrado and Pantanal with the notable exclusion of Amazonian isolates ( bootstrap = 100% ) . Interestingly , isolate FRN26 from the Pantanal , and isolate G41 from the Atlantic Forest formed a strongly supported sub-clade ( bootstrap = 100% ) . In contrast , nuclear phylogenies grouped G41 with Amazonian isolates . Also of note , isolates within sub-clusters were highly homogeneous . Analyses with MLST and maxicircle were congruent in relation to the isolates of Amazon , in which they formed a separate group that included a cluster with isolates from Cerrado and Atlantic Forest ( S17 Fig ) . The presence of genetically identical mitochondrial sequences despite a mutation rate one order of magnitude greater than that of nuclear genes provides support for the occurrence of multiple mitochondrial introgression events ( Fig 4 and S17 Fig ) . Additionally , these sequences correspond to geographically dispersed isolates , obtained from different biomes and hosts and vectors , further supporting the case for introgression . In total , 4595 alleles were identified , corresponding to 92 unique multilocus genotypes . Multiple ( ≥3 ) alleles were observed at 1 . 87% of markers . This is most likely attributable to aneuploidy in a small proportion of the loci ( S7 Table ) . Bayesian clustering applied to 92 strains revealed the existence of four discrete phylogenetic groups without apparent association to the biome of origin ( Fig 5 ) . For example , isolates from Atlantic Forest clustered across three groups ( Fig 5 , yellow , green and pink label ) , in which specimens from the state of Rio de Janeiro are genetically similar to those from Posse , Goias ( Cerrado biome ) and specimens from the states of Minas Gerais and Santa Catarina clustered together with samples from Pantanal . Moreover , TcI specimens of the state of Bahia are genetically more similar to samples from the states of Piauí ( Caatinga ) , Pará and Amazonas ( Amazon ) than to other samples from the Atlantic Forest biome . It is worth mentioning that , in general , samples from Cerrado , Caatinga and Amazon biomes were grouped together in two different groups ( Fig 5 , red and pink coloured groups ) . The DAPC analysis with the 92 strains yields five genetic clusters , evidenced by a slight ‘elbow’ in the distribution of the BIC values across optimal cluster numbers at K = 5 , once 22 principal components ( PCs ) were retained and analyzed ( representing 80% of the total variation ) ( S18 Fig ) . DAPC-derived populations were broadly congruent with patterns of nuclear clustering identified by NJ and Bayesian clustering analysis . The five DAPC clusters , showed in S1 Table , corresponded to: Population 1 that includes Caatinga ( n = 13 ) and Cerrado ( n = 4 ) ; population 2 , Atlantic Forest ( n = 4 ) , Pantanal ( n = 10 ) and Cerrado ( n = 1 ) ; Population 3 , Amazon ( n = 30 ) , Atlantic Forest ( n = 6 ) , Pantanal ( n = 1 ) and Caatinga ( n = 1 ) ; population 4 , Atlantic Forest ( n = 14 ) and Cerrado ( n = 3 ) and population 5 , the remaining parasites principally from opossums and primates in the Atlantic Forest ( n = 14 ) and bats and an opossum in Cerrado ( n = 3 ) . Similarly , the NJ tree ( Fig 6 ) reveals that parasites from the Atlantic Forest , Cerrado and Pantanal were generally admixed together . We observe no strict specific association between biomes , species or collection years and the clusters based on DAPC; however , the chi square contingency test ( p<0 . 05 ) can not completely exclude an association between these clusters and host/vector species , collection biome or dates . Cluster A , derived from MLST data , was congruent with one MLMT cluster , the equivalent maxicircle cluster ( S17 and S19 Figs paired trees ) . The isolate G41 ( Atlantic Forest ) , grouped with isolates from Amazonia for MLST but was grouped with Atlantic Forest isolates with MLMT analysis . Similarly , topological positions for FRN26/26 were different for MLST and maxcircle trees ( S17 Fig paired trees ) Population genetic parameters were calculated for strains grouped a priori according to their biome of origin , as well as a posteriori DAPC cluster assignments ( Table 2 and S8 Table ) . Consistent results are observed when strains are grouped according to DAPC-assigned clusters . Table 2 shows high levels of genetic heterogeneity in Amazon ( DAPC population 3 ) , as well as excess homozygosity , high numbers of private alleles per locus and a low standard deviation associated with DAS value . T . cruzi strains from Atlantic Forest , Cerrado and Caatinga displayed similar , but lower levels of diversity , with comparatively lower numbers of private alleles per locus . Three diverse populations ( Atlantic Forest , Cerrado and Caatinga ) were characterized by elevated standard deviations associated with DAS values and positive FIS values ( Table 2 ) . A hierarchical AMOVA demonstrated 83 . 1% of total genetic variation was present within populations , compared to 16 . 9% , among populations ( p<0 . 0001 for both ) . The observed subdivision between a priori populations suggests the existence of gene flow between T . cruzi from the Atlantic Forest and those of the Cerrado biome ( FST = 0 . 067 ) ( Table 3 ) . Gene flow was also inferred to have occurred between T . cruzi populations of Caatinga and Cerrado ( FST = 0 . 0982 ) ( Table 3 ) . The admixed character of these isolates was also supported by Bayesian assignment . More geographically-distant TcI populations display similar levels of subdivision , as observed between Caatinga and Pantanal , Caatinga and Atlantic Forest , Cerrado and Pantanal and Pantanal and the Atlantic Forest . T . cruzi isolates from the Amazon biome exhibited lower FST values than populations of all other biomes ( Table 3 ) . This observation is also supported by FST values calculated for the a posteriori populations ( S9 Table ) . At a local-level structure analysis ( i . e . when samples were grouped using the collection site as prior information; Table 4 ) , it is clear that some isolates from Atlantic Forest grouped with others from Cerrado and Pantanal due to the genetic similarity between samples of Rio de Janeiro and Possas , Goias ( FST = 0 . 04 ) , of Bahia and Tocantins ( FST = 0 . 13 ) , and of Santa Catarina and Mato Grosso do Sul ( FST = 0 . 09 ) . Similarly , samples from Cerrado ( Piaui ) and Amazon ( Para ) showed low levels of structure ( FST = 0 . 09 ) . The investigation of parasite population structure based on host taxonomy suggests that Didelphis marsupialis might play a role as the main disperser of TcI ( S10 Table ) , since its overall pairwise FST values were lower than the others ( FST ≤ 0 . 2; median = 0 . 15 ) . Finally , to determine the extent of spatial genetic structure , a Mantel test was conducted , demonstrating significant parasite isolation by distance across the sampled geographical range ( RXY = 0 . 384; P = 0 . 01 ) . ( S20 Fig ) .
The BPP values supporting those clusters that show incongruences varied widely between individual gene phylogenies . Similar patterns of incongruence have been previously observed in nuclear genes [37 , 87] . Such incongruence , where isolates differ in topological positions , are a classical marker in populations that have undergone genetic exchange . To investigate further , haplotypic phylogenies were constructed for each genetic locus in order to define heterozygous isolates and their potential homozygous allelic donors . ( Fig 3 , S12–S16 Figs ) The results indicate potential allelic recombinants in 3 of the 6 loci . Putative recombinant isolates possessed heterozygous allelic profiles , each present in two different homozygous putative donor isolates , situated in different phylogenetic clusters . Potential allelic recombinant isolates across 3 genes is suggestive of multiple genetic exchange events . PHASE is a Bayesian method for the reconstruction of haplotypes . It is generally considered one of the most accurate haplotype reconstruction methodologies . However , there are potential confounders , for example , population size and frequency of recombination have the potential to skew outcomes . Furthermore , one must be cautious when using PHASE to infer frequency of genetic exchange , as this is one of the assumptions of the method . Nevertheless , the presence of recombinants and potential “donor” genotypes inferred in three independent nuclear markers is confirmed by heterozygous and homozygous SNPs derived from nuclear sequences ( S4–S6 Tables ) . Together , these observations constitute evidence for the presence of genetic exchange at the nuclear level . Population structure of T . cruzi is frequently regarded as clonal [88] . This model does not exclude genetic exchange , but considers it to be infrequent [89] . However , exchange across DTUs has been demonstrated using MLST [18 , 36]; and intra-TcI genetic exchange in a single isolate has been observed in a cohort of Colombian samples [36] . Similarly , Messenger et al . [41] and Ramirez et al [82] observed multiple incongruence and introgression events within TcI on the basis of MLMT , MLST and maxicircle phylogenies , concluding that genetic exchange within DTU I is frequent . Genetic exchange is inferred in the current data set , however the frequency of genetic exchange is presently unknown and a topic of enthusiastic debate . In comparison with nuclear genes , remarkably low levels of intra DTU COII diversity were observed . Paradoxically , the mutation rate of mitochondrial genes is generally considered one order of magnitude higher than that of nuclear genes [90] . The spectrum of reduced diversity observed in maxicircle clades is consistent with introgression events as also reported in different TcI populations in South America [36 , 41] . Two major clades were observed , the first consisting of all samples from the Amazon biome , together with a few samples from other biomes . The second clade grouped all of the remaining samples . This pattern was congruent across both nuclear and mitochondrial loci , and is indicative of genetically discrete populations . MLMT analysis ( below ) suggests limited gene flow ( FST ) between the Amazon biome and other studied areas . Although nuclear and mitochondrial phylogenies shared some topological characteristics , there were substantial incongruences between them . For example , isolate G41 clustered with Amazonian isolates at the nuclear level , but associated in mitochondrial phylogenies with isolates of non-Amazonian origin ( S17 Fig ) . In the context of introgression , the discordance between nuclear and mitochondrial phylogenies is indicative of a prolonged and continuous association between populations from very distant localities [41] . This is consistent with the suggestion that genetic exchange in T . cruzi involves the independent exchange of kinetoplasts and nuclear genetic material [41] . Reciprocal nuclear genetic exchange among parasite strains undergoing mitochondrial introgression has not yet been detected , which may support an asymmetric , cryptic hybridization mechanism , or perhaps more likely , reflect the minor amount of nuclear genetic information sampled [81] . However , without the resolution of whole nuclear genome sequences , it is only possible to define the contributions of elements of meiosis , mitochondrial introgression and/or parasexual fusion [15 , 82 , 91] . The results presented here , include isolates from geographically distant sites ( approximately 1790 km ) and imply multiple introgression events occurring between different clades encompassing a large geographical area . MLMT , the most sensitive method for assessing diversity , identified 4 groups when collection sites were used to group TcI specimens ( Fig 5 ) or 5 clusters when no prior clustering was imposed . Three groups draw attention , one with isolates originating from Caatinga ( gray branch ) , another from Pantanal ( blue branch ) and a third , consisting of an admixture of Atlantic Forest and Cerrado ( Fig 6 , orange and red branch ) . The third group contained genotypes that occurred in primates , bats , Didelphis and Rhodnius spp . , in agreement with mitochondrial phylogenetic topology . There was a tendency for TcI isolates to cluster with other locally obtained isolates , which may reflect a sampling bias or clonal expansion . However , when samples were grouped according to their collection sites ( Table 4 ) , the analysis revealed specific examples of similar genotypes present across nearby states . Examples include Santa Catarina ( Atlantic Forest ) and Mato Grosso do Sul ( Pantanal ) , Bahia ( Atlantic Forest ) and Tocantins ( Cerrado ) , and Piaui ( Caatinga ) and Pará ( Amazon ) . In stark contrast , Amazon demonstrated significant intraspecific heterogeneity ( Table 2 and Table 3 ) and clustering indices suggest that parasites from Amazon ( DAPC population 3 ) have undergone long-term , undisturbed , sylvatic diversification . The relative lack of human impact , particularly in some municipalities in the state of Para , may account for allelic richness evolving over time in a biome with an abundance of host species . [39] . Interestingly , DAS values from three diverse populations ( Atlantic Forest , Cerrado and Caatinga ) suggest the presence of intra-population sub-clusters , which is likely a consequence of the fragmentation due to intense human activity in these areas . Significant gene flow is observed over vast distances , for example between Cerrado and Atlantic Forest ( Fig 1 and Table 3 ) . The most parsimonious explanation is host movement , particularly aerial dispersion with bats , as exemplified over large distances in African clades of Trypanosoma sp . [92] . In South America , bats are known to harbor diverse trypanosome genotypes [92 , 93] , but their role in biogeography and dispersion is not fully understood . Unfortunately , TcI samples from Chiroptera species were collected from a single location ( in Cerrado ) . A much more comprehensive effort to study Tc1 isolates in Chiroptera would be of interest to adequately address the nature of their role in dispersal in Brazil . Notwithstanding , we observed that D . marsupialis acts as a disperser of TcI genotypes across different biomes [94] , this is evidenced by generally low FST values in pairwise comparisons with samples obtained from other hosts ( FST ≤ 0 . 2 , S10 Table ) . Isolates from Atlantic Forest , Amazon and Cerrado showed significantly low heterozygosity levels , which may be due to gene conversion or under sampling used in the study . In this case , processes such as inbreeding are expected to shape the genetic background of populations [94] . Indeed , isolates from the Amazon biome presented low gene flow and moderate levels of inbreeding ( FIS = 0 . 194 ± 0 . 04 ) , relatively to other biomes , indicating a degree of genetic isolation . ( Table 2 ) . Our analyses of three classes of genetic markers revealed broadly similar patterns of intra-DTU diversity in Brazil . MLST and maxicircle marker analysis yielded two principal phylogenetic groups . One included all isolates from the Amazon region , with representatives from Cerrado and Atlantic Forest ( Fig 2 , clusters B and C ) . The second group included all other isolates from Atlantic Forest , Cerrado and part of Pantanal ( Fig 2 , cluster A ) . MLMT analysis comprising fast evolving markers , as expected , revealed the most diversity , five discrete populations and variable amounts of gene flow and fragmentation indicators . Among all biomes it is evident that Amazon harbors the most extensive diversity and comparatively low gene flow . High diversity and low fragmentation indicate a biome exposed to less ecological pressure and undisturbed sylvatic diversification . Generally , there was no clear evidence of specific host/vector associations . In particular , similar genotypes were represented in different vector/host species . For example , genotypes represented in cluster A ( Fig 2 ) consisted of closely related genotypes observed in a diversity of hosts species including didelphids , rodents , chiroptera , primates and triatomines scattered across diverse municipalities within the Atlantic Forest , Cerrado and Pantanal biomes . Additionally , this cluster included hosts whose habitat is principally arboreal , with Didelphis spp occupying all strata . The presence of Didelphis spp . in all clades and low associated FST values ( S10 Table ) is compatible with the hypothesis that they are bioaccumulators of multiple genotypes [83 , 94] , they are highly permissive to infection and are known to move between all ecological strata from terrestrial to arboreal . The genealogical relationship of isolates in cluster A in MLST was preserved across MLMT and mitochondrial analyses ( S17 and S19 Figs ) . Evidence from all markers reveals that similar genotypes are found across vast geographical distances , over ecological barriers , diverse habitats , and different hosts species . Noticeably , isolates G41 ( Atlantic Forest ) and 2896 , from Belem in the Amazon biome ( Figs 1 and 2 ) , have identical genotypes . Other examples include isolates 10272 and 11609 , which possess identical genotypes despite being separated by the Marajo bay ( a distance of 4500 km ) ; and isolates from Belem ( 2855 ) and Abaetetuba ( 11606 ) , which are genetically homogenous despite vast geographic separation . Human activity is likely to have an impact on the dispersal of genotypes . A case in point is provided by Combu and Murucutu , which are two island localities situated in the municipality of Belem ( Amazon ) that are sparsely occupied by humans and used primarily for açaí production [53] . They form a robust enzootic transmission cycle , and remote human infections are acquired by unwitting transport of infected triatomines in açai baskets [53] . Comparatively high indicators of gene flow between other biomes inferred by MLMT analysis are also compatible with the influence of human activity that may facilitate gene flow . Lima and collaborators [42] using MLMT , applied to Brazilian TcI , observed that isolates from Atlantic Forest and the Amazon formed distinct and separate clusters . Their proposition was that geographic distance separating biomes was the likely explanation for topological features . However , in this work , through the application of MLST , MLMT and maxicircle analysis , we find not only localized diversity but also genetic homogeneity over large distances . In summary , this study included a large number of samples and revealed extensive intra DTU diversity , an absence of strict associations to host/vector species , and similar genotypes circulating over vast areas . We provide evidence of genetic exchange based on phylogenetic incongruence among loci , haplotypic analysis of nuclear markers and also mitochondrial introgression . It is likely that gene flow between biomes is influenced by the movement of mammals and also facilitated by human activity . | T . cruzi is a zoonotic protozoan parasite infecting mammals and widely dispersed throughout endemic Latin America . It is known to possess considerable genetic diversity , comprising six discrete genetic lineages designated Discrete Typing Units ( DTUs ) TcI-TcVI . TcI is the most genetically diverse DTU and the most frequently sampled lineage in Brazil . We use a combination of high resolution molecular techniques to analyze the genetic diversity of Brazilian TcI isolates obtained from a wide geographical area encompassing five distinct biomes isolated from different mammal hosts and insect vectors . The results reveal significant genetic diversity and no clear association of genotypes with areas or host/vector species . Evidence from incongruent phylogenetic topologies based on nuclear and mitochondrial markers are indicative of genetic exchange and/or introgression events . The relevance of these findings in the context of population structure , ecology and epizootiology is discussed . | [
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion"
] | [
"taxonomy",
"biogeography",
"ecology",
"and",
"environmental",
"sciences",
"population",
"genetics",
"geographical",
"locations",
"biodiversity",
"phylogenetics",
"data",
"management",
"phylogenetic",
"analysis",
"population",
"biology",
"computer",
"and",
"information",
"sciences",
"geography",
"south",
"america",
"ecosystems",
"phylogeography",
"evolutionary",
"systematics",
"genetic",
"loci",
"brazil",
"people",
"and",
"places",
"ecology",
"forests",
"earth",
"sciences",
"genetics",
"biology",
"and",
"life",
"sciences",
"evolutionary",
"biology",
"terrestrial",
"environments"
] | 2018 | Dissecting the phyloepidemiology of Trypanosoma cruzi I (TcI) in Brazil by the use of high resolution genetic markers |
Norovirus is the leading cause of gastroenteritis worldwide . Despite recent developments in norovirus propagation in cell culture , these viruses are still challenging to grow routinely . Moreover , little is known on how norovirus infects the host cells , except that histo-blood group antigens ( HBGAs ) are important binding factors for infection and cell entry . Antibodies that bind at the HBGA pocket and block attachment to HBGAs are believed to neutralize the virus . However , additional neutralization epitopes elsewhere on the capsid likely exist and impeding the intrinsic structural dynamics of the capsid could be equally important . In the current study , we investigated a panel of Nanobodies in order to probe functional epitopes that could trigger capsid rearrangement and/ or interfere with HBGA binding interactions . The precise binding sites of six Nanobodies ( Nano-4 , Nano-14 , Nano-26 , Nano-27 , Nano-32 , and Nano-42 ) were identified using X-ray crystallography . We showed that these Nanobodies bound on the top , side , and bottom of the norovirus protruding domain . The impact of Nanobody binding on norovirus capsid morphology was analyzed using electron microscopy and dynamic light scattering . We discovered that distinct Nanobody epitopes were associated with varied changes in particle structural integrity and assembly . Interestingly , certain Nanobody-induced capsid morphological changes lead to the capsid protein degradation and viral RNA exposure . Moreover , Nanobodies employed multiple inhibition mechanisms to prevent norovirus attachment to HBGAs , which included steric obstruction ( Nano-14 ) , allosteric interference ( Nano-32 ) , and violation of normal capsid morphology ( Nano-26 and Nano-85 ) . Finally , we showed that two Nanobodies ( Nano-26 and Nano-85 ) not only compromised capsid integrity and inhibited VLPs attachment to HBGAs , but also recognized a broad panel of norovirus genotypes with high affinities . Consequently , Nano-26 and Nano-85 have a great potential to function as novel therapeutic agents against human noroviruses .
Human norovirus is recognized as the most important cause of outbreaks of acute gastroenteritis [1] . The virus is a non-enveloped single-stranded RNA virus within the Caliciviridae family . The human norovirus genome contains three open reading frames ( ORFs ) , where ORF1 encodes non-structural proteins , ORF2 encodes the capsid protein ( VP1 ) , and ORF3 encodes the minor capsid protein ( VP2 ) . The virion comprises of 90 VP1 dimers that form an icosahedral particle ( T = 3 ) 35–40 nm in diameter [2 , 3] . The VP1 can be expressed in insect cells and self-assembles into virus-like particles ( VLPs ) morphologically similar to native virions [4] . Smaller icosahedral particles ( 15–25 nm , T = 1 ) , presumably composed of 30 VP1 dimers , can also self-assemble in insect cells and were found in patient stool specimens [5 , 6] . The X-ray crystal structure of norovirus native-size VLPs showed that the VP1 can be divided into shell ( S ) and protruding ( P ) domains that are connected via a flexible hinge [3] . The S domain forms the scaffold of the capsid , while the surface exposed P domains contain the main determinants of antigenicity and host binding epitopes . Noroviruses are genetically diverse and can be divided into seven genogroups ( GI-GVII ) that are further subdivided into numerous genotypes [7] . The GII genotype 4 ( GII . 4 ) includes most epidemic and pandemic strains , while GII . 17 was recently attributed with major outbreaks in East Asia [8] . Norovirus illness is typically self-limiting and usually subsides in several days . However , chronic infections in vulnerable individuals , such as the young and elderly , can lead to additional complications and even death [9–11] . Currently there are no available vaccines or antiviral treatments for human noroviruses , despite their discovery over four decades ago [2] . Recently , two cell culture systems have shown that human norovirus can replicate in B-cells or stem cell-derived human enteriods [12 , 13] . However , norovirus pathogenesis is still poorly understood and the interaction with the host receptor ( s ) is unclear . Nevertheless , histo-blood group antigens ( HBGAs ) have been shown to be important binding factors for human norovirus infections [14–17] . HBGAs are found as soluble antigens in saliva and are expressed on epithelial cells , which suggest that noroviruses may encounter HBGAs several times during the course of the infection . Soluble HBGAs may interact with virion particles prior attachment to cells [13] or function as binding factors on cell surfaces [12] . Until recently , the norovirus capsid was presumed to bind two HBGA molecules per VP1 dimer , however two additional HBGA binding sites were identified on the VP1 dimer , indicating that the interaction with HBGAs is rather complex [18 , 19] . Interestingly , the presence of serum antibodies that block norovirus binding to HBGAs has been associated with a decreased risk of infection and illness [12 , 20 , 21] . Moreover , in a recent enteroid norovirus replication system inhibition in the blocking assay was correlated with neutralization in cell culture . A recent study suggested that antibodies targeting the HBGA pocket could inhibit norovirus replication by steric interference with the GI . 1 HBGA pocket [22] . A number of other studies have identified norovirus-specific monoclonal antibodies ( mAbs ) and single chain variable domains ( VHH or Nanobodies ) that could block norovirus VLP binding to HBGAs [20 , 23–27] . However , most of these antibodies and Nanobodies are genotype specific , which limits their therapeutic potential [28] . Apart from the HBGA binding site , other neutralizing epitopes likely exist . For example , upon binding to cell receptors , picornaviruses , which are structurally similar to noroviruses , initiate multiple structural rearrangements from the mature capsid to expanded intermediate forms , leading to externalization of the internal polypeptide , membrane fusion and release of viral RNA [29] . Neutralizing Nanobodies that interfere with the conformational rearrangement of the capsid were recently reported for poliovirus [30] . In that study , Nanobodies were used to trap transitional conformations of the viral capsid , which occur during cell entry and are required for the receptor binding . There is still limited information on norovirus particle attachment to cell surfaces and rearrangements during cell entry . Defining the structural dynamics of the norovirus particles during an infection could show transient conformations related to specific functions in the virus life cycle . These snapshots of the particle dynamics could be obtained from the reconstruction of the capsid protein complexes with antibody fragments or Nanobodies . Moreover , the structural analysis could offer insights into vulnerable regions on the capsid that could be targeted by inhibitors . Indeed , we recently discovered that a human norovirus specific Nanobody ( termed Nano-85 ) bound to intact norovirus VLPs and the Nanobody binding interaction caused the VLPs to disassemble [31] . Our results suggested that the Nano-85 binding epitope might represent a vulnerable region on the capsid that is important for the structural integrity . In the current study , we analyzed a novel panel of norovirus-specific Nanobodies in order to identify other vulnerable regions . The Nanobody binding epitopes were determined using X-ray crystallography and the specific binding interactions were correlated with a surrogate neutralization assay . We found that Nanobody binding could trigger capsid deformation and increase proteolytic degradation of capsid protein , ultimately exposing viral RNA . Our new findings showed that norovirus particles have vulnerable epitopes that were indispensable for capsid assembly , structural integrity and HBGA attachment .
The Nanobody binding specificities were initially confirmed with the immunization antigen ( i . e . , GII . 10 VLPs ) and the corresponding GII . 10 P domain ( S1 Fig ) . All six Nanobodies bound to the GII . 10 VLPs and P domain , which indicated that the S domain did not contain any Nanobody binding epitopes . Nano-42 , Nano-14 , Nano-26 , and Nano-4 showed the strongest binding capacities ( 0 . 02–0 . 2 μg/ml ) , whereas Nano-27 and Nano-32 had lower binding ability ( 1 . 5 and 0 . 2 μg/ml , respectively ) . Following these results , the cross-reactivities were analyzed with a panel of VLPs and P domains from various GI ( GI . 1 and GI . 11 ) and GII ( GII . 1 , GII . 2 , GII . 4 2006 and 2012 , GII . 10 , GII . 12 , and GII . 17 ) genotypes ( Fig 1 ) . Nano-85 exhibited the broadest recognition range , detecting GI . 11 VLPs and numerous GII P domains . Nano-26 also showed broad reactivity , detecting all GII genotypes . Nano-4 and Nano-42 showed limited cross-reactivity , while Nano-27 , Nano-32 , and Nano-14 were GII . 10 specific ( S1 Fig ) . In order to determine the HBGA blocking potential of the Nanobodies , a surrogate neutralization assays were performed using GII . 10 and GII . 4 VLPs . Three Nanobodies ( Nano-14 , Nano-32 , and Nano-26 ) inhibited the binding of GII . 10 VLPs to PGM in a dose-dependent manner ( IC50 = 1 . 7 to 6 . 6 μg/ml ) ( Fig 1E ) . Similarly , Nano-14 , Nano-26 , and Nano-32 inhibited binding to A-type saliva ( IC50 = 0 . 3 to 3 . 1 μg/ml ) and B-type saliva ( IC50 = 1 . 1 to 4 . 3 μg/ml ) ( S2A and S2B Fig ) . Nano-85 was relatively ineffective in blocking the GII . 10 VLPs to PGM or B-type saliva ( IC50 > 70 μg/ml ) and weakly blocked GII . 10 VLPs to A-type saliva ( IC50 = 12 μg/ml ) . Nano-4 , Nano-25 , Nano-27 , and Nano-42 demonstrated no inhibition of GII . 10 VLPs . Additionally , both Nano-26 and Nano-85 blocked GII . 4 VLPs from binding to PGM ( Fig 1F ) ( IC50 2 . 4 μg/ml and 3 . 1 μg/ml , respectively ) and B-type saliva ( IC50 0 . 7 μg/ml and 1 . 2 μg/ml , respectively ) ( S2C Fig ) . To demonstrate that Nano-26 and Nano-85 specifically inhibit VLP binding to HBGAs present in PGM and saliva , a blocking assay using synthetic HBGAs was performed ( S2D Fig ) . Nano-26 and Nano-85 blocked GII . 4 VLPs from binding to synthetic B-tri saccharide with IC50 ranging between 1 μg/ml to 10 μg/ml . Nano-4 and Nano-42 did not inhibit GII . 4 VLPs from binding to PGM . The thermodynamic properties of Nanobodies binding to GII . 10 P domains were analyzed using ITC ( Table 1 ) . Most of the Nanobodies ( Nano-4 , Nano-14 , Nano-26 , Nano-27 , and Nano-42 ) exhibited exothermic binding with nanomolar affinities ( S3 Fig ) . The binding reaction was driven by a large enthalpy change and was characterized with unfavorable entropy of the binding . This suggested that the net formation of non-covalent bonds between the Nanobody and the P domain was a major contributor to the binding . The stoichiometry indicated the binding of one Nanobody molecule per P domain monomer for all Nanobodies , except Nano-14 , where the ratio of P domain:Nanobody was 2:1 . Nano-32 binding was characterized by a positive enthalpy change associated with endothermic type of reaction ( S3 Fig ) . Instead , a large positive entropy was the main contributing factor to the ΔG . These different thermodynamic parameters were likely associated with the distinct binding epitope of Nano-32 , as presented below . The structures of GII . 10 P domain in complex with Nano-14 , Nano-26 , Nano-27 , Nano-32 , and Nano-42 were solved using X-ray crystallography ( Table 2 ) . Additionally , the X-ray crystal structure of GII . 17 P domain with Nano-4 was determined in order to explain Nanobody cross-reactivity binding interactions at the atomic level . We also solved a double complex structure of GII . 10 P domain with Nano-26/Nano-85 , which permitted a higher resolution than the GII . 10 P domain and Nano-26 complex alone , and explained how two distinct Nanobodies bound simultaneously to one P dimer . The overall structure of the P domains in all complex structures was reminiscent of unbound P domain with limited structural changes observed upon binding of the Nanobodies . All Nanobodies had typical immunoglobulin fold and interacted with the P domain primarily with CDR loops . The electron densities for Nano-4 , Nano-32 , Nano-26 , Nano-27 , and Nano-42 were well resolved , whereas for Nano-14 , the distant part of a Nanobody close to the two-fold crystallographic symmetry axis was partially disordered . Overall , we could separate six Nanobodies into three distinct binding regions on the P domain: termed top , side , and bottom . Nano-4 , Nano-26 , Nano-27 , and Nano-42 bound to the bottom; Nano-32 bound to the side; and Nano-14 bound on the top of the P domain ( Fig 2 and Table 3 ) . The binding sites of Nano-4 and Nano-27 partially overlapped the Nano-85 binding site . Moreover , Nano-42 bound with almost identical orientation as Nano-85 . To support our structural data and exclude the possibility of less probable orientations derived from the crystal packing we performed competitive ITC measurements . Nano-85 showed no binding to the P domain pre-incubated with Nano-4 , Nano-27 and Nano-42 , indicating that these Nanobodies competed for the same binding region on the P domain ( S4 Fig ) . On the contrary , when Nano-85 titrations were performed to the P domain pre-mixed with Nano-14 or Nano-26 the binding isotherm was reminiscent of the Nano-85 P domain measurement . These data implied that Nano-14 and Nano-26 bound to sites distinct from Nano-85 , whereas other Nanobodies competed with the Nano-85 epitope . Therefore , these P domain Nanobody complex structures clearly represented the precise Nanobody binding epitopes . The structure of GII . 10 P domain Nano-14 complex was solved to 1 . 8Å resolution . Nano-14 bound on the top of the P domain in the grove located between the two P domain monomers ( Fig 3A and 3B ) . A vast network of hydrogen bonds was formed between Nano-14 and both P domain monomers . The majority of interactions were built between one P domain monomer ( chain A ) and CDR3 of the Nano-14 ( Fig 3C ) . Six P domain residues ( chain A: Arg299 , Trp381 , Lys449 , Asp403 , and Glu333; chain B: Gln384 ) formed eleven direct hydrogen bonds with Nano-14 . Four electrostatic interactions were observed between Nano-14 and the P domain residues Arg299 and Glu382 ( chain A ) . The numerous hydrogen bonds and electrostatic interactions corresponded well with the large negative binding enthalpy ( S3 Fig ) . Five P domain residues ( His298 , Val361 , Ala363 , Arg299 , and Trp381 ) were involved in eight hydrophobic interactions with Nano-14 . Two additional interactions were observed: P domain His358 ( chain B ) formed a π-sulfur interaction with Nanobody Met106 , whereas P domain Glu333 ( chain A ) participated in a π donor hydrogen bond with Phe102 . The nine P domain residues involved in Nano-14 binding were predominantly variable ( see Fig 4 ) . This finding corresponded nicely with the ELISA data that showed Nano-14 was GII . 10 specific . Remarkably , the Nano-14 binding site , which is largely formed by CDR3 loop , extended between two HBGA binding pockets ( Fig 2 ) . Such strategic positioning of Nano-14 resulted in steric interference with the two conventional HBGA binding sites and the two newly identified HBGA binding pockets [18] . Moreover , three P domain residues ( Trp381 , Glu382 and Lys449 ) were directly involved in binding HBGAs [32] and Nano-14 , indicating a direct competition for the HBGA pocket . Importantly , analysis of the Nano-14 binding site with the ELISA blocking data provided a novel structural basis of GII HBGA binding interference ( see Fig 1 ) . The Nano-32 binding site was located on the side of the GII . 10 P domain in a cleft between two P domain monomers ( Fig 5A and 5B ) . In the P domain Nano-32 complex , several P domain loops were slightly shifted compared to the unliganded P domain ( S5 Fig ) ( chain A: residues 487–491 and 517–522; chain B: residues 309–314 , 287–300 , and 418–421 ) . Moreover , a P domain loop ( residues 343–352 ) was shifted ~4 . 3Å from the loop in the unliganded structure . Several residues within this loop were also disordered , suggesting a certain degree of P domain flexibility . The loop containing residues 295–300 was positioned identically in both monomers in contrast to the usual asymmetric orientation in unliganded structure [18] . These conformational rearrangements likely correlated with the major entropy change observed in ITC measurements ( Table 1 ) . Nano-32 was essentially held equally with two P domain monomers ( Fig 5C ) . Four P domain residues from chain A ( Arg287 , Asn344 , Trp343 , and Asp316 ) and two residues from chain B ( Arg492 and Thr519 ) formed seven direct hydrogen bounds with Nano-32 . Several P domain residues were also involved in electrostatic interactions ( chain A: Arg287 and Asp247; chain B: Glu236 ) and hydrophobic interactions ( chain A: Pro314; chain B: Val248 and Pro518 ) . Six P domain residues involved in Nano-32 binding were highly variable and five residues were conserved in GII . 4 and GII . 10 noroviruses ( Fig 4 ) . Although Nano-32 strongly inhibited binding of GII . 10 VLPs to HBGAs , none of the residues were shared between the HBGA pockets and the Nano-32 binding site . This result suggested that Nano-32 indirectly interfered with the HBGA pockets or utilized another mechanism to inhibit HBGA binding . We solved the structure of GII . 17 P domain Nano-4 complex , since the GII . 17 norovirus was of recent clinical concern; and we wanted to analyze the cross-reactive epitopes at the atomic level . According to the ELISA data , Nano-4 bound strongly to the GII . 17 VLPs . X-ray data for GII . 17 P domain Nano-4 complex was processed to 1 . 7Å resolution in C121 space group . Nano-4 bound to the bottom of the P domain in close proximity to the previously identified Nano-85 binding site ( Fig 2 and Fig 6A and 6B ) [31] . An extensive network of direct hydrogen bonds was formed between P domain residues ( Thr483 , Glu486 , Asp516 , Asn520 , Tyr523 , and Ser524 ) and Nano-4 ( Fig 6B ) . Two P domain residues were involved in hydrophobic interactions ( Tyr523 and Ala526 ) and five electrostatic interactions ( Arg482 , Glu486 , and Asp516 ) contributed to Nano-4 binding . Only three of nine P domain residues interacting with Nano-4 were variable ( Fig 4 ) . The six conserved residues provided a possible explanation for the broad cross-reactivity exhibited with Nano-4 ( Fig 1 ) . The Nano-4 binding epitope was located on the opposite side of the HBGA pocket , an observation that is supported by the lack of blocking potential in the surrogate neutralization assay . Nano-42 bound on the bottom of the P domain and closely overlapped with Nano-85 binding site ( Fig 6C ) . Five direct hydrogen bonds involved P domain residues ( Asp526 , Trp528 , Asn530 , and Thr534 ) and Nano-42 residues ( Fig 6D ) . Two hydrophobic interactions were formed between P domain residues Val529 and Ala536 and Nano-42 residues Tyr100 and Val54 , respectively . Interestingly , the Nanobody was also held by three additional hydrogen bonds mediated by an ethylene glycol molecule . Ethylene glycol interacted with P domain residues Arg484 and Asp526 on one side and Nano-42 residues Thr31 and Ser53 on the other side . Moreover , six water mediated bonds provided additional stabilization of the bound Nano-42 . Although Nano-42 binding residues were mainly conserved in GII noroviruses and were identical between GII . 4 2006 and 2012 strains , Nano-42 apparently distinguished these two strains in the ELISA cross-reactivity study ( Fig 1 ) . In addition , although the binding epitope of Nano-42 was rather similar to that of Nano-85 , Nano-42 did not inhibit VLP binding to HBGAs . Nanobodies were previously shown to aid the crystallization process by increasing protein stability and stabilizing flexible regions [33] . We have already utilized Nano-85 to obtain high-resolution complex structures with three different norovirus P domains [31] . Herein , we used Nano-85 to improve the resolution of the GII . 10 P domain Nano-26 complex structure and describe the synchronized binding of two Nanobodies . The initial structure of GII . 10 P domain Nano-26 complex was solved to ~3Å resolution . A single crystal of GII . 10 P domain Nano-85/Nano-26 double complex diffracted to 2 . 3Å in C121 space group . Binding epitopes and interactions of both Nanobodies were identical to those in the individual complexes [31] . Nano-26 bound at the bottom of the P domain , perpendicular to Nano-85 binding site ( Fig 7A ) . Nano-26 binding site comprised of residues from both P domain monomers , although the majority of the P domain interactions involved only one chain ( chain B ) . Nano-26 formed seven direct hydrogen bonds with one P domain monomer ( chain B: Asp269 , Leu272 , Gly274 , Gln471 , Glu472 , and Thr276 ) ( Fig 7B ) . Both P domain monomers were involved in hydrophobic interactions ( chain A: Ile231 , Pro488; and chain B: Tyr470 and Pro475 ) with Nano-26 . In addition , two electrostatic interactions contributed to the tight binding . Nano-26 binding residues were mainly conserved between GII genotypes , which correlated well with the broad recognition shown with ELISA ( Figs 1 and 4 ) . Although the binding site was distant from the HBGA binding pocket , Nano-26 had a high inhibition capacity in the blocking assay , which also suggested indirect HBGA interference . The Nano-27 binding epitope was located on the bottom region of the P domain ( Fig 8 ) . Interestingly , the binding site partially overlapped the Nano-4 binding site . Six P domain residues ( Arg484 , Gly491 , Arg492 , Thr493 , Glu496 , and Thr534 ) were involved in ten direct hydrogen bonds and two electrostatic interactions . Three residues ( Arg484 , Ala536 , and Pro537 ) were involved in four hydrophobic interactions with Nano-27 . The Nano-27 binding site comprised six conserved residues and two variable residues . The ELISA data showed that Nano-27 was strain specific , which indicated that certain variable residues likely play a crucial role in cross-reactivity ( Fig 4 ) . Similarly to Nano-4 , Nano-27 also failed to block VLP binding to HBGAs . We previously showed that Nano-85 was able to disassemble norovirus VLPs [31] . To explore if these six newly identified Nanobodies had a similar ability , we treated native-size VLPs with Nanobodies and examined the treated-particle morphology using EM . Overall , three distinct VLP structural modifications were observed with Nanobody treatment ( Fig 9 ) . In the first case , Nano-85 and Nano-26 treatment partially disassembled and deformed the native-size VLPs . Nano-85 treatment also produced a minor fraction of small-size VLPs ( 20–23 nm ) . In the second case , Nano-4 , and Nano-27 treatment induced a conformational transition from native-size VLPs ( 35–38 nm ) to the small-size VLPs . In case of Nano-42 , small and disassembled particles were equally present after treatment . In the third case , Nano-32 treatment produced large aggregates of apparently intact native-size VLPs . None of these effects were observed with Nano-14 treatment . To investigate a temperature dependence of the Nanobody treatment , we mixed GII . 10 VLPs with Nano-85 and Nano-26 at 4°C , room temperature , and 37°C for 30 minutes ( S6 Fig ) . Nano-85 treated VLPs showed a continuous degradation of native-size particles , producing small and/or partially broken particles as major intermediate forms . Nano-26 was more effective across the temperature range and almost completely altered the VLP integrity . The combination of Nano-85 and Nano-26 appeared to cause a more intense degradation of VLPs . The temperature dependence of Nano-85 induced morphological changes indicated the involvement of capsid “breathing” in the disassembly process . We also performed DLS measurements to quantitatively evaluate GII . 10 VLP heterogeneity after Nanobody exposure . Nano-14 treated VLPs had almost identical diameters to native-size particles ( 37 nm and 35 nm , respectively ) ( Fig 10A and 10B ) . Nano-32 treated VLPs displayed 10 , 000 times increased diameters , confirming the formation of the large aggregates observed using EM . Nano-26 and Nano-85 treated VLPs mainly formed VP1 protein aggregates , although a small peak corresponding to native-size particles remained . Nano-4 , Nano-27 , and Nano-42 treated VLPs showed peaks corresponding to small-size VLPs ( 21–23 nm ) . Overall , the DLS analysis corresponded well with the EM results and provided additional evidence that Nanobody treatment altered the capsid structural integrity . Two Nanobodies , Nano-26 and Nano-85 , exhibited broad cross-reactivities coupled with adverse effects on capsid integrity . To understand if these effects were relevant for clinically important norovirus strains , GII . 4 ( Sydney 2012 ) and GII . 17 VLPs were treated with Nano-26 and Nano-85 ( Fig 11 ) . Both Nanobodies lead to malformed and aggregated GII . 4 VLPs and produced only a few small-size particles . In the case of GII . 17 VLPs , Nano-26 treatment caused the formation of small-size VLPs , whereas Nano-85 seemed to have no notable effect on the particle size . These EM results were supported with the DLS measurements ( Fig 11 ) . Overall , these results suggested that effects of the Nanobody treatment might vary among different genotypes . To further evaluate the binding effects of Nano-26 and Nano-85 on GII . 4 ( 2012 ) VLPs , we performed time- , temperature- , and concentration-dependent DLS measurements ( S7 and S8 Figs ) . Nano-26 induced changes in the VLP size distribution after 30 seconds , whereas for Nano-85 15 minutes were required to observe the first noticeable effects ( S7 Fig ) . Fluctuations in VLP sizes were more evident at 37°C for both Nano-26 and Nano-85 after 15 minutes incubation ( S7A and S7B Fig ) . Nanobody effects were also concentration dependent , with minimum concentrations of 12 . 5 μM and 50 μM required for Nano-26 and Nano-85 , respectively ( S7C and S7D Fig ) . These results suggested that one Nano-26 molecule per VP1 dimer was sufficient to cause morphological changes , whereas Nano-85 required >2 times molar excess . In order to examine the Nanobody effects on norovirus virions , we implemented a modified RNA exposure assay and viral loads were quantified using real-time RT-PCR . Concentrated GII . 4 positive stool samples were treated with the broadly reactive Nano-26 and Nano-85 , while Nano-14 was used as a negative control and 250 mM citric buffer was used as a positive control . Treated samples were then subjected to RNAse digestion . Nano-26 , Nano-85 , and citrate treated stool samples showed reduced genome copy numbers compared to the Nano-14 control ( approx . 30 times for Nano-26 and Nano-85 and 250 times for citrate ) ( Fig 10C ) . These results suggested that the Nano-26 and Nano-85 opened the virions and released the viral RNA , which was degraded by RNAse . To evaluate the Nanobody effects on norovirus virions more directly , we used a stool sample where RNA degradation was not detected and performed RNA extraction with incomplete lysis step ( Fig 10D ) . Additional degradation caused by Nanobodies or citrate lead to an increased number of genome copies compared to untreated samples . Indeed , Nano-26 , Nano-85 , and citrate treated samples had higher RNA levels than in the control samples ( PBS or Nano-14 ) ( Fig 10D ) . Although , the fold increase was relatively small ( 5–7 times ) , the difference was significant . To further investigate if Nanobody treatment could render norovirus VLPs vulnerable to proteolytic cleavage , we subjected GII . 10 , GII . 4 , and GII . 17 VLPs to a 30-minute trypsin digestion after Nanobody exposure and observed the protein bands using SDS-PAGE ( S10 Fig ) . Nano-14 treated VLPs produced similar bands as the untreated VLPs . Nano-26 and Nano-85 treatment resulted in multiple cleavage products for GII . 10 and GII . 4 VLPs . In the case of GII . 17 VLPs , only Nano-26 treatment showed additional cleavage of the capsid protein . Overall , these results suggested that Nano-85 and Nano-26 caused the particles to become structurally unstable , more vulnerable to proteolytic cleavage , and viral RNA exposure .
Structural information of antibody and Nanobody binding sites can be instrumental for understanding the neutralizing and immuno-dominant epitopes as well as motion dynamics of the viral capsid . Numerous neutralizing mAbs have been identified in recent years with diverse neutralization mechanisms [34] . One of the most direct mechanisms is blocking the receptor binding sites . Such neutralizing mAbs and Nanobodies were previously identified for influenza virus , HIV , herpes simplex virus , rhinovirus , and others [35–41] . For example , in the case of HIV , with the aid of an extra long CDR3 loop , the neutralizing Nanobody D7 effectively competed for the CD4 binding site on gp120 protein [42] . Previously described Nanobodies and mAbs with therapeutic potential against human norovirus were also proposed to interfere with the HBGA binding site [20 , 22–27] . MAb termed NV8812 bound to a conformational epitope on the GI . 1 P domain and blocked the binding of norovirus VLPs to human and animal cell lines [24] . Four α-GI mAbs isolated from chimpanzees challenged with norovirus blocked VLP binding to carbohydrates and inhibited hemagglutination , although their precise binding sites were not described [20] . Recently , a GI . 1 specific mAb was discovered that sterically hindered the HBGA pocket [22] . In our study , we showed that Nano-14 overlapped with the GII . 10 HBGA binding sites and inhibited HBGA binding by steric interference and competition for the pocket . The blocking abilities of Nano-14 were also comparable to previously reported blocking Nanobodies ( IC50 = 0 . 34–2 . 0 μg/ml ) [23] , scFv fragments ( IC50 = 0 . 3–1 . 5 μg/ml ) [43] , and mAbs ( IC50 = 0 . 12–0 . 74 ) μg/ml [20 , 28] . Although exhibiting high inhibition capacity , these mAbs and Nanobodies tend to be strain specific . The use of mAbs or Nanobodies directed to the HBGA pocket may inherently suffer from the variations and constantly changing amino acids in this region . Therefore , there is a need to identify additional neutralization epitopes , which are less susceptible to sequence variations . Indeed , Nano-32 recognition epitope was distant from the HBGA binding pocket and blocked VLP binding to HBGAs . A similar phenomenon was previously discussed with the norovirus specific blockade mAb NVB71 . 4 , where neither particle disassembly nor steric hindrance could explain NVB71 . 4 blockade activity [25] . However , it was suggested that the NERK motif ( residues 310 , 316 , 484 , and 493 according to GII . 4 numbering ) could function as a conformational regulator through an allosteric effect [25] . Interestingly , two of these residues were directly involved in Nano-32 binding , suggesting a similar blockage mechanism as observed with mAb NVB71 . 4 . Nano-32 induced conformational rearrangement of several P domain loops , which in turn altered the hydrophobic landscape of the P domain surface . This rearrangement likely caused the particle aggregation leading to interference at the HBGA binding pocket . An inhibition mechanism by allosteric interference was previously described for highly neutralizing mAbs against HIV and dengue virus [44 , 45] . Also , a recent study showed that the PGT121 mAbs against HIV gp121 protein inhibited CD4 binding , although the binding epitope was remote from the CD4 binding site . Moreover , dengue virus neutralizing mAb 1A1D-2 bound to a partially occluded epitope on envelope glycoprotein E and promoted particle reorganization [45] . These changes in viral surface were likely responsible for the inhibitory properties by this mAb . To allow structural rearrangement to occur during viral entry and uncoating , the viral capsid proteins need to be exceptionally dynamic . Internal plasticity and motions of the capsid proteins can allow access to buried regions , which often play an important role in the viral life cycle [46] . Indeed , multiple antibodies against picornaviruses and flaviviruses that bind to normally inaccessible sites on the viral capsid were shown to be highly neutralizing [47–52] . For rhinoviruses and polioviruses , buried regions of internal VP4 protein are transiently exposed due to the capsid “breathing” and are targeted by neutralizing mAbs . These cryptic epitopes are often very conserved and therefore provide cross-serotypic neutralization . We previously identified a broadly reactive norovirus mAb [53] and Nano-85 that bound to a conserved region that was occluded in the context of native-size particles [31] . Here , we identified four novel Nanobodies ( Nano-4 , Nano-26 , Nano-42 , and Nano-27 ) that bound to the similar internal and poorly accessible epitopes as Nano-85 . In comparison with Nano-85 , the binding sites of Nano-27 and Nano-4 were located closer to the P domain crown . In context of the complete particle , this position had fewer steric clashes with neighboring P domains . The Nano-26 epitope was located at the bottom of the P domain , albeit perpendicular to Nano-85 binding site . Although completely different , Nano-26 recognition epitope was also conserved and poorly accessible ( Fig 7 ) . The time- and temperature-dependence of the Nanobody-induced degradation suggested an important role of conformational mobility and capsid “breathing” in Nano-85 and Nano-26 binding to these hidden epitopes [46] . Due to their small size and high affinities , the rapid binding of the Nanobodies provided a means to trap transiently exposed regions , otherwise buried in the native state of the particles . Trapping the particles in a particular conformation or otherwise inhibiting capsid “breathing” is a common antiviral strategy shared by many neutralizing mAbs , Nanobodies , and drugs against HIV , flaviviruses , picornaviruses , influenza , and others [54–60] . For example , several neutralizing Nanobodies against poliovirus and respiratory syncytial virus were shown to specifically stabilize either the native or expanded conformation of capsid , preventing it from further rearrangement necessary for the infection process [30 , 61] . It is plausible that in the case of norovirus Nanobodies described here , binding resulted in a stabilization of the particular P domain conformation , thus reducing the mobility and influencing the position on the S domain . The interaction between S and P domains was previously shown to control the size and stability of the GI . 1 norovirus capsid [62] . Superposition of P domain Nano-26 complex on the cryo-EM VLP structure revealed an extensive clash with the S domain ( Fig 12 ) . Nano-26 binding likely disrupted normal S-P domain orientations , which consequently resulted in particle disassembly . Nano-26 required less time and concentration to achieve particle disassembly than Nano-85 . This observation suggested that the restriction of a normal S-P domain relationship had a more destabilizing effect than interference with P-P domain interactions . Of note , only Nano-26 was able to influence the morphology of GII . 17 VLPs , whereas GII . 17 VLPs tolerated Nano-85 binding . Apparently , Nano-26 binding stabilized the S-P domain conformation that was incompatible with the morphology of native-size GII . 17 particles , but supported the formation of small-size VLPs . Furthermore , three other Nanobodies , Nano-27 , Nano-4 and Nano-42 , drove a shift from native-size GII . 10 VLPs to a smaller-size form . Likely , these Nanobodies could selectively stabilize the A/B conformation of the P dimer . The inability of the A/B dimer to reassemble into C/C dimers could lead to the formation of small particles , where all dimers are identical and resemble A/B dimer for T = 3 capsids . Interference with the capsid motions and integrity provides one possible explanation for the blocking properties of both Nano-26 and Nano-85 in the surrogate neutralization assay . Nanobody binding caused the loss of normal VLP morphology and the treated VLPs showed a reduced signal in the blocking assays . Indeed , chemically disassembled VLPs showed no binding in a PGM assay ( S10 Fig ) . These observations support the assumption that Nano-85 and Nano-26 inhibited the binding of norovirus VLPs to HBGAs by compromising capsid morphology instead of directly competing for the HBGA pocket . Interestingly , Nano-42 , Nano-27 , and Nano-4 , which stimulate the formation of small-size particles , did not interfere with the attachment to the HBGAs . It was previously shown that small-size VLPs effectively bound to the surface of CaCo2 cells and competed with the native-size VLPs [24] . Apparently , the small-size VLPs that resulted from Nanobody exposure were equally able to bind HBGAs . Intriguingly , our structural data indicated that closely overlapping epitopes are responsible for distinct functions . A striking example is Nano-42 and Nano-85 , which despite having almost identical binding footprints , showed distinct binding and blocking properties . Nano-42 seemed to be less effective in disassembling the VLPs compared to Nano-85 . Similar observations were previously reported for 80S poliovirus specific Nanobodies , where despite identical binding sites , the structures of the expanded virus differed in each complex [48] . Likewise , although Nano-4 and Nano-27 shared five of eight binding residues , Nano-27 was strain specific , whereas Nano-4 was cross-reactive . Even though the epitopes closely overlap with Nano-85 binding site , these Nanobodies did not exhibit blocking properties . Analysis of GII . 10 P domain residues involved in Nano-27 , Nano-4 , and Nano-42 binding suggested that residues 484 , 491–493 , and 496 might constitute the molecular switch responsible for preferential assembly of small particles . Thus , additional high-resolution structural information could be instrumental in understanding epitope-function relationships by providing the exact location and interactions of the binding partners . This information might remain elusive when more general epitope mapping methods are used . In addition to identification of functional epitopes on the norovirus capsid , our data provided insights of Nanobody potential neutralization properties in context of infectious norovirus virions . Recently , it was shown that silver dihydrogen citrate exposure compromises GII . 4 VLPs integrity and facilitates viral RNA degradation [63] . Similarly , we showed that Nanobody-induced morphological changes of norovirus capsid resulted in exposure of viral RNA from the norovirus virions in clinical samples . The naked RNA was especially vulnerable to RNAse digestion and a similar RNA degradation assay was shown to greatly reduce the infectivity of murine norovirus [64] . In addition to exposing the viral RNA , Nanobodies increased the susceptibility of capsid protein to proteases , which are abundant in the gut . Although the exact role of proteolytic cleavage in the norovirus life cycle is largely unknown , cleaved capsid protein was shown to lose the ability to bind HBGA and maintain capsid assembly [65] . In summary , we identified several Nanobodies that impaired normal capsid motions , assembly , and integrity with subsequent release of viral RNA . Four Nanobodies blocked norovirus binding to cell attachment factors ( HBGAs ) , utilizing three distinct inhibition mechanisms: steric occlusion of the HBGA binding site , allosteric interference , and violation of normal capsid morphology . Therefore , Nanobodies could act as broad inhibitors in multiple stages of the norovirus life cycle . The Nanobody capacity to inhibit human norovirus infections in the recently developed cell culture needs to be further evaluated . Nevertheless , the extensive evidence that interference with viral capsid dynamics could impair normal functioning suggested that Nanobodies could become effective norovirus therapeutics in future .
The norovirus P domains , GI . 1 ( Norwalk virus , Genbank accession number M87661 ) , GI . 11 ( Akabane , EF547396 ) , GII . 1 ( Hawaii , U07611 ) , GII . 2 ( Snow Mountain , AY134748 ) , GII . 4 ( Sydney-2012 , JX459908 and Saga4 2006 , AB447457 ) , GII . 10 ( Vietnam026 , AF504671 ) , GII . 12 ( Hiro , AB044366 ) , and GII . 17 ( Kawasaki308 , LC037415 were expressed in E . coli , purified and stored in GFB ( 25mM Tris-HCl pH7 . 6 , 0 . 3M NaCl ) [66] . The full-length capsid genes , GI . 1 ( AY502016 . 1 ) , GI . 11 , GII . 1 , GII . 2 , GII . 4 , GII . 10 , GII . 12 , and GII . 17 , were expressed in insect cells using the baculovirus expression system and stored in PBS [67 , 68] . Norovirus specific Nanobodies were produced at VIB Nanobody service facility , Belgium as previously described [31] . Briefly , a single alpaca was injected with GII . 10 VLPs . A VHH library was constructed from isolated peripheral blood lymphocytes and screened for the presence of antigen-specific Nanobodies using phage display . Thirty-five Nanobodies were isolated and allocated to 17 distinct groups based on a sequence alignment . Six Nanobodies ( Nano-4 , Nano-14 , Nano-26 , Nano-32 , Nano-42 , Nano-27 , and Nano-8 ) that represented different groups were analyzed in this study . The Nanobody genes were cloned to pHEN6C vector , expressed in WK6 E . coli cells , purified and stored in PBS or GFB . Nanobody titers to norovirus P domains or VLPs were quantified with direct ELISA ( 17 ) . Briefly , microtiter plates were coated with 7 μg/ml of GII . 10 P domains or 2 μg/ml of GII . 10 VLPs . For cross-reaction experiments , 15 μg/ml P domain and 4 μg/ml VLPs were coated on ELISA plates . The VLPs or P domain were detected with serially diluted Nanobodies and HRP-conjugated mouse α-His-tag monoclonal antibody . Absorbance was measured at 490 nm ( OD490 ) and all experiments were performed in triplicate . Pig gastric mucin ( PGM ) and saliva blocking assays were performed as previously described [69] . Briefly , ELISA plates were coated with 10 μg/ml PGM ( Sigma , Germany ) or with saliva type A or B diluted in PBS 1:2000 . Nanobodies were two-fold serially diluted in PBS containing 2 . 5 μg/ml GII . 10 VLPs ( for PGM assay ) , 0 . 5 μg/ml GII . 10 VLPs ( for saliva assay ) or 0 . 5 μg/ml GII . 4 2006 VLPs ( both PGM and saliva assay ) and incubated for 1 h at RT . The VLPs-Nanobodies mixture was added to the plates and bound VLPs were detected with a α-GII . 10 or α-GII . 4 VLPs rabbit polyclonal antibody . For synthetic HBGA blocking assay , 10 μg/ml synthetic blood type B trisaccharide amine derivative ( Dextra , UK ) was coated on Pierce maleic anhydride activated plates ( Thermo Fisher Scientific ) overnight at 4C . Serially diluted Nanobodies were pre-incubated with 5 μg/ml GII . 4 VLPs for 1h at RT . Following steps were performed as above . The binding of VLPs-only was set as a reference value corresponding to a 100% binding . The half maximal inhibitory concentrations ( IC50 ) values for Nanobody inhibition were calculated using GraphPad Prism ( 6 . 0a ) . Isothermal Calorimetry ( ITC ) experiments were performed using an ITC-200 ( Malvern , UK ) . Samples were dialyzed into the identical buffer ( GFB or PBS ) and filtered prior titration experiments . Titrations were performed at 25°C by injecting consecutive ( 1–2 μl ) aliquots of Nanobodies ( 100–150 μM ) into P domain ( 10–20 μM ) with 150 second intervals . The binding data was corrected for the heat of dilution and fit to a one-site binding model to calculate the equilibrium binding constant , KA , and the binding parameters , N and ΔH . Binding sites were assumed to be identical . For the competitive ITC measurements , the P domain was mixed with Nano-4 , Nano-42 , and Nano-27 in a 1:1 molar ratio . Titrations with Nano-85 were then performed as above . P domain and Nanobody complexes were purified by size exclusion chromatography ( 39 ) . The P domain and Nanobody complexes were crystallized using the following conditions: GII . 10 P domain Nano-26/Nano-85 [0 . 1 M sodium citrate , 40% ( w/v ) PEG600]; GII . 17 P domain Nano-4 [0 . 2 M calcium acetate , 10% ( w/v ) PEG8000 , 0 . 1 M imidazole ( pH 6 . 5 ) ]; GII . 10 P domain Nano-42 [0 . 2 M potassium iodide , 20% ( w/v ) PEG3350]; GII . 10 P domain Nano-14 [0 . 1 M sodium citrate ( pH 5 . 5 ) , 20% ( w/v ) PEG3000]; GII . 10 P domain Nano-32 [0 . 2 M magnesium formate]; and GII . 10 P domain Nano-27 [2 M sodium chloride , 0 . 1 M sodium acetate] . Crystals were grown in a 1:1 mixture of the protein sample and mother liquor at 18°C . Prior to data collection , crystals were transferred to a cryoprotectant containing the mother liquor in 30% ethylene glycol , followed by flash freezing in liquid nitrogen . X-ray diffraction data were collected at the European Synchrotron Radiation Facility , France at beamline BM30 , ID30A , ID23-1 A and processed with XDS [70] . Structures were solved by molecular replacement in PHASER Phaser-MR [71] using GII . 10 P domain ( PDB ID 3ONU ) or GII . 17 P domain ( 5F4M ) and a Nano-85 ( 4X7D ) as search models . Structures were refined in multiple rounds of manual model building in COOT [72] and refined with PHENIX [73] . Alternative binding interfaces derived from the crystal packing were analyzed using an online server PDBePISA . The orientation of the Nanobody with the highest interface surface area and contact with CDRs was selected as the biologically relevant interface . Atomic coordinates were deposited to the Protein Data Bank ( PDB ) . The norovirus VLP morphology was analyzed using negative stain electron microscopy ( EM ) as previously described [31] . Nanobodies ( 1 mg/ml ) and VLPs ( 1 mg/ml ) were mixed in 1:1 ratio and incubated for 1 h at room temperature . Prior to loading on carbon coated EM grids , all samples were diluted 30 times with distilled water . Grids were washed two times with distilled water and stained with 1% uranyl acetate . The grids were examined on a Zeiss 910 electron microscope ( Zeiss , Oberhofen , Germany ) at 50 , 000-fold magnification . VLP diameter was measured with ImageJ software using calibrated pixel/nm scale bar . The hydrodynamic diameters of treated and untreated norovirus VLPs were measured using dynamic light scattering ( DLS ) on ZetaSizer Nano ( Malvern Instruments , UK ) . Samples were diluted 1:50 with PBS up to a final volume of 1 ml . Three × 12 measurement runs were performed with standard settings ( Refractive Index 1 . 331 , viscosity 0 . 89 , temperature 25°C ) . The average result was created with ZetaSizer software . In order to determine the effects of the Nanobodies on native virions , we collected GII . 4 positive stool samples from two individuals with acute norovirus infection [74] . A 10% ( w/v ) stool suspension was prepared in PBS and clarified by centrifugation at 10 , 000 × g for 10 min . First stool sample was concentrated by ultracentrifugation at 285 , 000 × g for 3 h at 4°C . Then , 70 μl of the supernatant were treated with 150 μl of each Nanobody ( 1 mg/ml ) for 30 min at room temperature . Samples were digested with 50 U of RNAse One ( Promega , Germany ) for 30 min at 37°C . After treatment total RNA was extracted with QIAamp Viral RNA extraction kit ( Qiagen , Hilden , Germany ) . One step RT-qPCR was performed with previously published GII . 4 primers NKP2F ( 5’-ATGTTYAGRTGGATGAGATTCTC-3’ ) , NK2R ( 5’-TCGACGCCATCTTCATTCAC-3’ ) and probe RING2-TP ( 5’-FAM-TGGGAG GGCGATCGCAATCT-TAMRA-3’ ) using qScript XLT One-Step RT-qPCR ToughMix ( Quantabio , USA ) . For incomplete lysis , samples were diluted twice with PBS prior to RNA extraction with shortened incubation time . cDNA was synthesized using High Capacity cDNA Reverse Transcription Kit ( Applied Biosystems , Foster City , USA ) . qPCR with melt curve analysis was performed using SYBR Green Master Mix ( Bio-Rad , Hercules , USA ) . GII . 4 specific primers , sense JJV2F ( 5’-CAAGAGTCAATGTTTAGGTGGATGAG-3’ ) and antisense COG2R ( 5’- TCGACGCCATCTTCATTCACA-3’ ) were used for norovirus detection as previously described [63] . Viral load was quantified by comparison to a standard curve of GII . 4 norovirus RNA transcripts of a known concentration . Average values for two independent experiments for concentrated virus and three independent experiments for RNAse free stool are presented . Statistical analysis was performed using one-way ANOVA test . Differences were considered significant when P≤0 . 05 . To evaluate the impact of Nanobody binding on capsid susceptibility to proteolytic digestion norovirus VLPs ( 1 mg/ml ) were incubated with Nanobodies ( 1 mg/ml ) in 1:1 ratio for 30 min at 37°C . Then , trypsin-EDTA was added to final concentration of 10 μg/ml for 30 min at 37°C . The concentration of trypsin was chosen to yield only partial cleavage with visible intermediate products . After digestion , samples were loaded on the SDS-12% polyacrylamide gel and stained with coomassie stain . | We determined the binding sites of six novel human norovirus specific Nanobodies ( Nano-4 , Nano-14 , Nano-26 , Nano-27 , Nano-32 , and Nano-42 ) using X-ray crystallography . The unique Nanobody recognition epitopes were correlated with their potential neutralizing capacities . We showed that one Nanobody ( Nano-26 ) bound numerous genogroup II genotypes and interacted with highly conserved capsid residues . Four Nanobodies ( Nano-4 , Nano-26 , Nano-27 , and Nano-42 ) bound to occluded regions on the intact particles and impaired normal capsid morphology and particle integrity . One Nanobody ( Nano-14 ) bound contiguous to the HBGA pocket and interacted with several residues involved in binding HBGAs . We found that the Nanobodies delivered multiple inhibition mechanisms , which included steric obstruction , allosteric interference , and disruption of the capsid stability . Our data suggested that the HBGA pocket might not be an ideal target for drug development , since the surrounding region is highly variable and inherently suffers from lack of conservation among the genetically diverse genotypes . Instead , we showed that the capsid contained other highly susceptible regions that could be targeted for virus inhibition . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"chemical",
"bonding",
"medicine",
"and",
"health",
"sciences",
"chemical",
"characterization",
"body",
"fluids",
"crystal",
"structure",
"pathology",
"and",
"laboratory",
"medicine",
"engineering",
"and",
"technology",
"pathogens",
"immunology",
"condensed",
"matter",
"physics",
"microbiology",
"saliva",
"viruses",
"rna",
"viruses",
"nanoparticles",
"nanotechnology",
"crystallography",
"cross",
"reactivity",
"hydrogen",
"bonding",
"physical",
"chemistry",
"research",
"and",
"analysis",
"methods",
"caliciviruses",
"solid",
"state",
"physics",
"medical",
"microbiology",
"norovirus",
"microbial",
"pathogens",
"chemistry",
"viral",
"packaging",
"viral",
"replication",
"binding",
"analysis",
"physics",
"anatomy",
"virology",
"viral",
"pathogens",
"physiology",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"organisms"
] | 2017 | Nanobodies targeting norovirus capsid reveal functional epitopes and potential mechanisms of neutralization |
RIG-I-like receptor ( RLR ) plays a pivotal role in the detection of invading pathogens to initiate type I interferon ( IFN ) gene transcription . Since aberrant IFN production is harmful , RLR signaling is strictly regulated . However , the regulatory mechanisms are not fully understood . By expression cloning , we identified Pumilio proteins , PUM1 and PUM2 , as candidate positive regulators of RIG-I signaling . Overexpression of Pumilio proteins and their knockdown augmented and diminished IFN-β promoter activity induced by Newcastle disease virus ( NDV ) , respectively . Both proteins showed a specific association with LGP2 , but not with RIG-I or MDA5 . Furthermore , all of these components were recruited to NDV-induced antiviral stress granules . Interestingly , biochemical analyses revealed that Pumilio increased double-stranded ( ds ) RNA binding affinity of LGP2; however , Pumilio was absent in the dsRNA-LGP2 complex , suggesting that Pumilio facilitates viral RNA recognition by LGP2 through its chaperon-like function . Collectively , our results demonstrate an unknown function of Pumilio in viral recognition by LGP2 .
The host innate immune system is the first line of defense against invading pathogens . Pattern-recognition receptors ( PRRs ) detect pathogen molecules , termed pathogen-associated molecular patterns ( PAMPs ) , to initiate innate immune responses [1] , [2] , [3] , [4] , [5] . Viruses invade host cells to replicate their genome and produce new infectious virions . RIG-I-like receptors ( RLRs ) , including RIG-I , MDA5 and LGP2 , sense the invasion and generation of viral RNA PAMPs and trigger antiviral responses [6] , [7] . In the resting state , RIG-I and MDA5 exist in an autorepressed state , in which N-terminal caspase activation and recruitment domains ( CARDs ) are masked by the helicase domain; however , upon virus infection , these helicases are activated and oligomerized along with RNAs to form filament-like structures [8] , [9] . Signals from RLRs are relayed to an adaptor , IPS-1 ( also known as MAVS , VISA , Cardif ) [10] , [11] , [12] , [13] , [14] , [15] , which then recruits TRAF adaptors , protein kinases TBK-1 , IKK-i and IKK complex to activate transcription factors IRF-3 , -7 and NF-κB [16] , [17] . Knockout mouse studies have shown that RIG-I and MDA5 play a pivotal role in the detection of a series of RNA viruses in vivo [18] . RIG-I detects Sendai virus , NDV and influenza A virus , whereas viruses belong to picornaviridae are sensed by MDA5 . Although the mechanism underlying the differential sensing of different viruses by RIG-I and MDA5 is not completely understood , it is proposed that virus specificity comes from the dsRNA length and 5′-end structure of viral RNA [19] , [20] , [21] , [22] . LGP2 was originally thought to be a negative regulator because it lacks CARD , which is crucial for signal transduction . However , knockout and knock-in mouse studies have shown that LGP2 functions as a positive regulator via its ATPase activity [23] , consistent with its high affinity binding with dsRNA [7] , [24] . Recent studies have reported that RLR signaling is subject to numerous regulations [25] . TRIM25 positively regulates signaling through interactions with RIG-I and ubiquitination [26] . Riplet ( also termed RNF135 and REUL ) positively regulates RIG-I signaling through ubiquitination of RIG-I , independent of TRIM25 [27] , [28] . On the other hand , ubiquitin ligases , RNF125 [29] and A20 [30] , and deubiquitinating enzymes , DUBA [31] and CYLD [32] , are reported to function as negative regulators of RIG-I signaling . In addition to the ubiquitination of signaling peptides , involvement of the free ubiquitin chain has been proposed [33] . Furthermore , accumulating reports suggest the importance of the virus-induced stress response in antiviral innate immunity . In particular , viral infection induces antiviral stress granules ( avSGs ) , including RIG-I , MDA5 , LGP2 and viral RNA [34] , [35] , [36] , [37] . Our expression cloning for antiviral signal regulators identified Pumilio proteins . Pumilio proteins ( also termed PUF , Pumilio/FBF ) are evolutionary conserved from plants to mammals and were originally identified as translational repressors through direct binding to the specific sequence termed the Nanos response element ( NRE ) present within the 3′-UTR of target mRNAs , thereby regulating various processes: embryonic development , stem cell differentiation , cell cycle and mitochondrial biogenesis [38] , [39] , [40] , [41] , [42] . In this report , we describe a novel and non-translational function of Pumilio proteins in viral recognition by LGP2 .
We previously identified RIG-I by expression cloning of the human cDNA library using virus-inducible reporter gene activity as the readout [6] . This strategy allowed us to identify other candidate regulators of antiviral signaling , Pumilio proteins . Pumilio proteins share a highly conserved C-terminal Pumilio-homology domain ( PUM-HD ) [43] . In humans , two genes encoding PUM1 and PUM2 exist . Human PUM1 and PUM2 have similar domain structures and have high homology in their primary structure ( Fig . 1A ) . It was reported that PUM-HD is responsible for sequence-specific RNA binding , whereas the function of the N-terminal portion is unknown . We obtained two independent clones encoding full-length PUM1 and one clone of PUM2 ( missing coding amino acids 1-368 ) by expression cloning . We constructed expression vectors for full-length PUM1 and PUM2 and examined their effect on IFNB promoter activity in L929 cells . Overexpression of PUM1 and PUM2 augmented IFNB promoter activity induced by NDV infection ( Figure 1B ) . As IFNB promoter is regulated by both IRFs and NF-κB , we investigated whether PUM1 and PUM2 affect promoter activity regulated by IRFs ( p-55C1B ) and NF-κB ( p-55A2 ) . PUM1 and PUM2 augmented p-55C1BLuc activity ( Figure 1C ) , as well as p-55A2Luc activity ( Figure 1D ) , suggesting that PUM1 and PUM2 mediate the activation of both IRFs and NF-κB transcription factors . In accord with the increased IFN promoter activity , NDV RNA replication was suppressed by the overexpression of PUM1 and PUM2 24 h after NDV infection , suggesting that PUM1 and PUM2 share antiviral potential ( Figure 1E ) . It is known that translational repression by PUM-HD depends on H850 in PUM2 [44] and this residue was conserved between PUM1 ( H972 ) and PUM2 . Therefore , we constructed expression vectors for their alanine mutants and tested their antiviral activity . Interestingly , these mutants markedly enhanced NDV-induced IFNB promoter activity ( Figure 1F ) , suggesting that the enhancing function is independent of the translational repression function of Pumilio proteins . To further investigate the function of PUM1 and PUM2 in IFN induction , we performed siRNA-mediated knockdown . siRNA targeting for human PUM1 and PUM2 suppressed the expression of endogenous PUM1 and PUM2 protein , respectively ( Figure 2A ) . As expected , knockdown of endogenous PUM1 or PUM2 impaired the mRNA expression of IFNB1 and CXCL10 , one of the IFN-stimulated genes , in response to NDV infection ( Figure . 2B and C ) . We also examined IRF-3 phosphorylation in the PUM knockdown cells , as well as IRF-3 dimerization . As shown in Figure S1A and B , both phosphorylation and dimerization of IRF-3 were impaired in the Pumilio knockdown cells . Furthermore , the production of IFN-β protein was also reduced by PUM1 or PUM2 knockdown upon NDV infection ( Figure 2D ) . Conversely , NDV RNA copies were increased by knockdown of PUM1 or PUM2 ( Figure 2E ) . To rule out the possibility that Pumilio proteins regulate NDV-induced IFN production through affecting the expression level of RLRs , we examined the basal level of RLRs in the Pumilio knockdown cells . The knockdown of Pumilio proteins did not affect the basal and IFN-induced expression level of RLRs ( Figure S2 ) . Finally , we also tested synthetic oligonucleotides , such as poly I:C , in vitro-transcribed 5′pppRNA and poly dA:dT . In contrast to NDV infection , the knockdown of Pumilio proteins did not affect the IFN production in response to these stimuli ( Figure S1C ) . These results suggest that PUM1 and PUM2 positively regulate antiviral responses against NDV by controlling IFN production . It has been shown that NDV infection is mainly detected by one of the RLRs , RIG-I [18] . TRIM25 regulates the activation of RIG-I through ubiquitination of RIG-I [26] . In addition , IPS-1 is an adaptor protein essential for RLR signaling [14] , [15] . To elucidate the regulatory mechanism , the physical association of full-length PUM1 and PUM2 with RLRs , TRIM25 and IPS-1 was examined by co-immunoprecipitation . As shown in Figure 3A , LGP2 , but not RIG-I nor MDA5 was precipitated with PUM1 or PUM2 . No interaction of PUM1 and PUM2 with TRIM25 and IPS-1 was detectable , indicating that PUM1 and PUM2 selectively interact with LGP2 . We also investigated whether PUM1 and PUM2 interacted with each other . As shown in Figure S3 , PUM2 associated with PUM1 . This result suggested that PUM1 and PUM2 exist as heteromeric complex . LGP2 was shown to function as a positive regulator of RIG-I- and MDA5-mediated antiviral responses [23] . Specific associations between LGP2 and PUM1 and PUM2 prompted us to investigate the involvement of LGP2 in Pumilio-mediated transactivation . L929 cells were transfected with the expression vector for shRNA either non-targeted or targeted to LGP2 , then transactivation by PUM1 or PUM2 was examined ( Figure 3B ) . Knockdown of LGP2 markedly attenuated transactivation by PUM1 or PUM2 , suggesting that the observed physical interaction between LGP2 and PUM1 and PUM2 is relevant to the biological activity of these regulators . To elucidate the involvement of C-terminal PUM-HD in the association between LGP2 and PUM1 and PUM2 , expression vectors for PUM-HD ( PUM1dN and PUM2dN ) and the rest ( PUM1dC and PUM2dC ) were constructed ( Figure 3C ) . Co-immunoprecipitation using the mutants revealed that LGP2 interacted with PUM1 and PUM2 lacking PUM-HD as strongly as with the respective full-length proteins , while interaction with PUM-HD ( dN constructs ) was undetectable . Consistent with the lack of interaction between PUM-HD and LGP2 , PUM1H972A and PUM2H850A efficiently co-precipitated with LGP2 ( Figure 3D ) . We also determined the domain of LGP2 responsible for the interaction with Pumilio proteins . As shown in Figure S4 , LGP2 helicase domain is important for LGP2 to interact with Pumilio proteins . These results suggest that PUM1 and PUM2 interact with LGP2 through the N-terminal domain . To further elucidate the mechanism of transactivation by PUM1 and PUM2 , full-length and dN and dC mutants were tested for IFNB promoter activation ( Figure 3E ) . Full-length PUM1 and PUM2 but neither dC nor dN enhanced NDV-induced IFN-β reporter activity , indicating that both PUM-HD and the N-terminal portion are necessary for transactivation . It is reported that PUM1 and PUM2 are recruited to stress granules ( SGs ) upon stress responses , such as oxidative stress or starvation [45] . Previously , we reported that virus infection induces SG-like aggregates containing SG markers , RLR and several antiviral proteins , and termed the aggregate antiviral SGs ( avSGs ) [34] . avSGs are thought to function as a platform for detection of viral RNA by RLR and as action sites of antiviral proteins . We determined the cellular localization of PUM1 and PUM2 by immunostaining . In uninfected cells , PUM1 and PUM2 localized diffusely in the cytoplasm ( Figure 4A ) . NDV infection induced co-localization of PUM1 and PUM2 into cytoplasmic speckle-like aggregates ( Figure 4A ) . We confirmed that a SG marker , TIAR , localized with the speckles containing PUM1 in NDV-infected cells ( Figure 4B ) ; therefore , these aggregates correspond to avSG . We also confirmed that LGP2 localized in the avSG ( Figure 4C ) Both N- and C-terminal domains of Pumilio proteins are required for transactivation ( Figure 3E ) and the N-terminal domain is responsible for interaction with LGP2 . We therefore explored the function of C-terminal PUM-HD in terms of cellular localization . Flag-tagged PUM1dN and PUM2dN were expressed in cells and the cells were infected with NDV ( Figure 4D ) . In uninfected cells , these mutants were diffusely accumulated in nuclei and cytoplasm; however , upon viral infection , these proteins localized with avSG , suggesting that PUM-HD is responsible for the localization of PUM proteins in avSG . We also determined the cellular localization of PUM1dC and PUM2dC . As shown in Figure S5D , PUM1dC diffusely localized in the cytoplasm of NDV-infected cells , whereas PUM2dC was recruited to the avSGs in response to NDV infection . To explore the possibility that PUM1 and PUM2 are required for avSG formation , the effect of knockdown of PUM1 and PUM2 on avSG formation was examined . As shown in Figure S5A , knockdown of PUM1 or PUM2 expression did not alter avSG induction in NDV-infected cells , suggesting that PUM1 and PUM2 do not notably affect avSG assembly . We also examined the recruitment of Pumilio proteins and LGP2 in LGP2 KO cells and Pumilio knockdown cells , respectively . The knockdown of Pumilio proteins did not affect the localization of LGP2 ( Figure S5B ) . Furthermore , Pumilio proteins were recruited to the avSGs in response to NDV in LGP2 KO cells ( Figure S5C ) , indicating that Pumilio proteins were not involved in the recruitment of LGP2 to the avSGs and vice versa . Finally , to elucidate the mechanism of the enhancement of IFN gene expression by PUM1 and PUM2 , we examined dsRNA binding activity of LGP2 in the presence or absence of PUM1 and PUM2 . Electrophoresis mobility shift assay ( EMSA ) was performed using synthetic dsRNA and recombinant proteins . LGP2 bound to the probe , resulting in a slow-migrating complex . We confirmed that this slow-migrating band was a complex of the probe and LGP2 by a supershift experiment ( Figure 5A ) . PUM1 and PUM2 exhibited very weak binding with the probe ( Figure 5B ) . Interestingly , the LGP2-dsRNA complex intensity was increased with the addition of PUM1 and PUM2; however , the mobility of the complex is hardly affected . Dissociation constant for LGP2 in the absence and presence of PUM1 or PUM2 was determined by Scatchard plot analysis ( Figure 5C ) . The Kd value indicates that PUM1 and PUM2 increased the dsRNA-binding activity of LGP2 . We also purified recombinant PUM1 and PUM2 lacking PUM-HD ( PUM1dC and PUM2dC ) and subjected them to a binding assay ( Figure S6A and B ) . Essentially similar results were obtained , indicating that the N-terminal domain of PUM1 and PUM2 is sufficient for increasing the dsRNA binding affinity of LGP2 . Because the complex mobility in EMSA did not change in the presence or absence of Pumilio proteins , we examined whether the association between Pumilio proteins and LGP2 is affected in the presence or absence of dsRNA . Recombinant PUM1 and PUM2 proteins produced as GST fusion were mixed with recombinant Flag-tagged LGP2 in the presence or absence of dsRNA and pulled down with glutathione Sepharose ( Figure S7 ) . In the absence of dsRNA , we confirmed the association between PUM1 and PUM2 with LGP2; however , in the presence of dsRNA , this association was undetectable , suggesting that upon binding of LGP2 with dsRNA , PUM1 and PUM2 are released from the complex , consistent with the EMSA results . Taken together , we hypothesized that Pumilio proteins changed the conformation of LGP2 through physical associations to increase its dsRNA binding affinity ( Figure 6 ) .
We found that Pumilio proteins enhanced NDV-induced activation of the IFNB gene ( Figure 1 ) . Knockdown of PUM1 and PUM2 respectively attenuated gene activation ( Figure 2 ) , suggesting that PUM1 and PUM2 function non-redundantly . Furthermore , PUM1 and PUM2 accelerated the activation of IRFs and NF-κB transcription factors , suggesting their action in signal transduction , rather than post-transcriptional steps . Interestingly , PUM1 and PUM2 selectively interact with LGP2 but not with RIG-I , MDA5 , IPS-1 or TRIM25 ( Figure 3A ) . LGP2 knockdown diminished IFNB gene induction augmented by PUM1 and PUM2 ( Figure 3B ) , suggesting that PUM1 and PUM2 augment viral RNA sensing mediated by LGP2 . It was shown that LGP2 does not participate in the detection of synthetic oligonucleotides , such as poly I:C or in vitro-transcribed 5′pppRNA [23] . Consistent with this , the knockdown of PUM1 and PUM2 did not affect the IFN production induced by poly I:C , 5′pppRNA or poly dA:dT ( Figure S1C ) . LGP2 exhibits strong binding activity to dsRNA; however , it lacks CARD , through which the signal is relayed to IPS-1 . Therefore , it has been hypothesized that LGP2 cooperates with either RIG-I or MDA5 . Our findings uncovered a new mechanism of sensing viral RNA by LGP2 , PUM1 and PUM2 . Deletion analyses of PUM1 and PUM2 revealed that both N- and C-terminal regions are required for up-regulation ( Figure 3E ) . The N-terminal region is sufficient for interaction with LGP2 and to increase the binding affinity to dsRNA ( Figure 3D and S7 ) . The C-terminal region , also termed PUM-HD , is sufficient for translocation to avSG upon viral infection ( Figure 4D ) , although the underlying mechanism is unknown . PUM1 and PUM2 have been known to regulate mRNA translation through sequence-specific recognition of NRE within the target mRNA . It was proposed that PUM-HD consists of 8 repeats of the module and each module recognizes a single nucleotide in NRE [43] . It was shown that single amino acid substitution is sufficient to abolish NRE binding and the translational regulation of PUM2 [44] . On the other hand , we found that NRE binding-deficient mutants ( PUM1H972A and PUM2H850A ) augment virus-induced signaling as strongly as the respective wt protein ( Figure 1F ) , suggesting that Pumilio proteins facilitate two independent biological functions in translational and antiviral signal regulation . It is tempting to speculate that the repeated modules recognize component ( s ) associated with avSGs . Concerning the molecular mechanism , we found that PUM1 and PUM2 increased dsRNA binding affinity of LGP2 ( Figure 5 ) . However , interestingly , we did not observe the ternary complex of dsRNA , LGP2 and PUM1 or PUM2; furthermore , the interaction between LGP2 and PUM1 or PUM2 was lost when dsRNA was added ( Figure S7 ) . We therefore hypothesize that PUM1 and PUM2 have a cooperative function with LGP2 through physical association to increase its binding affinity to dsRNA ( Figure 6 ) . It has been shown that LGP2 facilitates RIG-I- and MDA5-mediated signaling [23] . Also it was shown that LGP2 and RIG-I interact [46] . In light of these observations , it is probable that increased dsRNA binding of LGP2 facilitates RLR signaling . Viral infection induces the formation of avSGs , including conventional SG markers , RIG-I , MDA5 , LGP2 , PKR , OAS , RNase L , DHX36 , TRIM25 , PUM1 and PUM2 , some of which are critical in sensing non-self viral RNA and triggering antiviral signaling . Unlike SGs induced by physical stress , viral RNA is accumulated in virus-induced avSGs [34] , [35] , [37] . In summary , these results support the idea that avSGs act as a critical platform for sensing and discriminating viral RNA as a defense mechanism against viral infections . Although the IFN system is absent in plants , Pumilio proteins participate in the antiviral response in plants [47] , suggesting that the principal mechanism of sensing non-self RNA is evolutionarily conserved .
L929 cells were maintained in minimal essential medium ( MEM ) ( nacalai tesque , ) containing 5% fetal bovine serum ( FBS ) . HEK293T and HeLa cells were maintained in Dulbecco's modified Eagle's medium ( DMEM ) ( nacalai tesque ) containing 10% FBS . The p-125 Luc , p-55C1B Luc , p-55A2 Luc , pU6i and pU6i-shLGP2 have been described previously [6] , [7] . pEF-Flag-PUM1 and PUM2 was obtained by subcloning cDNA into the empty vector pEF-BOS . Mutants were generated using the KOD -plus- Mutagenesis Kit ( TOYOBO ) . TRIM25 cDNA was purchased from OriGene . Negative control siRNA and siRNA targeting PUM1 or PUM2 were purchased from BONAC . siRNAs were transfected using Lipofectamine RNAiMAX ( Invitrogen ) according to the manufacturer's protocol . After 48 h , the cells were stimulated as indicated . Total RNA was isolated using Sepasol reagent ( nacalai tesque ) , treated with DNase I ( Roche ) and subjected to reverse transcription using a High-Capacity cDNA Reverse Transcription Kit ( Applied Biosystems ) . mRNA levels were monitored with the StepOne plus Real Time PCR System and TaqMan Fast Universal Master Mix ( Applied Biosystems ) . TaqMan primer and probe sets for 18S rRNA , human IFNB1 and human CXCL10 were purchased from Applied Biosystems . The RNA copy numbers of the gene of interest were normalized to that of internal 18S rRNA . NDV replication levels were monitored with Fast SYBR PCR Master Mix ( Applied Biosystems ) using the primers specific for the NDV F gene . Anti-Flag and anti-HA antibody were purchased from Sigma and Cell Signaling Technology , respectively . Anti-GST anti-β-actin , anti-c-Myc , anti-Pumilio1 , anti-Pumilio2 and anti-TIAR antibodies were from Santa Cruz Biotechnology ( Santa Cruz , CA , USA ) . Anti-IRF-3 , anti-RIG-I , anti-MDA5 and anti-LGP2 antibody were described previously [34] , [48] . Alexa 488- and 594-conjugated anti-rabbit or anti-goat IgG antibodies ( Invitrogen ) were used as secondary antibodies . The cells were fixed with 4% paraformaldehyde ( nacalai tesque ) for 10 min , permeabilized with an acetone: methanol ( 1∶1 ) solution , and blocked with 5 mg/ml bovine serum albumin ( BSA ) ( nacalai tesque ) for 30 min . The cells were incubated with the indicated primary antibodies overnight at 4°C , and then incubated with the relevant Alexa-conjugated antibodies at room temperature for 1 h . Nuclei were stained with DAPI ( nacalai tesque ) . The cells were analyzed with a microscope ( Leica microsystems ) . Luciferase assay was performed as described previously [7] . The Dual-Luciferase Reporter Assay System ( Promega ) was used according to the manufacturer's protocol . The indicated plasmids were transfected with HEK293T cells using Lipofectamine 2000 ( Invitrogen ) . The cell lysates were incubated with anti-Flag or anti-c-Myc antibody on ice for 30 min . The pre-washed Protein G Sepharose ( GE Healthcare ) was added to the mixture , which was rotated at 4°C overnight . After washing , the precipitates were eluted and separated by SPS-PAGE , followed by Western blotting . NDV was grown in the allantonic cavities of 9-day-old embryonated eggs . The cells were mock treated or infected with NDV at 37°C . The cell lysates were subjected to Native PAGE and Western blotting as described previously [6] , [48] . The cell culture supernatants were collected and subjected to ELISA with a human IFN-β ELISA kit ( TORAY , Tokyo , Japan ) according to the manufacturer's protocol . Recombinant LGP2 was produced as 6xHis-LGP2 fusion using baculovirus and High Five cells . 6xHis-LGP2 was bound to Ni Sepharose 6 Fast Flow ( GE Healthcare ) , and then eluted by elution buffer containing 50 mM Tris-HCl ( pH 8 . 0 ) , 150 mM NaCl , 1 . 5 mM DTT and 500 mM imidazole . The intact PUM1 and PUM2 were amplified by PCR and inserted into a modified pGex-6p-1 vector ( GE Healthcare ) . The C-terminal His6-tag was inserted using a KOD plus mutagenesis kit ( TOYOBO ) to produce N-terminal GST and C-terminal His6 tagged proteins . The vectors were transformed into an E . coli BL21 ( DE3 ) strain . Bacteria were first grown at 37°C in LB medium containing 100 µg/ml ampicillin at 160 rpm . Protein expression was induced by the addition of 0 . 1 mM IPTG when the absorbance at 600 nm was approximately 0 . 4 . The cells were then grown at 16°C for 16 h at 90 rpm . The cells were harvested by centrifugation and were suspended in a lysis buffer containing 50 mM Tris-HCl ( pH 8 . 0 ) , 500 mM NaCl , and 20 mM imidazole supplemented with protease Inhibitor Cocktail ( Roche Diagnostics ) and were lysed via sonication and centrifugation . The supernatant was suspended in Ni Sepharose 6 Fast Flow ( GE Healthcare ) , then the resin was washed with lysis buffer , and the protein was eluted by elution buffer containing 50 mM Tris-HCl ( pH 8 . 0 ) , 500 mM NaCl , and 500 mM imidazole . The protein was diluted by phosphate-buffered saline ( PBS ) and mixed with Glutathione Sepharose 4B ( GE Healthcare ) for 16 h . The mixture was washed with PBS and proteins were eluted by a buffer containing 50 mM Tris-HCl ( pH 8 . 0 ) , 150 mM NaCl , and 20 mM reduced glutathione . Recombinant LGP2 proteins were mixed with 32P-labeled synthetic dsRNA ( 25/25c ) [49] in a reaction mixture ( 20 mM Tris-HCl ( pH 8 . 0 ) , 1 . 5 mM MgCl2 , and 1 . 5 mM DTT ) in the presence or absence of recombinant Pumilio proteins . After incubation at 37°C for 15 min , the reaction mixture was applied to a 15% acrylamide gel ( TBE buffer ) and the radioactivity was detected with an Image Analyzer ( FUJIFILM , Tokyo , Japan ) . Recombinant LGP2 proteins were mixed with Pumilio proteins in a reaction mixture ( 20 mM Tris-HCl ( pH 8 . 0 ) , 1 . 5 mM MgCl2 , and 1 . 5 mM DTT ) in the presence or absence of synthetic dsRNA ( 25/25c ) at 37°C for 15 min . Pre-washed Glutathione Sepharose 4B ( GE Healthcare ) was added to the mixture and incubated at room temperature for 1 h . After washing , the precipitates were separated by SDS-PAGE , followed by Western blotting . | Mammals utilize innate immune system to counteract viral infections . The host pattern-recognition receptors , such as RIG-I-like receptors ( RLRs ) , sense invading pathogens and initiate innate immune responses . RLRs are composed of three RNA helicases , RIG-I , MDA5 and LGP2 , and detect a series of RNA viruses , such as influenza or hepatitis C virus , in the cytoplasm . Upon RNA virus infection , RLRs transmit signals through mitochondrial adaptor protein , IPS-1 , to activate transcription factor IRF-3/7 , resulting in the production of type I interferon ( IFN ) . Type I IFN plays a crucial role in innate immune system by inducing a hundreds of interferon-stimulated genes and its induction is tightly controlled at transcriptional and translational steps . Pumilio proteins are originally identified as translational repressor through direct binding to specific sequence motifs in the 3′ untranslated regions of specific mRNA , and regulate critical biological processes , such as development and differentiation . In this report , we identified human Pumilio proteins , PUM1 and PUM2 , as candidate regulators of IFN signaling . Our results demonstrated an unknown function of Pumilio in viral recognition by LGP2 . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"biochemistry",
"immune",
"system",
"proteins",
"pattern",
"recognition",
"receptors",
"developmental",
"biology",
"innate",
"immune",
"system",
"cytokines",
"proteins",
"molecular",
"development",
"biology",
"and",
"life",
"sciences",
"immune",
"receptors",
"immunology",
"immune",
"system"
] | 2014 | A Novel Function of Human Pumilio Proteins in Cytoplasmic Sensing of Viral Infection |
Psychophysical and neurophysiological studies have suggested that memory is not simply a carbon copy of our experience: Memories are modified or new memories are formed depending on the dynamic structure of our experience , and specifically , on how gradually or abruptly the world changes . We present a statistical theory of memory formation in a dynamic environment , based on a nonparametric generalization of the switching Kalman filter . We show that this theory can qualitatively account for several psychophysical and neural phenomena , and present results of a new visual memory experiment aimed at testing the theory directly . Our experimental findings suggest that humans can use temporal discontinuities in the structure of the environment to determine when to form new memory traces . The statistical perspective we offer provides a coherent account of the conditions under which new experience is integrated into an old memory versus forming a new memory , and shows that memory formation depends on inferences about the underlying structure of our experience .
Recent psychophysical studies have explored the dynamics of memory updating by presenting participants with sequences of stimuli and then probing their ability to discriminate between different stimuli in the sequence . The logic of these studies is that if the stimuli are assimilated into the same dynamical mode , then they will be perceived as being more similar , compared to a situation where they are segmented into different modes . For example , Wallis and Bülthoff [15] presented participants with a rotating face that gradually morphed into a different face . Compared to a condition in which the morphs were presented in a mixed ( scrambled ) order , participants in the gradual morph condition were more prone to perceive the final face as belonging to the same person as the original face . Similar findings were reported by Preminger and colleagues [16] , [17] using a variety of memory tests . These psychophysical observations are complemented by neurophysiological studies of spatial representation in the rodent hippocampus . Many neurons in the CA3 subfield of the hippocampus respond selectively when the animal is in a particular region of space , and are therefore known as “place cells” [18] . We can apply the same logic used in the aforementioned psychophysical studies to the hippocampal representation of space [19] , asking whether morphing one environment into another will lead to gradual changes in place cell firing rate ( indicating a gradually changing spatial memory ) or a global remapping of place fields ( indicating the formation of a new memory ) . Leutgeb et al . [20] and Wills et al . [21] had rats explore a set of enclosures whose shape varied from a square to a circle ( including intermediate shapes ) . Gradually changing the enclosure shape ( the “gradual” protocol ) resulted in gradual changes in place fields [20] , whereas presenting the same series of enclosures in a scrambled order ( the “mixed” protocol ) resulted in global remapping – enclosures that were more similar to the circle than to the square tended to elicit one set of place fields , and enclosures that were more similar to the square than to the circle tended to elicit a distinct set of place fields [21] . As with the psychophysical findings described above , these results highlight the importance of sequential structure in guiding memory organization; the same stimuli can elicit very different internal representations depending on the order in which they are presented . Using a Hopfield network to encode the input patterns , Blumenfeld et al . [22] proposed a “salience-weighted” modification of the standard Hebbian learning rule to model these findings . Intuitively , the salience weight encodes a prediction error or novelty signal that indicates the extent to which none of the network's existing attractors match the current input pattern . Formally , the salience weight is the Hamming distance between the input pattern and the network state after one step of dynamics; the salience weight is updated incrementally after each input pattern so as to smooth across recent history . A large salience weight promotes the formation of a new attractor based on the current input . For our purposes , the key idea to take away from this model is that prediction errors are useful signals for determining when to infer new memory modes ( see also [23]–[26] ) . In the network explored by Blumenfeld et al . , a new attractor is only formed if the prediction error is sufficiently large , but how large is “sufficient” ? In the next section , we place these ideas within a statistical framework , which allows us to specify the prediction error threshold in terms of probabilistic hypotheses about the environment . The essence of our approach is captured by the following generic assumption about the environment: Properties of the environment usually change gradually , but occasionally undergo “jumps” that reflect a new underlying state of affairs [11] , [12] . Returning to the temperature example , when you walk around outside , you may experience gradual changes in temperature over the course of the day . If you step into a building , the temperature may change abruptly . In predicting what the temperature will be like in 5 minutes , you might then generalize from one outdoor location to another , but not between the indoor location and outdoor locations . Thus , our generalizations depend strongly on how we segment our observations; cognitively speaking , one can view each segment as a memory trace that aggregates those observations assigned to the segment . The empirical data reviewed in the previous section are consistent with the idea that the brain is attuned to abrupt changes in the state of the environment . The problem of estimating the current state of a hidden variable given previous sensory measurements is known in engineering as filtering . The classic example of a filtering algorithm is the Kalman filter ( KF; [27] ) , which is the Bayes-optimal estimator under the assumption that the environment evolves according to a linear-Gaussian dynamical system ( LDS ) –i . e . , the state of the environment changes gradually and noisily over time . By design , this model cannot account for large sporadic jumps and periods of gradual change between them . One way to model jumps is to posit a collection of different “dynamical modes” , each corresponding to a slowly changing LDS , and allow the generative process to switch between them stochastically . This is known as a switching LDS , and its corresponding Bayes-optimal estimator is the switching KF . However , for real-world sensory measurements , it is not reasonable to specify the number of possible modes in advance . We therefore adopt a Bayesian infinite-capacity ( nonparametric ) generalization of the switching LDS based on the Dirichlet process [28] , which allows the number of modes to expand as necessary as measurements are collected ( another dynamical model that could capture jumps within a single mode is a random walk with a heavy-tailed distribution on step size , such as a Lévy flight [29] ) . The infinite-capacity prior over modes leads to an intuitive interpretation in terms of memory traces: Each mode clusters together a number of individual observations , and thus can be identified with a temporally extended episodic memory trace such as the memory of the temperature outside . The number of such modes is essentially unlimited . However , because in our model small numbers of modes have higher probability a priori , the result is that the memory system tries to account for its observations as parsimoniously as possible by using existing modes to explain multiple observations . This leads to potential modification of existing modes each time a new observation is assigned to them , and sporadic creation of new modes . Below we describe this model formally .
Let denote a set of sensory measurements at time t , arising from unobservable state variables , where k indexes modes . For instance , the observation may be the current temperature , and the state variables are the air pressure , cloud coverage , inside/outside location , air conditioner , thermostat status , and many other direct causes of temperature . Let denote the mode active at time t . This mode specifies particular state-space dynamics , for instance , a mode corresponding to being indoors with the air conditioning on ( which specifies the dependence of temperature on thermostat settings ) , another corresponding to air conditioning being off , another to being outside in the shade , etc . Our model assumes that measurements ( observations ) are generated according to the following stochastic process . For each time point t: This generative model is a simplification of the nonparametric switching LDS described in [28] . To summarize the generative model: The hidden state diffuses gradually until a jump occurs; this jump can be either to a previously activated mode , or to a new mode ( in which case a new starting point is drawn for that mode , from a Gaussian prior ) . The concentration parameter controls the probability that a new mode will be activated: Larger values of result in more modes , and if , there are no jumps and we obtain a special case of the standard LDS formulation . The stickiness parameter β encourages modes to persist over time; when , we recover the original Chinese restaurant process [33] . The diffusion variances control the rate of change within a mode: Larger values of result in faster change . The sensory noise variance controls the informativeness of the observations about the hidden state: As increases , the sensory measurements become noisier and hence convey less information about the hidden state . Given the generative model , the filtering problem is to infer the posterior distribution over the state variable for each mode given the history of sensory measurements . This computation is given by: ( 2 ) where . This corresponds to a “local” approximation [34]–[36] that maintains only a single high probability partition of previous observations to hidden causes . This partition is then used to calculate the probability of the current trial being drawn from each of the latent causes by combining the sticky Chinese restaurant process prior ( Eq . 1 ) and the likelihood ( conditional on the partition and the previous observations ) of the current state vector . Although we could have used more sophisticated methods ( e . g . , particle filtering ) to approximate the marginalization , this method works well on the examples we consider , and is much faster , making it easier to fit to behavioral data . We now describe how to compute each of the components in Eq . 2 . The conditional distribution is a Gaussian , with mean and covariance , updated according to: ( 3 ) for each dimension d , where the estimated mean and variance for a new mode k are and ( respectively ) and the step size ( or learning rate ) η , also known as the Kalman gain , is ( 4 ) Using the local approximation described above , the posterior over mode assignments is given by: ( 5 ) where the second term is the prior ( Eq . 1 ) , and the first term is the likelihood: ( 6 ) where “new mode” refers to the first mode that has never been active before time i . This completes the description of our inference algorithm , which we refer to as the Dirichlet process Kalman filter ( DP-KF ) . Viewed as a mechanistic psychological model , the DP-KF assumes that the memory system keeps track of two kinds of traces: episodic traces encoding the sensory stimulus at each time point ( ) , and more general traces that encode summary statistics of stimuli belonging to a common mode ( ) . These summary statistics are updated in an incremental , psychologically plausible manner using error-driven learning . Episodes are partitioned into modes by a competitive clustering process similar to mechanisms that have been proposed in many other psychological and neural models [23] , [24] , [34] , [37] , [38] . Eq . 6 operationalizes the idea that large prediction errors will lead to the inference of a new mode: For an old mode the Gaussian log-likelihood is inversely proportional to , the distance between the current observation and the state when the mode was last active , where is the time at which the old mode last occurred , while for a new mode the log-likelihood is proportional to ( with the constant of proportionality scaling these distances by the variances of the modes ) . Thus when is large relative to the DP-KF will tend to assign observation t to a new mode , analogous to the process by which Blumenfeld et al . 's [22] saliency-weighted learning rule creates a new attractor when the input pattern fails to match any of the existing attractors . ( Although the likelihood for a new mode depends on the absolute scale of , in our simulations this dependence was very weak , as the variance parameter was set to c = 1000 . ) Furthermore , because the variance of a mode grows with the length of time since its last occurrence ( ) , older modes will be more “tolerant” of prediction errors . Figure 1A illustrates the results of inference using our model with a one-dimensional sensory stimulus . Here we assumed and α = 1 . The sensory stimulus changed gradually , then underwent a jump , and then changed gradually again . On each time point we first inferred the hidden state based on past observations only ( these are the model predictions ) . Following that , the sensory measurement was observed , thereby allowing the computation of its likelihood and updating of the posterior distribution . As a result , model predictions lag behind the jump . Nevertheless , due to inferring a new mode after the jump , the DP-KF ( circles ) “catches up” with the sensory evidence after one trial , whereas the regular KF model ( squares ) takes much longer . This occurs because the KF smooths across the jump as all observations are assumed to be generated by one slowly diffusing mode , whereas the DP-KF achieves piecewise smoothness by segmenting the time series into two modes , thereby producing better predictions . Figure 1B shows the results of applying the DP-KF to the “gradual” and “mixed” experimental protocols described in the Introduction [15]–[17] , [20] , [21] , [39] . Here we used a sequence of one-dimensional measurements morphing between 0 and 1 . In the gradual protocol , the sensory measurement ( morphs ) increased monotonically with time , whereas in the mixed protocol the morphs were presented in scrambled order . To analyze the simulated data , we re-sorted the indices from the mixed condition to match the gradual condition and calculated the posterior probability of mode 1 for each morph . Consistent with the psychophysical and neurophysiological data [15]–[17] , [20] , [21] , [39] , the mixed protocol results in morphs being assigned to two different modes , whereas the gradual protocol results in all the morphs being predominantly assigned to a single mode . Note that even if each of the modes is already firmly ingrained ( through extensive experience with the morphs , as was the case in some of the experimental work we discussed ) , we still expect to see gradual or abrupt changes in the posterior probability of mode 1 depending on the morph sequence , since the sensory data are ambiguous with respect to the underlying dynamical mode . In other words , the time course of the posterior reflects uncertainty about which mode is currently active , and this uncertainty may change smoothly or abruptly depending on the stimulus sequence . We now describe an experiment designed to test a fundamental prediction of our model: if different modes correspond to different memories , inference of a new mode should protect the memory for old observations from retroactive interference due to new observations ( see Materials and Methods for more details ) . Figure 2 illustrates the task . We exposed human participants to sequences of simple visual stimuli ( lines ) whose orientation and length changed from trial to trial , and asked them , at the end of the sequence , to reconstruct from memory one of the stimuli from the beginning of the sequence . To ensure that participants were encoding the stimuli , and to provide data that can be compared to the model's trial-by-trial predictions for the purpose of model fitting , we also asked participants to actively predict the orientation and length of the next line . Each participant was exposed to sequences belonging to two conditions: in the “gradual” condition , the lines changed slowly , through small perturbations in orientation/length space; in the “jump” condition , this slow change was interrupted by a large change in the middle of the sequence ( Figure 3 ) . Importantly , we kept the overall distance ( in terms of orientation and length ) between the start and end points of each sequence approximately equal in both conditions . We reasoned that if participants used prediction errors to segment their observations into distinct modes , then they would infer two modes in the jump condition ( one for the first half and one for the second half of the sequence ) , but only one mode for the gradual condition . Segmenting the sequence would mean that the memory for the first half should be less biased by observations in the second half . We therefore hypothesized that reconstructions of early lines would be more veridical in the jump condition . By contrast , in the gradual condition , later observations would have been assigned to the initial mode , leading to alteration of that mode . Compared to the jump condition , reconstructions in the gradual condition should therefore be more similar to lines observed later in the block , and less similar to the target early lines . Example trajectories and reconstructions for a single participant are shown in Figure S1 . To test our hypothesis , for each sequence we calculated the Euclidean distance between the participant's reconstruction and the true line observed at the beginning of the block , as well as the distance from the line observed at the end of that block . The results , presented in Figure 4A , show that participants' reconstructions were closer to the last line ( ) , and farther from the first line ( ) in the gradual condition as compared to the jump condition . A two-way ( first/last × gradual/jump ) ANOVA confirmed that the interaction was significant ( ) . We interpret this result as showing that , in the gradual condition , participants inferred one mode , thereby causing lines from the second half to influence memory for the lines from the first half; by contrast , in the jump condition participants inferred separate pre-jump and post-jump modes , thereby protecting their memory of the pre-jump lines from being distorted by the post-jump lines . Assuming that the individual trace of each stimulus is noisy ( see Materials and Methods ) , it is reasonable for the memory system to use information from multiple trials to aid in reconstruction . In our model , this is accomplished at retrieval by “smoothing” over ( or blurring together ) the traces of trials that occurred nearby in time . This blurring removes noise under the assumption that stimuli change slowly over time and hence the underlying signal is temporally autocorrelated ( whereas the noise is not ) . Formally , this corresponds to a form of Kalman smoothing [40] . However , it is important to not smooth over instances that are very different from each other ( i . e . , across time points where an abrupt jump occurred and as a result the signal is no longer autocorrelated ) . Inference over multiple dynamical modes remedies this problem by segmenting the time series into parts that are each internally smooth; our smoothing algorithm operates within but not across these modes ( note that even when there is only a single dynamical mode , smoothing can still reconstruct individual stimuli , rather than blurring them all together , because a representation of each stimulus is available to the retrieval system ) . A formal description of this smoothing algorithm is given in the Materials and Methods . To test how well our proposed model fit participants' data throughout the experiment , we fit several variants of the DP-KF and KF models to participants' responses on prediction trials ( in which participants had to predict the next version of the line ) , holding out the responses on reconstruction trials for validation and comparison between the models ( see Materials and Methods for details of the model-fitting methods ) . Four model variants were constructed from the full model by restricting parameter values as follows: Figure 4B-E shows the predicted reconstruction biases for each of these models . Unlike our participants , neither the KF models nor the stationary DP-KF model showed a cross-over interaction between jump/gradual and start/end . In contrast , the DP-KF model showed a cross-over interaction effect ( ) . Thus among the four alternatives , only the DP-KF model adequately captured the experimental results . We quantitatively compared the fits of the different models in two ways . First , we performed cross-validation by splitting the blocks into two halves ( even- and odd-numbered blocks ) , fitting the model to the trial-by-trial prediction data for one half of the blocks and computing the predictive log-likelihood of data for the other half of the blocks . Figure 5A shows the predictive log-likelihood of each model relative to the stationary KF model . The KF and DP-KF models performed similarly ( a paired-sample t-test revealed no significant difference , ) , and significantly better than their stationary variants ( ) . Our second model-comparison metric was the predictive log-likelihood of participants' reconstructions . Note that the models were not fit to the reconstruction data , so there is no need to penalize for model complexity: overfitting the prediction-trials data due to too many degrees of freedom will automatically lead to poorer results when trying to predict the reconstruction trials . Figure 5B shows the predictive log-likelihood of each model relative to the stationary KF . According to this measure , the DP-KF model outperformed both the KF variants ( ) and performed marginally better than the stationary DP-KF ( ) . To illustrate the DP-KF model's accuracy in predicting reconstructions , we computed the Pearson correlation coefficient between the human and model reconstructions for each participant separately , Fisher z-transformed this value , and performed a t-test against 0 for all participants . Correlations for both orientation and length were significant ( each , two-tailed t-test; Figure 6 ) . Finally , in keeping with our theoretical predictions , we found that the number of modes ( K ) inferred by the fitted DP-KF model was , on average , higher in the jump condition than in the gradual condition ( ; Figure 7 ) .
Several authors have proposed neural implementations of the KF [51] , [52] . Wilson and Finkel [52] derived an approximation of the KF that can be computed by a recurrent neural network when the prediction error is small . Intriguingly , when the prediction error is large , their approximation ‘breaks down’ by creating two bumps in the posterior distribution ( rather than one as in the exact KF ) with each bump implementing an independent KF . Our theory suggests a normative account of this feature , since a network that creates multiple bumps is precisely what is required by the DP-KF algorithm . Pursuing this connection is an exciting direction for future research . Work on change detection [11] , [53]–[59] addresses a similar question: how does the brain detect a change in the statistics of sensory signals ? The study of Nassar et al . [56] , for example , showed that humans use the recent history of prediction errors to determine when a change has occurred . This work differs from our own in several ways . First , most existing change-detection theories assume stationary sensory statistics between jumps , whereas we allow for gradual change between jumps . Second , once a jump has occurred , theories of change detection assume that the statistics of earlier epochs are no longer relevant and can be discarded; in contrast , our model assumes that participants are able to retrieve statistics from earlier modes , and in general allows for the environment to return to earlier modes ( as noted above , our current experiment did not test this latter property of the model ) . Our work also intersects with research in cognitive psychology on the reuse of existing memory traces . For example , repeating items on a list tends to aid their recognition without degrading recognition of other items ( the null list-strength effect [60] ) . To explain this , Shiffrin et al . [8] assumed that repetition of items results in refinement of existing traces , rather than formation of new traces . Thus , there must be some reuse of memory traces . The question , then , is what counts as a repetition . Visually similar stimuli such as those used in our experiment may be judged by the memory system to be essentially the same item ( i . e . , a “repetition” ) . Our theory further asserts that small changes in these “repetitions” drive modification of existing memories , but not formation of new memories . This is similar to what Bower and Winzenz [7] dubbed the “reallocation hypothesis , ” according to which inputs are matched to memory traces and incorporated into an existing trace if the match is sufficiently high; otherwise , the input is routed to a new trace ( see also [9] ) . Interestingly , evidence suggests that failure to recognize a new context can sometimes lead to neither outcome: using an auditory statistical learning paradigm , Gebhart et al . [61] found that changes in structural information can go undetected without the aid of additional cues ( e . g . , sounds marking the transition between structures ) , preventing participants from learning new structures . This suggests that future models should incorporate a mechanism that allows some information to evade both old and new memories . The dynamically updated posterior posited by our model bears some resemblance to the drifting context vector posited by several models in the memory literature [62] , [63] . For example , the Temporal Context Model ( TCM ) introduced by Howard and Kahana [63] assumes that list items are bound to a context vector that is essentially an average of recently experienced items . In earlier work [64] , we operationalized the context vector as a posterior over latent “topics” that play the same role as modes in the present paper . In our current theory , items are bound to modes in much the same way that items are bound to the context vector in TCM . The connection to TCM also highlights the way in which episodic and semantic memory are deeply intertwined in contemporary theories: “episodic” traces of individual items become bound to “semantic” representations that average over multiple items [65] . Likewise in our model , episodic and semantic components are intertwined: a separate trace for each sensory stimulus is stored , but the traces are effectively blurred together by the smoothing operation during retrieval . Although the idea of separate episodic and semantic memory systems has been very influential [13] , it has been known since Bartlett's investigations [66] that semantic knowledge exerts strong constraints on many aspects of episodic memory [67] , [68] . A similar rapprochement has emerged in theories of category learning , where “episodic” ( exemplar ) and “semantic” ( prototype ) representations are combined to form varying levels of abstraction [35] , [69] , [70] . Another related line of work concerns the effects of novelty on memory . Our model predicts that a novel stimulus is more likely to be encoded in a separate trace compared to a familiar stimulus , making it less likely that the novel stimulus will suffer interference from other stimuli at retrieval . This prediction has been confirmed many times in the form of the von Restorff effect [71] . Note that while the von Restorff effect reflects proactive interference ( older memories interfering with the retrieval of newer memories ) and our experiment tested retroactive interference ( newer memories interfering with the retrieval of older memories ) , according to our model these are essentially due to the same process of grouping of different observations into temporally extended episodic memory traces . The idea of comparing gradual and abrupt changes as a means of influencing memory updating has also been explored in the motor control literature [72]–[74] . For example , Kagerer et al . [72] had participants make arm movements to a target and then introduced a perturbation ( by rotating the visual feedback ) either gradually or abruptly . Participants adapted to the perturbation; following the removal of the perturbation , participants exhibited an after-effect in which movement errors were in the direction opposite to the perturbation . Kagerer et al . found that the after-effect was smaller for participants in the abrupt condition than in the gradual condition . This pattern of results is consistent with the idea that two separate motor memories were formed in the abrupt condition , thereby allowing the pre-perturbation memory to be reinstated quickly . The larger after-effect in the gradual condition suggests that in that case the gradual perturbation led to modification of the original memory . Such modifications can be long-lasting: Yamamoto et al . [75] have shown that learning a gradually changing motor task produces a motor memory that can be recovered over a year later . Finally , we have recently reported related findings in the domain of Pavlovian fear conditioning [41] . Rats learned to associate a tone with a foot-shock . Subsequently , one group of rats were presented with the tone in the absence of shock ( standard ‘extinction’ of the tone-shock association ) . A second group of rats experienced the same number of tones , with the the tone-shock contingency only gradually reduced to zero ( that is , to full extinction ) . Although all rats showed similarly diminished fear of the tone at the end of the ‘extinction’ phase , rats in the standard extinction condition exhibited subsequent recovery of fear ( as is typically seen after extinction training ) , whereas rats in the gradual condition showed no evidence of fear recovery . These findings are consistent with the idea that the fear memory is more likely to be modified by extinction training in the gradual condition , thereby reducing the probability of later recovery .
In this paper , we empirically investigated a fundamental prediction that models of change detection make for memory . If , as we hypothesize , new experience is incorporated into old memories based on similarity , then abrupt change ( i . e . , dissimilar data ) should prompt the creation of a new memory trace , and thus protect old memories from being modified by new data , whereas gradual change will not . Our experimental results confirm this prediction , thereby providing support for a statistical account of how continuous experience is parsed into discrete memory traces . We conclude that memories are not simply a record of our ongoing experiences; the organization of memory traces reflects our subjective inferences about the structure of the world that surrounds us .
The experiment was approved by the Institutional Review Board at Princeton University . 32 undergraduate students received course credit or payment ( $12 per hour ) for participating in the experiment . The experiment was approved by the Institutional Review Board at Princeton University . The stimuli consisted of oriented line segments that changed in orientation and length on every trial . Each line segment was generated from the previous one by ( randomly ) adding or subtracting a fixed length ( 0 . 89 mm ) and a fixed angle ( 14 . 4° ) , thus generating a 45° ‘move’ in an orientation/length space in which one unit was 14 . 4° and 0 . 89 mm , respectively . ‘Moves’ were restricted so that the new line segment did not overlap with the previous line segment ( that is , there was no ‘backtracking’ in orientation/length space; see Figure 3 ) . Jumps were also at a 45° angle , but traversed a distance 4 times as long as the other steps ( i . e . , 3 . 6 mm length and 57 . 6° angle ) . Jumps always occurred ( if they did ) in the middle of the trajectory ( between trials 9 and 10 ) , and were unsignaled to the participant . Finally , in generating trajectories through orientation/length space , we required the Euclidean distance between the start and end points to lie within a narrow range ( 60–70% of the maximum possible distance ) regardless of the condition ( jump or gradual ) . Examples of jump and gradual trajectories are shown in Figure 3 . Participants played 12 blocks of the task ( 6 jump trajectories and 6 gradual trajectories , randomly interleaved ) . Each block consisted of a sequence of 18 prediction trials . A timeline showed participants the serial position of each trial in a block . On each prediction trial , participants used a mouse to adjust the orientation and length of a line on the screen so as to predict the next observed line . After making their prediction , participants were shown the true line and awarded points based on how accurate their prediction was . The prediction task was aimed at encouraging encoding of the different line segments in memory , and also provided data for fitting our models ( see below ) . At the end of the block , participants were given a reconstruction trial; on this trial , they were shown an arrow pointing toward a point on the timeline and asked to reconstruct the line segment they saw on that trial . Participants were always asked to reconstruct one of the first 3 trials in the block . No feedback was given on reconstruction trials . Let denote the estimated stimulus for time t given all observations up to the time of retrieval conditional on . Kalman smoothing [40] constructs this estimate through a backward recursion: ( 7 ) for each dimension d . In essence , smoothing combines the filtered estimate with information from the future propagated backward in time . We take to be the model's prediction for a participant's reconstruction of the stimulus shown at time t . Prior to model-fitting , the stimulus values ( length and orientation ) were rescaled to [0 , 100] . To model responses , we assumed that participants report the posterior mean , corrupted by anisotropic Gaussian noise ( with variances and , for length and orientation , respectively . Depending on the model variant , the noise variance r , the response noise variance v , the diffusion noise variance q , the stickiness parameter β and the concentration parameter α were treated as free parameters and fit to each participant's data by minimizing the negative log-likelihood of each participant's predictions using a numerical optimizer ( the routine fmincon in Matlab ) , while constraining parameters to lie in the appropriate range . To prevent implausibly large values of v , q and r , we constrained these to be less than 10 , 30 and 20 , respectively , although our results do not depend on these precise values . To avoid local minima , the optimization was run from 3 randomly chosen starting points . We assumed that responses were generated from the filtered state estimate ( or smoothed state estimate , in the case of retrieval ) , corrupted by Gaussian noise with anisotropic noise variance ( and ) . For the KF model , α was set to 0 . We set the prior covariances to be , instantiating an approximately uniform distribution over mode starting points . Reconstruction trials were not used in any of the fitting procedures . To model noise in the reconstruction process , we added a constant of 5 to the sensory noise variance ( rd ) . This value was chosen by hand , but the results were not sensitive to its precise value . | When do we modify old memories , and when do we create new ones ? We suggest that this question can be answered statistically: The parsing of experience into distinct memory traces corresponds to inferences about the underlying structure of the environment . When sensory data change gradually over time , the brain infers that the environment has slowly been evolving , and the current representation of the environment ( an existing memory trace ) is updated . In contrast , abrupt changes indicate transitions between different structures , leading to the formation of new memories . While these ideas fall naturally out of statistical models of learning , they have not yet been directly tested in the domain of human memory . In this paper , we describe a model of statistical inference that instantiates these ideas , and test the model by asking human participants to reconstruct previously seen visual objects that have since changed gradually or abruptly . The results of this experiment support our theory of how the statistical structure of sensory experiences shapes memory formation . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Conclusions",
"Materials",
"and",
"Methods"
] | [
"psychology",
"cognitive",
"psychology",
"social",
"sciences",
"cognition",
"computational",
"neuroscience",
"memory",
"biology",
"and",
"life",
"sciences",
"computational",
"biology",
"cognitive",
"science",
"neuroscience",
"learning",
"and",
"memory"
] | 2014 | Statistical Computations Underlying the Dynamics of Memory Updating |
This study aimed to compare the prevalence of Mycobacterium tuberculosis complex ( MTBc ) lineages between direct genotyping ( on sputum ) and indirect genotyping ( on culture ) , to characterize potential culture bias against difficult growers . Smear-positive sputa from consecutive new tuberculosis patients diagnosed in Cotonou , ( Benin ) were included , before patients had started treatment . An aliquot of decontaminated sputum was used for direct spoligotyping , and another aliquot was cultured on Löwenstein Jensen ( LJ ) medium ( 90 days ) , for indirect spoligotyping . After DNA extraction , spoligotyping was done according to the standard method for all specimens , and patterns obtained from sputa were compared versus those from the derived culture isolates . From 199 patient’s sputa , 146 ( 73 . 4% ) yielded a positive culture . In total , direct spoligotyping yielded a pattern in 98 . 5% ( 196/199 ) of the specimens , versus 73 . 4% ( 146/199 ) for indirect spoligotyping on cultures . There was good agreement between sputum- and isolate derived patterns: 94 . 4% ( 135/143 ) at spoligotype level and 96 . 5% ( 138/143 ) at ( sub ) lineage level . Two of the 8 pairs with discrepant pattern were suggestive of mixed infection in sputum . Ancestral lineages ( Lineage 1 , and M . africanum Lineages 5 and 6 ) were less likely to grow in culture ( OR = 0 . 30 , 95%CI ( 0 . 14 to 0 . 64 ) , p = 0 . 0016 ) ; especially Lineage 5 ( OR = 0 . 37 95%CI ( 0 . 17 to 0 . 79 ) , p = 0 . 010 ) . Among modern lineages , Lineage 4 was over-represented in positive-culture specimens ( OR = 3 . 01 , 95%CI ( 1 . 4 to 6 . 51 ) , p = 0 . 005 ) . Ancestral lineages , especially M . africanum West African 1 ( Lineage 5 ) , are less likely to grow in culture relative to modern lineages , especially M . tuberculosis Euro-American ( Lineage 4 ) . Direct spoligotyping on smear positive sputum is effective and efficient compared to indirect spoligotyping of cultures . It allows for a more accurate unbiased determination of the population structure of the M . tuberculosis complex . ClinicalTrials . gov NCT02744469
Tuberculosis ( TB ) , caused by bacteria of the Mycobacterium tuberculosis complex ( MTBc ) , remains a public health problem . Globally , over 8 million new patients with TB disease arise each year , including 2 million deaths . The vast majority ( 95% ) of global TB is detected in limited-resource countries [1] , including West-Africa . Each year in Benin , over 4000 cases of TB are detected , and the incidence of smear-positive pulmonary TB is 39 per 100000 inhabitants . Genotypic characterization is important in order to understand the population structure of the MTBc for better insights into endemic- and epidemic strains and to identify instances of nosocomial transmission or laboratory contamination . M . tuberculosis sensu stricto and M . africanum sub-species within the MTBc have been subdivided into 7 main lineages of human importance [2 , 3] . These 7 MTBc lineages are classified as ancestral ( or ‘ancient’ ) ( Lineages 1 , 5 , 6 ) [4 , 5] , intermediate ( Lineage 7 ) [3 , 4] and modern lineages ( Lineages 2 , 3 , 4 ) [4] . Lineage 5 ( M . africanum West African 1 ) and Lineage 6 ( M . africanum West African 2 ) are only found in West- and Central Africa , where they cause up to 40% of all TB [6 , 7] . Recent reports suggested a decrease in prevalence of M . africanum in some West-African countries [8–10] . Whether methodological issues explain the apparent disappearance of M . africanum has not been excluded to date . For the determination of the population structure of the MTBc , genotyping is usually applied on culture isolates [11] . M . africanum grows significantly slower than the other members of the MTBc ( M . tuberculosis sensu stricto ) [12] and cultures should be incubated for 90 days rather than the usual 56 days , before reporting a negative result [13] . However , even this extended incubation time may not permit recovery of M . africanum isolates at the same rate as M . tuberculosis , and thus bias the population structure derived from cultured isolates , especially in settings where M . africanum is endemic . Differences in expression of genes involved in metabolism pathways of the various MTBc lineages may also affect their growth in culture , as recently reported for M . africanum Lineage 6 which has an under-expression for the gene ( Dos R ) involved in adaptation to lower oxygen tension relative to Lineage 4 [14] . For isolation , of some MTBc species , including M . africanum , the need for pyruvate to support growth in culture [15] has been known for a long time [16] . Few studies evaluated genotyping , such as spoligotyping , directly on clinical specimens such as sputa [17 , 18] , sputum smears [19] , paraffin wax-embedded tissues [20] or mummified remains of human [20] . Only one study from Brazil , where M . africanum is not endemic , compared spoligotyping on sputum to spoligotyping from the respective isolates [21] . Moreover , to the best of our knowledge , no study has investigated whether the proportional prevalence of MTBc lineages differs among specimens with a positive culture versus culture-negative specimens . In this study , we determined the performance of spoligotyping on sputum ( ‘direct spoligotying’ ) relative to its yield on culture ( ‘indirect spoligotyping’ ) for genotypic characterization of MTBc , and evaluated for a potential culture bias against difficult-growers , even when incubation was prolonged to enhance detection of M . africanum .
This study is part of the BeniDiT study that has been approved by the national ethics committee of Benin , the Institutional Review Board of the Institute of Tropical Medicine of Antwerp , Belgium and the ethics committee of the University of Antwerp . It is registered on ClinicalTrials . gov under the registration number NCT02744469 . All sputa were anonymized before laboratory analyses . Smear-positive sputa from consecutive new TB patients diagnosed in the Centre National Hospitalier Universitaire de Pneumo-Phtisiologie in Cotonou , ( Benin ) were prospectively included ( Fig 1 ) , before patients initiated TB treatment . Laboratory analyses were conducted in the National Reference Laboratory for Mycobacteria ( Laboratoire de Référence des Mycobactéries ) in Cotonou , Benin . Spoligotype patterns were recorded in an Excel file using a binary code ( 1 for presence of a given spacer and 0 for the absence of a given spacer ) . Entered profiles were verified and validated by an independent person . The persons who typed and validated the data were blinded to the spoligotype pattern of the corresponding sputum or isolate . The Excel file was loaded into the TBlineage database http://tbinsight . cs . rpi . edu/run_tb_lineage . html [26] for lineage assignment . Sub-lineages ( spoligotype families ) were obtained by loading the Excel file with the spoligotype patterns in the SPOTCLUST database http://tbinsight . cs . rpi . edu/run_spotclust . html [27] . Data was analyzed using the statistical software Stata/IC 12 . 0 ( StataCorp ) . The two-group proportion test or the Fisher Exact test was used to analyze independent data . Mc Nemar Chi2 test was used to compare paired proportions . Two-sided p-values were calculated and for differences in proportion , odds ratios were calculated along with 95% confidence interval . Differences were considered statistically significant when p<0 . 05 .
From the 199 recruited TB patients and their sputum samples , 146 ( 73 . 4% ) yielded a positive culture , whereas 36 ( 18 . 1% ) remained negative and 17 ( 8 . 5% ) were contaminated . Spoligotype patterns were obtained for all the 146 culture isolates , and for 196 of the 199 sputa , yielding an overall success for direct spoligotyping of 98 . 5% . All of the extraction controls and amplification/hybridization controls yielded expected results , and repeat spoligotyping for discordant results between sputum and culture confirmed the original patterns . Stratified by culture result , direct spoligotyping reached a success of 100% ( 53/53 ) for culture-negative or contaminated sputa , and 98% ( 143/146 ) for culture-positive sputa . Microscopy was negative in 6 sediments after decontamination , while all the others had positive microscopy . Of the 6 microscopy negative sediments , 3 failed direct spoligotyping and 2 others had a negative culture . Spoligotype patterns were available for 98 . 5% of sputa versus 73 . 4% of cultures ( Table 1 ) . Comparison between respective direct and indirect spoligotypes showed 94 . 4% ( 135/143 ) agreement . In total three types of discrepancies were observed ( Fig 2 ) : mixed infection with one pattern found in sputum and the other found in the culture isolate ( n = 3 , discrepancy 5–7 ) , mixed infection with overlapping spoligotype patterns in sputum ( n = 2 , discrepancies 1 and 4 ) , and false negative ( missing ) spacers in sputum ( n = 3 , discrepancies 2 , 3 and 8 ) . Five ( 5 ) of these patterns led to inter-lineage discrepancies , and three ( 3 ) to intra-lineage discrepancies . For inter-lineage discrepancies , sub ( lineages ) observed in isolates are shaded grey . The five inter-lineage discrepant pairs ( discrepancy 4–8 ) showed patterns suggestive of a simultaneous presence of ancestral and modern lineages , while these yielded only the ancestral lineage in sputum and only the modern lineage in culture . Three ( discrepancy 5–7 ) of these five inter-lineage pairs showed this ( ancestral M . africanum in sputum and modern Lineage 4 in culture ) , without any other possible explanation , while the other two ( discrepancy 4 and 8 ) can also be interpreted as follows . Inter-lineage pair 8 and intra-lineage pairs 2 and 3 showed patterns suggestive of false negative spacers in sputum ( spacer present in isolate but absent in sputum ) . Intra-lineage pair 1 and inter-lineage pair 4 showed patterns suggestive of overlapping spoligotype signatures in sputum ( discrepancy 1 and 4 ) and/or in isolate ( discrepancy 1 ) . Discrepancy 4 suggested an overlapping of Lineages 2 and 4 signatures in sputum , with only the Lineage 2 grown in culture . Discrepancy 1 was suggestive of overlapping spoligotype signatures in sputum and in culture isolate that could be a mixture of Lineages 2 and 4 . The distribution of lineages in culture-positive sputa versus directly in sputum with unsuccessful culture differed , with Lineage 5 ( M . africanum West African 1 ) being significantly less prevalent among culture-positive sputa ( OR = 0 . 48 95%CI ( 0 . 24 to 0 . 94 ) p = 0 . 033 , Table 2 ) . This association became more significant when contaminated cultures were excluded from the analysis ( OR = 0 . 37 , 95%CI ( 0 . 17 to 0 . 8 ) , 21% vs 41 . 7% , p = 0 . 011 , Table 2 ) . Ancestral lineages ( Lineages 1 , 5 and 6 ) were significantly less present among culture-positive sputa ( OR = 0 . 33 , 95%CI ( 0 . 16 to 0 . 7 ) , 37 . 1% vs 63 . 9% , p = 0 . 004 , Table 2 ) . Lineage 4 ( M . tuberculosis Euro-American ) , a modern lineage , was most overrepresented in culture-positive sputa ( OR = 2 . 81 , 95%CI ( 1 . 30 to 6 . 03 ) 55 . 2% vs 30 . 5% , p = 0 . 008 , Table 2 ) . Excluding discrepant spoligotypes between direct and indirect spoligotype analysis , the association gained further statistical significance . The odds of detecting ancestral lineages in positive-cultures was 0 . 30 fold ( 95% CI ( 0 . 14 to 0 . 64 ) ; p = 0 . 0016 ) less in positive-cultures relative to negative cultures , especially Lineage 5 ( OR = 0 . 37 95%CI ( 0 . 17 to 0 . 79 ) ; p = 0 . 010 ) ( S1 Table ) . Modern lineages were inversely more represented in positive-culture specimens ( OR = 3 . 31 , 95%CI ( 1 . 57 to 6 . 99 ) , p = 0 . 0016 ) , especially Lineage 4 ( OR = 3 . 01 , 95%CI ( 1 . 4 to 6 . 51 ) , p = 0 . 005 ) ( S1 Table ) . The prevalence of L1 , L5 , L6 tended to be higher among culture-negative specimens ( respectively 8 . 3% , 41 . 7% , 13 . 9%; S1 Table ) than in culture-positive specimens ( 7 . 4% , 20 . 7% , 6 . 7%; S1 Table ) . In contrast the prevalence of L2 , L3 , L4 tended to be lower among culture-negative specimens ( 5 . 6% , 0% , 30 . 5% ) than in culture-positive specimens ( 6 . 7% , 1 . 5% , 57 . 0%; S1 Table ) . This justified the analysis in subgroup of ancestral and modern lineages . The distribution of sub-lineages ( families ) within Lineage 4 showed that LAM 10 , LAM 9 , LAM 1 , T1 , T2 , Haarlem 1 , Haarlem 2 , Haarlem 3 , X3 families were present in new TB patients in Cotonou . This distribution of Lineage 4 families did not differ significantly in culture-positive versus culture-negative sputa ( S2 Table ) . Almost all positive cultures were positive within 8 weeks of incubation , while prolonged incubation only yielded one additional positive culture . This was a Lineage 5/ M . africanum West African 1 strain . Among positive cultures , over half ( 5/9: 55 . 5% ) of the Lineage 6/ M . africanum West African 2 cultures became positive between 6 to 8 weeks of incubation , whereas most of positive cultures from other lineages specimens were positive within 6 weeks: 10/11 ( 90 . 9% ) for Lineage 1 , 10/10 ( 100% ) for Lineage 2 , 2/2 ( 100% ) for Lineage 3 , 83/85 ( 97 . 6% ) for Lineage 4 and 28/29 ( 95 . 6% ) for Lineage 5 . Despite the prolonged incubation period , over a third of specimens from each M . africanum lineage remained culture negative ( 34 . 1% for Lineage 5 and 35 . 7% for Lineage 6 ) , while for other lineages , none ( Lineage 3 ) or fewer specimens ( 21 . 4% for Lineage 1 , 16 . 7% for Lineage 2 , 11 . 5% for Lineage 4 ) remained negative ( Table 3 ) . The sediment smear of the culture negative specimens from Lineage 1 and 2 had low AFB-grading or were negative whereas nearly all ( 14/15 ) the culture negative specimens from Lineage 5 had high smear grading ( S3 Table ) .
Our results show that indirect spoligotyping provided spoligotype profiles for all 146 culture-positive specimens ( 73 . 4% ) , while direct spoligotyping provided spoligotyping profiles for 50 more sputa ( + 25 . 1% of all 199 specimens , 95% CI ( 18 . 1% to 32 . 1% ) ) that would not otherwise be genotyped in the absence of an isolate . Direct spoligotyping on sputum after semi-automated DNA extraction using Maxwell DNA tissue purification kit , has a high sensitivity ( 98 . 5% ( 196/199 ) ) to detect MTBc genotypes . The 98 . 5% ( 196/199 ) overall availability of spoligotype profiles in our study is higher than the 90 . 9% ( 159/175 ) found on smear-positive sputa by Goyal et al . in Ghana ( p = 0 . 001 ) [18] and the 49 . 1% ( 28/57 ) found by Heyderman et al . in Zimbabwe [17] . This could be explained by the variability of methods used for DNA extraction from sputa and/or the variability in PCR reagents mix . The overall availability of spoligotype profiles on sputa in our study ( 98 . 5% ) is also higher than the 77 . 7% ( 41/53 ) found by Suresh et al . and 90 . 5% ( 19/21 ) by Zanden et al . on smears [19 , 28] , which likely have less mycobacterial DNA than a 200 μL sputum sample . The fact that- within mixed infections- ancestral lineages are found with direct spoligotyping on sputum , suggests that the load of ancestral lineage bacilli in vivo exceeds the load of the modern lineage bacilli , with subsequent out-competition in culture by the latter . Sarkar et al . also found that Lineage 4 grows more rapidly ( in liquid medium ) than other lineages including Lineage 1 , an ancestral lineage [29] . Moreover , Gehre et al . found that Lineage 6 , another ancestral lineage , grows more slowly than MTBc lineages other than M . africanum in liquid medium [12] . Sputum provided the most representative population distribution of lineages of the MTBc in new TB patients in Cotonou , with more TB due to ancestral lineages , including M . africanum . This distribution did not alter when the three isolates which sputum failed direct spoligotyping were added ( two from Lineage 4 and one from Lineage 5; Table 2 ) . The ‘most true’ distribution is the one combining profiles obtained directly from sputum , complemented by profiles on isolates from failed direct spoligotyping , and includes: 8 . 0% ( 16/199 ) for Lineage 1 , 5 . 6% ( 11/199 ) for Lineage 2 , 1% ( 2/199 ) for Lineage 3 , 51 . 8% ( 103/199 ) for Lineage 4 , 25 . 1% ( 50/199 ) for Lineage 5 , 8 . 5% ( 17/199 ) for Lineage 6 , or 41 . 7% for ancestral lineages , and 33 . 7% for M . africanum ( Table 2 ) . This distribution would have been different if smear-negative specimens were also genotyped , as it had been previously reported that M . africanum is more likely to be found in lower grade smear-positive specimens [30] , and Lineage 6 is associated with HIV infection [31] , which is in turn associated with smear-negativity [32–34] . The comparison of the distribution of MTBc lineages in a similar population , also consisting of consecutive smear-positive new pulmonary TB patients aged at least 15 years old of Cotonou in year 2005–2006 on cultured isolates [9 , 35] , to the one obtained in our study indirectly on cultured isolates from similar patients in Cotonou 10 years later , showed that the previous prevalence of Lineage 1 ( 7 . 7% ) , Lineage 2 ( 10 . 3% ) , Lineage 3 ( 0% ) , Lineage 6 ( 6 . 2% ) are similar to our findings in this study ( respectively: 7 . 5% , 6 . 8% , 1 . 4% and 6 . 2% ) . Yet the prevalence of Lineage 4 ( 42 . 3% in year 2005–2006 ) has increased to 58 . 2% ( difference: +15 . 9% ) , and Lineage 5 prevalence ( 30 . 9% in year 2005–2006 ) has decreased to 19 . 9% ( difference: -11 . 0% ) . While we demonstrate that the present L5 prevalence of 19 . 9% on indirect genotyping is an underestimate , even the present ‘true’ L5 prevalence of 25 . 1% on direct genotyping would constitute a decline from the L5 prevalence of 30 . 9% on indirect genotyping in 2005–2006 . Other authors also reported a decrease of M . africanum [8 , 9] . Our results show that rates of culture isolation from smear-positive pulmonary TB patients are lower for Lineages 5 and 6 of the MTBc , despite prolonged incubation of cultures for 90 days [13] . Extending the incubation time beyond 6 weeks enhanced isolation of Lineage 6 ( between 6–8 weeks ) yet did not further augment the isolation rate . Ancestral lineages , especially Lineage5/M . africanum West African 1 are ‘difficult-growers’ in culture relative to modern lineages , such as Lineage 4 . The decreased odds of ancestral lineages to grow in culture could partly be due to culture procedures ( culture medium or decontamination method ) that were originally developed for modern lineages prevalent in Europe . Ofori-Anyinam et al . reported that Lineage 6 as compared to Lineage 4 , is more adapted to microaerobic growth [14] which may be the reason for its impaired growth on solid media such as LJ used in this study . Furthermore Gehre et al . found that Lineage 6 has mutations in genes that lead to its attenuated growth in vitro[12] . Such genetic analyses need to be conducted on Lineage 5 in order to understand the reasons for its difficult growth in vitro . Further studies should also be conducted on other lineages to find out the genetic basis of their in vitro growth pattern . To the best of our knowledge , this is the first demonstration that ancestral lineages are underrepresented in positive cultures . Direct spoligotyping is thus more appropriate for unbiased determination of MTBc population structure in settings where ancestral lineages , including M . africanum , are common . The implications of our findings also affect MTBc population structures generated with different typing methods , including whole genome sequencing . Such studies tend to be culture-based , given the ongoing limitations of sequencing entire MTBc genomes directly from clinical material . While direct genome sequencing is technically feasible given sufficient coverage , in practice the associated costs are prohibitive . Studies to date have shown limited coverage , precluding SNP cut-offs for molecular epidemiological studies [36] . Optimized methods to sequence genomes directly from clinical material are thus urgently needed . One strength of this study is the prolonged incubation time , to maximize the yield of M . africanum in culture . Other strengths include the paired design for the comparison of direct spoligotyping versus indirect spoligotyping and the inclusion of multiple controls and blinding of operators , and the fact that the study was conducted in a setting where M . africanum is prevalent . A limitation is that only LJ medium was used , and we do not know whether other medium , such as liquid medium ( known to enable the growth of more non-tuberculous mycobacteria ) may also favor the growth of ancestral MTBc lineages . This study was conducted only on fresh unshipped acid-fast bacilli positive sputa from new TB patients . Culture positivity may be worse if sputa had to be shipped from peripheral laboratories to a reference or central laboratory where spoligotyping can be done . Another limitation is that the number of specimens with Lineages 1 , 2 , 3 , 6 among culture negative specimens under-powered the estimation of any difference in the prevalence of these individual lineage among culture-negative versus culture-positive specimens . So , although no evidence of such difference in prevalence among culture-negative versus –positive specimens was found in Lineages 1 , 2 , 3 , 6 in the present study , such difference could be tested for in settings with higher prevalence of these lineages . In conclusion , ancestral lineages especially M . africanum West African 1 ( Lineage 5 ) , are less likely to grow in culture , unlike modern lineages especially M . tuberculosis Euro-American ( Lineage 4 ) . Direct spoligotyping on sputum is effective , and saves effort and time compared to indirect spoligotyping of cultures . It has an important gain in sensitivity , especially for ancestral lineages that may not yield a positive culture , allowing a more precise unbiased determination of the population structure of the MTBc . It can also be used for specimens from patients under TB treatment and other specimens in which culture may be negative or contaminated . While differences in culture isolation technique and reliance on indirect spoligotyping may partially account for the reduction in the prevalence of M . africanum observed in several West African countries [8 , 9] , comparison of our findings with the genotyping study from Cotonou 10 years ago suggests that the decline in M . africanum is not explained by the lower sensitivity of culture isolation . The potential decline of M . africanum lineages will be addressed in more depth in a larger ongoing study on the population structure of the M . tuberculosis complex in Benin , in which direct genotyping will be applied , given the findings presented in this manuscript . Further studies must be conducted to investigate whether culture procedures ( medium , decontamination ) can be optimized for growth of ancestral lineages . Additional studies should address the frequency and role , if any , of a mixed infection between an ancestral- and modern lineage in the faster spread of modern lineages [4] and disappearance of ancestral lineages [8 , 9] . | The vast majority ( 95% ) of tuberculosis ( TB ) patients worldwide live in low-income countries , including in West-Africa . Typing the bacteria responsible for TB ( tuberculosis; Mycobacterium tuberculosis complex ) is important for targeted TB control . Typing is usually performed on isolates obtained after the culture isolation of TB bacteria in the sputa from patients . However , cultures can be false negative , and some ‘ancestral’ strains , only found in West-Africa ( Mycobacterium africanum ) , require more time ( 90 days versus the usual 56 days ) to grow in culture . To characterize potential culture bias against such “difficult growers” , we compared the performance of direct typing ( on sputum ) relative to its yield on culture isolates . We found that ancestral types of TB bacteria were significantly less likely to grow in culture despite the 90-day incubation . This suggests that typing results of cultured isolates are not representative of the diversity in the population of TB bacteria causing disease in patients . Typing sputum directly is effective and can be used for a more precise , unbiased determination of the proportion of different TB bacteria in a population . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"africans",
"root",
"structure",
"plant",
"anatomy",
"medicine",
"and",
"health",
"sciences",
"body",
"fluids",
"tropical",
"diseases",
"ethnicities",
"bacterial",
"diseases",
"plant",
"science",
"geology",
"molecular",
"biology",
"techniques",
"genotyping",
"bacteria",
"extraction",
"techniques",
"petrology",
"research",
"and",
"analysis",
"methods",
"sputum",
"infectious",
"diseases",
"mucus",
"tuberculosis",
"actinobacteria",
"molecular",
"biology",
"people",
"and",
"places",
"dna",
"extraction",
"anatomy",
"plant",
"roots",
"mycobacterium",
"tuberculosis",
"physiology",
"earth",
"sciences",
"sediment",
"biology",
"and",
"life",
"sciences",
"population",
"groupings",
"sedimentary",
"geology",
"organisms"
] | 2017 | Genotypic characterization directly applied to sputum improves the detection of Mycobacterium africanum West African 1, under-represented in positive cultures |
Human GWAS of obesity have been successful in identifying loci associated with adiposity , but for the most part , these are non-coding SNPs whose function , or even whose gene of action , is unknown . To help identify the genes on which these human BMI loci may be operating , we conducted a high throughput screen in Drosophila melanogaster . Starting with 78 BMI loci from two recently published GWAS meta-analyses , we identified fly orthologs of all nearby genes ( ± 250KB ) . We crossed RNAi knockdown lines of each gene with flies containing tissue-specific drivers to knock down ( KD ) the expression of the genes only in the brain and the fat body . We then raised the flies on a control diet and compared the amount of fat/triglyceride in the tissue-specific KD group compared to the driver-only control flies . 16 of the 78 BMI GWAS loci could not be screened with this approach , as no gene in the 500-kb region had a fly ortholog . Of the remaining 62 GWAS loci testable in the fly , we found a significant fat phenotype in the KD flies for at least one gene for 26 loci ( 42% ) even after correcting for multiple comparisons . By contrast , the rate of significant fat phenotypes in RNAi KD found in a recent genome-wide Drosophila screen ( Pospisilik et al . ( 2010 ) is ~5% . More interestingly , for 10 of the 26 positive regions , we found that the nearest gene was not the one that showed a significant phenotype in the fly . Specifically , our screen suggests that for the 10 human BMI SNPs rs11057405 , rs205262 , rs9925964 , rs9914578 , rs2287019 , rs11688816 , rs13107325 , rs7164727 , rs17724992 , and rs299412 , the functional genes may NOT be the nearest ones ( CLIP1 , C6orf106 , KAT8 , SMG6 , QPCTL , EHBP1 , SLC39A8 , ADPGK /ADPGK-AS1 , PGPEP1 , KCTD15 , respectively ) , but instead , the specific nearby cis genes are the functional target ( namely: ZCCHC8 , VPS33A , RSRC2; SPDEF , NUDT3; PAGR1; SETD1 , VKORC1; SGSM2 , SRR; VASP , SIX5; OTX1; BANK1; ARIH1; ELL; CHST8 , respectively ) . The study also suggests further functional experiments to elucidate mechanism of action for genes evolutionarily conserved for fat storage .
Human Genome-Wide Association Scans ( GWAS ) have been successful in discovering many genetic loci that are significantly associated with Body Mass Index ( BMI ) . These associations have been replicated across consortia consisting of many large and independent studies , sometimes numbering into hundreds of thousands of subjects . From a statistical point of view , the evidence that single nucleotide polymorphisms ( SNPs ) are tagging something real is overwhelming . However , progress has been slowed in moving beyond the discovery phase to a deeper understanding the biological significance of these findings , due to difficulties isolating the driving causal variants or even identifying the acting genes tagged by these GWAS variants . Most of the findings are not in gene coding regions . Indeed , many are intergenic , suggesting that much of the underlying modes of action of these loci may be regulatory . Unfortunately , our limited biological understanding of the regulome has hampered further progress . While the field has begun serious annotation of the regulatory regions of the genome with initiatives and resources such as ENCODE , RoadMap and GTEx , the annotation is still far from complete and the answers that are emerging are complex . As a result , many publications annotate the statistically significant SNPs simply with the “closest” gene , even though trans-acting regulatory sequences certainly exist[1] , and some enhancers have been shown to regulate multiple genes[2] . Recently , it was reported that rs1421085 T-to-C intron of the well-known obesity-associated FTO gene , disrupts a conserved motif for the ARID5B repressor , which leads to derepression of a potent preadipocyte enhancer and a doubling of the transcription factors IRX3 and IRX5 expression during early adipocyte differentiation . Irx3-deficient mice showed a 25–30% reduction of body weight , primarily through the loss of fat mass and increase in basal metabolic rate . Hypothalamic expression of a dominant-negative form of Irx3 reproduces the metabolic phenotypes of Irx3-deficient mice . Thus , IRX3 has been suggested as a functional long-range target of obesity-associated variants within FTO and represents a novel determinant of body mass and composition , by regulating the process of thermogenesis as they can prevent the process in which energy is turned into heat , thus stored as fat [3 , 4] . The above research supports the idea that an intronic location of an associated SNP does not even establish that the genetic effect is on that gene . Carrying out functional mapping of these GWAS-associated regions can provide valuable information to sort out which genes are causal to adiposity , and possibly provide biological insight into their action . While mouse models of obesity can serve as powerful platforms to functionally probe a small number of candidate genes , this approach is expensive and time consuming , limiting the number of genes that can be readily assessed . However , many important biochemical pathways involved in growth , metabolism , fat storage and retrieval are ancient and are therefore well conserved across the animal kingdom from C . elegans and Drosophila to rodents and humans . For example , forward genetic screens in C . elegans and Drosophila have identified conserved genes that regulate triglyceride storage[5 , 6] . Readily available genetic tools in Drosophila , including mutations and inducible RNA interference ( RNAi ) , coupled with the short life span , offer the opportunity for high-throughput functional screening of candidate genes , such as those proximal to GWAS putatively regulatory variants . A Drosophila genetic approach was recently used to follow up a small-scale GWAS for Alzheimer pathology[7] and type 2 diabetes mellitus and related metabolic disorders[8] . Capitalizing on this approach , we conducted a high throughput functional screen in Drosophila of all nearby genes to 78 BMI SNPs from two recently published GWAS meta-analyses to see if we could make progress in identifying the possible genes of action for these novel loci , as outlined in Fig 1 .
To further validate the Drosophila functional results , we queried several bioinformatic resources , including the Mouse Genome Informatics ( MGI ) website ( http://www . informatics . jax . org/ ) , which catalogues publication results of mouse experiments , as well as the International Mouse Phenotyping Consortium ( IMPC ) website ( http://www . mousephenotype . org/ ) , which also contains unpublished as well as published extensive phenotype characterizations screens of knock out ( KO ) experiments ( the results are summarized in S2 Table ) . At the IMPC website , at this time , there are results of whole organism knockdown experiments for only 8 of our 36 human-fly obesity genes ( many more are planned in the future ) . One of these , SBK1 , was pre-weening lethal as a whole organism knockdown , and thus could not be evaluated . For the other seven , there were extensive phenotypic characterizations of adult mice , including Dexa fat mass evaluations . Three of these genes showed highly statistically significant fat mass differences between the KO and WT mice: SETD1A ( P = 2 . 46E-08 ) , TCF7L2 ( P = 5 . 97E-10 ) and FOXO3 ( P = 3 . 41E-05 ) while two were nearly statistically significant different from WT: ZNF704 ( P = 0 . 086 ) and YPEL3 ( P = 0 . 082 ) and two showed no differences from WT ( CLUAP1 and PDK4 ) . Thus , 5 of the 8 of our genes that have thus far been interrogated by whole body KO in mice in the IMPC confirm our results . Obesity-related phenotypes have been previously published for two of the three most significant genes . TCF7L2 KO mice were shown to be leaner and have improved glucose tolerance[9] , and further , TCF7L2 was shown to negatively regulate adipocyte differentiation[10] . FOXO3 has been shown to be downregulated in the brains of high-fat diet induced obese mice[11] and mRNA of FOXO3 levels were associated with chicken growth traits , including fat body weight[12] . The MGI website listed confirmatory mouse-obesity evidence for several of the same genes as the IMPC , but no additional ones . In literature search of PubMed , we found suggestive or validation evidence for 5 more of the 36 human-fly obesity genes . Whole body siRNA of NTSC2 in mice showed increased lipolysis [13] and VTI1A was shown to interact with GLUT4 in adipocytes in mice[14] . More directly , PARK2 KO mice show decreased fat absorption and are leaner on high-fat diets[15] . VASP KO mice have reduced body weight and increased brown adipocytes[16] . They also show increased triglyceride accumulation in liver[17] . Finally , mouse expression and fly KDs confirm our findings for NUDT3 by Williams et al . 2015[18] . NUDT3 was significantly up regulated in the reward and feeding related regions of the hypothalamus and amygdala of the mouse brain . In all , we found validation evidence supporting obesity phenotypes in mice for 10 of the 36 genes found in our fly validation screen of cis genes near human BMI loci ( ZNF704 , SETD1A , VTI1A , TCF7L2 , FOXO3 , NUDT2 , PARK2 , VASP , and YPEL3 ) . However , it should be noted that few of the published or unpublished catalogued experiments , if any , actually reproduce the exact conditions of our fly screen . The IMPC KO screen , as well as most of the published mouse data available are on whole organism knock outs , whereas in our fly screen , we used tissue-specific drivers to confine the knock down only to the brain and to adipose tissue . This is an extremely important difference . Whole organism knock out animals typically experience a wide range of phenotypes across many systems and organs , and in some cases , very severe defects ( some are even lethal ) . Thus , the lack of concordance in obesity phenotypes in such experiments does not mean that our fly experiment has been properly tested for replication in the mouse and it failed . Rather , the fact that we already have found suggestive or strong evidence for 10 of the 36 of the fly genes in the mouse , demonstrates the utility of using the fly as a high-throughput functional screen to help us identify which genes the non-coding human GWAS statistical loci might be regulating , as an important step to moving from statistical association to mechanism of action . We used the GTEX resource to see if there is any evidence that the 78 BMI loci are eQTLs for any of the 36 human-fly obesity genes , in the relevant human tissues ( S2 Table ) . We considered 6 human tissues available in GTEX: Adipose Subcutaneous , Adipose Omentum , and Liver ( corresponding to the fly fat body ) , and Brain Hypothalamus , Brain Hippocampus and Pancreas ( corresponding to the fly brain—the last because clusters of cells in the brain of flies secrete insulin ) . We find 19 of the 36 gene-loci pairs show significant eQTL evidence in at least one relevant human tissue .
The interpretation of the effect of a candidate gene is straightforward when a human gene—single fly ortholog exist . But ~30% of human genes do not have fly orthologs and therefore cannot be evaluated . The gene ontology matches between human and Drosophila genes in many cases is difficult , especially when many human homologs exist for a single Drosophila gene or there are many Drosophila orthologs ( many-to-many ) . For these cases , we selected the best ortholog for KD; but assigning observed functions to specific genes was more difficult . The sensitivity of the model system might be limited due to the how accurately Drosophila models human obesity as measured via BMI . Also , our tissue-specific KDs interrogated gene effects only in the brain and the fat body , so we would miss effects that operated through other organs or tissues . While many fundamental processes in energy regulation are likely to be conserved , there will be complexities of human physiology that are not modeled well in insects . For selected genes , future studies will require validation in mammalian models .
The quantitative percent body fat ( %BF ) distributions in the adult flies were tested against the corresponding controls using an analysis of variance model as implemented in PROC GLM ( SAS v 9 . 4 . ) , with Dunnett’s Multiple Comparisons test , which corrects for multiple comparisons when common control sets are used for multiple experimental conditions , to provide a 5% experiment-wise error rate . | Human Genome Wide Association Studies have successfully found thousands of novel genetic variants associated with many diseases . While these undoubtedly point to new biology , the field has been slowed in exploiting these new findings to reach a better understanding of exactly how they confer increased risk . Many , if not most , appear to be regulatory not coding variants , so their immediate consequence is not obvious . A real rate limiting step is even identifying which gene these variants might be regulating , and in what tissues they are operating to increase disease risk . In the absence of any other information , a first order assumption is that they may be more likely to be regulating a nearby gene , and such variants are often initially annotated by the “nearest” gene until their function is more definitively validated . Exploiting the idea that many genes may have conserved function across species , we conducted a high-throughput screen of fruit-fly orthologs of human genes nearby 78 well validated GWAS variants for human obesity , in order to more precisely identify the gene ( s ) of action . We systematically knocked down the function of each of these nearby genes in the brain and fat-body of the flies , raised them on a standard diet , and compared their percent body fat with control flies , in order to validate which genes showed a fat response . 43% of the time when fly orthologs existed in the region , we were able to identify the causal gene . Interestingly , nearly half the time ( 46% ) , it was not the nearest gene but another nearby one that regulated fat . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"genome-wide",
"association",
"studies",
"invertebrates",
"body",
"weight",
"medicine",
"and",
"health",
"sciences",
"rna",
"interference",
"animals",
"animal",
"models",
"drosophila",
"melanogaster",
"model",
"organisms",
"physiological",
"parameters",
"experimental",
"organism",
"systems",
"genome",
"analysis",
"obesity",
"epigenetics",
"drosophila",
"research",
"and",
"analysis",
"methods",
"genome",
"complexity",
"lipids",
"genetic",
"interference",
"fats",
"gene",
"expression",
"genetic",
"loci",
"insects",
"arthropoda",
"biochemistry",
"rna",
"eukaryota",
"nucleic",
"acids",
"genetic",
"screens",
"gene",
"identification",
"and",
"analysis",
"physiology",
"genetics",
"biology",
"and",
"life",
"sciences",
"genomics",
"computational",
"biology",
"introns",
"organisms",
"human",
"genetics"
] | 2018 | A high throughput, functional screen of human Body Mass Index GWAS loci using tissue-specific RNAi Drosophila melanogaster crosses |
A severe neurological disorder , Guillain-Barré syndrome ( GBS ) is the leading cause of acute flaccid paralysis . Enhanced surveillance of GBS in Latin America and the Caribbean ( LAC ) following the 2015–2016 Zika virus ( ZIKV ) epidemic presents an opportunity to estimate , for the first time , the regional incidence of GBS . For this systematic review and meta-analysis , we searched nine scientific databases and grey literature from January 1 , 1980 to October 1 , 2018 . Sources with primary data on incident GBS cases in LAC within a well-defined population and timeframe , published in English , Spanish , Portuguese , or French , were included . We calculated the annual GBS incidence rates ( IRs ) and 95% confidence intervals ( CIs ) for each source based on published data . Following an assessment of heterogeneity , we used random-effects meta-analysis to calculate the pooled annual IR of GBS . The study is registered with PROSPERO , number CRD42018086659 . Of the 6568 initial citation hits , 31 were eligible for inclusion . Background annual GBS IRs in Latin America ranged from 0 . 40 in Brazil to 2 . 12/100 , 000 in Chile . The pooled annual IR in the Caribbean was 1 . 64 ( 95% CI 1 . 29–2 . 12 , I2<0 . 01 , p = 0 . 44 ) . During the ZIKV epidemic , GBS IRs ranged from 0 . 62 in Mexico to 9 . 35/100 , 000 in Martinique . GBS increased 2 . 6 ( 95% CI 2 . 3–2 . 9 ) times during ZIKV and 1 . 9 ( 95% CI 1 . 1–3 . 4 ) times during chikungunya outbreaks over background rates . A limitation of this review is that the studies included employed different methodologies to find and ascertain cases of GBS , which could contribute to IR heterogeneity . In addition , it is important to consider that data on GBS are lacking for many countries in the region . Background IRs of GBS appear to peak during arboviral disease outbreaks . The current review contributes to an understanding of the epidemiology of GBS in the LAC region , which can inform healthcare system planning and preparedness , particularly during arboviral epidemics . Registered with PROSPERO: CRD42018086659 .
A rare but severe autoimmune neuropathy , Guillain-Barré syndrome ( GBS ) is the most common type of acute flaccid paralysis ( AFP ) [1] . Often preceded by infections such as Campylobacter jejuni , about 25% of patients require mechanical ventilation [2] . Prognosis varies greatly based on GBS type and urgent care availability [2–5] . Many patients report residual deficits , including pain , limited mobility , and fatigue , years after disease onset [4 , 6–8] . Mortality rates range from 3−7% [3 , 9] , although they can be higher in settings with limited access to intensive care [10 , 11] . In 2008 , the total annual cost of GBS in the U . S . alone was estimated at $1 . 7 billion ( 95% CI $1 . 6 to 1 . 9 billion ) [12] . The median global incidence rate ( IR ) of GBS was estimated at 1 . 10 per 100 , 000 person-years ( range , 0 . 81–1 . 89 ) [13] . However , this estimate was based on data from studies conducted in Europe and North America [13] . Worldwide , there are large variations in the incidence of GBS , ranging from 0 . 38 ( 95% CI 0 . 25–0 . 56 ) to 2 . 53 ( 95% CI 1 . 87–3 . 56 ) per 100 , 000 , with most studies reporting annual IRs between 1 . 1 and 1 . 8 per 100 , 000 [14] . Prior to the 2015–2016 Zika virus ( ZIKV ) epidemic in Latin America and the Caribbean ( LAC ) , there were few published studies on the incidence of GBS in the region , with an exception among children . As part of polio eradication efforts , AFP in children under 15 years of age has been a notifiable event in all LAC countries since the 1980s [15–17] . Using polio eradication surveillance data , in 2010 , Landaverde et al . estimated 0 . 82 cases of GBS per 100 , 000 among children under 15 years of age ( range , 0 . 72–0 . 90 ) [18] . ZIKV is an enveloped positive-strand RNA member of the Flavivirus genus in the Flaviviridae family . Other flaviviruses include dengue , yellow fever , West Nile virus and Japanese encephalitis virus , many of which are associated with neurological disease . Like these viruses , ZIKV is principally transmitted by a mosquito bite and is thus described as an arthropod-borne virus or ‘arbovirus’ . Its primary vector is the Aedes aegypti mosquito , which transmits the virus between humans and is widespread in tropical regions [19] . ZIKV is known to be neurotropic; infection halts proliferation of neural progenitor cells and may induce cell death , leading to ZIKV-related microcephaly [20] . Beyond congenital Zika syndrome , direct viral invasion as well as a parainfective or postinfective autoimmune response may contribute to GBS pathogenesis [21 , 22] . An association between GBS and ZIKV was first established in a case-control study in French Polynesia [23] . During the 2015–2016 ZIKV epidemic , many countries in LAC reported increases in GBS cases , particularly in the beginning of 2016 [24 , 25] . In 2016 , the World Health Organization ( WHO ) concluded that ZIKV infection was a plausible trigger for GBS [26] . The chikungunya ( CHIKV ) [27 , 28] and dengue ( DENV ) viruses [29] , two other arboviruses that are endemic in parts of the LAC region , have also been investigated as possible GBS antecedent infectious agents . Enhanced GBS surveillance [30] and increased research present a unique opportunity to assess the background IR of GBS in the LAC region in the aftermath of recent arboviral epidemics . This review aims to assess the background population-wide incidence of GBS in Latin America and the Caribbean through a synthesis of observational studies . For the purposes of this study , “background incidence” refers to the rates reported during time periods with no arboviral epidemic . To our knowledge this is the first systematic review on GBS in this region . A secondary aim of this review is to ascertain the incidence of GBS during arboviral disease outbreaks .
Our protocol followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses protocols ( PRISMA-P ) guidelines and the Meta-analyses of Observational Studies in Epidemiology ( MOOSE ) Checklist ( see S1 and S2 Files ) [31] . We developed the search word criteria by reviewing PubMed MeSH and Embase Emtree subject headings and keywords . We searched nine electronic databases . Our full search terms are provided in the appendix . We conducted a grey literature search on OpenGrey and OAIster and hand-searched relevant journals and the bulletins of ministries of health cited in bibliographic references of the articles selected for full-text review ( see S3 File ) . We also reviewed the references of all included articles . Inclusion criteria were peer-reviewed and government publications that presented primary data on incident GBS cases in LAC within a well-defined population and timeframe and that were published in English , Spanish , Portuguese , or French between January 1 , 1980 to October 1 , 2018 . The search timeframe was decided upon to update the worldwide review by McGrogan [14] from 1980 to 2008 . Review papers were excluded . After removing duplicates , two authors ( AC , DLV ) independently screened articles by title and abstract based on the inclusion criteria and agreed on those for full-text review . A standard extraction form was developed and tested for reliability . Disagreements were resolved by these two reviewers . One author ( AC ) extracted the following items from the included articles onto the extraction form: study design , country , region , data collection year ( s ) , population size , age , sex , GBS type , GBS diagnostic criteria , incident cases , statistical measures , and circulating arboviral diseases ( ZIKV , DENV , or CHIKV ) . Three other authors ( DLV , DCO , YT ) extracted data from 30% of the records for quality control . All data were standardized to annual mean IR by dividing the number of GBS cases reported by the number of weeks of data capture and multiplying the result by 52 weeks . The result was entered as annual GBS cases into the software program , and together with the base population , was used to calculate the annual IR per 100 , 000 persons . When necessary , we attempted to contact the authors to obtain clarifying information . Three authors ( AC , DCO , YT ) assessed studies for risk of bias using a tool developed for prevalence studies [32] . To reduce publication bias [33] , we searched the grey literature , institutional websites , and conference abstracts . The study protocol was registered with PROSPERO ( CRD42018086659 , available at https://www . crd . york . ac . uk/prospero/display_record . php ? RecordID=86659 ) . We performed the meta-analysis using the metaprop command in Stata version 15 ( College Station , Texas , USA ) . Annualized GBS cases and population sizes were entered for every selected study . Given high heterogeneity , we performed sub-group analyses by: geography ( Latin America versus the Caribbean; Southern Cone , Central and North America , and the Caribbean ) ; time ( before and during an epidemic outbreak ) ; population ( all and under 15 years of age ) ; and case ascertainment ( administrative data using ICD codes only and medical record review ) . Double arcsine transformation and random-effects models were used to calculate pooled IR estimates of GBS [34] .
Among the studies in Latin America that reported background GBS IRs in all ages , the highest were reported in Chile ( 2 . 12 per 100 , 000; 95% CI 1 . 90–2 . 36 ) [44] and Argentina ( 2 . 06 per 100 , 000; 95% CI 0 . 43–6 . 03 ) [39] , and the lowest in Brazil ( 0 . 40 per 100 , 000; 95% CI 0 . 20–0 . 71 ) [42] ( Fig 2 ) . High heterogeneity in Latin America and in other subgroup analyses precluded us from pooling the proportions . The pooled annual IR in the Caribbean was 1 . 64 per 100 , 000 ( 95% CI 1 . 29–2 . 12 , I2 = 0 . 00 , p = 0 . 44 ) ( Fig 3 ) . There were large annual fluctuations in GBS IRs in any given geographic location . In Chile , over a 12-year period , IRs ranged from 1 . 61 per 100 , 000 ( n = 250 cases ) in 2001 to 2 . 35 per 100 , 000 ( n = 402 cases ) in 2010 [44] . Over a 14-year period , Dourado et al . [42] reported a range of 0 . 12 to 0 . 66 per 100 , 000 in Rio Grande do Norte , Brazil . Another multi-year study from Brazil estimated mean annual IRs of 0 . 4 per 100 , 000 ( range , 0 . 3–0 . 6 ) [43] . In the Caribbean island of Aruba , over an 11-year period , Suryapranata et al . reported a range of 1 to 11 cases of GBS , the latter occurring during an outbreak of C . jejuni ( IRs ranged 1 . 0–10 . 37 per 100 , 000 ) . There was less fluctuation in IRs in Argentina , ranging from 0 . 56 per 100 , 000 in 2011 to 0 . 76 per 100 , 000 in 2010 over a 7-year period [40] . However , the IR of GBS among members of a private health maintenance organization in Buenos Aires averaged 1 . 99 per 100 , 000 over an 8-year period [39] . Of the studies reporting background IRs in children under 15 , the lowest rates were in Brazil at 0 . 40 per 100 , 000 [18] , followed by Paraguay at 0 . 72 per 100 , 000 [45] , and the highest rates in El Salvador at 3 . 86 per 100 , 000 [18] , Chile at 1 . 63 per 100 , 000 [18] , and Honduras at 1 . 37 per 100 , 000 [37] . The age-specific distribution of GBS in the pediatric population varied widely across countries . Two multi-country studies found the highest GBS IR in the 1−4 age group: Olivé et al . reported that 47% of the GBS cases were in children 1−4 years of age [35] and Silveira et al . reported an average IR of 0 . 86 per 100 , 000 ( 95% CI 0 . 78–0 . 89 ) in this age group as compared to 0 . 52 per 100 , 000 ( 95% CI 0 . 49–0 . 53 ) among 5−14 year-olds [36] . In Paraguay , the IRs ranged from 1 . 7 per 100 , 000 among 1−4 year-olds to 0 . 1 per 100 , 000 among 10−14 year-olds [45] . In Brazil , a similar trend was reported , with 40% of the cases reported in children under 5 [41] . However , in Chile the IRs were higher among 5−9 year-old children ( 2 . 23 per 100 , 000 ) than among 0−4 year−olds ( 2 . 17 per 100 , 000 ) [44] . GBS distribution by age in the general population did not follow a consistent pattern across countries . In Chile , the IR showed a bimodal distribution with a peak in the youngest ages ( 2 . 23 per 100 , 000 in 5−9 year-olds ) , and increased from 1 . 22 per 100 , 000 among 20−29 year-olds to 4 . 30 per 100 , 000 among 70−79 years-olds [44] . In São Paulo , Brazil , GBS was most common among 15−40 year-olds ( 0 . 15 per 100 , 000 ) , and the IR was lowest in the over 60 ( 0 . 60 per 100 , 000 ) and under 15 ( 0 . 80 per 100 , 000 ) age groups [43] . In Rio Grande do Norte , Brazil , half of the cases were recorded among under 20 years-olds [42] . In Argentina , 37% of the GBS cases were reported in children under 14 years of age [40] . Reporting of higher IRs among children might be due to higher case detection resulting from vaccination safety and polio eradication surveillance efforts . Studies consistently found a higher burden of GBS among males . In studies of children 15 years of age and younger , the highest male-to-female ratio was documented in El Salvador ( 1 . 8:1 ) [35] and the lowest in Brazil ( 1 . 2:1 ) [41] and Honduras 1 . 3:1 [37] . Among all ages , the highest male-to-female ratio was documented in Aruba ( 2 . 3:1 ) [46] , followed by Argentina ( 1 . 6:1 ) [40] , and the lowest in Brazil ( 1 . 3:1 ) [42] . We found that 17 studies reported the GBS IR during arboviral disease outbreaks in seven countries; 14 during the 2015–2016 ZIKV epidemic , one during the 2014 CHIKV outbreak in the French West Indies , and 1 during DENV , CHIKV and ZIKV outbreaks in Martinique ( See Fig 4 ) . The IR in the French West Indies during the 2014 CHIKV epidemic was 3 . 45 per 100 , 000 persons [27] , representing a two-fold increase from 1 . 77 per 100 , 000 during 2011–2013 ( p = 0 . 006 ) [27] . The IR in Martinique increased by 4 . 4 times to 9 . 35 per 100 , 000 during the ZIKV epidemic from a mean annual IR of 2 . 12 per 100 , 000 during 2006–2015 [Incidence rate ratio ( IRR ) ( 2016 vs . 2006–2015 ) = 4 . 52; 95% CI 2 . 80–7 . 64] [62] . In Puerto Rico , the IR increased by 2 . 1 times during the ZIKV epidemic from 1 . 7 to 3 . 5 per 100 , 000 [IRR ( 2016 vs . 2012–2015 ) = 2 . 06; 95% CI 1 . 51–2 . 85] [48 , 63] . In South America , there was substantial heterogeneity in the GBS IRs during the 2015−2016 ZIKV epidemic for the population as a whole , ranging from 7 . 63 per 100 , 000 in Barranquilla , Colombia [58] , to 0 . 71 per 100 , 000 in the Brazilian state of Piauí [54] . In a multi-year study , Barcellos et al . reported a 2 . 7 ( IRR ( 2015–2016 vs . 2008–2015 ) = 2 . 7; 95% CI 2 . 38–3 . 07 ) increase in the rate of GBS hospitalizations in Brazil’s Northeast region during the peak of the ZIKV epidemic as compared to the mean rate in the eight years preceding the epidemic [60] . In Colombia , Machado et al . reported over a two-fold increase in GBS diagnoses during the peak of the ZIKV epidemic as compared to baseline rates ( IRR ( 2016 vs . 2014–2015 ) = 2 . 29; 95% CI 1 . 69–3 . 14 ) [61] . In Veracruz , Mexico , del Carpio Orantes reported a GBS IR of 0 . 62 per 100 , 000 during the ZIKV epidemic in 2016 [49] . We performed sub-group analyses by case-ascertainment ( i . e . , administrative data based on ICD codes only versus both administrative data and medical record review ) to assess if the IR heterogeneity was due to this factor . We found significant within-group heterogeneity but not between-group heterogeneity , indicating that case ascertainment did not significantly bias the IRs . This held true when we performed a region-wide analysis , and when we analyzed Latin American and the Caribbean sub-regions separately ( See S1 Fig ) .
Data on GBS , both historical and current , are lacking for many countries in the region . With the exception of studies using AFP data , most are recent and from ZIKV-affected countries , and baseline data are often lacking . Publication bias in favor of significant results may have limited the availability of studies that found no association between arboviral epidemics and GBS . On the other hand , ZIKV’s high public visibility most likely led to increased scientific research and publications on GBS as well as surveillance ( i . e . , detection ) bias , particularly in countries affected by the emerging arbovirus . Few studies reported data from Central America and Mexico . Therefore , estimates for this sub-region have a high risk of bias . All but two studies reported effect sizes for the outcome of interest . However , we addressed this limitation by calculating 95% confidence intervals for all studies . Methodological differences in case finding limit comparability across studies . The specificity and sensitivity of administrative data varies by setting [48 , 77] . Rates of GBS based on administrative records or passive surveillance systems , without medical record review , may be prone to over-reporting [48] and under-reporting [35 , 78] , respectively . We opted to include surveillance and administrative data because of the paucity of data available . The high cost of capture-recapture methods to compare sensitivity of case identification and ascertainment methods of a rare syndrome limits the feasibility of such studies to assess GBS incidence nationwide . Several researchers have accepted the use of administrative data as a viable low-cost option to assess background GBS IRs [40 , 44] . A limitation of this review is that studies employed different methodologies to ascertain GBS cases . Diagnosing GBS is complex and based on clinical observation and electrophysiological studies . Specialists who can properly diagnose GBS may be unavailable in resource-limited settings . Natural annual variations in incidence may be masked or exacerbated by imperfect case identification . A strength of this research is that all included studies are population based . In addition , many of the studies are multi-year in duration , with a mean of 7 . 2 years for studies that analyzed background GBS IRs . This gives us a robust estimate of GBS incidence that balances out natural fluctuations in annual IRs . Another strength is the application of a thorough search methodology and inclusion of most languages spoken in LAC . Inclusion of ministry of health bulletins data served to balance publication bias . Some of the included studies were not carried out with the sole purpose of measuring GBS IRs . While the studies that focused on IR calculations tended to adjust the rates based on population age structures , others did not . This introduces a source of variability . Some studies reported incidence by person-years and others as annual incidence . We addressed this issue by calculating all IRs . Since GBS is triggered by a variety of antecedent infections , baseline incidence of GBS is critical for detecting and monitoring infectious disease outbreaks . The LAC region has been a pioneer in monitoring of GBS in children over the last 30 years [16] . Countries such as Colombia and Brazil have monitored GBS as part of DENV eradication programs [57 , 79] . The ZIKV epidemic and the reported increases in GBS in the Americas have made GBS a notifiable condition in many countries . As the ZIKV epidemic has spread beyond the Americas [80–82] , it is important that those countries are particularly prepared for GBS surveillance and management . Enhanced surveillance and increased research have provided us with new data to assess GBS incidence in this region . Because of its severity and lethality in the absence of adequate care , investments are needed to provide information on GBS to populations at-risk and to build healthcare providers’ capacity to diagnose GBS and follow appropriate care protocols [83 , 84] . GBS poses an additional burden to health care systems , particularly in resource-limited settings . | A severe neurological disorder , Guillain-Barré syndrome ( GBS ) is the leading cause of acute flaccid paralysis . This is the first systematic review on GBS incidence in Latin America and the Caribbean before and during arboviral disease outbreaks . There is a large sub-regional and annual fluctuation in the incidence of GBS . Background annual GBS incidence rates ( IRs ) in Latin America ranged from 0 . 40 in Brazil to 2 . 12/100 , 000 in Chile . The pooled annual IR in the Caribbean was 1 . 64 ( 95% CI 1 . 29–2 . 12 , I2<0 . 01 , p = 0 . 44 ) . During the ZIKV epidemic , GBS IRs ranged from 0 . 62 in Mexico to 9 . 35/100 , 000 in Martinique . GBS increased 2 . 6 times during ZIKV and 1 . 9 times during chikungunya outbreaks over background rates . GBS is a costly disease , which can result in long-term disability and high mortality rates in resource constrained healthcare settings . Because GBS can be triggered by arboviral infections , baseline incidence of GBS is critical for detecting neglected tropical disease outbreaks . The current review contributes to an understanding of the epidemiology of GBS in the LAC region , which can inform healthcare system planning and preparedness . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"chikungunya",
"infection",
"pathogens",
"immunology",
"geographical",
"locations",
"microbiology",
"tropical",
"diseases",
"viruses",
"north",
"america",
"clinical",
"medicine",
"rna",
"viruses",
"neglected",
"tropical",
"diseases",
"caribbean",
"infectious",
"diseases",
"south",
"america",
"medical",
"microbiology",
"microbial",
"pathogens",
"arboviral",
"infections",
"guillain-barre",
"syndrome",
"brazil",
"people",
"and",
"places",
"diagnostic",
"medicine",
"colombia",
"clinical",
"immunology",
"flaviviruses",
"viral",
"pathogens",
"autoimmune",
"diseases",
"biology",
"and",
"life",
"sciences",
"viral",
"diseases",
"organisms",
"zika",
"virus"
] | 2019 | Incidence of Guillain-Barré Syndrome (GBS) in Latin America and the Caribbean before and during the 2015–2016 Zika virus epidemic: A systematic review and meta-analysis |
The recurrent fixation of newly arising , beneficial mutations in a species reduces levels of linked neutral variability . Models positing frequent weakly beneficial substitutions or , alternatively , rare , strongly selected substitutions predict similar average effects on linked neutral variability , if the product of the rate and strength of selection is held constant . We propose an approximate Bayesian ( ABC ) polymorphism-based estimator that can be used to distinguish between these models , and apply it to multi-locus data from Drosophila melanogaster . We investigate the extent to which inference about the strength of selection is sensitive to assumptions about the underlying distributions of the rates of substitution and recombination , the strength of selection , heterogeneity in mutation rate , as well as the population's demographic history . We show that assuming fixed values of selection parameters in estimation leads to overestimates of the strength of selection and underestimates of the rate . We estimate parameters for an African population of D . melanogaster ( ŝ∼2E−03 , ) and compare these to previous estimates . Finally , we show that surveying larger genomic regions is expected to lend much more discriminatory power to the approach . It will thus be of great interest to apply this method to emerging whole-genome polymorphism data sets in many taxa .
The fixation of beneficial mutations can strongly reduce levels of closely linked neutral variation – the so-called genetic hitchhiking effect [1] . This prediction has been used to search for positive selection by looking for regions of the genome with reduced variability [e . g . , 2] . The hitchhiking model most often used is of a single selective sweep , where the location and timing of selection are assumed to be known [3] . This single sweep model has been of great value in understanding the effect that a single selective event has on patterns of polymorphism , as a function of the strength of selection and location of the beneficial mutation [e . g . , 1] , [4] , [5] . However , this model is somewhat disconnected from the problem of detecting selective sweeps in the genome , for which locations and timings are not known a priori , and should be treated as random variables . Kaplan et al . ( 1989 ) described a “recurrent hitch-hiking” ( RHH ) model , where the expected number of sweeps ( per base pair , per 2N generations ) is 2Nλ with sweeps occurring at random locations in the genome [6] . The RHH model is most commonly considered for the case of genic selection on new mutations entering the population [e . g . , 6]–[8] . Under this model , several patterns expected under the single sweep model no longer apply . For example , the single sweep model predicts coalescent histories with long internal branches , as some lineages may escape the recent coalescent event via recombination . This results in the widely employed prediction of an excess of high-frequency derived alleles flanking the fixed site [5] . Under RHH models however , the probability of such a history is small , as sweeps are on average old and high frequency derived mutations have thus likely drifted to fixation [9] . Wiehe and Stephan ( 1993 ) showed that under a RHH model , for a given recombination rate , the expected level of heterozygosity at linked sites relative to neutral expectations is dependent upon the compound parameter ( s ) ( 2Nλ ) , where 2Nλ is the rate of fixation of beneficial mutations and s is the average strength of selection [7] . This result implies that that the two parameters are confounded ( much like the effective population size , Ne , and mutation rate , μ , in θ = 4Neμ ) as their effect on expected levels of diversity depends on their product . In D . melanogaster and D . simulans , lower than expected levels of nucleotide diversity are observed in regions of reduced recombination [10] and in the coding sequences of rapidly evolving proteins [11] , [12] . These findings are compatible with either strong but infrequent positive selection ( i . e . , large s and small 2Nλ ) or weak but common positive selection ( i . e . , small s and large 2Nλ ) [7] , [11]–[13] . A number of methods have been proposed for quantifying s and 2Nλ ( separately ) using divergence and polymorphism data [e . g . , 11]–[12] , [14]–[17] . These approaches typically make strong assumptions regarding the possible distribution of selection coefficients , the number of adaptive substitutions between species , or the timing of selection . For example , Li and Stephan ( 2006 ) examined 250 non-coding regions from an East African population of D . melanogaster [18] . Using a likelihood approach , they estimate that approximately 160 beneficial mutations have fixed in this population over the last ∼60 , 000 years ( corresponding to ) , with mean selection coefficient ŝ∼0 . 002 . This inference is achieved by effectively assuming that the timing of all sweeps is known ( and the time since the sweep , τ = 0 ) . Under a recurrent sweep model , this assumption may bias the estimation of s and 2Nλ . Additionally , as this method relies on first fitting a demographic model to non-coding DNA polymorphisms , it is possible that the effects of purifying selection on the site frequency spectrum of non-coding DNA [19]–[20] may strongly affect the estimates . Using synonymous polymorphism data in D . melanogaster , and divergence to D . simulans , at 137 X-linked loci , Andolfatto ( 2007 ) employed a maximum likelihood approach to estimate the joint parameter 2Nλs , followed by a McDonald-Kreitman-based method to separately estimate 2Nλ and s [11] . Based on these calculations , Andolfatto estimated that most beneficial amino acid substitutions are very weakly advantageous on average ( with average ŝ∼1 . 2E−5 and ) . Macpherson et al . ( 2007 ) , using polymorphism data from D . simulans ( and divergence to D . melanogaster ) , propose a method to infer the rate and strength of selection from the spatial scale of variation in polymorphism and divergence [12] . In contrast to Andolfatto's estimates , Macpherson et al . estimate a much stronger average selection coefficient ( ŝ∼0 . 01 ) and less frequent selection ( ) . However , they note that their method is more likely to detect strong selection , so the effects of many weakly beneficial mutations may be missed . By evaluating a wide array of recurrent selection models across a variety of sampling schemes , with parameters relevant for both Drosophila and human populations , we demonstrate here that there are differences in the predictions of weak and strong selection models , both in the spatial distribution of variability levels and the distribution of polymorphism frequencies ( also called the site frequency spectrum , hereafter SFS ) . We propose a polymorphism-based approximate Bayesian ( ABC ) estimator that is most closely allied to the approach of Macpherson et al . ( 2007 ) , but is also applicable to sub-genomic multi-locus data of the kind that has most often been collected [e . g . , 11] , [21]–[22] , and incorporates more information from the data . Fundamentally , this estimation procedure is based on the principle that while models may predict the same average affects , the variance of many common summary statistics varies greatly between models . We show that highly accurate estimation will be possible with large-scale genome polymorphism data , and that the approach is robust to both mutation and recombination rate heterogeneity .
As pointed out by Macpherson et al . ( 2007 ) , there is reason to anticipate that region size may be key in uncoupling the strength of selection ( s ) from the rate of beneficial fixation ( 2Nλ ) ( see Table 1 for a summary of terms ) . Intuitively , because only a very strong sweep is capable of severely reducing larger regions - on the order of 100 kb for instance - regions may be observed with very little variation under this model . However , because selection is rare , other regions will appear close to neutral . Conversely , weak selection serves to homogenize variation as it occurs with much greater frequency . For example , for an effective population size of 106 and ρ = 4Nr = 0 . 1/bp , the expected waiting time between sweeps is 68 , 000 generations , for s = 1E−04 and 2Nλ = 5E−04 , for a region size of 104 base pairs . For the same population parameters , but s = 0 . 01 and 2Nλ = 5E−06 , the expected waiting time between sweeps is 532 , 000 generations . Considering that most signatures of selection are dissipated by 400 , 000 generations for these parameters [9] , [23] , this demonstrates that if selection is strong and rare on average , there will likely be a large variance across the genome , from strongly swept to essentially neutral looking regions ( Figure 1 ) . Capturing this variance is dependent upon the size of the sampled region as , while many values of s may reduce a 500 bp region for instance , only large selection coefficients are capable of reducing a 100 kb region , suggesting that larger region sizes should afford greater discriminatory power . In order to more precisely determine this ‘region size’ effect , we examined 500 bp , 1 kb , 2 kb , 5 kb , 10 kb , 25 kb , 50 kb , and 100 kb regions using simulated data ( Figure 2A ) . First examining L = 500 bp regions ( matching existing empirical datasets , e . g . , [11] , [21] ) , we observe that there is relatively little difference in the coefficient of variation ( CV ) of π between RHH models of strong and weak selection ( Figure 2 ) , consistent with previous observations that s and 2Nλ are difficult to estimate separately with data of this kind [13] . Examining larger regions , the CV is essentially unchanged under weak selection models once regions larger than 25 kb have been sequenced . Conversely , the CV continues to grow rapidly under a strong selection model , producing a four-fold difference in the CV at 50 kb of sequence relative to weak selection models , and over a five-fold difference at 100 kb for these parameters , for Drosophila-like parameters ( θ = 0 . 01/site; ρ/θ = 10 ) . The difference between strong and weak selection models in Figure 2 does not appear to be attributable to the total amount of surveyed sequence between the 100 kb and 500 bp regions . By comparing the distribution observed when considering ten 100 kb regions vs . two thousand 500 bp regions ( and thus the same number of segregating sites on average ) we still observe a large difference in CV at the scale of 100 kb , and little difference between models at the scale of 500 bp ( results not shown ) . We found that the relative point at which the region size benefit plateaus is a function of θ , ρ/θ , 2Nλ and s . We examined the effect of doubling the recombination rate ( such that ρ/θ = 20 ) , and find that the CV is reduced under all models relative to ρ/θ = 10 , and that the models begin to differentiate at smaller region sizes ( Figure 2B ) . These effects are a result of the fact that the expected size of the swept region will decrease as the recombination rate increases [6] . Additionally , using human-like parameters ( θ = 0 . 002/site , ρ/θ = 1 ) , we find that the pattern of an increasing CV with region size is still observed to some extent . However , the CV is much larger on average even under neutrality when ρ/θ = 1 , and the models are more similar to one another with human-like parameters ( Figure 2C ) than with Drosophila-like parameters ( Figures 2A and B ) . This implies that weak and strong selection models will be more difficult to distinguish in humans . It is noteworthy that for large surveyed regions , more strongly negative values of Fay and Wu's H-statistic ( i . e . , SFS skewed towards high-frequency derived alleles ) and Tajima's D-statistic ( i . e . , SFS skewed towards rare alleles ) are observed under strong selection models ( Figure 3 ) , suggesting that differences in the polymorphism site frequency spectrum may also be used to distinguish between models if large enough regions are surveyed . Though this differs qualitatively from the conclusions of Przeworski ( 2002 ) , simulations demonstrate that this is attributable to a modeling difference ( results not shown ) , as we here allow sweeps within the sampled region ( following [24] ) . This discrepancy between modeling approaches will thus only become greater as region sizes increase . The above results suggest that focusing on variability across loci may distinguish models of strong , rare sweeps from those of frequent , weak sweeps . Thus , we here implement an approximate Bayesian ( ABC ) approach to estimate the strength of selection ( s ) , the rate of fixation of beneficial mutations ( 2Nλ ) and the neutral population mutation rate ( θ = 4Neu ) under a recurrent hitchhiking model . We begin by employing the observed mean and standard deviation of heterozygosity ( π ) , which is closely related to previously published estimation procedures [e . g . , 11]–[12] . In order to evaluate this approach , we tested the performance using simulated data . Figure S1 shows distributions of maximum a posteriori ( MAP ) estimates of s , 2Nλ , and θ under two different models ( strong rare and weak frequent selection ) , for 50 kb and 500 bp regions . In these simulations , s , 2Nλ and ρ have fixed values indicated with the vertical dotted line . We find that this π-based estimation performs reasonably well , particularly when the size of surveyed regions is large and selection is strong . For 500 bp regions , MAP estimates are accurate within an order of magnitude . However , distributions of MAP estimates are typically widely dispersed , particularly when selection is weak ( Figure S1; Table S1 ) . Additionally , estimation of s , 2Nλ , and θ is generally upwardly biased . Under the best conditions - large region sizes and strong selection - the performance of the estimator is greatly improved ( RMSE ( ŝ ) = 0 . 179 , and the relative bias , RB ( ŝ ) = −0 . 281 ) . Given the computational efficiency of ABC , it is straightforward to explore multiple combinations of test statistics , in order to determine whether incorporating additional information from the site frequency spectrum or spatial distribution of sites may significantly improve the accuracy of estimation . We found that the incorporation of the mean and variance of several common summary statistics did not significantly improve or alter estimation , owing to correlations with π ( results not shown ) . However , other statistics such as θH [25] , and ZnS [26] are only weakly correlated with π ( results not shown ) . As such , it may be anticipated that the addition of these statistics may provide additional information , which would allow for further discrimination between models . This intuition appears to be accurate . The addition of the mean and SD of ZnS and θH particularly , and the number of segregating sites ( S ) to a lesser extent , appear to improve the performance of the method considerably . For strong selection , even at the 500 bp scale , the addition of multiple summary statistics reduces the bias and RMSE by half relative to π-based estimation ( Table S1 ) , thereby improving the accuracy of estimation ( Figure 4 ) . This result suggests a distinct advantage to utilizing these additional summary statistics , particularly when surveying larger regions . Though the parameters s , 2Nλ , and ρ are fixed in the above simulations , these parameters likely vary among genomic regions in real data . While it is attractive to assume a fixed parameter model given its simplicity , if the true model is in fact one in which parameters are drawn from distributions , this may lead to a bias in estimation owing to misspecification of the model . We consider a variety of examples – those in which s and 2Nλ are drawn from exponentials , and ρ is drawn from an exponential or normal . When comparing between fixed and distributed models – the mean of the distribution is equal to the fixed value used previously ( i . e . , if in the fixed model s = 0 . 01 , the distribution model to which it would be compared would have s exponentially distributed with mean 0 . 01 ) . Figure S2 documents the effect of modeling parameters drawn from distributions on the relative CV of π ( compare to Figure 2 ) . As expected the relative CV is inflated compared to the fixed parameter model , which may lead to biases in estimation if unaccounted for . In order to consider the effect of model misspecification on parameter estimation , datasets are simulated under a model where parameters were drawn from distributions , yet priors are constructed assuming that these parameters have fixed values . Misspecification of the model in this way leads to an upward bias in the estimate of selection coefficients , and a downward bias in the estimated rate of selection ( Figure 5 ) . To account for this misspecification , the priors must be appropriately constructed , by allowing each locus within a given replicate dataset to also be drawn from distributions ( see Methods ) . As shown in Figure 5 , while the distribution of MAP estimates are more greatly dispersed when compared with Figure 4 ( e . g . , under a fixed model the RMSE ( ŝ ) = 7 . 9E−06 for strong selection and large regions , and under a distributed model the RMSE ( ŝ ) = 1 . 11 ) , the mean of the distribution nonetheless accurately reflects the means of s , 2Nλ , and θ ( for the above two models , the RB ( ŝ ) are 0 . 12 and 0 . 57 , respectively; Table S1 ) . Additionally , for all estimated parameters , the relative bias is reduced for 50 kb relative to 500 bp regions . For comparison , an alternate distributed parameter model was considered . As opposed to s being drawn from an exponential distribution for each locus , we model s being drawn from an exponential distribution for each selective event . Results between the two models are similar , though this case results in consistently smaller RMSEs ( results under this alternative model , mirroring Figure 5 , are given in Table S1 ) . This result suggests that this alternative distribution model is intermediate between the two extreme cases examined here - fixed models and distributed locus-by-locus models . Despite the overall improvement gained by modeling distributed parameters in general , an important limitation is the assumption that the shape of the underlying distribution of each parameter is known . The above simulations however , continue to assume a constant mutation rate among regions . In reality , the mutation rate may vary among loci , which may be a potential source of bias for the method [11]–[12] . Thus , in order to consider the possible effects of mutation rate variation , the distribution of variation at synonymous sites among loci in the Andolfatto ( 2007 ) dataset ( see below ) was taken as a proxy for mutation rate variation . We estimated the parameters for a Γ-distribution using the distribution of synonymous site divergence estimates across loci . Modeling this observed distribution with simulated data ( i . e . , Γ ( 200 , 2 . 5 ) ; Figure S3 ) , we found that the estimation was not affected and results resemble those of a fixed θ model ( Figure S1 , Figures 4–5 ) . This result suggests that the variation in mutation rate observed in D . melanogaster is not widely dispersed enough to impact estimation , and is thus not likely to be biasing our estimated parameter values . As there is relatively little variance at synonymous sites observed among regions in the Andolfatto ( 2007 ) dataset , data was simulated in which θ is much more widely dispersed ( i . e . , Γ ( 10 , 50 ) ) , in order to determine the possible bias introduced by more extreme mutation rate variation . Importantly , under this model , estimation based upon and SD ( π ) becomes strongly biased in the direction of estimating larger selection coefficients , as heterogeneity in mutation rate is artificially inflating the variance among loci ( Figure S3 ) . However , when estimation is based upon the means and SDs of π , S , θH , and ZnS , results appear robust to mutation rate variation ( for π-based estimation , the RB of ŝ = 8 . 95 , for all statistics the RB = 0 . 51; Table S1 ) . This is owing to the fact that while π is greatly impacted by this heterogeneity , other statistics , such as ZnS , have standard deviations that vary greatly between RHH models , yet are largely unaffected by mutation rate variation within any given model . Importantly , we only here consider regional variation in mutation rates and not site-to-site variation within genes ( e . g . , CpG in mammals ) . In summary , we propose that our estimator of recurrent hitchhiking model parameters that incorporates information from multiple summary statistics performs reasonably well . This method is preferable to a π-based approach both because it is more accurate and more robust to variation in mutation rate . The overall performance of the method will be greatly improved by the availability of genome-scale polymorphism data . An important point relevant to all of these models is that relatively simple adaptive models have been considered , and additional complexities such as recently increased or decreased rates of adaptation , variation in dominance of beneficial mutations , or selection from standing variation , have yet to be incorporated . Here we apply our approach to the multi-locus data set of Andolfatto ( 2007 ) , who surveyed 137 X-linked regions from an East African population of D . melanogaster [11] . Though our performance evaluation of the method suggests that regions of this size are not ideal for estimation ( the average region length in this dataset is 680 bps ) , they indicate at least the possibility of distinguishing weak from strong selection models , though such small regions cannot assure accurate parameter estimation . We estimated selection parameters both from 1 ) priors where these parameters are drawn from distributions ( exp ( s ) , exp ( 2Nλ ) and N ( ρ , ρ/2 ) , and 2 ) in order to compare to previous estimation methods , priors that assume fixed values of s , 2Nλ and ρ . The strength of selection for each sweep , s , is drawn from an exponential distribution ( see Methods ) . We ignore variation in θ among loci as we have shown that this is not expected to significantly impact estimation ( see above ) . Shown in Figure 6 are marginal posterior distributions for selection parameters ( assuming distributed parameters , ŝ = 2E−03 , , and per site ) . Consistent with simulated data , parameter estimations assuming fixed values leads to considerably larger estimates of ŝ , and reduced estimates of ( Figure 6 , ŝ = 0 . 01 , , and per site ) . It is thus important to emphasize that estimation will be sensitive to the underlying models chosen for the priors . Given that we expect these parameters to vary among loci , we consider the former estimate to perhaps be better , with the caveat that we lack precise knowledge of how these parameters are actually distributed ( see Methods for more details ) . Interestingly , the large estimate of compared to previous studies [11]–[12] suggests a stronger mean reduction in genome variation due to hitchhiking ( ∼50% ) . Finally , it is additionally noteworthy that estimation does not necessarily need to be performed using the marginal posteriors as we have implemented here . For example , Figure S4 compares estimation between joint and marginal posteriors for our empirical dataset , and finds that while the estimates are similar , they are not identical . Understanding these differences , and better determining if estimation based upon joint posteriors may have any advantages , is a topic of future investigation . An important consideration we have not addressed thus far is the impact of non-equilibrium demography , which may closely resemble sweep-like patterns of variation and may be expected to bias the estimator [e . g . , 27]–[28] . For instance , a strong population bottleneck exhibits many characteristics of a selection model – greatly increasing the variance of summary statistics , and specifically producing very negative values of the H-statistic [29]–[32] . In order to assess the potential bias induced by demography on the proposed estimator , we model two simple bottleneck models ( BN1 and BN2 ) and a growth model ( see Methods ) . BN1 and the growth model were fit to match the observed mean π and Tajima's D . BN2 was chosen specifically to match the observed CV ( π ) . Under all three models , the posterior distributions are localized around weaker selection coefficients , and larger rates , than we estimate from the observed data , with estimation based upon distributed priors ( MAP estimates given in Table 2 ) . This result suggests both that , while the estimator is obviously sensitive to non-equilibrium demography , our empirical data is not easily explained by any of the demographic models considered ( with the empirical estimates falling outside of the 95% CIs for the demographic models considered ) . This is particularly encouraging given that one of the bottleneck models , BN2 , was chosen specifically to match the CV ( π ) that was observed for this dataset . Clearly , to minimize demographic effects , populations should be carefully chosen when possible . The dataset we have analyzed is from a putatively ancestral East African population that is believed to have been relatively demographically stable compared to non-African populations , which show signatures of a recent and severe bottleneck [18] , [31]–[32] . Characterizing biases induced from a wider range of demographic models is a topic of future study , and will be important before performing estimation in other populations and species . One promising direction will likely take advantage of the observed correlation between πs and Ka [11]–[12] , which is difficult to explain under neutral demographic models . The incorporation of divergence data of this sort may increase the robustness of the estimator to non-equilibrium perturbations [12] . Several other studies have attempted to estimate parameters under a recurrent hitchhiking model , and a discussion of how our estimates compare with those studies is of considerable interest . As previous studies assumed fixed values of s , 2Nλ and ρ , it is most appropriate to first compare these estimates with our “fixed value” estimation . Li and Stephan ( 2006 ) employed a sliding window likelihood ratio test using multi-locus polymorphism data and estimate that ŝ∼0 . 002 and [18] , which is similar to our estimates ( Table 3 ) . Their approach has a number of notable differences with ours: they co-estimate a growth model within their estimation procedure , use non-coding DNA rather than synonymous sites , and assume that all detectable sweeps have fixed immediately prior to sampling ( i . e . , τ = 0 ) . Given that our values of 2Nλs are quite similar , so is the expected level of reduction in genome variability ( Table 3 ) . Macpherson et al . ( 2007 ) used large-scale polymorphism data from six lines of D . simulans and estimate a strong average selection coefficient ( ŝ∼0 . 01 ) [12] , which is identical to our fixed value estimate . The bigger difference is in our estimates of 2Nλ , with our estimate being ∼4× larger . However , given that the dataset examined here is from D . melanogaster , there is no reason to necessarily anticipate that these estimates should match . It is noteworthy that our estimated selection coefficient is an order of magnitude smaller ( and our estimate of the rate an order of magnitude larger ) when we assume that s , 2Nλ and ρ are drawn from distributions rather than taking fixed values . Despite this , our estimated selection coefficient under the distributed model is still almost two orders of magnitude larger than Andolfatto's ( 2007 ) estimate [11] . Andolfatto's estimates of s and 2Nλ are particularly relevant , as we here examine the same dataset and arrive at quite different conclusions . The discord between estimates may arise from the fact that Andolfatto's estimate of s relies on estimating 2Nλ using the McDonald-Kreitman statistical framework [33]–[34] . However , we note that with short surveyed fragments , our estimator of s is somewhat upwardly biased ( Figure 5 ) so it will be interesting to apply our method to larger genomic regions when that data becomes available . Additionally , while Andolfatto ( 2007 ) and Macpherson et al . ( 2007 ) estimate a 20% average reduction in genome-wide variability , we estimate a considerably larger reduction ( 50% ) , which is more consistent with the estimate of 2Nλs of Li and Stephan ( 2006 ) . This may to some extent explain Andolfatto's observation that the observed Tajima's D at synonymous sites is more negative than predicted by his estimates of s and 2Nλ . When we model a recurrent hitchhiking model with our estimated parameters , the average Tajima's D is −0 . 3 , which is close to the observed average ( −0 . 28 ) . While a negative mean Tajima's D is usually interpreted in the context of demographic models ( such as population growth , see for example [18] ) , it may instead imply that recurrent hitchhiking may be having a larger genome wide impact than previously appreciated . While common/weak and rare/strong recurrent positive selection result in similar average levels of genome variation on average ( for 2Nλs = constant ) , rare/strong selection greatly increases the variance of common summary statistics relative to common weak selection . We demonstrate , using an ABC approach based upon this observation , that the rate and the strength of selection may accurately be estimated jointly . Though there is some power to differentiate parameters using existing data , our results strongly suggest that genome scale data will afford much better discriminatory power . Our study also highlights that learning more about how parameters such as s , 2Nλ and ρ are distributed among loci will be crucial for accurate parameter estimation .
We use the recurrent selective sweep coalescent simulation machinery described in [24] , with a modification to account for the stochastic trajectories of positively selected mutations in finite populations [11] , [35]–[36] . Briefly , sweeps are occurring in the genome at a rate determined by 2Nλ = Λ , where λ is the rate of sweeps per generation [6] , [8] . Following [24] , selective sweeps are allowed both within the sampled region , as well as at linked sites . This distinction is significant , because for large simulated regions the probability of a sweep within the region may not be negligible for large Λ . The rate of sweeps within a region is thus MΛ , and as each sweep may affect up to s/rbp ( from [6] , [37]; which is equivalent to 4Ns/ρbp ) , the rate considering both the sequenced and flanking regions becomes , where ρbp is the scaled recombination rate between base pairs and M is the size of the region in base pairs ( see [6] , [37] for details ) . With this , the expected waiting time between sweeps is in 2N generations . For the purposes of testing the proposed estimator , we evaluated models for Ne = 106 , θ = 4Nμ = 0 . 01/site , and ρ = 4Nr = 0 . 2/site ( r = 5E−08 per site per generation ) and 0 . 1/site ( r = 2 . 5E−08 per site per generation ) in order to replicate Drosophila-like parameters ( [32]; corresponding to values of ρ/θ = 20 and 10 , respectively ) . The product sλ was set at 2 . 5E−13 in the case of ρ/θ = 10 , and to 5E−13 for ρ/θ = 20 . To replicate human-like parameters , we consider Ne = 104 , θ = 0 . 002/site , and ρ = 0 . 002/site ( r = 5E−08 per site per generation; corresponding to ρ/θ = 1 ) and sλ was set at 5E−11 . In all cases , the sample size ( n ) = 25 , and neutral variation is reduced to 60% of the neutral expectation . These calculations may be made from Eq . ( 5 ) of [7] , which predicts the expected heterozygosity at linked neutral sites , ( 1 ) where θ is the neutral population mutation rate , r is the unscaled recombination rate in Morgans per base pair per generation , κ is a constant ∼0 . 075 , γ = 2Nes ( where s is the selection coefficient ) , and λ is the rate of adaptive substitutions per site per generation . In most cases , simulated datasets consist of 10 50 kb regions or 1000 500 bp regions ( which correspond to the same number of surveyed sites ) . 10 , 000 replicate datasets were generated under each model . When simulating distributed rather than fixed values of s , 2Nλ , θ , and ρ , values for each region are drawn from a distribution ( exp ( s ) , exp ( 2Nλ ) , N ( ρ , ρ/2 ) or exp ( ρ ) . Thus , the value is fixed for an individual locus , but varies among loci . An alternative model was additionally examined , in which s is not fixed per locus , but rather is drawn from an exponential distribution for each selective event . These two separate models were chosen for two distinct purposes: 1 ) an exp ( s ) per locus is chosen for the performance simulations as it results in a large variance between loci . Thus , alongside the fixed parameter model , these comparisons represent two extremes; 2 ) an exp ( s ) per sweep is chosen when analyzing the empirical and demographic data , as we believe it better approximates biological reality ( representing a model first introduced by Fisher ) . While the true underlying distributions are unknown , there is some biological data to draw from . For instance , observed Ka among genes [11] is nearly exponentially distributed , implying that an exp ( 2Nλ ) is a reasonable approximation . We model a normally distributed recombination rate for Drosophila-like parameters since heterogeneity in recombination rates is not believed to be large [38] . Additionally , recombination rate variation is minimized in the Andolfatto ( 2007 ) dataset analyzed here , as high recombination regions of the X were surveyed . For human-like parameters , we model an exponential recombination rate because recombination rate heterogeneity is more extreme [39] . When comparing between fixed and distributed models , a fixed value of s = 0 . 01 for example , is compared with a distributed model in which 0 . 01 is the mean of the exponential distribution from which the loci are drawn . In order to assess any bias which may be associated with variable mutation rates between regions , models were tested in which θ/locus is drawn from a Γ-distribution . Two Γ-distributions are examined , one matching the observed CV of synonymous site divergences among loci in the Andolfatto ( 2007 ) dataset analyzed here ( Γ ( 200 , 2 . 5 ) ) , and one in which θ is very widely dispersed ( Γ ( 10 , 50 ) ) . In order to consider the performance of our method under non-equilibrium demographic models , we fit a simple bottleneck and growth model to the empirical data based on observed values of and the average Tajima's D ( 0 . 025/site and −0 . 28 , respectively ) . Under both models , simulation parameters are thus scaled to mimic the observed values of these two statistics . As with above , n = 12 , ρ = 0 . 1 , θ = 0 . 01 and Ne = 106 . Course grids under both models were simulated using the program ms [40] . We estimate a growth model in which growth rates were set to α = 50 at time t = 0 . 5 4N generations in the past , where N ( t ) = N0exp−αt . We estimate a bottleneck model that posits a stepwise reduction to 0 . 0001 of the population's former size beginning at tb = 0 . 5 and lasting 0 . 01 4N generations ( BN1 ) . In addition , a bottleneck model was selected to fit another feature of the data , the observed CV ( π ) ( population reduced to 5 . 1% of its former size at time tb = 0 . 19 and lasting 0 . 01 4N generations; BN2 ) . Estimation is performed using priors generated under a model in which parameters are distributed between loci ( and s is distributed per sweep ) , as we argue that to be a more biologically relevant scenario compared to fixed parameter models . To estimate the parameters s , 2Nλ , and θ , we relied upon their relationship with the means and standard deviations of common summary statistics . We take an approximate Bayesian ( ABC ) approach [41]–[44] to obtain marginal posterior distributions ( estimation is also possible using joint posterior distributions , an example of this is discussed in the Results and given as a Supplement ) . Calculating our summary statistics ( the means and SDs of π , S , θH and ZnS ) from the observed data , and from simulated data with parameters drawn from uniform priors , we implement the regression approach of [42] . Briefly , this involves fitting a local-linear regression of simulated parameter values to simulated summary statistics , and substituting the observed statistics into a regression equation . The prior distributions used were s∼Uniform ( 1 . 0E−06 , 1 . 0 ) , 2Nλ∼Uniform ( 1 . 0E−07 , 1 . 0E−01 ) , and θ∼Uniform ( 0 . 0001 , 0 . 1 ) , and the tolerance , δ = 0 . 001 . Under a fixed selection parameter model , each draw from the prior represents the parameter value that is in common among all loci in a given dataset ( i . e . , 1000 500 bp regions , or 10 50 kb regions ) . Under a distributed parameter model , each draw from the prior represents the mean of the distribution from which each locus in a given dataset will be drawn ( or in the case of the alternative for modeling selection coefficients , a value of s is drawn for each sweep – see ‘simulation of the recurrent hitchhiking model’ ) . In order to determine the optimal combination of information , estimation was performed using all combinations of the mean and standard deviations of π , the number of segregating sites ( S ) , θH , Tajima's D , Fay and Wu's H , and ZnS . The combination of π , S , θH , and ZnS was found to result in highly accurate and unbiased MAP estimates . Two statistics were utilized to evaluate the MAPs of ŝ , and . First , in order to measure any biases , the relative bias ( RB ) was determined from 1000 MAP estimates , as RB = Mean ( Xˆ−X ) /X . Second , in order to measure deviations from the expected values , the relative mean square error ( RMSE ) was determined as RMSE = Mean ( Xˆ−X ) 2/X2 . The necessary code , and instructions for performing estimation , can be found at: http://www . molpopgen . org/ . We use the137 X-linked coding loci surveyed in [11]; Genbank accession numbers EU216760-EU218523 . All loci were surveyed in 12 lines of D . melanogaster from a Zimbabwe population . For this analysis , only synonymous sites were considered . We summarized the mean average pairwise diversity , , its standard deviation , SD ( π ) , and the coefficient of variation , , as well as the means and SDs of the number of segregating sites , S , θH [25] , Tajima's D [45] , Fay and Wu's H [5] , and ZnS [26] , for synonymous sites across loci . Levels of synonymous polymorphism positively correlate with rates of divergence at synonymous sites [11] . To account for this , we also used partial regression corrected values of π at synonymous sites that account for variation in Ks [11] . We found that this had very little effect on and SD ( π ) in this particular case . | Understanding the process of adaptive evolution requires quantifying the extent to which beneficial mutations contribute to differences between species . However , fundamental parameters of adaptation , such as the rate and strength of beneficial mutations , are poorly understood and have historically been difficult to estimate from data . In particular , distinguishing a high rate of weakly selected substitutions from a low rate of strongly selected substitutions has been problematic . Here , we introduce a new method to estimate the parameters of adaptive evolution from multi-locus population genetic data . We conduct simulations to show that this method is able to discriminate the rare/strong model from the frequent/weak model . Applying this method to an African population sample of Drosophila melanogaster , we estimate selection parameters and find that recurrent adaptive evolution has reduced genome variability by ∼50% on average . The availability of genome-scale population genetic data will lend considerable discriminatory power to the approach . Thus , this new approach represents an important step towards characterizing the nature of adaptive evolution in natural populations . | [
"Abstract",
"Introduction",
"Results/Discussion",
"Methods"
] | [
"evolutionary",
"biology/genomics",
"evolutionary",
"biology/evolutionary",
"and",
"comparative",
"genetics"
] | 2008 | An Approximate Bayesian Estimator Suggests Strong, Recurrent Selective Sweeps in Drosophila |
Determining distances to objects is one of the most ubiquitous perceptual tasks in everyday life . Nevertheless , it is challenging because the information from a single image confounds object size and distance . Though our brains frequently judge distances accurately , the underlying computations employed by the brain are not well understood . Our work illuminates these computions by formulating a family of probabilistic models that encompass a variety of distinct hypotheses about distance and size perception . We compare these models' predictions to a set of human distance judgments in an interception experiment and use Bayesian analysis tools to quantitatively select the best hypothesis on the basis of its explanatory power and robustness over experimental data . The central question is: whether , and how , human distance perception incorporates size cues to improve accuracy . Our conclusions are: 1 ) humans incorporate haptic object size sensations for distance perception , 2 ) the incorporation of haptic sensations is suboptimal given their reliability , 3 ) humans use environmentally accurate size and distance priors , 4 ) distance judgments are produced by perceptual “posterior sampling” . In addition , we compared our model's estimated sensory and motor noise parameters with previously reported measurements in the perceptual literature and found good correspondence between them . Taken together , these results represent a major step forward in establishing the computational underpinnings of human distance perception and the role of size information .
The perception of distances by monocular vision is fundamentally ambiguous: an object that is small and near may create the same image as an object that is large and far ( Figure 1A ) . More precisely , the monocular image size of the object ( , visual angle ) does not uniquely specify the physical distance ( ) , because and the object's physical size ( , diameter ) are confounded , . Subjectively we are not usually aware of this visual ambiguity because we perceive object distances unambiguously across a variety of conditions – this work examines how humans perform distance disambiguation by studying whether and how haptic size information is applied to these judgments . Despite previous evidence that adults [1] and infants [2] use object size information , like familiar size , to disambiguate ( Figure 1B ) the otherwise ambiguous visual information , debate exists [3] , summarized by [2] . Recently , Battaglia et al . [4] reported that the brain merges image and haptic sensations in a principled fashion to unambiguously infer distance . Incorporating haptic size information is particularly interesting because it requires sophisticated causal knowledge of the relationship between distance , size , and the multisensory sensations available to the brain to overcome size/distance ambiguity . Bayesian models provide the exact machinery needed to capture the size-distance perceptual ambiguity , the knowledge required to interpret noisy sensations , and how noisy sensations should be merged with prior knowledge to draw statistically sound perceptual estimates of object distances . This work uses Bayesian models to explicate , test , and confirm/deny a variety of hypotheses about the role of size information in human distance perception . Our results provide a significantly more comprehensive , quantitative account of the underlying computational processes responsible for incorporating size information into distance perception than any previous report . We formulated a family of Bayesian perception/action models , whose model structure and parameters encoded different assumptions about observer's internal knowledge and computations . We analyzed Battaglia et al . 's [4] data within this context , and used statistical model-selection methods to infer the most probable model and associated parameters for explaining their data . By committing to a full probabilistic model of observers' sensation , perception , and decision-making processes , we leveraged Battaglia et al . 's [4] data to uncover properties of: 1 ) the image and haptic sensory noise , 2 ) the observer's prior knowledge about size and distance , their causal relationship with the sensations , and how they are applied during perceptual processing , and 3 ) the decision-making strategy by which observers' perceptual inferences yielded psychophysical measurements . Important elements obscured from Battaglia et al . 's [4] original analyses were revealed: the present findings answer four key questions about how size influences human distance perception ( described in Model section ) . Using a full observer model allows us to transcend simplistic debates about whether humans are “optimal vs . sub-optimal” by providing a more textured account of perceptual phenomena that quantifies the sensory quality , what internal knowledge is involved , how they are merged and exploited , and how decisions result . This allows vague questions like “Is perception Bayesian ? ” to be reformulated into more precise ones like “To what degree does the brain encode uncertainty and apply structured knowledge to perceptual inference ? ” Our family of candidate observer models treat the world , observer , and observer's responses as one coherent interrelated physical system , which are represented in the models' structures and parameters using formal probabilistic notation . The fundamental assumptions are that world properties ( and ) generate pieces of sensory evidence , or cues , ( , and the haptic size information ) , and the observer's perceptual process uses probabilistic ( i . e . sensitive to various sources of noise and uncertainty ) inference to compute the posterior distribution over the distance given sensory cues , and ( Figure 1 ) . The literature [5]–[8] reports many similarities between behavior prescribed by optimal Bayesian inference models , and humans' use of sensory cues , prior knowledge , and decision-making for perceptual inference . The perceptual task used by [4] is well-suited to Bayesian modeling because of important effects of uncertainty and especially the use of auxiliary information ( in this case , ) for disambiguating hidden causes ( i . e . ) . In fact , disambiguation of hidden causes using indirectly-related data is a key , beneficial feature of Bayesian inference , termed “explaining-away” [9]; we hypothesize that human distance perception in the presence of auxiliary size cues is consistent with probabilistic explaining-away . Battaglia et al . 's [4] experimental task asked participants to intercept a moving ball , and treated their interception distances as perceptual distance judgments . Specifically , participants intercepted the ball as it moved at some distance , after a brief exposure to the ball that in some cases offered the ability touch the ball and feel its physical size and in other cases did not provide explicit size sensations . Our candidate observer models also make distance judgments using the sensory input available to human participants , so a direct comparison between human and model behaviors is possible . We derived all our candidate models from a base , ideal observer model ( IO ) that contains internal knowledge about the distributions of sensory noise that corrupt the sensations and , has knowledge about the prior distributions over and , the relationship between , , and , and the relationship between and ( Figure 1 , lower-right insets , black arrows ) . In Bayesian parlance these pieces of knowledge fall under the rubric of generative knowledge , or background information about the data's generative process that can aid in inferring the underlying causes . The IO estimates by computing and selecting the that maximizes it ( “maximum a posteriori” , MAP , decision rule ) . This computation requires merging image-size and haptic cues , as well as prior distance and size knowledge , in a manner Bayes' rule prescribes to yield optimal information about ( Figure 1's caption illustrates the inference process ) . We formulated this IO , as well as the other candidate observer models by enumerating all combinations of the following hypothetical questions: 1 ) Does the observer use the haptic size cue ? , 2 ) Does the observer know the haptic cue's reliability , and integrate the cue appropriately ? , 3 ) Does the observer know the image-size cue's reliability , and integrate the cue appropriately ? , 4 ) Does the observer perform MAP estimation , or rather estimate the distance by averaging a limited number of samples drawn from the posterior ? The models were designed to allow standard model-selection methods to decide which hypothetical candidate model , and associated parameters , were best-supported by the experimental human data . Thus we were able to select the most accurate hypothesis , among the field we pre-specified , as the best explanation for how human distance processing uses size information . Moreover , we compared the resultant parameter estimates with measurements reported by other studies , and found they conform with previous findings regarding perception's computational dynamics , which provides independent verification of our conclusions' validity . Our results indicate humans incorporate haptic size information for distance perception , consistent with Bayesian explaining-away . We also found that all but one participant underestimated the haptic cue's reliability ( specifically , they overestimated its sensory noise variance ) and integrated the haptic information to a lesser degree than the IO prescribed , similar to the human underuse of auditory information for spatial localization reported by [10] . We found that participants' priors over size and distance were comparable to the experiment's actual random size and distance parameter distributions , implying participants applied knowledge of probable stimulus parameters in their perceptual processing ( possibly learned or assumed during the experiment ) . Last , the sample-averaging estimation model , as opposed to the MAP-estimator , best-accounted for participants' distance judgments , a finding consistent with a growing body of results from perceptual studies that suggest perceptual judgments result from posterior sampling processes [11]–[13] .
The scene properties relevant for object distance perception are the object's physical distance and physical size; the relevant sensory cues they generate are visual angle and felt ( “haptic” ) size . As noted in the Introduction , visual angle is proportional to the ratio of size and distance; so , taking the of each of these variables transforms this relationship into a linear sum ( below ) . Our sensation model uses this -transformed representation for two reasons: 1 ) Weber-Fechner phenomena support a noise model in which the standard deviation linearly scales with signal magnitude ( which can be accomplished with independent noise in log-coordinates ) , and 2 ) this -linear approximation is analytically tractable , as we will show . So we assume a linear Gaussian model , meaning the scene properties are a priori Gaussian distributed , and the sensory and motor noise are additive , zero-mean Gaussian , and the sensory generative process is linear , in the log domain . Log-distance , log-size , log-visual angle , and log-haptic size are represented as: , , , , respectively . The relationship between , , and ( by “small angle approximation” to ) is:and between and is:where and represent image-size and haptic sensory noise with standard deviations ( SDs ) and , respectively . The notation indicates that the parameter represents a property of the scene; this is distinct from the observer's knowledge about the scene , defined in the next section with no tilde . It follows that the distribution of sensory cues conditioned on the scene properties are: ( 1 ) ( 2 ) We assume observers' internal prior probabilities over and are: Battaglia [14] derives model observers for perceptual inference in linear Gaussian contexts under a variety of assumptions – we co-opt the “explaining-away” derivations ( Sec . 3 . 4 in [14] ) for the current size/distance perception context . All model observers are assumed to use their knowledge of the world , i . e . the sensory noise ( and ) and prior distributions ( , , , and ) , to compute beliefs about . These beliefs are represented as the posterior distribution , ( which is Gaussian ) : ( 3 ) where , ( 4 ) For those familiar with “standard” cue combination , Eqs . 3 and 4 are similar to the “optimal cue combination” formulae in [15] , and in fact by looking closely at the Bayes' net in the lower right of Fig . 1B , one can see that the subgraph composed of variables , , and represents the standard two-cue “cue combination” situation . However , our present situation is distinct from [15] because we focus on data fusion in conditions where one cue ( ) is only indirectly related to the desired property ( ) by its ability to disambiguate another cue ( ) . The intuition for the weights in Eq . 4 is as follows . Because provides information about to improve inference of , the numerator of assigns sensory cue more influence when prior knowledge of and are weaker ( higher and ) . Similarly , 's numerator dictates that is more influential when information about is weaker ( higher ) . Interpreting is less straightfoward , but essentially holds that when information about is poor , because both the prior over and sensory cue are weak ( higher and ) , then is more exclusively influential for inferring , whereas if either prior knowledge about or sensory cue are strong , and that information jointly guide inference of . Last , 's numerator assigns stronger influence to prior knowledge of only when the sensory cues and prior knowledge of are weak . Human observers who do use for distance perception are modeled above by Eq . 4 . The hypothesis that observers do not use , either because is unavailable or because they are not capable , is formulated: ( 5 ) where , ( 6 ) Eq . 5 is algebraically equivalent to taking in the formulation in Eq . 4 . Whether humans do ( Eq . 3 ) or do not ( Eq . 5 ) use to make distance judgments is the first of our hypothesis questions ( see Table 1 ) . Also , whether humans know the true sensory noise magnitude i . e . whether they use vs . , and/or vs . , are the second and third of our hypothesis questions ( Table 1 ) . The model observer uses beliefs about to select a position at which to intercept the moving ball . We assume that participants attempt to minimize the difference between their judged distance and the true distance , which for Gaussian distributions may equivalently correspond to minimizing a MAP , mean-squared , or symmetric Heaviside loss functions . However accessing their perceptually-inferred information about is not necessarily trivial: we consider that they may select the maximum probability , i . e . ( or ) ) , as their judgment of distance , or instead draw a number , , of independent samples from or and compute their sample mean as a judgment , is the fourth ( and last ) of our hypothesis questions ( Table 1 ) . These distinct models may imply different neural representations for posterior beliefs about distance , which we address in the Discussion . Additionally our models all include an element of motor noise , the small degree of error between judged and the experimentally-measured , due to motor imprecision when performing an interception . For consistency with known parameters of motor control , we selected an additive , Gaussian motor noise term , that was added to the distance judgment to form . We combine the sensation , perception , and decision-making models described above to define a set of coherent model observers that input sensations , combine them with internal knowledge to form beliefs about distance , and form decisions that are output as interception responses in the experimental task . By varying the models' structure and parameters we encoded the four hypothesis questions in the Introduction ( subsequently referred to as “Q I , II , II , IV” ) to form the candidate observer models ( Table 1 ) : In total 12 distinct candidate models spanned the possible combinations of the four questions ( the reason the total is 12 , instead of 16 , is because for candidate models that do not include the use of haptic information [Q I] , the question of whether the observer knows the haptic cue noise magnitude or not [Q II] is inconsequential and those models are redundant ) . First , we describe how the model observers predict responses in the experimental interception task and illustrate responses produced by each model . Second , we describe how the model's parameters were inferred given each participants' response data . Third , we show how we computed the human data likelihood under each model and how we quantitatively compare them to determine which model provides the best account of the human data .
The central result of our study is quantitative selection of the model that best explains the data , which we determine by comparing the models' DIC scores , to answer the four hypothetical questions posed above . Figure 5A shows raw DIC scores , and Figure 5B shows the difference between the best model's DIC ( indicated by circle on x-axis ) and the other models' DIC scores . We defined DIC significance as described in the previous subsection: models whose DIC differed by greater than 10 were deemed “sigificantly” different ( dashed horizontal line and * in Figure 5B ) and greater than 15 deemed “highly significantly different” ( solid horizontal line and ** in Figure 5B ) ; this is a conservative modification of the criteria mentioned in [17] . We found that all participants incorporate haptic size information to make their distance judgments ( Q I ) . Also , we found 5 of 6 participants misestimated their haptic size noise and thus incorporated the haptic information less than optimally prescribed , while one participant applied the haptic cue in proportion to its reliability ( Q II ) ; the following section addresses the nature of the misestimation . All participants incorporated the visual image-size cue optimally , in accordance with its noise magnitude ( Q III ) . All participants used a sample-averaging strategy over MAP decision-making ( Q IV ) . With respect to Q IV , the DIC scores were always worse for the MAP model versus the sample-averaging model , by an average DIC difference of 129 ( Figure 5 ) , so we exclusively focus on the sample-averaging models ( odd numbers ) for the remaining discussion . Figure 5 depicts each participant's DIC scores for each sample-averaging model , the left graph shows the absolute DIC values and the right graph shows the differences between best model DICs and the other models' DICs . Participant 6 was an author . Participant 3's DIC differences between Model 7 and Models 5 and 11 was not significant under our conservative criteria , however Model 7 was still better by DICs of and , respectively , which is considered marginally significant under typical uses of AIC/DIC [17] . Participant 5 , the only participant whose DIC favored the hypothesis that the haptic noise magnitude was correctly known ( Q II ) , had the worst DIC scores across participants , as well as substantially different parameter value estimates from the other participants ( see next paragraphs ) . Upon closer inspection of participant 5's data , it was qualitatively the noisiest: in Battaglia et al . 's [4] simple regression analysis of this data their statistical analysis determined participant 5's data was so significantly different from the other participants' that it ought to be excluded as an outlier . The reason we included it in the current analysis was to determine whether there was still some patterns the previous analysis had not detected . Though the parameters still yield meaningful values , because of the major differences between raw DIC scores , the DIC-favored model , the parameter value estimates , and the general noisiness of the response data , we strongly suspect this participant either was not focusing on performing this task , was randomly selecting answers on a large fraction of the trials , and in general should be distinguished in further analysis due to these aberrations: so , we report participant 5's parameter estimates separately from the other “inlier” participants . Figure 6 shows Participant 1's model-predicted compared against the actual values , for the best model , 7 , as well as several that differ by one assumption ( Table 1 ) . The spread in the dots is due to sensory noise and the random posterior sampling process , how neatly the actual data falls within the ranges predicted by a particular model ( black error bars ) is indicative of the model's explanatory quality . Notice the pattern of more varied no-haptic vs . haptic in Models 7 , 11 , 5 , a direct prediction of the sampling models over MAP . Though MAP decisions incur more bias in the no-haptic condition , they actually have less trial-to-trial variance than in the haptic condition . This is due to the fact that the prior does not vary between trials , while the more informative haptic cue does . A possible concern is that participants learned to use the haptic cue during the course of the experiment , and that the weak DIC scores of Models 1–4 in comparison to Models 5–12 actually reflect the effects of associative learning rather than knowledge the participants brought into the experiment . We evaluated this possibility by performing the same DIC analysis on data from only the first day to test whether Models 5–12 were still favored over their 1–4 counterparts . The results unequivocally confirm the results on the data from the final 3 days above: for every participant , the DIC analysis across the models shows that the no-haptic models ( 1–4 ) have worse DIC scores than their haptic model counterparts ( 5–12 ) . The best no-haptic models' DICs are below the best haptic models' DICs by margins of { , , , , , } for Participants 1 through 6 , respectively . In fact , removing the sampling models , even the no-haptic models ( 1 and 3 ) with the best DIC scores still have worse scores than the haptic models ( 5 , 7 , 9 , and 11 ) with the worst DIC scores . This firmly supports conclusion that the haptic cue is used even on the first day of trials . Though it might seem that given the 6 to 10 “free” parameters in our general observer model , we could “fit” any data , we are actually inferring the best parameters and using the posterior's expected values rather than the most probable a posteriori parameters . Moreover DIC acknowledges the possibility for overfitting and counters it by penalizing overfits through the complexity term , thus affirming that the chosen model's structure and parameters are accurate and robust explanations of the humans' judgments . Moreover because we encoded different hypotheses within the models we could clearly distinguish those hypotheses best-supported by the data . Lastly , despite the possibility that we could fit a variety of data , the remainder of this section shows that the individual inferred parameter values are consistent with known perceptual parameters measured in other studies . A secondary result of this work , beyond providing answers to the 4 hypothetical questions , is that the inferred parameter values ( ) our analysis yielded can be meaningfully interpreted . Though there is no guarantee that the inferred parameters are unique , they offer an indication of what the analysis finds probable . All reported parameters are MCMC expections , from which we compute meansSEs across participants and report the values in log coordinates . First , we present the SDs in terms of Weber fractions for the sensory noise , with discrimination thresholds corresponding to . The image-size noise SD , , and assumed noise SD , , were coupled in the best-fit models ( 7 and 11 ) for all participants . Their values correspond to Weber fractions of to ( meanSE of ) for the inlier participants , and for participant 5 . This is comparable to the Weber fractions of measured in humans by [18] for parallel line separation discrimination , and by [19] for line length discrimination . Because our task did not involve interval-wise discrimination of pairs of stimuli , but rather absolute perception , it is to be expected that our noise magnitudes will be slightly higher . The haptic noise SDs , , and assumed haptic noise SD , , were uncoupled in the inlier participants' best-fit model ( 7 ) , and coupled for participant 5's best fit model . The inlier participants' haptic noise SDs correspond to Weber fractions of between and ( meanSE of ) , and for participant 5 . A Weber fraction of was measured in humans by [20] for haptic size discrimination of objects between and mm in width using a similar haptic stimulus presentation apparatus , but with two fingers gripping the object rather than one finger probing the size . Because two fingers are likely to provide a more precise size measurement and because their participants performed interval discriminations of pairs of objects , our somewhat elevated Weber fraction are reasonable values . The inlier participants overestimated their haptic noise SDs , with their assumptions corresponding to Weber fractions of to ( meanSE of ) . The consequences of overestimating haptic noise are that the observers do not achieve the level of disambiguation possible by fully incorporating the haptic cue , and apply prior knowledge about the ball's size and distance relatively more heavily ( Figure 3 ) . Our analysis provided information about the observer models' prior knowledge , and found it strikingly similar to the sample statistics of the experimental stimuli's distances and sizes , with slightly higher SDs ( remember the stimuli were uniformly distributed in the mm domain ) . The meanSE estimated prior distance mean and SD parameters , and , across all participants were and log-mm , respectively; the experimental distance mean and SD were and log-mm , respectively . The mean estimated prior size mean and SD parameters , and , across participants were and log-mm , respectively; the experimental size mean and SD were and log-mm , respectively . This indicates participants learned the range of possible stimuli presented in the experiment and applied that knowledge toward improving their judgments , to the effect of lowering the posterior variance ( Figure 3 ) . To further investigate the source of participants' prior knowledge , we ran our full analysis on only the first day of participants' trials , to measure what difference between inferred parameters exist between early and later in the experiment . We found that participants' first-day priors for and were and log-mm , respectively; and , participants' first-day priors for and were and log-mm , respectively . So , the prior means did not shift significantly ( in terms of SE interval overlap ) , but the prior SD values did . It appears that participants rapidly learned the prior means , which are more easily estimable from experience and also may be assumed to some extent ( the true prior distance mean is at the center of the virtual workbench , and the balls' sizes were directly observable in haptic condition trials ) . However , participants appeared to use more diffuse prior and parameters early in the experiment , which is consistent with making weaker prior assumptions about the range of distance/size variation ( top row of Figure 3 ) . Our analysis provided estimates of participants' motor noise SD , whose meanSE across participants was log-mm , which amounts to a SD of mm at a reach distance of mm , and mm at reach distance of mm , the extremal distances presented in the experiment . A value of log-mm was reported [21] under similar reaching conditions . The sample-averaging models generally outperformed the MAP estimate models ( Q IV ) with respect to DIC scores . The inlier participants had values between and ( meanSE of ) , and participant 5's estimate was . Of course in our model must be integer-valued , but these real valued estimates are robust means across our MCMC analysis samples . An alternative interpretation of is that it is an exponent applied to the posterior distribution , from which one sample is then drawn after renormalizing . For a Gaussian distribution , because , drawing sample , , from yields a Gaussian-distributed: . And , drawing samples , , from the unexponentiated and averaging yields a sample mean with the same distribution: Between the DIC analysis and the validity of the inferred parameters , we conclude that model 7 is both structurally and parametrically accurate . This strongly supports model 7 and its encoded hypotheses as a coherent computational account of the underlying processes responsible for size-aided distance perception .
We conclude that humans can use haptic size cues to disambiguate and improve distance perception , but that the degree to which they incorporate haptic size information is lower than the ideal observer prescribes . We also conclude that the distance responses are best explained as a process of drawing several samples from the posterior distribution over distance given sensations , and averaging them to form a distance estimate . This behavior is broadly consistent with a Bayesian perceptual inference model in which mistaken generative knowledge about haptic cues is used , and beliefs about distance are accessed by drawing samples from an internal posterior distribution . The brain's use of sensory cues for disambiguating others has been reported in a variety of perceptual domains , and broadly falls under the category “perceptual constancy” . Constancy effects , like the present distance constancy , involve situations in which an observer cannot unambiguously estimate a scene property due to confounding influences from other “nuisance” properties , and so leverages “auxiliary” cues ( in this study , haptic size ) to rule out inconsistent possibilities . Auxiliary disambiguation effects , like constancy , have other names in the literature , like “cue promotion” [22] , “simultaneous contrast” [23] , and “taking-into-account” [24] . Many studies have reported “size constancy” , distance cues disambiguating object size perception [25]–[32] , so it is not entirely surprising that size cues can conversely disambiguate distance perception . Humans underestimating non-visual cue reliabilities and thus integrating them less strongly has been measured before by [10] , [32] . There are several potential reasons for this phenomenon , one idea that has recently garnered support [33]–[36] is that sensory cues are used in accordance with their causal relationships to the unobserved scene properties: when the brain believes cues are unrelated to the desired scene property , it down-weights or outright ignores them . In the present study , this would mean the brain is unwilling to fully apply the haptic size cues because they might originate from a source independent of the ball , for instance imagine the hand touched a ball behind a photograph of a different ball; of course , such miscorrespondences are uncommon in nature , but examples like “prism adaptation” demonstrate the brain can accommodate and recalibrate in such situations . Another possibility is non-visual cues to spatial properties may be experienced far less frequently in life , and had fewer opportunities on which to be calibrated , so they are mistrusted . Our finding that all our observers' responses are best modeled as sampling the posterior is consistent with recent studies and ideas about the representation and computation of probability in the brain . Using posterior sampling to generate responses in a choice task should manifest as probability matching of the options , a common finding in many behavioral tasks , including a perceptual audio-visual cue-combination task [13] . Sampling has also been used to provide a novel explanation for perceptual switching to multistable displays [11] , [37] . Moreover , sampling provides an interpretation of neural activity in population codes and makes difficult probabilistic computations simple to neurally implement ( see review by [38] ) . Although Bayesian decisions are usually modeled as maximizing the posterior , maximization is not the best decision rule in all instances . MAP's optimality depends on both the task and the veridicality of the decision maker's posterior distribution . MAP assumes the decision maker's goal is to maximize the number of correct responses and that the posterior is based on the correct generative model for the data . When the posterior is not correct , basing responses on sampling provides exploration that can be used to improve the decision maker's policy . This idea has been extensively explored within reinforcement learning , where exploration is frequently implemented using a softmax decision strategy [39] where choices are stochastically sampled from an exponentiated distribution over the values of a set of discrete options . This idea can be generalized to the case of continuous decision variables . The value of an estimate is based on the reward function for the task . In our decision task , participants were “correct” whenever their choices fell within a narrow region relative to their posterior distribution . Approximating the experimental reward function as a delta function , the optimal strategy is to maximize the posterior . However , if we need to improve our estimate of the posterior , then it is important to estimate the error . Sampling from the posterior gives a set of values that can be used to compute any performance statistic , making it a reasonable strategy when an observer is needs information needed to learn - i . e . to assess and improve performance . Though our models posit observers draw samples directly from the posterior and averaging , any decision rule that is sensitive to the posterior variance may produce similar predictions – for instance , it is possible that participants internally exponentiate the posterior and draw exactly one sample ( detailed in Results ) . This means that for greater exponents , the posterior is more greatly sharpened; as the exponent approaches infinity , the posterior approaches a delta function located at the MAP estimate ( after re-normalizing ) . This is a general strategy used in many machine learning domains to transition neatly between posteriors , MAP estimates , and “watered down” versions of the posterior . However we find this account unappealing because it implies that drawing more than one samples is less attractive to the observer's underlying perceptual mechanics than performing posterior exponentiation . Also , though our models assume posterior sample-averaging is a source for behavioral response variance ( Figure 4 ) , another possibility is that observers have uncertainty in the parameter values that characterize their generative knowledge itself , and actually draw samples of generative parameters instead of using deterministic parameter estimates . For instance , when combining haptic cues they may sample from an internal distribution over haptic reliability ( ) . This could be a strategy for learning when the brain is uncertain about internal generative model parameters; because the observer receives feedback , and presumably wishes to calibrate the internal perceptual model , varying behavior by using different samples of internal model parameters avoids redundant feedback associated with similar behavioral responses to similar input stimuli . Using a full probabilistic model of observers' sensation , perception , and decision-making processes provide us with answers to the four key questions we posed in the Model section . This study's analysis of data reported by [4] resulted in a much more comprehensive account of the computations responsible for distance and size perception . By formally characterizing a set of principled computational perception hypotheses , and choosing the best theoretical account of the measured phenomenology using Bayesian model selection tools , we demonstrated the power , robustness , and flexibility of this coherent framework for studying human cognition , and obtained deeper understanding of distance perception . | Perceiving the distance to an object can be difficult because a monocular visual image is influenced by the object's distance and size , so the object's image size alone cannot uniquely determine the distance . However , because object distance is so important in everyday life , our brains have developed various strategies to overcome this difficulty and enable accurate perceptual distance estimates . A key strategy the brain employs is to use touched size sensations , as well as background information regarding the object's size , to rule out incorrect size/distance combinations; our work studies the brain's computations that underpin this strategy . We modified a sophisticated model that prescribes how humans should estimate object distance to encompass a broad set of hypotheses about how humans do estimate distance in actuality . We then used data from a distance perception experiment to select which modified model best accounts for human performance . Our analysis reveals how people use touch sensations and how they bias their distance judgments to conform with true object statistics in the enviroment . Our results provide a comprehensive account of human distance perception and the role of size information , which significantly improves cognitive scientists' understanding of this fundamental , important , and ubiquitous behavior . | [
"Abstract",
"Introduction",
"Model",
"Results",
"Discussion"
] | [
"computer",
"science",
"psychology",
"reasoning",
"social",
"and",
"behavioral",
"sciences",
"cognitive",
"psychology",
"psychophysics",
"sensory",
"perception",
"computerized",
"simulations",
"neuroscience"
] | 2011 | How Haptic Size Sensations Improve Distance Perception |
Idiopathic chronic diarrhea ( ICD ) is a leading cause of morbidity amongst rhesus monkeys kept in captivity . Here , we show that exposure of affected animals to the whipworm Trichuris trichiura led to clinical improvement in fecal consistency , accompanied by weight gain , in four out of the five treated monkeys . By flow cytometry analysis of pinch biopsies collected during colonoscopies before and after treatment , we found an induction of a mucosal TH2 response following helminth treatment that was associated with a decrease in activated CD4+ Ki67+ cells . In parallel , expression profiling with oligonucleotide microarrays and real-time PCR analysis revealed reductions in TH1-type inflammatory gene expression and increased expression of genes associated with IgE signaling , mast cell activation , eosinophil recruitment , alternative activation of macrophages , and worm expulsion . By quantifying bacterial 16S rRNA in pinch biopsies using real-time PCR analysis , we found reduced bacterial attachment to the intestinal mucosa post-treatment . Finally , deep sequencing of bacterial 16S rRNA revealed changes to the composition of microbial communities attached to the intestinal mucosa following helminth treatment . Thus , the genus Streptophyta of the phylum Cyanobacteria was vastly increased in abundance in three out of five ICD monkeys relative to healthy controls , but was reduced to control levels post-treatment; by contrast , the phylum Tenericutes was expanded post-treatment . These findings suggest that helminth treatment in primates can ameliorate colitis by restoring mucosal barrier functions and reducing overall bacterial attachment , and also by altering the communities of attached bacteria . These results also define ICD in monkeys as a tractable preclinical model for ulcerative colitis in which these effects can be further investigated .
The incidence of inflammatory bowel diseases ( IBD ) is highest in industrialized regions wherein helminth infections have been largely eliminated , raising the hypothesis that helminths may protect against intestinal inflammation underlying the disease [1] . Indeed , there is evidence that experimental helminth treatment can ameliorate symptoms in IBD patients [2] , [3] and in mice [4] , [5] , [6] , [7] . This effect may be attributable to the induction of immunoregulatory networks [8] , [9] and mucosal repair by a TH2-type immune response [10] , [11] . The pig whipworm , T . suis , is currently being used in clinical trials to alleviate symptoms in IBD patients , but the mucosal immune responses in treated patients has not been investigated [11] . Recently , we characterized the mucosal responses of an ulcerative colitis ( UC ) patient who self-infected with the human parasite T . trichiura [12] . These studies led to the proposal that the TH2 immune response activated for parasite clearance , as well as the induction of IL-22 expression , may promote mucosal healing in UC patients by increasing mucus production and turnover of intestinal epithelial cells [11] . In a different study , the mucosal responses of celiac disease patients were examined in detail after experimental treatment with human hookworm infection , showing a suppression of TH17 responses and upregulation of IL-22 [13] , [14] , [15] . This growing body of clinical evidence suggests that a better understanding of mucosal responses activated by helminth infections may contribute to new IBD therapies . Non-rodent , preclinical models of IBD are lacking , and would greatly benefit the development of such novel therapeutics . ICD , leading to progressive weight loss and dehydration [16] , frequently afflicts captive juvenile rhesus macaques ( Macaca mulatta ) and is a leading cause of death at primate research centers . The clinical management of these animals is difficult because the condition is often refractory to available treatment . Intestinal inflammatory pathology during ICD can be similar to that found in UC and it has been suggested that ICD may be an informative model for UC [17] , [18] . As in the case of UC , inflammation is more severe in the colon than in the jejunum and it has been shown that dysregulation of IL-6-STAT3 activation may be an important inflammatory mediator [19] . In this study , we sought to characterize the mucosal inflammatory response driving ICD in the rhesus monkey , and to determine whether helminth exposure can modulate this inflammatory response and lead to clinical improvement . It is conceivable that the protective effects of helminth treatment for IBD patients may be partly attributable to alterations in the microbial communities of the intestinal tract [11] , [20] . Changes to the intestinal mucosa by helminth infection [21] are likely to have a major impact on the gut microbial environment [11] . H . polygyrus infection has been shown to have major effects on the microbiota of mice , especially increasing the abundance of the Lactobacillaceae family [20] . Colonization of the colon with T . muris has been shown to be dependent on the gut microbiota [22] . Uncultured bacterial communities can now be analyzed using DNA barcoding and deep sequencing approaches to investigate how interactions between communities can influence disease pathogenesis [23] , [24] . In a previous cross-sectional study , McKenna et al . investigated the microbiome of healthy macaques in comparison to macaques with chronic diarrhea and after SIV infection [25] , finding that bacterial diversity was significantly lower and the family Campylobacteraceae was more common in monkeys with colitis . In this study , we have measured the quantity of attached bacteria to the intestinal mucosa pre-and post-helminth treatment , and further characterized the composition of bacterial communities by deep sequencing in order to determine whether helminth treatment alters the communities of attached bacteria within the intestinal tract of macaques with ICD .
Five juvenile rhesus macaques diagnosed with ICD were enrolled ( see Methods for selection criteria ) and monitored daily before and after oral administration of 1000 T . trichiura ova ( Figure 1 ) . The extent of diarrhea ( Figure 1 ) was scored daily and an improvement in fecal consistency ( Figure 1A ) , accompanied by weight gain ( Figure 1B ) , was observed in four out of the five monkeys ( TC01–TC04 ) following T . trichiura treatment . Subject TC05 continued to have severe diarrhea and weight loss , and eventually had to be euthanized . Importantly , patent infection was not established in any of the monkeys despite symptomatic improvements and eggs were never detected in the feces . Spontaneous remission does not typically occur in untreated animals , which maintain similar fecal consistency scores in other experiments ( Figure S1 ) . To assess changes to the intestinal mucosa before and after helminth treatment , colonoscopies were performed and mucosal biopsies were collected from the five animals prior to treatment , as well as from two healthy , age-matched controls . A second colonoscopy was performed on the ICD subjects 14 weeks after oral administration of T . trichiura ova . Peripheral blood mononuclear cells ( PBMC ) and lamina propria mononuclear cells ( LPMC ) harvested from colon biopsies were stimulated with PMA and ionomycin for intracellular cytokine staining and flow cytometry analysis ( FACS ) . The proportion of colonic CD4+ T cells producing IL-4 , but not those producing IFNγ , was significantly higher ( p<0 . 001 ) following T . trichiura treatment ( Figure 2A ) , revealing a localized TH2 response . In PBMCs , we did not detect more CD4+ T cells producing IL-4 after treatment ( data not shown ) . All five ICD subjects demonstrated this response , indicating that the lack of clinical improvement in subject TC05 was not due to the absence of a TH2 response . The colonic mucosa of ICD subjects also showed a higher proportion of Ki67+ CD4+ T cells ( Figure 2B ) , indicative of ongoing inflammation . The four clinical responders showed a diminished fraction of such cells following T . trichiura treatment , while their proportion actually increased in subject TC05 . The proportion of FoxP3+ CD4+ regulatory T ( Treg ) cells was greater in the colonic mucosa of the ICD subjects compared to healthy controls and was markedly reduced in three out of four clinical responders following T . trichiura treatment ( Figure 2C ) . This observation suggests that the local expansion of Tregs in these subjects was reflective of active ongoing inflammation . We isolated RNA from pinch biopsies paired with the samples used for FACS analysis . Whole genome expression profiling of colon biopsy samples identified 185 transcripts that distinguished the inflamed mucosa of ICD subjects ( prior to T . trichiura treatment ) from healthy colon tissue ( Figure 3A , Figure S2 , Table S1 ) . The genes upregulated in ICD samples included classical TH1-type inflammatory mediators such as inducible nitic oxide synthase ( NOS2 ) , chemokines ( CXCL9 , CXCL10 , CXCL11 ) , and serum amyloid A ( SAA1 , SAA3 , SAA4 ) . IBD-associated genes implicated in mucosal injury and defense [including regenerative factors ( REG1 , REG3 ) , trefoil peptides ( TFF1 ) , and defensins ( MNP2 , ROAD1 , ROAD2 ) ] were also upregulated in ICD samples . This transcriptional signature is consistent with previous studies on UC patients [12] , [26] and could be driven by attached mucosal microbiota [26] . Pathway analysis indicates enrichment for genes in the biological processes that include the JNK cascade , MAPKKK cascade , response to IFN-g , IκB Kinase/NF-κb cascade and macrophage activation as shown in Figure 3B . Changes in gene expression induced by helminth treatment were also evaluated and 99 transcripts were found to be significantly altered following T . trichiura treatment ( Figure 3B , Figure S3 , Table S2 ) . Notably , many of the IBD-associated genes identified in ICD samples were downregulated following T . trichiura treatment . Furthermore , post-treatment samples demonstrated the induction of TH2-type immune response pathways , including those associated with IgE signaling ( FCER1A , MS4A2 ) , mast cell activation ( CPA3 , CMA1 ) , TH2 and eosinophil recruitment ( CCL17 , CCL18 , CCL26 ) , alternative activation of macrophages ( ALOX5 , ALOX15 ) , cytokine signaling ( IL5RA , IL9R , POSTN ) , and worm expulsion ( RELMB ) . This shift was confirmed by quantitative reverse transcription ( qRT ) -PCR ( Figure 4 ) . KEGG pathway analysis showed significant ( P = 0 . 0136 ) enrichment for genes in the arachidonic acid pathway ( ALOX5 , ALOX15 , LTC4S , PTGS1 ) indicating the production of prostaglandins , which are important regulators of the inflammatory response [27] . The transcriptional profile of subject TC05 clustered separately from the four clinical responders , a distinction primarily driven by a group of immunoglobulin-related transcripts present at much lower levels in this subject ( Figure S4 and Table S3 ) . TC05 did not demonstrate a shift from a TH1-type to TH2-type gene expression pattern following T . trichiura treatment as seen in the four responders , resulting in the close hierarchical clustering between pre- and post-treatment samples from this animal . This observation was confirmed by qRT-PCR ( Figure 4 ) , as TC05 did not show reduced expression of certain type-1 inflammatory genes ( NOS2 , IL1A ) or increased expression of TH2-type response genes ( CCL18 , CALPAIN13 , RELMB ) following T . trichiura treatment . Since IBD may be driven by an aberrant immune response against commensal gut bacteria due to changes in bacterial attachment to the intestinal mucosa [28] , [29] , we quantified the abundance of several bacterial taxa in colon biopsy samples using quantitative PCR for 16S ribosomal RNA genes . Biopsies from each ICD subject demonstrated a higher abundance of multiple taxa compared to healthy controls , revealing a broad increase in bacterial attachment ( Figure 5A ) . Bacterial attachment was reduced following T . trichiura treatment , suggesting that the defective mucosal barrier was partially restored . Surprisingly , the bacterial loads in biopsies from TC05 were only slightly higher than those found in control biopsies and only slightly reduced post treatment , suggesting that increased bacterial attachment may not have driven colitis in this subject . One possibility would be that subject TC05 is more sensitive to bacterial translocation even with small quantities of attached bacteria because of an inherently leaky mucosal barrier . Consistent with this possibility , only TC05 showed an increase in serum soluble CD14 ( a correlate of microbial product translocation across the mucosal barrier [30] ) following T . trichiura treatment ( Figure 5B ) . In addition to an increased bacterial load , changes to bacterial taxa may be linked with disease pathogenesis . To investigate the composition of microbial communities attached to the intestinal mucosa through culture-independent methods , deep sequencing analysis was performed on the variable region 4 ( V4 ) region of bacterial 16S rRNA [31] . On average , 8439±1373 ( SD ) sequences were obtained per sample . The microbial diversity within each sample ( α-diversity ) was compared between biopsies taken from control macaques and from macaques pre-treatment and post-treatment by measuring the Shannon index and phylogenetic diversity ( PD ) through rarefaction curves ( Figure 6A and S5 ) . Shannon diversity was higher in samples post-treatment compared to pre-treatment samples [although not statistically significant ( P = 0 . 0752 ) ] and similar to control samples , suggesting that helminth treatment promoted the restoration of the diversity of mucosal microbial communities . We then performed a Principal Coordinates Analysis ( PCoA ) on the unweighted UniFrac distances between samples to analyze differences between microbial communities ( β-diversity ) . We observed clustering of control samples with post-treatment samples and the separation of these samples with pre-treatment samples along the PC2 axis ( Figure 6C ) . Clustering patterns generated with Unweighted Pair Group Method with Arithmetic mean ( UPGMA ) trees were consistent with the PCoA plot ( Figure S5 ) . Interestingly , the non-responder TC05 was found to cluster separately from the other macaques both pre- and post-treatment . By overlaying taxonomic information onto the PCoA plot , we could illustrate that the phyla Bacteroidetes , Firmicutes , and Tenericutes contribute towards the clustering of control and post-treatment samples , whereas Cyanobacteria contributes towards the clustering of the pre-treatment samples ( Figure 6C ) . Indeed , it was striking to find that in three out of the five macaques pre-treatment samples , the bacterial phylum Cyanobacteria was vastly increased in abundance , representing between 10 . 8–32 . 9% of total sequences ( Figure 6B ) . Importantly , this taxon is no longer a prominent population post T . trichiura treatment . A single operational taxonomic unit ( OTU ) , most closely related to plastids from plant organisms ( Chloroplast: Streptophyta ) , constituted 99 . 2–99 . 6% of sequences from the Cyanobacteria phyla . BLASTN analysis of the sequence against the nr ( NCBI ) nucleotide database confirmed sequence identity with Pinus chloroplast . To further identify bacterial clades with statistically significant differences between control , pre-treatment , and post-treatment samples , we used the LEfSe method [32] for comparison between these groups ( Figure 7 and S6 ) . By defining pre-treatment and post-treatment samples as two classes of microbial communities , we found that Tenericutes and Bacteroidetes are more abundant in control and post-treatment samples ( consistent with the PCoA plot ) whereas the unclassified bacteria taxon ZB2 appears to be slightly enriched in pre-treatment samples ( Figure 7A–C ) . When we compared control with post-treatment samples ( Figure 7D and S6 ) , we found that Tenericutes is the only taxon more abundant in post-treatment samples , indicating that it could be induced in response to T . trichiura treatment .
The data presented here suggest that ICD in juvenile rhesus macaques is an IBD-like disease in which the mucosal barrier is compromised , allowing increased bacterial attachment that contributes to persistent inflammation and dysbiosis of the mucosal microbiota . This study revealed that T . trichiura treatment induces a TH2-type immune response in the intestinal mucosa of ICD subjects that was associated with symptomatic improvement . This positive clinical response was inversely correlated with cellular markers of mucosal T cell inflammation , such as Ki67+ T cells . In contrast to the TH2-type response , Treg expansion was noted under conditions of mucosal inflammation and was reduced by T . trichiura treatment , suggesting that the presence of Tregs in the intestinal tract is more indicative of ongoing intestinal inflammation than of helminth exposure . Based on these results , we speculate that T . trichiura promotes mucosal healing in the setting of ICD by activating mucus production and epithelial cell turnover , aspects of TH2-type immunity aimed at expelling the parasite [9] , [11] , [21] , thereby reducing the attachment of immunostimulatory bacteria to the colonic epithelium and restoring diversity to the mucosal micobiota ( Figure 8 ) . However , additional studies are needed to determine the mechanism of action for Trichuris-mediated amelioration of colitis , and functional studies to establish causation would have to be performed in a suitable mouse model to confirm these hypotheses . One weakness of this study is the absence of a control arm to ensure that the macaques with ICD did not undergo spontaneous remission . Future studies to confirm the findings presented here should have a sham-treated control arm , which would also enable a blinded assessment of stool frequency . The association of clinical improvement with weight gains , which is an objective measure , supports the observation that T . trichiura treatment did lead to improvement in symptoms . Additionally , our transcriptional profiling experiments clearly show that clinical improvement is associated with dramatic changes in mucosal gene expression patterns . The number of biopsies that could be collected from these juvenile macaques also limited our study . Histopathological observations that could be correlated with gene expression patterns would have provided additional insights into the relationship between cellular infiltrates and mucosal responses . While ICD shares some similar features with UC , inflammation is typically microscopic and cannot be determined objectively by visual inspection during endoscopy . The mucosal gene expression patterns ( e . g . , REG1 , REG3 , NOS2 , TFF1 , SAA1 , and SAA3 ) that distinguish ICD and healthy macaques are strikingly similar to signatures observed in microarray studies of UC patients [12] , [26] , [33] , [34] as well as in studies of germ free mice repopulated with various bacterial taxa [35] , [36] , suggesting that a response to gut bacteria may be the predominant driving force for this inflammatory signature . Importantly , differences in gene expression patterns were apparent between the four clinical responders and the single non-responder in this study . Of note , the non-responder showed lower expression of immunoglobulin-related genes ( Figure S4 ) , suggesting that the etiology of disease in this animal may have stemmed from a defect in the mucosal B cell compartment , as has been shown previously to precipitate colitis through the absence of regulatory B cells [37] . Recently , MyD88 signaling in B cells was found to prevent the lethal dissemination of intestinal bacteria after dextran sulfate sodium ( DSS ) treatment in mice [38] . The IL-23 , IL-17 and IL-22 network plays a critical role in intestinal homeostasis and IBD pathogenesis [39] . We previously reported that IL-22-producing CD4+ cells were a prominent feature of T . trichiura infection in a UC patient . Unfortunately , our efforts to stain for IL-17- and IL-22-producing CD4+ cells from the macaque pinch biopsies were unsuccessful ( data not shown ) . When we investigated the expression levels of the cytokines in these pathways pre- and post-T . trichiura treatment , we found no significant trends ( Figure S7 ) . Since the role of these cytokines in ICD of macaques is completely unknown , this may represent an important difference with human IBD , but it is difficult to draw any conclusions from these results at the moment . The effects of T . trichiura treatment on improving symptoms of colitis occurred despite the absence of a chronic and active infection . Ova were never detected in fecal samples from any of the macaques , indicating the lack of a patent infection , and adult worms were not seen during the post-treatment colonoscopy . The lack of patency is most likely due to a species barrier , since the T . trichiura ova used for inoculation were from a human subject [12] . Notably , this parallels the use of T . suis ova ( TSO ) as a therapeutic intervention in human subjects with autoimmune diseases , which is dependent on a species barrier to eliminate the parasites after dosing . After hatching , Trichuris larvae molt several times in the intestines before maturing into adult worms . Thus , the larval stages could have elicited a mucosal response leading to a positive clinical outcome . The absence of egg production does not preclude the presence of adult forms or mature larval forms , although none were observed during the endoscopies . Despite the absence of adult forms in humans , TSO clearly induces a strong TH2 response in treated subjects , as indicated by increased serum IL-4 , eosinophilia , and IgE antibodies [40] , [41] . Although we did not detect more IL-4-producing CD4+ T cells by intracellular cytokine staining in post-treatment PBMCs , this is not indicative that a systemic Th2 response was absent in the macaques . In our previous study [12] , responses in PBMCs could only be detected after antigen-specific expansion of CD4+ cells with T . trichiura antigen . We propose that the treatment of colitis ( but not necessarily other autoimmune diseases ) with Trichuris sp . may be more dependent on a TH2 response than immunoregulation . Since we did not detect ova in the feces of the treated macaques , it is also possible that the clinical improvements were spontaneous and not related to T . trichiura treatment . However , we find this unlikely since the TH2 response in the intestinal mucosa at a cellular and molecular level is remarkably consistent among all of the treated macaques . It is important to note the changes in absolute bacterial quantities as well as the composition of mucosal microbiota following helminth exposure . With both measures , post-treatment samples were much more similar to samples from healthy controls . Perhaps the most consistent observation made in studies on the microbiota of IBD patients is the overall reduction in microbial diversity [23] . There is also reduced diversity of the intestinal microbiota in rhesus macaques with colitis , as noted previously [25] as well as in this study , supporting the use of rhesus macaques with ICD as a preclinical model for IBD . We noted a striking expansion of Cyanobacteria in three ICD subjects , which disappeared post-treatment . Cyanobacteria have been previously noted to reside in the intestinal tract of mice and humans [42] as well as that of macaques [25] . However , there have not been any previous reports , including studies on the mucosal microbiota from IBD patients [26] , where this taxon has represented an abundance of up to 32 . 9% of 16S sequences . This expansion may reflect a unique environment on the intestinal mucosa of juvenile rhesus monkeys with chronic diarrhea . Notably , a recent publication ( including data from the Human Microbiome Project ) indicated that , in healthy human beings , the taxon Cyanobacteria∶Chloroplast∶Streptophyta was enriched in non-mucosal sites relative to mucosal body sites and also enriched under conditions with high oxygen availability [32] . Like Salmonella typhimurium [43] , perhaps the activation of neutrophils to produce reactive oxygen species favors the expansion of a Cyanobacterium more adapted to an aerobic environment or to alterations in redox conditions . It is also quite possible that these sequences are merely a representation of plant organisms in poorly digested food , which could be more abundantly attached to the mucosa of monkeys suffering from diarrhea . Since Bacteroidetes is the most predominant phylum in the normal gut , the reduction in proportional abundance in ICD subjects may reflect an expansion of non-Bacteroidetes phyla amongst the mucosal microbiota during ICD . Indeed , quantification of absolute Bacteroidetes abundance by RT-PCR clearly shows a reduction post-treatment . These results reflect a typical problem for this type of 16S compositional data of microbial communities . What is more interesting is the unexpected expansion of Tenericutes after helminth exposure , even relative to control macaques , indicating that bacteria of this phylum could be preferentially expanded in a TH2-type mucosal environment . Little is known about the role of Tenericutes in the intestinal tract . In one previous study with mice , dextran sodium sulfate ( DSS ) -driven colitis was shown to decrease the abundance of this phylum [44] . Further studies on the relationship between intestinal helminths and Tenericutes are warranted . ICD-afflicted juvenile rhesus macaques may represent a useful preclinical model in which to further study the pathology , diagnosis , and treatment of UC in humans under certain circumstances . Most notably , the transcriptional signature of the inflamed mucosa in these macaques closely resembles that of UC patients . Given the prevalence of ICD at primate research centers , this could be an important resource to develop new therapeutics for IBD in humans , especially when addressing the need of a non-rodent model during preclinical testing . However , the costs and ethical issues of using non-human primates may limit the usefulness of this animal model for IBD research towards very specific circumstances when an outbred non-rodent model is required . While very similar , there are also distinct differences between the macaque immune system and the human immune system [45] , [46] that may limit the use of ICD afflicted rhesus macaques as a model for IBD research . Our findings describe the immune mechanisms that may mediate the therapeutic effect of Trichuris sp . in the setting of colitis and highlight the role of TH2-type immunity in promoting mucosal repair . Instead of acting through an immunoregulatory mechanism , Trichuris sp . may trigger a TH2-driven expulsion mechanism related to increased mucus production and turnover of epithelial cells , thereby reducing the quantity of attached bacteria among the mucosal microbiota . As part of this process , the dysbiosis observed in the mucosal environment during colitis could also revert back to a normal equilibrium .
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 use and care of all animals followed policies and guidelines established by the University of California , Davis Institutional Animal Care and Use Committee ( IACUC ) and CNPRC ( Animal Welfare Assurance #A3433-01 ) . The protocol for this trial was approved by the University of California , Davis IACUC . Veterinary care was provided to all animals in order to minimize pain and distress . The California National Primate Research Center ( CNPRC ) , which houses over 5 , 000 nonhuman primates , is a part of the National Primate Research Centers Program and is accredited by the Association for the Assessment and Accreditation of Laboratory Animal Care , International ( AAALAC ) . All animal facilities are maintained in compliance with United States Department of Agriculture specifications . Rhesus macaques are maintained in small or large social groups . All animals were housed indoors at the CNPRC . ICD cases were identified by recurrent episodes of diarrhea ( during 45 or more days in a 90 day period ) without evidence for ( or a history of ) known causes of infectious colitis , as demonstrated by three negative cultures for bacterial pathogens and negative stool examination and immunofluorescence assays for protozoan and helminthic parasites . The diarrhea was refractory to antibiotic and antiparasitic treatment . Animals were weighed every 8–14 days and their stool evaluated according to a standardized four-point scale for fecal consistency ( Figure 1 ) . Animals were fasted for 36 hours prior to colonoscopy and 30 mL/kg of polyethylene glycol-electrolyte solution ( PEG-ES; GoLYTELY brand ) was provided twice the day before the procedure by mixing 67 gm with 1 liter of citrus-flavored water ( Tang , Kraft Foods , Northfield , IL ) and making the solution available for the animals to drink by a hanging bottle in the cage . Blood was collected by venipuncture into citrate tubes prior to colonoscopy . Five pinch biopsies were collected during colonoscopy from the proximal ascending colon . Three biopsies were collected into culture media for ex vivo analysis and two biopsies were collected into RNAlater ( Qiagen ) for nucleic acid extraction . Colon biopsy specimens were treated with 0 . 25 mg/ml collagenase type II ( Sigma-Aldrich ) for 30 minutes with constant shaking at room temperature . Digested tissue was dispersed over a 70-micron nylon mesh filter . Cell suspensions were washed twice with RPMI containing 15% fetal calf serum . Whole blood was collected into ACD-containing tubes ( BD Biosciences ) and PBMC were isolated by density centrifugation . Biopsy cells and PBMC ( 1×106 ) were resuspended in 200 µl of complete R-10 [RPMI 1640 medium ( Invitrogen ) supplemented with 10% fetal calf serum ( Hyclone ) , 50 U/ml penicillin , 50 µg/ml streptomycin , and 2 mM L-glutamine] , and stimulated with phorbol myristate acetate ( 10 ng/ml ) and ionomycin ( 1 µg/ml ) in the presence of brefeldin A ( GolgiPlug , BD Pharmingen ) for five hours at 37°C . For biopsy cells , amphotericin B ( Gibco ) was also added to the culture media . Cell surface staining and intracellular cytokine staining were performed with Fix/Perm and PermWash solutions from BD and eBioscience , according to the manufacturer's instructions . Biopsies were collected into RNAlater ( Qiagen ) and homogenized in TRIzol ( Invitrogen ) . RNA was collected in the aqueous extraction phase and column purified using an RNeasy kit ( Qiagen ) . Sample preparation , labeling , and array hybridizations were performed according to standard protocols from the UCSF Shared Microarray Core Facilities and Agilent Technologies ( http://www . arrays . ucsf . edu and http://www . agilent . com ) . Total RNA quality was assessed using a Pico Chip on an Agilent 2100 Bioanalyzer ( Agilent Technologies , Palo Alto , CA ) . RNA was amplified and labeled with Cy3-CTP using the Agilent low RNA input fluorescent linear amplification kits following the manufacturer's protocol ( Agilent ) . Labeled cRNA was assessed using the Nandrop ND-100 ( Nanodrop Technologies , Inc . , Wilmington DE ) and equal amounts of Cy3 labeled target were hybridized to Agilent Rhesus Macaque ( V2 ) whole genome 4x44K Ink-jet arrays ( Agilent ) . Hybridizations were performed for 14 hours , according to the manufacturer's protocol ( Agilent ) . Arrays were scanned using the Agilent microarray scanner ( Agilent ) and raw signal intensities were extracted with Feature Extraction v10 . 1 software ( Agilent ) . Each data set was normalized using the quantile normalization method [47] with no background subtraction . A one-way ANOVA linear model was fit to the comparison to estimate the mean M values and calculated B statistic , false discovery rate and p-value for each gene for the comparison of interest . All procedures were carried out using functions in the R package . For DAVID [48] and PANTHER [49] pathway analyses , EnsEMBL and Genbank accession numbers associated with significantly different M . mulatta genes were converted to official gene symbols , along with a background list consisting of all genes used as input to the analyses . Functional annotation charts were generated using the default parameters . Reported p-values are Bonferroni-corrected . For these analyses , the human pathways and ontologies were used as they are more fully developed than those available for macaques . Tissue samples were homogenized in TRIzol ( Invitrogen ) . RNA was collected in the aqueous extraction phase , and DNA was harvested from the interphase and phenol-chloroform organic phase . RNA was column purified using an RNeasy kit ( Qiagen ) . cDNA was generated using an Omniscript Reverse Transcription kit ( Qiagen ) with oligo-dT primers in the presence of RNasin Plus RNase inhibitor ( Promega ) . DNA was collected by ethanol precipitation and washed according to the manufacturer's instructions . PCR reactions were carried out with Taqman primer/probe sets ( Applied Biosystems ) in a StepOne Plus machine ( Applied Biosystems ) . The abundances of major intestinal bacterial groups were measured by qPCR of extracted DNA , using the MyIQ single-color real-time PCR detection system ( Bio-Rad , Hercules , CA ) and group-specific 16S rRNA gene primers that have been previously published and verified [50] . Samples were normalized to controls and eubacteria . Plasma sCD14 levels were measured using the Quantikine ELISA kit ( R&D Systems; Minneapolis , Minnesota , USA ) , as described in the manufacturer's protocol . Plasma samples were diluted 1/500 and run in duplicate . DNA samples isolated from pinch biopsies were PCR amplified for sequencing through a validated protocol [31] . The V4 region of the 16S rRNA gene was amplified with region specific barcoded primers [51] and sequenced on a MiSeq sequencer [31] along with other barcoded samples . Reads shorter than 140 bp were discarded . The QIIME suite of analysis tools was used to filter and analyze the sequence data [52] . Sequences were assigned to OTUs with a threshold of 97% pair-wise identity and then classified taxonomically using the Ribosomal Database Project ( RDP ) classifier . A representative sequence for each OTU was aligned using PyNAST and used to build a phylogenetic tree for α-diversity and β-diversity measurements . Alpha rarefaction was performed using the Shannon index and phylogenetic diversity . Beta diversity was calculated in QIIME , using default metrics of weighted and unweighted UniFrac distances between samples , and visualized using Principal Coordinate Analysis ( PCoA ) plots and Unweighted Pair Group Method with Arithmetic Mean ( UPGMA ) trees . Relative abundance of microbial phyla was determined in QIIME by grouping OTUs by macaque sample . To identify significant differences in bacterial taxa between groups , we utilized the linear discriminant analysis ( LDA ) effect size ( LEfSe ) algorithm [32] through the Galaxy Framework [53] , [54] online . LEfSe uses the Kruskal-Wallis ( KW ) sum-rank test to detect features with significantly different abundances between classes , and performs LDA to estimate the effect size of each differentially abundant feature . So that we could recover all features detected by the KW test , we did not set a threshold for LDA in the analysis . | Young macaques kept in captivity at Primate Research Centers often develop chronic diarrhea , which is difficult to treat because it is poorly understood . This disease shares many features with ulcerative colitis , which is an autoimmune disease affecting the intestinal tract of humans . Recently , parasitic worms have been used in clinical trials to treat inflammatory bowel diseases in humans with positive results , but very little is known about how worms can improve symptoms . We performed a trial where we treated macaques suffering from chronic diarrhea with human whipworms , collecting gut biopsies before and after treatment . We found that 4 out of the 5 treated macaques improved their symptoms and studied the changes in their gut immune responses , as they got better . We found that after treatment with worms , the monkeys had less bacteria attached to their intestinal wall and a reduced inflammatory response to the gut bacteria . Additionally , the composition of gut bacteria was altered in the sick macaques and was restored close to normal after treatment with whipworms . These results provide a potential mechanism by which parasitic worms may improve the symptoms of intestinal inflammation , by reducing the immune response against intestinal bacteria . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"medicine",
"pathology",
"immunology",
"microbiology",
"host-pathogen",
"interaction",
"gastroenterology",
"and",
"hepatology",
"veterinary",
"science",
"infectious",
"diseases",
"veterinary",
"diseases",
"veterinary",
"parasitology",
"veterinary",
"microbiology",
"biology",
"microbial",
"ecology",
"immune",
"response",
"veterinary",
"pathology"
] | 2012 | Therapeutic Helminth Infection of Macaques with Idiopathic Chronic Diarrhea Alters the Inflammatory Signature and Mucosal Microbiota of the Colon |
The three-dimensional structures of proteins are stabilized by the interactions between amino acid residues . Here we report a method where four distances are calculated between any two side chains to provide an exact spatial definition of their bonds . The data were binned into a four-dimensional grid and compared to a random model , from which the preference for specific four-distances was calculated . A clear relation between the quality of the experimental data and the tightness of the distance distribution was observed , with crystal structure data providing far tighter distance distributions than NMR data . Since the four-distance data have higher information content than classical bond descriptions , we were able to identify many unique inter-residue features not found previously in proteins . For example , we found that the side chains of Arg , Glu , Val and Leu are not symmetrical in respect to the interactions of their head groups . The described method may be developed into a function , which computationally models accurately protein structures .
Most biological activities of the living cell are directed or regulated by proteins . These diverse functions are due to proteins three-dimensional structures consequent of the physical interactions of their amino acid residues [1] . As the backbone of all amino acids is identical , side chains ( and associated cofactors ) dictate structure . Therefore , the knowledge of how side chains interact with each other and with the backbone will enable the computational prediction of protein structures , and the design of their shapes and functions [1] . Computational methods have been used in several research fields for the assessment and prediction of protein structures [2]–[8] such as: fold recognition , threading [9] , binding [10]–[19] , de novo design [20]–[24] and the prediction of protein stability [25]–[27] . There are two main issues to be considered while computationally modeling protein structure , one is the conformational search and the other is the scoring function . Currently rigorous functions to describe the physical interaction between molecules are computationally demanding [20] , [28] . Therefore , three types of approximation are currently used: physical-based , empirical or a combination of both methods . The first is based on the fundamental analysis of forces between atoms [12] , [29]–[31] . The second is a knowledge-based scheme that provides a shortcut by assuming predictable [32] , [33] , though theoretically questionable , potentials derived from databases of protein structures and sequences . The third method is a hybrid that combines the two schemes [34] . The majority of scoring functions are a sum of pair-wise interactions , which are assumed to be independent . Yet this approximation was proven to be inaccurate both computationally [35] and experimentally [36] . The cooperativity of the residue contacts have been modeled partially by three and four-body interaction and by modeling protein local environment with no significant advantage compared to the pair-wise methods [37]–[41] . The key concept of the knowledge-based potential ( KBP ) is collecting features from protein structure databases relative to random predictions [42]–[44] . Statistical potentials can be categorized on the basis of different aspects: residue-level potentials [9] versus atomic-level potentials [2] , [29] , [45] , [46] . Examples of knowledge-based features used include: solvent accessibility [47] , local environment [40] , [41] , atom contact area , sequence fragments [34] , bond angle [24] , [48] and distance [2] , [4] , [8] , [14] , [26] , [42] , [45] , [49]–[51] ( which is the most abundantly used ) . A good random model is crucial for the success of this method as being a reference state . There remains , however , a lack of consensus on how to define random models for KBP [13] , [26] , [28] , [52] . One method is to shuffle the residues or atoms in the database and then recollect the data [46] , [53] . A different scheme is the distance-scaled finite ideal gas reference state , which assumes the spatial distribution in the reference state that should be scaled as power distance ( i . e . rα ) [26] . The expected number of atom pairs in a given distance shell is proportional to that observed in the database regardless of the atom type . A few issues have been raised criticizing statistical potentials , including the argument that the topology of the proteins included in the database “remembers” the database it was derived from [54] , the Boltzmann distribution assumption [32] and the reproducibility of the function using the knowledge-based method [33] . Knowledge-based distance potentials vary in the representation of the amino acid , which can be at the residue level or at the atom level . At the residue level the side chain is represented by one object; this simple model is used to reduce the computational cost , though , it was empirically shown that such simplifications reduce the accuracy of the resulting statistical potentials [2] , [55]–[57] . Examples include residue centroid [45] , [58] ( center of mass ) , where a pseudo atom is calculated for each residue . Another representation for the side chain is the “volume block” for parts of the amino acid molecule [59] . In the case of atom-level representation for high-resolution models , a representative atom [60] ( i . e . CA or CB ) or several atom sets for each residue can be selected [61] . The collected distances are binned to a distance distribution , and the prediction power of the model is improved in accordance with the bin resolutions [5] . Here we report a new high-resolution distance method for precise description of the residue interaction geometry . We show the disadvantage of considering similar atoms as one group . Several observations regarding the reference model are described . Finally , we demonstrate that our four-distance description has predictive power on residue contact geometry .
High-resolution protein structures ( ≤2 Å ) were taken from the PISCES server [62] . A 90% non-redundant dataset comprising of 6830 structures was chosen . The low-resolution protein structures ( resolution 2 . 5 Å–3 . 0 Å ) were also taken from the PISCES server . Non-redundant NMR protein structures with 60% identity were taken from the OCA server ( 1877 structures ) [63] . To increase the number of distance measurements , two NMR models were chosen randomly for each structure . An interaction between two residues is defined in terms of four distances between two pairs of atoms ( Figure 1 ) . The following notation “R1_R2_A1_A2_B1_B2” is used throughout this study , where R1 and R2 are the interacting residues , A1 and A2 is the pair of atoms in residue R1 , B1 and B2 is the pair of atoms in residue R2 . For example , Lys_Val_CB_CD_CG1_CG2 means that for the interacting residue pair Lys and Val the representative atoms of Lys are CB and CD and the representative atoms of Val are CG1 and CG2 . To choose the pair of atoms for each residue pair , all the possible combinations were enumerated . The four Euclidean distances were collected for all of them and a four-distance set was kept only if one of the distances was <5 Å . This cutoff enabled us to discard the huge amount of data comprising distanced non-interacting pairs . A representative four-atom set was chosen for each residue pair . The criterion for choosing this representation was to maximize the chances that at least one interactive distance of less than 5 Å will be represented for each amino acid pair ( Table S1 ) . This allows the possibility that the same amino acid is in contact with different partners via different atom pairs . For example , the Lys–Val contact was defined as Lys_Val_CB_CD_CG1_CG2 while the contact of Lys with Asp was defined as , Lys_Asp_CD_CE_OD1_OD2 . Side chain backbone contacts were grouped regardless of the identity of the backbone amino acid . The representative backbone atoms are the alpha carbon and the carboxyl oxygen . For example , for Arg_Any_CG_NH2_CA_O side chain of Arg can be in contact with backbone atoms from any residue . The random model was generated from the high-resolution dataset ( see above ) . Each residue conformation was replaced randomly by a rotamer from the rotamer library [64] , keeping the same amino acid identity . No considerations for rotamer probabilities or conformational clashes were taken . The distances were collected as for the real data . However , to reduce the noise level of the random data , a 25-fold excess of random to real data was used . The four distance data were arranged in a 4-D histogram . Each contact pair was binned at a resolution of 0 . 5 Å , from 0 to 10 Å forming histograms of 214 bins ( i . e . 194481 ) . The probability ( frequency normalized to one ) for each four-distance combination was the number of measurements in a selected bin divided by total number of measurements . To test the suitability of the derived data for modeling residue-residue interactions a simulation was performed . Two residues with arbitrary side chain conformations were built . Then , these residues were randomly placed relative to each other 106 times . Four distances between atoms defining the residue-residue interaction were calculated for each random placement and a score was assigned in accordance with Equation 2 . The orientation of two residues , for which the best score was attained was identified and examined .
The 4-D description of residue-residue interactions is a more restricted form of the single distance potential extensively used . This higher order description allowed us to challenge common beliefs of the symmetric nature of some of the amino acids . For example , the Arg nitrogen atoms NH1 and NH2 were usually considered to be exchangeable with one another [55] . The notation of atoms for Arg in X-ray structures of proteins is as shown in Figure 1: in the guanidinium plane NH2 is trans to CD [65] . In Figure 2 we present the distances between the Arg–Asp atom pair as collected from the PDB . Comparing the two distance peaks of NH1–OD1 ( d11 ) and NH2–OD1 ( d21 ) ( Figure 2B ) shows a non-equal distribution of the two populations: d21 at 3 Å and d11 at 5 Å is much more populated than d21 at 5 Å and d11 at 3 Å . This observation shows that the guanidinium group in the Arg residue is not symmetric [55] , which is a result of the notation of Arg in the PDB: in this conformation NE can serve as a hydrogen bond donor ( compare Figure 2A to 2B ) . In the conformation where d21 is 5 Å and d11 is 3 Å the CD atom sterically interferes in forming the second hydrogen bond of an incoming residue with NE . Figure 2E gives the interaction distances for the atom pairs NH1–OD2 and NH2–OD2 , showing that OD1 and OD2 are symmetric while NH1 and NH2 are not . It is interesting to note that it is rare to find OD1 and OD2 being located at an equal distance from either NH1 or NH2 ( Figure 2B , 2E and 2F ) . In fact , one distance is 3 Å while the other is 3 . 6 Å . This means that at a given Asp–Arg interaction only one of the OD atoms forms a hydrogen bond with the NH group of Arg , while the second distance is beyond the hydrogen bond threshold and may be of an electrostatic nature . To the contrary , the distances of both NH1 and NH2 with either OD1 or OD2 peak at 3 Å ( Figure 2C and 2D ) . Another striking example of an amino acid , which is considered as symmetric , is Val . Defining the atom notation as an optical enantiomer with CB acting as a pseudo chiral center and the hydrogen pointing away from the viewer , then going clockwise from CA the two methyl groups are always CG1 and next CG2 [65] . In Figure 3A one can see that the shorter d21 distance ( CG2–CG1 ) of 4 Å is preferred over the d11 distance ( CG1–CG1 ) of 6 Å . Figure 3B ( panels 1 to 4 ) depicts the possible orientations where d11 and d21 are 4 Å or 6 Å . The observation that the B1 , B3 conformations are preferred over the B2 , B4 conformations require an explanation . Figures 3C and 3D suggest that CG1 is generally closer to the backbone oxygen ( negative partial charge ) while CG2 is closer to the backbone nitrogen ( positive partial charge ) . In the B1 conformation there are three short bonding distances , two with opposite partial charges and one with a similar charge . Conversely , in B2 there are two distances with a similar partial charge , and one with an opposite partial charges . For the same reason , B3 is preferred over B4 . The definition we propose in the 4 dimensional matrices exploits this asymmetry because the two atoms are treated separately . The two examples brought here demonstrate how defining interactions as a 4-D matrix provides high-resolution structural insight . Additional examples of the asymmetry of the same atom types behaving differently in different contexts include His , Ile , Phe , Glu , Leu and Pro ( Figure 4 , panels A–F ) . We have noticed different distance distributions for the two nitrogen atoms of His . The NE2–NE2 distance has a higher occupancy at 3 . 5 Å ( which is a good hydrogen bond distance ) compared to ND1–ND1 ( Figure 4A ) . For Ile , the CD1–CG2 distance is more occupied at 4 Å compared to the CG2–CG2 distance ( Figure 4B ) . The origin for this discrepancy might be the better availability of the CD1 and NE1 atoms over CG and CD1 for the Ile and His residues respectively , as being farther away from the backbone on the residue side chain . In the Phe 4-D distribution , shorter d12 distance ( CE1–CE2 ) of 4 Å is preferred over the d11 distance ( CE1–CE1 ) of 6 Å ( Figure 4C ) . This example holds also for Tyr . In the histogram of the Glu–Pro contacts it seems that OE1 is preferred over OE2 bonding to the Pro CD atom . Phe and Glu are similar residues in the sense of having a symmetrical side chain , the atom names were defined by the second chi angle [65] . As can be seen in the inset of Figure 4C , the atom that makes the smaller chi angle is assigned number 1 [65] ( in the figure it is CD1 ) . The higher preference for interaction of Phe CE2 atom is a result of the higher exposure of this atom to the solvent , and thus to an incoming bond . Conversely , the less exposed atom of Glu , namely OE1 , is preferred over OE2 , though we have no good explanation for this phenomenon . In the case of Leu , the atom notation is similar to Val . We have noticed that the CD1–CD1 distance is more occupied at 4 Å compared to CD2–CD2 distance ( Figure 4E ) . It is more difficult to explain this situation , since both CD1 and CD2 are closer to the backbone oxygen than to the backbone nitrogen ( data not shown ) . The mean distance values for the CD1 , CD2 atoms to the backbone oxygen are 4 . 36 Å and 4 . 15 Å respectively , and to the backbone nitrogen 5 . 03 Å and 4 . 87 Å . Favoring the contact of CD1–CD1 might be a result of an unfavorable close contact between the two backbone oxygens . The last example we report is Pro , the atom notation of this residue is trivial . We have noticed different distance disruptions for two Pro carbon atoms CD and CB . In the distribution of Asn and Pro , the OD1–CD distance is more occupied compared to the OD1–CB distance at 3 . 5 Å ( Figure 4F ) . The origin for this discrepancy might be the proximity of CD to the backbone nitrogen . Similar results were seen for Asp , Glu and Gln residues in contact with Pro . X-ray crystallography and NMR are the two main methods for protein structure determination at high resolution; in both , computational energy minimization is required at various stages of structure calculation . Since our distance potential was originated from high-resolution data , and the atom positions are highly restricted due to the 4-D data description of each residue pair , we argue that our new potential provides a realistic view of side chain orientations that can be used to evaluate computational minimization techniques . Figure 5 shows the NH1–OD1/NH2–OD2 distances for the Arg–Asp pair , based on distances collected from structures obtained from either high or low-resolution X-ray data , or from NMR . The high-resolution X-ray data gave the sharpest peaks , followed by the low-resolution data . Additionally we collected high-resolution X-ray data where the diffractions were collected at a temperature of 275–300 Kelvin . The 4-D data collected from NMR structures did not show any clear peaks , and resembled the distribution obtained from random data . It is argued that the proteins in NMR solutions are more flexible since the data is collected at room temperature . However , this should not affect the contact geometries , moreover , it seem that high resolution data collected at higher temperature do not change the distributions significantly ( compare Figure 5B and 5C ) . What distinguishes high resolution X-ray data from low-resolution X-ray data and more so from NMR is the extent of minimization methods dictating the structure . This would suggest that the current minimization methods do not produce the “real” inter-residue contacts as provided by high-resolution X-ray structures . In the case of high-resolution structures the atom positions can be deduced more precisely from the electron density , which reduces the need for inaccurate minimization protocols [5] . Our initial aim in acquiring the 4-D distributions was to generate a knowledge-based potential . The standard KBP is built using Equation 1 and 2 . Equation 1 is the conditional probability of a distance set in case where the amino acid sequence is known . The probability is defined as the product of the data collected ( the probability of the four distances for a given amino acid pair ) and the probability of an amino acid pair in the protein . Since the energy is estimated as the ratio of Preal and Prand ( Equation 2 ) and the real and random probabilities were generated from the same data set , the probability of the amino acid pair term cancels out . Equation 2 is an inverse Boltzmann relation [42] , where ΔE is a pseudo energy gap obtained from the log ratio of the real and random distributions . As we do not attempt to predict experimental results , we defined kBT as unity . To calculate Prand we used Equation 1 on the set of randomized structures as described in Methods . Our random model corresponds to the situation , in which conformations of side chains are not dictated by forces characteristic to real proteins . The random model would correspond to the starting point in the conformational search , in which residue side chain conformations are assigned arbitrarily . Thus this model is designed to maximize the difference between real interactions and the initial state of the system . According to Equation 2 , three possible relations are found between the real and random distributions; in the case of a favorable interaction , Preal is higher than Prand , for an unfavorable interaction the opposite is observed . The third possibility is that both distributions are equal ( neutral conformation ) with a pseudo energy score of zero . The random distribution acts as a reference that distinguishes the preferred conformational states in the PDB database . ( 1 ) ( 2 ) A general phenomenon , which is not exclusive to the 4-D data , stems from the normalization of both the real and random distributions to one ( or the same number of events are used for both sets ) . As bond distances in the real distribution peak at short distances ( which reflect the bound state ) , while the random distribution of inter-residue distances will increase monotonically ( Figure 6A ) , the curves of the two distributions have to cross each other at larger distances ( as the integral of both curves equals to one ) . This causes positive pseudo energy ( i . e . unfavorable conformation according to Equation 2 ) at larger distances , while no real interaction is expected ( namely zero energy [13] ) . We observed the same unphysical pseudo energy gap values even when generating a random model accounting for both rotamer probabilities and atom clashes ( data not shown ) . Because the number of counted interactions is growing with the square of the distance , even a small repulsive term at extended distances will have a large effect on the total energy calculated . We devised two solutions for this problem; the first was to raise the distance cutoff of the collected data to 30 Å , resulting in a much-reduced negative energy term ( see Figure 6B versus 6A ) . This is better seen in Figures 6C and 6D , where the log ratio of the real/random is drawn , showing values much closer to zero for the raised cutoff . Secondly , we defined a distance cutoff for plausible interactions at 5 Å ( at least one of the four distances has to be <5 Å for it to be counted ) . This distance cutoff removes all residue pairs with no direct interaction between them . Binning data at intervals of 0 . 5 Å gives a total number of 194481 bins up to a residue-residue distance of 10 Å . However , the number of “real” counts as extracted from the database was only between 1836 and 108094 for the least ( Cys–Glu ) and most frequent ( Leu–Leu ) amino acid pairs respectively , with a mean value of 108094 counts for all 190 interactions ( see Table S1 for the entire dataset ) . Thus , most bins were actually empty , or were occupied by a very small number of events . This is demonstrated in Figure 7 , where the amount of data is defined as the product of bin count times the number of bins with that count . For example , if N bins are occupied each by 5 events , the total number of events in this group is 5 N . To determine the distribution of events in different bins , the data was normalized by the total number of measurements in the histogram . The plot in Figure 7 shows the normalized data ( y-axis ) in bins with a particular number of counts ( x-axis ) for real and random distributions . Comparing the real and random bin occupancy distributions clearly shows that the random is dominated by low bin occupancies , while high counts are reserved for the real data . For low number of events per bin , the real and random data overlap . For example , the average count of events per bin for the Arg–Asp pair is 3 . 2±9 , with some bins having up to 431 counts , whereas in the random data no bin has more than 19 counts . This suggests that the data generated from the bins with low occupancy are more prone to error ( for example , if a bin of real data has 5 events , and the random has 2 , the two occupancies are within the error of one another ) . The low number of average bin counts is a major issue of the 4-D method . To evaluate whether the derived 4-D data can be used for modeling , a simple simulation was performed ( see Methods for details ) . Figure 8 shows the 4-D histograms of an optimal side chain placement for Arg and Asp according to Equation 2 . It is important to mention that the carboxyl group of Asp is always placed close to the NH2 atom of Arg . The same geometry for this contact is observed in high-resolution protein structures as was discussed above . The ability to reproduce the geometry of side chain/side chain contacts as observed in the PDB demonstrates that constructed 4-D histograms might be a useful tool for accurate protein structure modeling .
Knowledge-based potentials have become very popular and successful in recent years . For a KBP to be general , each term in it should have the maximal information content , and the number of terms should be minimized . Too many terms will make the energy landscape too ragged and cause overcounting . For example , a function that contains both a van der Waals term and an environment term may overcount the London dispersion force . We argue that the 4-D description introduced here has higher information content than the standard 1D distance usually used in KBPs , therefore providing a better definition for the relations ( forces ) between residues . The downside of a more exact , multi dimensional description of a structure is the ruggedness of the resulting energy surface , which cannot be exhaustively probed using discrete rotamers . Moreover , to properly define all the interactions requires an amount of data currently not available . Therefore , implementing the four-distance description for sampling requires reducing the potential ruggedness , which can be achieved by smoothening the data ( Potapov , V . , Cohen , M . & Schreiber , G . ; unpublished data , 2009 ) . However , for scoring , the ruggedness is of lesser importance , thus the original histograms can be used . The high information content of amino acid interactions was demonstrated by the observation that many such interactions are asymmetric , a surprising fact by itself . The asymmetry involved interactions of Arg , Glu , Val , Phe , His , Leu , Pro and Ile . To verify that the asymmetry is not a result of inconsistent naming we verified that the atom names were assigned according to the IUPAC definitions [65] . For seven residues , in which branches are identical , we found that substantial amount of atom name assignments in the PDB do not follow the conventions; 24% of the cases for Phe , 23% for Tyr , 18% for Glu , 15% for Asp , 8% for Arg , 0 . 8% for Leu , and 0 . 01% for Val . We could not attribute this to particular structural refinement software or the date of deposition . This kind of asymmetry was not detected previously , using 1D data or physical forcefields . However , the asymmetry was also not detected in KBP , which use information on bond lengths , bond angles , and dihedral angles for pairs , triplets , and quandruplets of bonded atoms . The reason for this may be the lower information content of other KBPs . A recent forcefield developed by Ma et al . [59] is actually a 3D potential where the geometry between the volume blocks is defined by two angles and a distance between two planes . This method also cannot detect the asymmetry of the residue atoms since it is a type of united atom representation . For example , NH1 and NH2 belong to the same volume block . The reasoning behind the asymmetry given here is mostly intuitive . We do not argue that there is a chemical difference between different residue atoms . Though we argue that a KBP that is based on the PDB must maximize the information content extracted from it , thus , this asymmetry must be taken into account . The advantage of KBP over physical based potentials is the fact that unknown factors can emerge from the constructed potentials , with no actual known physical explanation . One example for such is the hydrogen bond geometry reported by Kortemme et al . [66] . The physical model predicted the angle between the hydrogen acceptor and the atom covalently bound to the acceptor to be 180° , while in the PDB most of these angles are closer to 120° . A KBP taking advantage of the full extent of information in the PDB database may be better in modeling protein structures . More exact calculations in the future will be needed to produce a more satisfying explanation for the observed asymmetries . However , the advantage of KBPs over physical forcefields is that one can use observations to model protein structures , even if they are not fully understood . Since most side chains contain more than two atoms , we had to decide which of the atoms to use to generate the 4-D data . For example the structurally simple Val-Val pair has three different atom pairs CB–CG1 , CB–CG2 and CG1–CG2 . Thus , 6 different 4 atoms distances could be generated . More generally , the 202 different atom pairs on the 19 amino acids ( excluding Gly ) can be paired using 20503 different 4-atom distributions . To choose the 190 representative pairs used for the 4-D database ( one per residue pair ) we considered three different approaches; first , manually choosing the pair that best represents the residue ( e . g . CG1–CG2 for Val or OD1–ND2 of Asn ) . The second approach was to maximize the Kullback–Leibler divergence between the real and the corresponding random distributions . Since the Kullback–Leibler divergence measures the difference between two probabilities the pair with the highest difference may have the highest information content . The third approach was to search for the pair with the highest number of accepted distance measurements ( at least one distance less than 5 Å ) . The problem of limited data is clearly demonstrated in Figure 6 , that shows that most bins are empty , or contain only low bin counts that are statistically within the numbers found when using random side chain rotamers . This is a main problem with a KBP , as the division between two small numbers ( random and real bin occupancy ) still generates a value different from 1 , and thus assigns an energy term to that bin . To minimize this problem , we currently use the most frequent atom pairs as described above . Using this logic to choose atom pairs also seems to produce the best results when using the 4-D matrix for side chain modeling ( Potapov , V . , Cohen , M . & Schreiber , G . ; unpublished data , 2009 ) . This may be the result of the limited amount of data in the PDB , and thus may change once many more structures will be available . | Knowledge of high-resolution structures of proteins is an invaluable source of information for molecular biologists . Since obtaining structures experimentally is a laborious process , using computational methods to model correctly protein structure is highly beneficial . As protein structures are stabilized by the specific contacts formed by amino acid side chains , detailed understanding of inter-residue interaction is essential . Here we report a novel concept to analyze contact geometry , in which four distances are calculated between any two residues to describe their interactions in great detail . It allowed us to extract information from the Protein Data Bank , which until now was overlooked . This concept can be used to develop a computational method to accurately model protein structures . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"computational",
"biology/protein",
"structure",
"prediction",
"computational",
"biology/macromolecular",
"structure",
"analysis",
"computational",
"biology"
] | 2009 | Four Distances between Pairs of Amino Acids Provide a Precise Description of their Interaction |
Humans can meaningfully report their confidence in a perceptual or cognitive decision . It is widely believed that these reports reflect the Bayesian probability that the decision is correct , but this hypothesis has not been rigorously tested against non-Bayesian alternatives . We use two perceptual categorization tasks in which Bayesian confidence reporting requires subjects to take sensory uncertainty into account in a specific way . We find that subjects do take sensory uncertainty into account when reporting confidence , suggesting that brain areas involved in reporting confidence can access low-level representations of sensory uncertainty , a prerequisite of Bayesian inference . However , behavior is not fully consistent with the Bayesian hypothesis and is better described by simple heuristic models that use uncertainty in a non-Bayesian way . Both conclusions are robust to changes in the uncertainty manipulation , task , response modality , model comparison metric , and additional flexibility in the Bayesian model . Our results suggest that adhering to a rational account of confidence behavior may require incorporating implementational constraints .
People often have a sense of a level of confidence about their decisions . Such a “feeling of knowing” [1 , 2] may serve to improve performance in subsequent decisions [3] , learning [1] , and group decision-making [4] . Much recent work has focused on identifying brain regions and neural mechanisms responsible for the computation of confidence in humans [5–7] , nonhuman primates [8–10] , and rodents [11] . In the search for the neural correlates of confidence , the leading premise has been that confidence is Bayesian , i . e . , the observer’s estimated probability that a choice is correct [1 , 12–14] . In human studies , however , naïve subjects can give a meaningful answer when you ask them to rate their confidence about a decision [15]; thus , “confidence” intrinsically means something to people , and it is not a foregone conclusion that this intrinsic sense corresponds to the Bayesian definition . Therefore , we regard the above “definition” as a testable hypothesis about the way the brain computes explicit confidence reports; we use Bayesian decision theory to formalize this hypothesis . Bayesian decision theory provides a general and often quantitatively accurate account of perceptual decisions in a wide variety of tasks [16–18] . According to this theory , the decision-maker combines knowledge about the statistical structure of the world with the present sensory input to compute a posterior probability distribution over possible states of the world . In principle , a confidence report might be derived from the same posterior distribution; this is the hypothesis described above , which we will call the Bayesian confidence hypothesis ( BCH ) . The main goal of this paper is to test that hypothesis . Recent studies have attempted to test the BCH [19 , 20] but , because of their experimental designs , are unable to meaningfully distinguish the Bayesian model from any other model of confidence . Recent work has proposed possible qualitative signatures of Bayesian confidence [21] . However , the observation ( or lack thereof ) of these signatures provides an uncertain amount of evidence in favor of ( or against ) the Bayesian model , and the signatures are therefore not useful for determining which computations underlie confidence reports [22] . To objectively and quantitatively determine whether confidence ratings appear to be Bayesian , we use a formal model comparison approach . We test the predictions of the BCH as we vary the quality of the sensory evidence and the task structure within individuals . We compare Bayesian models against a variety of alternative models , something that is important for the epistemological standing of Bayesian claims [23 , 24] . We find that the BCH qualitatively describes human behavior but that quantitatively , even the most flexible Bayesian model is outperformed by models that take uncertainty into account in a non-Bayesian way .
During each session , each subject completed two orientation categorization tasks , Tasks A and B . On each trial , a category C was selected randomly ( both categories were equally probable ) , and a stimulus s was drawn from the corresponding stimulus distribution and displayed . The subject categorized the stimulus and simultaneously reported their confidence on a 4-point scale , with a single button press ( Fig 1a ) . Using a single button press for choice and confidence may prevent post-choice influences on the confidence judgment ( [25] , but see [26] ) and emphasized that confidence should reflect the observer’s perception rather than a preceding motor response . The categories were defined by normal distributions on orientation , which differed by task ( Fig 1b ) . In Task A , the distributions had different means ( ±μC ) and the same standard deviation ( σC ) ; leftward-tilting stimuli were more likely to be from category 1 . Variants of Task A are common in decision-making studies [27] . In Task B , the distributions had the same mean ( 0° ) and different standard deviations ( σ1 , σ2 ) ; stimuli around the horizontal were more likely to be from category 1 . Variants of Task B are less common [28–30] but have some properties of perceptual organization tasks; for example , a subject may have to detect when a stimulus belongs to a narrow category ( e . g . , in which two line segments are collinear ) that is embedded in a a broader category ( e . g . , in which two line segments are unrelated ) . Subjects were highly trained on the categories; during training , we only used highest-reliability stimuli , and we provided trial-to-trial category correctness feedback . Subjects were then tested with 6 different reliability levels , which were chosen randomly on each trial . During testing , correctness feedback was withheld to avoid the possibility that confidence simply reflects a learned mapping between stimulus orientation and reliability and the probability of being correct [30–33] . Because we are interested in subjects’ intrinsic computation of confidence , we did not instruct or incentivize them to assign probability ranges to each button ( e . g . , by using a scoring rule [34–36] ) . If we had , we would have essentially been training subjects to use a specific model of confidence . To ensure that our results were independent of stimulus type , we used two kinds of stimuli . Some subjects saw oriented drifting Gabors; for these subjects , stimulus reliability was manipulated through contrast . Other subjects saw oriented ellipses; for these subjects , stimulus reliability was manipulated through ellipse elongation ( Fig 1c ) . We found no major differences in model rankings between Gabor and ellipse subjects , therefore we will make no distinctions between the groups . For modeling purposes , we assume that the observer’s internal representation of the stimulus is a noisy measurement x , drawn from a Gaussian distribution with mean s and s . d . σ ( Fig 1d and 1e ) . In the model , σ ( i . e . , uncertainty ) is a fitted function of stimulus reliability . A Bayes-optimal observer uses knowledge of the generative model to make a decision that maximizes the probability of being correct . Here , when the measurement on a given trial is x , this strategy amounts to choosing the category C for which the posterior probability p ( C ∣ x ) is highest . This is equivalent to reporting category 1 when the log posterior ratio , d = log p ( C = 1 ∣ x ) p ( C = 2 ∣ x ) , is positive . In Task A , d is d A = 2 x μ C σ 2 + σ C 2 . Therefore , the ideal observer reports category 1 when x is positive; this is the structure of many psychophysical tasks [37] . In Task B , however , d is d B = 1 2 log σ 2 + σ 2 2 σ 2 + σ 1 2 - σ 2 2 - σ 1 2 2 ( σ 2 + σ 1 2 ) ( σ 2 + σ 2 2 ) x 2; the observer needs both x and σ in order to make an optimal decision . From the point of view of the observer , σ is the trial-to-trial level of sensory uncertainty associated with the measurement [38] . In a minor variation of the optimal observer , we allow for the possibility that the observer’s prior belief over category , p ( C ) , is different from the true value of ( 0 . 5 , 0 . 5 ) ; this adds a constant to dA and dB . We introduce the Bayesian confidence hypothesis ( BCH ) , stating that confidence reports depend on the internal representation of the stimulus ( here x ) only via d . In the BCH , the observer chooses a response by comparing d to a set of category and confidence boundaries . For example , whenever d falls within a certain range , the observer presses the “medium-low confidence , category 2” button . The BCH is thus an extension of the choice model described above , wherein the value of d is used to compute confidence as well as chosen category . There is another way of thinking about this . Bayesian models assume that subjects compute d in order to make an optimal choice . Assuming people compute d at all , are they able to use it to report confidence as well ? We refer to the Bayesian model here as simply “Bayes . ” We also tested several more constrained versions of this model . The observer’s decision can be summarized as a mapping from a combination of a measurement and an uncertainty level ( x , σ ) to a response that indicates both category and confidence . We can visualize this mapping as in Fig 2 , first column . It is clear that the pattern of decision boundaries in the BCH is qualitatively very different between Task A and Task B . In Task A , the decision boundaries are quadratic functions of uncertainty; confidence decreases monotonically with uncertainty and increases with the distance of the measurement from 0 . In Task B , the decision boundaries are neither linear nor quadratic . At first glance , it seems obvious that sensory uncertainty is relevant to the computation of confidence . However , this is by no means a given; in fact , a prominent proposal is that confidence is based on the distance between the measurement and the decision boundary , without any role for sensory uncertainty [10 , 11 , 39] . Therefore , we tested a model ( Fixed ) in which the response is a function of the measurement alone ( equivalent to a maximum likelihood estimate of the stimulus orientation ) , and not of the uncertainty of that measurement ( Fig 2 , second column ) . We also tested heuristic models in which the subject uses their knowledge of their sensory uncertainty but does not compute a posterior distribution over category . We have previously classified such models as probabilistic non-Bayesian [40] . In the Orientation Estimation model , subjects base their response on a maximum a posteriori estimate of orientation ( rather than category ) , using the mixture of the two stimulus distributions as a prior distribution . In the Linear Neural model , subjects base their response on a linear function of the output of a hypothetical population of neurons . We derived two additional probabilistic non-Bayesian models , Lin and Quad , from the observation that the Bayesian decision criteria are an approximately linear function of uncertainty in some measurement regimes and approximately quadratic in others . These models are able to produce approximately Bayesian behavior without actually performing any computation of the posterior . In Lin and Quad , subjects base their response on a linear or a quadratic function of x and σ , respectively . A comparison of the Lin and Quad columns to the Bayes column in Fig 2 demonstrates that Lin and Quad can approximate the Bayesian mapping from ( x , σ ) to response despite not being based on the Bayesian decision variable . All of the models we tested were variants of the six models described so far ( Bayes , Fixed , Orientation Estimation , Linear Neural , Lin , Quad ) . Each trial consists of the experimentally determined orientation and reliability level and the subject’s category and confidence response ( an integer between 1 and 8 ) . This is a very rich data set , which we summarize in Fig 3 . We find the following effects: performance and confidence increase as a function of reliability ( Fig 3a , 3b , 3h and 3i ) , and high-confidence reports are less frequent than low-confidence reports ( Fig 3e and 3f ) . Note Fig 3c and 3d especially; this is the projection of the data that we will use to demonstrate model fits for the rest of this paper . We use this projection because the vertical axis ( mean button press ) most closely approximates the form of the raw data . Additionally , because our models are differentiated by how they use uncertainty , it is informative to plot how response changes as a function of reliability , in addition to category and task . Recently , a measure of the degree of association between accuracy and confidence , meta-d′ , has been developed [41 , 42] . While it can be useful for characterizing individual differences , we do not include it in our analyses or display it in Fig 3 . That is because one strength of our experimental design is that we parametrically vary stimulus strength and stimulus reliability; this differs from papers in which meta-d′ plays a central role because , in those papers , the stimulus is often only a binary category . We used Markov Chain Monte Carlo ( MCMC ) sampling to fit models to raw individual-subject data . To account for overfitting , we compared models using leave-one-out cross-validated log likelihood scores ( LOO ) computed with the full posteriors obtained through MCMC [43] . A model recovery analysis ensured that our models are meaningfully distinguishable ( Methods ) . Unless otherwise noted , models were fit jointly to Task A and B category and confidence responses .
Our study has several limitations . For instance , because of our short presentation time , we cannot say much about how our results generalize to tasks that require integration of evidence over time [8 , 53–55] . Additionally , because our stimuli are very low-level , we cannot say much about high-level stimuli like faces [56] . Also , we only considered explicit confidence ratings , which differ from the implicit confidence that can be gathered from humans ( e . g . , by presenting two tasks and asking the subject to choose which one they feel more confident about completing correctly [57 , 58] ) or from nonhuman animals [13] ( e . g . , by measuring how frequently they decline to make a difficult choice [8] , or how long they will wait for a reward [11] ) . It is possible that implicit confidence might be more Bayesian than explicit confidence; Barthelmé and Mamassian [58] conduct an implicit confidence experiment and rule out some heuristic models . However , their experimental task is substantially different from the one presented here . In their experiment , the stimulus feature of interest ( orientation ) only takes on two values rather than varying parametrically , so it requires a different class of heuristic models . Future studies of the difference between implicit and explicit confidence should use experiments that are able to distinguish the models presented here , which has not been done . Like the present study , Aitchison et al . [19] found evidence that confidence reports may emerge from heuristic computations . However , they sampled stimuli from only a small region of their two-dimensional space , where model predictions may not vary greatly . Therefore , their stimulus set did not allow for the models to be strongly distinguished . Furthermore , although they tested for Bayesian computation , they did not test for probabilistic computation ( whether observers take sensory uncertainty into account on a trial-to-trial basis [40] ) as we do here . Such a test requires that the experimenter vary the reliability , not only the value , of the stimulus feature of interest . Navajas et al . [49] suggested that confidence reports are best described as a weighted average of precision and the probability of being correct . However , their model uses the estimated probability of being correct under a non-Bayesian decision rule [22] . They did not show the fit of a Bayesian model , and therefore their study does not constitute a true test of whether confidence is Bayesian . Here , we tested and rejected the hypothesis that confidence is a weighted average of precision and the posterior probability of being correct under a Bayesian decision rule . Sanders et al . [20] reported that confidence has a “statistical” nature . However , their experiment was unable to determine whether confidence is Bayesian or not [17] , because the stimuli varied along only one dimension . Aitchison et al . [19] note that , to distinguish models of confidence , the experimenter must use stimuli that are characterized by two dimensions ( e . g . , contrast and orientation as in this experiment , or contrast and crowding as in Barthelmé and Mamassian [58] ) . This is because , when fitting models that map from an internal variable to an integer confidence rating , it is impossible to distinguish between two internal variables that are monotonically related ( in the case of Sanders et al . [20] , the measurement and the posterior probability of being correct ) . Therefore , the only alternative model proposed by Sanders et al . [20] is based on reaction time , rather than on the presented stimuli . In detection and coarse discrimination tasks , Lau , Rahnev , and colleagues report that subjects overestimate their confidence in the periphery and for unattended stimuli . The authors have proposed a signal detection theory model in which high eccentricity or lower attention induces higher noise , and the confidence criterion may not change at all [39 , 59–63] . As a result , more probability mass will “spill over” the criterion to the high-confidence regime . How do these findings relate to ours ? At a qualitative level , they are consistent in that confidence does not seem Bayesian . However , in detection and coarse discrimination tasks , it is not possible to distinguish between fixed-criterion and probabilistic models [64] , and their data cannot be used to infer that the criterion is fixed . The paradigms in the present paper are able to distinguish such models because of the parametric manipulation of orientation , the stimulus feature of interest; indeed , we find strong evidence against the Fixed criterion model . It remains to be seen whether claims that confidence can be systematically dissociated from perceptual performance [1 , 51 , 65–69] are consistent with the account presented here , in which the brain adjusts confidence criteria based on uncertainty but in a non-Bayesian manner . Another form of non-Bayesian confidence ratings is the recent proposal that , in confidence judgments , only the “positive evidence” in favor of the chosen option matters , instead of the “balance of evidence” between two options [31 , 53 , 56 , 70 , 71] . In our tasks , this form of suboptimality would entail that confidence is derived from the ( log ) likelihood of the chosen category , instead of from the ( log ) likelihood ratio . This does not seem consistent with our data; for example , the likelihood of category 2 decreases as the absolute value of the stimulus and , correspondingly , the measurement , increases ( Fig 7b ) . However , confidence for a category 2 decision steadily increases with the absolute value of the stimulus ( Fig 7a ) . More work is needed to understand whether alternative models could explain the “positive evidence” data , and if not , what causes the difference with our results . What do our findings tell us about the neural basis of confidence ? Previous studies have found that neural activity in some brain areas ( e . g . , human medial temporal lobe [7] and prefrontal cortex [72] , monkey lateral intraparietal cortex [8] and pulvinar [10] , rodent orbitofrontal cortex [11] ) is associated with behavioral indicators of confidence , and/or with the distance of a stimulus to a decision boundary . However , such studies mostly used stimuli that vary along a single dimension ( e . g . , net retinal dot motion energy , mixture of two odors ) . Because measurement is indistinguishable from the probability of being correct in these classes of tasks , neural activity associated with confidence may represent either the measurement or the probability of being correct [19] . In addition to the recommendation of Aitchison et al . [19] to distinguish between these possibilities by varying stimuli along two dimensions , we recommend fitting both Bayesian and non-Bayesian probabilistic models to behavior . In view of the relatively poor performance of the Bayesian models in the present study , the proposal [12] to correlate behavior and neural activity with predictions of the Bayesian confidence model should be viewed with skepticism . Our results raise general issues about the status of Bayesian models as descriptions of behavior . First , because it is impossible to exhaustively test all models that might be considered “Bayesian , ” we cannot rule out the entire class of models . However , we have tried to alleviate this issue as much as possible by testing a large number of Bayesian models—far more than the number of Bayesian and non-Bayesian models tested in other studies of confidence . Second , Bayesian models are often held in favor for their generalizability; one can determine the performance-maximizing strategy for any task . Although generalizability indeed makes Bayesian models attractive and powerful , we do not believe that this property should override a bad fit . One could take two different views of our heuristic model results . The first view is that the heuristics should be taken seriously as principled models [73]; here , the challenge is to demonstrate that they describe behavior in a variety of tasks and can be motivated based on underlying principles . The second view is that these are descriptive models simply meant to demonstrate that a simple model can provide a good fit to the data; here , the heuristics are benchmarks for more principled models , and the challenge is to find a principled model that fits the data as well as the heuristics . We lean towards the second view and interpret our results as demonstrating that the purest form of the Bayesian confidence hypothesis does not describe human confidence reports particularly well . However , one might still conclude , after examining the fits of the Bayesian model , that the behavior is “approximately Bayesian” rather than “non-Bayesian . ” As written , this is a semantic distinction because it relies on one’s definition of “approximate . ” However , it can be turned into a more meaningful question: Are the differences between human behavior and Bayesian models accounted for by an unknown principle , such as an ecologically relevant objective function that includes both task performance and biological constraints ? Although there are benefits associated with veridical explicit representations of confidence [74–76] , there are also neural constraints that may give rise to non-Bayesian behavior [23 , 24] . Such constraints include the kinds of operations that neurons can perform , the high energy cost of spiking [77 , 78] , and the cost of neural wiring length [79 , 80] . A search for ecologically rational constraints on Bayesian computation benefits from a positive characterization of the deviations from Bayesian computation , in the form of heuristic models such as Lin and Quad . Specifically , one could define neural networks with various combinations of constraints , and train them as if they were psychophysical subjects in our tasks . After training , one could fit behavioral models to them; this approach has already shown that the output from such neural networks is sometimes best described by heuristic models [81] . Using model ranking as a measure of similarity , one could determine which network architecture and training procedure produces confidence behavior that is most similar to that of humans . This could reveal which constraints are responsible for the specific deviations from Bayesian computation that we have observed .
The experiments were approved by the University Committee on Activities Involving Human Subjects of New York University . Informed consent was given by each subject before the experiment . This control experiment was identical to experiment 1 except for the following modifications: This experiment was identical to experiment 1 except for the following modifications: | Humans are able to report a sense of confidence in decisions that we make . It is widely hypothesized that confidence reflects the computed probability that a decision is accurate; however , this hypothesis has not been fully explored . We use several human behavioral experiments to test a variety of models that may be considered to be distinct hypotheses about the computational underpinnings of confidence . We find that reported confidence does not appear to reflect the probability that a decision is correct , but instead emerges from a heuristic approximation of this probability . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"statistical",
"noise",
"ellipses",
"decision",
"making",
"statistics",
"social",
"sciences",
"geometry",
"neuroscience",
"cognitive",
"psychology",
"mathematics",
"probability",
"distribution",
"cognition",
"research",
"and",
"analysis",
"methods",
"gaussian",
"noise",
"animal",
"cells",
"behavior",
"research",
"assessment",
"probability",
"theory",
"psychology",
"cellular",
"neuroscience",
"cell",
"biology",
"normal",
"distribution",
"research",
"validity",
"neurons",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"physical",
"sciences",
"cognitive",
"science"
] | 2018 | Comparing Bayesian and non-Bayesian accounts of human confidence reports |
Huntingtin is a large HEAT repeat protein first identified in humans , where a polyglutamine tract expansion near the amino terminus causes a gain-of-function mechanism that leads to selective neuronal loss in Huntington's disease ( HD ) . Genetic evidence in humans and knock-in mouse models suggests that this gain-of-function involves an increase or deregulation of some aspect of huntingtin's normal function ( s ) , which remains poorly understood . As huntingtin shows evolutionary conservation , a powerful approach to discovering its normal biochemical role ( s ) is to study the effects caused by its deficiency in a model organism with a short life-cycle that comprises both cellular and multicellular developmental stages . To facilitate studies aimed at detailed knowledge of huntingtin's normal function ( s ) , we generated a null mutant of hd , the HD ortholog in Dictyostelium discoideum . Dictyostelium cells lacking endogenous huntingtin were viable but during development did not exhibit the typical polarized morphology of Dictyostelium cells , streamed poorly to form aggregates by accretion rather than chemotaxis , showed disorganized F-actin staining , exhibited extreme sensitivity to hypoosmotic stress , and failed to form EDTA-resistant cell–cell contacts . Surprisingly , chemotactic streaming could be rescued in the presence of the bivalent cations Ca2+ or Mg2+ but not pulses of cAMP . Although hd− cells completed development , it was delayed and proceeded asynchronously , producing small fruiting bodies with round , defective spores that germinated spontaneously within a glassy sorus . When developed as chimeras with wild-type cells , hd− cells failed to populate the pre-spore region of the slug . In Dictyostelium , huntingtin deficiency is compatible with survival of the organism but renders cells sensitive to low osmolarity , which produces pleiotropic cell autonomous defects that affect cAMP signaling and as a consequence development . Thus , Dictyostelium provides a novel haploid organism model for genetic , cell biological , and biochemical studies to delineate the functions of the HD protein .
Huntington's disease ( HD ) is an autosomal dominant neurodegenerative disorder whose clinical manifestations include psychiatric disturbances , cognitive decline and characteristic involuntary movements , typically diagnosed in mid-life [1] , [2] . HD is caused by a CAG trinucleotide repeat expansion mutation ( >35 units ) that produces an elongated version of a normally polymorphic polyglutamine segment in huntingtin [3]–[5] , a large 350 kDa ubiquitously expressed HEAT ( huntingtin , elongation factor 3 , the A subunit of protein phosphatase 2A , and TOR1 ) repeat protein . Several lines of evidence indicate that the HD mutation confers a CAG length-dependent ‘gain-of-function’ that likely produces its distinctive neuropathology through a modulatory effect on some structural or functional feature of huntingtin [6]–[9] . HD homozygote individuals with two mutant HD genes , and therefore no wild-type huntingtin , develop the characteristic movement disorder with timing comparable to that seen in typical HD mutation heterozygote individuals , indicating the absence of a strong dosage effect . Moreover , complete deficiency of huntingtin causes developmental abnormalities and embryonic lethality in the mouse that can be fully rescued by mutant huntingtin , indicating that these fundamental normal functions of huntingtin are not abrogated by the HD mutation . Thus , defining the disease-producing ‘gain-of-function’ - either a polyglutamine-length dependent increase or deregulation of a normal huntingtin activity or the introduction of a novel polyglutamine-length dependent activity , will require an understanding of the protein's normal function ( s ) . Huntingtin is present throughout eukaryotic evolution except in fungi and plants and shows no close primary sequence homology to any other protein [6] . Therefore , one approach to huntingtin function is to investigate its orthologs in tractable experimental models . Manipulation of the HD gene homologs in model organisms has revealed that huntingtin is essential for normal embryonic development both in the mouse and in the zebrafish [10]–[13] , is dispensable for Drosophila development [14] and is implicated in a variety of functions ranging from vesicle trafficking to chromatin silencing and gene expression [15]–[18] . However , though murine embryonic stem cells lacking huntingtin are viable in tissue culture , permitting multi-cellular developmental studies in vitro [10] , investigation of huntingtin function in single cells versus multicellular stages of development would be expedited by an experimental organism with a short life-cycle , comprising cellular and multicellular developmental stages . The soil amoeba Dictyostelium discoideum is a bona fide multicellular eukaryotic organism with a haploid genome and a relatively simple developmental program that serves as a model for basic biological research [19] and is emerging as a valuable tool for understanding gene function and pathogenic mechanisms in a variety of human disorders [20]–[24] . During development , Dictyostelium undergoes a series of coordinated morphological and physiological changes that are initiated by starvation and progress in defined stages over a 24 hour period . Within the first 6 hours of development , cells secrete , and undergo chemotaxis toward cyclic adenosine monophosphate ( cAMP ) to form aggregation centers . The secretion of cAMP promotes a G protein-coupled receptor signal/response cascade resulting in the formation of loose mounds comprising up to 100 , 000 cells [25] , [26] . As development continues , cells within the mound are directed to differentiate into either prestalk or prespore cells , leading to morphological changes that yield a multicellular stalk , supporting a ball of encapsulated dormant spores [27] . To explore huntingtin function in both single cells and multicellular structures of the same organism , we have characterized a Dictyostelium ortholog ( hd ) of the human HD gene , have generated a viable hd-null mutant and have delineated a number of consequent phenotypes . We present evidence to suggest that huntingtin is a multifunctional protein that plays a protective role during hypoosmotic stress , regulates cell shape and homophilic cell-cell adhesion through a cytoskeletal–based mechanism , modulates pre-spore/spore cell fate determination and affects spore dormancy in Dictyostelium . Interestingly , hd− cells were found to have a requirement for bivalent cations in the medium for optimal chemotaxis and cell aggregation . Our findings establish Dictyostelium as a valuable experimental eukaryotic organism for exploring in biochemical detail huntingtin's normal function ( s ) , as defined by the pleiotropic effects of huntingtin deficiency throughout the developmental life cycle .
The Dictyostelium discoideum genome , examined via dictyBase ( www . dictybase . org ) [28] , contains a single gene ( DDB_G0272344 ) with evident sequence homology to human huntingtin . The hd locus is comprised of four exons , located on chromosome 2 . Analysis of GenBank with psi-BLAST [29] , [30] placed the product of hd firmly within the huntingtin family ( Figure 1A ) , with a length of 3 , 095 amino acids comparable to the 3 , 144 amino acids of human huntingtin . Structural analysis of huntingtin proteins has identified the presence of numerous HEAT and HEAT-like repeats thought to produce a large α-helical solenoid rather than a typical globular protein [6] , [15] , [31] , [32] . Dictyostelium huntingtin is also predicted to be α-helical across most of its length ( Figure 1B ) and , as in other organisms , BLASTP searches of GenBank revealed no significant sequence alignments with other Dictyostelium proteins . Interestingly , Dictyostelium huntingtin contains a polyglutamine tract of 19 residues ( Figure 1B ) that is comparable in size to that found in normal-range human huntingtin but is encoded by the trinucleotide repeat CAA interrupted by a single CAG codon , is not followed by a polyproline domain , is located further downstream of the initiator methionine ( at residue 533 rather than residue 18 as in the human protein ) , and may reflect the unusually high number of predicted proteins ( ∼34% ) that contain homopolymer tracts of 15 residues or more in Dictyostelium [33] . To determine the spatial and temporal pattern of huntingtin mRNA expression throughout the life cycle ( Figure 2A ) ; we cloned the intergenic region up to the closest upstream gene ( 608 bp ) and used it to replace the actin15 promoter in pTX-GFP and to direct expression of the green fluorescent protein ( GFP ) , which was detected in all cell types of transformed AX3 wild-type Dictyostelium cells during growth and development ( Figure 2B , 2C , 2D ) , a pattern that fits well with the developmental and cell-type specific mRNA expression data acquired through extensive microarray and RNA-seq analysis which is freely available in dictyBase ( DDB_G0272344 ) [34] . To further characterize the function of huntingtin in Dictyostelium , we examined the existence of alternate transcripts using exon specific primers and reverse transcriptase polymerase chain reaction ( RT-PCR ) . No alternatively spliced forms of mRNA could be detected using various primer sets specific for exons ( exon 1–exon 4 ) during vegetative growth and development of AX3 cells ( Figure 2E ) . To explore huntingtin function , we utilized homologous recombination to create multiple independent Dictyostelium strains carrying a disruption of the hd gene in the haploid genome ( Figure 3A ) . The disruption cassette targeted hd such that a double-recombination event would remove 278 bp of exon 2 and insert a blasticidin resistance ( Bsr ) selection cassette ( Figure 3B , 3C ) . Following transformation of parental AX3 cells and blasticidin selection , 150 independent resistant clones were isolated and those ( 16 ) with the properly targeted hd gene disruption were identified by PCR amplification of the expected genomic DNA products ( Figure 3D ) . Proper targeting was also confirmed by Southern blot using DIG-labeled Bsr specific probes as well as by RT-PCR amplification of cellular mRNA which demonstrated that hd mRNA was absent ( Figure 3E , 3F ) . Clonal isolates from multiple independent mutants possessing the same disruption and developmental phenotype ( glassy sori and round spores ) were obtained . One of these clones ( httE13 ) was used in all subsequent experiments and is referred to as hd− for clarity . We did not observe differences in the growth properties of hd− and wild-type cells when grown as adherent cultures in tissue culture dishes ( data not shown ) . However , in contrast to wild-type cells , axenically grown shaking cultures of hd− cells consistently grew faster with a doubling time of ∼10 hours compared to the ∼12 hour doubling time of wild-type cells ( Figure S1 ) , suggesting a physiological impact of the absence of huntingtin at the level of single cells . Plating of hd− cells under non-nutrient , low ionic strength phosphate buffer ( KK2; ∼40 mOsmol/L ) , elicited a cell rounding phenotype that suggested an actin-cytoskeleton defect , which we examined by staining for F-actin . The morphology of hd− cells grown as an adherent culture in HL-5 medium was similar to that of wild-type cells ( Figure 4A ) and phalloidin staining showed a similar distribution of F-actin in extended pseudopods in both genotypes ( Figure 4C ) . However , removal of nutrients and the addition of KK2 buffer resulted in a rapid reduction in pseudopod formation ( ∼20 minutes ) and ultimately a failure to polarize only in the hd− cells ( Figure 4B ) . The rounded cell phenotype of hd− cells was accompanied by a redistribution of F-actin from the cell cortex to the cytosol ( Figure 4D ) . As the extreme cell rounding observed in low osmotic buffer hinted at a potential defect in contractile vacuole osmoregulation , we assessed the ability of hd− cells to respond to osmotic stress . When wild-type cells were placed in water , several small contractile vacuoles formed within 30 minutes and , over time , the cells compensated for the sudden change in their osmotic environment and gradually regained their original size , shape and cell-substratum adhesion ( Figure 4E; Video S1 ) . In contrast , hd− cells failed to form any visible vacuoles , but instead became round , swollen , lost their adhesion to the substratum and underwent complete lysis within 5–6 hours ( Figure 4F; Video S2 ) , indicating that contractile vacuole activity was dramatically compromised in hd− cells under conditions of hypotonic stress . No difference was observed between wild-type and hd− cells under hyperosmotic ( 400 mM sorbitol ) conditions ( data not shown ) . To determine whether the mutant cells were capable of carrying out the full program of multicellular development , we assayed wild-type and hd− cells by plating on filters . Wild-type cells formed numerous streaming aggregation centers by 6 hours that went on to form tight mounds by 12 hours , early culminants by 18 hours and fruiting bodies by 24 hours as expected ( Figure 5 ) . By contrast , hd− mutant cells formed delayed aggregates , and by 12 hours had constructed comparatively loose mounds with supernumerary prestalk tips , which subsequently went on to form multiple slugs and small fruiting bodies ( Figure 5 ) . Thus , whereas wild-type cells had completed the developmental program by ∼24 hours , the development of hd− cells was delayed and asynchronous , resulting in the appearance of aborted intermediates and the formation of small fruiting bodies that develop glassy sori . A closer examination of the effects that loss of huntingtin has on the earliest stages of development revealed deficits in both the aggregation properties and cell-cell cohesiveness of wild-type and hd− cells . When assayed on non-nutrient KK2 agar or buffer-soaked nitrocellulose filters , wild-type cells migrated as streams to form aggregates by 6–8 hours ( Figure 6A ) . The hd− cells did not form streams that led to distinct aggregation territories , but rather form delayed aggregation territories by accretion ( Figure 6B ) . We next examined the behavior of cells deposited at different cell densities under starvation buffer ( non-nutrient phosphate buffer KK2 , pH 6 . 2 ) in order to get a better view of the streaming defect . When plated at 1×105 cells/cm2 , wild-type cells formed long streams of polarized , elongated cells that congregated into aggregation signaling centers ( Figure 6C; Video S3 ) . In contrast , hd− cells adopted the rounded shape noted previously and ultimately failed to polarize , form aggregation centers or initiate streams ( Figure 6D; Video S4 ) . Yet , when assessed at high cell densities ( 5×106 cells/cm2 ) , hd− cells managed to form aggregation centers , largely by accretion; however the mounds appear polarized ( regular shape on one side , irregular with dark clumping cells on the other side ) and many cell were excluded ( Figure 6E , 6F ) . The observed defects suggest two possibilities; 1 ) hd− cells are defective in cAMP signaling , or 2 ) the extreme sensitivity of hd− cells to conditions of low osmolarity negatively affects development . The secretion of periodic cAMP pulses is required to initiate Dictyostelium development [35] and so the addition of timed pulses of exogenous cAMP should facilitate early development in hd− cells if they are defective in cAMP signaling . Pulsing wild type cells in KK2 buffer with 75 nM cAMP every 6 minutes for a period of 4–5 hours accelerated the onset of chemotactic streaming , but equivalent cAMP pulses did not rescue the rounding or chemotactic streaming of hd− cells ( data not shown ) . We next assessed what effect the addition of bivalent cations ( higher osmolarity ) would have on cAMP signaling and chemotactic aggregation of hd− cells under KK2 buffer . Interestingly , the addition of 1 mM CaCl2 ( Figure 6G , 6H ) , 1 mM MgCl2 or 1 mM MgSO4− ( data not shown ) rescued endogenous cAMP signaling and chemotactic aggregation of hd− cells ( Video S5 ) . However , the streams formed by hd− cells in the presence of bivalent cations were shorter and had a tendency to break and re-form when compared to wild type streams ( Figure S2A , S2B ) . In addition , rescue of cAMP signaling and chemotactic aggregation of hd− cells in the presence of bivalent cations was inhibited by the addition of 1 mM ethylene glycol tetraacetic acid ( EGTA ) ( Video S6 ) . To further assess cAMP signaling in hd− cells , we explored their aggregation properties on KK2 agar containing caffeine ( 1 , 2 . 5 and 5 mM ) , a compound shown to inhibit cAMP signaling in Dictyostelium [36] . Hd− mutant cells were equally sensitive and exhibited similar developmental phenotypes to wild type cells when treated with caffeine . Aggregation and development of wild type and hd− cells was severely inhibited at 5 mM caffeine and at 1 mM caffeine both wild type and hd− cells were capable of forming fruiting bodies ( data not shown ) . We next assessed aggregation and development of wild type and hd− cells on KK2 agar containing EGTA ( 1 mM or 2 mM ) . In contrast to caffeine treatment , development of wild type cells in the presence of 1 mM EGTA was unaffected ( Figure 6I ) whereas hd− cells failed to form chemotactic streams and arrested development as loose aggregates ( Figure 6J ) . Taken together , our data suggest that hd− cells are not primarily defective in cAMP signaling but rather have a hypersensitivity to conditions of low osmolarity which affects cAMP signaling and thus development . The cell-cell adhesion properties of hd− cells in the absence or presence of 10 mM ethylenediaminetetraacetic acid ( EDTA ) mirrored those observed for wild-type cells during the first 3 hours of development ( Figure 7 ) . However , after ∼4 hrs of development , wild-type cells began to attain EDTA-resistant cell-cell contacts whereas hd− cells not only failed to acquire EDTA-resistant cell-cell contacts but also began to display an overall loss in cell-cell cohesion as judged by an increase in the number of single cells assayed in the absence of EDTA ( Figure 7 ) . Even by 24 hours , neither EDTA-resistant adhesion in hd− cells nor recovery of their cohesion properties in suspension developed ( data not shown ) . Thus , stringent conditions exacerbated hd− cell developmental deficits , revealing aberrant cell polarization , aggregation and motility consistent with abnormal cell shape , cytoskeletal function and cell-cell adhesion properties . Though streaming and aggregation of hd− cells were abnormal and overall development was delayed , multiple prestalk tips formed atop most of the developing hd− mounds . Each of these tips elongated into a finger structure , suggesting the possibility of aberrant prestalk/prespore cell differentiation ( Figure 8A , 8B ) , and eventually produced a fruiting body with a glassy sorus containing a reduced spore complement when developed on KK2 agar or with a lawn of Klebsiella on SM agar ( Figure 8C , 8D ) . In contrast to wild-type fruiting bodies , the numbers of spores in hd− sori decreased as the fruiting bodies aged , resulting in a glassy sorus; this suggested a potential defect in spore formation or alternatively , spore dormancy . We used electron microscopy to examine the fine structure of the elliptical wild-type spores and round mutant hd− spores from 72 hr fruiting bodies ( Figure 8E , 8F ) . The EM images revealed the premature germination of spores in the sori of hd− fruiting bodies . In stark contrast to the dormant wild-type spores , the hd− sori contained a mixture of apparently dormant spores with three wall layers , swollen spores , swollen spores with thin spore walls about to release amoebae , and nascent amoebae ( Figure 8F ) . Our data suggest that shortly after fruiting body formation hd− spores are unable to maintain dormancy , or alternatively , fail to achieve dormancy possibly through poor differentiation properties and germinate within the sorus . Consequently , we assessed the ability of hd− cells to differentiate into stalk and spore cells in low cell density monolayer culture [37] . In response to 10–100 nM of the stalk cell inducer , differentiation inducing factor 1 ( DIF-1 ) , hd− mutants were able to produce stalk cells as efficiently as wild-type cells ( data not shown ) . However , in response to the sporulation inducer 8-Br-cAMP ( 5–15 mM ) , a cell permeable analogue of cAMP , sporulation in the mutants was delayed ∼24 hours compared to wild-type cells ( Figure 8G ) . hd− cells collectively formed spore cells at ∼40% efficiency relative to wild-type controls after 48 hours differentiation ( Figure 8G ) . At least some spores produced by development of hd− organisms were capable of germination and formation of viable amoebae . The spore coats of both wild-type and hd− cells also stained brightly with Calcofluor , demonstrating the normal presence of cellulose in the mutant spores ( data not shown ) . To test the relative viability of the hd− mutant spores , we compared their germination rates with wild-type spores . Spores harvested from sori of each cell line were mixed with Klebsiella , deposited on SM agar plates and the number of plaques that formed was scored . Spores harvested from wild-type sori displayed an average survival rate of 82±3 . 4% , whereas hd− spores demonstrated a dramatically reduced mean survival rate of 20±2 . 2% ( Figure 9 ) . This did not appear to be due to reduced resilience of hd− mutant spore coats , as spore viability of both wild-type and hd− mutant spores was not differently altered by treatment with the non-ionic detergent NP-40 ( 0 . 5% ) and brief heat treatment ( 45°C ) ( Figure 9 ) . To distinguish whether the developmental defects observed in hd− cells were cell autonomous or non-cell autonomous , we followed the cell fate of hd− cells during chimera development in mixed cell cultures . Mutant hd− cells and wild-type cells carrying an act15/GFP construct , which marks all cells , were prepared and mixed in ratios of 1∶1 , 1∶3 or 1∶9 with unmarked cells of the opposite genotype ( or of the same genotype as controls ) . As expected , the deficits revealed by hd− cells in assays at the single cell stage ( e . g . , cell rounding , actin cytoskeleton rearrangement , sensitivity to osmotic shock ) or early in development ( cell-cell adhesion ) were not rescued by the presence of wild-type cells in co-culture ( data not shown ) . Similarly , in developing co-aggregates of cells developed under starvation buffer , hd− cells failed to populate the central region of the developing aggregate , consistent with the inability of wild-type cells to correct the observed failure in the initiation of intracellular cAMP signaling pathways required to trigger aggregation ( Figure 10A ) or , possibly , the onset of cell-cell adhesion properties in hd− cells . We then studied cell pattern organization of live cells during the slug stage , where prestalk cells primarily populate the anterior region and prespore cells occupy the slug posterior [38] . In the context of an excess of unmarked wild-type cells ( 3∶1 ratio ) , GFP-marked hd− cells within chimeric slugs were distributed within the prestalk zone ( pstO ) , rearguard and anterior-like cells ( ALC ) , which suggests that they have a greatly reduced ability to form prespore cells in the presence of wild-type cells ( Figure 10B ) . In the complementary experiment , GFP-marked wild-type cells co-developed with hd− cells populated primarily the prespore region at cell ratios of 1∶1 or lower ( Figure 10C ) . In both comparable control experiments , GFP fluorescence was observed throughout the entire slug . Furthermore , this trend was also seen in terminally differentiated chimeric fruiting bodies with wild-type cells predominantly found in the spore mass whereas hd− cells occupied the tip and lower cup region , stalk and basal disc ( Figure 10D , 10E ) . Thus , although hd− cells were capable of forming spores when developed as a pure population ( albeit poorly , with reduced viability ) , the presence of wild-type cells in a chimeric organism did not rescue but rather reduced the contribution of hd− cells to spore formation .
Huntingtin is notable for its role in HD , where an expanded polyglutamine tract near the amino terminus of human huntingtin produces late-onset progressive neurodegeneration , likely through a modulatory effect of the polyglutamine region on the structure and/or function of the protein . In this study , we have investigated the fundamental function ( s ) of huntingtin in the haploid eukaryote Dictyostelium discoideum , by characterizing deficiency phenotypes in the single and multicellular phases of development of this social amoeba . The deletion of the hd gene , using targeted homologous recombination , was compatible with cell growth but produced pleiotropic cell autonomous phenotypes that demonstrated that mutant cells could not efficiently complete the processes necessary for coordinated synchronous development of a multicellular organism . The pleiotropic effects of huntingtin deficiency could be the result of loss of a single activity or could be separate effects due to the protein having multiple functions . Further experimentation will be required to resolve these options , but the multiple phenotypes , at both single cell and multicellular stages , offer several routes to further explore the issue . Our findings suggest that huntingtin participates , directly or indirectly , in actin cytoskeleton-membrane dynamics that affect cell shape . The mechanical properties of the cytoplasm are important determinants of cell shape and permit cells to change the cytoplasm from a rigid to a dynamic actin network that can greatly influence cell motility [39] . Huntingtin under certain conditions appears to act as an essential facilitator of this process . Whereas hd− cells displayed apparently normal cell shape , pseudopod formation , F-actin localization and random motility in nutrient-rich media , when assayed in the absence of nutrients ( under developmental buffer ) , they failed to produce membrane extensions , were abnormally round and lacked cortical F-actin , although at high cell density , rounded hd− cells managed to form delayed aggregation territories by accretion . These observations could also reflect the loss of an excitatory signaling pathway during development of hd− cells , as the phenotypes resemble those exhibited by Dictyostelium null mutants for adenylyl cyclase that overexpress the catalytic subunit of protein kinase A ( PKA ) [40] or cells expressing the constitutively active inhibitory heterotrimeric G-protein Gα9 [41] . Addition of exogenous pulses of 75 nM cAMP at 6 minute intervals over a period of 4–5 hours did not rescue the aggregation defect in our hd− cells ( data not shown ) , though it is plausible that high cell density partially rescues mound formation by increasing cell-cell contacts . However , cAMP signaling and subsequent chemotactic aggregation could be rescued through the addition of bivalent cations ( Ca2+ or Mg2+ ) to the medium , which suggests the possibility that hd− cells exhibit depleted cationic stores . Our findings support the notion that huntingtin may facilitate proper Ca2+- Mg2+-dependent actin cytoskeleton remodeling that determines cell shape , but that the detection of this role critically depends upon the conditions in which the cells are evaluated . A second function for huntingtin is maintenance of cellular integrity under conditions of osmotic stress . Contractile vacuoles are intracellular membrane organelles involved in osmoregulation , function as a highly efficient acidic Ca2+-store that is required for cAMP-induced Ca2+-influx and are found in free-living amoebae and protozoa [42]–[44] . Normally , wild-type cells growing in culture media contain a few moderately active contractile vacuoles that maintain the osmotic balance of the cell . The contractile vacuole system functions as a “bladder” during conditions of hypoosmotic stress , where it collects fluid ( water and neutral amino acids ) in a network of tubular channels that associate with the cortical actin network to allow for transient fusion with the plasma membrane and expulsion of its contents to the extracellular environment [45] , [46] . The behavior and survival rate of hd− cells in response to hyperosmotic conditions is similar to wild-type cells ( data not shown ) . However , contractile vacuole activity in response to hypoosmotic stress is completely abolished in hd− cells , and renders the cells sensitive to conditions of low ionic strength . These effects were consistently associated with a retraction of F-actin from the cell cortex to the cytosol , suggesting that in the absence of huntingtin , inefficient actin cytoskeleton remodeling may underlie the failure of the tubular network to convert into the contractile vacuole . This could involve abnormal formation of the vacuole and failure to discharge , or altered regulation of F-actin binding proteins , as there are examples of both . Cells deficient for MEGAP1 and MEGAP2 , members of the Pombe/Cdc15 homology ( PCH ) family of proteins involved in actin cytoskeletal reorganization , are sensitive to hypoosmotic conditions because they are defective in the tubulation and the associated emptying of contractile vacuoles [47] . Whereas MEGAP-null cells fail to form tubules from vacuoles and therefore accumulate the latter , we suggest that hd− cells cannot effectively osmoregulate due to the inability of the tubule system to convert to vacuoles . Indeed , Dictyostelium double mutants lacking two F-actin crosslinking proteins , α-actinin and gelation factor , display a general weakening of the cortical cytoskeleton and , like hd− cells , do not exhibit the normal polarized morphology of wild-type cells during aggregation and are sensitive to hypoosmotic shock [48] . Our data further suggest that the defective CV system in hd− cells renders cells not only sensitive to extreme hypoosmotic shock , but secondarily affects intracellular ion homeostasis as the CV has been shown to function as a major Ca2+-store [42] , [44] as well as being tightly linked to Ca2+/Mg2+-containing mass dense granules [43] . As a consequence , unless these cations are provided exogenously hd− cells appear unable to initiate cAMP-induced Ca2+-transients that may act in a feedback loop to positively reinforce cAMP relay [44] leading to chemotaxis . Moreover , in contrast to wild type cells , development of hd− cells on non-nutrient agar was blocked by EGTA and , interestingly , was not differentially sensitive to low concentrations of caffeine ( data not shown ) suggesting that a defect in cAMP signaling , if present , is relatively minor in comparison to their sensitivity to low osmolarity and/or cation chelation . We posit that huntingtin likely acts very early in establishing CV activity or integrity under hypoosmotic conditions . It may act indirectly , or it may function directly as a regulatory scaffold in this setting to drive assembly of the cortical cytoskeleton with CV-associated proteins during the conversion of tubule to vacuoles . Importantly , several proteins associated with the contractile vacuole network and involved in membrane protein trafficking are related to human proteins , suggesting that some functions of the contractile vacuole network have been preserved during the evolution of higher eukaryotes [45] , [49] . The cell-cell adhesion properties of hd− cells are similar to wild-type cells during the first 3 hours of development , but as development proceeds hd− cells fail to acquire EDTA-resistant contacts . This suggests that hd− cells retain the ability to form early functional adhesion sites but not the EDTA-resistant sites characteristic of aggregating chemotactic cells . Cell-cell contact plays an important role in differentiation and gene expression in Dictyostelium [50]–[52] . EDTA-resistant cell-cell binding is mediated by the expression of glycoprotein gp80 at the onset of aggregation [53] , [54] . Interestingly , the expression of gp80 is greatly augmented by pulsatile cAMP signaling and the formation of EDTA-sensitive cell-cell contacts [51] , [55] . Taken together , the apparently normal EDTA-sensitive adhesion properties of hd− cells , their inability to initiate signaling centers at densities where wild-type cells can , and the rescue of cAMP relay and streaming by the addition of exogenous cations suggest that hd− cells potentially lack the Ca2+- Mg2+-dependent excitatory intracellular signal transduction pathways that upregulate gp80 EDTA-resistant adhesion complexes , or that they are deficient in transporting gp80 to the cell surface . We also provide compelling evidence to suggest huntingtin modulates prespore cell fate choice . In the context of either an equal ratio or excess ratio ( 3∶1 and 9∶1 ) of unmarked wild-type cells , GFP-marked hd− cells within chimeric slugs are distributed within the prestalk zone ( pstO ) , and as rearguard and anterior-like cells ( ALC ) which suggests they have a significantly reduced ability to form prespore cells when challenged in the presence of wild-type cells . This could be due to a defect in the mutant cells in cell-cell interaction , either through extracellular signaling or direct contact relative to the efficient homotypic interactions of wild-type cells . Indeed , this difference in differentiation might be explained by the adhesion defect in hd− cells as , in chimeras with wild-type , gp80-null cells are also directed preferentially toward the prestalk pathway [56] . Differential differentiation of huntingtin deficient cells during development has also been observed in vertebrates . In zebrafish , reduced expression of huntingtin differentially targets development of telencephalic neurons compared to mid- and hind-brain [57] . In mouse chimeras , Hdh−/− cells also preferentially colonize the hypothalamus , midbrain , and hindbrain during relative to the telencephalon and the thalamus during early development [58] . Thus , like these latter neuronal populations , Dictyostelium cells require huntingtin for the proper development of viable spores in the presence of wild-type cells . The culmination of Dictyostelium multicellular development permits survival of the organism through the production of environmentally resistant spores and the complex organization of actin filaments in Dictyostelium spores contributes to their ellipsoid shape and dormancy [59] . hd− mutants produced abnormally round spores , which exhibited decreased viability , suggesting that they were poorly differentiated or had a defect in the actin-cytoskeleton . Moreover , electron microscopic examination of the fine structure of spores from wild-type and hd− fruiting bodies suggested that the formation of the glassy sorus was likely a result of the premature germination of spores and the resulting death of newly emerged amoebae . Spore dormancy is maintained via active biological processes including high osmolarity , actin dynamics , production of the germination inhibitor discadenine and active PKA which in concert function to insure a viable supply of spores [60]–[62] . Since hd− amoebae are sensitive to changes in osmolality , display aberrant F-actin staining and show compromised sporulation when subjected to constitutive PKA activation via 8-Br-cAMP in vitro , it is plausible that in the absence of huntingtin , imprecise cytoskeletal architecture and signaling from PKA required to maintain or enter dormancy might also be defective at this late stage of development . Our data do not provide a simple definition for a single normal function for huntingtin , but , as huntingtin deficiency in Dictyostelium produces pleiotropic effects throughout the life cycle , our findings are consistent with the consensus from mammalian studies that huntingtin is a multifunctional protein that can impact upon many biochemical processes . In Dictyostelium , defects in CV activity or integrity leading to a disruption in ion homeostasis affecting the cytoskeleton , cell shape and cell-cell adhesion would all predictably interfere with many aspects of chemotactic aggregation and development . Defining the degree to which the phenotypes reported here are connected by common underlying biochemical deficits or , alternately , reflect different functions of huntingtin will require detailed molecular investigation . However , the existence of a Dictyostelium ortholog of human huntingtin , the viability of the null hd− mutant , and its discrete , readily assayed deficiency phenotypes indicate that this haploid organism provides an effective genetic model system to identify molecular and cellular processes affected by the loss of huntingtin function . While the latter is of fundamental biological interest considering the unique nature of this ancient large α-solenoid HEAT protein , delineating which of these functions are conserved in mammals and determining whether they are altered by expansion of the polyglutamine tract in human huntingtin will also provide much needed insights into the mechanism by which mutant huntingtin triggers HD pathogenesis .
Wild-type Dictyostelium discoideum AX3 cells were grown in association with Klebsiella aerogenes on SM plates , axenically in tissue culture plates or as shaking cultures in HL-5 medium ( Formedia ) at 21°C . To assess growth rates , shaking cultures ( 150 rpm ) of AX3 or hd− null mutant cells were suspended in HL-5 at a density of 1×104 cells/mL and grown at 21°C . Cells were counted in triplicate using a hemocytometer . For synchronous development exponentially growing cells ( ∼2×106 cells/mL ) were washed three times from the nutrient source with DB buffer ( 5 mM Na2HPO4 , 5 mM KH2PO4 , 1 mM CaCl2 , 2 mM MgCl2 , pH 6 . 5 ) by centrifugation at 1500 rpm for 5 min , resuspended at a density of 5×107 cells/mL , deposited on black nitrocellulose filters supported by filter pads soaked in DB buffer and developed at 21°C . Growth curves were determined for the hd− strain , and parental ( wild-type ) strain AX3 . Shaking ( 150 rpm ) cultures were grown axenically in HL-5 at 21°C . Exponentially growing cultures were used to inoculate four 50 mL cultures for each strain , starting at 1×105 cells per mL and counted daily using a hemocytometer . For development under buffer ( KK2 ) , cells were grown in 6-well tissue culture dishes in HL-5 to a density of ∼5×105 cells/cm2 . Cells were washed twice with KK2 and then allowed to develop under KK2 or KK2 supplemented with 1 mM CaCl2 , 1 mM MgCl2 , 1 mM MgSO4 or were provided with 75 nM pulses of cAMP every 6 minutes for a period of 4 hours [41] . For development and assessment of streaming behavior cells were washed twice with KK2 and then deposited on agar at a density of ∼5×105 cells/cm2 on non-nutrient 1 . 5% agar plates ( KK2 buffer pH 6 . 4 ) or KK2 supplemented with caffeine ( 1 , 2 . 5 and 5 mM ) or 1 mM EGTA . The targeting construct for disruption of the hd gene was made by standard cloning procedures in the floxed-Bsr gene disruption vector pLPBLP [63] . PCR was used to amplify genomic sequences flanking and within the coding region of the hd gene . The 5′ targeting region of hd was amplified using primers 5′CCCGGTACCATGGATCTTATTCG3′ and 5′-CCCAAGCTTCCAATGATAATATA3′ which incorporate restriction sites ( underlined ) for KpnI and HindIII , respectively to facilitate cloning into the vector pLPBLP . The 3′ targeting region of the hd gene was amplified using primers 5′CCCCTGCAGTTCTCCACCAATCT3′ and 5′CCCGGATCCGTTATATGATCGG3′ which incorporate restriction sites ( underlined ) for PstI and BamHI , respectively to facilitate directional cloning into pLPBLP . For electroporation , 5×106 cells in 100 µL of ice cold buffer H-50 ( 20 mM HEPES , 50 mM KCl , 10 mM NaCl , 1 mM MgSO4 , 5 mM NaHCO3 , 1 mM NaH2PO4 ) was mixed with 10 µg of linearized gene-targeting DNA and electroporated twice , waiting for about 5 sec between pulses , using a Bio-Rad Gene Pulser ( 0 . 85 kv , 25 µF ) ( Bio-Rad , Hercules , CA ) [64] . The next day , the media was replaced with fresh HL-5 supplemented with 10 µg/ml blasticidin . Single colonies were collected and replica-plated into multiple 96-well plates . Genomic DNA was extracted exactly as described [65] , and targeted gene disruptions were identified initially by several PCR reactions using a combination of primers ( Table S1 ) . Clonal isolates from multiple independent mutants possessing the same disruption and sporulation phenotype were obtained . One of these clones ( httE13 ) was used in all subsequent experiments . Southern blot hybridizations using Bsr DIG-labeled probes were used for definitive confirmation of gene disruption . The Bsr fragment from pLPBLP was agarose gel purified and used to make a DIG-labeled probe using the DIG-High Prime starter kit from Roche . Hybridization was carried essentially as described in the DIG-DNA Detection manual ( Roche ) . Blots were washed at room temperature and then at 68°C , equilibrated in detection buffer , incubated in disodium 3- ( 4-methoxyspiro {l , 2-dioxetane-3 , 2′- ( 5′-chloro ) tricyclo[3 . 3 . 1 . 1]decan}-4-yl ) phenyl phosphate ( CSPD ) for 15 minutes at 37°C and developed using enhanced chemiluminescence XOMAT Kodak film . The analysis of total RNA for hd alternate splice variants was performed using exon-specific primers and RT-PCR . Total RNA was collected from vegetative cells and at 6-hour increments during development using the Qiagen RNeasy kit followed by on-column DNase digestion ( Qiagen ) . The hd transcript was detected using the Qiagen One-Step RT-PCR kit as per the manufacturer's recommendations using primer pairs listed in Table S2 . Growing cells were harvested and deposited into Lab-tek chambered cover glass ( 8 well ) at 1×105 cells/cm2 and allowed to grow overnight in HL-5 at 21°C . For F-actin staining , the media was aspirated and cells were fixed with 4% formaldehyde in PDF buffer ( 20 mM KCl , 11 mM K2HPO4 , 13 . 2 mM KH2PO4 , 1 mM CaCl2 , 2 . 5 mM MgSO4 , pH 6 . 4 ) at room temperature for 15 minutes , permeabilized with 0 . 025% TX-100 in PDF buffer for 15 minutes , followed by a brief wash in PDF buffer . Cellular F-actin was stained by incubating the fixed cells with Texas-Red-conjugated phalloidin ( Molecular Probes , Eugene , OR ) at a concentration suggested by the manufacturer for 25 minutes at room temperature and washed three times with PBS prior to viewing . To assess the effects of nutrient removal , the HL-5 was aspirated and replaced with KK2 buffer ( 16 . 5 mM KH2PO4 and 3 . 8 mM K2HPO4 ) . After starving cells for 1 hr , the cells were fixed and processed for F-actin staining as described above . Nuclei were stained with Hoechst 33342 . Images were taken on an inverted NIKON microscope TE2000 ( NIKON Instruments , Dallas , TX ) with either 4× , 20× , 40× objectives or a 63× 1 . 4 NA PlanFluor oil immersion objective and Quantix camera ( Roper Scientific , AZ ) controlled by NIS Elements and processed with Adobe Photoshop software ( Adobe , San Jose , CA ) . The viability of wild-type and hd− spores were assayed by harvesting approximately 2×108 cells and plating onto 10 cm non-nutrient 1 . 5% agar plates ( KK2 buffer pH 6 . 4 ) . Spores from mature fruiting bodies were harvested using sterile pipette tips containing 10 µL of spore buffer ( 40 mM KH2PO4 , 20 mM KCl , 2 . 5 mM MgCl2 ) , washed twice by centrifugation at 12 , 500 g for 2 minutes at room temperature and counted with a hemocytometer . Aliquots of 100 spores were heated to 45°C for 10 minutes , incubated with 0 . 5% NP40 detergent ( Sigma-Aldrich , St Louis , MO ) for 5 minutes or incubated with spore buffer alone for 5 minutes as a control . Spores were washed and plated in triplicate onto SM-5 agar plates in a suspension of bacteria ( either E . coli B/R or K . aerogenes ) and grown for 7 days at 21°C . The viability of spores was assessed by counting the number of clear plaques formed on the bacterial lawns for each treatment . Relative viability was measured as the percentage of hd− plaques formed compared to wild-type cells for all conditions . The submerged-monolayer assay was modified from an assay described previously [66] . For stalk cell induction , vegetative cells were washed once with KK2 buffer and three times with stalk buffer ( 10 mM morpholineethanesulfonic acid [MES] , 2 mM NaCl , 10 mM KCl , 1 mM CaCl2 , 5 mM cAMP and 200 µg of penicillin-streptomycin per ml , pH 6 . 2 ) . The cells were then plated into 24-well tissue culture plates at a density of 104 cells/cm2 and incubated in the presence or absence of 10–100 nM differentiation-inducing factor ( DIF-1 ) . A calcofluor solution ( 0 . 01% ) was added to the wells for 5 min . , removed and the cells were observed immediately by microscopy . Only the cells that were vacuolated and stained by calcofluor were counted as stalk cells [67] . For spore cell induction , wild-type cells and hd− cells were washed once with KK2 buffer and three times with spore buffer ( 10 mM MES , 20 mM NaCl , 20 mM KCl , 1 mM CaCl2 , 1 mM MgCl2 ) and then incubated in spore buffer supplemented with 5–15 mM 8-Br-cAMP for 24 and 48 hours [68] and visualized in bright field microscopy . Relative sporulation percentage was calculated as described for the stalk cell assay . Assays were performed three independent times , each with three replicates . In Dictyostelium , chimeras are readily produced by mixing cells of different genetic backgrounds and allowing them to co-aggregate to form a chimeric mound [69] . For chimera aggregation , development and tracing cell lineages , wild-type and hd− cells were transfected with pTX-GFP and selected in 10 µg/mL G418 . GFP-marked and unmarked cells were mixed at varying percentages prior to deposition under buffer or development on non-nutrient agar as described above . To examine spores under electron microscopy , greater than 2×108 wild-type and huntingtin-null spores were harvested and resuspended in 2% glutaraldehyde ( EM grade ) in spore buffer and incubated for 1–2 hours at 21°C . After this fixation , the spores were washed in spore buffer and processed for electron microscopy ( EM ) . Electron micrographs were prepared using the CHGR Microscopy Core . Dictyostelium expression vector pTX-GFP was digested with restriction enzymes SalI and KpnI , releasing its actin15 promoter and the first 11 codons of the ORF containing an 8× His tag [70] . A region of DNA sequence including 595 bp 5′ to hd ORF and the first 42 bp of the ORF was amplified from D . discoideum ( AX3 ) genomic DNA using forward primer DdHtt_F608+SalI ( 5′- AATATGTGTCGACCTACAGTTATTAAATAAATTGCAATAAAGGTGC-3′ ) and reverse primer DdHtt_R_Pro+KpnI ( 5′-TCCTCTGTGGTACCTGGTGATGCTGATAATATATCTAATCCACG-3′ ) . The product was digested with restriction enzymes SalI and KpnI to facilitate ligation into the vector upstream and in frame with the GFP ORF sequence . The resulting vector therefore expresses the first 14 amino acids of hd fused upstream of GFP , under the control of the predicted minimal hd promoter sequence . The vector was transformed into AX3 cells as previously described [71] . Cell-cell adhesion was assessed as previously described [72] . Briefly , cells were grown axenically to a density of 2–5×106 cells/mL , washed twice in ½ volume of ice cold Soerensen buffer ( SB ) , resuspended in 0 . 4 initial volume of SB and vortexed , then immediately counted in order to adjust the concentration of cells to 5×106 cells/mL . Cells were incubated in an Erlenmeyer flask at 150 rpm and 21°C and samples were collected at various time points over a period of 6 hours . For each collection point , cells were incubated in the presence or absence of 10 mM EDTA for 30 minutes and fixed with 2% glutaraldehyde . Single cells were counted using a hemocytometer . All experiments were performed in duplicate and the mean value for single cells in duplicate samples , expressed as percentage of total cells was plotted over time . Control ( wild-type ) or hd− cells were grown HL-5 and the medium was replaced with low ionic buffer ( KK2 ) , water ( hypotonic ) or 400 mM sorbitol in KK2 ( hyperosmotic ) . At various time points samples were taken , diluted into KK2 phosphate buffer containing K . aerogenes and plated onto SM agar plates to assay for viability . Cell lysis upon osmotic shock was also observed visually by brightfield microscopy . All experiments were performed in triplicate . | Genetic evidence in humans and mouse models of Huntington's disease suggests that the disease mutation confers a deleterious gain-of-function on huntingtin that acts through the deregulation of some aspect of the protein's normal function ( s ) . While huntingtin's function is poorly understood , its evolutionary conservation makes investigation of its physiological role in lower organisms an attractive route that has yet to be fully exploited . Therefore , we have used Dictyostelium discoideum to study the consequences of huntingtin ( hd ) deficiency . Developing Dictyostelium cells chemotax to form a multicellular slug that forms a fruiting body , comprising dormant spores encased above dead stalk cells . We found that hd− cells were hypersensitive to hypoosmotic stress . When starved , hd− cells aggregate by accretion , showed disorganized F-actin , and failed to form EDTA-resistant cell–cell contacts . Surprisingly , chemotactic signaling was rescued with Ca2+ or Mg2+ but not pulses of cAMP . Development of hd− mutants produced small fruiting bodies with round , defective spores , and when mixed with wild-type cells they didn't differentiate into spores . Our results are consistent with mammalian studies that show huntingtin is a multifunctional protein involved in many biochemical processes; and , importantly , they establish Dictyostelium as a valuable experimental organism for exploring in biochemical detail huntingtin's normal function ( s ) . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"cellular",
"structures",
"subcellular",
"organelles",
"cellular",
"stress",
"responses",
"mechanisms",
"of",
"signal",
"transduction",
"dictyosteliomycota",
"microbiology",
"signaling",
"in",
"selected",
"disciplines",
"adenylyl",
"cyclase",
"signaling",
"cascade",
"cell",
"differentiation",
"gene",
"function",
"camp",
"signaling",
"cascade",
"developmental",
"biology",
"model",
"organisms",
"developmental",
"signaling",
"cell",
"movement",
"signaling",
"cell",
"growth",
"molecular",
"development",
"molecular",
"genetics",
"cytoskeleton",
"dictyostelium",
"discoideum",
"signaling",
"in",
"cellular",
"processes",
"slime",
"molds",
"cell",
"adhesion",
"protozoan",
"models",
"signal",
"initiation",
"biology",
"molecular",
"biology",
"signal",
"transduction",
"pka",
"signaling",
"cascade",
"cell",
"biology",
"signaling",
"gene",
"identification",
"and",
"analysis",
"genetics",
"molecular",
"cell",
"biology",
"genetics",
"and",
"genomics",
"cell",
"fate",
"determination",
"signaling",
"cascades"
] | 2011 | Deficiency of Huntingtin Has Pleiotropic Effects in the Social Amoeba Dictyostelium discoideum |
Positive feedback plays a key role in the ability of signaling molecules to form highly localized clusters in the membrane or cytosol of cells . Such clustering can occur in the absence of localizing mechanisms such as pre-existing spatial cues , diffusional barriers , or molecular cross-linking . What prevents positive feedback from amplifying inevitable biological noise when an un-clustered “off” state is desired ? And , what limits the spread of clusters when an “on” state is desired ? Here , we show that a minimal positive feedback circuit provides the general principle for both suppressing and amplifying noise: below a critical density of signaling molecules , clustering switches off; above this threshold , highly localized clusters are recurrently generated . Clustering occurs only in the stochastic regime , suggesting that finite sizes of molecular populations cannot be ignored in signal transduction networks . The emergence of a dominant cluster for finite numbers of molecules is partly a phenomenon of random sampling , analogous to the fixation or loss of neutral mutations in finite populations . We refer to our model as the “neutral drift polarity model . ” Regulating the density of signaling molecules provides a simple mechanism for a positive feedback circuit to robustly switch between clustered and un-clustered states . The intrinsic ability of positive feedback both to create and suppress clustering is a general mechanism that could operate within diverse biological networks to create dynamic spatial organization .
The formation of local , high density regions of signaling molecules ( referred to below as “clusters” ) can switch cellular pathways between “off” and “on” states and direct downstream processes [1] . This transition may require careful regulation , particularly when an “on” state initiates large-scale cellular changes , such as observed in migration , cell division , or immune responses [2] , [3] , [4] , [5] , [6] . Experimental and theoretical studies have demonstrated that positive feedback plays a central role in pattern formation . Positive feedback can amplify and reinforce spatially asymmetric distributions of signaling molecules in single cells . This amplification , however , is indiscriminate; stochastic fluctuations could cause switches between “off” and “on” states to occur at undesired times , and sites of activation to occur in undesired locations [7] , [8] . Additional mechanisms may be combined with positive feedback for regulating pattern formation , including coupled inhibitors [9] , long-range negative feedback [10] , tight regulation of input noise [11] , or sequestration of components required for positive feedback [12] . Here , we wondered whether mechanisms existed within positive feedback circuits themselves to enable both the robust repression of noise required to maintain an “off” state , and the reliable establishment and persistence of distinct , high-density clusters of signaling molecules required to maintain an “on” state . First , it has been shown that positive feedback can attenuate the effects of noise . Previous studies have demonstrated that nonlinear models of positive feedback can give rise to bistable , temporal responses , which in turn set thresholds for activation below which an “off” state can be robustly maintained [13] . ( The coupling of multiple positive feedback loops can also act to robustly maintain an “on” state in the presence of noisy input [14] , [15] . ) However , these investigations were focused on temporal transitions between “off” and “on” states , and not on the emergence of spatial patterning . Second , it has been shown that positive feedback circuits can create clusters of signaling molecules through amplification of stochastic fluctuations [5] , [16] , [17] . In particular , discrete simulations of diffusing and interacting molecules [16] , motivated by activated GTPase Ras clustering on the cell membranes of lymphoid cells [6] , showed that positive feedback resulted in spatial clustering of slowly diffusing , activated molecules . In that model , clusters spread outward until the entire cell membrane was covered and the spatial patterning was lost to a homogeneous activated state . Another stochastic model [17] , motivated by eukaryotic gradient sensing , showed that patches of the phosphoinositide PIP3 could accumulate near activated receptors on the surface of a cell . A coarsening process then occurred with smaller patches eventually being absorbed into larger patches . While positive feedback was shown to initiate cluster nucleation and growth in these studies , mechanisms for buffering the onset of nucleation and limiting the spread of clusters were not considered . An important case of cluster formation is cell polarity , in which the formation of a single , asymmetric accumulation of signaling molecules , such as Rho GTPases , serves to define a unique cellular axis . Many previous theoretical studies [17] , [18] , [19] , [20] , [21] , [22] , [23] have provided insight into possible mechanisms by which a wide variety of eukaryotic cell types , including budding yeast , mammalian neutrophils , and amoeba can spontaneously polarize in the absence of spatial cues [2] , [24] , [25] . We previously considered a simple , positive feedback circuit , inspired by the ability of Cdc42 to polarize spontaneously in latrunculin-treated yeast [26] . In that model , molecules stochastically transitioned between inactive ( cytosolic ) or active ( membrane-bound ) states; and activated molecules , diffusing laterally along the membrane , recruited inactive molecules to their membrane locations . It was shown that polarity emerged from this positive feedback circuit for intermediate ranges of signaling molecule numbers . While stochastic events and diffusion eventually led to the dispersal of a cluster , at steady state the process was recurrent and a new site of polarity would eventually re-form on the membrane . In that study , the circuit operated by mass action for any fixed number of signaling molecules . However , for varying numbers of molecules , the strength of positive feedback was scaled to maintain a constant average fraction of signaling molecules on the membrane , so the circuit could not be in an “off” state . Hence , the repression of a clustered state , and transition from “off” to “on” state was not–and could not be–considered for varying numbers of molecules . Though , clustering could effectively be shut off by having so few total molecules that stochastic activation events rarely occur , or by varying other model parameters [26] . Here , in one unified model , we investigate the ability of positive feedback to reliably repress or create localized signaling domains . An essential difference between our previous and current models is that positive feedback now operates entirely through mass action kinetics ( i . e . rate constants are not rescaled by total numbers of signaling molecules ) . In principle , removing the constraint that held the average fraction of signaling molecules constant could potentially significantly alter emergent behavior . This is indeed the case ( Table 1 compares the present model to previous work , including our own , and indicates key differences in behaviors ) . In particular , we find that when the density of molecules is below an easily computable threshold , all signaling molecules are expected to be inactive; hence , no clusters of activated signaling molecules form , and cells are buffered against the onset of cluster formation regardless of the constant presence of noise . Above the threshold , increasing densities leads to increasing numbers of activated molecules . This process can be applied to many cell-biological settings , and we investigate clustering of molecules in the case of cell polarity , as well as for 2-D membranes or in 3-D volumes where the inactive and active forms of the signaling molecules are not segregated to spatially distinct compartments . Taken together , we find seemingly opposing effects for noise in this positive feedback circuit: at low densities of signaling molecules , biochemical noise is ignored in an “off” state; at intermediate densities , biochemical noise drives the formation of single , polarized clusters of signaling molecules to create an “on” state; and at high densities , biochemical noise overwhelms polarization to create a spatially homogeneous “on” state .
Here , we investigate emergent behaviors of a “minimal” positive feedback circuit based on mass action kinetics interactions between two states of a signaling molecule ( Figure 1A ) . In this conceptual model , a single molecular species spontaneously transitions between inactive and active signaling forms , while positive feedback allows activated molecules to recruit and activate nearby inactivated molecules . While many molecular networks containing positive feedback have been identified ( see Table 2 ) , detailed knowledge of their components and interactions is often incomplete . In our analysis , specific details of molecular mechanisms are elided to better focus on identifying fundamental properties of positive feedback that may be operating within diverse biological contexts . We first consider the situation where inactive or active forms of a molecule are exclusively associated with localization to the cytosol or membrane ( respectively ) ( Figure 1B ) . Here , our model is based on the following assumptions: We begin by discussing biological motivation for a two-state model of positive feedback . We next mathematically analyze the model described above . Finally , we discuss alternative cellular settings to the first assumption , in which the inactive and active forms intermingle in the cytosol or on the membrane , and the inactive form has a finite speed of diffusion . This positive feedback model is applicable to diverse biological systems . In a particular biological setting , the two states of the signaling molecules could be distinguished by many mechanisms , such as biochemical modifications ( e . g . phosphorylation , or GDP/GTP association ) and/or cellular localization ( e . g . membrane or cytosolic compartments ) ; exchange between these forms may be regulated by additional molecular components . For example , on the membrane , the activated small GTPase Ras is observed to form dynamic nanoclusters [27] , [28] . Ras activation via the Ras activator SOS ( Son of Sevenless ) has been demonstrated to contain a positive feedback loop [6] . In the nucleus , unphosphorylated splicing factors ( SFs ) self-organize into dynamic nuclear speckles . Speckle formation is modulated by the self-interaction ( binding ) of slow moving unphosphorylated SFs , whereas self-interaction is diminished in the fast-diffusing , phosphorylated state [29] , [30] , [31] , [32] . Finally , within a cell , clustering may involve molecules cycling between membrane and cytosol . Examples include myristoylated alanine-rich C kinase substrate ( MARCKS ) proteins [33] that colocalize with patches of PIP2 on the plasma membrane in their dephosphorylated form [34] . An important biological application for our model is provided by proteins involved in cell polarity , such as Cdc42 and other Rho family GTPases . Like other GTPases , Cdc42 cycles between an active GTP-bound form ( that is localized to the membrane ) and an inactive GDP-bound state ( that can be on the membrane or in the cytosol ) . GDI ( guanine dissociation inhibitor ) molecules extract the GDP-bound form of Rho family proteins from the membrane to the cytosol , where they are sequestered in an inactive pool . In the budding yeast , Saccharomyces cerevisiae , active Cdc42 localizes to a single zone on the plasma membrane marking the bud assembly site [2] . Dynamic Cdc42 GDP/GTP cycling is required for the polarization response in S . cerevisiae [35] , [36] , and continual exchange between membrane and cytosol is hypothesized to be essential for generating robust cell polarity [19] , [20] , [26] . Cycling of Cdc42 between the membrane and cytosol may be described by our two-state positive feedback circuit . First , active Cdc42 promotes the recruitment of more Cdc42 to the polarity zone , resulting in a positive feedback loop ( reviewed in [2] ) . Second , on the timescale of cell polarization , the amount of Cdc42 can be considered to be roughly constant . Third , the inactive cytosolic pool can be considered well mixed assuming: ( i ) the amount of GDI in the cell is not a limiting factor [37] , [38]; and ( ii ) switching between membrane and cytosolic states is rapid [39] . Then , we can use the fast exchange ( rapid equilibrium approximation ) between the membrane and cytosolic GDP-bound forms to derive a net “effective diffusion coefficient” [40] for the inactive forms . This coefficient is given by a weighted average of the membrane and cytosolic diffusion coefficients , for membrane association and dissociation rates and membrane and cytosolic diffusion coefficients and . The diffusion rates in the membrane are typically 100–1000× slower than in the cytosol [41] , resulting in . Thus , Cdc42 satisfies all three criteria of our positive feedback model . More generally , this simple two-state model could be used to approximate membrane/cytosol cycling of Rho GTPases or other signaling molecules , where an active membrane-bound form undergoes slow diffusion , while an inactive cytosolic pool is well mixed . To provide insight into emergent model behaviors , we made use of numerical simulations . We first simulated the positive feedback circuit for a 1-D circular membrane to facilitate easy visualization of signaling molecule behaviors in time and space . We quantified the total number of signaling molecules on the membrane as well as the frequency of polarization , determined by whether ≥50% of the molecules were contained within a contiguous region ( ranging from 15 to 25% ) of the membrane . To compare model behavior with our previous study of stochastic polarity [26] , we varied the total number of molecules , ( see Table 3 for model parameters ) . Cells were initially seeded with 10% of the molecules randomly position on the membrane . Consistent with previous findings , for large the distribution of activated signaling molecules was largely homogeneous while for intermediate self-organized clustering occurred . However , in contrast with our previous model ( Figure S1 ) , we observed a clear “off switch”: below a critical number of signaling molecules , all molecules were localized to the cytosol; above this critical number clustering occurred ( Figure 2A–B ) . To test the robustness of repression below this threshold , halfway through a numerical simulation we abruptly moved 50% of the molecules in the cytosolic pool to a small region covering 10% of the membrane , then restarted the simulation ( Figure 2C ) . Again , below the critical number , polarization was always immediately lost after restimulation , indicating that the maintenance , as well as the establishment of polarization is prevented . Finally , we varied the level of “input noise” to the system by varying the spontaneous on rate over 5 orders of magnitude . Throughout this range , a switch-like transition between no-clustering and clustering was observed ( Figure S2 ) . Thus , these simulations suggested that as increased the behaviors of the positive feedback model transitioned from the repression , to the emergence , and finally to the homogenization of polarization . With these observed behaviors as motivation , we next describe the mechanisms that underlie the behavior of this model in three steps . First we consider the time evolution of the number of particles in the cytosol , . Assuming that this number can be treated as a continuous variable , whose evolution is determined by a differential equation , we find switch-like behavior . Namely , when the density of particles in the cell lies below a certain critical value all particles remain in the cytosol , thus preventing polarization of the membrane; for larger values of the density a fraction of the particles will move to the membrane , enabling polarization . Second , we remove the assumption that is a continuous variable , and more accurately model the number of cytosolic molecules in terms of a stochastic process . We again find that switch-like behavior emerges ( Protocol S1 , Section 4 ) . Third , to determine the range of parameters in which polarization occurs , we extend the stochastic model to include the membrane-bound particle positions . Assuming that can be described by a continuous variable , our model dictates that its rate of change ( Figure 1C ) is given by: ( 1 ) We may also keep track of the density of molecules in the cytosol , . Rescaling time ( so that ) , equation ( 1 ) can be rewritten more simply as: ( 2 ) where the constants are given by: ( 3 ) The first constant , , is the total density of molecules in a cell; since some molecules may be membrane-bound , is the upper bound for the cytosolic density ( that is , ) . The second constant , , also has dimensions of density , and is dependent on parameters of the positive feedback system itself , but is independent of the cell's volume or the number of molecules it contains . The third constant , , is dimensionless and reflects a ratio of spontaneous on-to-off rates . As will be shown subsequently , these three constants play critical roles in determining when polarization can and cannot occur . Polarity cannot emerge when the spontaneous on rate is comparable to the off rate: high molecular flux due to undirected , spontaneous on-events will overwhelm the ability of positive feedback to create localized regions of high density [26] . Thus , we analyze the system behavior when the spontaneous on rate is small relative to the off rate , that is , when . By considering the steady states of equation ( 2 ) , we can understand why the “off” state is buffered from noise in the small molecule number regime . When there are no spontaneous activation events ( i . e . , , the evolution of the cytosolic density given by ( 2 ) reduces to: ( 4 ) The right hand side is a simple quadratic expression . Hence , at steady state , the cytosolic density is predicted to be at one of two values , or , defined in equation ( 3 ) . If , then all molecules are sequestered in the cytosol ( i . e . , ) . No molecules are available for the membrane , and clustering is repressed by default . If , then molecules are in the cytosol and the remaining molecules are on the membrane . Then , the circuit is in a permissive state for clustering , and whether or not clustering can occur is determined by other relationships among the parameters [26] . It follows from equation ( 4 ) that the smaller of these two steady states is stable , while the larger is unstable ( Figure 3A ) . What determines which of these steady states a cell will be in ? A key distinguishing factor is that depends on the molecule number whereas does not . This makes a natural parameter to vary , whose effects can easily be observed by comparing the ratio of the two steady state roots , ( analogous to the basic reproductive ratio [42] in theories of epidemiology , discussed later ) . As increases from small to large values , clustering goes from being repressed to being possible ( Figure 3A ) . Switching occurs at the “critical” density ( in a so-called transcritical bifurcation ) when the two roots are equal ( ) . ( Note that density-dependent switching is not observed if feedback is scaled to maintain a constant fraction of activated molecules [26]; the analogous ratio of roots is independent of . ) What other parameters of this positive feedback could cells modulate to regulate repression of clustering ? As can be seen from , decreasing the positive feedback rate or the recruiting volume , or increasing the membrane dissociation rate or the cell volume will expand the range of densities where clustering is repressed . Taken together , when molecular density is below an easily computable threshold , a cell will be in a repressed state for clustering . The “off” state can also be buffered when spontaneous activation events are allowed to occur ( i . e . ) ( Figure 3B ) . Equation ( 2 ) has two distinct steady states , given by solutions of ( 5 ) When spontaneous activation events are relatively infrequent , i . e . when is small , we expect the roots of equation ( 5 ) to be close to the steady states of equation ( 4 ) . For equation ( 5 ) , the smaller root is always smaller than and , and corresponds to a stable steady state , while the larger root is always larger than and , and corresponds to an unstable steady state . Because the larger root is always greater than , only the smaller root is physically relevant ( there cannot be more particles than ) . There is no exchange of stability , and equation ( 2 ) has a unique , stable steady state . However , when is small , the smaller root still exhibits switch-like behavior near ( due to the proximity of the bifurcation point; this phenomenon is described in bifurcation theory as an imperfect transcritical bifurcation with as imperfection parameter [43] ) . Of course , the actual number of molecules in any given cell is finite . It is known that stochastic fluctuations may drive cellular behavior to new dynamic states not seen in deterministic models [26] , [44] , [45] , [46] . We next investigate whether the switching behavior shown in the continuous setting would also hold in a stochastic setting . A more detailed description of the probabilities for the time evolution of the number of molecules in the cytosol is given in terms of a one-step continuous time stochastic process [47] . In a stochastic process , the steady state is described by its stationary distribution . This distribution specifies the probability , , that the cytosolic pool contains exactly molecules for a randomly chosen cell from a large ensemble of cells , or for one cell inspected at a randomly chosen time from a sufficiently long time interval . The stationary distribution is obtained by solving the master equation ( Protocol S1 , Section 4 and [47] ) . The switching in the preceding continuum approximation also appears in the stochastic analysis when the ratio of on-to-feedback rates , , is small . We find that the stationary distributions undergo a qualitative change as increases above . More precisely , when all probability density centers around , while for the stationary distribution is essentially a Poisson distribution with expectation ( Figure 3C ) . Interestingly , the stationary distribution shows bimodality near the transition point ( Figure 3C , inset ) while the deterministic solution is unimodal . The ability of stochasticity to induce bimodality to a deterministic mass action equation was also recently reported for cellular signaling in phosphorylation-dephosphorylation cycles [45] . As we discuss next , the ratio cannot be too large if clustering in the “on” state is also desired . The preceding analysis focused on the overall numbers of molecules in inactive or active states . We next examine the spatial distribution of the active signaling molecules . We perform this analysis in the stochastic setting , as the continuous setting modeled by partial differential equations leads to a homogeneous steady state ( Protocol S1 , Section 6 ) . The basis for stochastic cluster formation , previously described [26] , [48] , remains valid in this current work for any specified set of parameters . However , the feedback reaction rate in the present work depends differently on and , as a consequence , the range of parameters that permit polarization are changed . Here , we provide a new approach for analyzing the mechanism responsible for the emergence of clusters and calculate the parameter ranges in which polarization is possible ( Protocol S1 , Section 5 ) . The phenomenon of polarization is defined by the property that a large fraction of the membrane-bound signaling molecules cluster within one small region of the membrane . In general , molecules will be distributed unevenly on the membrane due to stochastic fluctuations . Recruitment and disassociation will not deterministically amplify such asymmetries; at equilibrium , each membrane-bound signaling molecule will recruit or disassociate with equal probabilities causing these effects to cancel at first approximation . However , the stochastic nature of the process can cause imbalances in the molecular distribution to undergo a neutral drift in which eventually a small region of the membrane contains most molecules while the remainder is largely depleted . When is small , on-events are infrequent and we can analyze the clustering mechanism in between on-events by grouping the membrane-bound signaling molecules into “clans” and tracking their genealogy [26] . Initially the clans are defined by dividing the membrane into a large number of small regions and declaring all molecules in any such region to form one clan . Clan genealogy is defined by assigning each newly recruited molecule to the clan of its recruiter , and by erasing the clan identity of any spontaneously dissociating molecule . As time progresses , clans will shrink and grow in population size , but once a clan has lost its last member it becomes extinct and cannot return . As long as no on-events occur , the number of clans cannot increase . Our analysis shows that if the time interval between two on-events is sufficiently long then , the expected time in which only half the original clans survive is ( Protocol S1 , Section 7 ) ( 6 ) If the membrane is initially partitioned into clans , then after time ( 7 ) only one clan is expected to remain ( Figure 4A ) . A range of parameters for which single clans emerge within localized domains can be computed explicitly , even when a small number of on-events are allowed ( Protocol S1 , Section 5 ) . First , the frequency of on-events must be low enough that there is enough time to allow all clans but one to become extinct before the on-events significantly contribute to a new fraction of membrane-bound population of molecules . If ( 8 ) then all but a ( small ) fraction ( e . g . ) of the total molecules on the membrane will belong to a single clan . Second , if membrane diffusion is slow enough then all members of the surviving clan will be located in a small neighborhood of the site of its ancestral clan ( Figure 4B ) . This will be true if the number of molecules is bounded by ( 9 ) where is the membrane diffusion constant and the constant depends on the size of the initial regions ( Protocol S1 , Section 5 ) . Taken together , estimates ( 8 ) and ( 9 ) indicate when membrane-bound molecules will have redistributed with high probability to form a single localized cluster ( Figure 4C ) . Note that equation ( 8 ) is a conservative estimate; polarization was observed even when was above this bound ( Figure S2 ) . Finally , we tested applications of this positive feedback circuit to generate molecular clusters in 2-D or 3-D cellular settings , motivated by possible applications of our model framework to clustering on membranes [49] and in the nucleus [50] . We used the freely available spatial stochastic particle simulator Smoldyn version 2 . 15 [51] to implement an alternative version of this circuit ( Figure 5A; see Protocol S1 , Section 8 for differences ) . In particular , we removed the assumption that the inactive form exists in a spatially homogenous pool , and assumed a finite rate of diffusion , , for the inactive molecules , which was still faster than the diffusion rate of the active molecules ( i . e . , ) . We tested the model in three different biologically motivated spatial settings , in which: ( 1 ) active molecules diffuse on the surface of a sphere , and recruit inactive forms from its interior; ( 2 ) active and inactive molecules both diffuse within the same 2D compartment , such as plasma membrane; and ( 3 ) active and inactive forms both diffusing within the same 3D volume , such as within the nucleus . In all settings , we observed transitions from a buffered “off” state , to one or several localized clusters , to a homogeneous “on” state as the number of molecules increased ( Figure 5B and Video S1 ) . For a fixed , intermediate number of molecules we observed competition between clans , until a single recurrent cluster remained ( Video S2 ) . Adjusting for the dimensions of the model parameters and ( see Protocol S1 , Section 8 ) , the transitions for this alternative model ( Figure 5C and Figure S3 ) were in close agreement with the analytically computed phase plane .
It has long been appreciated that positive feedback plays a key role in intracellular signaling [13] , [52] and , in particular , the ability of molecules to self-organize into highly localized clusters in the membrane or cytosol of cells . Positive feedback can cause clustering to occur in the absence of localizing mechanisms such as pre-existing spatial cues ( e . g . chemoattractants ) , diffusional barriers ( e . g . septins at the base of the primary cillium ) or molecular cross-linking . What prevents positive feedback from amplifying inevitable biological noise when an un-clustered “off” state is desired ? And , what limits the spread of clusters when an “on” state is desired ? In theory many additional mechanisms could be postulated . Here , we find that a minimal model of a positive feedback circuit has the intrinsic ability both to suppress and amplify noise: below a critical number of signaling molecules , clustering switches off; above this threshold , highly localized clusters are recurrently generated . Interestingly , positive feedback only produces spatially localized clusters in the stochastic regime , when one assumes a finite number of molecules and the presence of biological noise . In a continuum limit positive feedback alone is not sufficient for pattern formation , and many reaction-diffusion models have shown the need for additional mechanisms , such as long range negative feedback or substrate depletion [9] , [10] . The loss of small clusters and emergence of a dominant cluster for finite numbers of molecules is partly a phenomenon of random sampling , and is somewhat analogous to the fixation or loss of neutral mutations in finite populations [53] . That clustering may not be observable in the continuous limit [26] suggests , when analyzing signal transduction networks , finite sizes of populations cannot be ignored . In homage to classic work in population genetics , we refer to our model as the “neutral drift polarity model . ” Similar threshold behaviors in autocatalytic processes appear in diverse settings [54] . Interestingly , our minimal model of positive feedback can also be recast as a well-studied mathematical model for the spread of an epidemic . In this setting: the cytosolic molecules correspond to susceptible individuals ( S ) ; membrane-bound molecules correspond to infectious individuals ( I ) ; and the recruitment of new molecules by membrane-bound molecules corresponds to the spread of infection when an infected and a susceptible individual come in contact . Collectively , these interactions are referred to as an SIS model , and form a system of equations similar to ours [55] . The dimensionless parameter , described earlier , can be interpreted as the expected number of susceptible individuals that can become infected though contact with an infected individual [42] . Our results are analogous to the property that when the spread of infection is repressed , whereas for the disease is endemic in the population . The differences between the deterministic and the stochastic SIS model in a spatially homogeneous setting have been well-characterized [56] , [57] , [58] . In particular , the endemic steady state is only quasi-stationary , as is an absorbing steady state ( once there are no infective individuals left , there is no new source of infection ) , and stochastic fluctuations always result in eventual disease extinction [59] . Our positive feedback model differs from the SIS model in the inclusion of spontaneous transitions from the susceptible ( inactive ) pool to the infectious ( active ) pool . This significantly changes the long-term behavior of the system . Our results suggest that by including the effect of variable spread rates for susceptible and infectious individuals , introduction of new infections , and finite population sizes , parameter regimes exist in the SIS model where recurrent spatial patches of infected individuals can occur . The neutral drift polarity model considered in this paper is a simple conceptual model that encapsulates the mechanism for particle clustering . Although this model captures the generic features of the emergence of cell polarity , and can be mathematically analyzed both in the deterministic and the stochastic regime , simplifying assumptions were made for mathematical tractability . First , in our theoretical treatment we assumed that the inactive forms are well-mixed throughout the cell . This assumption was weakened for the Smoldyn implementation , where a high but finite rate of diffusion was assumed for the inactive molecules . Second , the reaction volumes and are assumed to be constants , so that the spontaneous-on and feedback rates , and ( respectively ) , are independent of the density of membrane-bound molecules . This may not be a reasonable approximation for regions of high molecule density when the reaction volumes frequently overlap and mass-action kinetics no longer apply . Third , recruitment of inactive molecules to the membrane in our circuit is modeled by simple mass action between active and inactive forms . In reality , positive feedback loops are more complicated and can involve additional molecular components . For example , in budding yeast , active Cdc42 recruits the adaptor protein Bem1 , which in turn recruits/activates the Cdc42 GEF ( guanine nucleotide exchange factor ) Cdc24 [2] . Several biological predictions come out of our work . First , the ability to switch this positive feedback circuit on and off suggests that it could be placed as a primer , upstream of other signaling circuits , to initiate subsequent physiological processes . Such a role has been proposed in the context of Cdc42 polarization in yeast , in which a cytoskeleton-independent positive feedback circuit–with some similar features to the one studied here–acts as a primer for a second actin-dependent positive feedback circuit [2] , [36] . Second , our results suggest that over- or under-expression of a reporter probe to monitor a feedback system can have the unintended effect of eliminating the spatial organization that it was intended to observe . Third , our numerical simulations in 2D and 3D suggest that a positive feedback provides a “minimal” model for repressing or initiating molecular aggregation and microdomain formation , such as observed on lipid membranes [1] , in the nucleus [50] and in cytosolic puncta [4] . Fourth , our results provide a natural interpretation of ( and prediction for ) heterogeneity of cells in clustered states by providing a link between numbers of signaling molecules per cell and probabilities of observing off/on states or cluster formation within the population . Our work points to the intrinsic ability of positive feedback to give rise to spatial clustering . We propose that a positive feedback circuit , operating in the stochastic regime , can create a robust switch that can prevent spurious activation in an “off” state , and can be switched “on” or “off” by simply varying molecular density . As positive feedback loops form a common motif in many signal transduction networks , our work reveals a design principle based on neutral drift dynamics that may lie at the heart of diverse network functions . Additional mechanisms could be coupled to this basic positive feedback module to fine-tune the ability of biological systems to create sharp localized clusters . Finally , the discrete nature of molecular processes means that there can be significant fluctuations from mean behavior described by deterministic models , and stochastic models will be required to capture those effects .
Derivations of estimates and formulas used in Figures 1–4 are given in the Main Text and Protocol S1 . In previous work [26] , feedback was scaled to maintain a constant fraction of membrane molecules regardless of the total number of molecules . The relationship between the previous constants to the current constants is as follows: . Simulations in Figure 4 were performed using Matlab version R2009a on a unit as previously described [26] . Parameter values are as shown in Table 3 . All simulations in Figure 5 were performed using the stochastic particle simulator Smoldyn version 2 . 15 . The algorithm for bimolecular reactions in Smoldyn is based on the Smoluchowsky theory of diffusion-limited chemical reactions [51] . We note that the model implemented in Smoldyn differs from the theoretical treatment of the positive feedback circuit in several ways ( see Protocol S1 , Section 8 ) . Parameter values and Smoldyn code for Figure 5 is given in Protocol S1 . | A large body of work has focused on the ability of positive feedback in biological networks to create either switches in time ( i . e . , cells are either in an “on” or an “off” state ) or form patterns in space ( i . e . , spatial organization in cells and tissues ) . Here , we propose a stochastic “neutral drift polarity model” by which positive feedback alone is sufficient to create switch-like behaviors both in time and space for finite molecule numbers . Our theory predicts that below a critical density of signaling molecules , positive feedback robustly maintains an off state; exceeding this threshold switches on the recurrent emergence of highly localized signaling clusters . Cluster formation requires only this minimal positive feedback circuit , and does not require additional mechanisms such as diffusion barriers , spatial cues , or biochemical inhibitors . This mechanism is general , and could be applied to a variety of cellular signaling systems to create clusters in the membrane , cytosol , or organelles . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"systems",
"biology",
"mathematics",
"theoretical",
"biology",
"biology",
"genetics",
"and",
"genomics"
] | 2011 | A Density-Dependent Switch Drives Stochastic Clustering and Polarization of Signaling Molecules |
Competence is a transiently differentiated state that certain bacterial cells reach when faced with a stressful environment . Entrance into competence can be attributed to the excitability of the dynamics governing the genetic circuit that regulates this cellular behavior . Like many biological behaviors , entrance into competence is a stochastic event . In this case cellular noise is responsible for driving the cell from a vegetative state into competence and back . In this work we present a novel numerical method for the analysis of stochastic biochemical events and use it to study the excitable dynamics responsible for competence in Bacillus subtilis . Starting with a Finite State Projection ( FSP ) solution of the chemical master equation ( CME ) , we develop efficient numerical tools for accurately computing competence probability . Additionally , we propose a new approach for the sensitivity analysis of stochastic events and utilize it to elucidate the robustness properties of the competence regulatory genetic circuit . We also propose and implement a numerical method to calculate the expected time it takes a cell to return from competence . Although this study is focused on an example of cell-differentiation in Bacillus subtilis , our approach can be applied to a wide range of stochastic phenomena in biological systems .
Competence is the ability of a cell , usually a bacterium , to bind and internalize transforming exogenous DNA . Under stressful environments , such as nutrient limitations , some cells enter competence while other cells commit irreversibly to sporulation . Entry in competence is a transient probabilistic event that facilitates copying of the exogenous DNA [1] , [2] . It has been shown that among a group of cells only a randomly chosen fraction enters in competence [3] , [4] . Proper modeling and correctly accounting for noise in the model of this phenomenon is crucial to understanding the underlying biological explanation . The few cells that enter competence express a high concentration of the key regulator ComK , which activates hundreds of genes , including the genes encoding the DNA-uptake and recombination systems [5]–[7] . Competence is understood as a bistability pattern [4] , [8] and the nonlinear system describing the competence regulatory circuit is an excitable dynamical system . Auto-activation of the regulator ComK is responsible for the bistable response in competence development . Auto-activation of ComK , is essential and can be sufficient to generate a bistable expression pattern [9]–[11] . Specifically , the concentration of an inducer must cross a certain threshold to start the positive feedback . Different experimental studies concluded that an auto activation of ComK is the only needed factor for bistability to occur in the expression of this protein [9] , [11] , [12] . In [9] , Smits et . al discuss the factors that determine the required threshold for the activation of ComK and deduce that other transcription factors can raise or lower the threshold . Although many proteins are involved in the regulation of competence , there are two main proteins that play a major role . Süel et al . [13] propose a deterministic model driven by an additive noise to describe the dynamics of competence regulation . We use the reduced order Stochastic Differential Equation model ( SDE ) presented in [13] to develop a discrete stochastic model for competence . Calculating the probability and the expected time for entering and returning from competence , requires solving for the splitting probabilities and the first moment passage time . The problem of calculating the first passage time has been studied heavily in the literature for the stochastic difference equations , Fokker Planck equations and some special cases of the CME ( separable kernels or single specie ) . For a detailed treatment of this topic see [14]–[17] and references therein . Researchers usually use Monte-Carlo simulations to calculate the distribution of the first passage time when working with he CME ( e . g . see [18] and references therein ) . We propose in this work , an alternative approach that makes it possible to calculate the states in which the system will be as time evolves . The main idea here is to aggregate regions of the state space over which specie evolve into absorbing states . This technique is useful in analytically computing the distribution of the first passage time , by providing a way to deal with the infinite dimension of the state space over which the system evolves . The contributions of this paper are threefold . First , it provides a new method to calculate exact probabilities of biological phenomena where transient behaviors such as competence , which is the topic we chose to study here , occur . Second , it shows how to calculate sensitivities of the probabilities of passing to the transient state with respect to the system's parameters . Third , it gives a methodology to calculate the expected time that it takes a cell to return from its transient state . All these methods can be used to analyze any biological system that has the characteristic of switching between two states , while staying for a while in the unstable state . In this paper we start by describing the chemical reactions and the deterministic model . We then generate the Chemical Master Equation ( CME ) of our proposed discrete stochastic model . The CME characterizes the evolution of the probability density of the different discrete states . We simulate it using the Stochastic Simulation Algorithm ( SSA ) and show how the solution can be approximated using the Finite State Projection method ( FSP ) . We then conduct a sensitivity analysis studying the effect that the various system parameters have on the probability with which a cell enters in competence . This analysis shows the usefulness of our proposed numerical method in analyzing the roles of the different affinity , transcription and degradation rates , etc . , in driving the cellular switching ( between competence and vegetative states in this case ) . Finally , we analyze the roles of these parameters in determining the expected time a cell stays in competence .
Competence is a physiological state that enables cells to bind and internalize transforming DNA . This state is accompanied by blockage of the essential cell's functions , and since this state is driven by the transcriptional factor ComK , it is no surprise that ComK synthesis is subject to a number of finely tuned regulatory circuits [19] . The gene regulatory model for competence has been presented and described in [13] . Entrance of a cell in competence is controlled by a set of molecular interactions . Initially ComK and ComS are present in the cell at basal levels . The transcriptional factor ComK activates its own expression through positive feedback . The MecA complex is a multiprotein assembly that includes the ClpP-ClpC proteases . Bound to MecA , ComK is degraded under the action of the ClpP-ClpC proteases . In stressful environments , the level of ComS is high and that favors entrance into competence since ComS competes with ComK to bind to MecA . Inhibition of the binding of ComK to MecA by competitive binding with MecA-ComS allows a higher number of free ComK molecules to be present , which finally triggers the positive feedback that further raises the number of ComK molecules driving the cell in competence . This rise in the number of ComK is specific to competence . Once the number of ComK molecules reaches a certain level , it acts as an inhibitor for ComS through repression . The increase in the level of ComK will also favor the binding of MecA-ComK complex which degrades ComK through the ClpP-ClpC proteases , starting the return from competence . At this point ComS is below its basal level because of the aforementioned repression from ComK . The level of ComK starts to decrease by degradation . The degradation has two effects: ( 1 ) the decrease in the level of ComK will affect the transcriptional auto regulatory positive feedback loop of ComK , and ( 2 ) the absence of ComK in high levels , favors the synthesis of ComS by releasing the ComK-mediated ComS repression . This continues until the cell eventually exits the state of competence . The above mentioned molecular interactions are described [13] by the following chemical reactions:The rate equations describing the dynamics of the molecular reactions between the species model are the following: ( 1 ) ( 2 ) ( 3 ) ( 4 ) where K , S , , and are the concentrations of ComK , ComS , MecA , MecA-ComK and MecA-ComS respectively . We give in Table 1 the values and the description of each of the parameters in Eq . 1–4 . If one further assumes that the reactions of degradation of and are much faster than the other reactions , and can then be eliminated through time scale separation [13] , [20] and the conservation law:giving the following reduced model for the dynamics of competence: ( 5 ) ( 6 ) whereand In their paper Süel et al . [13] analyze the excitable dynamical system described above . They present a phase diagram where they study the nullclines and the vector field of the dynamical system . Their analysis gives insight about the vegetative and competent states analyzed in this work . As we already stated , under the same conditions some cells enter into competence while other cells do not . Entry in competence is a random event , and in order to properly model the cell's behavior , we need to include the effect of noise on the dynamics of competence . In their analysis Süel et al . [13] account for stochasticity by adding white gaussian noise terms in Eq . 6 . This drives the excitable dynamical system presented in Eq . 5–6 into long excursions when the noise magnitude is large enough . These long excursions correspond to a high level of ComK indicating entry into a state of competence . The problem with this approach is that reaching a competent state is highly dependent on the magnitude of the additive noise . The dynamics of the system in Eq . 5–6 are such that if the initial number of molecules of ComK and ComS is in the neighborhood of the fixed point of the dynamical system described in Eq . 5–6 , the number of molecules for both species will stay in the vicinity of that point without taking long excursions . If on the other hand , the number of molecules is driven beyond a threshold , the dynamical system in Eq . 5–6 will have a totally different behavior . The number of molecules of ComK will increase significantly because of ComK auto-activation through positive feedback; In other words , the cells will enter in competence . Here we would like to analyze the stochastic behavior of the dynamics of the competence regulatory circuit taking into account the internal noise in the environment of the cell without having a direct control on the magnitude of the noise driving the regulatory circuit . To do so , we model the stochasticity in the chemical reactions using the CME . We look at the problem at the molecular level and propose four reactions to model the system in Eq . 5–6 . The four reactions are: ( 7 ) with the following reaction rates:These reactions will serve as the starting point for developing and simulating a discrete stochastic model for competence in the next section . In order to compute the probability of entering into competence we use the CME to describe the stochastic chemical kinetics . Once we derive the CME , we simulate it using the Monte-Carlo based SSA . We then use the FSP method to obtain a finite dimensional solution to the infinite dimensional CME . In the CME , the state vectors indicate the number of molecules of each of the two species of interest: ComK and ComS . The CME describes the evolution of the probability that the number of molecules of each of the species has a given value . The dynamics of the evolution of the probability density vector are directly related to the chemical reactions . Starting from a number of molecules , the probability of being at molecules at time has the following dynamics: ( 8 ) where is the propensity vector and it represents the change that reaction will have on the number of molecules of each of the species . For example reaction increases ComK by one molecule and leaves the number of molecules of ComS unchanged so the propensity vector is , denotes the probability that the reaction will occur in the next infinitesimal time interval . Written in vector form , the CME becomes ( 9 ) where corresponds to the number of reactions that the species would go through . Let be a vector of the possible states of the system . Let be the corresponding vector of probabilities of the states in computed at time . evolves according to the equation ( 10 ) In general , may be infinite , resulting in an infinite dimensional system . Getting the exact value for the solution to the CME is not generally an easy task . In this part we introduce the SSA that is normally used to simulate Eq . 9 . The SSA is a Monte-Carlo based algorithm that generates sample paths for the underlying stochastic process . Gillespie introduced this algorithm in 1977 [21] . Reactions are modeled as a random event whose occurrence depends in a non linear manner on the number of molecules through the reaction rates . The algorithm can be summarized as follows: What we described above is a a basic summary of the algorithm , interested readers are referred to [21] for more details . The CME derived in Eq . 9 describes the evolution of the probability density vector of the number of molecules . Using SSA to get an estimate of the probability of entering into competence is easy to implement . However , a large number of simulations is required for a reasonably accurate estimate to be obtained . Aside from being time consuming , the algorithm has the drawback of lacking an accurate bound on the estimation error . In addition , analyzing the effect that different parameters have on the probability with which a cell enters in competence , requires the repetition of a large number of SSA simulations while changing those parameters of interest . This is numerically very costly . An alternative method in dealing with the CME is to compute an analytical expression for the probability of being in each state . The FSP method introduced in [22] provides a way to compute these probabilities . The probability density vector described in Eq . 9 allows molecules to evolve on an infinite lattice ( Fig . 1 ) and therefore gives an infinite dimensional system . The idea behind FSP is to choose a suitable subset of the lattice in which one retains all the states and chemical reactions ( transitions ) found in the original system , while aggregating the remaining states in the lattice into one absorbing state . Transitions that drive the states outside the region are retained , while those that allow return to the selected finite region are deleted ( see Fig . 1 for illustration ) . The finite state projection method gives the probability of being at any of the states inside the specified region at any point in time [22] . In this problem we are interested in finding the probability with which the pair ( ComK , ComS ) enters a region corresponding to the cell entering a state of competence . The sum of the probabilities of a cell being in and of the probability of being anywhere else in the state space has to equal one at all times . Moreover , if we divide the state space of the two proteins ComK and ComS in two regions , then the probability of the cell being inside the first region without ever leaving it , and the probability of leaving the first region once within a time should sum to one . These properties make FSP a very well suited numerical method to solve our problem . In the finite model all the states outside the projection region are aggregated into one absorbing state: ( see Fig . 1 for an illustration ) . The probability vector at time is given as in Eq . 10 by ( 11 ) where is an infinite matrix and is the initial distribution of the probabilities , that is a vector with infinite entries , where each entry corresponds to a probability with which the system starts with a given number of molecules . Using FSP we can project the infinite system in Eq . 11 into the following finite system: ( 12 ) In this case , A becomes a finite matrix , and is the finite vector of projected states . We build the finite matrix as followswhere and are the terms appearing in Eq . 8 . If denotes the underlying stochastic process , gives the probability of being in any of the states listed in during the time , conditioned on the event of never leaving the inside region for any time . We can rewrite the probability as the conditional probability ( 13 ) where is the state to which the outside region is aggregated . Remember that is an absorbing state . The probability of being inside the region without ever leaving it during the interval and the probability of visiting once should sum to one . Therefore ( 14 ) Eq . 14 gives the probability of entering the region at least once within a time . The boundary of the region that is aggregated into the absorbing state , is chosen to include the states with a high number of ComK molecules . This indicates that the systems reaching the absorbing state corresponds to the cell being in a state of competence . Denoting by and by it can be seen that the probability of competence at time , , is given by ( 15 ) where is chosen so that the columns of the state transition matrix add up exactly to zero . One advantage of having an analytical solution of the probability of competence is that we can use the solution to run a sensitivity analysis with respect to different model parameters . This makes it possible to shed light on the importance and roles that the different parts of the regulatory circuit play in reaching competence . We start this section by introducing the equations we used to compute the sensitivity for the probability with respect to all the parameters . We then compare answers obtained by this method to estimates of sensitivities that we obtained using a finite difference method . Recall that and suppose that we are interested in looking at the sensitivity of with respect to a parameter , which could be any of the parameters presented in Table 1 . The entry in is given by , where , is an vector with in the entry and zero everywhere else . We have from Eq . 9 that . Letting take values in the set of parameters , and using the fact that , we get that where is defined to be . Similar equations hold for . The sensitivity of the probability with which a cell enters in competence evolves according to the following dynamical system: ( 16 ) Solving the above linear system , we obtain the sensitivity of the exit probability to all the parameters . We evaluate the solution at the nominal values given in Table 1 . The results are reported in Table 2 . For comparison , we calculated the same terms computed above by using a finite difference method . The sensitivity of the probability of entering competence with respect to the various parameters is calculated according to the formula , where denotes the normalized sensitivity and denotes the nominal value of the parameter of interest . In order to change study the sensitivity to each parameter , we update the value with small steps using the equation below ( 17 ) In summary , the sensitivity results presented in Table 2 are calculated using two different methods: Method #1: We solve the double order system in Eq . 16 . This results in more accurate answers but is more computationally expensive . Method #2: We use the solutions for the original system describing the evolution of the probabilities of the states presented in Eq . 15 in addition to the numerical approximation method presented in Eq . 17 with . This method is less accurate than the first but is considerably faster to implement . We study here the time it takes for a cell to return from a state of competence to its original vegetative state . We use once again the analytical solution of the CME to conduct this analysis . We use a similar concept to the one explained earlier , with the difference that in this case , we aggregate into an absorbing state the region of the state space that corresponds to the vegetative state , indicating that the cell returned from competence . We also assume that the cell starts from a state of competence and that it is allowed to return from that state , i . e . , competent states are no longer absorbing in this case . Starting from competence corresponds to starting from a pair ( ComK , ComS ) that falls anywhere in the region . We assume that the cell can be at any state in equally likely . This assumption translates to setting the initial probability vector in a way that gives equal probability to all the states in . Return from competence is mapped to the region defined by . We set the initial probability vector to take the value at the entries corresponding to the states in and zero everywhere else . Here is the cardinality of the competence region in . Having defined a region to be the region in the state space corresponding to return from competence , we aggregate all the states of return from competence into one absorbing state . Hence , for the purpose of this calculation , once a trajectory ‘returns’ from competence , it cannot go back to it . Having described the dynamics of the probability for return from competence in a similar manner to the description we had presented for the probability of entering in competence , we find the probability of returning from competence as a function of time by solving a set of differential equation just like we did earlier . We still need to deal with the infinite dimensions of the original model . For this purpose we add another absorbing state . This state is an aggregation of the region outside the finite state space that we consider , , into a single state . The finite state space is chosen so that the probability of reaching in the time interval of interest remains small . This small probability gives an upper bound on the approximation error due to the reduction of the infinite system into a finite one , as can be seen in the FSP algorithm [22] . Define to be the probability of returning from competence within . Denote by , the probability of returning from competence at time , and by , the probability of exiting to the outside region at time . The system becomes: ( 18 ) Now consider a partition of the interval as follows:We can approximate the expected value of return time as follows: ( 19 )
Now that we presented the SSA , and FSP method , we first use the SSA algorithm to compare the reduced model in Eq . 5–6 to the full model in Eq . 1–4 both presented in [13] . In order to do this , we simulate both models using SSA and compare the probability of entering in competence as the parameters presented in Table 1 were changed . We show the results for the parameters , and for demonstration purposes , but we note that the behavior of the full and reduced model were very close for all the parameters . We then compare the results given by the SSA and the FSP method , when applied to the reduced model . We show in Figs . 5 and 6 , these results . We show next the insights our numerical methods allowed us to have about how the molecules involved in competence , affect the time a cell spends in this state . In Fig . 7 we show how changing the parameter affects the time a cell stays in competence . This parameter corresponds to the saturation expression rate of the ComK positive feedback . The plot shows results obtained by both FSP and SSA . We can see that the plots exhibit similar behaviors , keeping in mind that such a calculation requires a lot more SSA simulations . In addition to giving more accurate results , the FSP approach allows us to combine multiple points from which we consider the cell as being in competence , while a different set of SSA simulations should be run for each different initial condition ( starting number of molecules ) . Combining initial conditions is extremely useful in this case , since we care more about regions that the states go through than about specific points . It is not crucial to know the specific number of molecules of ComK or ComS when the cell is entering and returning from competence . We saw earlier that increasing will increase the probability of cells entering competence . We now know that it will also keep the cell in competence for a longer time . Competence is an exhausting but occasionally necessary state for the cell . In this work we develop the CME accounting properly for the internal noise driving the competence switching dynamical system . The stochastic behavior of cell switching to competence has been studied in the literature . For example in their work , Süel et al . [13] account for the stochasticity by introducing an additive noise term to their model . The intensity of the noise and its distribution were parameters that are determined by the authors . In this work , we accounted for noise in its natural intrinsic form , eliminating therefore any controlled excitation of the excitable system . We applied FSP to come up with an analytical solution , whereas other researchers always reverted to Monte-Carlo simulations , in their analysis . Finding an analytical solution made it possible for us to describe to a great extent the role of each of the molecules in driving cells into and out of competence . We discuss our results below . We start by addressing the roles of the different expression and degradation rates in a cell entering competence . Fig . 5 shows that an increase in the saturating expression rate of ComK positive feedback increases the probability of entering in competence . Fig . 7 also shows that it makes returning from competence slower . Although Figs . 5 and 6 show that ComK and ComS have similar roles in driving a cell into and back from competence , Table 2 suggests that changes in ComS affected by the values of the expression and degradation rates of ComS affect the probability of entering and staying in competence more than changes in ComK affected by the values of . This leads to the expectation that the genetic circuits controlling ComS levels need to be much more sophisticated and complex than those regulating ComK in order to keep ComS concentration at specific values . Our normalized sensitivity analysis showed that increasing the basal expression rate and the saturating expression rate of ComK has an almost canceling effect to increasing the degradation rate of Comk as far as the probability of entering in competence is concerned . It also showed that the expression and degradation rates of ComS , had a similar canceling effect . This means that each of these molecules plays a dual role . As it turned out , while the expression rate of Comk drives the cell in competence , its degradation rate brings it back to its vegetative state . Similarly , a high concentration of ComS drives the cell in competence by competing over free MecA with ComK molecules , leaving more ComK molecules free . On the other hand , a decrease in ComS is necessary to return from competence as we will see next . This is true because low levels of ComS allow free MecA molecules to bind to ComK decreasing therefore the level of ComK molecules . We saw as well that high levels of ComK and ComS drive the cell into competence with probability 1 . This is in agreement with experimental results reported in [24] , where Leisner et al use an approximate SDE model in which they account for noise by introducing an additive gaussian noise term , in contrast to our approach which uses CME directly . We now study the roles of the different molecules in the return from competence . Figs . 7 and 8 suggest that the degradation rate has a larger effect than when it comes to the expected time for which a cell stays in competence . We found similar results for and . This implies that once a cell is in a state of competence , the degradation rate acts fast bringing it back to its vegetative state . The degradation rate is faster than the rate at which the free molecules try to keep the cell in competence . Fig . 8 suggests that increasing the value of will decrease the time for which a cell stays in competence . We also know from Table 2 that an increase in diminishes the probability with which Bacillus subtilis enters in competence . Our calculations also show that an increase in has a similar effect to an increase in in the sense that they both decrease the probability with which a cell enters in competence and the expected time it takes for a cell to return form competence . Recall that is the degradation rate of ComK , and is the degradation rate of ComS . Also recall that whenever the number of ComS molecules is sufficiently small , more MecA molecules will be free to bind with ComK decreasing therefore the number of ComK molecules . Similarly , a higher ComK degradation rate , will lead to a decrease in the number of ComK molecules . A lower number of ComK molecules drive the cell back to its vegetative state and/or decreases its probability of entering in competence . This explains the similarity in the effect of and on the probability of entering competence and the expected return time . In this paper we developed a discrete stochastic model for competence in Bacillus subtilis . We performed simulations of the model using Monte Carlo based SSA and verified that the reduced order model gave a valid approximation of the full model . We then applied the recently developed FSP method to the reduced model and computed the probability of competence , where competence was defined in terms of a trajectory leaving a pre-defined region of the state space . Having the analytical solution , we were able to conduct a sensitivity analysis of the probability with which a cell enters in competence as the model parameters vary . We were also able to compute interesting terms such as the expected time it takes for a cell to return from competence . This paper presented numerical methods that are applicable to many biological systems that exhibit a transient switching behavior . These methods were shown to be very useful in studying the genetic circuit regulating competence in a bacteria , and in answering questions about exact probabilities of stochastic events in this bistable biological behavior . They were also useful in studying sensitivities of these probabilities when expression rates , degradation rates , repression rates or activation rates of proteins were changed . Finally , the methods introduced in this paper showed how to calculate the expected time for return from transient states . Many other terms characterizing different transient physiological behaviors , such as the number of molecules that are most likely to enter in the transient states , and the return trajectories that are most likely to be taken can be computed using similar approaches to the one discussed here . Our approach should be easily extendible to analyze many biological system exhibiting a bistable switching behavior . | When exposed to stress , organisms react by taking actions that help them protect their DNA . ComK protein is a key regulator which activates hundreds of genes , including the genes encoding the DNA-uptake and recombination systems . In Bacillus subtilis , stress in the environment activates a sequence of chemical reactions that , driven by cellular noise , stochastically increases the level of ComK in some bacterial cells driving them from their original vegetative state into a competent state . Entrance into and exit from competence are stochastic switching events that the cell undergoes . In this work , we present a novel numerical method that allows the analysis of stochastic events in biological systems . We illustrate our method by computing the probability with which Bacillus subtilis enters in competence . We also present a method to analyze the sensitivity of stochastic events . We use this method to study the sensitivity of the probability of entrance in competence with respect to various gene expressions and degradation rates . We finally present a numerical method to calculate the expected time it takes a cell to return from competence . Although we studied the competence regulatory genetic circuit , our approach can be applied to a variety of stochastic events in biological systems . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results/Discussion"
] | [
"computational",
"biology/molecular",
"dynamics",
"computer",
"science/systems",
"and",
"control",
"theory",
"computational",
"biology/systems",
"biology",
"computational",
"biology/signaling",
"networks"
] | 2010 | Analysis of Stochastic Strategies in Bacterial Competence: A Master Equation Approach |
The baculovirus is a classic example of a parasite that alters the behavior or physiology of its host so that progeny transmission is maximized . Baculoviruses do this by inducing enhanced locomotory activity ( ELA ) that causes the host caterpillars to climb to the upper foliage of plants . We previously reported that this behavior is not induced in silkworms that are infected with a mutant baculovirus lacking its protein tyrosine phosphatase ( ptp ) gene , a gene likely captured from an ancestral host . Here we show that the product of the ptp gene , PTP , associates with baculovirus ORF1629 as a virion structural protein , but surprisingly phosphatase activity associated with PTP was not required for the induction of ELA . Interestingly , the ptp knockout baculovirus showed significantly reduced infectivity of larval brain tissues . Collectively , we show that the modern baculovirus uses the host-derived phosphatase to establish adequate infection for ELA as a virion-associated structural protein rather than as an enzyme .
Viruses and other parasites are known to usurp or alter the behavior of their hosts for their own benefit . This type of behavior modification by animal and even plant viruses is widely observed in arthropod hosts [1] , [2] . One of the earliest documented examples of such behavior modification is Wipfelkrankheit or tree-top disease of caterpillars [3] . A hallmark of this disease is enhanced locomotory activity ( ELA ) that causes the diseased caterpillars to migrate to the upper foliage of the host plant where they die . We now know that the causative agent of Wipfelkrankheit is a large , double-stranded DNA virus in the family Baculoviridae . Baculoviruses form a large group of arthropod-specific pathogens that commonly attack lepidopteran insects [4] . The baculovirus genome is large , 80 to over 160 kbp , and generally encodes more than 100 potential genes of which more than 10% appear to be derived from an ancestral host [5] . Baculoviruses produce two types of progeny during their infection cycle: the budded virus ( BV ) and occlusion-derived virus ( ODV ) . BVs are involved in spread of the virus within an infected host . ODVs on the other hand are occluded within an occlusion body ( OB ) that protects and transmits the ODV from insect-to-insect via oral infection [6] , [7] . At a late stage of infection , baculovirus-infected lepidopteran larvae often display ELA [3] , [8] , [9] and climb to the top of the host plant where they die and liquefy after death . It is believed that this behavior results in the dispersal of progeny OBs over a larger surface area thus improving the chance of virus transmission to other hosts . We have previously identified a protein tyrosine phosphatase ( ptp ) gene of the baculovirus , Bombyx mori nucleopolyhedrovirus ( BmNPV ) that induces wandering-like ELA in the silkworm B . mori . This gene was identified by behavioral screening of silkworms against a library of gene knockout mutants of BmNPV . Interestingly , the BmNPV ptp gene appears to have been acquired by an ancestral BmNPV from an ancestral silkworm [8] . Unlike silkworms that are infected with wild-type BmNPV , silkworms that are infected with a ptp-deleted BmNPV do not show ELA . The protein encoded by ptp , PTP , shows dephosphorylation activity with protein and RNA as substrate [10]–[13] , however , the role that PTP plays in the induction of ELA is still unknown . Recently , a knockout mutant of the baculovirus Lymantria dispar nucleopolyhedrovirus ( LdMNPV ) has also been shown to exhibit reduced ELA in comparison to wild-type LdMNPV in the European gypsy moth [9] . Specifically , gypsy moths infected with an ecdysteroid UDP-glucosyltransferase ( egt ) gene deletion mutant of LdMNPV show reduced climbing behavior . Here we surprisingly show that baculovirus PTP induces ELA as a structural protein , and not as an enzyme . Furthermore , we show that baculovirus infection of brain tissues appears to be important for the induction of ELA .
We previously reported that a ptp gene deletion mutant of BmNPV ( BmPTPD ) does not induce ELA in larval B . mori at a late stage of infection [8] . This suggested that baculovirus-induced ELA involves the dephosphorylation of an unknown protein or RNA target by baculovirus PTP . To test this hypothesis , we generated BmPTP-C119S ( Figure 1A ) , a mutant virus that expressed a PTP that was nearly deficient in phosphatase activity ( Supplementary Figure S1A ) . This mutagenesis was based on previous studies showing that mutation of cysteine 119 to serine ( C119S ) in the P-loop motif of the closely related PTP of Autographa californica NPV ( AcMNPV ) almost completely abolishes phosphatase activity [10] , [12] . To our surprise BmPTP-C119S induced ELA in 5th instar B . mori in a manner similar to that induced by wild-type BmNPV ( Figure 1B ) . This indicated that the phosphatase activity of PTP is not required for the induction of ELA . In order to determine whether the full-length PTP protein or RNAs transcribed from the ptp locus is required for the induction of ELA , we next generated BmPTP-Y9stop and BmPTP-E93stop ( Figure 1A ) . These BmNPV mutants each carried a ptp gene with a point mutation in the coding region that generated a premature stop codon . These mutations likely had little effect on the structure of the expressed mRNAs , however , the expressed proteins were only 9 or 93 amino acid residues in length . In a manner similar to that observed with BmPTPD , BmPTP-Y9stop and BmPTP-E93stop were both unable to induce ELA in larval B . mori ( Figure 1B ) . This indicated that the PTP protein itself is required for the induction of ELA but not mRNAs transcribed from the ptp gene . Our mutagenesis experiments indicated that the phosphatase activity of PTP is not required for the induction of ELA , thus we next used a yeast two-hybrid ( Y2H ) screening system to identify proteins that likely interact with PTP . The Y2H screening identified 5 clones which potentially interact with PTP ( Figure 2A ) . Four of the clones ( #12h-3 , -4 , -11 , and -16 ) contained BmNPV-derived sequences whereas one clone ( #2d-2 ) contained a B . mori-derived sequence of unknown function . The protein encoded by clone #12h-3 showed the strongest interaction with PTP ( Figure 2A ) . This clone contained a nearly full-length ( 1572 nts ) open reading frame ( ORF ) corresponding to ORF1629 of BmNPV ( Supplementary Figure S2A ) . ORF1629 encodes a WASP-like protein that localizes at one end of the nucleocapsid structure [14] , [15] . The deduced protein encoded by the clone #12h-11 corresponded to the C-terminal 347 amino acid residues ( 64% of full-length ) of ORF1629 and exhibited a moderately strong interaction with PTP ( Supplementary Figure S2A ) . Further analysis in yeast revealed that ( i ) the C-terminal of ORF1629 is critical for interaction with PTP ( Supplementary Figure S2A ) , and ( ii ) the N-terminal 90 amino acid residues of PTP do not interact with ORF1629 . The inability of the N-terminal of PTP to interact with ORF1629 is consistent with our locomotion assay results showing that BmPTP-E93stop did not induce ELA ( Figure 1B ) . In order to examine whether PTP interacts with ORF1629 in BmNPV-infected BmN cells , we next generated BmPTPD-wt . BmPTPD-wt is a derivative of BmPTPD that expresses FLAG-tagged PTP under an authentic ptp promoter immediately upstream of the polh gene ( Supplementary Figure S3B–C ) . The authentic promoter of ptp was identified by 5′-RACE analyses ( Supplementary Figure S3A ) . Immunoprecipitation experiments with anti-FLAG antibody and cell extracts from cells infected with BmPTPD-wt clearly showed that PTP interacts strongly with ORF1629 ( Figure 2B ) , confirming the results of the Y2H experiments . Because ORF1629 is a known structural protein , we speculated that PTP is also a structural protein that is associated with the BV envelope or capsid . Western blot analysis of BV-derived proteins that were fractionated into envelope and capsid components showed that PTP is primarily localized in the BV envelope ( Figure 3A ) . In order to examine whether the loss of PTP has any effects on the structural properties of the BV , we investigated the relative levels of ORF1629 and GP64 in BVs that were isolated from BmN cells infected with BmNPV , BmPTPD or BmPTPDR ( a repair virus of BmPTPD ) by western blot analysis . GP64 is a major envelope protein of BV that is essential for cell-to-cell infection [16] . The amounts of both GP64 and ORF1629 were reduced in the envelope and capsid , respectively , of BmPTPD BVs , in comparison to the corresponding BV fractions of BmNPV and BmPTPDR ( Figure 3B ) . These findings indicated that disruption of ptp results in the formation of abnormal BVs with potentially reduced virus infectivity and/or replication . The role of PTP in the productive infection of BmNPV was investigated in BmN cells and silkworm larvae at 3 and 4 days postinfection ( d p . i . ) , respectively . These time points were chosen because the production of BV and OB of wild-type BmNPV peak at these times . In BmN cells , BmPTPD produced about 50% fewer BVs and OBs in comparison to wild-type BmNPV at 3 d p . i . ( Supplementary Figure S4 ) . The reduction in BmPTPD OB production in BmN cells was consistent with that found in Sf9 cells infected with a ptp-deleted AcMNPV [17] , [18] . Similar reductions in BV and OB production were also observed in BmN cells infected with BmPTP-Y9stop- and BmPTP-E93stop ( Supplementary Figure S4 ) . In contrast , BmPTP-C119S and BmPTPDR produced wild-type levels of BV and OB in BmN cells ( Supplementary Figure S4 ) . These results indicated that the expression of full-length PTP is required for the production of wild-type levels of BV and OB in BmN cells . We next examined if the dramatic drop in BV and OB production that was found in BmN cells also occurred in BmPTPD-infected silkworm larvae . The production of BV and OB in the hemolymph of BmPTPD-infected larvae was less than 50% of that found in wild-type BmNPV-infected larvae at 4 d p . i . ( Figure 4A–B ) . In contrast , the production of BV and OB in larvae infected with BmPTP-C119S or BmPTPDR was similar to that found in BmNPV-infected larvae ( Figure 4A–B ) . These in vivo findings were essentially identical to those found in vitro and suggested again that the PTP protein , but not its enzymatic activity , is essential for normal BV and ODV production . In addition , we did not observe significant differences in the median lethal dose ( LD50 ) of BmNPV , BmPTPD , BmPTPDR , and BmPTP-C119S in 5th instar B . mori ( Supplementary Table S1 ) , suggesting that the absence of PTP does not alter the virulence of BmNPV . In order to investigate why BmPTPD produced fewer progeny , the expression profiles of a series of known baculovirus early and late gene products were examined by western blot analysis . Western blot analysis clearly showed that the expression of both early ( DBP , BRO , and LEF3 ) and late ( V-CHIA ) proteins was delayed in BmPTPD-infected BmN cells ( Figure 4C ) . These results indicated that loss of PTP caused a delay in the infection cycle , a delay that presumably led to the production of fewer BVs and OBs . In order to examine the effects of ptp deletion in larval B . mori in greater detail , we measured the expression of viral early/late ( gp64 ) and very late ( polh ) class genes in 16 tissues that were isolated from BmNPV- or BmPTPD-infected larvae . qRT-PCR analyses showed that , with the exception of a few tissues ( i . e . , corpora allata , prothoracic glands , and hemocytes ) , the relative expression levels of gp64 and polh were much lower in BmPTPD-infected larvae than in BmNPV-infected larvae ( Figure 5A ) . The expression of polh in the brain of BmPTPD-infected larvae at 4 d p . i . showed the most dramatic ( 67% ) reduction ( Figure 5B ) . These findings were consistent with the western blot analyses indicating that the replication cycle of BmPTPD was generally delayed but went further to show that the reduction in virus replication was most pronounced in the larval brain .
The manipulation of the behavior of caterpillars by baculoviruses has been known for over 100 years as Wipfelkrankheit . Recent developments in molecular biological and genomic tools have led to the identification of two baculovirus genes , ptp and egt , that are involved in altering host behavior [8] , [9] . Interestingly , the baculovirus appears to have obtained both ptp and egt from an ancestral host . Hoover et al . [9] hypothesize that the egt gene product ( EGT ) , a protein that is known to inactivate 20-hydroxyecdysone , controls the climbing behavior of NPV-infected gypsy moth larvae by hormonal regulation . On the other hand , the mechanistic action of how the ptp gene establishes behavioral control of host larvae remains elusive . In this study , we attempted to unravel this intriguing mystery by dissecting the functions of ptp/PTP in BmNPV-infected silkworms . We surprisingly found that the phosphatase activity of PTP appears not to be required for the behavioral control . In addition , we found that BmNPV induces ELA only when ptp mRNA is translated as a full-length protein , suggesting a non-enzymatic role for PTP . Our analyses revealed that PTP is a structural component of BV that is required for the production of mature BVs with full infectivity . We also found that loss of PTP dramatically reduces virus gene expression in several host tissues , especially in the brain . Previous biochemical studies show that baculovirus PTP has the ability to remove phosphate groups from protein and RNA substrates [10]–[13] . In this study we confirmed that BmNPV PTP is also a functional phosphatase ( Supplementary Figure S1A ) . Biologically , PTP and baculovirus LEF-4 have been predicted to play coordinated roles in 5′ cap formation of baculovirus late mRNAs [18] . A double-knockout mutant of ptp and lef-4 of AcMNPV , however , does not show defects in mRNA cap formation and replicates normally in cultured cells [18] . Thus , the overall biological significance of the phosphatase activity of baculovirus PTP is still unknown . Interestingly , the ptp gene is conserved in Group I NPVs ( e . g . , BmNPV ) but not in Group II NPVs ( e . g . , LdMNPV ) . Group II NPVs , however , are also able to induce ELA even though they lack the ability to produce PTP . This conundrum can be explained by the presence of the egt gene , a gene that is found in both Group I and II NPVs [4] . PTP and EGT appear to induce different types of ELA . PTP is involved in wandering-like ELA that is dramatically enhanced by light and shows positive phototropism [8] , whereas EGT is involved in the induction of vertical climbing behavior [9] . Thus , the baculovirus-induced “wandering” and “climbing” behaviors appear to be regulated by different viral genes but appear to work in concert in Group I NPVs to improve transmission of the virus . In addition , Group I and Group II NPVs have unique BV envelope structures ( e . g . , Group I NPVs use GP64 as an envelope fusion protein for host cell attachment whereas in Group II NPVs use the F protein [19] ) . These structural differences may lead to unique tissue tropism and modes of ELA induction by Group I and Group II NPVs . Additionally , there may be other baculovirus genes that are involved in induction of ELA but their roles in ELA may be difficult to identify if they are essential for other viral functions or if host-derived genes can partially substitute for their functions . Modern baculoviruses have apparently captured a number of essential and non-essential ‘auxiliary’ genes from ancestral host insects by horizontal gene transfer [5] . The authentic biological function of these captured genes or their products may be maintained , modified or lost in modern baculoviruses so that they confer selective advantages . The viral fibroblast growth factor ( vfgf ) gene is a clear example of a captured ancestral host gene whose authentic function has been maintained during evolution [20] , [21] . The protein encoded by vfgf , vFGF , transmits its signaling via a host FGF receptor that , when activated , causes the migration of hemocytes to virus-infected tissues . vFGF is thus able to usurp the host's signaling pathway in order to recruit hemocytes which , following infection , can disseminate the virus to other tissues and increase systemic infection . BmNPV ptp is another example of a captured ancestral host gene [8] . In the case of BmNPV ptp , however , the biological importance of the PTP protein appears to have changed over time from a protein with enzymatic significance to one that has structural significance for establishing infection in larval tissues that are critical for the induction of ELA . In this study , we show that PTP binds strongly to ORF1629 , a baculovirus structural protein that is phosphorylated during the infection cycle [22] . We hypothesize that in modern baculoviruses , the ability of PTP to bind ORF1629 or some other target became more important because of the role it plays in increasing virus transmission . In contrast , the ability of PTP to dephosphorylate a potential protein or RNA substrate appears not to be as important or perhaps taken over by a host phosphatase or only required when the virus has to replicate under unusual conditions . Alternatively , the PTP protein may have a dual function as a structural protein in the induction of ELA and as a phosphatase enzyme perhaps during earlier stages of infection , in specific tissues , or different host developmental stages . The ptp gene is thus the first example of a host-derived gene whose product is utilized by the modern baculovirus in a completely different manner from how it was likely utilized in the ancestral host . Wandering is a normal ELA behavior that occurs towards the end of the last larval instar that causes caterpillars to search for an appropriate location to undergo metamorphosis [23] . Wandering behavior is regulated by a combination of internal ( e . g . , larval size , hormones ) and external ( e . g . , photoperiod ) processes . In larval Manduca sexta the brain exerts a net inhibitory influence that prevents wandering behavior during the caterpillar feeding stage [24] , [25] . At the hormonal level , wandering is induced by exposure to the 20-hydroxyecdysone which causes the brain to become excitatory during the wandering stage . We hypothesize that baculovirus infection of caterpillar brain also leads to an excitatory state leading to the induction of the wandering-like ELA that we observe in BmNPV-infected silkworms . Electrophysiological studies of the locomotory patterns in the brain and subesophageal ganglion from baculovirus-infected larvae will allow us to understand what occurs in the central nervous system during virus-induced ELA . Our current hypothesis suggests that the baculovirus plays a direct role in the induction of ELA by infecting the brain . However , other more subtle factors such as baculovirus-induced changes in host energy metabolism , signal transduction , sensitivity to light or gravity , etc . may also play roles in the induction of the various types of ELA . In conclusion , we show here that PTP functions to induce wandering-like ELA in baculovirus-infected caterpillars as a structural protein and likely not as an enzyme . Notably , we found that virus propagation was markedly reduced in brain tissues when ptp was deleted from the BmNPV genome . These results tell an amazing story of how the modern baculovirus has evolved to use a captured host gene in a different way from how it was likely used by the ancestral host . Collectively , we conclude that PTP augments baculovirus infection of the brain and possibly other tissues that play critical roles in the induction of ELA .
Larval B . mori were reared as described previously [26] . BmN ( BmN-4 ) cells were cultured at 27°C in TC-100 medium supplemented with 10% fetal bovine serum [26] . The T3 strain of BmNPV was used as the wild-type virus . The construction of BmPTPD ( a ptp deletion mutant ) and BmPTPDR ( a repair virus of BmPTPD ) , have been reported previously [8] ( see Figure 1A ) . The titers of BmNPV and mutant BmNPVs were determined by plaque assay on BmN cells [26] . BmNPV genomic DNA containing ptp and its flanking regions were cloned into pcDNA3 . 1 ( - ) and used as a template to generate mutations in the ptp gene . Mutagenesis was performed by overlapping PCR [26] and confirmed by DNA sequencing . The resultant plasmids were transfected with Bsu36I-digested BmPTPD DNA ( a Bsu36I restriction endonuclease site is uniquely found within the lacZ gene cassette of BmPTPD ) into BmN cells using Lipofectin reagent ( Invitrogen ) . Five days after transfection , the medium was collected and stored at 4°C until use . Three recombinant BmNPVs expressing PTP-C119S ( BmPTP-C119S ) , PTP-Y9stop ( BmPTP-Y9stop ) , and PTP-E93stop ( BmPTP-E93stop ) ( Figure 1A ) were isolated by the identification of plaques that did not express β-galactosidase [26] . The presence of the mutated ptp genes in these constructs was confirmed by polymerase chain reaction ( PCR ) using primers ptpF1 and ptp_B ( Supplementary Table S2 ) . BmPTPD-wt , a repair mutant of BmPTPD that expresses a FLAG-tagged PTP under an authentic ptp gene promoter ( inserted immediately upstream of the polh gene ) was generated by a two step process . Firstly , the FLAG-tagged ptp gene driven by the authentic ptp gene promoter ( identified by 5′-RACE ) was amplified by PCR using BmNPV DNAs and primers ptpEPS1 and ptpEPS3 ( Supplementary Table S2 ) . The amplicon was inserted into the transfer vector pBmEPS1 [27] , and the recombinant transfer plasmid was transfected with Bsu36I-digested BmNPV-abb [27] genomic DNA into BmN cells using Cellfectin reagent ( Invitrogen ) [28] . A recombinant BmNPV ( T3-wt ) expressing the FLAG-tagged PTP was plaque-purified by the identification of plaques that were OB-positive . In the second step , the authentic ptp gene of T3-wt was disrupted by transfection of T3-wt genomic DNA with a plasmid carrying a lacZ gene cassette flanked by ptp gene sequences [8] into BmN cells using Cellfectin reagent ( Invitrogen ) . BmPTPD-wt ( Supplementary Figure S3 ) , a recombinant BmNPV expressing FLAG-tagged PTP under the authentic ptp gene promoter ( but not expressing authentic PTP ) was identified by the formation of plaques expressing β-galactosidase [29] and by PCR using the primer sets BmEPS_F1/BmEPS_R1 and ptpF2/ptpG2 ( Supplementary Table S2 ) . Expression of FLAG-tagged PTP by BmPTPD-wt was confirmed by western blot analysis with anti-FLAG antibody ( Sigma ) . Locomotion assays were performed as reported previously with minor modifications [30] . Briefly , 5th instar B . mori ( 24 larvae per treatment ) were starved for several hours , injected with 50 µl of a viral suspension containing 1×105 PFU , and returned to the artificial diet at 27°C . Infected larvae were photographed at 3 h intervals from 84 to 132 h postinfection ( h p . i . ) . At each 3 h interval , the 24 infected larvae ( separated into 4 groups of 6 larvae ) were placed in the center of a piece of paper marked with concentric circles ( the radius of each circle was 5 mm greater than the previous circle , with a maximum radius of 100 mm ) . Photographs were taken with a digital camera at 1 min intervals until 5 min after release . The coordinates of each larva , at the midpoint of the third and fourth abdominal segments , was determined at each time point after release using ImageJ software ( Rasband WS ( 2006 ) ImageJ . Bethesda , Maryland: U . S . National Institutes of Health , rsb . info . nih . gov/ij/ ) . The distance moved during each 1 min-long interval was determined and summed up to derive total locomotory distance in 5 min . The locomotory distance of dead larvae was designated as zero . Yeast two-hybrid ( Y2H ) screening was performed using the PROQUEST two-hybrid system ( Gibco BRL ) as described previously [31] . The Y2H screening used a cDNA library that was generated from BmNPV-infected BmN cells as described previously [31] , as well as a cDNA library that was constructed using mRNAs purified from epidermal tissues from BmNPV-infected larvae ( 2 d p . i . ) . BmN cells were infected with BmNPV , BmPTPD , or BmPTPDR at an MOI of 5 and harvested at 48 h p . i . Biochemical fractionation of the BmN cells was performed as described previously [32] . Procedures for the isolation of BVs and fractionation of BV components were reported previously [33] . SDS-PAGE and western blotting were performed using anti-FLAG antibody , anti-GP64 antibody ( Santa Cruz Biotechnology ) , anti-ORF1629 antibody [14] ( a gift from George F . Rohrmann ) , anti-LEF3 antibody [34] , [35] ( a gift from Eric B . Carstens ) , anti-BRO antibody [36] , anti-DBP antibody [37] , and anti-actin antibody ( Santa Cruz Biotechnology ) as described previously [20] . Immunoprecipitation experiments were performed as described previously [20] . Fifth instar B . mori ( 4 larvae per virus ) were inoculated with virus and reared as described above . OBs that were released into the hemolymph , at 96 h p . i . , were quantified from individual larva using a hemocytometer as described previously [26] . Hemolymph BV titer was determined by plaque assay on BmN cells [26] . Fifth instar B . mori were inoculated and reared as described above . Total RNA was prepared using Trizol reagent ( Invitrogen ) from 16 tissues ( brain , corpora allata , central nerve , prothoracic gland , fat body , trachea , hemocyte , testis , ovary , anterior silk gland , middle silk gland , posterior silk gland , midgut , Malpighian tubule , integument , and muscle ) that were dissected from BmNPV- or BmPTPD-infected , 5th instar B . mori ( 5 to 30 larvae/tissue ) at 1 , 2 , 3 , and 4 d p . i . For the experiments shown in Figure 5B , total RNA was prepared using Trizol reagent from four tissues ( brain , central nerve , fat body , and trachea ) that were dissected from four individual 5th instar larvae . First-strand cDNAs were synthesized from 0 . 2 µg of total RNA , and qRT-PCR was performed using Power SYBR Green PCR master mix ( Applied Biosystems ) using primers that were previously described [29] . PCR was performed using the StepOne real-time PCR system ( Applied Biosystems ) [29] . Statistical analysis was performed using Prism 5 software ( Graphpad ) . One-way analyses of variance ( ANOVA ) was performed with post hoc Tukey's test comparing each of the treatment group means with the mean of the control group . Locomotion assay data were subjected to Kruskal-Wallis analysis with post hoc Dunn's test . Student's t-test was used to compare values obtained in the qRT-PCR experiments ( Figure 5B ) . | Pathogens are known to usurp or alter the behavior of their hosts for their own benefit . Such behavior modification by animal and even plant viruses is widely observed in insect hosts . One of the earliest documented examples of such behavior modification is Wipfelkrankheit , a baculovirus-induced disease that causes caterpillars to migrate to the upper foliage of food plants where they die . Two baculovirus genes , ptp and egt , are involved in the induction of enhanced locomotory activity ( ELA ) such as climbing behavior in baculovirus-infected caterpillars . Here we dissect the functional role that baculovirus protein tyrosine phosphatase ( PTP ) , the protein encoded by ptp , plays in the induction of ELA . We surprisingly found that PTP functions as a virus-associated structural protein and not as an enzyme in regard to the induction of ELA . We show that PTP plays a crucial role in virus infection of brain tissues , and hypothesize that this infection results in pathogen control of insect behavior . Since ptp was likely captured from an ancestral host by horizontal gene transfer , our findings tell an amazing story of how the modern baculovirus uses a captured host gene in a completely different way from how it was likely used in the ancestral host . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"pest",
"control",
"biology",
"microbiology",
"evolutionary",
"biology",
"agriculture"
] | 2012 | The Baculovirus Uses a Captured Host Phosphatase to Induce Enhanced Locomotory Activity in Host Caterpillars |
Circadian rhythms enable organisms to synchronise the processes underpinning survival and reproduction to anticipate daily changes in the external environment . Recent work shows that daily ( circadian ) rhythms also enable parasites to maximise fitness in the context of ecological interactions with their hosts . Because parasite rhythms matter for their fitness , understanding how they are regulated could lead to innovative ways to reduce the severity and spread of diseases . Here , we examine how host circadian rhythms influence rhythms in the asexual replication of malaria parasites . Asexual replication is responsible for the severity of malaria and fuels transmission of the disease , yet , how parasite rhythms are driven remains a mystery . We perturbed feeding rhythms of hosts by 12 hours ( i . e . diurnal feeding in nocturnal mice ) to desynchronise the host’s peripheral oscillators from the central , light-entrained oscillator in the brain and their rhythmic outputs . We demonstrate that the rhythms of rodent malaria parasites in day-fed hosts become inverted relative to the rhythms of parasites in night-fed hosts . Our results reveal that the host’s peripheral rhythms ( associated with the timing of feeding and metabolism ) , but not rhythms driven by the central , light-entrained circadian oscillator in the brain , determine the timing ( phase ) of parasite rhythms . Further investigation reveals that parasite rhythms correlate closely with blood glucose rhythms . In addition , we show that parasite rhythms resynchronise to the altered host feeding rhythms when food availability is shifted , which is not mediated through rhythms in the host immune system . Our observations suggest that parasites actively control their developmental rhythms . Finally , counter to expectation , the severity of disease symptoms expressed by hosts was not affected by desynchronisation of their central and peripheral rhythms . Our study at the intersection of disease ecology and chronobiology opens up a new arena for studying host-parasite-vector coevolution and has broad implications for applied bioscience .
The discovery of daily rhythms in parasites dates back to the Hippocratic era and a taxonomically diverse range of parasites ( including fungi , helminths , Coccidia , nematodes , trypanosomes , and malaria parasites [1–6] ) display rhythms in development and several behaviours . Yet , how rhythms in many parasite traits are established and maintained remains mysterious , despite their significance , as these traits underpin the replication and transmission of parasites [7] . For example , metabolic rhythms of Trypanosoma brucei have recently been demonstrated to be under the control of an oscillator belonging to the parasite , but the constituents of this oscillator are unknown [8] . In most organisms , endogenous circadian oscillators ( “clocks” ) involve transcription-translation feedback loops whose timing is synchronised to external cues , such as light-dark and feeding-fasting cycles [9 , 10] but there is generally little homology across taxa in the genes underpinning oscillators . Multiple , convergent , evolutionary origins for circadian oscillators is thought to be explained by the fitness advantages of being able to anticipate and exploit predictable daily changes in the external environment , as well as keeping internal processes optimally timed [11 , 12] . Indeed , the 2017 Nobel Prize in Physiology/Medicine recognises the importance of circadian oscillators [13 , 14] . The environment that an endoparasite experiences inside its host is generated by many rhythmic processes , including daily fluctuations in the availability of resources , and the nature and strength of immune responses [15 , 16] . Coordinating development and behaviour with rhythms in the host ( or vector ) matters for parasite fitness [17] . For example , disrupting synchrony between rhythms in the host and rhythms in the development of malaria parasites during asexual replication reduces parasite proliferation and transmission potential [18 , 19] . Malaria parasites develop synchronously during cycles of asexual replication in the host’s blood and each developmental stage occurs at a particular time-of-day . The synchronous bursting of parasites at the end of their asexual cycle , when they release their progeny to infect new red blood cells , causes fever with sufficient regularity ( 24 , 48 , or 72 hourly , depending on the species ) to have been used as a diagnostic tool . Malaria parasites are assumed to be intrinsically arrhythmic and mathematical modelling suggests that rhythms in host immune effectors , particularly inflammatory responses , could generate rhythms in the development of malaria parasites via time-of-day-specific killing of different parasite developmental stages [20 , 21] . However , the relevant processes operating within real infections remain unknown [22] . Our main aim is to use the rodent malaria parasite Plasmodium chabaudi to ask which circadian rhythms of the host are involved in scheduling rhythms in parasite development . In the blood , P . chabaudi develops synchronously and asexual cycles last 24 hours , bursting to release progeny ( schizogony ) in the middle of the night when mice are awake and active . We perturbed host feeding time ( timing of food intake ) , which is known to desynchronise the phase of rhythms from the host’s central and peripheral oscillators , and we then examined the consequences for parasite rhythms . In mammals , the central oscillator in the brain ( suprachiasmatic nuclei of the hypothalamus , SCN ) , is entrained by light [10 , 23] . The SCN is thought to shape rhythms in physiology and behaviour ( peripheral rhythms ) by entraining peripheral oscillators via hormones such as glucocorticoids [24] . However , oscillators in peripheral tissues are self-sustained and can also be entrained by several non-photic cues , such as the time-of-day at which feeding occurs [25 , 26] . Thus , eating at the wrong time-of-day ( e . g . diurnal feeding in nocturnal mice ) leads to altered timing of oscillators , and their associated rhythms in peripheral tissues . This phase-shift is particularly apparent in the liver where an inversion in the peak phase of expression of the circadian oscillator genes Per1 and Per2 occurs [26] . Importantly , eating at the wrong time-of-day does not alter rhythmic outputs from the central oscillator [25] . In murine hosts with an altered ( diurnal ) feeding schedule , the development rhythms of parasites remained synchronous but became inverted relative to the rhythms of parasites in hosts fed at night . Thus , feeding-related outputs from the hosts peripheral timing system , not the SCN , are responsible for the timing ( phase ) of parasite rhythms . We also reveal that the inversion of parasite rhythms corresponds to a phase-shift in blood glucose rhythms . That parasites remain synchronous during the rescheduling of their rhythm coupled with evidence that immune responses do not set the timing of parasite rhythms , suggests parasites are responsible for scheduling their developmental rhythm , and may express their own circadian rhythms and/or oscillators . Furthermore , our perturbed feeding regimes are comparable to shift work in humans . This lifestyle is well-known for increasing the risk of non-communicable diseases ( cancer , type 2 diabetes etc . [27] ) but our data suggest the severity of malaria infection ( weight loss , anaemia ) is not exacerbated by short-term desynchronisation of the central and peripheral oscillators .
First , we examined the effects of changing the time of food intake on the phasing of circadian rhythms in host body temperature and locomotor activity ( Fig 1 ) . Body temperature is a commonly used phase marker of circadian timing because core body temperature increases during activity and decreases during sleep [28 , 29] . Mice were given access to food for 12 hours in each circadian cycle , either in the day ( LF , light fed ) or night ( DF , dark fed ) . All food was available ad libitum and available from ZT 0–12 ( ZT refers to ‘Zeitgeber Time’; ZT 0 is the time in hours since lights on ) for LF mice , and from ZT 12–24 for DF mice . All experimental mice were entrained to the same reversed photoperiod , lights on: 7pm ( ZT 0/24 ) , lights off: 7am ( ZT 12 ) , for 2 weeks prior to starting the experiment ( Fig 1 ) . We found a significant interaction between feeding treatment ( LF or DF ) and the time-of-day ( day ( ZT 0–12 ) or night ( ZT 12–24 ) ) that mice experience elevated body temperatures ( χ2 ( 5 , 6 ) = 75 . 89 , p < 0 . 0001 ) and increase their locomotor activity ( χ2 ( 5 , 6 ) = 39 . 57 , p < 0 . 0001; S1 Table ) . Specifically , DF mice have elevated body temperature and are mostly active during the night ( as expected ) whereas LF mice show no such day-night difference in body temperature and locomotor activity , due to a lack of night time elevation in both measures where food and light associated activity are desynchronised ( Fig 2 ) . We also find the centres of gravity ( CoG; a general phase marker of circadian rhythms , estimated with CircWave ) , are slightly but significantly earlier in LF mice for both body temperature ( approximately 2 hours advanced: χ2 ( 3 , 4 ) = 28 . 17 , p < 0 . 0001 ) and locomotor activity ( approximately 4 hours advanced: χ2 ( 3 , 4 ) = 27 . 32 , p < 0 . 0001 ) ( S1 Table ) . Therefore , the LF mice experienced a significant change in the daily profile of activity , which is reflected in some phase advance ( but not inversion ) relative to DF mice , and significant disruption to their body temperature and locomotor activity rhythms , particularly during the night . Because an altered feeding schedule does not affect the phase of the SCN [25] , our data suggest that rhythms in body temperature and locomotor activity in LF mice are shaped by both rhythms in feeding and the light-dark cycle [30] . Finally , the body weight of LF and DF mice did not differ significantly after 4 weeks ( χ2 ( 3 , 4 ) = 0 . 02 , p = 0 . 9 ) and both groups equally gained weight during the experiment ( S1 Fig ) , corroborating that LF mice were not calorie restricted . Having generated hosts in which the phase relationship between the light-entrained SCN and food-entrained rhythms are altered ( LF mice ) or not ( DF mice ) , we then infected all mice with the rodent malaria parasite Plasmodium chabaudi adami genotype DK ( Fig 1 ) from donor mice experiencing a light-dark cycle 12 hours out of phase with the experimental host mice . After allowing the parasite’s developmental rhythms to become established ( see Materials and Methods ) we compared the rhythms of parasites in LF and DF mice . We hypothesised that if parasite rhythms are solely determined by rhythms driven by the host’s SCN ( which are inverted in the host mice compared to the donor mice ) , parasite rhythms would equally shift and match in LF and DF mice because both groups of hosts were entrained to the same light-dark conditions . Yet , if rhythms in body temperature or locomotor activity directly or indirectly ( via entraining other oscillators ) contribute to parasite rhythms , we expected that parasite rhythms would differ between LF and DF hosts . Further , if feeding directly or indirectly ( via food-entrained oscillators ) drives parasite rhythms , we predicted that parasite rhythms would become inverted ( Fig 1 ) . In the blood , P . chabaudi parasites transition through five developmental stages during each ( ~24hr ) cycle of asexual replication ( Fig 3A ) [6 , 31] . We find that four of the five developmental stages ( rings , and early- , mid- , and late-trophozoites ) display 24hr rhythms in both LF and DF mice ( Fig 3B , S2 Table , S2 Fig ) . The fifth stage—schizonts—appear arrhythmic but this stage sequesters in the host’s tissues [32 , 33] and so , are rarely collected in venous blood samples . Given that all other stages are rhythmic , and that rhythms in ring stages likely require their parental schizonts to have been rhythmic , we expect schizonts are rhythmic but that sequestration prevents a reliable assessment of their rhythms . The CoG estimates for ring , and early- , mid- , and late-trophozoite stages are approximately 10–12 hours out-of-phase between the LF and DF mice ( Fig 3B and 3C , S2 Table ) . For example , rings peak at approximately ZT 10 in LF mice and peak close to ZT 23 in DF mice . The other stages peak in sequence . Schizogony ( when parasites burst to release their progeny ) occurs immediately prior to reinvasion , therefore we expect it occurs during the day for the LF mice and night for DF mice [7] . The almost complete inversion in parasite rhythms between LF and DF mice demonstrates that feeding-related rhythms are responsible for the phase of parasite rhythms , with little to no apparent contribution from the SCN and/or the light: dark cycle . Changing the feeding time of nocturnal mice to the day time has similarities with shift work in diurnal humans [34] . This lifestyle is associated with an increased risk of acquiring non-communicable diseases ( e . g . cancer , diabetes ) [35] and has been recapitulated in mouse models [e . g . 36 , 37 , 38] . In contrast , in response to perturbation of their feeding rhythm , infections are not more severe in hosts whose circadian rhythms are desynchronised ( i . e . LF hosts ) . Specifically , all mice survived infection and virulence ( measured as host anaemia; reduction in red blood cells ) of LF and DF infections is not significantly different ( comparing minimum red blood cell density , χ2 ( 3 , 4 ) = 0 . 11 , p = 0 . 74; S3A Fig ) . As described above , changes in body mass were not significantly different between treatments ( S1 Fig ) . Using a longer-term model for shift work may reveal differences in infection severity , especially when combined with the development of non-communicable disease . There are no significant differences between parasite densities in LF and DF hosts during infections ( LF versus DF on day 6 post infection , χ2 ( 3 , 5 ) = 0 . 66 , p = 0 . 42 , S3B Fig ) . This can be explained by both groups being mismatched to the SCN of the host , which we have previously demonstrated to have negative consequences for P . chabaudi [18] . Our previous work was carried out using P . chabaudi genotype AJ so is not directly comparable to our results presented here , because DK is a less virulent genotype [39] . Instead , a comparison of our results to data collected previously for genotype DK , in an experiment where SCN rhythms of donor and host mice were matched ( see Materials and Methods; infections were initiated with the same strain , sex , and age of mice , the same dose at ring stage ) reveals a cost of mismatch of donor and host entrainment . Specifically , parasite density on day 6 ( when infections have established but before parasites start being cleared by host immunity ) is significantly lower in infections mismatched to the SCN ( LF and DF ) compared to infections matched to the SCN ( χ2 ( 3 , 5 ) = 16 . 71 , p = 0 . 0002 , mean difference = 2 . 21e+10 parasites per ml blood ) ( see S4A Fig ) . In keeping with a difference in parasite replication , hosts with matched infections reach lower red blood cell densities ( χ2 ( 3 , 5 ) = 18 . 87 , p < 0 . 0001 , mean difference = 5 . 29e+08 red blood cells per ml blood ) . The mismatched and matched infections compared above also differ in whether hosts had food available throughout the 24-hour cycle or for 12 hours only ( LF and DF ) . Restricting food to 12 hours per day does not affect host weight ( S1 Fig ) and mice still undergo their main activity bout at lights off even when food is available all the time . Therefore , we propose that rather than feeding duration , mismatch to the host SCN for as few as 5 cycles is costly to parasite replication and reduces infection severity . Because peripheral and SCN driven rhythms are usually in synchrony , we suggest parasites use information from food-entrained oscillators , or metabolic processes , to ensure their development is timed to match the host’s SCN rhythms . Instead of organising their own rhythms ( i . e . using an “oscillator” whose time is set by a “Zeitgeber” or by responding directly to time-of-day cues ) , parasites may allow outputs of food-entrained host oscillators to enforce developmental rhythms . Previous studies have focused on rhythmic immune responses as the key mechanism that schedules parasite rhythms ( via developmental-stage and time-of-day specific killing [20 , 21] ) . Evidence that immune responses are rhythmic in naïve as well as infected hosts is increasing [15 , 16] , but the extent to which peripheral/food-entrained oscillators and the SCN drive immune rhythms is unclear . Nonetheless , we argue that rhythms in host immune responses do not play a significant role in scheduling parasites for the following reasons: First , mismatch to the host’s peripheral rhythms ( which occurs in DF mice but not LF mice as a feature of our experimental design ) does not cause a significant reduction in parasite number ( S3B Fig ) , demonstrating that stage-specific killing cannot cause the differently phased parasite rhythms in LF and DF mice . Second , while changing feeding time appears to disrupt some rodent immune responses [40 , 41] , effectors important in malaria infection , including leukocytes in the blood , do not entrain to feeding rhythms [42 , 43] . Third , inflammatory responses important for killing malaria parasites are upregulated within hours of blood stage infection [44] so their footprint on parasite rhythms should be apparent from the first cycles of replication [19] . In contrast , rhythms of parasites in LF and DF mice do not significantly diverge until 5–6 days post infection , after 5 replication cycles ( S3 Table , Fig 4 ) . Fourth , an additional experiment ( see Materials and Methods ) reveals that rhythms in the major inflammatory cytokines that mediate malaria infection ( e . g . IFN-gamma and TNF-alpha: [45 , 46 , 47 , 48] ) follow the phase of parasite rhythms ( Fig 5 ) , with other cytokines/chemokines also experiencing this phenomenon ( S5 Fig ) . Specifically , mice infected with P . chabaudi genotype AS undergoing schizogony at around midnight ( ZT17 ) , produce peaks in the cytokines IFN-gamma and TNF-alpha at ZT21 and ZT19 respectively ( following a significantly 24h pattern: IFN-gamma p = 0 . 0055 , TNF-alpha p = 0 . 0015 ) . Whereas mice infected with mismatched parasites undergoing schizogony around ZT23 ( 6 hours later ) , experience 3–6 hour delays in the peaks of IFN-gamma and TNF-alpha ( IFN-gamma: ZT0 , TNF-alpha: ZT1; following a significantly 24h pattern: IFN-gamma p = 0 . 0172 , TNF-alpha p = 0 . 0041 ) . Thus , even if parasites at different development stages differ in their sensitivity to these cytokines , these immune rhythms could only serve to increase synchrony in the parasite rhythm but not change its timing . More in-depth analysis of LF and DF infections provides further support that parasites actively organise their developmental rhythms . We examined whether parasites in DF mice maintain synchrony and duration of different developmental stages during rescheduling to the host’s SCN rhythms . Desynchronisation of oscillators manifests as a reduction in amplitude in rhythms that are driven by more than one oscillator ( e . g . parasite and host oscillator ) . No loss in amplitude suggests that parasites shift their timing as a cohort without losing synchrony . Parasite rhythms in LF and DF mice did not differ significantly in amplitude ( χ2 ( 6 , 7 ) = 1 . 53 , p = 0 . 22 , S4A Table ) and CoGs for sequential stages are equally spaced ( χ2 ( 10 , 18 ) = 11 . 75 , p = 0 . 16 , S2 Table ) demonstrating that parasite stages develop at similar rates in both groups . The rhythms of parasites in LF and DF mice were not intensively sampled until days 6–8 PI , raising the possibility that parasites lost and regained synchrony before this . Previously collected data for P . chabaudi genotype AS infections mismatched to the host SCN by 12 hours that have achieved a 6-hour shift by day 4 PI also exhibit synchronous development ( S4B Table and S6 Fig ) , suggesting that parasites reschedule in synch . That parasite rhythms do not differ significantly between LF and DF mice until day 5–6 post infection ( Fig 4 ) could be explained by the parasites experiencing a phenomenon akin to jet lag . Jet lag results from the fundamental , tissue-specific robustness of circadian oscillators to perturbation , which slows down the phase shift of individual oscillators to match a change in ‘time-zone’ [10] . We propose that the most likely explanation for the data gathered from our main experiment for genotype DK , and that collected previously for AJ and AS , is that parasites possess intrinsic oscillators that shift collectively , in a synchronous manner , by a few hours each day , until they re-entrain to the new ‘time-zone’ . Because there is no loss of amplitude of parasite rhythms , it is less likely that individual parasites possess intrinsic oscillators that re-entrain at different rates to the new ‘time-zone’ . The recently demonstrated ability of parasites to communicate decisions about asexual to sexual developmental switches [49] could also be involved in organising asexual development . If parasites have evolved a mechanism to keep time and schedule their rhythms , what external information might they synchronise to ? Despite melatonin peaks in lab mice being brief and of low concentration [50 , 51] , the host’s pineal melatonin rhythms have been suggested as a parasite time cue [52] . However , we can likely rule pineal melatonin , and other glucocorticoids , out because they are largely driven by rhythms of the SCN , which follow the light-dark cycle and have not been shown to phase shift by 12 hours as a result of perturbing feeding timing [25]; some glucocorticoid rhythms appear resistant to changing feeding time [53] . Whether extra-pineal melatonin , produced by the gut for example [54] , could influence the rhythms of parasites residing in the blood merits further investigation . Body temperature rhythms have recently been demonstrated as a Zeitgeber for an endogenous oscillator in trypanosomes [8] . Malaria parasites are able to detect and respond to changes in environmental temperature to make developmental transitions in the mosquito phase of their lifecycle [55 , 56] , and may deploy the same mechanisms to organise developmental transitions in the host . Body temperature rhythms did not fully invert in LF mice but they did exhibit unusually low ( i . e . day time ) temperatures at night . Thus , for body temperature to be a time-of-day cue or Zeitgeber it requires that parasites at early developmental stages ( e . g . rings or early trophozoites ) are responsible for time-keeping because they normally experience low temperatures during the day when the host is resting . The same logic applies to rhythms in locomotor activity because it is very tightly correlated to body temperature ( Pearson’s correlation R = 0 . 85 , 95% CI: 0 . 82–0 . 88 ) . Locomotor activity affects other rhythms , such as physiological oxygen levels ( daily rhythms in blood and tissue oxygen levels ) , which can reset circadian oscillators [57] and have been suggested as a time cue for filarial nematodes [4] . Feeding rhythms were inverted in LF and DF mice and so , the most parsimonious explanation is that parasites are sensitive to rhythms related to host metabolism and/or food-entrained oscillators . Malaria parasites have the capacity to actively alter their replication rate in response to changes in host nutritional status [58] . Thus , we propose that parasites also possess a mechanism to coordinate their development with rhythms in the availability of nutritional resources in the blood . Further work could explore whether parasites use information via the kinase ‘KIN’ to regulate their timing [58] . KIN shares homology with AMP-activated kinases ( AMPK ) , mammalian metabolic sensors implicated in both circadian timing and metabolic regulation [59] . Glucose , and other sugars that require metabolising , suppresses the activation of AMPK and its subsequent nutrient-sensing signalling cascade , with KIN proposed to act as a nutrient sensor to reduce parasite replication rate in response to calorie restriction during malaria infection [58] . Rhythms in blood glucose are a well-documented consequence of rhythms in feeding timing [60] and glucose is an important resource for parasites [61] . We performed an additional experiment to quantify blood glucose rhythms in ( uninfected ) LF and DF mice ( Fig 6A and 6B ) . Despite the homeostatic regulation of blood glucose , we find its concentration varies across the circadian cycle , and is borderline significantly rhythmic in DF mice ( p = 0 . 07 , peak time = ZT17 . 84 , estimated with CircWave ) and follows a significantly 24-hour pattern in LF mice ( p < 0 . 0001 , peak time = ZT8 . 78 ) . Glucose rhythms/patterns are shaped by feeding regime ( time-of-day: feeding treatment χ2 ( 18 , 32 ) = 45 . 49 , p < 0 . 0001 ) . Specifically , during the night , DF mice have significantly higher blood glucose than LF mice ( t = 3 . 41 , p = 0 . 01 , mean difference 20 . 6mg/dl±7 . 32 ) and there is a trend for LF mice to have higher blood glucose than DF mice during the day ( t = -0 . 94 , p = 0 . 78 , mean difference 7 . 9mg/dl±9 . 86 ) . Titrating whether glucose availability is high or low would only provide parasites with information on whether it is likely to be day or night , and a 12-hour window in which to make developmental transitions should erode synchrony , especially as glucose rhythms are weak in DF mice . Instead , parasites may use the sharp rise in blood glucose that occurs in both LF and DF mice after their main bout of feeding as a cue for dusk ( S5 Table; regions with solid lines connecting before and after feeding in Fig 6 ) , using KIN as a sensor [58] . In line with the effects of feeding timing we observe in mice , a recent study of humans reveals that changing feeding time can induce a phase-shift in glucose rhythms , but not insulin rhythms [43] . Alternatively , parasites may be sensitive to fluctuations in other factors due to rhythms in food intake , such as amino acids [62] or other rhythmic metabolites that appear briefly in the blood after feeding , changes in oxygen consumption , blood pressure or blood pH [63 , 64] . In summary , we show that peripheral , food-entrained host rhythms , but not central , light-entrained host rhythms are responsible for the timing of developmental transitions during the asexual replication cycles of malaria parasites . Taken together , our observations suggest that parasites have evolved a time-keeping mechanism that uses daily fluctuations in resource availability ( e . g . glucose ) as a time-of-day cue or Zeitgeber to match the phase of asexual development to the host’s SCN rhythms . Why coordination with the SCN is important remains mysterious . Uncovering how parasites tell the time could enable an intervention ( ecological trap ) to “trick” parasites into adopting suboptimal rhythms for their fitness .
We compared the performance of parasites in our main experiment ( in which infections were initiated with parasites from donor mice that were mismatched to the host’s SCN rhythms by 12 hours ) , to the severity of infections when infections are initiated with parasites from donor mice that are matched to the host’s SCN rhythms . Twelve infections were established in the manner used in our main experiment ( eight-week-old male mice , strain MF1 , intravenously infected with 1 x 107 P . chabaudi DK parasitised RBC ) , except that donor SCN rhythms were matched to the experimental host’s SCN rhythm and hosts had access to food day and night . Densities of parasites were quantified from blood smears and RBC density by flow cytometry on day 6 and 9 PI , respectively . We chose to compare parasite density in matched infections to LF and DF infections on day 6 PI because parasites are approaching peak numbers in the blood ( before host immunity starts to clear infections ) and their high density facilitates accurate quantification when using microscopy . This experiment probes whether host immune responses mounted during the early phase of malaria infection could impose development rhythms upon parasites . We entrained N = 86 eight-week-old female mice , strain MF1 , to either a reverse lighting schedule ( lights on 7pm , lights off 7am , N = 43 ) or a standard lighting schedule ( lights on 7am , lights off 7pm , N = 43 ) . Donor mice , infected with P . chabaudi genotype AS , were entrained to a standard lighting schedule to generate infections matched and 12 hours mismatched relative to the SCN in the experimental mice . Mice were intravenously injected with 1 x 107 parasitised RBC at ring stage . Genotype AS has intermediate virulence [39] and was used to ensure immune responses were elicited by day 4 PI . We terminally sampled 4 mice every 3 hours over 30 hours starting on day 4 PI , taking blood smears , red blood cell counts and collecting plasma for Luminex cytokine assays . Cytokines were assayed by the Human Immune Monitoring Centre at Stanford University using mouse 38-plex kits ( eBiosciences/Affymetrix ) and used according to the manufacturer’s recommendations with modifications as described below . Briefly , beads were added to a 96-well plate and washed in a Biotek ELx405 washer . 60uL of plasma per sample was submitted for processing . Samples were added to the plate containing the mixed antibody-linked beads and incubated at room temperature for one hour followed by overnight incubation at 4°C with shaking . Cold and room temperature incubation steps were performed on an orbital shaker at 500–600 rpm . Following the overnight incubation , plates were washed as above and then a biotinylated detection antibody was added for 75 minutes at room temperature with shaking . Plates were washed as above and streptavidin-PE was added . After incubation for 30 minutes at room temperature a wash was performed as above and reading buffer was added to the wells . Each sample was measured as singletons . Plates were read using a Luminex 200 instrument with a lower bound of 50 beads per sample per cytokine . Custom assay control beads by Radix Biosolutions were added to each well . We staged the parasites from the blood smears collected from the infections used to assay cytokines ( above ) to investigate their synchrony during rescheduling . The infections from mismatched donor mice began 12 hours out of phase with the host SCN rhythms and the CoG for ring stage parasites reveals they had become rescheduled by 6 hours on day 4 PI . We focus on the ring stage as a phase marker–for the analysis of synchrony in these data and the divergence between LF and DF parasites–because rings are the most morphologically distinct , and so , accurately quantified , stage . In a third additional experiment , we entrained 10 eight-week-old male mice , strain MF1 , to a standard lighting schedule for 2 weeks before randomly allocating them to one of two feeding treatments . One group ( N = 5 ) were allowed access to food between ZT 0 and ZT 12 ( equivalent to the LF group in the main experiment ) and the other group ( N = 5 ) allowed access to food between ZT 12 and ZT 0 ( equivalent to the DF group ) . After 10 days of food restriction we recorded blood glucose concentration every 2 hours for 30 hours , using an Accu-Chek Performa glucometer . We used CircWave ( version 1 . 4 , developed by R . A . Hut; available from https://www . euclock . org ) to characterise host and parasite rhythms , and R v . 3 . 1 . 3 ( The R Foundation for Statistical Computing , Vienna , Austria ) for analysis of summary metrics and non-circadian dynamics of infection . Specifically , testing for rhythmicity , estimating CoG ( a reference point to compare circadian rhythms ) for host ( body temperature , locomotor activity , blood glucose concentration ) and parasite rhythms , and amplitude for parasite stage proportions , was carried out with CircWave for each individual infection . However , the cytokine data display high variation between mice ( due to a single sample from each mouse ) so we calculated a more robust estimate of phase than CoG by fitting a sine curve with a 24h period ( using CircWave ) and finding the maxima . Linear regression models and simultaneous inference of group means ( using the multcomp R package ) were run with R to compare summary measures that characterise rhythms , parasite performance , glucose concentration and disease severity . R was also used to construct and compared linear mixed effects models using which included mouse ID as a random effect ( to account for repeated measures from each infection ) to compare dynamics of parasite and RBC density throughout infections , and glucose concentration throughout the day . All procedures were carried out in accordance with the UK Home Office regulations ( Animals Scientific Procedures Act 1986; project licence number 70/8546 ) and approved by the University of Edinburgh . Euthanasia was performed using anaesthesia ( combination of Medetomidine and Ketamine ) followed by cervical dislocation and rigor mortis as confirmation of death . | How cycles of asexual replication by malaria parasites are coordinated to occur in synchrony with the circadian rhythms of the host is a long-standing mystery . We reveal that rhythms associated with the time-of-day that hosts feed are responsible for the timing of rhythms in parasite development . Specifically , we altered host feeding time to phase-shift peripheral rhythms , whilst leaving rhythms driven by the central circadian oscillator in the brain unchanged . We found that parasite developmental rhythms remained synchronous but changed their phase , by 12 hours , to follow the timing of host feeding . Furthermore , our results suggest that parasites themselves schedule rhythms in their replication to coordinate with rhythms in glucose in the host’s blood , rather than have rhythms imposed upon them by , for example , host immune responses . Our findings reveal a novel relationship between hosts and parasites that if disrupted , could reduce both the severity and transmission of malaria infection . | [
"Abstract",
"Introduction",
"Results",
"and",
"discussion",
"Materials",
"and",
"methods"
] | [
"medicine",
"and",
"health",
"sciences",
"parasite",
"replication",
"tropical",
"diseases",
"parasitic",
"diseases",
"parasitic",
"protozoans",
"biological",
"locomotion",
"parasitology",
"circadian",
"oscillators",
"protozoans",
"physiological",
"parameters",
"body",
"temperature",
"chronobiology",
"malarial",
"parasites",
"biochemistry",
"circadian",
"rhythms",
"eukaryota",
"physiology",
"biology",
"and",
"life",
"sciences",
"malaria",
"organisms"
] | 2018 | Timing of host feeding drives rhythms in parasite replication |
Phenotypic variance heterogeneity across genotypes at a single nucleotide polymorphism ( SNP ) may reflect underlying gene-environment ( G×E ) or gene-gene interactions . We modeled variance heterogeneity for blood lipids and BMI in up to 44 , 211 participants and investigated relationships between variance effects ( Pv ) , G×E interaction effects ( with smoking and physical activity ) , and marginal genetic effects ( Pm ) . Correlations between Pv and Pm were stronger for SNPs with established marginal effects ( Spearman’s ρ = 0 . 401 for triglycerides , and ρ = 0 . 236 for BMI ) compared to all SNPs . When Pv and Pm were compared for all pruned SNPs , only BMI was statistically significant ( Spearman’s ρ = 0 . 010 ) . Overall , SNPs with established marginal effects were overrepresented in the nominally significant part of the Pv distribution ( Pbinomial <0 . 05 ) . SNPs from the top 1% of the Pm distribution for BMI had more significant Pv values ( PMann–Whitney = 1 . 46×10−5 ) , and the odds ratio of SNPs with nominally significant ( <0 . 05 ) Pm and Pv was 1 . 33 ( 95% CI: 1 . 12 , 1 . 57 ) for BMI . Moreover , BMI SNPs with nominally significant G×E interaction P-values ( Pint<0 . 05 ) were enriched with nominally significant Pv values ( Pbinomial = 8 . 63×10−9 and 8 . 52×10−7 for SNP × smoking and SNP × physical activity , respectively ) . We conclude that some loci with strong marginal effects may be good candidates for G×E , and variance-based prioritization can be used to identify them .
Gene-environment ( G×E ) interactions may contribute to complex diseases , but their detection has proven challenging; hence , a variety of approaches have been developed to enhance power . Most G×E analyses focus on loci that are strong biological candidates [1] or those with highly significant marginal effects [2] . The latter approach is attractive because these loci are available in many large cohorts , and can be conveniently followed-up with interaction analyses if environmental data are accessible . Moreover , selecting SNPs with strong and reproducible marginal effect signals is a pragmatic data-reduction step that may improve power [3] , although this approach risks omitting other promising candidates [4] . In a linear regression setting , the presence of interaction effects drives phenotypic variance heterogeneity by genotype [3 , 5] . Exploiting variance heterogeneity as a signature of interactions is appealing because , unlike standard approaches for assessing G×E interactions , no explicit information about environmental exposures is needed [6] and multiple exposures can be simultaneously considered . Here we explored whether loci identified in large-scale genome-wide association studies ( GWAS ) of blood lipids and body mass index ( BMI ) are strong candidates for G×E interactions by comparing genome-wide variance heterogeneity P-value distributions generated using Levene’s test against P-value distributions for marginal effects and explicit G×E interaction effects ( for smoking and physical activity ) .
We assessed between-genotype variance heterogeneity for up to 1 , 927 , 671 directly genotyped or imputed SNPs ( HapMap II CEU reference panel [7] ) that passed quality control ( QC ) . Meta-analyses of Levene’s test summary statistics [8] were performed for BMI ( n≤44 , 211 participants ) , and blood concentrations of high-density lipoprotein cholesterol ( HDL-C ) ( n≤34 , 315 ) , low-density lipoprotein cholesterol ( LDL-C ) ( n≤34 , 180 ) , total cholesterol ( TC ) ( n≤34 , 318 ) and triglycerides ( TG ) ( n≤34 , 110 ) . We then obtained marginal effects results for the same index traits and SNPs from publicly available GWAS summary data from the GIANT ( Genetic Investigation of ANthropometric Traits ) Consortium [9] and GLGC ( Global Lipids Genetics Consortium ) [10 , 11] . We compared the genome-wide marginal effects with between-genotype variance heterogeneity results for each of the five cardiometabolic traits by calculating the association between marginal effects ( Pm ) and variance heterogeneity ( Pv ) P-values using the rank-based Spearman correlation ( ρ ) . This was done using a set of 42 , 710 pruned SNPs produced using the--indep-pairwise command in PLINK ( see Materials and Methods ) to account for linkage disequilibrium ( LD ) among variants . As shown in Table 1 ( see also Fig 1A and S1 Table ) , the Spearman’s ρ for the association between Pm and Pv for all pruned SNPs was of very small magnitude and only statistically significant for BMI . The exclusion of SNPs based on progressively more conservative Pm thresholds ( Pm<0 . 05; Pm<10−4; previously established loci with Pm<5×10−8 in external datasets ) , saw corresponding improvements in the magnitude of these correlations , which were statistically significant for all traits except TC when focusing on previously established loci . The BMI correlation at the Pm<0 . 05 threshold , as well as the test of equality with ρ for all SNPs , was statistically significant , suggesting concordance between marginal and variance signals at a nominal level of significance . The odds ratio ( OR ) for a SNP to have both Pm<0 . 05 and Pv<0 . 05 as compared to Pv≥0 . 05 was 1 . 33 ( 95% CI: 1 . 12 , 1 . 57 ) for BMI while the 95% CIs of ORs for other traits included 1 . On the other hand , the P-value for a non-zero ρ for TG was statistically significant when focusing on the established loci and at Pm<10−4 , suggesting concordance between marginal and variance signals at more conservative Pm thresholds . We further compared Pm with interaction P-values from exposure-specific ( smoking and physical activity ) genome-wide interaction tests for BMI ( Pint ) ; this was only done for BMI owing to the requirement for an adequately powered external dataset ( such a dataset was accessible through the GIANT consortium ) ( Table 2 ) . Marginal effects GWAS were performed by strata of smokers vs . non-smokers and physically active vs . inactive participants ( n = 210 , 316 European-ancestry adults [12] ) respectively , and a heterogeneity test [12] was used to generate exposure specific Pint distributions . Spearman ρ for the pruned set of SNPs in the SNP × physical activity and the SNP × smoking analyses were low and not statistically significant ( Table 2 ) . We also compared Pint values and Pv values for BMI . Spearman’s ρ for the pruned set of SNPs were low and not statistically significant . We next tested if the number of previously established marginal effect SNPs ( Pm<5×10−8 ) that were also nominally significant ( Pv<0 . 05 ) for variance heterogeneity was greater than expected by chance ( Tables 3 and 4 , Fig 1 ) . For 4 out of the 5 index traits , we observed enrichment at the lower end of the Pv distribution ( Pv<0 . 05 ) for the established GWAS-derived lead SNPs . Thus , the nominally significant regions of the Pv distributions were generally enriched for GWAS-derived loci . We also performed enrichment analyses to test if previously established marginal effects SNPs ( Pm<5×10−8 ) are enriched for nominally significant ( Pint<0 . 05 ) interactions in the SNP × physical activity or SNP × Smoking analyses , but no enrichment was observed ( Table 3; Fig 1B ) . By contrast , for the physical activity and smoking interaction tests ( using all pruned SNPs ) , the lower end of the Pint distribution ( Pint<0 . 05 ) was enriched with SNPs that were nominally significant in the Levene’s test analysis ( Pv<0 . 05 ) ( Table 4 ) . This enrichment translated into an OR of 1 . 08 ( 95% CI: 1 . 01 , 1 . 14 ) for a SNP to have Pint<0 . 05 given Pv<0 . 05 vs . Pv≥0 . 05 for SNP × physical activity interaction . The corresponding OR for the SNP × smoking interaction test was not significant ( OR = 1 . 02; 95% CI: 0 . 96 , 1 . 08 ) . Finally , in the pruned SNP-set we used the Mann–Whitney U test to probe for systematic differences in Pv and Pm ranks . P-values were ordered from least significant to most significant , and the lowest 100th centile ( i . e . the most significantly associated SNPs ) was compared to the remaining 99th percentile for each of the five traits . For BMI , SNPs in the lowest 100th centile of the Pm distribution had markedly higher Pv ranks ( i . e . more significant Pv ) than the remaining SNPs ( PMann–Whitney = 1 . 46×10−5; Table 5 ) . Even when excluding previously established lead SNPs ( Pm<5×10−8 ) for BMI ( or SNPs +/-500kb proximal ) , SNPs from the lowest 100th centile of the Pm rank-ordered distribution had higher Pv ranks than the remaining SNPs ( PMann–Whitney = 4 . 30×10−4; Table 5 ) . Conversely , no difference in Pv ranks was observed for SNPs from the lowest 100th centile of the Pm rank-ordered distribution for the four blood lipid traits; this may reflect trait-specific G×E effects or differences in statistical power by trait . No differences in Pv ranks between SNPs from the lowest 99th centile of the Pm rank-ordered distribution compared to SNPs from the 98th to 1st centiles of the distribution were observed for any trait ( PMann–Whitney>0 . 05; Table 5 ) . Similarly , no difference in Pm ranks was observed for SNPs from the lowest 100th centile of the Pv rank-ordered distribution for any traits ( PMann–Whitney>0 . 05; Table 6 ) . To assess whether a trait with a non-normal distribution ( e . g . BMI ) or strong marginal associations could cause spurious association between the marginal and variance signals , we recapitulated the analysis pipeline ( correlation analysis , enrichment analysis , comparisons of rank Pm and Pv values ) in simulations described in the Materials and Methods . Careful assessment of results emanating from these simulations did not reveal evidence of type I error inflation caused by the non-normal distribution of an outcome trait nor strong marginal effects . For instance , we extracted correlation P-values of Pm , Pv and Pint generated from 5 , 000 simulations . QQ-plots of the 5 , 000 correlation P-values , 2 , 500 binomial P-values , and 2 , 500 Mann-Whitney U test P-values revealed no inflation ( S1A–S1C Fig , S2A and S2B Fig and S3A and S3B Fig , respectively ) . Repeating these analyses on subsets of SNPs with low Pm values did not materially change the results .
Collectively , our analyses highlight a few variants with genome-wide significant marginal effects that may be strong candidates for G×E interactions owing to their strong concurrent variance heterogeneity P-values . For BMI , such SNPs are also overrepresented in the nominally significant part of the Pv distribution . FTO is an excellent example , as it conveys strong marginal effects [13] , exhibits high between-genotype heterogeneity here ( Tables 2 and 3 and Fig 1B ) and elsewhere [5] , and reportedly interacts with physical activity , diet and other lifestyle exposures [2 , 14 , 15] and is associated with macronutrient intake [16 , 17] . Although variance heterogeneity tests are potentially powerful screening tools for G×E interactions , like most interaction tests , they may be bias prone . For example , apparent differences in phenotypic variances across genotypes may be caused by scaling , particularly when the phenotypic means also differ substantially [18] , such that the per-genotype means and variances for index traits are correlated . However , where necessary we transformed variables , and the correlations between Pm and Pv were generally weak , excluding this as a likely source of bias . Using simulated data , we investigated whether the non-normal distribution of a trait can cause a spurious association between marginal and variance signals , which we show is highly improbable . Through further simulations , we assessed whether SNPs with large marginal effects inflate Pv , but observed no inflation , indicating that large genetic marginal effects do not artificially inflate variance heterogeneity to a meaningful extent , and SNPs with low Pm and low Pv-values are thus likely to be strong candidates for G×E interactions , at least in the case of BMI . It might also be that combining populations from ancestral ( e . g . , hunter-gatherers ) and contemporary environments increases variance heterogeneity owing to diversity in population substructure rather than G×E interactions per se [19] . However , this seems unlikely here , as the cohorts examined are from Westernized European-ancestry populations . There are several additional explanations for between-genotype variance heterogeneity , such as variance misclassification that can occur when the index variant is located within a haplotype containing rare functional variants that convey strong marginal effects [5] . Hence , although variance heterogeneity tests represent a useful data-reduction step , before conclusions are drawn about the presence or absence of G×E interactions , index variants should be validated by testing their interactions with explicit environmental exposures , as we did here with smoking and physical activity . However , genome-wide G×E interactions datasets are not comprised of functionally validated G×E interactions , as no such resource is currently available for human complex traits . This limitation inhibits the extent to which causal effects can be attributed to the top-ranking loci and their interactions with smoking or physical activity . We conclude that the common approach of prioritizing loci with established genome-wide significant association signals without further discrimination for G×E interaction analyses might be useful , but the efficiency of such analyses could be substantially improved by focusing on variants with low P-values for both variance heterogeneity and marginal effects . We provide these rankings here to facilitate this approach .
We performed a genome-wide search for SNPs whose associations with the following traits are characterized by high between-genotype variance heterogeneity: BMI , TC , TG , HDL-C and LDL-C . The variance heterogeneity analyses were performed using Levene’s test [20] in up to 44 , 211 participants of European descent from seven population-based cohorts . Descriptions of these cohorts are presented in S2 Table . To minimize bias that might result from unequal sample sizes between SNPs when calculating the correlations between the P-values from the marginal ( Pm ) and variance heterogeneity ( Pv ) meta-analyses , we restricted the sample size for analyses to 26 , 000 participants for BMI and to 24 , 000 participants for lipid traits ( S4 Fig ) . A detailed summary of sample sizes , genotyping platforms , genotype calling algorithms , sample and SNP quality control filters , and analysis software for all participating cohorts are provided in S2 and S3 Tables . For each individual , SNPs were imputed using the CEU reference panel of HapMap II [7] ( S2 Table ) . We excluded SNPs with low imputation quality ( below 0 . 3 for MACH , 0 . 4 for IMPUTE , and 0 . 8 for PLINK imputed data ) , Hardy-Weinberg equilibrium P <10−6 , directly genotyped SNP call rate < 95% , and minor allele frequency ( MAF ) < 1% . We identified SNPs that have been robustly associated ( P<5x10-8 ) with the five cardiometabolic traits in European ancestry populations: 77 SNPs associated with BMI discovered by GIANT [9]; and 58 SNPs associated with LDL-C , 71 SNPs associated with HDL-C , 74 SNPs associated with TC , and 40 SNPs associated with TG [10 , 11] discovered by GLGC . We used Levene’s test [20] to identify SNPs that show heterogeneity of phenotypic variances ( σi2 ) across the three genotype groups at each SNP locus ( i = 0 , 1 , or 2 ) . We first log10 transformed all five traits followed by a z-score transformation by subtracting the sample mean and dividing by the sample standard deviation ( SD ) , and further Winsorized the z-score values at 4 SD . The transformed phenotype Y was then used to calculate Z , defined by the absolute deviation of each participant’s phenotype from the sample mean of his or her respective genotype group at a given SNP locus . For each trait , participating cohorts provided the necessary summary statistics for each genotype at each marker [8] . Specifically , the per genotype group counts ( n0s , n1s , n2s ) , per genotype means ( Z¯0s , Z¯1s , Z¯2s ) , and per genotype group variances of Z ( σ0s2 , σ1s2 , σ2s2 ) were centrally collected and meta-analyzed . The minimum number of observations per genotype group required is 30 participants per cohort . Meta-analyses were performed using the following formula , derived previously [8]: L= ( N−3 ) ( 3−1 ) ⋅ ( ∑i=02γi⋅ ( ∑sZ¯is⋅ωis ) 2− ( ∑i=02∑sZ¯is⋅ωis⋅γi ) 2 ) ∑i=02 ( ∑s ( σZis2⋅ωis−σZis2N⋅γi+Z¯is2⋅ωis ) ⋅γi− ( ( ∑sZ¯is⋅ωis ) 2⋅γi ) ) Where N is the combined sample size , Z¯is and σZis2 are the sample mean and variance of Z in the ith genotype group of the sth study , respectively . When combining summary-level data to calculate the Levene’s test statistics L , the following natural weights ωis and γi were calculated: ωis=nis∑snis and γi=niN , where ni the sum of genotype counts in the ith genotype group across all participating cohorts . These weights are determined by the frequency of the marker amongst the cohorts , such that the sum of both weights is equal to 1 , i . e . ∑sωis=1 and ∑iγi=1 . The meta-analysis Levene’s test P-value is obtained by comparing L to an F-distribution with df1 = 2 and df2 = N-3 . Marginal effects P-values for BMI and the relevant lipid traits were obtained from publically available GWAS summary data from the GIANT [9] and GLGC [10 , 11] consortia , respectively ( all cohorts included here in the Levene’s meta-analysis were also included in the GIANT and GLGC datasets ) . To illustrate our findings , we rank-ordered the P-values ( from lowest to highest ) from both marginal effects and variance effects analyses for all 1 , 927 , 671 SNPs so that the lowest P-value for a given trait was assigned a rank equal to the lowest 100th centile . These rank-scaled distributions for Pm for all five traits are presented in Fig 1 . We calculated Spearman’s correlations for each of the five cardiometabolic traits between Pm and Pv . This was done using a pruned set of SNPs . Pruning was performed in the TwinGene cohort using the--indep-pairwise 50 5 0 . 1 command in PLINK [21] by calculating LD ( r2 ) for each pair of SNPs within a window of 50 SNPs , removing one of a pair of SNPs if r2>0 . 1; we proceeded by shifting the window 5 SNPs forwards and repeating the procedure . Spearman’s correlations were computed for categories of SNPs: i ) all pruned SNPs , ii ) the subset of SNPs that was nominally significant ( Pm<0 . 05 ) in the marginal effects analysis , iii ) the subset of SNPs with Pm<10−4 in the marginal effects analysis , and iv ) SNPs that were previously established in conventional marginal effects GWAS meta-analyses ( Pm<5×10−8 ) . We also compared Spearman’s correlations between these categories of SNPs using the test for equality of two correlations [22] . Next , we performed enrichment analyses to test if there was a higher number of established SNPs in the nominally significant variance P-value ( Pv<0 . 05 ) distribution than expected by chance under the binominal distribution . We also tested if there is a difference in Pv ranks for SNPs from the lowest 100th centile of the Pm rank-ordered distribution for all five traits and the rest of SNPs in the pruned set of SNPs using the Mann–Whitney U test , including and excluding established SNPs ( or SNPs that were +/-500kb from the reported lead SNP ) . This analysis was repeated for SNPs from the 99th centile vs SNPs from 1st to 98th centiles of the Pm rank-ordered distribution . The same Mann–Whitney U tests were used to study differences in Pm ranks for SNPs from the lowest 100th and 99th centiles of the Pv rank-ordered distribution and the rest of SNPs in the pruned set of SNPs . All analyses were performed using Stata 12 ( StataCorp LP , TX , USA ) , unless specified otherwise . We used now published data from 210 , 316 European-ancestry adults ( from the GIANT consortium ) pertaining to marginal effects meta-analyses for BMI that had been performed separately by strata of smoking ( 45 , 968 smokers vs . 164 , 355 non-smokers ) [23] . The genetic marginal effect estimates , calculated separately within each of the two strata , were compared using a heterogeneity test [12] to infer the presence or absence of SNP × smoking interaction effects . The same analyses were performed using physical activity as a binary stratifying variable in up to 180 , 287 European-ancestry adults ( 42 , 065 physically active vs . 138 , 222 physically inactive ) [24] . We calculated Spearman correlations between the P-values derived from the marginal effects meta-analysis and the Pint from the interaction effects meta-analysis ( i . e . , the between-strata heterogeneity test for SNP × smoking and SNP × physical activity interactions from the GIANT consortium ) ; these tests were undertaken for all SNPs and those SNPs that were nominally significant ( Pm<0 . 05 ) in the marginal effects analysis . We then performed enrichment analyses to test if the numbers of nominally significant ( Pint<0 . 05 ) GWAS-derived SNPs from both SNP × physical activity and SNP × smoking analyses were greater than expected by chance under the binomial distribution . We further calculated the OR of having Pint<0 . 05 given Pv<0 . 05 versus Pv≥0 . 05 both SNP × physical activity and SNP × smoking interaction analyses in a pruned set of TwinGene SNPs produced using the—indep-pairwise 50 5 0 . 8 command in PLINK [21] . Thereafter , we calculated the average rank for each SNP’s ranking on the Pint rank-ordered distributions from the SNP × smoking and SNP × physical activity interaction analyses and performed enrichment analysis using these average ranks with >95th centile instead of Pint<0 . 05 as the cut-off . We simulated genetic data for 44 , 000 individuals from a pruned set of 50 , 335 SNPs with allele frequencies , effect estimates and Pm values drawn from the GIANT consortium . We generated an outcome trait by summing the products of the simulated allele counts and effect estimates over all SNPs for each individual , and subsequently added a randomly generated non-normal error term such that the trait resembles the observed distribution of the transformed BMI trait used in the main ( real data ) analyses . We also simulated a fixed binary interacting factor with 30% prevalence . Using this simulated dataset , we calculated Pm , Pv and Pint values for each SNP and undertook i ) pairwise Spearman correlation analyses between Pm , Pv and Pint values ( 5 , 000 simulations ) , ii ) enrichment analysis using binomial tests ( 2 , 500 simulations ) and iii ) Mann-Whitney U tests to determine systematic differences in Pv and Pm ranks ( 2 , 500 simulations ) . Following the same pipeline , we created additional simulated datasets narrowing down SNPs to i ) those with Pm values from the lowest percentile ( n = 504; highest Pm = 5×10−3 ) and to ii ) genome-wide significant SNPs ( n = 71; Pm<5×10−8 ) , and tested the pairwise Spearman correlation for Pm , Pv and Pint values ( 1 , 000 simulations for both sets ) . Simulations were run using the statistical software R ( v . 3 . 3 . 2 ) . [25] | Most contemporary studies of gene-environment interactions focus on gene variants that are known to bear strong and reliable associations with the traits of interest . The strategy is intuitive because it helps limit the number of tests performed by focusing on a relatively small number of gene variants . However , this approach is predicated on an implicit assumption that these loci are strong candidates for interactions owing to their established relationships with the index traits . The counter-argument is that , because these loci have highly consistent signals within and between populations that vary by environmental characteristics , the probability that these variants interact with other factors is low . The current analysis tests whether variants with strong marginal effects signals ( i . e . , those prioritized through conventional genome-wide association analyses ) are strong or weak candidates for gene-environment interactions . Here we describe analyses focused on lipids and BMI that test this hypothesis by comparing marginal effect signals with variance effect signals and those derived from explicit genome-wide , gene-environment interaction analyses . We conclude that for BMI , there are features of the top-ranking marginal effect loci that render them stronger candidates for interactions than is true of variants with weaker marginal effects signals . These findings are likely to help optimize the efficiency of future gene-environment interaction analyses by providing evidence-based rankings for strong candidate loci . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"genome-wide",
"association",
"studies",
"medicine",
"and",
"health",
"sciences",
"quantitative",
"trait",
"loci",
"sociology",
"social",
"sciences",
"physical",
"activity",
"mathematics",
"statistics",
"(mathematics)",
"genome",
"analysis",
"research",
"and",
"analysis",
"methods",
"public",
"and",
"occupational",
"health",
"genomic",
"signal",
"processing",
"lipids",
"proteins",
"mathematical",
"and",
"statistical",
"techniques",
"lipoproteins",
"statistical",
"methods",
"consortia",
"genetic",
"loci",
"cholesterol",
"biochemistry",
"signal",
"transduction",
"cell",
"biology",
"meta-analysis",
"genetics",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"genomics",
"cell",
"signaling",
"computational",
"biology",
"human",
"genetics"
] | 2017 | Ranking and characterization of established BMI and lipid associated loci as candidates for gene-environment interactions |
Trypanosoma cruzi , an intracellular protozoan parasite that infects humans and other mammalian hosts , is the etiologic agent in Chagas disease . This parasite can invade a wide variety of mammalian cells . The mechanism ( s ) by which T . cruzi invades its host cell is not completely understood . The activation of many signaling receptors during invasion has been reported; however , the exact mechanism by which parasites cross the host cell membrane barrier and trigger fusion of the parasitophorous vacuole with lysosomes is not understood . In order to explore the role of the Low Density Lipoprotein receptor ( LDLr ) in T . cruzi invasion , we evaluated LDLr parasite interactions using immunoblot and immunofluorescence ( IFA ) techniques . These experiments demonstrated that T . cruzi infection increases LDLr levels in infected host cells , inhibition or disruption of LDLr reduces parasite load in infected cells , T . cruzi directly binds recombinant LDLr , and LDLr-dependent T . cruzi invasion requires PIP2/3 . qPCR analysis demonstrated a massive increase in LDLr mRNA ( 8000 fold ) in the heart of T . cruzi infected mice , which is observed as early as 15 days after infection . IFA shows a co-localization of both LDL and LDLr with parasites in infected heart . These data highlight , for the first time , that LDLr is involved in host cell invasion by this parasite and the subsequent fusion of the parasitophorous vacuole with the host cell lysosomal compartment . The model suggested by this study unifies previous models of host cell invasion for this pathogenic protozoon . Overall , these data indicate that T . cruzi targets LDLr and its family members during invasion . Binding to LDL likely facilitates parasite entry into host cells . The observations in this report suggest that therapeutic strategies based on the interaction of T . cruzi and the LDLr pathway should be pursued as possible targets to modify the pathogenesis of disease following infection .
The Low-Density Lipoprotein receptor ( LDLr ) ( UniProtKB: P01130 ) is a cell surface glycoprotein that plays a critical role in cholesterol homeostasis [1] . LDLr is the patriarch of an entire class of receptors called LDL receptor related proteins ( LRPs ) that contain similar structural modules [2] . The mature LDLr is a modular type I transmembrane protein of 839 amino acids and is composed of a number of functionally distinct domains that can function independently of each other [3] , [4] . The N-terminus of the receptor contains three types of extracellular modules consisting of cysteine-rich repeats , three epidermal growth factor precursor ( EGFP ) regions , and O-linked oligosaccharides followed by a membrane spanning domain . The C-terminus domain of the receptor contains a signal sequence ( NPXY ) that is needed for receptor binding to clathrin pits and internalization [5] . The most important physiologic ligand for the receptor is Low Density Lipoprotein ( LDL ) . Members of the LDLr superfamily bind a variety of ligands including lipoproteins , proteinases and proteinase-inhibitor complexes , and transport them into endosomes in the cell [6] . The functional properties of LDLr family members include clustering of receptors into clathrin-coated pits mediated by adaptor proteins , a pH sensitive ligand uncoupling mechanism , and recycling of the receptors back to the cell surface after dissociation of ligands . The transcription of LDL receptor is regulated by intracellular cholesterol and extracellular stimuli such as TNFα , IL-1β , TGF-β and insulin [7]–[9] . The signaling pathways leading to activation of Protein Kinase C ( PKC ) , Protein Kinase A ( PKA ) and intracellular Ca2+ mobilization are also involved in LDLr expression [10] . LDL-containing immune complexes upregulate LDLr transcription . Interestingly , Pseudomonas exotoxin A and a minor group of rhinoviruses have been reported to utilize LDLr members to enter into host cells [11] . Chagas disease , caused by the obligate intracellular parasite Trypanosoma cruzi , affects millions of people in Mexico , Central and South America . During acute infection clinical progressions may include myocarditis and/or meningoencephalitis , although most patients are asymptomatic . Chronic manifestations include irreversible cardiomyopathy and megasyndromes [12]–[14] . Current antiparasitic treatments are not effective for chronic infection . T . cruzi invades a wide variety of mammalian cells including macrophages , smooth muscle cells , striated muscle cells , fibroblasts , cardiomyocytes , and adipocytes [15] , [16] . In its vertebrate host this parasite is transmitted from cell to cell by non-replicating motile trypomastigotes which are capable of invading host cells; following invasion trypomastigotes transform into amastigotes which replicate intracellularly . In contrast to many intracellular pathogens that avoid contact with host cell lysosmes , T . cruzi requires the low pH environment of lysosomes to initiate egress from the parasitophorous vacuole and delivery to the host cell cytoplasm where replication begins [17]–[19] after approximately 24 hours post-invasion . The molecular mechanism ( s ) of invasion by this parasite and the associated regulatory pathways have been the subject of intense investigation for many years . Two models of invasion , a lysosomal dependent , and a phosphotidylinositol phosphates ( PIPs ) pathway have been suggested for T . cruzi invasion . The lysosomal dependent pathway postulates that T . cruzi elicited signals evoke the early recruitment of host cell lysosomes to the cytosolic face of the plasma membrane at the parasite attachment site where the localized fusion of lysosomes provide membrane for the nascent parasitophorous vacuole [20] . The proposed lysosome independent PIP dependent parasite entry pathway is based on host cell PI3K signaling as a key regulator of T . cruzi invasion [20] , [21] . However , neither of these models explains the precise mechanism ( s ) by which this parasite traverses the host cell permeability barrier and interacts with the lysosomal compartment . Recent studies have demonstrated that T . cruzi activates many cell membrane receptors such as Toll-like receptors ( TLRs ) , kinins ( B1/B2 sub types ) , receptor tyrosine kinases , TGF and EGF receptors and that the activity of these receptors is required for optimal parasite binding and/or invasion [22]–[25] . In addition , invasion results in the activation of ERK/MAPK signaling pathways . Taken together these studies reveal that T . cruzi activates many signaling pathways involving diverse receptors on the host cell surface in preparation for internalization . Our study highlights , for the first time , that LDLr is involved in the trafficking of lysosomes to the parasitophorous vacuole containing trypomastigotes and that inhibition or disruption of LDLr affects the intracellular parasite load . We also report that LDLr expression is upregulated in infected mouse hearts and LDL/LDLr is associated with the amastigotes ( pseudocysts ) in the heart tissue of infected mice . The accumulation of LDL and LDLr in the heart probably contributes to the pathogenesis of chagasic heart disease . The LDL/LDLr pathway could represent a new therapeutic target for modulating T . cruzi infection .
The Brazil strain of T . cruzi was maintained in C3H/He mice ( Jackson Laboratories , ME ) . Six to 8 week old male CD-1 mice were obtained from Charles River Laboratories ( Wilmington , MA ) and infected IP with 5×104 trypomastigotes . The serum and heart tissues were collected at 15 , 20 and 30 days p . i . The mice were anesthetized with isoflurane and about 75 µl of blood is collected from the orbital venous sinus . The mice were then observed for recovery from anesthesia and returned to their cages . The parasitemia was determined using a Neubauer hemocytometer . Hearts were fixed in 10% buffered formalin and paraffin sections were stained with IFA . The animal experiments were approved by the Institutional Animal Care and Use 200Committees ( IACUC ) of Albert Einstein College of Medicine ( No . 20100204 ) . Parasites were also maintained in L6E9 myoblasts as previously described [26] . Human Foreskin Fibroblast ( HFF ) ( ATCC CRL 1475 ) , rat cardiomyocyte H9c2 and 3T3-L1 ( ATCC CL 173 ) cell lines are maintained in our laboratory using standard methods as previously published [27] , [28] . All the cell culture reagents used in these experiments were obtained from Cellgro ( Mediatech Inc . ) , primary antibodies were obtained from Abcam ( MA ) and secondary fluorescence antibodies were obtained from Invitrogen ( CA ) unless other suppliers are specifically mentioned in the text . For each experiment a minimum of 4 mice were used per group and each experiment has been repeated thrice . Cell lysates were prepared as previously described [29] . An aliquot of each sample ( 40 µg protein ) was subjected to 7 . 5% SDS-PAGE and the proteins were transferred to nitrocellulose filters for immunoblot analysis . LDLr specific rabbit monoclonal antibodies ( 1∶2000 dilution , ab52818 Abcam , MA ) and horseradish peroxidase- conjugated goat anti-rabbit immunoglobulin ( 1∶5000 dilution , Amersham Biosciences ) were used to detect specific protein bands ( explained in Figure Legends ) using a chemiluminescence system [29] . GAPDH ( 1∶5000 dilution , mouse monoclonal Ab8245 , MA and secondary antibody horseradish peroxidase conjugated goat anti-rabbit 1∶2000 dilution , Amersham Biosciences ) was used to normalize protein loading . Fibroblasts were cultured on cover slips to 80% confluence and then infected with trypomastigotes ( 3 . 1×106/cm2 surface area of culture plates ) for 10 , 20 and 30 minutes . The fibroblast cultures were fixed with 4% paraformaldehyde , permeabilized with 0 . 1% Triton ×100 ( 30min ) and stained for LDLr/clathrin/PIP/LAMP2/LDL using specific primary antibodies rabbit monoclonal to LDLr ( Ab52818 ) , mouse monoclonal to Clathrin ( 1∶200 , Ab2731 ) , mouse monoclonal to PIP2 ( 1∶300 , Ab11039 ) , rat monoclonal LAMP1 ( 1∶150 , Hybridoma bank , 1D4B ) , rat monoclonal to LAMP2 ( 1∶150 , Hybridoma bank , ABL93 ) and goat anti human LDL ( 1∶10 , Sigma L 8016 ) respectively used with the concentrations as recommended by the manufacturers and Alexa fluor 594 ( goat anti rabbit or anti mouse IgG 1∶500 dilution; Invitrogen , CA ) , or Alexa 488 ( goat anti rabbit or anti mouse or anti rat IgG 1∶500 dilution; Invitrogen CA ) . The cells were stained with DAPI ( blue ) to detect nuclei following manufacturer's protocols ( www . abcam . com/technical ) . Images were obtained and analyzed by fluorescence microscopy using an inverted Olympus IX71 with a HQ2 CCD camera and a Nikon Microphot-FXA with Spot camera software . IFA of paraffin embedded tissues were performed as previously published [30] . HFF cells were pretreated with recombinant proprotein convertase subtilisin/kexin type 9 ( 0 . 5ug of PCSK9/cm2 surface area ) for 1h at 37°C . PCSK9 treated and untreated cells were incubated with trypomastigotes ( 3 . 1×106/cm2 surface area ) for 1h at 37°C . The cells were washed ( 5 times in phosphate buffered saline ( PBS , 7 . 2 ) , fixed with 4% paraformaldehyde , permeabilized with 0 . 1% Triton X-100 ( 30 min incubation ) and stained with DAPI . The number of parasites/2000 host cells was counted under the microscope ( 40× ) . The total number of parasites counted in PCSK9 untreated cells was considered as 100% . Fibroblast cells were incubated with parasites ( MOI 5∶1 ) for 1h , washed to remove unbound parasites , fed with fresh medium and incubated at 37°C . At 4 , 15 and 24h post infection , the cells were fixed with 4% paraformaldehye , blocked in 3%BSA , incubated with anti-parasite mouse serum ( serum of infected CD1 mice 1∶20 dilution ) and secondary antibody fluorescent Alexa 480 ( green ) to stain parasites bound to the cell surface . Then the cells were permeabilized with 0 . 2% Triton X-100 , blocked in 3%BSA , incubated with anti- parasite mouse serum and secondary antibody fluorescence Alexa 594 ( red ) to stain intracellular parasites . Trypomastigotes ( 1 . 8×106 ) were washed twice in PBS and incubated with 5 µg of recombinant hu-LDLr ( 2148-LD/CF , R&D Systems , Inc . ) for 1h at room temperature ( final concentration 10ng/ul ) . The incubation mixture was centrifuged at 5000 rpm for 5min , washed twice in DMEM ( Dulbecco's Modified Eagle Medium ) and surfaced on lysine treated cover slips for 20 min before fixing with 4% paraformaldehyde ( 30min ) . IFA was performed for bound parasites using LDLr specific monoclonal antibody ( 1∶100 dilution , for 1h at 37°C ) and IgG goat Alexa fluor 594 ( 1∶500 dilution for 1h at 37°C ) . As an alternative method we used recombinant hu-LDLr dye conjugate ( Alexa fluor 488 , prepared as per Invitrogen protocol ) to incubate with trypomastigotes for 1hr at room temperature . The parasites were then centrifuged at 5000 rpm for 5 min , washed with DMEM ( 2 times ) and surfaced on lysine treated cover slips for 20 min before fixing with 4% paraformaldehyde . GAPDH dye conjugate was prepared as above and used as control . Total RNA was extracted from the heart tissue of CD1 infected mice using Trizol reagent ( Invitrogen ) . Further cleaning up of RNA was performed using RNeasy minikit ( QIAGEN Sciences , Maryland ) according to the manufacturer's instructions . Reverse transcription of total RNA and the quantitative PCR was carried out as described earlier using iQ5 BioRad system [16] . The LDLr mRNA levels were detected using PCR arrays designed by SABiosciences ( PAMM-030 ) following manufacturer's instructions . Wild type and LDLr KO cells ( mouse embryonic fibroblast ) [31] were incubated with trypomastigotes ( 3 . 1×106/cm2 surface area ) for 1h at 37°C . The cells were washed four times in PBS to remove unbound parasites and incubated in DMEM containing 10% FBS for 68 hrs at 37C . Parasite load in these cells was quantitated by real-time PCR as previously described [29] . Serum collected at day 15 , 20 and 30 p . i . from CD1 infected and uninfected mice were used to quantitate serum LDL , HDL and triglyceride levels using E2HL-100 ( EnzyChrom AF HDL and LDL/VLDL Assay Kit ) following manufacturer's protocol .
Endocytosis of LDL receptor ( LDLr ) in association with calcium mobilization , its subsequent trafficking to lysosomes , and the release of ligands at low pH are processes reminiscent of those involved in T . cruzi invasion . Interestingly , some rhinoviruses use LDLr members to enter into host cells [11] . We hypothesized that T . cruzi may utilize host LDLr to enter the host cells . The association of LDLr with T . cruzi invasion and infection was therefore investigated using in vitro infection of human fibroblast cells ( HFF ) and murine cardiomyocytes ( H9c2 ) . Immunoblot analysis of LDLr protein levels in cell lysates from infected HFF and H9c2 cells using LDLr specific monoclonal antibodies demonstrated that HFF and H9c2 cells incubated with T . cruzi had a two-fold increase in LDLr protein levels within 1h post-infection ( p . i . ) ( Figure 1A and B ) . The monoclonal LDLr antibodies used in these studies were raised against the synthetic peptide corresponding to residues from the C-terminus of the human LDLr . To determine the distribution of LDLr proteins in uninfected and infected cells immunofluorescence analysis ( IFA ) was performed using LDLr specific monoclonal antibodies ( Figure 1C ) . IFA of uninfected fibroblasts demonstrated an even distribution of LDLr around the cell membrane . In contrast , there was a clustering of LDLr in infected cells ( 10 min p . i . ) at the cell membrane suggesting a role for LDLr in infection . These antibodies did not cross react with T . cruzi alone ( Figure S1 ) . To investigate the role of LDLr in parasite trafficking , we pre-incubated HFF cells with exogenous recombinant proprotein convertase subtilisin/kexin type 9 ( PCSK9 ) , an enzyme which binds to the extracellular domain of LDLr and induces LDLr degradation [32] . The PCSK9 pretreated cells demonstrated a 42% reduction in T . cruzi invasion compared to untreated cells ( Figure 2A ) . Immunoblot analysis of PCSK9 pretreated cells demonstrated no significant difference in LDLr levels between uninfected and infected HFF cells ( Figure S2 ) . To determine whether parasites can invade the cells in the absence of LDLr , we infected embryonic fibroblast cells derived from LDLr KO mice [31] . Immunoblot analysis confirmed the absence of full length LDLr ( 120 kDa ) protein in KO cells compared to wild type cells ( Figure 2B ) . It should be noted , however , that these LDLr-KO cells express a truncated LDLr as previously demonstrated by Ishibashi [31] . This truncated version ( based on the original gene KO ) lacks the domains for binding LDL and for internalization . When cells were incubated with T . cruzi for 1h and assayed at 68 hrs there was a 62% reduction in total parasite load in LDLr-KO cells compared to wild type using a qPCR method ( Figure 2C ) . Parasites bound strongly to the cell surface in these LDLr KO cells and could not be dislodged even with extensive washings with PBS . A kinetic study of parasite invasion using a double staining IFA method demonstrated that parasites rapidly invaded wild type cells , but were retarded ( 70% ) from invading the LDLr-KO cells ( Figure 2D ) . Binding versus internalization assays performed after 15 hrs of incubation with parasites using double staining IFA showed significantly less internalized parasites in KO cells compared to wild type ( Figure S3 ) . We observed a reduced LDLr protein ( truncated ) expression in KO cells compared to wild type full length LDLr ( data not shown ) . IFA demonstrated that clustering of the disrupted LDLr still occurred at the cell membrane and that the disrupted LDLr remained associated with internalized parasites in infected KO cells similar to that of wild type ( Figure 2E ) . Parasite binding was noted to occur in the vicinity of this truncated LDLr in these KO cells ( Figure 2F ) . We do not know whether this association of truncated protein with the parasite is involved in parasite internalization ( 30% compared to wild type ) or if other members of LDLr family are involved in the absence of full length LDLr . Overall , these results supported the hypothesis that LDLr plays an important role in mediating T . cruzi invasion . The activation of various cell surface receptors and signaling pathways by T . cruzi has been reported by other investigators [22]–[25] . To examine if trypomastigotes utilize LDLr to activate certain signaling pathways to enter the host cells or if LDLr-assisted pathways are involved in internalization , we performed IFA to observe the localization of LDLr in infected cells . IFA of HFF cells incubated with trypomastigotes demonstrated an association of LDLr with internalized parasites within 10 min of incubation ( Figure 3A ) . It was previously reported that T . cruzi has an affinity for binding to LDL [33] . We performed IFA of thoroughly washed parasites using LDL specific antibodies and observed no signals for the presence of LDL ( data not shown ) . To ascertain whether trypomastigotes can directly bind to LDLr or use LDL as a bridge to bind to LDLr , we performed binding studies using recombinant human LDLr ( huLDLR Ala22-Arg788 ) . We incubated washed parasites with recombinant LDLr and carried out IFA as described in experimental procedures . IFA using monoclonal LDLr antibodies demonstrated the association of recombinant LDLr with the parasites ( Figure 3B ) . An alternative experiment with fluorescent labeled LDLr also confirmed direct binding between parasites and the LDLr ( Figure 3C ) . Fluorescent labeled GAPDH ( control protein ) did not bind to parasites and the monoclonal LDLr antibody did not bind to the parasite in the absence of preincubation of parasites with LDLr . The endocytosis of LDLr and its associated adaptor proteins , including clathrin and PIPs , has been extensively studied . Accumulation of PIP2 and PIP3 at the penetration site of trypomastigotes has been reported [34] . To further confirm the association of PIPs to LDLr during parasite invasion we performed double staining IFA with antibodies specific for LDLr and PIP2/3 . IFA demonstrated the co-localization of LDLr and PIPs with invaded parasites in infected HFF cells ( Figure 4A ) . It had been previously reported that inhibition of dynamin ( a protein involved in clathrin-mediated endocytosis ) , drastically diminished T . cruzi entry in both phagocytic and non-phagocytic cells [35] . To investigate the involvement of clathrin in T . cruzi invasion , we performed IFA of 30 min infected HFF cells using clathrin specific monoclonal antibodies . The results as demonstrated in Figure 4B corroborate the involvement of clathrin in the invasion of extracellular parasites into host cells . These antibodies did not cross react with T . cruzi alone ( Figure S1 ) . The functional properties of LDLr family members include clustering of receptors into clathrin-coated pits mediated by adaptor proteins , a pH sensitive ligand uncoupling mechanism , and recycling of the receptors back to the cell surface after dissociation of ligand in endosomes [1] . We wanted to ascertain whether parasites dissociate from LDLr as soon as they enter host cells or co-exist while fused with endosomes/lysosomes in order to gain entrance to the acidic environment important for transformation from trypomastigotes to amastigotes . Therefore , we employed double staining IFA using lysosome associated membrane proteins ( LAMP-1 and 2 ) specific antibodies ( Figure 4C and 4D ) . Fibroblast cells infected with trypomastigotes for 30 min showed the presence of lysosomes surrounding the invaded parasites in association with LDLr . These antibodies did not cross react with T . cruzi alone ( Figure S1 ) . To investigate the role of LDLr in a T . cruzi infected mouse model , we infected CD1 mice with trypomastigotes and analyzed for LDLr mRNA levels in heart tissue 15 day p . i . qPCR demonstrated a significant increase in the mRNA levels of LDLr up to 8000 fold in infected heart tissue compared to control mice ( Figure 5A ) . A serum lipid analysis was performed to examine any changes in LDL , HDL and triglyceride levels in infected mice compared to uninfected mice ( Figure 5B ) . LDL levels in infected mice significantly decreased with time ( 22% by 15 day p . i . and 50 . 5% by 30 day p . i . ) compared to control mice . IFA of paraffin embedded heart tissue of infected mice ( 15 day p . i . ) using LDLr specific monoclonal antibodies ( Figure 5C ) demonstrated the co-localization of LDLr to the specific area surrounding invaded parasites in heart tissue . We also performed IFA of LDL using LDL specific polyclonal antibodies in these tissues and the results confirmed an accumulation of LDL along with LDLr around intracellular parasites in the hearts of infected mice ( Figure 5D ) . These results demonstrate that in addition to the increase in LDLr mRNA levels in infected tissues there was also a localization of LDLr as well as LDL to areas of pseudocysts containing thousands of parasites .
LDLr family members share similar structural homology and are involved in lipoprotein and other ligand endocytosis events . Internalization of ligands by LDLr is a complex process which requires a vast assembly of structural coat components and a host of accessory proteins to drive the endocytic machinery . Many signaling pathways and secondary messengers are involved in this process . The mechanisms involved in LDLr endocytosis are similar to that of T . cruzi internalization such as , calcium mobilization , fusion with endosomes/lysosomes , and the requirement of an acidic pH environment . We therefore explored the possible involvement of LDLr in T . cruzi invasion . The in vitro and in vivo observations in the current manuscript confirm that T . cruzi utilizes the LDLr in their host cell invasion process . The interaction between T . cruzi and host cells has been extensively reviewed [36]–[38] . Earlier reports demonstrate that a variety of host receptors become activated during T . cruzi binding and invasion . For example , activation of TLR2 mediated Rab-5 in T . cruzi invasion has been explored . TNF-α , interleukins and cytokines are regulated by TLR-2 activation . Two other receptors namely , “transforming growth factor β receptor and bradykinin receptor” have also been reported to be involved in T . cruzi infection [23] , [25] . Melo-jorge and Pereira Perrin demonstrated the involvement of receptor tyrosine kinases during T . cruzi invasion [24] . Based on these reports , we propose that T . cruzi binds to host cell membrane receptors and activates many signaling pathways mainly involved in cell proliferation , PI3kinase activation , MAPK signaling and transcription factors , since all these components are known to positively regulate LDLr transcription [7]–[10] . Our results demonstrate that the parasites directly bind to LDLr and inhibition or disruption of LDLr resulted in a reduced rate of invasion . This mechanism of invasion is associated with PIPs and clathrin . It had been previously reported that inhibition of dynamin ( a protein associated in clathrin-mediated endocytosis ) , drastically diminished T . cruzi entry in both phagocytic and non-phagocytic cells [35] . Earlier reports have demonstrated the accumulation of PIP2/3 around the parasite penetration site and parasitophorous vacuoles and that inhibition of PI3 kinase resulted in decreased parasite entry [21] , [34] , [39] . Here , we report that phosphotidylinositol bis phosphate ( PIP2 ) co-localized to the LDLr/parasite complex ( Figure 4A ) . These data are consistent with the earlier observation of a lysosome independent pathway for parasite invasion [17]–[20] . It is probable that T . cruzi uses the sorting motif in the cytoplasmic tail of LDLr to recruit the host lysosomes to the site of invasion , which provides the acidic environment to the parasite for its transformation to amastigotes . IFA demonstrated the co-localization of lysosomes around the parasite associated LDLr complex ( Figure 4C ) . While these results are consistent with a role for LDLr in T . cruzi internalization and trafficking to host lysosomes , the exact mechanism through which LDLr recruits the lysosomes to the site of invasion will require further studies . Overall , these observations indicate that both of the current models that exist for T . cruzi invasion ( i . e lysosome-dependent and PIPs dependent ) are part of the same model in which the LDLr complex machinery connects and completes the process of invasion by this pathogenic microbe . Our studies employing wild type and LDLr KO cells suggest that the presence of full-length LDLr facilitates the binding and internalization of parasites . Disruption of LDL binding domains retarded both parasite binding and invasion . IFA revealed that parasites could still associate with the truncated LDLr expressed in LDLr-KO cells . The KO lacks the LDL binding domain but contains other functional regions of the LDLr including the C terminus . The monoclonal LDLr antibodies employed were raised against the synthetic peptide corresponding to residues from the C-terminus of the human LDLr . The NPXY motif at the C-terminal sequence of LDLr is involved in the internalization signaling [5] and persists in the KO construct . Further investigations will be necessary to determine if other members of the LDLr family are also involved in parasite invasion in the absence of full length LDLr or if the truncated LDLr itself was involved in the reduced rate ( 30% of wild type ) of parasite internalization seen in the LDLr KO cells . Acutely infected mice displayed a significant decrease in plasma LDL levels . In addition , LDL was increased at areas where parasites were present in the heart . The infection-associated increase in phospholipids , triglycerides , and fatty acids could contribute to the pathogenesis of chagasic heart disease [40] . Our data strongly suggest that LDLr and its family members play an important role in T . cruzi invasion and the subsequent lysosomal recruitment that facilitates transformation of trypomastigotes into amastigotes . LDL may facilitate parasite entry and also contribute to LDL-parasite immune complexes regulating LDLr levels [41] . Further research on the mechanism by which this parasite interacts with the host LDLr/clathrin complex is justified . In addition , the observations in this report suggest that therapeutic strategies based on the interaction of T . cruzi and the LDLr pathway should be pursued as possible targets to modulate the consequences of infection . | Trypanosoma cruzi , an intracellular protozoan parasite that causes Chagas disease in humans and results in the development of cardiomyopathy , is a major health problem in endemic areas . This parasite can invade a wide variety of mammalian cells . The mechanisms by which these parasites invade their host cells are not completely understood . Our study highlights , for the first time , that the Low Density Lipoprotein receptor ( LDLr ) is important in the invasion and the subsequent fusion of the parasitophorous vacuole with host lysosomes . We demonstrate that T . cruzi directly binds to LDLr , and inhibition or disruption of LDLr significantly decreases parasite entry . Additionally , we have determined that this cross-linking triggers the accumulation of LDLr and phosphotidylinositol phosphates in coated pits , which initiates a signaling cascade that results in the recruitment of lysosomes , possibly via the sorting motif in the cytoplasmic tail of LDLr , to the site of adhesion/invasion . Studies of infected CD1 mice demonstrate that LDLs accumulate in infected heart and that LDLr co-localize with internalized parasites . Overall , this study demonstrates that LDLr and its family members , engaged mainly in lipoprotein transportation , are also involved in T . cruzi entry into host cells and this interaction likely contributes to the progression of chronic cardiomyopathy . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"infectious",
"diseases/neglected",
"tropical",
"diseases",
"biochemistry/cell",
"signaling",
"and",
"trafficking",
"structures",
"infectious",
"diseases/protozoal",
"infections",
"immunology/immunity",
"to",
"infections"
] | 2011 | Trypanosoma cruzi Utilizes the Host Low Density Lipoprotein Receptor in Invasion |
Sterol biosynthesis is a crucial pathway in eukaryotes leading to the production of cholesterol in animals and various C24-alkyl sterols ( ergostane-based sterols ) in fungi , plants , and trypanosomatid protozoa . Sterols are important membrane components and precursors for the synthesis of powerful bioactive molecules , including steroid hormones in mammals . Their functions in pathogenic protozoa are not well characterized , which limits the development of sterol synthesis inhibitors as drugs . Here we investigated the role of sterol C14α-demethylase ( C14DM ) in Leishmania parasites . C14DM is a cytochrome P450 enzyme and the primary target of azole drugs . In Leishmania , genetic or chemical inactivation of C14DM led to a complete loss of ergostane-based sterols and accumulation of 14-methylated sterols . Despite the drastic change in lipid composition , C14DM-null mutants ( c14dm− ) were surprisingly viable and replicative in culture . They did exhibit remarkable defects including increased membrane fluidity , failure to maintain detergent resistant membrane fraction , and hypersensitivity to heat stress . These c14dm− mutants showed severely reduced virulence in mice but were highly resistant to itraconazole and amphotericin B , two drugs targeting sterol synthesis . Our findings suggest that the accumulation of toxic sterol intermediates in c14dm− causes strong membrane perturbation and significant vulnerability to stress . The new knowledge may help improve the efficacy of current drugs against pathogenic protozoa by exploiting the fitness loss associated with drug resistance .
Leishmaniasis is a group of parasitic diseases infecting 10–12 million people in 88 countries [1] . It is caused by protozoan parasites of the genus Leishmania and transmitted through the bite of sandflies . During their life cycle , Leishmania parasites alternate between motile promastigotes which live in the midgut of sandflies and non-motile amastigotes which reside in the phagolysosome of mammalian macrophages . Depending on parasite species and host genetic factors , symptoms of leishmaniasis include localized skin sores , diffuse cutaneous lesions , severe mucosa destruction , and deadly visceral infections ( kala azar ) which damage the spleen , liver , and bone marrow [2] . Current treatments are often toxic , difficult to administer , and not cost-effective [3] . With drug resistance on the rise and no safe vaccine available , it is necessary to maintain a steady stream of new inhibitors and new biochemical targets to control these dangerous pathogens [4] . In eukaryotes , sterol biosynthesis is a vital pathway and an important source of antimicrobial targets . It consists of three stages: 1 ) the synthesis of isopentenyl pyrophosphate from acetyl CoA or an alternative carbon source such as leucine in trypanosomatids [5]; 2 ) the condensation of isopentenyl pyrophosphate and dimethylallyl pyrophosphate to form squalene; and 3 ) the cyclization of squalene into lanosterol , which is then converted into final products such as cholesterol , ergosterol , and phytosterol ( Fig . S1 ) [6] , [7] . Along with sphingolipids , sterols are tightly packed into ordered membrane microdomains or lipid rafts , which can be isolated as detergent resistant membrane fractions ( DRMs ) serving as scaffolds to support membrane integrity and signal transduction [8] , [9] . In Saccharomyces cerevisiae , ergosterol synthesis is implicated in cell growth , ethanol resistance [10] , heat shock response [11] , and gene expression [12] . In mammals , cholesterol is a vital constituent of cell membrane and a key component of lipoprotein particles . It is also the precursor for the synthesis for various steroid hormones [13] . Precise functions of sterol synthesis in protozoa , however , are not well-characterized . Similar to fungi , trypanosomatid pathogens including Trypanosoma brucei , Trypanosoma cruzi , and various Leishmania species synthesize C24-alkylated , ergostane-based sterols [14] ( Fig . S1 ) . Although the early steps of sterol synthesis ( prior to zymosterol ) are conserved in most eukaryotes , structural differences between mammalian enzymes and microbial enzymes can be exploited to produce selective drugs . Enzymes involved in the late steps of sterol pathway could also be valuable targets because mammalian cells do not synthesize ergostane-based sterols . Indeed , multiple classes of compounds targeting sterol biosynthesis exhibit good anti-trypanosomatid activities in vitro although their efficacies in vivo are often unsatisfactory . Examples include 3- ( biphenyl-4-yl ) -3-hydroxyquinuclidine which blocks the activity of squalene synthase ( E . C . 2 . 5 . 1 . 21 ) [15] , terbinafine which inhibits squalene epoxidase ( EC 1 . 14 . 99 . 7 ) [16] , [17] , various azole drugs which target sterol 14-alpha-demethylase ( C14DM , EC 1 . 14 . 13 . 70 ) [18]–[20] , and azasterol which interferes the C24-alkylation of sterol precursor [21] , [22] . Amphotericin B ( Amp B ) is another antifungal which binds to ergosterol or other ergostane-based sterols leading to pore formation on the plasma membrane [23] , [24] . It possesses potent anti-Leishmania activity and is widely used as the drug of choice to treat antimony-resistant parasites [25] . Despite the promise , the underlying mechanism of how the alteration in sterol composition leads to growth retardation and/or parasite death is not well understood , which hinders the development of new and improved treatments [26]–[28] . The primary target of azole drugs is C14DM ( known as CYP51 in animals and ERG11 in yeast ) , an evolutionarily conserved , heme-dependent , cytochrome P450 enzyme present in fungi , plants , mammals , and trypanosomatids [29] ( Fig . S1 ) . The reaction catalyzed by C14DM consists of three steps: the initial oxygenation of 14α-methyl group ( –CH3 ) to 14α-alcohol ( –CH2OH ) , further oxidation to 14α-aldehyde ( –CHO ) , and finally the elimination of formic acid leading to the formation of C14-15 double bond in the sterol core [30] . Mouse C14DM is essential for embryogenesis , as deletion of this gene leads to embryonic lethality at day 15 [31] . In S . cerevisiae , null mutants of ERG11 require exogenous ergosterol to survive and only grow in the absence of oxygen or in the presence of a suppressor mutation in sterol C5-desaturase ( ERG3 , an enzyme upstream of C14DM ) [12] , [32]–[34] . While C14DM appears to be indispensable in mammals and fungi , azole drugs exhibit higher affinity for fungal enzymes over mammalian orthologs which contributes to their selectivity [35] . The C14DMs from several trypanosomatids have been cloned and biochemically characterized [36]–[38] . Significant efforts have been devoted to identify new and better C14DM inhibitors as anti-T . cruzi agents [18] , [39]–[41] . Biochemical and structural studies of the C14DM from Leishmania infantum indicate that this enzyme prefers C4-monomethylated sterol substrates ( such as 4 , 14-dimethyl zymosterol ) , although it also metabolizes C4-dimethylated sterols ( e . g . lanosterol ) and C4-desmethylated sterols ( e . g . 14α-methylzymosterol ) with lower efficiency [38] ( Fig . S1 ) . This type of substrate preference is similar to the C14DMs in plants and T . brucei [37] , [42] but distinct from that in T . cruzi which favors C4-dimethylated sterols [38] , [43] . Meanwhile , the C14DMs in mammals and fungi provide rapid demethylation of sterol substrates without obvious restriction ( regarding C4-methylation ) [38] . The goal of our study is to address the following important yet still unanswered questions about sterol metabolism in Leishmania: Is C14DM essential for the promastigote stage ( found in sandflies ) and amastigote stage ( found in mammals ) ? What is the role of sterol synthesis in the organization of plasma membrane ? Is C14DM the primary target of azoles ? How to improve the efficacy of current sterol synthesis inhibitors ? To answer these questions , we generated and characterized a C14DM-null mutant in Leishmania major . Our results suggest that inactivation of C14DM severely disrupts the membrane stability of Leishmania parasites , probably due to the accumulation toxic sterol intermediates . Although this is not lethal by itself , it leads to extreme vulnerability to heat stress . The new knowledge will not only provide novel insight into the physiological role of sterol synthesis in Leishmania parasites , but also can guide the development of new treatments or to improve the efficacy of current antileishmanial drugs .
L . major C14DM was identified from the TriTrypDB ( gene ID: LmjF . 11 . 1100 ) showing 28–33% identity to the C14DMs from human , fungi , and Mycobacterium tuberculosis ( Fig . S2 ) . Its syntenic orthologs are present in the genomes of L . braziliensis , L . infantum , L . mexicana , T . brucei , and T . cruzi . The L . major C14DM protein ( 479 aa ) contains a potential N-terminal signal peptide ( M1-F24 ) and motifs predicted to mediate sterol substrate binding ( Y102–V113 ) and heme binding ( G415–G424 ) [44] ( Fig . S2 ) . Using the targeted gene replacement approach [45] , we were able to generate null mutants of C14DM ( c14dm− ) in L . major promastigotes . Southern-blot confirmed the loss of endogenous C14DM alleles in three independent c14dm− clones ( Fig . 1A–C ) . Add-back parasites ( c14dm−/+C14DM ) were generated by introducing pXG-C14DM ( a high copy number plasmid containing C14DM ) into the mutants ( Fig . 1B–C ) . To determine the localization of C14DM , a C14DM-GFP fusion protein was constructed and expressed in c14dm− promastigotes . The integrity and functionality of C14DM-GFP were confirmed by western blot and later by lipid analysis ( Fig . S3 ) . Immunofluorescence microscopy revealed a significant overlap between GFP fluorescence and the anti-BiP staining [46] , but less so with the mitochondrial marker Mitotracker ( Fig . 1D–H ) . These data suggest that C14DM is mainly located in the endoplasmic reticulum ( ER ) although a fraction of it may also reside in the mitochondrion . This localization is consistent with its predicted role in sterol biosynthesis and previous reports on C14DMs from S . cerevisiae and rat liver [47] , [48] . In S . cerevisiae , deletion of C14DM ( Δerg11 ) led to ergosterol auxotrophy and cell death under aerobic conditions , possibly due to the production of oxygenated sterol intermediates [32]–[34] . Surprisingly , L . major c14dm− promastigotes were fully viable in culture during the replicative log phase although their doubling time ( ∼12 hours ) was longer than that of wild type ( WT ) parasites ( ∼7 hours ) ( Fig . 2A–2B ) . Despite the slower growth rate , these mutants reached similar densities as WT parasites ( 2 . 3–3 . 0×107 cells/ml ) in stationary phase ( Fig . 2A ) . In late stationary phase , c14dm− mutants had slightly more dead cells and produced less metacyclics ( the non-replicative but highly infective forms [49] ) than WT promastigotes ( Fig . 2B–C ) . In addition , more round cells were detected in c14dm− mutants ( 20–30% ) than in WT parasites ( 5–10% ) in both log and stationary phase ( Fig . 2D ) . DNA staining revealed that 15–32% of c14dm− promastigotes had two kinetoplasts ( containing mitochondrial DNA ) and two nuclei ( 2K2N ) , whereas only 3–8% of WT parasites were 2K2N ( Fig . 3 ) . Similar results were observed in a cell cycle analysis of permeabilized promastigotes labeled with propidium iodide ( Fig . S4 ) . These data suggest that sterol synthesis is involved in maintaining normal cytokinesis in Leishmania . Defects manifested by c14dm− ( growth delay , altered morphology , reduced metacyclogenesis , and overabundance of 2K2N cells ) were completely reversed when C14DM expression was restored ( c14dm−/+C14DM in Figs . 2–3 and Fig . S4 ) . Therefore , although C14DM is not required for promastigote survival or proliferation in culture , it is involved in the control of cell shape , differentiation , and division in L . major . The effect of C14DM deletion on sterol synthesis was assessed by gas chromatography-mass spectrometry ( GC-MS ) . Briefly , promastigote lipids were examined by total ion current ( “A” in Fig . S5–Fig . S8 ) , selected ion spectra ( “B–F” in Fig . S5–Fig . S8 ) , and the full mass spectra of major sterol species were acquired by electron ionization ( Fig . S9 ) . In WT parasites , four major sterol species were identified based on their retention time , formula weights , and electron impact mass spectra as the following: 5-dehydroepisterol ( 50–57% ) , ergosterol ( 22–28% ) , cholesta-5 , 7 , 24-trienol ( 6–10% ) , and cholesterol ( 3–5% ) ( Fig . 4A , 4E , Fig . S5 and Table S1 ) . Deletion of C14DM led to a complete loss of ergostane-based sterols ( 5-dehydroepisterol , ergosterol , and episterol ) and cholesta-5 , 7 , 24-trienol , but the level of cholesterol ( salvaged from the medium ) was not significantly affected ( Fig . 4B , 4E , Fig . S6 , and Table S1 ) . Meanwhile , c14dm− mutants possessed a new , highly conspicuous sterol peak with a retention time of 14 . 76–14 . 78 on GC spectrum ( Fig . 4B and Fig . S6 ) . Selected ion analysis revealed that this peak was comprised of two lipid species with formula weights of 398 . 6 and 412 . 6 ( Fig . S6C and 6F ) . Based on the role of C14DM in sterol synthesis , these lipids are predicted to be 14-methyl fecosterol ( FW = 412 . 6 , XII in Fig . S1 ) and 14-methyl zymosterol ( FW = 398 . 6 , XI in Fig . S1 ) . Together these 14-methylated sterols constitute >95% of total sterols in c14dm− ( Fig . 4B , 4E , and Table S1 ) . Very similar results were observed when WT parasites of L . major , L . donovani , L . mexicana , and L . amazonensis were cultured in the presence of ITZ ( 3 . 3–200 nM ) , a C14DM inhibitor , for 2 days ( Fig . 4C , 4E , Fig . S7 , and Table S1 ) . Parasites with episomal C14DM expression ( c14dm−/+C14DM and c14dm−/+C14DM-GFP ) had WT-like sterol composition , not elevated amounts of ergostane-based sterols ( Fig . 4D , 4E , Fig . S8 , Fig . S3B , and Table S1 ) . This may reflect a limitation of substrates and/or a feedback regulation mechanism . It is also worth mentioning that deletion or overexpression of C14DM had no significant impact on the overall abundance of total sterols in Leishmania promastigotes ( Table S1 ) . To test whether C14DM is the primary target of ITZ in Leishmania , promastigotes were inoculated in 0–10 µM of ITZ and culture densities were determined after 48 hours . For L . major WT and c14dm−/+C14DM parasites , a dose-dependent response was observed ( Fig . 5A ) ; the IC25 , IC50 , and IC90 ( concentrations required to inhibit growth by 25% , 50% , or 90% ) were estimated to be 0 . 12 µM , 0 . 40 µM , and 10 µM , respectively ( Fig . 5A and Table 1 ) . Based on our sterol analysis , ITZ could shut down C14DM in WT L . major at fairly low concentrations ( 50 nM–0 . 2 µM ) but only caused mild growth retardation ( Fig . 4C , Table S1 and unpublished data ) . For c14dm− parasites , ITZ had negligible effect on growth at ≤0 . 2 µM , but did cause dose-dependent inhibition at >0 . 2 µM similar to WT parasites ( Fig . 5A ) . For these mutants , the IC25 , IC50 , IC90 were around 0 . 60 µM , 2 . 0 µM , and 10 µM , respectively ( Fig . 5A and Table 1 ) . These data suggest that ITZ's mode of action is two-fold: at low concentrations ( <0 . 2 µM ) , the drug mainly exerts its effect by blocking C14DM; and at high concentrations ( >0 . 2 µM ) , it affects other targets beyond C14DM ( Fig . 5A ) . Amp B is another widely-used antifungal/antiprotozoal compound . It binds membrane sterol leading to the formation of channels and subsequent cell lysis . As shown in Fig . 5B and Table 1 , c14dm− mutants were extremely resistant to Amp B as their IC values were 10–100 times higher than those of WT and c14dm−/+C14DM parasites . These findings support the notion that Amp B targets ergostane-based sterols more efficiently than cholesterol-like sterols , which confers selectivity [50] , [51] . Similar to ITZ , Amp B exhibited a biphasic inhibition on L . major growth: a C14DM-dependent phase at low concentrations ( <0 . 1 µM ) and a C14DM-independent phase at high concentrations ( Fig . 5B ) . Together , these data indicate that: 1 ) loss of ergostane-based sterols ( through genetic or chemical inactivation of C14DM ) is not detrimental to promastigotes in culture; and 2 ) ITZ and Amp B have additional targets in Leishmania beyond the sterol synthesis pathway . Sterols are key stabilizers of biological membranes . Along with sphingolipids , they promote the formation of ordered membrane microdomains or lipid rafts [9] , [52] . The unusual sterol profile in c14dm− prompted us to investigate whether sterol synthesis affects the expression and organization of membrane-bound , GPI-anchored virulence factors such as lipophosphoglycan ( LPG ) and GP63 ( an abundant metalloprotease ) . In both log phase and stationary phase , c14dm− mutants had much less LPG than WT and c14dm−/+C14DM cells based on western-blot ( 10–25% , Fig . 6A–C ) . This was not due to increased shedding/secretion , as the LPG in c14dm− culture supernatant was also low ( Fig . 6A–B ) . Similar results were observed by immunofluorescence microscopy and flow cytometry using an anti-LPG antibody ( Fig . S10 ) . Meanwhile , these mutants contained more GP63 than WT ( ∼two-fold increase ) in log phase but not in stationary phase or metacyclic promastigotes ( similar to WT in these stages; Fig . 6A , B , D and Fig . S11 ) . Therefore , changes in sterol composition do affect the steady state level of GPI-anchored virulence factors . We also assessed the abundance of LPG and GP63 in liquid-ordered membrane microdomains by examining the DRMs . In mammalian cells and trypanosomatids , GPI-anchored macromolecules tend to be segregated in DRMs at 4°C ( but not 37°C ) , which may reflect their association with cholesterol/sphingolipid- rich domains ( lipid rafts ) [9] , [53] . In WT parasites , LPG was enriched in DRM in late stationary phase ( 35–38% of total LPG ) but not in log phase ( only 9–14% of total LPG ) ( Fig . 7A and G ) , indicative of a plasma membrane remodeling process during promastigote development as previously proposed [54] . Differing from LPG , GP63 had a clear association with DRM ( 50–60% of total GP63 ) in both log phase and stationary phase ( Fig . 7C and H ) , suggesting that it is a constitutive component of lipid rafts . As a control , the cytosolic protein HSP83 was not found in DRM ( Fig . 7E ) [55] . Importantly , loss of C14DM reduced the DRM-association of GP63 in log phase and stationary phase ( from 50–60% in WT to 20–32% in c14dm− , Fig . 7C , D , and H ) ; c14dm− mutants also had less LPG in DRM than WT during stationary phase ( 12–15% in c14dm− versus 35–38% in WT; Fig . 7A , B , and G ) ; and restoration of C14DM expression reversed these defects ( Fig . 7G and H ) . Collectively , these data indicate that ergostane-based sterols are critical not only for the synthesis and/or trafficking of GPI-anchored virulence factors , but also for their association with liquid-ordered microdomains . To determine whether ergosterol synthesis is required for Leishmania survival in mammals , metacyclics were isolated from stationary phase promastigotes and injected into the footpads of BALB/c mice . As indicated in Fig . 8A , WT and c14dm−/+C14DM parasites caused rapid progression of lesions and all the mice had to be euthanized within 100 days post infection due to severe pathology . In contrast , mice infected by c14dm− did not show any disease for the first 120 days and it took them 180–200 days to develop large lesions ( ∼2 . 0 mm ) . The parasite loads in c14dm−-infected mice were also significantly lower than those infected by WT or c14dm−/+C14DM parasites at the same time ( Fig . 8B ) . Similar results were obtained when BALB/c mice were infected with lesion-derived amastigotes of WT , c14dm− , c14dm−/+C14DM parasites ( Fig . 8C–D ) . While LPG is an important virulence factor for L . major promastigotes , it is not required for the infectivity of amastigotes [56] . Thus , the reduced virulence of c14dm− cannot be solely attributed to LPG deficiency . Besides mouse infection , we also examined the ability of c14dm− mutants to parasitize primary murine macrophages in vitro . Comparing to WT and c14dm−/+C14DM parasites , c14dm− mutants survived poorly in BALB/c macrophages ( Fig . S12 ) . Together , these findings demonstrate that C14DM is extremely important for Leishmania to effectively survive , proliferate , and cause disease in the mammalian host . The fact that c14dm− mutants could still cause disease ( at a reduced capacity nonetheless ) suggests sterol synthesis is not absolutely essential for Leishmania during the mammalian stage . To investigate the effect of C14DM-deletion on the sterol composition of amastigotes , we isolated WT and c14dm− amastigotes from footpad lesions ( Fig . S13 ) and examined their lipid contents by GC-MS ( Fig . S14–15 ) . For comparison , we also extracted lipids from uninfected mouse footpad tissue ( Fig . S16 ) and promastigotes ( Fig . S17–18; the RT values here were different from the ones in Figs . 4 and S5–S8 because a different GC column was used ) . Cholesta-3 , 5-diene was added as an internal standard to both amastigote and promastigote samples ( 1 . 0×109 molecules/amastigote and 2 . 0×107 molecules/promastigote; RT = 11 . 00 in Fig . S14–18 ) . In WT amastigotes , a very high level of cholesterol was evident ( RT = 12 . 87 in Fig . S14A–B ) , whereas the ergostane-based sterols ( ergosterol , 5-dehydroepisterol , and episterol ) were almost undetectable ( Fig . S14C–D; similar to uninfected mouse tissue in Fig . S16 ) . Clearly , some of the cholesterol was not directly associated with amastigotes ( instead from mouse cells; Fig . S13 and Fig . S16 ) . Nonetheless , the lack of endogenous sterols suggests that de novo sterol synthesis is significantly downregulated in amastigotes . This was in sharp contrast to WT promastigotes which had much more ergostane-based sterols than cholesterol ( Fig . S17 and Fig . S5 ) . C14dm− amastigotes also contained an overwhelming amount of cholesterol ( Fig . S15A–B ) , much more abundant than the 14-methyl sterols ( Fig . S15D–E ) which were dominant in promastigotes ( RT = 13 . 69 in Fig . S18 ) . Since Leishmania parasites do not synthesize cholesterol [14] , [57] , these results suggest that amastigotes acquire the majority of their sterols from the host rather than de novo synthesis . Next we investigated whether sterol synthesis was involved in resistance to heat , acidic pH , and reactive oxygen intermediates/reactive nitrogen intermediates ( ROIs/RNIs ) . In order to establish infection in mammals , Leishmania parasites must overcome these stress conditions . To examine if C14DM is required for heat tolerance , stationary phase parasites were incubated at either 27°C ( the regular promastigote culture temperature ) or 37°C ( mimicking the mammalian body temperature ) . Most parasites were alive at 27°C as expected ( Fig . 9A ) . At 37°C , however , 73–90% of c14dm− promastigotes were dead in 12 hours whereas the vast majority of WT and c14dm−/+C14DM cells remained viable ( Fig . 9B ) . Similar to c14dm− , WT parasites grown in the presence of ITZ from log phase to stationary phase were hypersensitive to 37°C condition ( Fig . 9C ) . In contrast , if WT parasites were cultured without ITZ to stationary phase and then treated with ITZ ( which would not significantly affect sterol synthesis since most stationary phase cells were non-replicative ) , they did not show such defects ( Fig . 9D ) . Therefore , it is the alteration of sterol composition ( rather than other effects from ITZ ) that is responsible for this hypersensitivity to high temperature . To determine if the function of C14DM on heat resistance is conserved in other Leishmania species , we grew L . mexicana , L . amazonensis and L . donovani parasites in sub-lethal concentrations of ITZ ( 3 . 3 nM for L . mexicana , 25 nM for L . amazonensis , and 81 nM for L . donovani ) which were sufficient to shut down ergostane-based sterol synthesis but only inhibit growth by ∼25% ( Fig . S19A , Table S1 and S2 ) . Similar to c14dm− , these ITZ-treated parasites were extremely vulnerable to heat ( Fig . S19B ) . Therefore , C14DM likely plays similar roles in multiple Leishmania species . We also tested the ability of c14dm− mutants to withstand oxidative , nitrosative and acidic pH stress as previously described [58] . As shown in Fig . S20A–B , these mutants were slightly more sensitive to SNAP ( a nitric oxide releaser ) than WT and c14dm−/+C14DM parasites ( although the difference was not statistically significant ) , which might be due to their low LPG abundance ( Fig . 6 ) [56] . Meanwhile , their resistance to H2O2 and acidic pH were normal ( Fig . S20C–F ) . Since sterols could function as stabilizers in lipid bilayer [59] , we examined whether alteration in sterol composition affects the cell membrane fluidity of c14dm− mutants , which may be linked to their heat sensitivity and defects in forming DRM/rafts . To do so , promastigotes were labeled with TMA-DPH ( a cationic lipophilic probe that diffuses into the outer leaflet of lipid bilayer ) for 20 min at 4°C , 25°C , 37°C , or 45°C ( >95% of cells were alive by propidium iodide staining ) . Plasma membrane fluidity was then determined by measuring the fluorescence depolarization of TMA-DPH as previously described [60] . As indicated in Fig . 9E–F , WT and c14dm−/+C14DM parasites maintained their membrane fluidity at a reasonably stable level when the temperature rose from 4°C to 45°C ( a high anisotropy value means the membrane is more rigid or less fluid ) . In contrast , the plasma membrane of c14dm− mutants became much more fluid at elevated temperatures ( Fig . 9E–F ) . Therefore , defects in sterol synthesis may compromise cell membrane stability and rigidity at high temperatures , resulting in hypersensitivity to heat .
In this study , we investigated the role of C14DM in L . major , a vector-borne protozoan parasite responsible for cutaneous leishmaniasis . C14DM catalyzes the heme-dependent oxidative removal of 14α-methyl group from sterol intermediates , a key step in sterol biosynthesis . Deletion of C14DM in L . major results in a complete loss of ergostane-based sterols and significant accumulation of 14-methylated sterol intermediates . This drastic change of sterol composition leads to increased plasma membrane fluidity , failure to form normal DRM/lipid rafts , and extreme vulnerability to heat . Nonetheless , c14dm− mutants are fully viable and replicative as promastigotes in culture with only minor imperfections in growth rate , morphology and cytokinesis . They do exhibit marked defects in the synthesis and/or trafficking of GPI-anchored virulence factors and are more resistant to antifungals such as ITZ and Amp B . The infectivity of c14dm− mutants is greatly reduced but not completely abolished , suggesting that inhibition of C14DM by itself is not sufficient to eliminate L . major infection . It is rather surprising that Leishmania promastigotes remain viable and proliferative without C14DM . In the absence of endogenous sterols , c14dm− mutants mainly accumulate 14-methylated intermediates . Similar results were observed when parasites were exposed to sub-lethal concentrations of azoles ( Table S1 ) [57] , [61] . Since the overall level of sterols is similar between WT and c14dm− parasites ( Table S1; only the composition is altered ) , it appears that 14-methylfecosterol and 14-methylzymosterol could partially compensate the loss of ergostane-based sterol . Other membrane lipids such as sphingolipids , glycerophospholipids , and cholesterol ( salvaged from the environment ) may also help stabilize the plasma membrane in Leishmania . However , the aberrant sterol composition in c14dm− does have serious consequences as these mutants fail to maintain proper membrane rigidity at elevated temperatures , which probably contributes to their hypersensitivity to mild heat ( although other mechanisms may also be involved ) . Inactivation of C14DM also seems to interfere with the formation of liquid-ordered microdomains , as the DRMs from c14dm− is depleted of GP63 and LPG which should be enriched in lipid rafts . One possibility is that protrusion of axial 14α-methyl group from the planar 4-ring core structure decreases the interaction between sterols and phospholipid side chains [62] , [63] . Consequently , compared to regular sterols ( which possess a smooth α-side ) , 14α-methylated sterols may be less efficient at promoting the condensation of lipid bilayer , leading to increased membrane fluidity ( especially at elevated temperatures ) in c14dm− [61] , [64] [65] , [66] . Additionally , the loss of ergostane-based sterols ( besides the accumulation of 14α-methylated sterols ) may also contribute to these membrane defects . It has been reported that ergosterol is more effective at promoting the liquid-ordered phase than lanosterol ( which also contains the 14α-methyl group ) [67] , [68] . The altered shape of c14dm− mutants is likely caused by increased membrane permeability due to high fluidity , allowing more water penetration as previously shown in S . cerevisiae treated with fluconazole [69] . C14dm− mutants also have more 2K2N cells which have completed DNA replication but are slow to finish division , consistent with their prolonged doubling time . In mammalian cells , cholesterol starvation induced growth arrest at G2 phase and polyploidy formation [70] , [71] . In S . cerevisiae , sterol depletion led to growth arrest at G1 stage [12] . Addition of cholesterol and ergosterol at hormonal amounts reversed these effects in mammalian cells and yeasts , respectively [12] . This indicates that in addition to its membrane function , sterols also possess a signaling role in fungi and mammals . As promastigotes , c14dm− mutants cannot be rescued by exogenous ergosterol when provided at nM-µM range , suggesting that: 1 ) the accumulation of 14-methylated sterol intermediates ( rather than the lack of ergostane-based sterols ) is primarily responsible for the defects in membrane stability , heat resistance and replication; or 2 ) the uptake of ergosterol by Leishmania is insufficient although cholesterol can be incorporated into the membrane . Loss of C14DM also affects the synthesis and/or trafficking of major GPI-anchored virulence factors , as c14dm− mutants contain less LPG but more GP63 ( only in the log phase ) than WT parasites . Previous studies suggest that the synthesis of LPG and GP63 starts with a common pool of alkyl-acyl-PIs with long alkyl chains ( C24:0/C26:0 ) , followed by differential glycosylation and fatty acid remodeling in separate compartments [72] , [73] . Alteration in sterol composition may compromise the vesicular trafficking or the proper compartmentalization of these pathways , causing abnormality in GPI-molecule synthesis . The hypersensitivity of c14dm− mutants to heat is probably a key contributing factor to their severely reduced virulence in mice . The LPG deficiency could partially explain the virulence defect of c14dm− promastigotes but is unlikely to be a major factor for amastigotes since LPG is not required during the mammalian stage of L . major [74] . Besides heat tolerance , the synthesis of ergostane-based sterols is likely needed for other purposes . In L . amazonensis , ketoconazole ( another azole drug targeting C14DM ) treatment induced the appearance of large multivesicular bodies , increased amounts of lipid droplets and acidocalcisomes ( calcium- and phosphate-rich organelles ) [75] , and alterations in the distribution and appearance of mitochondrial cristae [76] , [77] . L . amazonensis parasites exposed to 22 , 26-azasterol , a sterol methyltransferase inhibitor , also exhibited profound morphological changes including mitochondrial swelling , increased number of acidocalcisomes , and the appearance of large , membranous bodies reminiscent of autophagic vesicles [22] . Comparing to Leishmania promastigotes , intracellular amastigotes show a global decrease in the uptake and utilization of glucose and amino acids , but are more dependent on mitochondrial metabolism ( for TCA cycle and glutamine synthesis ) [78] . Thus , perturbation of mitochondrial structure/function may be an underlying mechanism for the anti-proliferative effect of sterol synthesis inhibitors . Importantly , after a delay of 70–120 days , c14dm− -infected mice started to show symptom ( footpad swelling ) and eventually produced lesions similar to WT-infected mice . Mutant parasites were also capable of proliferation after the initial delay . Promastigotes and amastigotes isolated from c14dm− -infected mice were still attenuated ( Fig . 8 ) , suggesting that this is not due to reversion or compensatory mutations . Hence , despite their profound defects , c14dm− mutants are still somewhat virulent . One possibility is that Leishmania amastigotes salvage huge amounts of host lipids including cholesterol and sphingolipids [79]–[81] , which may alleviate the loss of de novo synthesis and/or accumulation of toxic sterols . This is supported by our amastigote lipid analysis which showed significant accumulation of cholesterol ( host-derived ) and only trace amount of endogenous sterols ( Fig . S13–18 ) . The fact that c14dm− mutants are viable as promastigotes and infective in mice ( at a reduced capacity ) suggests that inhibition of C14DM by itself may not be sufficient to cure Leishmania infection . In vitro , azole drugs such as ketoconazole , fluconazole , itraconazole , and posaconazole have shown activity against the growth of Leishmania and Trypanosoma cruzi ( responsible for Chagas disease ) , yet their in vivo efficacies remain somewhat unsatisfactory [39] , [82]–[85] . These drugs are often limited by poor pharmacokinetics ( difficulties in formulation , delivery and bioavailability ) [86] and emergence of resistance ( e . g . increased drug efflux and mutations in the target gene ) [86] . Findings from our study suggest that the efficacy of azoles may improve if they are used in combination with localized heat treatment . Thus , although C14DM inhibition only exerts modest anti-Leishmania effect , it does make parasites vulnerable to other physical or chemical perturbations . For Leishmania promastigotes , ITZ ( and possibly other azoles ) treatment seems to be mimic the effect of C14DM deletion at low concentrations but it clearly inhibits other unknown targets at high concentrations ( Fig . 5A and Fig . S19A ) . Based on our findings , it may be worthwhile to explore whether inhibitors of sphingolipid/phospholipid synthesis can exacerbate the membrane instability of c14dm− mutants . If so , combined inhibition of multiple lipid synthesis pathways may have synergistic effect on parasite survival . Our findings also indicate that mutations in C14DM can confer significant resistance to Amp B , although the fitness costs associated with such mutations could be therapeutically exploited . In summary , genetic or chemical inactivation of C14DM in Leishmania results in dramatic change in sterol composition , leading to DRM/raft disruption , increased membrane fluidity , and impairment in the synthesis and/or trafficking of GPI-anchored molecules . Ablating C14DM is not detrimental in L . major , perhaps due to the compensatory effect of other lipids , but does render parasites extremely vulnerable to heat . These findings may guide the development of new therapies which would improve the efficacies of current treatments and exploit the fitness cost of drug resistant strains . In addition , future studies will determine the mechanistic basis of c14dm− -associated defects , e . g . whether they are mainly caused by membrane perturbations or dysregulation of intracellular pathways . The viability of c14dm− mutants also provides a valuable platform to study the roles of DRM/rafts in crucial events such as vesicular trafficking and signaling . Finally , the interaction between Leishmania amastigotes and host cells at sterol metabolism , e . g . de novo synthesis vs salvage is another important topic worthy of further studies .
BALB/c ( female , 7–8 weeks old ) mice were purchased from Charles River Laboratories International ( Wilmington , MA ) . All procedures involving mice were approved by the Animal Care and Use Committee at Texas Tech University ( PHS Approved Animal Welfare Assurance No . A3629-01 ) . Mice were housed and cared for in the facility operated by the Animal Care and Resources Center at Texas Tech University adhering to the Guide for the Care and Use of Laboratory Animals ( the 8th Edition , NRC 2011 ) for animal husbandry . Reasonable efforts were made to minimize animal suffering . Anesthesia was applied through intra-peritoneal injection of ketamine hydrochloride ( 100 mg/kg ) /xylazine ( 10 mg/kg ) . Euthanasia was achieved by asphyxiation through controlled flow of pure CO2 . Zymosterol , lanosterol , cholesterol , ergosterol and 5-dehydroergosterol were purchased from Avanti Polar Lipids ( Birmingham , AL ) as standards ( to determine retention times ) in gas chromatography-mass spectrometry ( GC-MS ) studies . Cholesta-3 , 5-diene was purchased from Sigma-Aldrich ( St . Louis , MO ) as an internal standard for quantitation in total ion current chromatograms . Itraconazole ( ITZ ) was purchased from LKT Laboratories , Inc . ( St . Paul , MN ) . Amphotericin B ( Amp B ) and 30% H2O2 were purchased from EMD Chemicals , Inc . ( San Diego , CA ) . 1- ( 4-Trimethylammoniumphenyl ) -6-Phenyl-1 , 3 , 5-Hexatriene p-Toluenesulfonate ( TMA-DPH ) was purchased from Life Technologies Corporation ( Grand Island , NY ) . All other chemicals were purchased from VWR International or Fisher Scientifics unless specified otherwise . The predicted open reading frame ( ORF ) of L . major C14DM ( LmjF . 11 . 1100 ) was amplified by PCR from L . major genomic DNA using primers #170/#171 . The resulting 1 . 44 Kb DNA fragment was digested with BglII and cloned in the pXG vector [87] as pXG-C14DM ( B294 ) . A modified C14DM-ORF was amplified with primers #170/#333 to remove the stop codon and cloned into the pXG-'GFP+ vector [87] to generate pXG-C14DM-GFP ( B321 ) , which was used to generate a C-terminal GFP fusion protein for the localization study . The upstream and downstream flanking sequences ( ∼1 Kb each ) of C14DM ORF were amplified with primers #172/#173 and primer #174/#175 , respectively . These two PCR products were cloned together into pUC18 . Genes conferring resistance to the puromycin ( PAC ) and blasticidin ( BSD ) were inserted between the upstream and downstream flanking sequences to generate pUC-KO-C14DM::PAC ( B292 ) and pUC-KO-C14DM::BSD ( B293 ) . Primers used in this study were summarized in Table S3 . All DNA constructs were confirmed by restriction enzyme digestion and sequencing . L . major LV39 clone 5 ( Rho/SU/59/P ) , L . ( L ) amazonensis ( MHOM/BR/77/LTB0016 ) , L . ( L ) mexicana M379 ( MNYC/BZ/62/M379 ) and L . donovani 1S2D ( MHOM/SD/62/1S ) promastigotes were cultured at 27°C in M199 medium ( pH 7 . 4 ) with 10% fetal bovine serum and additional supplements [88] . In general , log phase promastigotes refer to replicative parasites at densities lower than 1 . 0×107 cells/ml , and stationary phase promastigotes refer to non-replicative parasites at densities higher than 2 . 0×107 cells/ml . The infective metacyclic parasites ( metacyclics ) were isolated from stationary phase promastigotes using the density centrifugation method [89] . To generate C14DM-null mutants ( c14dm− or ▵C14DM::PAC/▵C14DM::BSD ) , the C14DM alleles from wild type L . major parasites ( WT ) were sequentially replaced by PAC and BSD resistance genes using the homologous recombination-based approach as previously described [90] . To confirm the loss of C14DM , genomic DNAs were digested with SacI , resolved on a 0 . 7% agarose gel , transferred to a nitrocellulose membrane , and hybridized with a [32P]-labeled DNA probe recognizing either the C14DM ORF or a ∼500-bp upstream region of C14DM . Blots were then visualized by radiography . The c14dm− mutants were maintained in media containing 10 µg/ml of puromycin and 10 µg/ml of blasticidin . To restore C14DM expression , pXG-C14DM or pXG-C14DM-GFP was introduced into c14dm− by electroporation and stable transfectants were referred to as c14dm−/+C14DM or c14dm−/+C14DM-GFP , respectively . Three independent c14dm− mutant clones were generated and their phenotypes were nearly identical . Therefore , c14dm− #1 and its add-back control were described in this study . To measure promastigote growth , parasites were inoculated in complete M199 medium at 1 . 0×105 cells/ml . Culture density was determined at designated times using a hemacytometer . Percentages of round cells ( defined as those with the long axis shorter than twice the length of the short axis ) and dead cells were determined by microscopy and flow cytometry , respectively , as previously described [91] . To assess thermal tolerance , stationary phase promastigotes were incubated at either 27°C ( the regular temperature ) or 37°C/5%CO2 and cell viability were determined after 0–12 hours [58] . To measure sensitivity to oxidative and nitrosative stress , stationary phase promastigotes were incubated in various concentrations of H2O2 or S-nitroso-N-acetylpenicillamine ( SNAP ) [92]; cell density and viability were determined after 48 hours . To determine sensitivity to acidic pH , stationary phase promastigotes were inoculated in a pH 5 . 0 medium at 2 . 5×107 cells/ml and culture densities were determined after 48 hours [58] . To test drug sensitivity , promastigotes were inoculated in M199 medium at 2 . 0×105 cells/ml in the presence of ITZ ( 0–10 µM ) or Amp B ( 0–10 µM ) . Culture densities were determined after 48 hours . To collect whole cell lysates , promastigotes were washed once in PBS and resuspended at 5 . 0×107 cells/ml in 1× SDS sample buffer . Supernatants were collected from log and stationary phase cultures after centrifugation . To generate detergent resistant membrane fractions ( DRMs ) , promastigotes were washed once in PBS and extracted with 1% of TritonX-100 ( at 1 . 0×108 cells/ml ) for 10 minutes at 4°C or 37°C . Detergent-soluble and -insoluble fractions were separated by centrifugation at 14 , 000 g for 2 minutes . An equal volume of 2 × SDS sample buffer was added to the detergent soluble fraction and two volumes of 1 × SDS sample buffer were added to the detergent insoluble fraction [53] . Samples were boiled for 5 minutes before SDS-PAGE . After transfer to PVDF membranes , blots were probed with either mouse-anti-LPG monoclonal antibody WIC 79 . 3 ( 1∶1000 ) [93] , or mouse-anti-GP63 monoclonal antibody #235 ( 1∶1000 ) [94] , followed by a goat anti-mouse IgG conjugated with HRP ( 1∶2000 ) . For C14DM-GFP , blots were probed with a rabbit anti-GFP HRP-conjugated antibody ( 1∶5000 ) . For loading controls , blots were probed with a mouse-anti-α-tubulin antibody or a rabbit anti-Leishmania HSP83 antibody . A FluorChem E system ( Protein Simple ) was used to detect and quantify signals . For LPG/GP63 localization , formaldehyde-fixed parasites were attached to poly-lysine coated cover slips and permeabilized with ice-cold ethanol . Cells were labeled with either mouse-anti-LPG antibody WIC79 . 3 or mouse-anti-GP63 antibody ( both at 1∶2000 dilution in 2% bovine serum albumin prepared in PBS ) for 20 minutes , and then incubated with a goat anti-mouse IgG-FITC ( 1∶1000 dilution ) for 20 minutes . For C14DM-GFP localization , c14dm−/+C14DM-GFP parasites were labeled with a rabbit anti-T . brucei BiP antiserum ( 1∶10 , 000 ) [46] for 30 minutes and then incubated with a goat anti-rabbit IgG-Texas Red antibody ( 1∶1000 dilution ) for 30 minutes . For mitochondrial staining , 1×106 parasites were centrifuged at 1000 g for 10 minutes , resuspended in 350 nM of Mitotracker Red 580 ( Life technologies ) in darkness; after 30 minutes , cells were washed in PBS once , and fixed with 3 . 7% formaldehyde; cells were then transferred to poly-L-lysine coated coverslips by centrifugation ( 462 g for 5 minutes ) , washed by 50% methanol , and stained with 1 . 0 µg/ml of Hoechst 33342 for 10 minutes . Images were acquired using an Olympus BX51 Upright Fluorescence Microscope equipped with a digital camera . Flow cytometry analyses for cell viability , DNA content , and surface LPG expression were performed as previously described [53] [95] , [96] , using a BD Accuri C6 flow cytometer . Total lipids were extracted according to a modified Folch's protocol [97] . Briefly , promastigotes were resuspended in chloroform: methanol ( 2∶1 ) at 1 . 0×108 cells/ml and vortexed for 30 seconds . An internal standard , cholesta-3 , 5-diene ( FW = 368 . 84 ) , was added to cell extract at 2 . 0×107 molecules/cell ( or 1 . 2 µg/108 promastigotes ) . Cell debris was removed by centrifugation ( 1000 g for 10 minutes ) and the supernatant was washed with 0 . 2 volume of 0 . 9% NaCl . After centrifugation , the aqueous layer was removed and the organic phase was dried under a stream of air . Lipid samples were then dissolved in methanol at the equivalence of 1 . 0×109 cells/ml . Lesion amastigotes were purified from infected mice as previously described [98] . Amastigote lipids were then extracted following the same procedure as promastigote samples except that the internal standard ( cholesta-3 , 5-diene ) was provided at 1 . 0×109 molecules/amastigote ( due to the high cholesterol level ) or 30 µg/5×107 amastigotes/footpad . Lipid from uninfected mouse footpads also contained the internal standard ( 30 µg/footpad ) . Electron impact GC/MS analyses of sterol lipids were performed on a Thermo Scientific ISQ ( San Jose , CA ) single-stage quadrupole mass spectrometer with Trace GC controlled by Thermo Xcalibur 2 . 1 software . The extract ( 1 µL ) was injected in a splitless mode and analyzed by GC on a Phenomenex ( Torrance , CA ) ZB-50 column ( 15 m , 0 . 32 mm id , 0 . 5 µm film thickness ) . The initial temperature of GC was set at 100°C for 2 min , increased to 200°C at a rate of 50°C/min , and then raised to a final temperature of 300°C at a rate of 10°C/min ( and then maintained at 300°C for 10 min ) . Temperatures of the injector , transfer line of the GC column , and of the ion-source were set at 280°C , 280°C , and 220°C , respectively . The full scan mass spectra ( 50 to 500 Dalton ) or total ion current chromatograms were acquired at a rate of 1 scan/0 . 2 sec . Electron ionization mass spectra of major Leishmania sterols were performed at 70 eV . Pure sterol standards ( zymosterol , lanosterol , cholesterol , ergosterol and 5-dehydroergosterol ) were also analyzed to obtain their electron impact mass spectra and GC retention times . Bone marrow derived macrophages were isolated from BALB/c mice as previously described [58] . Macrophage infection was performed using metacyclic promastigotes ( opsonized with C57BL6 mouse serum ) at a ratio of five parasites per macrophage [99] . Footpad infections of BALB/c mice were performed as previously described [58] using metacyclic promastigotes ( 2 . 0×105 cells/mouse ) or lesion-derived amastigotes ( 2 . 0×104 cells/mouse ) [100] . Lesion size ( the thickness of infected footpad minus the thickness of uninfected footpad ) was measured weekly using a Vernier caliper . Parasite numbers in the infected footpad were determined by the limiting dilution assay [101] . The plasma membrane fluidity of live Leishmania promastigotes was determined by measuring the fluorescence depolarization of TMA-DPH , as previously described for T . brucei [60] . Parasites were washed once with and resuspended in PBS at a density of 5 . 0×106 cells/mL . TMA-DPH was added to a final concentration of 0 . 5 µM and allowed to stain the cell membrane for 20 min at 4°C , 25°C , 37°C , or 45°C in the dark . Anisotropic values were acquired using a T-mode Photon Technology International ( Lawrenceville , NJ ) C61/2000 spectrofluorimeter . Samples were excited at 358 nm , and emission was read at 430 nm , with 10-nm excitation and emission slit widths . Temperature was maintained by means of the PerkinElmer LS55 Biokinetics accessory . Data were corrected for light scattering with an unlabeled sample of cells , and anisotropy was calculated according to the equation r = ( IVV − GIVH ) / ( IVV + 2GIVH ) , where r is the anisotropy value , IVV is the emission intensity acquired with the excitation- and emission-polarizing filters set vertically , G is the instrument correction factor , and IVH is the emission intensity acquired with the excitation-polarizing filter set vertically and the emission-polarizing filter set horizontally . Data points shown are the average of triplicate measurements with standard deviations . Most experiments ( except for the Southern blot in Fig . 1 ) were repeated at least three times . The difference between two groups was determined by the Student's t test using Sigmaplot 11 . 0 ( Systat Software Inc , San Jose , CA ) . P values indicating statistical significance were grouped into values of <0 . 05 and <0 . 01 . | Leishmania parasites are transmitted through the bite of sandflies causing a spectrum of serious diseases in humans . Current drugs are inadequate and no safe vaccine is available . These parasites produce different types of sterols from humans , making the sterol synthesis pathway a valuable target of selective inhibitors . However , functions of sterols and sterol synthesis in protozoa are poorly understood , which hinders the development of new and improved treatments . In this study , we investigated the role of sterol C14α-demethylase , a key enzyme in sterol metabolism and the primary target of azole drugs . Loss of sterol C14α-demethylase completely altered the sterol composition in Leishmania , leading to increased membrane fluidity , failure to maintain lipid rafts , and hypersensitivity to heat stress . Despite these defects , null mutants of sterol C14α-demethylase were viable during the promastigote stage ( found in sandflies ) and could still cause disease in mice ( although at a reduced capacity ) . Our findings provide direct evidence to support the role of specific sterols in membrane stability and stress response . The new knowledge may also help the development of new treatments or improve the efficacy of current drugs against pathogenic protozoa . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"biochemistry",
"lipids",
"metabolism",
"kinetoplastids",
"biology",
"and",
"life",
"sciences",
"protozoology",
"microbiology",
"sterols",
"lipid",
"metabolism",
"parasitology",
"parasite",
"physiology"
] | 2014 | Sterol Biosynthesis Is Required for Heat Resistance but Not Extracellular Survival in Leishmania |
Successful host colonization by bacteria requires sensing and response to the local ionic milieu , and coordination of responses with the maintenance of ionic homeostasis in the face of changing conditions . We previously discovered that Mycobacterium tuberculosis ( Mtb ) responds synergistically to chloride ( Cl- ) and pH , as cues to the immune status of its host . This raised the intriguing concept of abundant ions as important environmental signals , and we have now uncovered potassium ( K+ ) as an ion that can significantly impact colonization by Mtb . The bacterium has a unique transcriptional response to changes in environmental K+ levels , with both distinct and shared regulatory mechanisms controlling Mtb response to the ionic signals of K+ , Cl- , and pH . We demonstrate that intraphagosomal K+ levels increase during macrophage phagosome maturation , and find using a novel fluorescent K+-responsive reporter Mtb strain that K+ is not limiting during macrophage infection . Disruption of Mtb K+ homeostasis by deletion of the Trk K+ uptake system results in dampening of the bacterial response to pH and Cl- , and attenuation in host colonization , both in primary murine bone marrow-derived macrophages and in vivo in a murine model of Mtb infection . Our study reveals how bacterial ionic homeostasis can impact environmental ionic responses , and highlights the important role that abundant ions can play during host colonization by Mtb .
Ions are a fundamental component of organisms , playing roles in myriad biological processes . The ionic milieu surrounding a bacterium during infection varies with location and the immune response [1–7] , and successful host colonization thus requires proper sensing and response to the local ionic milieu , and the ability to coordinate responses with the maintenance of ionic homeostasis in the face of changing conditions . For Mycobacterium tuberculosis ( Mtb ) , pH ( [H+] flux ) is well-established as a critical signal during host colonization [8 , 9] , with the bacteria concurrently able to robustly maintain intrabacterial pH [10] . We have since identified chloride ( Cl- ) as an ionic cue important for Mtb [2] , but it remains an open question what other abundant ions may flux during Mtb colonization , and how bacterial ionic homeostasis may affect response to environmental ionic cues . Mtb is a bacterium of immense public health importance , being the leading cause of death from infectious diseases worldwide [11] . Uncovering facets of the local environment that the bacteria respond to , and how fundamental aspects of Mtb biology relate to environmental response and colonization success , is critical for comprehension of Mtb-host interactions and has the potential to reveal novel nodes that can be targeted for therapeutic purposes . Our previous discovery of Cl- as a novel environmental signal that Mtb responds to in synergy with pH indicated the significant role that flux and homeostasis of abundant ions can play during Mtb infection and disease [2] . In this context , potassium ( K+ ) , the most abundant intracellular cation in both mammalian and bacterial cells , stands out as a compelling candidate for study . In mammalian systems , K+ has fundamental roles in processes as diverse as renal function , muscle contraction , and neuronal information transmission [12–15] . There is also increasing appreciation for the importance of K+ in non-excitable cells , including in epithelial and immune cells in the lung [16 , 17] . In contrast , studies on K+ in bacterial systems have predominantly focused on its role in osmoprotection [18] . However , results from a few studies have begun to hint at the broader importance of K+ in influencing bacterial pathogenicity and disease outcome . For example , higher environmental K+ concentrations ( [K+] ) increased expression of virulence factors and host cell invasion by the food-borne pathogen Salmonella enterica [19] . Inactivation of the Trk K+ uptake system in S . enterica further resulted in defects in effector protein secretion through the bacterium’s type III secretion system , and attenuation in virulence in both murine and chick models of infection [20] . A bacterial K+ uptake system has been shown to be essential for Helicobacter pylori gastric colonization [21] , and K+ has also been shown to be part of a signal cascade that enables coordination of bacterial metabolic states in a Bacillus subtilis biofilm [22] . Mtb possesses two K+ uptake systems and one K+ efflux system [23–25] , but the role of K+ in Mtb infection biology has been largely unstudied . How does Mtb respond transcriptionally to local changes in [K+] , and how may this response be linked to the bacterial response to other environmental cues ? What is the impact of perturbation of bacterial K+ homeostasis on successful host colonization by Mtb ? In this study , we show that Mtb has a unique transcriptional response to changes in environmental [K+] . While the pH and Cl- response regulator PhoPR does not affect Mtb response to external K+ , the novel transcriptional repressor Rv0500A regulates bacterial response not just to pH and Cl- , but also K+ . By using a K+-responsive fluorescent reporter Mtb strain in macrophage infections , we find that K+ in the phagosome is not limiting , and instead increases during phagosome maturation . Disruption of Mtb K+ homeostasis by deletion of the Trk K+ uptake system ( ΔceoBC ) results in dampening of the bacterial response to pH and Cl- , revealing the links between environmental ionic responses and bacterial ionic homeostasis . Finally , we show that the ΔceoBC mutant is attenuated for host colonization , both in primary murine bone marrow-derived macrophages and in vivo in a murine model of Mtb infection .
To elucidate the global transcriptional response of Mtb to changes in environmental [K+] , we performed RNA sequencing ( RNAseq ) on Mtb exposed for 4 hours to K+-free 7H9 media , and compared it to samples in standard 7H9 media ( [K+] = 7 . 35 mM ) . We found that 43 genes were upregulated and 66 genes downregulated ( >2-fold , p<0 . 05 ) in response to low [K+] ( S1 and S2 Tables ) . The upregulated gene list represented a unique gene set , with slight overlap with genes upregulated in response to high [Cl-] , but otherwise little overlap when compared to other known environmental regulons ( Table 1 ) [2 , 8 , 26–28] . Among the strongest upregulated genes were those belonging to the kdpFABC operon . These genes encode the inducible K+ uptake ATPase KpdFABC , and their robust upregulation served to affirm their function in low environmental [K+] [24] . Interestingly , several genes related to iron uptake were repressed upon Mtb exposure to low environmental [K+] ( S2 Table ) . These included five mycobactin genes ( mbtB , mbtC , mbtD , mbtI , and mbtK ) , as well as the iron-regulated transporter genes irtA and irtB . Comparison to the previously reported gene set downregulated under conditions of high iron revealed a significant overlap ( Table 1 ) [27] . Oxidative stress is known to induce the expression of iron uptake genes , including the mbt genes [26] , and we accordingly also observed an overlap between the low [K+] downregulated gene set and genes upregulated upon exposure to oxidative stress ( Table 1 ) . Quantitative real time PCR ( qRT-PCR ) of candidate genes confirmed the RNAseq data ( Fig 1 ) . In addition , no expression differences in the candidate genes were observed in sodium ( Na+ ) -free media , reinforcing the specificity of the K+ response regulon ( S1 Fig ) . These transcriptional data reveal a significant response of Mtb to limiting external [K+] , with a unique signature of upregulated genes . The unexpected overlap with the iron and oxidative stress response regulons in the downregulated gene set provides an intriguing base for future studies . The transcriptional data above demonstrated the very robust induction of the kdpFABC operon upon exposure to low environmental [K+] [24] . To aid further study of Mtb response to K+ , we thus generated a fluorescent K+-responsive Mtb strain by cloning the promoter region of kdpF immediately upstream of GFP in a replicating plasmid , and transforming this kdpF’::GFP construct into Mtb . As expected , reporter GFP fluorescence was induced upon growth in low [K+] , with signal intensity increasing with decreasing [K+] ( Fig 2A ) . Testing of the kdpF’::GFP reporter in several other conditions known to be environmental cues for Mtb ( acidic pH , high [Cl-] , high osmolarity , and hypoxia ) revealed no fluorescence induction , illustrating its specificity ( Fig 2B and 2C ) . The lack of response to osmolarity is intriguing , as the KdpFABC system is canonically upregulated during exposure to high environmental osmolarity , in addition to limiting [K+] [29–31] . There is however precedence for a Kdp system being non-responsive to osmolarity , with this phenotype demonstrated for Kdp homologs in the cyanobacterium Anabaena and the extremophile Deinococcus radiodurans [32 , 33] . In agreement with our results , previous studies examining the response of Mtb to osmolytes ( NaCl , KCl , and sucrose ) had not observed changes in kdpFABC gene expression [2 , 34] . Mtb is extremely robust at coping with osmolarity , and it is possible that it utilizes distinct systems for osmolarity response . Indeed an osmosensory pathway regulated by the eukaryotic-like serine/threonine protein kinase PknD has been described for Mtb [34] . Expression of kdpFABC is expected to be modulated by the two-component system KdpDE , based on homology to model systems such as Escherichia coli [35] , and from Mtb studies examining regulatory partners of KdpD and the effect of over-expression of kdpE [24 , 36] . As expected by this regulation , induction of kdpF’::GFP signal in the presence of low environmental [K+] was lost in a ΔkdpDE Mtb mutant , with genetic complementation successfully restoring the WT phenotype ( Fig 2D ) . These data demonstrate the robustness of the kdpF’::GFP reporter as a specific and sensitive K+ reporter for Mtb , and support its utility for analyses of Mtb K+ response . With the kdpF’::GFP K+ reporter in hand , we next sought to determine if regulation of Mtb response to K+ is linked to the bacterial response to other abundant environmental ions , in particular H+ ( acidic pH ) and Cl- . Acidic pH is well-established as a critical environmental signal for Mtb during host colonization [8 , 10 , 37] , and we had discovered that the bacteria also respond transcriptionally to external Cl- , with significant overlap in the pH and Cl- response regulons [2] . Mtb response to pH and Cl- is synergistic [2] , and regulation of Mtb response to pH and Cl- is closely linked , with both shared activators and repressors . In particular , we had previously found that the two-component regulator PhoPR is a key activator of Mtb response to both acidic pH and Cl- , with attenuation of response upon PhoPR inactivation [2] . Conversely , in an initial screen for other genes controlling Mtb response to Cl- that utilized transposon libraries generated in the background of our previously described Cl- and pH-sensitive rv2390c’::GFP reporter Mtb strain [2] , we identified Rv0500A as a novel repressor of Mtb response to acidic pH and Cl- . Specifically , inactivation of Rv0500A was found to enhance expression of the rv2390c’::GFP reporter upon exposure to the inducing signals of Cl- and/or acidic pH ( Fig 3A ) . Expression of the K+-sensitive kdpF’::GFP reporter in a phoP transposon mutant ( phoP::Tn ) revealed no difference in signal induction upon exposure of the bacteria to low [K+] as compared to WT Mtb , indicating that the PhoPR regulator does not contribute to regulation of Mtb K+ response ( Fig 3B ) . In contrast , kdpF’::GFP expression was enhanced in a rv0500A transposon mutant ( rv0500A::Tn ) when the bacteria were grown in K+-free media ( Fig 3C ) , similar to the enhancement of rv2390c’::GFP reporter signal observed during exposure of the rv0500A::Tn mutant to high [Cl-] and/or acidic pH media ( Fig 3A ) . We note that the rv2390c’::GFP reporter does not respond to low external [K+] ( S2 Fig ) , in accord with its previously described high specificity for Cl- and pH [2] . To further explore the role of Rv0500A in Mtb response to ionic signals , we expressed and purified recombinant Rv0500A ( S3 Fig ) , and tested for its ability to directly bind the rv2390c and kdpF promoters . As shown in Fig 4A , a clear shift was observed in electrophoretic mobility shift assays ( EMSAs ) conducted with Rv0500A and the rv2390c promoter , demonstrating binding of the protein to the DNA fragment . In contrast , we did not observe a shift when the kdpF promoter was used ( Fig 4B ) . EMSAs with separated 5’ and 3’ sections of the rv2390c promoter showed a distinct shift with the 3’ but not the 5’ section , further illustrating the specificity of Rv0500A binding ( Fig 4C ) . Together , these data demonstrate that Rv0500A binds directly to specific DNA regions , and indicate that Rv0500A regulation of kdpF may be indirect , as opposed to its direct effect on rv2390c expression . Our findings demonstrate that Mtb response to the different ionic signals of K+ , Cl- , and pH are controlled by both distinct and shared regulatory mechanisms , and illustrate the complexity of Mtb integration of its response to disparate environmental cues . These data further reveal Rv0500A as a novel transcription regulator that represses Mtb response to these abundant ionic signals , and additional studies focused on analysis of Rv0500A function are currently in progress . pH and Cl- are key signals present in the Mtb phagosome [2 , 8 , 37] , and a previous study had indicated that [K+] in the lysosome decreases as a counterbalance to the influx of H+ [38] . While K+ levels in Mtb-containing phagosomes had been queried in an earlier intriguing study utilizing x-ray fluorescence microscopy , conclusions regarding changes in [K+] could not be drawn from the values obtained , and the resolution of the method had precluded differentiation of levels in the bacterium itself versus in the phagosomal compartment [39] . In the context of neutrophil phagosomes , there has been much debate as to whether [K+] increase is important upon generation of the oxidative burst [40–43] . Given our findings of both distinct and shared regulatory mechanisms underlying Mtb response to acidic pH , Cl- , and K+ , we next utilized our kdpF’::GFP reporter Mtb strain as a unique tool for interrogating the relative environmental [K+] the bacteria are exposed to during macrophage infection . For this assay , we used our previously established dual fluorescent reporter strategy , in which a constitutively expressed smyc’::mCherry is placed downstream of the environmental-responsive reporter ( here kdpF’::GFP ) , enabling visualization of all bacteria during infection [2 , 37 , 44] . Infection of primary murine bone marrow-derived macrophages with this dual kdpF’::GFP , smyc’::mCherry reporter Mtb strain revealed a decrease in kdpF’::GFP reporter signal over the course of eight days in both resting and activated macrophages ( Fig 5A and 5B ) . This downregulation in kdpF’::GFP signal was further emphasized when the reporter Mtb were grown in low environmental [K+] prior to infection ( Fig 5C and 5D ) . These data strongly indicate that K+ is not limiting within the Mtb phagosome , similar to conclusions reached for the Salmonella-containing vacuole that were based on transcriptional studies showing no upregulation of bacterial K+ transport systems in intracellular Salmonella [45] . Our findings further suggest that the outward K+ flux countering acidification of lysosomes previously observed does not function in the context of the Mtb phagosome [38] , in accord with Mtb prevention of complete acidification of its phagosome [46 , 47] . To further examine [K+] changes in the maturing phagosome with an independent approach , we utilized the K+-sensitive compound Asante K+ Green 4 ( APG4 ) [22 , 48] , which we covalently linked to 3 μm silica beads , together with Alexa Fluor 594 ( AF594 ) as a calibration fluorophore [2] . These K+ sensor beads fluoresced specifically as [K+] increased , with no APG4 fluorescence induced by changes in sodium concentration or pH ( Fig 6A and 6B ) . In agreement with the kdpF’::GFP reporter assay above , tracking of APG4/AF594 bead fluorescence upon addition to primary murine bone marrow-derived macrophages showed that K+ was not limiting in the maturing macrophage phagosome; instead , phagosomal [K+] increased as the compartment matured in both resting and activated macrophages ( Fig 6C ) . Inhibition of H+-ATPase activity with concanamycin A inhibits phagosome acidification and had previously been shown to affect phagosome maturation [49] . We found that concanamycin A treatment resulted in very significant loss of the increase observed in APG4/AF594 signal in mock-treated samples , further supporting the conclusion that [K+] increases during phagosome maturation ( S4 Fig ) . As the oxidative burst has been implicated as a driving factor in [K+] increase within neutrophil phagosomes [40] , we next asked if macrophages deficient in NADPH oxidase function ( gp91phox-/- ) would exhibit the increase in phagosomal [K+] observed in WT macrophages . We found that gp91phox-/- macrophages retained their ability to increase phagosomal [K+] , indicating that the oxidative burst is not required for this phenotype in macrophages ( Fig 6D ) . These findings add to the knowledge base of changes in the ionic milieu during macrophage phagosome maturation , and further studies are required to delineate the channels/transporters that mediate this K+ flux . Having established that [K+] increases during macrophage phagosome maturation , similar to increases in [Cl-] and [H+] [2] , and in light of the interplay in Mtb response to these ions illustrated by our results above with Rv0500A , we next sought to determine if disruption of Mtb K+ homeostasis would affect bacterial response to Cl- and pH . Mtb has two K+ uptake systems: ( i ) Kdp , a high affinity , inducible system , and ( ii ) Trk , a constitutive , low-moderate affinity system [23 , 50 , 51] . Given our finding that K+ is not limiting within the Mtb phagosome , we focused our studies here on the Trk K+ uptake system , encoded by the ceoB and ceoC genes . A ΔceoBC Mtb mutant was constructed via homologous recombination , and the Cl- and pH-responsive rv2390c’::GFP reporter introduced into the strain [2] . In agreement with the role of CeoBC as a constitutive , low-moderate affinity K+ uptake system , tests with the kdpF’::GFP reporter indicated that the ΔceoBC mutant is slightly more sensitive to low environmental [K+] ( S5 Fig ) . The ΔceoBC mutant had no growth defect in standard broth , or in high [Cl-] , acidic pH , or low [K+] media , as compared to WT Mtb ( Fig 7A ) . Strikingly however , the ΔceoBC mutant demonstrated significantly decreased reporter GFP signal when grown in defined broth conditions with high [Cl-] and/or acidic pH , which was restored to WT levels upon genetic complementation ( ceoBC* ) ( Fig 7B ) . To establish that an attenuated response to environmental [Cl-] and pH by the ΔceoBC mutant was a broader phenotype not restricted to rv2390c response , we performed qRT-PCR on several other Cl- response regulon genes [2] . As shown in Fig 7C , the transcriptional response of the ΔceoBC mutant was dampened in each case , although the magnitude of reduction varied from gene to gene . To test that the phenotypes observed were due more directly to disruption of bacterial K+ homeostasis versus possible downstream effects on the bacterial membrane from deletion of an ion transport system , we first assessed the ability of the ΔceoBC mutant to maintain intracellular pH during growth in acidic conditions . As previously described , the pH-sensitive , cell-permeable , dye 5-chloromethylfluorescein diacetate ( CMFDA ) is rendered cell-impermeant once in the cytoplasm of cells , and thus enables measurement of intracellular pH in a ratiometric manner , as fluorescein is pH sensitive when excited at 490 nm , but pH insensitive when excited at 450 nm ( emission at 520 nm in both cases ) [52] . We found that the ΔceoBC mutant maintained intracellular pH as well as WT Mtb , even when exposed to a pH 5 . 7 environment ( Fig 7D ) . Treatment of the bacteria with nigericin , an ionophore , resulted in the expected decrease in excitation 490 nm/450 nm ( emission 520 nm ) fluorescence ratio in the presence of external acidity , and served as an assay control ( Fig 7D ) . Next , we tested for defects in maintenance of membrane potential . The cationic dye Rhodamine 123 allows relative determination of cellular membrane potential , with increased uptake , and consequently greater fluorescence decay over time , in cells with higher membrane potential [53 , 54] . In standard pH 7 growth media , WT and ΔceoBC Mtb had similar membrane potential ( Fig 7E ) . Exposure of Mtb to an acidic external pH resulted in the expected membrane depolarization , but here again no difference in membrane potential was observed between WT and ΔceoBC Mtb ( Fig 7E ) . Membrane potential of the ΔceoBC mutant was also not different to that of WT Mtb in conditions of low environmental [K+] ( Fig 7E ) . The effect of protonophore ( carbonyl cyanide m-chlorophenylhydrazone , CCCP ) addition was analyzed as a control , and resulted in the expected membrane depolarization and decrease in relative fluorescence decay in each condition ( Fig 7E ) . These results reveal the novel concept that bacterial K+ homeostasis can impact on Mtb response to ionic signals in its local environment , independent of broader effects on maintenance of membrane potential . This new concept highlights the complexity of bacterial K+ biology , and the intimate relationship between ionic homeostasis and ionic signal response . It further raises the intriguing suggestion that targeting of Mtb K+ homeostasis will have far-reaching adverse consequences for the bacterium that extend beyond K+ balance alone . The changing [K+] in the phagosome and the observed impairment in bacterial response to pH and Cl- in the ΔceoBC mutant suggests that this mutant may be attenuated in colonization of macrophages . To test this , we infected primary murine bone marrow-derived macrophages with WT , ΔceoBC , or ceoBC* Mtb , and tracked bacterial growth over time . We found that the ΔceoBC mutant was significantly attenuated in its ability to colonize macrophages , with the bacterial load recovered 6 days post-infection only about half that obtained from WT or ceoBC*-infected samples ( Fig 8A ) . In agreement with our results , we note that ceoB mutants had previously been found to be under-represented in a transposon site hybridization-based screen for Mtb genes important for survival/growth in macrophages [55] . Utilization of the rv2390c’::GFP , smyc’::mCherry reporter in the various strains further showed that rv2390c’::GFP signal was decreased in the ΔceoBC mutant as compared to that in WT or ceoBC* Mtb during infection of activated macrophages ( Fig 8B and 8C ) . These reporter data suggest that the impaired response of the ΔceoBC mutant to Cl- and pH observed in broth is similarly exhibited during macrophage infection . Finally , to examine the effect of disrupting Mtb K+ homeostasis on the bacterium’s ability to colonize a whole animal host , we infected C57BL/6J WT mice with WT , ΔceoBC , or ceoBC* Mtb , and analyzed bacterial load and histopathology two and four weeks post-infection . The ΔceoBC mutant was significantly attenuated for host colonization at both time points , with bacterial loads 3–4 fold less than that obtained from infections with WT or ceoBC* ( Fig 9A ) . Histopathology in ΔceoBC-infected animals was correspondingly less severe , with decreased cellular infiltration and better maintenance of clear alveolar air spaces ( Fig 9B ) . These results demonstrate a role for the CeoBC K+ uptake system in Mtb pathogenesis , and underscore the importance of K+ homeostasis for the bacterium in successfully establishing host colonization . From the macrophage phagosome to the granuloma , Mtb is exposed to a complex environmental milieu that plays a significant role in its infection biology . Signals present reflect cellular and tissue location , as well as the host immune status [1–4] , and Mtb must coordinate its response to disparate signals to ensure adaptation and continued survival and growth . Our discovery that disruption of the CeoBC K+ uptake system results in attenuation of Mtb response to external pH and Cl- levels , independent of changes in membrane potential , reveals the close links between ionic homeostasis and bacterial response to diverse environmental cues . Together with our previous findings demonstrating the synergistic response of Mtb to acidic pH and Cl- [2] , these results highlight the need to examine Mtb ionic homeostasis and response beyond the study of individual signals in isolation . Our discovery of Rv0500A as a shared repressor affecting Mtb response to pH , Cl- , and K+ begins to shed mechanistic insight into this key facet , and studies to elucidate the role of Rv0500A during infection are ongoing . Interestingly , a global transcription factor study by Rustad et al . identified over-expression of glnR ( rv0818 ) , a master nitrogen metabolism regulator [56 , 57] , as inducing transcription of kdpFABC in standard broth conditions where this K+ uptake system is not normally upregulated [36] . It will thus also be intriguing in future studies to assess the connections that may exist between K+ signals and nitrogen metabolism proposed by this data . In addition , the observed attenuation of the ΔceoBC mutant in host colonization suggests that uncovering how Mtb integrates ionic homeostasis and response to environmental cues represents an important area for understanding mechanisms that enable effective bacterial colonization of the host , essential for complete comprehension of Mtb pathogenesis . Our findings are likely also applicable to other bacterial pathogens , given the widespread presence of the Trk ( CeoBC ) K+ uptake system , which in the case of Salmonella has also been shown to play an important role during infection [20] . More broadly , studies on the role of ions in host-pathogen interactions have most often focused on scarce ions that can be sequestered by the host , such as iron , zinc , and manganese [6 , 7 , 58 , 59] . These ions are clear targets in understanding bacterial pathogenesis , given both their scarcity and their importance for bacterial and host physiology . Much less studied however has been the role that abundant ions , like K+ and Cl- , can play in infectious disease . While such ions are not a resource that bacteria have to “fight” their host for , we propose that their very abundance , inability to be sequestered by the host , varying levels in different cellular compartments and upon immune activation , and critical roles in host biology , make them ideal targets for exploitation by pathogens . Our findings of the change in [K+] during macrophage phagosome maturation , and of the specific transcriptional response of Mtb to K+ , builds on our previous results with Cl- [2] , and raise questions about how bacteria may exploit [K+] flux during infection . S . enterica presents a noteworthy example , given the reported increase in virulence factor expression upon exposure to higher environmental [K+] [19] . Might external K+ levels also affect the ability of Mtb to adapt to its local environment ? It further remains to be elucidated when Mtb may experience limiting environmental [K+] , and the possible link to iron and oxidative stress suggested by our transcriptional data is an intriguing finding that merits further study . Of note , K+ levels have been implicated in macrophage immune responses , with inhibition of K+ channels reducing nitric oxide induction [60] , and low intracellular K+ activating the NALP3 inflammasome [61] . K+ may thus function as a signal both for the host and the bacterium , and further studies are needed to determine crosstalk and balance in these processes during host-pathogen interactions . We propose that further studies pursuing the novel concepts highlighted by our results here will provide important insight into how fundamental aspects of both Mtb and host biology are linked during the infection process , and reveal unique nodes that can be perturbed to shift the balance of infection .
All animal procedures followed the National Institutes of Health “Guide for the Care and Use of Laboratory Animals” standards . The Institutional Animal Care and Use Committee at Tufts University reviewed and approved all animal protocols ( #B2016-37 ) , in accordance with guidelines established by the Association for Assessment and Accreditation of Laboratory Animal Care , the US Department of Agriculture , and the US Public Health Service . Mtb cultures were propagated as previously described [37] . Strains for in vitro assays were in the CDC1551 background , while strains for CFU enumeration in macrophages and in vivo were in the Erdman background . Mtb mutants and their complements were constructed as previously described [2] . The ΔceoBC mutation consisted of a deletion beginning 102 bp from the ceoB start codon through 73 bp from the ceoC stop codon , while the entire open reading frame was deleted for ΔkdpDE . The phoP::Tn mutant from BEI Resources ( NR-14776 ) has been previously described [2 , 37] . The rv0500A::Tn mutant was isolated from a Tn library generated in the background of a CDC1551 ( rv2390c’::GFP ) strain via use of a mariner-based ϕMycoMarT7 phage [62 , 63] , with the Tn insertion 125 bp into the open reading frame . Log-phase ( OD600 ~ 0 . 6 ) Mtb was used to inoculate standing , vented , T-25 flasks containing either 10 ml of standard 7H9 , pH 7 medium ( [K+] = 7 . 35 mM ) or K+-free 7H9 , pH 7 medium at OD600 = 0 . 3 . K+-free 7H9 was made by replacing monopotassium phosphate with monosodium phosphate . To make Na+-free 7H9 , disodium phosphate was replaced with dipotassium phosphate , monosodium glutamate with monopotassium glutamate , and sodium citrate with potassium citrate . The media was supplemented with 0 . 5% bovine serum albumin , 0 . 2% dextrose , and 14 . 5 mM KCl , and brought to pH 7 with KOH . Samples were collected 4 hours post-exposure and RNA isolation carried out as previously described [8] . Two biological replicates per condition were used for RNA sequencing . Library preparation using Ribo-Zero rRNA removal ( bacterial ) and TruSeq Stranded kits ( Illumina ) were performed by the Tufts University Genomics Core Facility . Barcoded samples were pooled and run on a single lane on an Illumina HiSeq 2500 ( High Output v4 ) with single-end 100 bp reads . Data were analyzed using the SPARTA program [64] . qRT-PCR experiments were performed as previously described , except cDNA was synthesized from 250 ng of RNA without prior amplification [2 , 37] . The Cl- and pH-sensitive rv2390c’::GFP reporter has been previously described [2] . A similar approach was used to construct the K+-sensitive reporter kdpF’::GFP , with the 754 bp region immediately upstream of kdpF amplified by PCR and fused to GFPmut2 [2 , 37] . The resulting kdpF’::GFP construct was then introduced into a vector containing the constitutively expressed smyc’::mCherry , and transformed into Mtb . Selection was carried out on 7H10 agar containing 25 μg/ml kanamycin , 50 μg/ml apramycin , or 50 μg/ml hygromycin B as appropriate . Broth culture assays were conducted by growing Mtb in standing vented T-25 flasks in 10 ml of 7H9 buffered media as previously described [2] , with addition of 250 mM NaCl or 300 mM sucrose as needed . To vary [K+] , a specified amount of KCl was added to the K+-free 7H9 . Antibiotics were added as necessary to maintain selection . The hypoxia assays were performed as previously described [2] . Mtb was fixed in 4% paraformaldehyde ( PFA ) in phosphate buffered saline ( PBS ) prior to fluorescence analysis . GFP was read on a BD FACSCalibur with subsequent data analysis using FloJo ( Tree Star , Inc ) . The rv0500A open reading frame from CDC1551 was cloned into pET-28a ( Novagen ) to construct an in-frame N-terminal 6x-His tag fusion . A sequence verified construct was then transformed into E . coli BL-21 ( DE3 ) . For expression , a single transformant colony from an LB plate containing 50 μg/ml kanamycin was picked and grown overnight at 37°C on a roller drum in 2 . 5 ml of LB broth + 50 μg/ml kanamycin . 1 L of 2YT ( 5 g NaCl , 10 g yeast extract , and 15 g tryptone per liter ) with 50 μg/ml kanamycin was subsequently inoculated with 2 ml of the overnight culture and grown shaking at 37°C to an OD600 of 0 . 75 . To induce protein expression , 1 mM isopropyl-β-D-1-thiogalactopyranoside ( IPTG ) was added , and the culture grown overnight , shaking , at 18°C . Induced culture was pelleted the next day and resuspended in 25 ml lysis buffer ( 500 mM NaCl , 50 mM Tris pH 7 . 5 , 15 mM imidazole , 10% glycerol ) . The resuspended cells were flash frozen in liquid nitrogen , thawed , and lysed via sonication . The insoluble fraction was pelleted and the cleared lysate mixed with 1 ml of Ni-NTA agarose ( Macherey-Nagel ) and incubated overnight at 4°C on a nutator . Purified protein was eluted the next day using high-imidazole buffer ( 500 mM NaCl , 50 mM Tris pH 7 . 5 , 200 mM imidazole , 10% glycerol ) ( elution repeated 8 times in total ) . Samples were run on a 20% SDS-PAGE gel , visualized with Coomassie Brilliant Blue R-250 ( Bio-Rad ) staining , and imaged using the 700 nm channel of an Odyssey CLx imaging system ( LI-COR ) . Protein concentration was quantified using a Bradford assay ( Bio-Rad ) . To assess DNA binding of Rv0500A , promoter regions for rv2390c ( full length—704 bp; 5’ section—403 bp; 3’ section—301 bp ) and kdpF ( 754 bp ) were amplified using IRDye 700 labeled primers ( Integrated DNA Technologies ) and the PCR products purified using a QIAquick PCR purification kit ( Qiagen ) . Indicated amounts of purified His-Rv0500A were mixed with no more than 40 fmoles of DNA in EMSA buffer ( 20 mM Tris-HCl , pH 8 , 50 mM KCl , 2 mM MgCl2 , 5% glycerol , 0 . 5 mM EDTA , 1 mM DTT , 0 . 05% Nonidet P-40 , 25 μg/ml salmon sperm DNA [65] ) in 11 μl final volume reactions . After incubation at room temperature for 20 minutes , the reactions were run on a non-denaturing 8% Tris-glycine gel in HEPES-imidazole buffer ( 35 mM HEPES , 43 mM imidazole ) . The gel was then imaged using the 700 nm channel of an Odyssey CLx imaging system ( LI-COR ) . Mtb grown to mid-log phase in standard 7H9 broth , pH 7 , was pelleted and resuspended to OD600 = 0 . 6 in 10 ml of PBS ( 1 . 54 mM KH2PO4 , 2 . 71 mM Na2HPO4 , 155 mM NaCl , pH 7 . 3 ) containing 0 . 05% Tween 80 and 10 μM 5′-chloromethylfluorescein diacetate ( Invitrogen ) for 1 hour , protected from light . Bacteria were washed twice in their respective buffer type , before resuspension to OD600 = 1 . 2 in 500 μl of appropriate buffer ( PBS , pH 7 . 3 or pH 5 . 7; K+-free PBS , pH 7 . 3 or pH 5 . 7 ) . To generate K+-free PBS , NaH2PO4 was substituted for KH2PO4 . For control samples , 10 μM of the ionophore nigericin was added . 100 μl samples were transferred to a 96-well clear bottom black plate ( Corning Costar ) , with each condition read in triplicate . After equilibration , top read fluorescence at Ex . 450/10 nm , Em . 520/10 nm , and Ex . 490/10 nm , Em . 520/10 nm , was measured using a Biotek Synergy Neo2 microplate reader . The ratio of fluorescence signal at Ex . 490 nm , Em . 520 nm , versus that at Ex . 450 nm , Em . 520 nm , was used to determine relative intrabacterial pH . Data were normalized as a percentage of the fluorescence ratios observed in PBS , pH 7 . 3 , for a given strain [52] . Mtb grown to mid-log phase were resuspended to OD600 = 0 . 6 in ( i ) 7H9 , pH 7 , ( ii ) 7H9 , pH 5 . 7 , or ( iii ) K+-free 7H9 , pH 7 , with each containing 0 . 5 mg/l rhodamine 123 ( Sigma ) [53] . 50 μM carbonyl cyanide m-chlorophenylhydrazone ( Sigma ) was added for control samples . 150 μl quadruplicate aliquots of the bacterial suspensions were then transferred to a 96-well clear bottom black plate . Fluorescence decay was tracked every 10 minutes for 4 hours with a Biotek Synergy Neo2 microplate reader ( bottom read Ex . 485/10 nm , Em . 527/10 nm ) , and normalized to initial probe fluorescence for each strain and condition . C57BL/6J wild type and gp91phox-/- mice ( Jackson Laboratories ) were used for isolation of bone marrow-derived macrophages . Cells were maintained in DMEM supplemented with 10% FBS , 15% L-cell conditioned media , 2 mM L-glutamine , 1 mM sodium pyruvate , and penicillin/streptomycin as needed , in a 37°C incubator at 5% CO2 . Macrophage infections with Mtb were performed as previously described [2 , 37] . For CFU enumeration , macrophages were lysed with water containing 0 . 01% sodium dodecyl sulfate . The strain expressing kdpF’::GFP was pre-induced for 6 days in K+-free 7H9 , pH 7 , prior to infection as needed . Fixed Mtb-infected macrophages were stained with DAPI and Alexa Fluor 647 phalloidin ( Invitrogen ) for visualization of nuclei and F-actin respectively . Stained samples were mounted with ProLong Diamond antifade , and images acquired with Leica LAS X software using a Leica SP8 confocal laser scanning microscope , with 0 . 5 μm z-steps . 3D reconstruction and image analyses were carried out with Volocity software ( PerkinElmer ) as previously described [2 , 66 , 67] . K+ sensor beads were generated by first covalently linking human IgG ( Sigma ) and fatty acid-free bovine serum albumin ( Sigma ) to 12 . 5 mg of carboxylated , 3 μm silica beads ( Kisker Biotech ) as previously described [2 , 68] . Beads were then washed once with coupling buffer ( 0 . 1 M sodium borate , pH 8 . 0 ) , and twice with PBS . 20 mg of EDC ( Sigma ) and 25 μg of Asante Potassium Green 4 , TMA+ salt ( TEFLabs ) were added to the beads in 1 ml of PBS , and the mixture incubated with agitation for 2 hours at room temperature , protected from light . APG4 beads were washed three times with PBS , before resuspension in 1 ml of coupling buffer containing 25 μg of Alexa Fluor 594-SE ( Invitrogen ) , and incubation with agitation for an additional 1 hour at room temperature , protected from light . APG4/AF594 beads were washed twice with coupling buffer , and once with PBS , before final resuspension in 500 μl of PBS . Induction of APG4 fluorescence by K+ was tested by placing APG4/AF594 beads in assay buffer ( 1 . 54 mM NaH2PO4 , 2 . 71 mM Na2HPO4 , 157 . 87 mM NaCl , 5 mM dextrose , 1 mM CaCl2 , 0 . 5 mM MgCl2 ) supplemented with specified concentrations of K+ gluconate , in a 96-well clear bottom black plate . Na+ gluconate supplementation was used for Na+ response tests . For tests with [Na+] less than 157 . 87 mM , NaCl was excluded from the assay buffer . Tests for response to pH was conducted in assay buffer with 100 mM Bis-Tris added , and pH adjusted with HCl . Each condition was tested in quadruplicate , and bottom reads of Ex . 510 nm/Em . 540 nm ( APG4 ) and Ex . 590 nm/Em . 617 nm ( AF594 ) taken after equilibration with a Biotek Synergy H1 microplate reader . For intraphagosomal K+ bead assays , primary murine bone marrow-derived macrophages ( 2x105/well ) were seeded in a 96-well clear bottom black plate . Where needed , 100 U/ml IFNγ ( PeproTech ) and 10 ng/ml LPS ( List Biological Laboratories ) were used for activation of macrophages . 100 nM concanamycin A ( AdipoGen Life Sciences ) was added to the media where indicated . Macrophages were washed three times with pre-warmed assay buffer , before addition of APG4/AF594 beads ( ~2–5 beads/macrophage ) . A Biotek Synergy H1 microplate reader was used for fluorescence tracking , with four-six replicate wells/condition and bottom reads taken every two minutes for four hours . Temperature was maintained at 37°C throughout the assay . C57BL/6J wild type mice ( Jackson Laboratories ) were infected intranasally with 103 CFUs of Mtb ( 35 μl ) , under light anesthesia with 2% isoflurane . After sacrifice with CO2 , the left lobe and accessory right lobe were homogenized in PBS containing 0 . 05% Tween 80 and serial dilutions plated on 7H10 agar plates containing 100 μg/ml cycloheximide for CFU determination . The remaining three right lobes were fixed in 4% PFA in PBS , and used for histology examination via standard hematoxylin and eosin ( H&E ) staining ( Tufts Comparative Pathology Services ) . H&E stained slides were imaged on a Nikon Eclipse E400 equipped with a SPOT Insight color digital camera . RNA sequencing data has been deposited in the NCBI GEO database ( GSE120725 ) . | Tuberculosis is the leading cause of death from infectious diseases globally , and knowledge of the fundamental biology underlying successful host colonization by Mycobacterium tuberculosis ( Mtb ) is critical for understanding its pathogenicity . Our study focuses on one such important facet—the ability of Mtb to sense and respond to changing environmental ionic signals , and its relation to maintenance of bacterial ionic homeostasis . Potassium ( K+ ) is the most abundant ion in both mammalian and bacterial cells , but much remains unknown about how it may affect host-pathogen interactions . We show here that Mtb has distinct gene expression changes in response to variation in environmental K+ levels . Importantly , disruption of a Mtb K+ uptake system reduces the ability of Mtb to respond to other key ionic cues , and results in diminished host colonization . Our study reveals novel roles of K+ during Mtb-host interactions , and suggests the broader importance of abundant ions in bacterial pathogenicity . | [
"Abstract",
"Introduction",
"Results",
"and",
"discussion",
"Materials",
"and",
"methods"
] | [
"blood",
"cells",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"vesicles",
"pathology",
"and",
"laboratory",
"medicine",
"electrophoretic",
"mobility",
"shift",
"assay",
"pathogens",
"immunology",
"membrane",
"potential",
"microbiology",
"electrophysiology",
"tuberculosis",
"drug",
"discovery",
"physiological",
"processes",
"homeostasis",
"phagosomes",
"pharmacology",
"molecular",
"biology",
"techniques",
"cellular",
"structures",
"and",
"organelles",
"bacteria",
"bacterial",
"pathogens",
"research",
"and",
"analysis",
"methods",
"white",
"blood",
"cells",
"gene",
"mapping",
"animal",
"cells",
"medical",
"microbiology",
"microbial",
"pathogens",
"actinobacteria",
"molecular",
"biology",
"drug",
"discovery",
"cell",
"biology",
"drug",
"research",
"and",
"development",
"mycobacterium",
"tuberculosis",
"physiology",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"macrophages",
"restriction",
"fragment",
"mapping",
"organisms"
] | 2019 | Potassium response and homeostasis in Mycobacterium tuberculosis modulates environmental adaptation and is important for host colonization |
Human waste is used as an agricultural fertilizer in China and elsewhere . Because the eggs of many helminth species can survive in environmental media , reuse of untreated or partially treated human waste , commonly called night soil , may promote transmission of human helminthiases . We conducted an open cohort study in 36 villages to evaluate the association between night soil use and schistosomiasis in a region of China where schistosomiasis has reemerged and persisted despite control activities . We tested 2 , 005 residents for Schistosoma japonicum infection in 2007 and 1 , 365 residents in 2010 and interviewed heads of household about agricultural practices each study year . We used an intervention attributable ratio framework to estimate the association between night soil use and S . japonicum infection . Night soil use was reported by half of households ( 56% in 2007 and 46% in 2010 ) . Village night soil use was strongly associated with human S . japonicum infection in 2007 . We estimate cessation of night soil use would lead to a 49% reduction in infection prevalence in 2007 ( 95% CI: 12% , 71% ) . However , no association between night soil and schistosomiasis was observed in 2010 . These inconsistent findings may be due to unmeasured confounding or temporal shifts in the importance of different sources of S . japonicum eggs on the margins of disease elimination . The use of untreated or partially treated human waste as an agricultural fertilizer may be a barrier to permanent reductions in human helminthiases . This practice warrants further attention by the public health community .
For centuries , people have collected human waste and used the material , called night soil , to fertilize agricultural crops [1 , 2] . The use of human fecal waste as an agricultural fertilizer has the potential to improve crop yields without the expense , environmental risk or transportation infrastructure required of synthetic fertilizers . However , if the waste is not properly treated , the practice may promote fecal-borne diseases [3] . High temperatures , high pH , desiccation and the introduction of additives such as leaf litter and ash can reduce pathogen loads in human waste products , allowing use on agricultural commodities . But fresh human waste or waste that has been stored under suboptimal conditions , may contain human pathogens , presenting risks to individuals that handle night soil and to those who may come into contact with food , soil or water in areas where night soil is applied [3] . Helminth eggs are particularly hardy and may require longer storage and/or higher temperatures and pH for destruction relative to other pathogens found in human stool [4] . The use of night soil may be particularly hazardous in the context of schistosomiasis , a water-borne infection that causes anemia , fibrosis of the liver and kidneys and impairs growth and cognitive development [5 , 6] . The eggs of Schistosoma japonicum , the parasite that causes intestinal schistosomiasis in Asia , and S . mansoni , which causes intestinal schistosomiasis in the Americas , Africa and the Middle East , are excreted in stool . Upon contact with water , eggs hatch into miracidia and then must mature within a snail host before the parasite can infect humans . Schistosome eggs can survive for several days ( S . mansoni ) or weeks ( S . japonicum ) in excreted stool [4] . Schistosome host snails thrive in irrigation ditches and rice paddies [7–9] . Thus , the application of human waste to agricultural crops can transport schistosome eggs from stool pits directly to snail habitat , facilitating the schistosome life cycle . The use of night soil may be a barrier to permanent schistosomiasis control in China . Chinese public health officials are attempting elimination of schistosomiasis using a multi-pronged control effort that includes anti-helminthic treatment , reduction of snail populations and improvements to water and sanitation infrastructure [10] . While schistosomiasis disease burden has declined in China and transmission has been eliminated in five provinces , pockets of transmission persist and reemergence has been detected in previously controlled areas [11–13] . The challenge of sustained interruption of schistosomiasis is increasingly relevant outside of China as endemic countries adopt antihelminthic treatment programs and look towards permanent reductions in schistosomiasis infections and morbidity [14] . The reemergence and persistence of schistosomiasis may depend , in part , on the internal potential of an area , the set of conditions that make an area hospitable ( or inhospitable ) to the schistosome lifecycle such as the availability of snail habitat , human water contact patterns and waste disposal practices [15 , 16] . The output of S . japonicum eggs into the environment and efforts to reduce egg output through improved sanitation have been previously shown to impact long-term disease patterns in endemic areas [15] . Improving sanitation access is a priority for the schistosomiasis control program in China , and the importance of safe sanitation is recognized by global programs to reduce human helminthiases [10 , 14] . However , night soil use has not been a focus of schistosomiasis control activities . Here , we evaluate the association between night soil use and human schistosomiasis in a region where schistosomiasis reemerged and persisted in the presence of ongoing disease control programs . Because information about the use of human waste as an agricultural fertilizer is limited in the public health literature , we first document night soil practices in 36 villages in rural China . We then test the hypothesis that greater night soil use in a village is associated with increased human S . japonicum infection prevalence .
This research was conducted in two rural counties in Sichuan , China where schistosomiasis reemerged following the reduction of human and bovine infection prevalence below 1% , the Chinese Ministry of Health threshold for transmission control [12] . In 2007 , we selected 53 villages in 3 counties where schistosomiasis had reemerged for a longitudinal study of social and environmental determinants of schistosomiasis reemergence [17] . One county was excluded from follow-up because it was severely impacted by the 7 . 9 magnitude earthquake that struck Sichuan May 12 , 2008 . The analysis presented here includes the 36 villages in the two counties followed through 2010 . The names and exact locations of the study villages have been withheld to protect the privacy of study participants and promote candid reporting . In rural Sichuan , populations fluctuate due to rural-to-urban migration , as well as marriages , births and deaths . We therefore employed an open cohort design , conducting a census in 2007 and 2010 in order to identify all residents living in the study villages , age 6 and older . All individuals identified in each census were recruited for S . japonicum infection testing and household surveys . In 2007 , we identified 2 , 891 eligible residents . In 2010 , we identified 2 , 287 residents , including 1 , 875 identified in the previous census and 412 new residents . Of the 1 , 016 people identified in 2007 that were no longer residents in 2010 , 760 had left their village for work , 203 had left to attend school , 33 had died and 20 had left for marriage . The research protocol was approved by the Sichuan Institutional Review Board and the University of California , Berkeley , Committee for the Protection of Human Subjects . All participants provided written , informed consent before participating in this study . All children provided assent and their parents or guardians provided written , informed permission for them to participate in this study . Everyone testing positive for S . japonicum was notified and provided treatment with 40 mg/kg praziquantel by the anti-schistosomiasis control station . Household interviews . In the summer of 2007 and 2010 , the head of each household was invited to complete a structured interview about agricultural practices , sanitation access and socio-economic status ( SES ) . Survey instruments were pilot tested in the study region and administered by trained public health workers fluent in the local dialect . We interviewed 1 , 156 households in 2007 and 951 households in 2010 . Participants reported all crops planted in the past 12 months , the quantity of night soil applied to each crop and whether chemical fertilizers were used . The amount of night soil used by each household was calculated as the total quantity of night soil applied to all crops in the past 12 months . An additional metric of night soil use was also developed for supplemental analyses in light of evidence that improved toilet designs reduce the number of viable helminth eggs in effluent [18 , 19] . Households with a working anaerobic biogas digester or triple compartment septic tank were classified as having improved sanitation , and night soil applied by such households was classified as coming from improved sanitation sources . Night soil applied by all other households was classified as coming from unimproved sources . SES was assessed using an asset-based approach . Household assets can provide a more stable estimate of long-term wealth than monetary income in agrarian regions , where income is episodic and includes agricultural products [20 , 21] . The index included ownership of eight durable goods ( car , motorcycle , tractor , computer , television , air conditioner , washing machine and refrigerator ) , reported by the head of household . Principal components analysis was used to create an aggregated SES score , deriving weights from the first principal component which explained 28% of the total variance between measures . Multiple imputation by chained equations was used to impute missing household survey data . Multiple imputation avoids biases due to exclusion of incomplete cases and reduces the variance introduced by the uncertainties of imputation through the generation of multiple datasets [22 , 23] . The method assumes that data are missing at random—that missingness can be explained by the other , observed variables used in the imputation . We imputed the quantity of night soil used , the number of bovines owned by the household , SES score , whether the household had an improved toilet and the area of land cultivated by the household . Because the volume of night soil used and the number of bovines per household were highly right-skewed , predictive mean matching was used for the imputation of these variables . We imputed the missing data using the aforementioned variables , including observations from prior/future time points and county of residence , generating 20 imputed datasets . Imputed values were used for the 19% of households missing each of these survey measures , including the 482 households not interviewed and 13 households with incomplete survey data . S . japonicum infection testing . Everyone identified in each census was invited to submit three stool samples on three consecutive days in November/December of 2007 and 2010 for S . japonicum infection testing . Each sample was examined using the miracidium hatching test [24] . Briefly , 30 g of stool were suspended in aqueous solution , strained with copper mesh to remove large particles and strained with nylon mesh to concentrate schistosome eggs . The sediment was re-suspended in water and left undisturbed in a room with ambient temperatures between 28 and 30°C . Samples were examined two , five and eight hours after sample preparation for the presence of miracidia . One stool sample per person was also examined using the Kato-Katz thick smear procedure [25] . Three slides were prepared from each sample , using 41 . 7 mg stool per slide . Slides were allowed 24 hours to clear and were examined for S . japonicum eggs by trained technicians using a dissecting microscope . Stool samples were delivered to county laboratories daily for analysis . A person was classified as infected if any test was positive for S . japonicum . Because night soil use by one household may impact the S . japonicum infection risk of other village residents , we wanted to include both village- and household-level night soil use in our statistical models . However , using a village-level measure that is an average of all household-level measures results in a given individual’s household night soil use appearing in the model twice—once as part of the average , and again as an individual-level variable . This endogeniety leads to theoretical challenges . Using an outcomes based causal framework , we typically define the effect of an exposure , X , on an outcome as the difference in mean outcomes when the population is uniformly at X = a vs . X = b ( where a and b are any combination of exposure levels , only one of which is observable ) for some target population with a specific distribution of confounders [26] . But the overlap of household and village-level variables makes it difficult to evaluate changes in one without changes in the other . To avoid this problem , we defined village-level night soil use as the average amount of night soil applied by all households in the village excluding the index household . This allows for theoretical considerations of the effect of changes in village-level night soil use holding an individual’s household-level night soil use constant and vice versa . We employed a two-step approach to evaluate the relationship between night soil use and human infection . First , we examined the association between village-level night soil use and human S . japonicum infection using a multi-level , fixed-effect logistic regression model , modeling village-level night soil use as a categorical variable to allow for non-linear relationships between the explanatory variable and the outcome . Tests for trend were conducted by treating the categorical variable as ordinal . Models were run separately for each study year . Potential confounding variables were selected a priori based on prior evidence and the plausibility of a relationship with both the outcome of interest and night soil use . Models adjusted for participant age , county , household night soil use , bovine ownership , village bovine density ( the mean number of bovines per household ) , area of land cultivated by the household in the past year , agricultural intensity in the village ( mean area cultivated in the past year per household ) , household SES score and village SES ( mean household SES score ) . Village SES , bovine density and agricultural intensity were estimated separately for each household , excluding the index household , as described above . Occupation was not included as a potential confounder , as greater than 95% of adults in the region reported their occupation as farmer in 2007 [17] . Given the large set of potential confounders and the uncertainties in variable selection , we defined a reduced set of variables that we strongly suspected to be confounders ( age , sex , county and household-level night soil use ) and ran all models twice , once using the reduced set and once using the full set of confounders . We accounted for correlation within villages using generalized estimating equations and exchangeable working correlation , calculating robust variance estimates [27] . Second , we evaluated the potential impact of interventions to reduce the use of night soil on schistosomiasis . To do this , we estimated a parameter akin to attributable risk using a population intervention model approach [28 , 29] that we call the intervention attributable ratio . The intervention attributable ratio is defined as E ( YA ) /E ( Y ) , where E ( YA ) is the expected prevalence of infection in the study population if night soil use were eliminated , and E ( Y ) is the observed prevalence of infections at the observed levels of night soil use . For the purpose of the model , we assume that all infections were acquired recently ( a reasonable assumption given the frequency of mass and targeted chemotherapy in this population ) , that our measure of night soil use is relevant to the infection risk period , and that the observed statistical associations between night soil use and schistosomiasis prevalence represent a causal relationship . We explore the strengths and limitations of these assumptions in the discussion . G-computation was used [28–30] . We fit a fixed-effect logistic regression model assuming an independent correlation structure to allow for a population-level prevalence estimate . Based on the results of the first analysis , we included the limited set of potential confounders , modeled each year separately and modeled village night soil use as a continuous variable . This model was used to calculate infection probabilities for each individual surveyed when household and village night soil use were reduced to zero . Inference was estimated by bootstrapping: the population was sampled with replacement by village to obtain a 36 village population , the model was re-fit , E ( Y ) was estimated as the observed infection prevalence in the resampled population and E ( YA ) was estimated as the predicted infection prevalence when household and village night soil use were set to zero , as above , repeating this procedure 1 , 000 times . The 2 . 5th and 97 . 5th percentile values used to estimate the 95% confidence interval . All analyses were conducted using Stata version 12 ( College Station , TX ) .
Human waste was used as an agricultural fertilizer in 56% and 46% of households surveyed in 2007 and 2010 , respectively . Night soil was applied to all major summer and winter crops and by households across SES categories ( Table 1 ) . In 2007 , 99% of households that planted one or more crops used chemical fertilizers . Night soil was usually gathered from sources within the household: in 2007 , 8% of households with improved sanitation and 7% of households without improved sanitation reported using night soil from other households . Night soil was collected from all toilet types . Of those that used night soil in 2010 , 91% of households with an anaerobic biogas digester , 94% with a triple compartment septic tank and 96% with an unimproved toilet reported removing waste from their toilet system to use as an agricultural fertilizer . The mean quantity of night soil applied per household declined from 67 buckets per household in 2007 to 32 buckets per household in 2010 ( Table 2 ) . Access to improved sanitation increased over the study period , but most night soil applied came from unimproved sources: 79% and 72% of all night soil was applied by households without improved sanitation in 2007 and 2010 , respectively . We tested 2 , 005 people in 36 villages for S . japonicum infection in 2007 and 1 , 365 people in the same villages in 2010 ( 69% and 60% of the eligible populations identified in the census in 2007 and 2010 , respectively ) . Participation in the infection surveys was higher among older populations and , particularly in 2010 , among residents in County 2 ( S1 Table ) . S . japonicum infection prevalence was 8 . 4% in 2007 and 7 . 4% in 2010 . Among infected individuals , infection intensity was low . The mean eggs per gram of stool among infected was 24 in 2007 and 18 in 2010 . S . japonicum infections were detected in 28 villages in 2007 and 20 villages in 2010 . Categorical analysis . Village night soil use was associated with increased prevalence of human S . japonicum infection in 2007 but not in 2010 . Table 3 shows the relationship between night soil use and S . japonicum infection when village night soil use was categorized into quartiles . In 2007 , individuals in areas where village night soil use was in the highest quartile were 10 . 8 times more likely to be infected with S . japonicum compared to the lowest quartile , adjusting for the full set of potential confounders ( 95% CI 3 . 25 , 35 . 87 ) . Infection prevalence increased with village night soil use in 2007 ( test for trend , adjusting for the full set of potential confounders , p = 0 . 009 ) . Village night soil use was not associated with S . japonicum infection in 2010 . Household-level night soil use was not associated with S . japonicum infection regardless of year and the set of confounders included in the model . Estimates and inference were similar using the limited vs . full set of confounders . Estimates and inference were also similar when we limited the analyses to the 1 , 577 individuals in 2007 and 747 individuals in 2010 with complete infection testing data ( 3 stool samples examined using the miracidium hatching test and 3 Kato-Katz slides examined ) ( S2 Table ) . Intervention attributable ratio . The estimated intervention attributable ratio was 0 . 51 ( 95% CI 0 . 29 , 0 . 88 ) in 2007 and 1 . 44 ( 95% CI 0 . 85 , 2 . 01 ) in 2010 . If night soil use were eliminated , we estimate infection prevalence would be reduced by 49% of the observed infection prevalence in 2007 . No reductions in infection prevalence were estimated if the same intervention were to have occurred in 2010 . In 2007 , the amount of night soil applied in a village from both improved and unimproved sources applied was positively associated with S . japonicum infection ( Fig . 1 ) . The prevalence of S . japonicum infection was higher in areas where night soil from improved or unimproved sources were in the highest vs . lowest quartiles ( unimproved sources OR 3 . 56 , 95% CI: 1 . 19 , 10 . 63; improved sources OR 4 . 87 , 95% CI 1 . 79 , 13 . 26 ) . The positive trend was significant across quartiles of night soil from improved sources ( test for trend , p = 0 . 008 ) but not for unimproved sources ( test for trend , p = 0 . 118 ) . In 2010 , there was no association between S . japonicum infection and village-level night soil from improved or unimproved sources .
While access to antihelminthic treatment remains a major global challenge [39] , programs such as China’s national schistosomiasis control program and the multinational Schistosomiasis Control Initiative have greatly reduced morbidity and infection intensities in the regions where these interventions have been implemented [13 , 40] . In Sichuan , for example , a survey of 20 villages conducted in 2000 found S . japonicum infection prevalence was 29% and mean infection intensity , 26 EGP , with village prevalence and intensity as high as 73% and 104 EPG [9] . Today , schistosomiasis transmission has been interrupted in 41 counties in Sichuan [41] . The study sites described here provide an example of the schistosomiasis transmission pockets that remain—areas where few individuals are infected and those that are infected are shedding few parasites . We suspect the dynamics of transmission in such areas differ from areas with high infection prevalence and intensity [42] . Despite the low infection prevalence and intensities , schistosomiasis in our study sites has been remarkably robust to efforts to eliminate schistosomiasis infections . Following the reemergence of schistosomiasis in the region , the study area has been the focus of control efforts including snail control through the application of molluscicides to snail habitat ( applied up to 3 times annually from 2007 to 2010 ) , health education , improved sanitation construction and both mass and targeted chemotherapy in addition to the treatment provided following positive infection testing in our infection surveys . These control activities have yielded only modest declines in infection prevalence and intensity—infection prevalence declined from 8 . 4% in our study population in 2007 to 7 . 4% in 2010 and infection intensity from 2 . 1 to 1 . 3 EPG—a decline far less dramatic than observed following the initiation of control programs in highly endemic areas . In these areas of disease emergence and persistence , interventions that can permanently interrupt the transmission cycle are needed . Improvements to sanitation have been one of the strategies adopted by the Chinese schistosomiasis control program , and the World Health Organization and others have recognized the importance of improvements to water and sanitation in the control of human helminthiases [10 , 14 , 43] . Improvements to sanitation infrastructure can reduce the potential of a schistosome to complete its life cycle by preventing human fecal waste from contaminating snail habitat and thereby preventing schistosome infections in snails . The Chinese government has subsidized sanitation improvements in schistosomiasis endemic areas and the increase in access can be observed in the study sites: improved sanitation access increased from 16 . 5% of households in 2007 to 26 . 7% in 2010 . However , the extraction of human waste from stool pits to apply as an agricultural fertilizer may compromise sanitation investments . In our study of 36 villages in rural China , we found that human waste is commonly used as an agricultural fertilizer . This practice was strongly associated with schistosomiasis prevalence in 2007 but not in 2010 . The inconsistent relationship may be due to the stochastic nature of schistosomiasis transmission on the margins of disease elimination . Our findings suggest a possible link between schistosomiasis and the use of human waste as an agricultural fertilizer . Further evaluation of the relationship between human helminthiases and night soil use is warranted , particularly in areas where helminthiases persists despite disease control programs . | Many people use human waste as an agricultural fertilizer , often called “night soil . ” If the waste is not properly treated , the use of night soil may promote the spread of infectious diseases . We suspected that night soil use may facilitate the spread of the water-borne disease , schistosomiasis , as some schistosomiasis eggs can survive in the environment for weeks . We conducted a study in 36 villages in rural China in order to see if the amount of night soil used in a village was associated with schistosomiasis . The study was conducted in an area where schistosomiasis reemerged and persisted despite an aggressive disease control program . We found half of households reported using night soil—it was used on all major crops and by people across the socio-economic spectrum . We also found that night soil use was strongly associated with schistosomiasis infection in 2007 , but not in 2010 . Our findings show the use of human waste as an agricultural fertilizer is common in our study region and may increase schistosomiasis infections . The extent to which night soil is used and risks of this practice should be evaluated as part of disease control programs targeting schistosomiasis and other human helminthiases . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [] | 2015 | Associations between Schistosomiasis and the Use of Human Waste as an Agricultural Fertilizer in China |
The disease cryptococcosis , caused by the fungus Cryptococcus neoformans , is acquired directly from environmental exposure rather than transmitted person-to-person . One explanation for the pathogenicity of this species is that interactions with environmental predators select for virulence . However , co-incubation of C . neoformans with amoeba can cause a “switch” from the normal yeast morphology to a pseudohyphal form , enabling fungi to survive exposure to amoeba , yet conversely reducing virulence in mammalian models of cryptococcosis . Like other human pathogenic fungi , C . neoformans is capable of microevolutionary changes that influence the biology of the organism and outcome of the host-pathogen interaction . A yeast-pseudohyphal phenotypic switch also happens under in vitro conditions . Here , we demonstrate that this morphological switch , rather than being under epigenetic control , is controlled by DNA mutation since all pseudohyphal strains bear mutations within genes encoding components of the RAM pathway . High rates of isolation of pseudohyphal strains can be explained by the physical size of RAM pathway genes and a hypermutator phenotype of the strain used in phenotypic switching studies . Reversion to wild type yeast morphology in vitro or within a mammalian host can occur through different mechanisms , with one being counter-acting mutations . Infection of mice with RAM mutants reveals several outcomes: clearance of the infection , asymptomatic maintenance of the strains , or reversion to wild type forms and progression of disease . These findings demonstrate a key role of mutation events in microevolution to modulate the ability of a fungal pathogen to cause disease .
Pathogens across all major microbial groups – viruses , bacteria , fungi and protists – have representative species that owe their success to rapid change during infection or within a population . Microevolution is thus essential to pathogenesis , yet due to its stochastic nature it can be difficult to study and the underlying mechanisms challenging to elucidate . Cryptococcus neoformans is a fungal pathogen that is acquired directly from the environment through inhalation of desiccated yeast cells or the sexual basidiospores . The fungus is found world wide and it causes disease predominantly in immunocompromised individuals , especially AIDS patients [1] , [2] . The global mortality rate is estimated at 624 , 000 per annum [3] . The closely-related species C . gattii causes disease mostly in healthy individuals , and is responsible for an ongoing and expanding outbreak of cryptococcosis in the Pacific Northwest of Canada and the United States [4] , [5] , [6] . Both Cryptococcus species are extensively studied , have a suite of experimental resources , and serve as general models for understanding pathogenesis and its evolution in pathogenic eukaryotes [7] , [8] . Cryptococcus species undergo microevolution in vitro , within animal models , and during the course of disease in humans [9] , [10] , [11] , [12] . A current hypothesis is that the Cryptococcus species are pre-selected for virulence within mammalian animals because of interactions with predatory microbes like amoeba or nematodes [13] , [14] , [15] , [16] . Evidence for this comes from studies on interactions with non-mammalian hosts . C . neoformans has been co-isolated with three different Acanthamoeba species [15] , [17] , [18] and these amoebae can take up the fungus by phagocytosis [19] , [20] . Genes that are essential for mammalian virulence are also required for virulence in non-mammalian models [21] . Furthermore , screens of insertional mutants of the fungus with the nematode Caenorhabditis elegans identified fungal genes required for nematode viability: deletion of these genes also reduces virulence in mouse models of cryptococcosis [22] , [23] . Additional support for the hypothesis comes from passage of C . neoformans through a slime mold host , Dictyostelium discoideum , since this produces strains with increased virulence in mice [24] . However , one caveat is that Acanthamoeba species ingest and kill C . neoformans . Surviving subpopulations can be isolated , including one common class that has pseudohyphal cells rather than the normal yeast shape [17] . The pseudohyphal strains were avirulent in animal models [17] , [25] , [26] . Strikingly , the pseudohyphal phenotype is not always stable . One of the eight pseudohyphal strains originally isolated , after wild type strains were exposed to amoeba , was inoculated into mice and it demonstrated wild type virulence [17] . Closer examination revealed , to quote directly , that “a high percentage of the cells in the inoculum of this isolate had reverted to the encapsulated yeast form” [17] , a description of an unstable trait governed by epigenetics or microevolution . In repeat experiments , including use of different wild type strains of C . neoformans and a different species of Acanthamoeba , pseudohyphal isolates were again obtained: the phenotype exhibited instability in some , but not all , strain backgrounds [15] , [25] . More recently , a similar pseudohyphal morphology was reported from in vitro experiments , and that it too could revert back to wild type at a high frequency ( e . g . an average of 1 revertant per 1600 colonies ) [27] . This high frequency of phenotypic change is referred to as phenotypic switching . The basis for this rapid evolution in C . neoformans remained unknown . The pseudohyphal morphology of strains from amoeba or phenotypic switching appear similar to those of C . neoformans strains with loss-of-function mutations in the RAM/MOR pathway of genes [28] . This pathway is conserved in eukaryotes and characterized primarily in Saccharomyces cerevisiae where the abbreviation is from Regulation of Ace2p activity and cellular Morphogenesis [29] , [30] . In Schizosaccharomyces pombe the pathway is known as MOR , for the Morphogenesis-related NDR kinase network . The pathway comprises six components , centered around the action of two protein kinases Cbk1 and Kic1 and the accessory proteins Sog2 , Hym1 and Mob2 . The large size and physical interaction of Tao3 with Cbk1 and Kic1 suggests that Tao3 may act as a scaffold protein [29] , [31] . Within the fungi , the pathway can have dramatically different effects on cell type , for instance promoting cell polarity in the ascomycete yeast S . cerevisiae whereas inhibiting it in the basidiomycete yeast C . neoformans [28] , [29] . It is required for virulence in plant and human pathogenic fungi [32] , [33] , [34] , [35] , and polymorphisms in KIC1 recently emerged from genome wide association studies between clinical and non-clinical isolates of S . cerevisiae [36] . The role of the RAM pathway in pathogenesis has been most thoroughly analyzed in Candida albicans , in which components mediate cell separation and polarity , and thus mutations block filamentation , impair biofilm formation and surface adhesion , and reduce virulence [34] , [37] , [38] , [39] , [40] , [41] , [42] . Mutation of the RAM pathway in C . neoformans causes a pseudohyphal morphology and other phenotypic changes [28] , although effects on virulence had not been tested . We hypothesized that the RAM pathway is affected in strains that “switch” morphology either in vitro or upon exposure to amoeba , and that the analysis of this process would provide insight into the mechanism of microevolution . In this study , we identify mutations within a RAM pathway gene in original pseudohyphal mutants derived from amoeba and phenotypic switching . We recapitulate the isolation of pseudohyphal strains by both means , and find that these strains all have mutations in the RAM pathway . A strain derived from amoeba with a point mutation in the MOB2 gene was unable to cause disease in a murine model , unless during the course of infection the mutation reverted to wild type . Switching back to yeast morphology relies on a multifactor system for microevolution that is also driven by DNA mutations . These findings demonstrate that DNA mutation contributes to fungal pathogenesis .
The C . neoformans species consists of two varieties , var . grubii ( serotype A ) and var . neoformans ( serotype D ) . Pseudohyphal strains have been isolated from both varieties ( table S1 ) . Three isolates of C . neoformans var . grubii from the 1970s that originated after exposure to amoeba ( strains C , D and E; ATCC 42343–42345 [17] ) and one C . neoformans var . neoformans isolate from the 1990s that originated from a phenotypic switching study ( strain F7 [27] ) were compared to defined deletion strains of the RAM pathway genes . The strains had similar pseudohyphal cell morphologies distinguishing them from wild type yeast cells ( Fig . 1A ) . Most strains showed decreased growth at mammalian body temperature ( Fig . 1B ) . Consistently , all were highly sensitive to the immunosuppressive chemical FK506 ( Fig . 1B ) . This drug inhibits the calcineurin pathway , the impairment of which is synergistically lethal with RAM pathway mutation in C . neoformans [28] . The similarity in phenotypes between the strains isolated from amoeba and phenotypic switching with defined deletion strains suggests the same genes or pathways are affected , and that this could be the RAM pathway . We hypothesized that the historical pseudohyphal strains represent loss of the RAM pathway . The nature of the deficiency was sought in the strains that had been isolated from amoeba and the phenotypic switched isolate . Epigenetic or microevolutionary changes can arise from a suite of different causes . As a first approach towards gene identification , constructs containing wild type copies of four of the genes in the pathway ( MOB2 , CBK1 , KIC1 and SOG2 ) were introduced into strains C , D , E and F7 with the endeavor to identify the affected gene by expressing additional copies . None of the four genes restored growth to the wild type yeast form , whereas these constructs complemented deletion strains . A fifth RAM pathway gene , HYM1 , has not emerged as a RAM component in C . neoformans from random mutant screens , and has thus far eluded gene replacement experiments . The sixth component of the pathway , TAO3 , is large and it was a challenge to generate a vector for this gene for transformation experiments . However , the lack of evidence for a direct role by the other members of the RAM pathway provoked closer examination of the TAO3 gene in the pseudohyphal strains . The TAO3 gene was sequenced from the four historical pseudohyphal strains . The three strains isolated after exposure to amoeba ( strains C , D and E ) all contained an identical predicted g-t base pair substitution , that results in a codon for glutamic acid being substituted for a premature stop codon ( Fig . 2A; dataset S1 ) . The wild type strain for the three amoeba derived isolates is unknown , but is likely to be strain G ( ATCC 42437 ) based on co-deposition of this isolate with C , D and E to the American Type Culture Collection . The TAO3 gene was sequenced from strain G to confirm that the stop codon is not present . The TAO3 gene from the strains C , D , E and G strains , with the exception of the stop codon in three isolates , was identical in sequence to the gene from C . n . var . grubii strain H99 sequenced by the Broad Institute ( CNAG_03622 ) . Sequence analysis of strain F7 , the phenotypic switched strain , has an allele of TAO3 bearing a predicted a-t bp substitution that also causes a premature stop codon ( Lys-Stop , Fig . 2B and dataset S1 ) . The wild type parent for F7 is strain ATCC 24067A , and TAO3 was amplified and sequenced from this strain . The wild type strain does not contain the stop codon . Thus , the historical pseudohyphal strains derived from exposure to amoeba or phenotypic switching are tao3 mutants . While the stop codons within the TAO3 gene are consistent with impaired RAM function , we aimed to demonstrate that these mutations cause the pseudohyphal phenotype . A Mendelian genetic segregation approach was first considered . However , strain F7 and the three amoeba-derived strains are infertile and thus genetic linkage tests were not feasible . Next , a construct was generated by overlap PCR that introduces a silent point mutation into the coding region adjacent to the Glu-stop codon , and at the same time introduces a new BglII restriction enzyme site ( Fig . 3A , B ) . The rationale behind this approach was that the chance of two identical nucleotide changes occurring in independent strains is highly unlikely , and this construct could be used to distinguish reconstituted strains from reverted strains . The TAO3BglII construct was transformed into strain D using a biolistic apparatus that facilitates homologous gene replacement events . After transformation the cells were plated on FK506-containing media . Strains were isolated with the wild type yeast morphology , and when the TAO3 fragment was amplified and the PCR product digested with BglII , a subset of strains now contained a novel BglII restriction enzyme site ( Fig . 3C , D ) . This indicates successful gene targeting and reconstitution to a wild type copy of the gene , and that the pseudohyphal phenotype is due to mutation of TAO3 . Additional pseudohyphal strains were sought in order to explore the basis of this trait and whether or not it was specific to the TAO3 gene . Two C . n . var . grubii candidate TAO3 mutant strains isolated in a previous study [28] , but not further characterized , were examined through PCR and DNA sequencing ( Fig . 2; dataset S1 ) . One has a T-DNA insertion in the promoter of the TAO3 gene , and the other has a deletion of 7 bp ( gcgtagc ) . Two pseudohyphal T-DNA mutants were isolated during other experiments . One contains an insertion in the KIC1 gene and the other in TAO3 . In addition , strains with complete deletion of TAO3 were generated in three backgrounds ( C . n . var . grubii KN99α and strain G , and C . n . var . neoformans ATCC 24067A ) . Overlap PCR products were created to replace the TAO3 ORF with the nourseothricin acetyltransferase cassette via biolistic transformation and homologous recombination . The tao3 point mutant strains have the same phenotype as complete deletion or T-DNA insertion alleles , consistent with complete loss-of-function . We screened and isolated 13 spontaneous pseudohyphal strains in the C . n . var . neoformans ATCC 24067A strain , six using a UV-induction method [27] and seven by plating colonies for random mutation events ( table S1 ) . The strains were transformed with the wild type copies of CBK1 , KIC1 , MOB2 or SOG2 to test for complementation , and/or the TAO3 gene from these strains was sequenced . Eight contained changes in the TAO3 sequence , as illustrated in Fig . 2 . Complementation experiments implicated mutations in the KIC1 , MOB2 , SOG2 and CBK1 genes in the remaining five , and these mutations were identified by sequencing those genes ( Fig . 2 and dataset S1 ) . The SOG2 gene in one had a mutation in which a stretch of 16 bp ( tgcacaacgcaactct ) in the fifth exon was duplicated and inserted adjacent to the original sequence . This insertion would result in a frame shift mutation . The cbk1 mutant bears an a-g mutation in the 3′ g splicing site of an intron . Likewise , the two kic1 mutants are both bp substitutions within splice sites . MOB2 was sequenced from the mob2 mutant , and has a bp deletion that will cause a translational frameshift . In summarizing , TAO3 is most often mutated in spontaneous pseudohyphal strains but other genes in the RAM pathway can also be affected . The provenance of the C , D and E strains with pseudohyphal morphology that were isolated in the 1970s by exposing C . neoformans to amoeba is not well documented . Isolating RAM mutants using amoeba was tested . The original Acanthamoeba polyphaga strain from mouse feces and the A . palestinensis strain from pigeon guano , used previously to isolate pseudohyphal colonies , were not saved in a culture collection . The ATCC 30234 strain of A . castellanii was co-isolated with C . neoformans by Aldo Castellani [18] , and is commonly used to study the interactions of amoeba with other pathogenic microbes . First , C . neoformans wild type and RAM mutants were exposed to this amoeba in the expectation that a similar interaction would occur as had been described in the 1970s . The outcome of the fungus-amoeba interaction depended on strain background and medium type . In some combinations , such as that illustrated in Fig . 4 , the amoeba consumed the wild type strain whereas the RAM pathway mutants were resistant . Second , the wild type strains G ( C . n . var . grubii ) and ATCC 24067A ( C . n . var . neoformans ) were exposed to the amoeba on proteose peptone agar , and examined 2–3 weeks later for the presence of surviving colonies . A subset was of colonies comprised of pseudohyphal cells . The nature of the mutation within the RAM pathway for those strains was sought through complementation experiments and gene sequencing . Mutations were identified again in the TAO3 gene , as well as MOB2 ( Fig . 2 , dataset S1 ) . These findings thus extend the diversity of the C . neoformans/amoeba interaction to include both C . neoformans varieties and a third Acanthamoeba species . In doing so , the results provide further evidence that the RAM pathway is integral to pseudohyphal morphology , and the isolation of these mutants reflects a common underlying ability across divergent C . neoformans strains . RAM pathway mutants are sensitive to FK506 , providing a simple means to select for strains that revert to wild type . Revertants were sought from tao3 mutant strains D and F7 . A high rate of spontaneous resistance to FK506 was observed for strain F7 , but not all of these FK506 resistant strains had reverted to the wild type yeast cell morphology . Rather , the strains had acquired FK506 resistance through some other means , possibly through mutation of the gene encoding the FKBP12 protein to which FK506 physically binds [43] . The frequency of reversion and FK506 resistance was noticeably lower in strain D than F7 . Part of the TAO3 gene was sequenced from >30 revertants derived from F7 with the wild type yeast morphology: four types of changes were identified ( Fig . 5A ) . In the first , an a-t bp mutation had restored the original lysine codon . In the second , an a-c bp change replaced the stop codon with a glutamine codon . In the third , the adjacent nucleotide had mutated ( a-t ) to change the stop codon to a leucine codon . In the fourth , the original stop codon was still present . In addition , another reversion event was also common , leading to a suppression of the RAM phenotype and partial return to the wild type phenotype . These strains were distinctive because of the yellow-colored colonies , with cells exhibiting a lemon-shape and inefficient cell separation . Sequence analysis revealed that the stop codon mutation was still present in TAO3 in these types of strains . In contrast to the revertants isolated from strain F7 , when the region containing the stop codon in TAO3 was sequenced from 24 revertants of strain D , all 24 still contained the stop codon . Thus , mutation within TAO3 allows reversion back to wild type , although another mechanism ( s ) can account for some reversion events . The pseudohyphal strain DM03 contains a 16 bp duplicated region within the SOG2 gene , and as such is a different type of mutation compared to the bp substitutions in the TAO3 gene described above . After selection on FK506 for wild type colonies from strain DM03 , three reversion types were observed ( Fig . 5B ) . In one , the duplicated piece of DNA was excised . In a second case , 4 bp were deleted downstream of the insertion event , returning the gene to the correct reading frame and producing an allele encoding 45 different amino acid residues . In the third , the original mutation was still present . Thus , excision of a duplication region or insertions or deletions correcting the open reading frame provide yet additional mechanisms to revert RAM mutants . Northern blot analysis was use to examine changes in transcript levels or sizes in RAM pathway genes in response to mutation or reversion in a selection of var . neoformans and var . grubii strains ( Fig . S1 ) . The KIC1 and TAO3 transcripts were below detectable levels . CBK1 and MOB2 have overall constitutive transcript levels . HYM1 and SOG2 showed variation in transcript levels , although there was no perfect correlation between loss of a RAM gene and upregulation . Of note , the mob2 mutant DM09 used in the virulence analysis described below exhibited altered transcript sizes , consistent with a mutation in a predicted intron splice site ( Fig . S1 ) . If mutation is the primary source of pseudohyphal strains , it was surprising that they arise at such a high frequency . This together with the observation of high rates of spontaneous resistance to FK506 in the strains in the ATCC 24067A strain background used in phenotypic switching experiments [27] , [44] , led us to test the mutation rate in this strain . ATCC 24067A is derived by laboratory passage from strain ATCC 24067 [45] . 20 separate cultures of strains ATCC 24067 and ATCC 24067A were plated onto medium to select for spontaneous uracil auxotrophy . The mutation rate for ATCC 24067 was 2 . 66 per 1×108 ( 95% confidence interval 1 . 78–4 . 75 ) . In contrast , in ATCC 24067A the rate was 67 . 88 per 1×108 ( 95% CI 57 . 06–79 . 38 ) . Thus , ATCC 24067A has greater than 25 fold higher mutation rate that its progenitor parent ATCC 24067 . To ensure that the uracil auxotrophs were due to mutations in the same gene , the URA5 gene enoding orotate phosphoribosyltransferase was amplified from 15 5-FOA resistant strains derived from separate starting colonies , and sequenced ( Fig . 6A; dataset S2 ) . 5-FOA resistance in fungi can result from mutation of either URA3 or URA5 homologs; prior studies suggest that URA5 is the main target in C . neoformans [46] . All 30 strains had mutations in URA5 . Comparing the mutation profiles for ATCC 24067 and ATCC 24067A revealed similarities , e . g . three in each strain background had the same t-c mutation . A main difference was in mutations involving more than one base pair . Two indels for ATCC 24067A were single bp . In contrast , for ATCC 24067 three alleles have large insertions or rearrangements identified by altered or absent PCR products ( data not shown ) . Another three uracil auxotrophs have insertions or deletions between 2 and 18 bp ( dataset S2 ) . We interpret this to imply a higher level of bp substitutions in the ATCC 24067A strain . Comparison of the ATCC 24067 and ATCC 24067A strains under stress conditions also revealed altered response to stress agents , especially oxidative stress agents and ethidium bromide ( Fig . 6B ) . We hypothesize that during laboratory passage ATCC 24067A acquired a mutation in a DNA repair pathway gene . The RAM pathway itself could potentially influence mutation rates , thereby enhancing the rate of reversion . Mutation rates were compared between 15 cultures each of ATCC 24067A and the tao3 point mutant strain F7 . Uracil auxotrophs were isolated at a rate of 42 . 00 per 1×108 ( CI 34 . 30–53 . 11 ) for the ATCC 24067A strain , and 15 . 74 per 1×108 ( CI 13 . 37–18 . 93 ) for the F7 strain . These results suggest that there is no increase in mutation rate in the RAM mutants , since the wild type strain had 2 . 6 times higher frequency for isolation of uracil mutants compared to the F7 strain . A caveat in this comparison is the challenge of accurate quantification of viable cells for RAM pathway mutants . Additional evidence was sought that pseudohyphal strains and reversions are due to mutation events . A “yellow” suppressor or partially-reverted strain was examined by Mendelian genetic analysis . The strain was isolated in the tao3::NAT deletion strain background ( strain AI235 ) by selection on FK506 . The AI235ya strain was crossed to a wild type of opposite mating type . 20 progeny were obtained: three pseudohyphal NATR FK506S , ten wild type yeast NATS FK506R , and seven “yellow” NATR FK506R ( Fig . 7 ) . These results show that the suppression phenotype is meiotically stable , and the progeny ratio is consistent with its segregation as a single genetic locus . This finding further illustrates that mutation leads to some types of reversion events that are yet to be defined . An alternative hypothesis for the high frequency of isolation of RAM mutants would be if the pathway or parts of it were hot spots for mutations . To test this , aliquots from the identical 20 cultures of strain ATCC 24067A used to isolate uracil auxotrophs were plated onto YPD . Six RAM mutants were identified from ∼143 , 000 colonies . These six came from four starting cultures . In two examples , pairs on strains were isolated from the same plate . Characterization of the pairs ( AI273–AI274 and AI275–AI276 ) revealed that they shared the identical mutation within TAO3 , reflecting an attached pair of pseudohyphal cells that were physically separated when spread on the plate . There are 225 codons in URA5 vs . 5750 codons combined for the six RAM pathway genes . Based on size alone we expected a 25 . 6 fold higher frequency of isolation of pseudohyphal strains compared to 5-FOA resistant strains , or 1 in every 57 , 546 colonies . The isolation of six mutants from 143 , 000 screened would be unlikely ( P<0 . 03; Poisson distribution ) . However , if the pairs of identical tao3 mutants are considered as one event , then four from 143 , 000 is not statistically significant ( P<0 . 24 ) . To circumvent bias due to cell separation in the RAM mutants , an alternative measure of mutation was taken . Uracil auxotrophs were isolated in the AI228 mutant , and the URA5 gene sequenced . AI228 has an a-g transition in an intron splice site in CBK1 . Strain AI228 ura#3 was isolated with a g-a transition in a highly-conserved glycine codon . We reasoned that the only way to revert the strain to wild type would be the perfect reversal of the mutation . 15 separate colonies were inoculated into liquid medium , cultured overnight , and from each 2 . 5×108 cells plated onto media supplemented with FK506 , to select for a mutation in CBK1 , and onto media without uracil , to select for a mutation in URA5 . No wild type yeasts were obtained . In contrast , nine uracil prototrophs were isolated , reflecting reversion in URA5 . These results indicate that at least one gene in the RAM pathway , CBK1 , is not a general hot-spot for mutation . If amoeba select for pseudohyphal strains of C . neoformans in the wild , then why are pseudohyphal strains not isolated on a regular basis ? We explored this question by testing isolation methods for C . neoformans and the fitness of the RAM pathway mutants under different growth conditions . First , we explored the ability of RAM pathway mutants to grow on medium that mimics an environmental substrate , pigeon guano , with which C . neoformans is associated in nature [47] . Both wild type and RAM mutants grew equally well on pigeon guano medium , suggesting that RAM mutants have equivalent growth as wild type on this substrate ( Fig . 8A ) . Second , RAM mutants were grown on bird seed agar . This medium is made from Guizotia abyssinica seed and is a standard medium for isolation of C . neoformans from environmental sources , aided by melanization of C . neoformans colonies [48] , [49] . In two serotype A genetic backgrounds , the RAM mutant strains were delayed in pigmentation and produced smaller colonies compared to wild type ( Fig . 8B ) . Melanin is a well-established virulence factor for C . neoformans . Another virulence trait is the biosynthesis of a polysaccharide capsule , which was found previously to be produced like wild type in RAM pathway mutants [28] . Third , phenotypes were explored under various stress conditions . No visual differences were observed between wild type and pseudohyphal strains growing on YPD at pH 4 . 5 or pH 8 , or on YPD supplemented with high levels of salt NaCl , detergent sodium dodecyl sulfate , antifungal flucoazole , or oxidative stress agents ( H2O2 and t-butyl hydroperoxide ) . One phenotype identified that did differ is altered colony integrity . The RAM pathway mutant cells were easily dispersed by washing ( Fig . 8C ) , possibly a detrimental trait leading to reduced protection of the cells within a structured colony in the wild . Mutation of the CBK1 homolog in the basidiomycete fungus Ustilago maydis causes sterility [33] . C . neoformans RAM mutants also have reduced fertility in crosses in which both parents bear mutations , since no filaments , basidia or basidiospores are produced in crosses on V8 juice or Murashige-Skoog medium ( Fig . 8D ) . Thus , explanations for the lack of pseudohyphal C . neoformans isolated from the wild could include their discard due to reduced melanization , reduced growth on bird seed agar or at elevated temperatures , and unconventional cell morphology . Alternatively , while it is possible that RAM mutation confers benefits under some environmental conditions , under others the pseudohyphal strains are less fit thereby countering the advantages gained in avoiding predation by amoeba . The mechanism by which the RAM mutants of C . neoformans evade amoeba was explored using light microscopy , with confocal microscopy of GFP-expressing fungal strains and amoeba stained with FM4–64 used to confirm internalization of the fungal cells ( data not shown ) . These results indicated that one possible mechanism of action is less efficient phagocytosis of RAM pathway mutants by amoeba . The pseudohyphal strains , especially when in clusters of cells , are larger than the amoeba thereby forming a physical impediment to phagocytosis ( Fig . 4B ) . Consistent with this hypothesis , pseudohyphal cells are less frequently found inside amoeba . For instance , a mixture of wild type and tao3 mutants was exposed to amoeba . Only 3% of amoeba that harbored Cryptococcus had cells that were pseudohyphal ( n = 178 ) . In the cases in which pseudohyphal cells were present in amoeba , these were as single or two attached cells ( Fig . 4B ) . While a physical block may account in part for evasion of amoeba , other reasons for resistance , such as increased intracellular survival of pseudohyphal cells , cannot be excluded . Based on previous observations of the virulence of pseudohyphal C . neoformans and the role of the RAM pathway in virulence in other fungi , we hypothesized that our mutants would be attenuated or avirulent . Wild type strain G , the mob2 mutant strain DM09 that was isolated by exposing strain G to amoeba , and a complemented strain AI255 ( DM09+MOB2-NEO ) were used to test the role of the RAM pathway in virulence in wax moth larvae and mouse models . The mob2 mutant strain DM09 exhibits reduced growth at 37°C , which is predicted to influence virulence in mammalian hosts . To address virulence at a more permissive temperature , the three strains were inoculated into wax moth ( Galleria mellonella ) larvae and the larvae maintained at 30°C . 1×105 cells of wild type and complemented strains were used as inocula . Two inocula were used for DM09: 1×105 cell-clusters and one at 1/10 that concentration . Microscopic analysis of the cultures of the mob2 mutant and accompanying plating assays indicated that approximately 10 pseudohyphal cells formed the equivalent of one colony forming unit , due to the cell separation defect of the strain . The larvae inoculated with wild type or complemented strains started dying five days after inoculation , and 22 of 23 were dead by day nine ( Fig . 9 ) . In contrast , the larvae inoculated with the mob2 mutant strain survived longer , e . g . on day nine only two of the 21 larvae had died . The experiment was terminated when the surviving larvae , including the control group inoculated with PBS , formed cocoons . Log-rank statistical comparisons indicated that the differences in survival between wild type or complemented strains with the mob2 mutant were significant ( P<0 . 0001 ) . Thus , the RAM pathway controls fungal virulence in an insect model and reduced virulence is independent of temperature . Next , the three strains were tested in a mouse inhalation model of cryptococcosis . A subset of mice were sacrificed 24 and 96 h post-inoculation , and colony counts measured and lung tissue prepared for histology . The wild type proliferated during the three day interval , while the mob2 mutant strain maintained a level of ∼1×105 cfu per gram ( Fig . 10A ) . The colonies isolated from mice inoculated with the wild type were smooth and comprised of yeast cells . Colonies from mice inoculated with the mob2 mutant were all wrinkled and comprised of pseudohyphal cells . Yeast or pseudohyphal cells were evident in histological samples of the lungs at both 24 and 96 h from mice infected with the wild type and mob2 mutant , respectively ( Fig . 10B , C ) . These results show that C . neoformans pseudohyphal cells can penetrate the lung and survive at least four days . The remaining sets of inoculated mice were monitored daily for signs of cryptococcal disease . The mice infected with the wild type or complemented strains succumbed to disease and were sacrificed by day 26 , a time at which all of the mice inoculated with the mob2 mutant strain were alive and healthy . Interestingly , four of these mob2-inoculated mice developed symptoms of cryptococcosis and were sacrificed between days 38 and 49 post-inoculation ( Fig . 11A ) . Cells isolated from either the lungs or brains of these mice were the round yeast morphology like wild type strains ( Fig . 11C ) . The MOB2 gene was amplified from two colonies isolated from each organ . All sequences showed the original g nucleotide found in the wild type gene sequence ( Fig . 11C , dataset S3 ) . Histological examination of the lungs and brain of the diseased animals also revealed yeast cells , as well as extensive host tissue damage ( Fig . S2 ) . The interpretation of these data is that the mob2 mutation had reverted to wild type in the four animals during the course of infection . At day 70 , the remaining six mice inoculated with the mob2 mutant had no symptoms of disease , so were sacrificed and organ homogenates plated . Three mice had cleared the infection since they had no fungal cells present in either lung or brain tissue . One mouse ( number 5 ) had wild type morphology cells in the brain and lung ( Fig . 11B , C ) . Sequence of MOB2 from these cells showed the wild type gene sequence . Two mice ( numbers 7 and 10 ) had pseudohyphal strains only in the lung , and the brains were free of fungal cells ( Fig . 11B , C ) . When the MOB2 alleles in the pseudohyphal strains were examined by DNA amplification and sequencing , those strains contained the original splicing mutation in MOB2 ( Fig . 11C , dataset S3 ) . In summarizing the results from the mouse model , the mob2 mutant was attenuated for virulence ( Log-rank test P<0 . 0001 ) . However , the pseudohyphal cells can persist during infection , and reversion mutations occur stochastically over time to restore cell shape to the wild type and fully pathogenic form .
In this study we describe a new mechanism for microevolution in the human pathogenic fungus C . neoformans that controls the host range of the organism . Cryptococcus species have plastic genomes , with experiments showing changes in chromosome length over time , microevolution during human infection and in culture , and changes in chromosome numbers conferring azole drug resistance [9] , [10] , [45] , [50] , [51] . Microevolution modulates the polysaccharide capsule composition that surrounds the cell: this is the best-studied aspect about microevolution in the fungus [44] , [52] , [53] . The Cryptococcus genus is found in association with trees , soil , bird excreta and additional environments that are also homes to other microbial species [1] , [14] . One hypothesis is that selection for traits that defend against small predators , such as amoeba or nematodes , has led to species capable of causing disease in humans [13] , [14] , [15] , [16] . The evidence for amoeba-C . neoformans interactions date to over half a century ago with the work of Castellani , who showed that the species later named Acanthamoeba castellanii could kill C . neoformans cells [20] . In the 1970s , amoeba were again co-isolated with C . neoformans [15] , [17] . Subsequent studies found that a subset of C . neoformans colonies changed cellular morphology after exposure to Acanthamoeba species , from yeast to pseudohyphal cells , and that some isolates reverted rapidly to wild type yeast . Based on the more recent observation of phenotypic switching between pseudohyphal and yeast forms [27] and the discovery of a set of genes that produces a pseudohyphal phenotype when mutated [28] , we hypothesized that phenotypic switching involved the RAM pathway . Here we show that the RAM pathway is the integral component of “switching” in C . neoformans because it is mutated in pseudohyphal strains isolated from amoeba and spontaneously in culture . The largest gene in the pathway , TAO3 , most commonly bears point mutations leading to the introduction of premature stop codons . Some strains like F7 revert to wild type at a high frequency in vitro . Analysis of the mutated region in those strains reveals that there are multiple ways in which the strain can revert ( Fig . 5 ) . This may be through a bp substitution leading to a coding triplet being reformed . Alternatively , the stop codon may still be present . The basis for the latter situation is unknown . It could be mediated by stop codon read through , tRNA suppressor mutations , changes in downstream gene expression , or epigenetic phenomena . Mutation in another RAM pathway component may suppress the phenotype , as occurs with a specific residue in the Cbk1 kinase of S . cerevisiae to rescue mutations in other pathway components [54] , or be modified by interacting pathways such as seen for the Neurospora crassa cot-1 suppressors [55] , [56] . An alternative mechanism of reversion is illustrated by strain DM03 bearing a mutation in SOG2 . Excision of the duplicated region or deletion of another region downstream reverts Sog2 sequence back to wild type or in frame , respectively . Taken together , the conversion between yeast and pseudohyphal cells reported previously as a form of phenotypic switching is based on DNA mutations , rather than epigenetic changes . The effect of a defective RAM pathway on mammalian virulence was tested . The mob2 mutant strain used carries a bp substitution mutation within an intron splice site , and is phenotypically stable . Three strains were inoculated into mice . The wild type and complemented strains caused cryptococcal disease . In contrast , different outcomes were observed for mice infected with the mob2 mutant ( Fig . 11 ) . Four mice succumbed to disease , albeit weeks after those infected with the wild type and mutant had been sacrificed , and when their organs were harvested only yeast cells were recovered rather than the expected pseudohyphal cells that were used to inoculate the animals . The MOB2 gene was sequenced , and now had the wild type sequence ( Fig . 11C ) . Sacrifice of the remaining and asymptomatic animals at day 70 and characterization of fungal material in lung and brain tissue shows that three mice had cleared the infection , one carried wild type cells , and the other two still maintained the original mob2 mutant phenotype and genotype . These results are consistent with previous virulence studies using pseudohyphal strains selected by amoeba in which reversion back to wild type for some occurred at a high frequency within mouse models [17] , [25] , [26] . A third animal experiment has been performed with pseudohyphal strains , presumably also RAM mutants , in a rat tracheal model [27] . In this experiment , two of the four animals cleared the infection while the other two did not , potentially representing another case of reversion within the host . Morphological differentiation is important for Cryptococcus pathogenesis . Recently , a role has been assigned for a giant cell form during disease development [57] , [58] , while constitutive filamentation by altered regulation of the ZNF2 gene impairs virulence [59] . There are reports of pseudohyphal cells in histopathological samples of patients infected with C . neoformans [60] , [61] . One speculation is that pseudohyphal forms could allow escape of the fungus from mammalian or amoeba cells , in addition to escape of yeast cells from macrophages or A . castellanii by exocytosis [62] , [63] , [64] . How can “switching” occur at high frequency ? The formation of pseudohyphal strains relates to mutation rates in cells; switching also increases upon exposure to UV light [27] . Two factors influence frequency . The first is that the strain used in phenotypic switching studies has a 25 times higher mutation rate compared to a standard wild type strain . Second , because there are six genes in the RAM pathway there is a large amount of target DNA available for spontaneous mutation . Protein-coding sequence alone , the six genes comprise more than 17 kb of DNA , or ∼0 . 1% of the genome ( Fig . 2 ) . TAO3 , as the largest member ( 42% of total cumulative size ) , is therefore the most likely gene to be hit . The original historical isolates bear mutations in this gene . Of 14 unique mutations that we defined in the ATCC 24067A background , eight are in TAO3 , supporting this hypothesis . One challenge for quantitative analysis of the pseudohyphal strains is their defect in cell separation . Reversion to wild type has been estimated as high as 1 . 6×10−3 [27] , but this may be an over-estimate by up to an order of magnitude if the colony forming units were derived from attached cells . It is unknown why an organism like Cryptococcus , which is normally found in the environment , can cause disease in humans or other mammals . Further , the fungal behavior upon entering the human host that ends in life-threatening disease remains unclear . Three points are worth raising . First , people are exposed to C . neoformans during their lifetimes yet most do not develop disease . For instance , children in city environments where there is a high prevalence of pigeons as sources of C . neoformans become antigen positive at an early age [65] . Second , cases of re-activation from quiescent infections brought about by immunosuppression supports a hypothesis that the fungus enters a latent state [66] . Third , in comparing virulence of strains derived directly from the environment vs . a human host , many environmental strains do not cause disease in animal models although they persist in the lungs [67] , [68] , [69] , [70] . Collectively , a high rate of exposure to C . neoformans , to strains that may not necessarily be able to cause disease immediately , and the potential for latency provide the scenario in which microevolution of C . neoformans by DNA mutations could influence clinical outcomes . Multiple mechanisms can facilitate microevolution in pathogens . Among the fungi , mutations during infection lead to the emergence of antifungal drug resistance . However , these arise under conditions with a high fungal burden in the host , promoting the generation of strains from rare events . More broadly , there is evidence from diverse microbes that DNA is mutated to generate phenotypic variation . In the protist Trypanosoma brucei , mutation is required for evasion of the host immune response whereby double stranded breaks are generated and then repaired to generate antigenic diversity via the VSG genes [71] . Adaptation via mutation is implicated in bacterial disease progression . For example , 20% of Pseudomonas aeruginosa isolates from cystic fibrosis patients are “hypermutators” compared with 0% of environmental isolates [72] . Nevertheless , this trend in bacteria is not universal , as correlations between clinical isolates of E . coli and higher mutation rates have been found in some , but not all , studies [73] , [74] . A very different system to develop variation is the low fidelity of the human immunodeficiency virus' reverse transcriptase: along with other factors this results in rapid genetic changes within the human host and reduces immunological recognition of the virus . In contrast , evidence for mutations affecting the pathogencity of C . neoformans are rare at present . Curiously , a C . neoformans mutant in the MSH201 gene predicted to function in mismatch repair has a competitive advantage in mouse lungs compared to control strains [75] . This research points towards future directions into investigating the contribution mutation and mutation rates play in the ability of C . neoformans and other pathogenic fungi to adapt and cause disease . Specific directions are to assess mutation rates within the host and test if mutations that arise in the host result in more virulent strains . Second is to test for correlations between clinical and environmental isolates and rates of mutation . A third direction is to explore what role , if any , transcriptional , translational or epigenetic regulation of the RAM pathway plays in the interaction of C . neoformans with amoeba and mammalian cells .
Cryptococcus neoformans wild type strains used were KN99α ( var . grubii ) , G ( var . grubii , ATCC 42347 ) , ATCC 24067 , ATCC 24067A ( var . neoformans ) , and JEC21 ( var . neoformans ) . ATCC 24067A is derived from laboratory passage of ATCC 24067 . Historical pseudohyphal strains were F7 ( var . neoformans ) , and C , D , and E ( var . grubii; ATCC 42343-5 ) . Strains were kindly provided by Dr . Joseph Heitman and Dr . Bettina Fries . C . neoformans strains were cultured on yeast extract-peptone dextrose ( YPD ) ±2% agar medium , and stored as glycerol stocks at −80°C . The Acanthamoeba castellanii strain was obtained from the American Type Culture Collection ( ATCC 30234 ) and maintained according to ATCC instructions and stored at 4°C [76] . To isolate new pseudohyphal strains , ATCC 24067A was grown in overnight YPD cultures , then spread on YPD plates . A subset of plates were subject to a low dose of UV light in a UV transilluminator . Colonies were screened by eye for those with a dry appearance , which were streaked to isolate single colonies . To isolate RAM mutants using the A . castellanii amoeba , C . neoformans strains ATCC 24067A and G were inoculated in a cross pattern on a selection of different agar medium types ( potato dextrose , proteose peptone , YPD , V8 juice , Murashige-Skoog and trypan blue ) , and a drop of amoeba placed at the intersection , following the original protocol [17] . T-DNA insertional mutants were generated as described previously and RAM pathway mutants isolated based on colony morphology [28] . Wild type revertants were selected from RAM pathway mutants by plating on YPD agar supplemented with FK506 ( 1 µg/ml ) . Pigeon guano medium was 10% w/v of unfiltered pigeon guano ( collected under the I-35 overpass of Southwest Blvd , Kansas City , MO ) that had been homogenized in a coffee grinder and autoclaved with 4% agar . Bird seed agar was prepared as described [48] . Crosses were set up on 5% V8 juice or Murashige-Skoog agar [77] . Strains used and generated during this study are listed in Table S1 . Genomic DNA was extracted using a CTAB buffer [78] from 50 ml overnight cultures of strains . The TAO3 gene was amplified by PCR with two primer sets for each variety: ALID0013–ALID0061 and ALID0014–ALID0060 for var . grubii strains , and ALID0127–ALID0128 and ALID0129–ALID0138 for var . neoformans strains . SOG2 was amplified with primers ALID0123–ALID0162 . MOB2 was amplified with primers DM062–DM063 . CBK1 was amplified with primers ALID0977–ALID0978 . Part of the KIC1 gene was amplified with primers ALID1681–ALID1682 The PCR products were sequenced with the primers used for amplification and additional internal primers . Primer sequences used for amplification are listed in Table S2 . Tests were performed on pseudohyphal strains using vectors that complement the deletion mutants of mob2 , cbk1 , kic1 , and sog2 . The first three plasmids were generated in a previous study on the RAM pathway [28] . The SOG2-NEO1 construct was generated by amplification of SOG2 from strain JEC21 with primers ALID0123–ALID0162 , cloning into TOPO pCR2 . 1 ( Invitrogen , Life Technologies , Grand Island , NY ) , and a SpeI-XbaI fragment subcloning into the XbaI site of pPZP-NEO11 . The four genes were in plasmids enabling their introduction into C . neoformans cells via Agrobacterium-mediated transformation [79] . Transformants were selected on YPD medium containing cefotaxime ( 200 µg/ml ) and either nourseothricin ( 100 µg/ml ) or neomycin ( 200 µg/ml ) . The TAO3 gene in strain D was reconstituted by homologous recombination . A construct with an engineered BglII restriction enzyme site was generated by overlap PCR using primers ALID0061–ALID0227 and ALID0111–ALID0228 , and introduced into strain D cells plated on YPD+1 M sorbitol by biolistic transformation with a PDS-1000/He Particle Delivery System ( Bio-Rad , Hercules , CA ) , using standard methods [80] . Cells were allowed to recover for 3 h and transferred to YPD+FK506 ( 1 µg/ml ) . The TAO3 gene was deleted in the KN99α , ATCC 24067A and G strains . For var . grubii strains , the 5′ and 3′ flanks were amplified from genomic DNA of strain KN99α using primers damp5-ALID0013 and damp6–damp7 , respectively . For var . neoformans , the 5′ and 3′ flanks were amplified from strain ATCC 24067A using primers damp1–damp2 and damp3–damp4 , respectively . Nourseothricin acetyltransferase ( NAT ) was amplified from plasmid pAI3 using primers ai006–ai290 [79] . The primers ALID0013-damp7 or damp1–damp4 were used for overlap PCR . The SOG2 gene was deleted in strain KN99α . Primers ALID0483–ALID0484 and DM036–DM043 were used to amplify the 5′ and 3′ flanks , and ALID0483–DM036 used for overlap PCR with these and the NAT cassette . The DNA molecules were transformed into C . neoformans cells using the biolistic apparatus , cells allowed to recover for 3 h , and transferred to YPD medium containing nourseothricin ( 100 µg/ml ) . Correct gene replacement was confirmed by PCR analysis and Southern blotting with [32P]-dCTP-labelled fragments of the genes . Isolation of spontaneous uracil auxotrophs was used to measure mutation frequency . For comparison between strains ATCC 24067 , which was acquired from the ATCC , and ATCC 24067A , strains were grown on yeast nitrogen base ( YNB ) medium , then 20 separate cultures of each strain established at 1×105 cells/ml in YPD medium . After overnight culture in a roller drum incubated at room temperature , 5×107 or 1×108 cells were plated onto YNB supplemented with uracil ( 20 mg/L ) and 5-fluoroorotic acid ( 5-FOA; 1 g/L ) medium . The resulting colony numbers were analyzed to determine the mutation rate ( Lea-Coulson method of the median ) with FALCOR software [81] . To ensure mutations targeted the same gene in both strains , the URA5 gene was amplified with primers ALID0375–ALID0376 and sequenced . To compare mutation rates in URA5 to the RAM pathway genes , aliquots from the same 20 ATCC 24067A cultures used to isolate URA5 mutants were diluted and plated onto ten YPD plates . Dry colonies were screened visually and pseudohyphal morphology confirmed by microscopy . The nature of the mutation in these strains was identified by complementation tests and DNA sequence analysis . Statistical analysis used the Poisson distribution , testing the probability of isolating n or more pseudohyphal strains . Strain AI228 ura#3 ( cbk1 ura5 ) was inoculated into 15 YPD cultures , grown overnight , and 2 . 5×108 cells plated onto YPD+2 µg/ml FK506 and YNB . 50 ml cultures in liquid YPD medium were incubated overnight at 150 rpm at room temperature . The cells were frozen and lyophilized . Total RNA was extracted with TRIzol ( Invitrogen ) or TRI reagent ( Sigma-Aldrich , St . Louis , MO ) . For northern blots , 10 µg of RNA purified from each strain were resolved on 1 . 4% agarose/formaldehyde gels . RNA was blotted to Zeta-Probe membrane ( Bio-Rad ) . [32P]-labeled probes of the six RAM genes ( the primers used for amplification are in Table S2 ) were hybridized to blots . Blots were stripped and reprobed with actin ( ACT1 ) as a loading and RNA transfer control . RNA purified from wild type strain KN99α was also used to confirm the intron-exon boundaries of TAO3 , by sequencing cDNAs reverse transcribed with Superscript III ( Invitrogen ) . Wax moth larvae ( G . mellonella ) were purchased from Vanderhorst Wholesale ( Saint Mary's , OH ) . Overnight cultures of C . neoformans grown in liquid YPD were washed in phosphate buffered saline ( PBS ) , the concentration determined by counting cells with a hemocytometer , and diluted such that 1×105 cells of wild type and the complementation mob2+MOB2 strain were injected into the larvae as described previously [82] . For mob2 mutant strain DM09 , 1×105 and 1×106 cells were inoculated . Concentrations were confirmed by plating serial dilutions onto YPD agar plates . Groups of female A/JCr mice ( NCI-Frederick , MD ) were infected intranasally with 105 cfus of each strain , as previously described [83] . Inocula were confirmed by plating onto YPD agar . Animals that appeared moribund or in pain were sacrificed by CO2 inhalation . For cfu assays , lungs and brain were dissected from animals , homogenized in PBS , and plated onto YPD medium containing ampicillin and chloramphenicol . Colonies were determined after incubation for 3 d at 30°C . For histology , lung and brain samples were fixed and hematoxylin and eosin ( H&E ) stained . Survival data from the murine experiments were statistically analyzed between paired groups using the log-rank test in the PRISM program 4 . 0 ( GraphPad Software ) . P values of <0 . 01 were considered significant . The mouse experiments were performed in full compliance with a protocol approved by the University of Medicine and Dentistry of New Jersey Institutional Animal Care and Use Committee , and in compliance with the United States Animal Welfare Act ( Public Law 98–198 ) . The experiments were carried out in facilities accredited by the Association for Assessment and Accreditation of Laboratory Animal Care . Sequences of CBK1 , MOB2 and TAO3 from strain ATCC 24067A have been deposited to GenBank , under accessions HM770879 , JX297541 and GU903010 , respectively . | Many diseases are contracted from the environment , rather than from sick people . It is unclear why those species are able to cause disease , since the selective pressures in the environment are presumed to be very different from those found within the host . Cryptococcus neoformans is a fungus that causes life-threatening lung and central nervous system disease in approximately one million people each year . The fungus is inhaled from environmental sources . One hypothesis to account for C . neoformans virulence is that amoeba are predators for this fungus , and surviving strains are pre-selected to be virulent in the human host . On the other hand , experiments have found that amoeba eat C . neoformans . A pseudohyphal cell type can survive , and while protecting against amoeba these cells are unable to cause disease in mouse models . We predicted that the pseudohyphal morphology reflected a change in function of a pathway of genes , and found that all pseudohyphal isolates contain mutations within genes for this pathway . The pseudohyphal trait is unstable , with reversion to normal yeast growth by counter-acting mutations . These mutations can occur during the course of mammalian infection . Our results show that mutation events account for a microevolution system currently described as phenotypic switching , and that mutations , at least under experimental conditions , can regulate pathogen adaptation and influence its host range . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"microbial",
"mutation",
"genetic",
"mutation",
"microbiology",
"host-pathogen",
"interaction",
"gene",
"function",
"mutation",
"molecular",
"cell",
"biology",
"mutation",
"types",
"fungal",
"evolution",
"microbial",
"evolution",
"cell",
"growth",
"molecular",
"genetics",
"mycology",
"medical",
"microbiology",
"biology",
"pathogenesis",
"adaptation",
"cell",
"biology",
"natural",
"selection",
"gene",
"identification",
"and",
"analysis",
"genetics",
"evolutionary",
"biology",
"evolutionary",
"processes",
"genetics",
"and",
"genomics"
] | 2012 | DNA Mutations Mediate Microevolution between Host-Adapted Forms of the Pathogenic Fungus Cryptococcus neoformans |
The mosquito Aedes aegypti is the principal vector of dengue and yellow fever flaviviruses . Temephos is an organophosphate insecticide used globally to suppress Ae . aegypti larval populations but resistance has evolved in many locations . Quantitative Trait Loci ( QTL ) controlling temephos survival in Ae . aegypti larvae were mapped in a pair of F3 advanced intercross lines arising from temephos resistant parents from Solidaridad , México and temephos susceptible parents from Iquitos , Peru . Two sets of 200 F3 larvae were exposed to a discriminating dose of temephos and then dead larvae were collected and preserved for DNA isolation every two hours up to 16 hours . Larvae surviving longer than 16 hours were considered resistant . For QTL mapping , single nucleotide polymorphisms ( SNPs ) were identified at 23 single copy genes and 26 microsatellite loci of known physical positions in the Ae . aegypti genome . In both reciprocal crosses , Multiple Interval Mapping identified eleven QTL associated with time until death . In the Solidaridad×Iquitos ( SLD×Iq ) cross twelve were associated with survival but in the reciprocal IqxSLD cross , only six QTL were survival associated . Polymorphisms at acetylcholine esterase ( AchE ) loci 1 and 2 were not associated with either resistance phenotype suggesting that target site insensitivity is not an organophosphate resistance mechanism in this region of México . Temephos resistance is under the control of many metabolic genes of small effect and dispersed throughout the Ae . aegypti genome .
Aedes aegypti is the principal vector of Dengue Fever ( DENV ) and Yellow Fever ( YFV ) flaviviruses throughout tropical and subtropical regions of the world and 2 . 5 billion people are at risk for DENV infection [1] . Currently DENV vaccines have low efficacy [2] , [3] so that vector control remains the only option to reduce or prevent DENV transmission . Adult control depends largely on the use of pyrethroid insecticides . However , resistance to pyrethroids has been rising globally [4] , [5] , [6] , [7] , [8] , [9] . More sustained control can potentially be achieved through the placement of insecticides in water containers that are known to harbor developing Ae . aegypti larvae in and around human habitations . For larval control , the three most widely used compounds are Bacillus thuringiensis israelensis ( Bti ) , methoprene , and temephos . Globally , temephos is the most widely used of these three due to its very low vertebrate toxicity , relatively low cost , the fact that methoprene is a growth regulator with greatest effectiveness against older ( third and fourth instar ) larvae [10] and , because Bti must be ingested to be effective , it does not affect late larval or pupal stages when active feeding has ceased . Temephos is one of a few organophosphates registered to control Ae . aegypti larvae , and is the only organophosphate with any appreciable larvicidal use . Temephos was first registered in the United States for mosquito control in 1965 . It was quickly adopted as a larvicide because it was effective in polluted water , had a long residual activity , was available in several use-specific formulations , had a different mode of action than alternatives , and could be used on any larval instar . Temephos is toxic to many mosquito vector species that grow in a diversity of stagnant , saline , brackish and temporary water bodies . It remains an important management tool for mosquito abatement programs . The most widely used commercial preparation of temephos is Abate ( EPA Registration No . 8329-60 , Clarke Mosquito Control Products , Inc . , Roselle , IL ) . Temephos was used for 30 years before initial reports of resistance appeared in 1995 . Initial studies reported less than a 5-fold resistance ratio ( RR ) in Ae . aegypti collections from Falcon and Aragua states of Venezuela [11] . In 1995 , larvae from 34 strains of Ae . aegypti from 17 Caribbean countries were bioassayed and there were fairly high levels of temephos resistance in Tortola , British Virgin Islands ( RR = 10–12 ) and Antigua ( RR = 6–9 ) [12] . In 1999 a Tortola collection of Ae . aegypti was tested and a RR = 47 was identified [13] . After 13 generations of temephos laboratory selection , the RR increased to 181 fold [13] . Since 2000 , temephos resistance has been reported from Cuba and Venezuela [14] , [15] , Thailand [16] , the Brazilian states of Sao Paulo [17] , Espirito Santo , Rio de Janeiro [18] , Sergipe , Alagoas , [19] , Ceara [20] , and Paraiba [21] . Most recently reports have appeared from El Salvador [22] , Martinique Island in the French West Indies [23] , Argentina [24] , [25] , India [26] , Colombia [27] , and Trinidad [28] , [29] . Although resistance to temephos has been demonstrated in many areas of the world , it is the only remaining organophosphate larvicide with any appreciable use . As such , it is an important tool in resistance management programs that depend on alternative larvicides . Alteration in the registration status or availability of temephos would have a large negative impact on our ability to control DENV transmission globally . The purpose of the present study was to develop a better understanding of the genetics underlying temephos resistance in Ae . aegypti using QTL mapping in recently collected strains . A strain previously established from Solidaridad , Mexico was selected to have 290 fold higher temephos resistance than another strain that had been established from Iquitos , Peru . Parents from these two strains were reciprocally crossed to generate F1 siblings which were then intercrossed to generate an F2 . The F2 generations were not large enough to assay for temephos resistance and so an F3 was generated through additional sib mating . F3 larvae were exposed to a discriminating dose of temephos and then checked every two hours up to 16 hours . Dead mosquitoes were preserved for DNA isolation at each time point and those surviving longer than 16 hours were considered resistant .
Two strains of Aedes aegypti were used . A F3 strain collected from Iquitos , Perú was kindly provided by Dr . Amy Morrison ( University of California , Davis ) . A second strain raised during two generations in the lab was collected by the authors from the neighborhood of Solidaridad , in the city of Chetumal , in the state of Quintana Roo , México . Eggs were hatched in deoxygenated water from egg papers and then fed brewer's yeast . Adults were provided 10% ( w/v ) sucrose solution and were blood fed on citrated sheep blood in an artificial membrane feeder every three days . Incubators were set to a 14∶10 photoperiod , 30°C water temperature for larvae and 28°C for adult with a relative humidity of 85% . F2 or F3 offspring from the field constituted the FS0 generation in the selection experiments . FS0 larvae were bioassayed to estimate the concentration of temephos ( Chem Service , West Chester , PA ) necessary to kill 50% of larvae ( LC50 ) . Bioassays were performed in plastic cups containing 100 ml of water with five different concentrations of temephos in 1 mL ethanol as a solvent . Approximately 25 3rd-instar larvae were gently pipetted into each cup . Mortality was recorded every 15 minutes up to two hours . All larvae were then transferred into clean water and mortality was scored at 24 hours . Each bioassay was performed in triplicate to obtain ∼75 larvae per concentration . LC50 and confidence limits were calculated using the IRMA quick calculator software ( http://sourceforge . net/projects/irmaproj/files/Qcal/beta/QCal_ver_0 . 1_rev190 . msi/download ) which performs logistic regression [30] . Selection proceeded in three replicate lines for three generations . In the first round of selection 40–100 third instar larvae from each of the three replicates were exposed to an LC50 of 30 ng temephos/mL for two hours . Larvae were then transferred to clean water and mortality was recorded at 24 hours . Surviving larvae were transferred to 1 cubic foot rearing cages ( BugDorm-1 , Mega View Science , Co . ) and raised to adults who were then blood fed to obtain FS1 eggs . We performed an initial bioassay with ∼75 larvae in each of the subsequent FS1–FS3 generations of selection to calculate the new LC50 . From 40–100 larvae from each replicate were then exposed to the new LC50 . For the P1 mapping family , we crossed Solidaridad ( SLD ) FS3 and Iquitos ( Iq ) adults . Twenty P1♀SLD FS3×♂Iq and twenty reciprocal P1♀Iq×♂ SLD FS3 crosses were made . Larvae from each line were hatched and at the pupal stage , a female ( larger size ) from one strain was transferred to plastic cups in cardboard containers with a male pupa from the other strain . After adults emerged , they were allowed to mate for 3 days and the P1 male was frozen and held at −80°C . Females were blood fed three times with an artificial membrane feeder over the next ten days and the P1 female was then frozen and held at −80°C . Egg batches were maintained at room temperature for 7 days and then hatched by submersion in water followed by feeding them on Brewer's yeast ad libidum . For the F1 intercross families , one female and one male pupa from the same P1 family were allowed to emerge , mate and blood fed to eventually generate F2 progeny . F2 eggs from the largest F1 families were hatched and siblings were intercrossed in a single cage . Third instar larvae ( 200 total ) were exposed to 250 ng temephos/mL . After 2 hours , larvae that were unresponsive to prodding with a pipette tip were individually transferred to a labeled 1 . 5 mL microcentrifuge tube and frozen at −80°C . This was repeated every two hours for the next 16 hours . After 16 hours all remaining larvae were recorded as resistant . The DNA of the P1 and F1 parents , and the two sets of 200 F3 offspring was individually isolated following the salt extraction method [31] and then suspended in 200 µl of TE buffer ( 10 mM Tris-HCl , 1 mM EDTA pH 8 . 0 ) . The DNA was divided into 2–100 µl aliquots and stored at −80°C . A total of 23 single copy genes [32] , [33] and 26 microsatellite loci from [34] were amplified and analyzed . Each of these 49 genes has a known physical and linkage map position in the Ae . aegypti genome . A PCR mixture sufficient to perform 100 25-µl reactions was made by mixing 2 , 114 µL ddH2O , 250 µL 10×Taq buffer ( 500 mM KCl , 100 mM Tris-HCL pH 9 . 0 ) , 25 µL of 20 mM dNTPs , and 2 , 500 pm of each of the primers . This reaction mixture was set under a UV light source ( 302 nm ) for 10 min , after which 20 µl of Taq DNA polymerase was added . The mixture was then dispensed into a 96-well plate . Template DNA ( ∼100 ng ) was then added to each well , followed by a drop of sterilized mineral oil . Each set of reactions was checked for contamination by the use of a negative control containing all reagents except template DNA . Samples were stored at 4°C before electrophoresis . The contents of each well were tested for the presence of amplified products by loading 5 µl from each well onto a 1 . 5% ( w/v ) agarose gel made with Tris-Borate-EDTA buffer . DNA fragments were size fractionated by electrophoresis for 15–20 min at 112 V . Fragments were visualized by staining with Syber Green and viewing the gel over a UV transilluminator . SSCP analysis and silver staining procedures were previously published [31] . Polymorphic SSCP-markers were sequenced in the four P1 and F1 parents to test for SNPs and to determine the inheritance patterns of SNP alleles . Sequences were aligned using CLUSTALW [35] . Allele specific primers were designed at those loci in which genotypes were fully or partially informative in the P1 and F1 parents . Design of primers for melting curve PCR is previously published [36] . Allele specific fragments were detected by melting curve PCR in a CFX-96 Real time PCR detection system ( Bio-Rad , Hercules , CA ) . Table S1 provides previously unpublished oligonucleotide sequences for allele specific detection . Associations between genotypes at each marker locus and hours until death ( HTD ) phenotype were initially assessed with ANOVA using summary ( glm ( HTD∼“Marker locus name” ) ) in R2 . 15 . 2 [37] . Our null hypothesis was that HTD was equal in each genotype . Associations between death ( scored 0 ) or survival ( 1 ) ( DOA ) after 16 hours were initially assessed with Fisher's exact test ( table ( DOA , “Marker locus name” ) ) in R2 . 15 . 2 . The null hypothesis was that the proportions of surviving larvae were equal in each genotype class . When the ANOVA or Fisher's exact test yielded a probability below 0 . 05 , we examined the inheritance of the alleles at that locus . Our a priori hypothesis was that an excess of F3 individuals with an allele inherited from the SLD P1 parent would be resistant while an excess of F3 individuals with an allele inherited from the Iq P1 parent would die . Multiple Interval mapping ( MIM ) [38] was then performed using QTL Cartographer 2 . 5 [39] . Two separate MIM were done . First , mosquitoes were scored as 2 , 4 , 6 , 8 , 10 , 12 , 14 , 16 or 24 corresponding to hours until death . Second , F3 mosquitoes were scored as one if they survived to 16 hours or as zero if they died before 16 hours . In either case we created an initial model containing QTL map positions for markers at which ANOVA or Fisher's exact tests were significant . This model was then refined in MIM by 1 ) searching for new QTL , 2 ) estimating QTL effects , 3 ) obtaining and recording a summary , 4 ) optimizing QTL position , 5 ) searching for new QTL interactions , 6 ) testing for existing QTL main effects , 7 ) testing for existing QTL interaction effects , and 8 ) obtaining and recording a final summary . In addition , we used QTL Cartographer 2 . 5 to perform an initial MIM model selection on all markers using forward and backward selection with a significance level criterion of 0 . 01 . We then compared this model with the model based upon markers identified as significant by ANOVA or Fisher's exact tests . The models agreed in all four cases: ( 1 ) ♀ SLD FS3×♂Iq –HTD ( 2 ) ♀ SLD FS3×♂Iq –DOA , ( 3 ) P1 ♀ Iq×♂ SLD – HTD and ( 4 ) P1 ♀ Iq×♂ SLD – DOA .
The concentration of temephos sufficient to kill 50% of larvae ( LC50 ) was 50 ng temephos/mL water for the Iquitos strain . The Solidaridad FS0 strain initially had an LC50 of 27 ng temephos/mL water . Following three generations of temephos selection , the LC50 increased to 7 . 9 ug temephos/mL water in the Solidaridad strain . Thus the selected Solidaridad strain had ∼160 fold higher temephos resistance than the Iquitos strain . Among the SLD×Iq F3 larvae the LC50 was 6 . 5 ug temephos/mL water and was 1 . 9 ug temephos/mL water among the IqxSLD F3 larvae . The genetic markers used in constructing maps in both the SLDxIq and IqxSLD crosses are listed along with their linkage positions in Table S2 . Results of the ANOVA to test the null hypothesis that time until death is equal among genotypes are presented in Table 1 . Results of Fisher's Exact Test on proportions of surviving larvae among genotype classes appear in Table 2 . Loci with significant results are shown for all three chromosomes in Figure 1 . In the SLDxIq cross there were five QTL on chromosome 1 associated with HTD , four on chromosome 2 and four on chromosome 3 . In the same cross there were four QTL on chromosome 1 associated with DOA , four on chromosome 2 and four on chromosome 3 . In the IqxSLD cross there were three QTL on chromosome 1 associated with HTD , four on chromosome 2 and five on chromosome 3 . There was one QTL on chromosome 1 associated with DOA , two on chromosome 2 and three on chromosome 3 . The two families shared common QTL at loci 192TAAA1 and 88GAA1 on chromosome 1 , at loci 462GA1 and 1132CT1 on chromosome 2 and at locus 86AC1 on chromosome 3 . Between the two families there were six , six and nine QTL affecting HTD on chromosomes 1 , 2 , and 3 , respectively or 21 loci in total . In the two families there were four , five and six QTL affecting DOA on chromosomes 1 , 2 , and 3 , respectively or 15 loci in total . When the ANOVA or Fisher's exact tests yielded a probability below 0 . 05 , we examined the inheritance of the alleles at that locus . The last columns of Tables 1 and 2 indicate when the allele inherited from the SLD FS3 P1 parent were associated with resistance while the allele inherited from the Iq P1 parent was associated with susceptibility . Figure 2 plots HTD among larvae with the three possible genotypes . The first column of plots correspond to chromosomes 1 , 2 , and 3 in the SLDxIq cross . SLD alleles conferred slightly greater longevity for the first three marker loci on chromosome 1 but Aegi22 Iq homozygotes had greater longevity than heterozygotes ( Fig . 2A ) . In contrast , SLD alleles confer greater longevity for all marker loci on chromosome 2 ( Fig . 2B ) and the effects appear to be additive . On chromosome 3 , no general trend is evident ( Fig . 2C ) . Iq homozygotes confer slightly greater longevity at marker loci 69TGA1 and para . The opposite trend is seen in markers 766ATT1 and 86AC1 . The second column in Figure 2 corresponds to chromosomes 1 , 2 , and 3 in the Iq×SLD cross . Again , SLD alleles confer slightly greater longevity on chromosome 1 ( Fig . 2D ) . In contrast , on chromosome 2 SLD alleles at markers 328CTT1 , 462GA1 , and Arc4 confer only slightly greater longevity ( Fig . 2E ) while SLD alleles at the 1132CT1 locus appear to act as recessives in conferring much greater longevity . A similar pattern is seen in SLD alleles at 301ACG1 on chromosome 3 ( Fig . 2F ) . However , Iq homozygotes confer slightly greater longevity at marker loci CCEae2D , vitg , 201TTA1 and Apyr1 . Figure 3 plots proportion surviving past 16 hours among larvae with the three possible genotypes . In the SLDxIq cross SLD alleles conferred greater survival at the first three marker loci on chromosome 1 but Aegi22 Iq homozygotes had greater longevity than heterozygotes ( Fig . 3A ) . Note that these are the same markers as in Figure 2A , but with markers 192TAAA1 , and 88GAA1 . SLD alleles confer a 50% increase in survival . On chromosome 2 ( Fig . 3B ) , with the exception of Arc4 , SLD alleles at markers , 462GA1 , Carbox and 1132CT1 all greatly increase survival . SLD alleles at 462GA1 appear to act additively in increasing survival from zero in Iq homozygotes to 50% in heterozygotes to 100% in SLD homozygotes . Resistant alleles at markers Carbox and 1132CT1 are recessive with 75–80% greater survival in SLD homozygotes . As with HTD , on chromosome 3 there is no general trend ( Fig . 3C ) . Iq homozygotes confer slightly greater survival at marker loci 69TGA1 and para but the opposite trend is seen in markers 766ATT1 and 86AC1 . In the Iq×SLD cross ( Fig . 3D ) SLD alleles at marker 88GAA1 increase survival by 50% and SLD alleles appear recessive . Similarly , alleles at the 1132CT1 marker increased survival by 90% . Identical patterns were seen in the SLDxIq cross ( Fig . 3B ) . On chromosome 3 , Iq homozygotes confer slightly greater survival at marker loci CCEae2D , vitg , and 86AC1 . The results of Multiple Interval Mapping with the HTD and DOA phenotypes are shown for both crosses in Table 3 . Eleven QTL were identified in the SLD×Iq cross and these accounted for 68% of the phenotypic variance in HTD . There were nine QTL that accounted for 63% of the phenotypic variance in DOA . These nine were also all associated with HTD . The QTL that accounted for most ( 48% ) of the genetic variation in HTD were at 47 cM and 70 cM on chromosome 2 . The QTL that accounted for the most variation in DOA was at 62 cM on chromosome 2 . QTL at 30 cM and 70 cM on chromosome 1 affected both phenotypes . Genetic factors accounted for less of the variation in HTD and DOA phenotypes in the Iq×SLD cross . Eleven QTL were identified that accounted for 58% of the phenotypic variance in HTD . There were only two QTL that accounted for 31% of the variance in DOA and these were also associated with HTD . The QTL that accounted for most of the variation in HTD were at 57 cM on chromosome 1 , 64 cM on chromosome 2 and 43 cM on chromosome 3 . The only QTL that accounted for negligible variation in DOA was at 62 cM on chromosome 2 . QTL at 57 cM on chromosome 1 affected both phenotypes . QTL at 30 and 57 cM on chromosome 1 , and at 23 . 5 and 70 cM on chromosome 2 were common to both families
QTL mapping indicates that resistance to temephos is conditioned by many regions of the Ae . aegypti genome and therefore appears to behave as a classic quantitative genetic trait that is controlled by many loci each of minor effect . This pattern is supported by a recent parallel study in which we tracked changes in transcription of metabolic detoxification genes using the Ae . aegypti ‘Detox Chip’ microarray [40] during five generations of temephos selection [41] . We selected for temephos resistance in three replicates in each of six collections , five from México , and one from Peru . We used the esterase inhibitor DEF ( S . S . S-tributylphosphorotrithioate ) to show that esterases were the major metabolic source of resistance . However , the microarray data indicated that expression of many esterase genes increased with selection and that no single esterase was consistently upregulated among the six selected lines . Target site resistance in acetylcholine esterase genes is a very common mechanism of resistance to organophosphate and carbamate insecticides [42] . We therefore tested for a significant genotype -phenotype interaction with SNPs in the AChE-2 gene ( AAEL012141 ) at 40 . 7 cm on chromosome 1 and the AChE-1 gene ( EF209048 ) at 3p1 . 2 ( 30 . 4 cM ) on chromosome 3 [43] . Results in Table 1–3 show that no significant associations were detected . Similar studies of temephos resistance in field populations of Ae . aegypti also failed to detect insensitive acetylcholine esterase [44] despite the fact that these authors were able to generate recombinant clones that produced Ae . aegypti insensitive acetylcholine esterases in the laboratory [45] . Another possibility is that temephos in particular fails to select for insensitive acetylcholine esterases . Cuban investigators were able to select Ae . aegypti with 13-fold increase in insensitive acetylcholine esterase but using the carbamate insecticide propoxur [46] . Previous studies of esterase isozyme loci identified two genetically mapped loci associated with resistance to the organophosphate insecticide malathion . Elevated activity staining of Esterase-5 located at 57 cM at the base of Chromosome 1 [47] was reported [48] . This may correspond to the 57 cM QTL on chromosome 1 associated with marker 88GAA1 in both families in the current study . Similarly elevated activity staining of Esterase-6 located at 83 cM at the base of Chromosome 2 in the map of [47] was reported [49] , [50] . This may correspond to the QTL at 70 cM on chromosome 2 associated with marker 1132CT1 found in both families in the current study . We have no means to formally check these associations because neither the nucleotide nor amino acid sequences of proteins Esterase-5 and 6 are known . There are 49 currently identified carboxy/choline esterase genes [40] . With the recent publication of a physical map that contains 45% of the Ae . Aegypti genome [51] , [52] we had hoped to learn the physical locations of many of these esterases . However , other than AChE-1 and AChE-2 , there were only six other esterase genes that occurred in mapped supercontigs . These were CCEbe2o ( AAEL008757 ) on 2p3 . 4 ( also mapped in the present study see Figure 1 ) , CCEjhe2o ( AAEL004323 ) on 2q2 . 4 , and four ( CCEjhe1F ( AAEL005200 ) , CCEjhe2F ( AAEL005198 ) , CCEjhe3F ( AAEL005210 ) , and CCEjhe4F ( AAEL005182 ) ) all located in supercontig 1 . 145 at 2p4 . 4 . Whether these four are associated with the QTL at 5 . 8 cM on the top of Chromosome 2 in the Qi×SLD cross ( see Tables 1–2 ) is unknown at this time . Even though the selected Solidaridad strain had overall ∼160 fold higher temephos resistance than the Iquitos strain , this pattern wasn't uniform across the entire genome . This could have affected the locations and relative contributions of QTL . There are many instances in Tables 1 and 2 wherein the mosquitoes homozygous for markers from the “susceptible” Iquitos strain were more resistant than heterozygotes or homozygous for markers from the “resistant” SLD strain ( note especially the bottom of chromosome 3 for both HTD and DOA ) . This counterintuitive outcome is probably a result of using Iquitos mosquitoes taken directly from the field without selecting for a more susceptible phenotype . However , it could also be associated with negative fitness effects associated with resistance alleles in the SLD strain that became concentrated during selection . In our previous QTL mapping study [36] we found resistance to permethrin to be principally ( 91 . 8% of genetic effect in MIM ) under the control of target site insensitivity in the voltage gated sodium channel gene ( orthologue of paralysis in Drosophila [53] ) . We have shown that the genetic architecture underlying temephos resistance to be completely different with both families having up to 11 QTL affecting the HTD phenotype in both families and from 2–9 QTL affecting DOA . The practical implications of these findings are that selection for temephos resistance in the field is likely to involve many ( principally esterase ) loci . It is unlikely that the same genes will be involved in all field populations and that genetic drift may play a large part in determining which combinations of the 49 currently identified carboxy/choline esterase genes [40] become upregulated and assume responsibility for metabolic detoxification of temephos . | The mosquito Aedes aegypti is the principal vector of dengue and yellow fever flaviviruses . Due to a lack of effective drugs or vaccines , if an epidemic of dengue fever occurs in the near future , the first line of defense will involve the use of insecticides to suppress adult populations of Ae . aegypti . Unfortunately , the species has become resistant to most of the insecticides that can be safely applied . The authors have worked extensively on the mechanisms of resistance to the various insecticides commonly used for suppression of Ae . aegypti populations . Temephos is an organophosphate insecticide used globally to suppress Ae . aegypti larval populations but resistance has evolved in many locations . In this study we show that temephos resistance is under the control of many metabolic genes of small effect and dispersed throughout the Ae . aegypti genome . This information will be of general interest to field workers involved in the suppression of field populations of Ae . aegypti . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"biology",
"and",
"life",
"sciences",
"ecology",
"and",
"environmental",
"sciences",
"medicine",
"and",
"health",
"sciences"
] | 2014 | QTL Mapping of Genome Regions Controlling Temephos Resistance in Larvae of the Mosquito Aedes aegypti |
Pseudomonas aeruginosa causes severe sight-threatening corneal infections , with the inflammatory response to the pathogen being the major factor resulting in damage to the cornea that leads to loss of visual acuity . We found that mice deficient for macrophage migration inhibitory factor ( MIF ) , a key regulator of inflammation , had significantly reduced consequences from acute P . aeruginosa keratitis . This improvement in the outcome was manifested as improved bacterial clearance , decreased neutrophil infiltration , and decreased inflammatory responses when P . aeruginosa-infected MIF knock out ( KO ) mice were compared to infected wild-type mice . Recombinant MIF applied to infected corneas restored the susceptibility of MIF deficient mice to P . aeruginosa-induced disease , demonstrating that MIF is necessary and sufficient to cause significant pathology at this immune privileged site . A MIF inhibitor administered during P . aeruginosa-induced infection ameliorated the disease-associated pathology . MIF regulated epithelial cell responses to infection by enhancing synthesis of proinflammatory mediators in response to P . aeruginosa infection and by promoting bacterial invasion of corneal epithelial cells , a correlate of virulence in the keratitis model . Our results uncover a host factor that elevates inflammation and propagates bacterial cellular invasion , and further suggest that inhibition of MIF during infection may have a beneficial therapeutic effect .
Eye trauma and contact lens wear are the main factors that predispose to the development of infectious keratitis associated with vision loss and blindness [1] , [2] . The organism most often isolated from contact lens associated corneal ulcers is Pseudomonas aeruginosa [3] , [4] . Existing therapies often fail to control the excessive tissue damage that is induced during P . aeruginosa infection [5] . While antibiotic treatment reduces the bacterial burden , tissue damage still occurs as a result of an poorly-controlled local inflammation . Hence , new therapeutic modalities are needed to control the inflammatory response in addition to the antibiotic treatments . We hypothesized that the innate immunity factor–MIF ( NP 002406 ) –could promote the pathogenic consequences of P . aeruginosa infection by potentiating local inflammation , and , if so , could be a suitable drug target for treatment . MIF is an innate immunity molecule with ubiquitous tissue expression leading to induction of proinflammatory activities . MIF was originally described as a regulator of macrophage responses [6] . It directly or indirectly promotes expression of a large panel of pro-inflammatory cytokines including TNF-α ( P01375 ) , IFN-γ ( P 01579 ) , IL-1β ( NP 000566 ) , IL-2 ( AAA 59140 ) , IL-6 ( CAG 29292 ) , IL-8 ( CAG 46948 ) , MIP-2 ( AAF 78449 ) , NO , COX2 ( P 00403 ) , products of the arachidonic acid pathway and matrix metalloproteinases [7] . Interestingly , low levels of MIF can override the anti-inflammatory properties of glucocorticoids by reversing the inhibitory effect of glucocorticoid on production of TNF-α , IL-1β , IL-6 and IL-8 [8] . Detailed studies performed in the rat have shown that preformed MIF protein is released into the circulation within 6 hrs of LPS injection [9] . LPS toxicity is exacerbated by co-injection of recombinant MIF ( rMIF ) with LPS , whereas neutralization of MIF activity reduces the circulating levels of TNF-α by 50% and rescues mice from lethal LPS-induced endotoxic shock [10] , [11] . All these properties of the MIF molecule suggest that MIF has a prominent regulatory role related to inflammation , likely with a critical function as an effector molecule that is active early in the course of infection with a pathologic function when continued production exacerbates inflammation , giving rise to attendant tissue pathology . The contribution of MIF during responses to infections by a variety of pathogens , including bacteria , viruses , and parasites is currently an area of active research . Recent clinical correlative studies have demonstrated increased MIF levels and elevated MIF-dependent proinflammatory cytokines are produced during H1N5 influenza infection , dengue fever , and bacterial urinary tract infections [12] , [13] , [14] . These results demonstrate an important contribution of MIF to the pathogenesis of viral or bacterial induced inflammation and suggest a possible beneficial role of neutralizing MIF as an adjunctive therapeutic approach to treat the severe forms of disease . In the eye , the high levels of MIF protein expression and consequent inhibition of cellular migration has supported the conclusion that MIF contributes to the establishment of the eye's immune privilege status due to immunosuppressive activities [15] , [16] , [17] , [18] . While this could be important in the resting state , the strong impact MIF has on induction of inflammation in response to infection suggests a different role for this molecule during active infection . In this study , we tested whether inhibition of MIF in the eye led to a decrease in P . aeruginosa-induced pathology , and if this was associated with decreased PMN infiltration and thus decreased inflammation and corneal pathology . Overall , we found that inhibition of MIF had a pronounced salutary effect on eye pathology emanating from P . aeruginosa keratitis , which we associated with a better ability of PMN from MIF-deficient mice to mediate opsonic killing of this organism , making MIF a promising therapeutic target to control local inflammation in the context of P . aeruginosa corneal infection .
To determine the effect of MIF on P . aeruginosa-induced keratitis , MIF KO and C57Bl6 WT mice were infected with P . aeruginosa strain 6294 and the bacterial levels in the corneas of mice measured to monitor disease progression . At 24 h after onset of infection there were modest but significantly ( P = 0 . 04 ) lower bacterial levels in both the extracellular and intracellular samples from the corneas of the infected MIF KO mice when compared to C57Bl6 control animals ( Fig . 1A ) . However , the corneal pathology scores of MIF KO and C57Bl6 were comparable at 24 h . At 48 h after infection the differences between the MIF KO and C57Bl6 mice increased dramatically: there were about 100-fold fewer bacteria recovered from the corneal cell exterior in the infected MIF KO mice compared to the C57Bl6 controls . Even more dramatically , about 10 , 000 less intracellular bacteria were recovered from MIF KO mice when compared to C57Bl6 mice ( Fig . 1B ) . Consistent with the reduced bacterial levels , the MIF KO mice infected with P . aeruginosa strain 6294 had a significant decrease in corneal pathology ( P = 0 . 001 ) . As is known from published data [19] , if a P . aeruginosa corneal infection is left to proceed for a longer period of time in C57Bl6 mice , perforation of the cornea occurs . This outcome was observed in an additional group of C57BL6 control mice , but not in MIF KO mice in the C57Bk6 background ( Fig . 1C ) . This reduced pathology in the MIF KO mice was observed across a range of bacterial challenge doses from 1×106 ( data not shown ) to 1×107 cfu/mouse eye ( Fig . 1C ) . The improved outcomes in MIF KO mice was also obtained with P . aeruginosa strain PAO1 48 h following infection ( Figure S1 ) . Since MIF regulates the pro-inflammatory responses to LPS challenge in vitro as well as in several in vivo models , we examined how MIF might modulate the molecular responses to P . aeruginosa induced keratitis by measuring local cytokine production based on prior findings as to which mediators are regulated by MIF and a small-scale microarray analysis we performed ( data not shown ) on infected corneas from WT and MIF KO mice to suggest which cytokines might be candidates for a more thorough investigation . A panel of five cytokines , TNF-α , IL-1β , KC , IL-6 , and IL-10 , were quantified at the protein level in infected corneas harvested from individual animals . Infected MIF KO mice had significantly lower levels of TNF-α , IL-1β , IL-6 , and KC protein in corneal homogenates compared to corneas from infected C57Bl6 mice . There was no difference in the levels of the anti-inflammatory cytokine IL-10 ( Fig . 2 ) . To determine if differences in inflammatory responses translate into differences in the presence of infiltrating PMNs during infection , histologic observations were carried out . C57Bl/6 mice showed destruction of the epithelial layer of the cornea and elevated infiltration of PMNs in the aqueous chamber . In contrast , in the MIF KO mice , there were less infiltrating PMNs , less edema , and no destruction of the epithelial layer ( Fig . 3A and 3B ) when either strain 6294 or PAO1 were used to induce infections . The pathology in the C57Bl/6 corneas was consistent with an acute P . aeruginosa infection . The findings on bacterial levels and PMN infiltration into the P . aeruginosa-infected cornea indicated that MIF promoted PMN infiltration leading to more pathology while also inhibiting the clearance of this microbe from the cornea . To determine a potential cellular basis for the improved outcomes in MIF-deficient mice , we compared the opsonic killing activity of PMNs obtained from MIF-sufficient and MIF-deficient animals . MIF-deficient PMNs phagocytosed P . aeruginosa strain PAO1 more efficiently that MIF-sufficient PMNs , as evident by the decreased number of P . aeruginosa recovered from the PMN/bacteria mixture at the end of the experiment ( Fig . 3D ) . Since it is well documented that human epithelial cells participate in neutrophil recruitment in response to infection , we determined if MIF affects IL-8 synthesis by corneal epithelial cells [20] , [21] , [22] . Primary human corneal epithelial cells were grown to confluence , treated with either a siRNA to knock-down MIF or with a control ( control ) siRNA ( Fig . 4A ) . Seventy-two h after the knock-down cells were infected with P . aeruginosa strain 6294 ( Fig . 4B ) . We also evaluated the effect of adding recombinant MIF ( rMIF ) to cells treated with siRNA for MIF . Tissue culture supernatants were collected and IL-8 levels measured by ELISA . P . aeruginosa infection induced IL-8 expression in control siRNA-treated cells , whereas IL-8 production was much less pronounced in MIF siRNA-treated cells ( ANOVA , P<0 . 01 ) ( Fig . 4B ) . Treatment with rMIF reconstituted P . aeruginosa-induced IL-8 production in MIF-siRNA treated cells . Protein concentrations in cellular lysates were determined and no differences found , indicating no major changes in cellular numbers occurred during the experiments . We next analyzed the effect of a small molecule inhibitor of the MIF tautomerase activity , 4-IPP , on P . aeruginosa-induced inflammation . Published studies have demonstrated that 4-IPP forms covalent complexes with MIF , and , hence , inhibits MIF activity [23] . Treatment with 4-IPP almost completely abolished IL-8 synthesis . This effect of 4-IPP was observed at concentrations of 200 µM and 400 µM ( Fig . 4C ) . Since we were concerned that the 4-IPP treatment could induce cellular apoptosis due to the documented role of MIF in promoting cell survival [24] , [25] , we next analyzed lysates obtained from non-treated and P . aeruginosa strain 6294 infected cells in the presence or absence of 400 µM 4-IPP and probed for activated caspase 3 as a marker for apoptosis . At 6 h post infection we did not observe activation of caspase 3 , however some activation was detected at 24 h post-infection ( Figure S2 ) . Hence , we have performed the majority of our infection experiments within 6 h post-infection . When epithelial cells were stimulated by P . aeruginosa strains 6294 or PAO1 both IL-1β and IL-6 secretion was abolished in the presence of the MIF inhibitor ( Fig . 4D ) . As P . aeruginosa bacteria found in the cornea during infection are mostly intracellular , we determined if MIF modulated the uptake of bacteria by corneal epithelial cells using in vitro invasion assays . Inhibition of MIF by either siRNA treatment or by treatment with small molecule inhibitor 4-IPP , significantly decreased the internalized levels of P . aeruginosa in primary human corneal epithelial cells ( Fig . 5 ) . These findings suggest that MIF modifies epithelial responses facilitating bacterial entry in the epithelial cells and thus inhibits the niche that P . aeruginosa uses to survive and escape from PMNs recruited to the eye in response to infection . To verify that MIF was the major factor responsible for the reduced infectious complications of P . aeruginosa keratitis in MIF KO mice , reconstitution experiments were performed . One µg of rMIF administered topically onto the eye of infected MIF KO mice restored the wild-type response to infection with P . aeruginosa strain 6294 ( Fig . 6 ) . This increased bacterial burden in the reconstituted MIF KO correlated with elevated pathology scores , indicative of increased susceptibility to infection . To analyze the effect of MIF inhibition during P . aeruginosa-induced keratitis , cohorts of C57Bl6 mice were infected with 5×105 cfu/mouse of P . aeruginosa strain 6294 and treated topically with gentamicin starting 24 h post-infection . A separate cohort of infected C57Bl6 mice received both 4-IPP IP and gentamicin eye-drops on the same schedule . Animals were monitored for 5 days to analyze recovery from infection and the infected corneas were harvested to quantify the intracellular and extracellular bacteria . Mice that received the 4-IPP and gentamicin treatment cleared P . aeruginosa more efficiently than did the group treated only with gentamicin as demonstrated by the significantly decreased extracellular and intracellular cfu recovered from the corneas , and decreased pathology scores ( Fig . 7 ) .
Pseudomonas aeruginosa causes serious corneal disease , associated with acute ocular inflammation . Both bacterial and host factors contribute to the pathogenesis of P . aeruginosa keratitis , indicating that the optimal interventions for this infection will involve both control of bacterial levels and modulation of inflammation to limit pathology while not compromising bacterial clearance . Here , we demonstrated that a master regulator of inflammation–MIF–promoted both pathology associated with P . aeruginosa corneal infection and inhibited bacterial clearance as shown by outcomes in MIF-KO mice which had decreased bacterial burdens in the corneas , decreased inflammation , and decreased neutrophil infiltration compared to MIF-sufficient WT mice . These improved outcomes in MIF-deficient mice could be reversed by treating P . aeruginosa-infected eyes with rMIF and could be mimicked in WT mice treated with the MIF inhibitor 4-IPP . Thus , it appears that one host molecule makes a major contribution to both aspects of disease resulting from P . aeruginosa corneal infection , indicating a high potential for ameliorating the consequences of this infection by therapies that target the activities of MIF . To define mechanisms associated with the effects of MIF in P . aeruginosa keratitis we used in vitro studies to show that reduction of MIF levels by siRNA treatment or by the use of the tautomerase inhibitor of MIF , 4-IPP , results in decreased production of IL-8 , IL-1β , and IL-6 by human corneal epithelial cells after infection . Consistent with this finding , the absence of MIF in vivo also reduced inflammatory responses to P . aeruginosa during experimental murine corneal infection . Furthermore , we also found that PMNs from MIF-deficient mice had a better innate ability to kill P . aeruginosa in an in vitro opsonic-killing assay , indicating a likely cellular basis for the better control of infection in thee animals . Of note , Niederkorn and colleagues [18] found that MIF inhibits perforin release by NK cells and inhibits NK cell motility , indicating a contribution of MIF to the immune-privileged status of the eye . Thus , it appears in the basal state MIF limits cellular activation , but a consequence of this is that the responses to infection are dampened and the increased infiltration of PMN into the infected cornea that is needed to deal with the increased bacterial burdens results in more tissue pathology . A similar improved outcome from P . aeruginosa lung infection was reported for MIF-deficient mice [11] . Successful recovery form acute bacterial infection depends on timely recruitment of neutrophils . PMNs are present in the cornea and aqueous chamber in MIF deficient mice at reduced levels when compared to infected WT control mice , yet , there are sufficient numbers of PMNs recruited in the MIF KO to clear bacteria , as demonstrated by the decreased levels of P . aeruginosa measured at 48 h after the onset of infection . Hence , MIF deficiency both reduces pathology during infection and still allows for sufficient host responses to clear bacteria . Our data suggests that in infected mouse corneas , inflammatory mediators such as KC or MIP2 are initially released by corneal epithelial cells or keratocytes which in turn recruit PMNs to effectively handle bacterial infections in the eye [26] , a process that occurs in the infected MIF KO mice leading to resolution of infection . In contrast , in the infected C57Bl6 mice these mediators are produced at higher levels and become part of a classic hypersensitivity response that contributes to tissue damage and elevated pathology [27] , [28] . Taken together these data suggest that by modulating the level of inflammation occurring in the eye triggered by P . aeruginosa infection it is possible to promote expression of proinflammatory mediators sufficient to control infection while also preventing inflammation-induced corneal pathology and associated loss of vision .
Ethics Statement: All studies were performed in accordance with the Harvard Medical School Institutional Animal Care and Use Committee guidelines ( approval date 10/14/2008; number IACUC 02432 ) . Breeding pairs of mif knock-out ( KO ) mice were a kind gift from Dr . C . Gerard ( Children's Hospital Boston ) . Control mice ( C57Bl6 ) were obtained from Taconic Farms . Mice were housed and bred in the Channing Laboratory Animal Care Facilities . Mice were housed and bred in one of the Harvard Medical Area Animal Care Facilities . P . aeruginosa strains PAO1 and 6294 were used throughout these experiments . Effects from low dose ( 5×105 cfu/eye ) and high dose ( 1×107 cfu/eye ) inocula were evaluated to determine the doses to use in the in vivo infection studies . Generally , the bacterial strains were grown o/n at 37C on Tryptic Soy Broth agar plates prior to experiments . The experimental protocols were approved by the Institutional Animal Care and Use Committee of the Harvard Medical Area Office for Research Subject Protection and were consistent with the Association for Research in Vision and Ophthalmology guidelines for studies in animals . Infections were carried out as described previously [29] . Mice were anesthetized with ketamine and xylazine injections . Three 0 . 5 cm scratches were made on the cornea and an inoculum of P . aeruginosa delivered in 5 µl onto the eye . Mice remained sedated for about 30 min . For evaluation of corneal pathology , daily scores are recorded by an observer unaware of the experimental status of the animals based on a the following scoring system using a graded scale of 0 to 4 as follows: 0 , eye macroscopically identical to the uninfected contra-lateral control eye; 1 , faint opacity partially covering the pupil; 2 , dense opacity covering the pupil; 3 , dense opacity covering the entire anterior segment; and 4 , perforation of the cornea , phthisis bulbi ( shrinkage of the globe after inflammatory disease ) , or both . To determine the levels of bacteria in the cornea 24 or 48 h after infection , mice were sacrificed , eyes enucleated and corneas dissected from the ocular surface . To quantify the extracellular levels of P . aeruginosa , corneas were excised and suspended in PBS , vortexed , serial dilutions made and plated on P . aeruginosa selective cetrimide plates . To measure the intracellular levels of P . aeruginosa , the intact corneas were placed in F12 medium containing 5% FBS and 300 µg gentamicin/ml for 60 min , vigorously rinsed to wash away the antibiotic and the tissue then homogenized in 0 . 05% Triton X100 in 5% FBS-F12 , diluted in 5% FBS-F12 , and plated for bacterial counts . For treatment , groups of mice were infected with P . aeruginosa strain 6294 , treated topically daily for 5 days with 5 µl of a solution of 100 µg gentamicin/ml , along with 50 µl of 1 M 4-IPP injected IP . The 4-IPP was solubilized in DMSO/corn oil [23] . Bacteria were quantified 5 days after infection by homogenizing and plating the corneal tissue . Mice were infected with P . aeruginosa strain 6294 and 1 µg recombinant MIF ( rMIF ) . applied topically in 5 µl at the time of infection and subsequently every 6 h after the infection . Corneas were harvested after 48 h of infection , treated with 300 µg gentamicin/ml for 60 min , washed , homogenized with F12 medium containing 5% FBS and 0 . 05% Triton X100 , diluted and plated . Eyes were enucleated from euthanized mice and fixed in 4% paraformaldehyde then embedded in paraffin . Four µm sections were cut , and stained with hematoxylin-eosin to visualize tissue morphology following previously used techniques [30] . Primary corneal epithelial cells [31] were grown in 6-well plates in either keratinocyte SFM ( Invitrogen ) medium supplemented with antibiotics , EGF , and pituitary gland extract or in MEM/F12 mix supplemented with EGF , insulin , DMSO , and cholera toxin [31] . Confluent monolayers of corneal cells were treated for 1 h with 4-IPP then infected with different strains of P . aeruginosa ( PAO1 , 6294 ) at a MOI of 25 for 1 h . Following treatment for 1 h with 300 µg gentamicin/ml , cells were re-supplemented with growth medium . The keratinocyte SFM growth medium was supplemented with penicillin , streptomycin , EGF , and pituitary gland extract . 6-iodo-6-phenylpirimidine ( 4-IPP ) was used to inhibit MIF activity in vitro and was maintained during the infection and gentamicin treatments . Different concentrations of 4–IPP were used , ranging from 100 µM to 400 µM . The tissue culture supernatants were collected at 6 h and 24 h after infection and total cell lysates were prepared in RIPA buffer ( 50 mM Tris-HCl , pH 7 . 5 , 150 mM NaCl , 1% Triton X-100 , 1% sodium deoxycholate , 1% SDS , and Complete Mini Protease Inhibitor Cocktail ( Roche Diagnostics ) . P . aeruginosa strains 6294 or PAO1 were grown to mid-log phase in TSB for 2 h at 37C and used for invasion assays . Bacterial cells were washed and suspended in F12 medium . Human primary corneal epithelial cells were grown to confluence in 6 well plates in F12 medium and infected with the P . aeruginosa strains at a MOI of 25 for 3 h . After the infection , cells were washed with F12 , then treated with 300 µg or 400 µg gentamicin/ml for 1 h to kill the extracellular bacteria depending on the activity of the antibiotic . After washing away the antibiotic , cells were lysed in 1% Triton X 100 in MEM and the lysates were diluted then plated on cetrimide plates to enumerate intracellular bacteria . Commercially available ELISA assays ( R&D Systems ) were used to determine the levels of cytokines ( IL-8 , IL-6 ) produced by P . aeruginosa infected primary human corneal cells . Mouse cytokines were measured using a Meso Scale Discovery ( MSD ) multiplex 7-spot electrochemiluminescence ( ECL ) assay and outputs measured by an ultra low noise charge-coupled device ( CCD ) Imager 2400 ( Meso Scale Discovery , Gaithersburg , MD , USA ) . The cytokines measured included interleukin ( IL ) -1β , IL-6 , IL-12p70 , IL-10 , IFNγ and the alpha chemokine neutrophil attractant and activator CXCL1/GRO ( also known as KC ) . The MSD ECL platform has been previously validated against cytokine standards recommended by WHO and U . K . National Institute for Biological Standards and Control ( NIBSC ) and by comparison to traditional ELISA [32] . Production of MIF by the primary corneal epithelial cells was reduced by MIF-specific siRNA treatment . Briefly , cells were transfected with MIF-specific or control siRNA ( On-target plus SMART pool; Dharmacon ) by incubating with a transfection mixture that consists of 50 nM siRNA ( Dharmacon ) and 6 µl Oligofectamine ( Invitrogen ) in 1 ml of Opti-MEM ( Gibco ) for 48 h , after which the cells were maintained with regular growth medium . Epithelial cells were infected with P . aeruginosa ( strain 6294 or strain PAOI ) 48 h to 72 h after siRNA treatment . A non-siRNA treated control was also included to determine if there is an effect on inflammatory gene expressing from the control siRNA . Murine PMNs were purified from bone marrow by flushing the cells out of the tibias and femurs with 3 ml HBSS ( Hanks balanced salt solution ) buffer ( Invitrogen ) . Cells were pelleted by centrifugation at 380 g for 10 min at tabletop centrifuge; resuspended in 1 ml HBSS and overlayed on the top of a Histopaque gradient composed of 3 ml Histopaque 1119 ( Sigma ) and 3 ml Histopaque 1077 ( Sigma ) . Cells were centrigufed for 30 min at 700 g , room temperature , without a brake . Neutrophils were collected from the interface of the Histopague 1077 and 1119 , transferred to a fresh tube and washed three times by adding 10 ml og HBSS . Viable cells were enumerated by tryptan blue exclusion technique . Opsonophagocytic assays were performed by mixing 100 µl of 2 . 5×106 PMN with 100 µl of 400 fold dilution of P . aeruginosa grown in TSB to reach mid-log phase as estimated by OD 650 reading of 0 . 4 . 100 µl of Rabbit complement sera ( Sigma ) and 100 µl of freshly harvested mouse sera from either C57Bl6 or MIF KO mice were added to the sample mixture as appropriate . The reaction was incubated for 90 min at 37°C with rotation . After completion of the incubation an aliquot was taken and lyzed with 0 . 05% Tween 20/TSB . Bacteria were enumerated by plating on P . aeruginosa selective cetrimide plates . | Pseudomonas aeruginosa can induce infections that lead to a rapid loss of visual function . The current therapy includes antibiotic treatment which reduces the bacterial burden; nevertheless , tissue damage occurs as a result of an uncontrolled inflammation . Therefore , new therapeutic approaches are needed that will combine antimicrobial treatment with anti-inflammatory treatment . Our experiments have uncovered that mice deficient for macrophage migration inhibitory factor ( MIF ) recovered from acute bacterial infection more efficiently than wild-type control mice . This improvement was manifested as improved bacterial clearance , and decreased inflammatory responses in the MIF knockouts when compared to P . aeruginosa-infected wild-type control mice . We find that treatment of the infected mice with small molecule inhibitor of MIF activity after the onset of the infection promoted recovery from disease . This approach will facilitate generation of novel treatment strategies for bacterial keratitis . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"immunology/immunity",
"to",
"infections",
"immunology/innate",
"immunity"
] | 2010 | Inhibition of Macrophage Migration Inhibitory Factor Ameliorates Ocular Pseudomonas aeruginosa-Induced Keratitis |
In allosteric regulation , an effector molecule binding a protein at one site induces conformational changes , which alter structure and function at a distant active site . Two key challenges in the computational modeling of allostery are the prediction of the structure of one allosteric state starting from the structure of the other , and elucidating the mechanisms underlying the conformational coupling of the effector and active sites . Here we approach these two challenges using the Rosetta high-resolution structure prediction methodology . We find that the method can recapitulate the relaxation of effector-bound forms of single domain allosteric proteins into the corresponding ligand-free states , particularly when sampling is focused on regions known to change conformation most significantly . Analysis of the coupling between contacting pairs of residues in large ensembles of conformations spread throughout the landscape between and around the two allosteric states suggests that the transitions are built up from blocks of tightly coupled interacting sets of residues that are more loosely coupled to one another .
Allosteric transitions , in which binding of an effector molecule to one site of a protein is coupled to a conformational change at a distant site , are fundamental to biological regulation . Although the first models were proposed more than 40 years ago [1] , [2] , developing a mechanistic understanding of allostery continues to be an active and vigorous area of research [3] , [4] . For a small number of allosteric proteins , X-ray crystal structures of ligand bound and ligand free states have illuminated the structural transitions underlying allostery [5]–[10] . However , the small number and static nature of these structures present several important challenges for structural biology that may be approached using computational methods . First , it may be possible to predict the structure of the one allosteric state starting from the structure of the other state . Meeting this challenge requires both an efficient method for conformational sampling in the neighborhood of the starting state and a sufficiently accurate energy function . Predicting the bound state from the unbound state is more challenging because it requires solving both the docking problem and the allosteric conformational change problem simultaneously . Predicting the unbound state from the bound structure is more straightforward and hence is a natural first step toward addressing the general prediction challenge . A successful approach would be extremely useful for predicting the conformational changes that occur in an allosteric protein for which only the structure of the bound state is available . A second challenge is to determine the mechanisms controlling allosteric regulation by identifying how individual residues are involved in allosteric transitions . Normal mode analysis of elastic network models [11]–[13] , a nonlinear elastic model [14] , network modeling of contact rearrangements [15] , and statistical coupling of local unfolding [16] , [17] have all been applied to protein structures to investigate mechanisms of conformational switching . These methods work best for identifying global motions , geometrical differences , or residue stability . NMR and other data suggest that most allosteric proteins are essentially two state systems , with bound and unbound states , but not intermediate states , populated at equilibrium [18] , [19] . Since states intermediate between the observed bound and unbound states are higher in free energy and cannot be readily observed experimentally , it is difficult to map the free energy landscape between the two states using experimental methods . One computational approach has been to use a multiple basin model to map the free energy landscape and approximate the transition between states [20] , [21] . However , this method only considers Cα atoms and utilizes knowledge of the structural end points as references in the potential function . Insight into residue couplings has come from studies of evolutionary covariance [22]–[24] , but this method can only be applied to systems with a large and diverse set of sequences . All-atom molecular dynamics simulations [25]–[28] can show residue couplings in great detail , but only when conformational transitions occur in the nanosecond timescale . The Rosetta high-resolution structure prediction methodology [29] has shown considerable progress in the related problem of predicting the structure of a protein based on the structure of a homologue [30] . The recently developed “rebuild and refinement” sampling methodology combines complete remodeling of the protein structure in specific regions [30] with global optimization of the entire protein structure using the Rosetta all-atom refinement protocol and energy function [29] . Because of the stochastic nature of the search , and the very large number of local minima on the rugged all-atom landscape , different models end up in different minima and these collectively create a map of the energy landscape in the neighborhood of the starting structure . Previously , this high-resolution refinement has been applied with the assumption that there is a single state to find , and it remains unclear whether the method has sufficiently high resolution to distinguish between two low-energy conformations in an allosteric protein . Here we employ Rosetta to address the twin challenges of allostery: prediction and mechanism . We apply the high-resolution refinement method to the problem of finding an alternative conformation of a protein , which in this case represents the alternative allosteric state . Here we assume that multiple states exist , e . g . bound and unbound , and then ask whether Rosetta can identify the alternative state . We report that Rosetta can reproduce conformational transitions for three proteins in which significant allosteric structural changes occur , particularly when provided information on which regions change the most in the allosteric transition . Exploring the energy basins near each starting structure identifies state-dependent residues that control protein function . Mapping the energy minima suggests that energetically coupled residue pairs switch together in groups ( blocks ) that are weakly coupled to each other .
We began by testing the extent to which the Rosetta high-resolution structure prediction methodology can predict the ligand free structure of an allosteric protein starting from the structure of the ligand bound form . We focus here on three allosteric proteins that undergo significant conformational changes upon effector binding: CheY , Integrin αL I-domain , and Ras . We initially selected 8 proteins ( see Table 1 ) but restricted our efforts to these three proteins for the following reasons . Three of the others involved relatively small loop rearrangements induced directly by a ligand rather than global conformational changes induced by an allosteric effector . In the SH2 domain and FixJ , the energy difference between conformational states was too small for the Rosetta energy function to identify the correct conformation , while β-lactoglobulin involved a single loop difference where the deep energy minimum near the alternative structure wasn't sampled . The final two proteins , Troponin C and S100A6 , involved calcium-binding sites for which the electrostatic interactions proved hard to model with the Rosetta energy function ( Figure S1 ) . In this first set of calculations , all loop regions were stochastically rebuilt in the “rebuild” portion of the “rebuild and refinement” protocol described in ref [30] . 100 , 000 independent Monte Carlo “rebuilding and refinement” simulations were initiated from the bound conformations following removal of the ligand . Plots of energy vs root-mean-square deviation ( rmsd ) to the native structure ( left panel of Figure 1 ) show that the deep energy minimum surrounding the native structure is sampled to some extent for Ras and CheY , as indicated by a minimum about 1 Å rmsd ( the typical noise within a state ) from the unbound state . However , this is not seen for the I-domain because regions with secondary structure differ in the two crystal conformations but were not allowed to be rebuilt in our initial calculations . To make the sampling problem more tractable while modeling secondary structure movements , we limited the rebuilding step in the “rebuild and refinement” protocol to loop and secondary structure regions that significantly change structure in going from the bound to the unbound state ( the entire protein is allowed to move in the following all-atom refinement step – see methods ) . The 20 lowest-energy structures were clustered based on their pairwise rmsd and the lowest-energy structure from the largest cluster was compared to both the starting and alternative structures . For three proteins ( CheY , the αL I-domain , and Ras ) , the lowest-energy structure of the largest cluster was closer to the alternative conformation than the initial structure , and energy versus rmsd plots reveal an energy minimum at the unbound conformation ( center panel of Figure 1 ) . Additionally , the largest cluster of the 20 lowest-energy structures contained at least 4 models , suggesting that sampling is converging toward the alternative conformation . That is , with the specification of the regions in which major conformational changes take place , the rebuild and refinement protocol can sample the alternative state and the energy function has sufficient accuracy to distinguish the unbound state based on its lower energy . In addition to identifying low-energy structures that are near the crystal conformation of the unbound state , subregions with the largest conformational difference between states were predicted to within an accuracy of between 0 . 3–3 . 4 Å ( Cα-rmsd ) to the alternative state ( indicated by black arrows in Figure 1 , and Table S1 ) . The structural changes in CheY involve a shift of helix α1 and rearrangements of the loops L7 & L9 near the FliM binding pocket ( indicated by ** ) . Removal of a disulfide bond ( allosteric effector indicated by * ) in the αL I-domain that mimics the activated state allows the α7 helix to shift upward more than 6 . 5 Å and the loop between strand β5 and helix α6 to move toward the active conformation of the ICAM-1 binding site ( indicated by ** ) . In Ras , loop L4 moves away from the allosteric effector ( located at * ) and toward the alternative state , and the helix α2 near the protein-binding site ( indicated by ** ) is formed , although it has not fully moved into position . Because of high intrinsic variability in loop regions , we independently measured the RMSD only over the regular secondary structure elements , as described in Supplemental Table S2 and the accompanying description of the methods . In all three cases , the secondary structure elements were predicted on average even better than the overall structures , and the only regions of the secondary structure which remained closer to the starting structure than the alternative structure were those that differed very little between the states to begin with . Thus , Rosetta is most successful in predicting structural changes in secondary structure elements . The crystal conformations of both states show a number of structural differences . Although many individual residues change conformation or contacts when an allosteric protein switches between states , only a small number of these changes may be critical to conformational switching [26] . To identify critical changes , we generated a set of 500 structures near each crystal structure by using Monte Carlo methods to perturb the backbone angles slightly and optimize side chain rotamer conformations , followed by energy minimization of each structure . We first identified pairs of residues for which the mean difference in pairwise interaction energy ( prE ) was greater than 1 Rosetta energy unit between the 500 structures in the two ensembles surrounding each state . Since these contacts differ consistently between conformations in the two states , we call them “state dependent” . Averaging interaction energies over conformations in the two states eliminates the set of contacts that differ between the two structures not because of the change in conformational state but because of differences in crystal packing interactions . The left panel of Figure 2 shows the state dependent prE differences ( orange ) and the remaining ( non state dependent ) differences ( blue ) mapped on to the three-dimensional structure of CheY , the I-domain , and Ras . CheY , the I-domain and Ras contain 128 to 180 residues , 27 to 82 of which formed pairwise interactions that had different energies in the two crystal structures . However , of these , only 10 to 20 formed state dependent interactions according to analysis of the ensemble of states ( Table 2 ) . Random mutagenesis [31]–[37] and mutations found in clinical samples [38]–[41] have identified a number of residues that alter protein function in the three proteins . Mutations identified by site-directed mutagenesis studies were not included since they are designed to target regions believed by the researchers to be important , which would cause an undesirable bias for our purpose . As shown in Table 2 , there are a higher fraction of residues important for function among the residues with state dependent energy differences than in the protein as a whole . On average , using ensembles to identify state-dependent residues provided a 1 . 9-fold enrichment in the number of function-altering residues . A lesser ( 1 . 4-fold ) enrichment was observed if the crystal structure differences were used to identify function-altering residues . We also identified state dependent side chain χ1 angles ( dihedral angle rotation around the Cα–Cβ bond ) based on mean differences in the χ1 angle between ensembles ( right panel of Figure 2 ) . Ensemble calculations identified 5 , 19 , and 11 residues ( CheY , the I-domain , and Ras ) with mean side-chain angle ( χ1 ) differences greater than 46° between states ( Table 2 ) . Comparison between the calculated χ1 differences and the experimental data showed the state dependent residues contain a higher fraction of function-altering residues than in the protein overall ( Table 2 ) . To examine how pairwise interactions are coupled during switching between the states , we generated models starting from the unbound state to map the neighboring landscape more thoroughly . Maps of the energy landscapes for CheY , the I-domain and Ras were created by combining the “rebuild and refinement” calculations starting from the bound and unbound structures ( left panel of Figure 3 ) . Each point on this landscape represents a single model , the axes are the rmsd values to the starting and alternative structures , and the colors represent the all-atom energy , graded on a continuum from lowest ( blue ) to highest ( red ) . A clear minimum is evident in the vicinity of the unbound state in all three cases , as indicated by a cluster of low-energy structures near 1 Å rmsd from the unbound state and over 1 Å rmsd from the bound state . Each structure on this landscape represents a distinct local minimum—the lowest energy structure sampled in an individual simulation . The two-dimensional view of the energy landscape suggests we have sampled the conformational space of both states and have reasonable coverage of intermediate conformations . To determine what residues switch conformational states together , we evaluated the association between pairwise contacts ( see methods ) . Some residues are strongly correlated and evidently switch states together whereas others switch independently . The correlated pairwise interactions appear as blocks when grouped using hierarchical clustering ( middle panel of Figure 3 ) . Within a block , the pairwise interactions show a stronger association than between blocks . In the context of the three-dimensional protein structure , the blocks comprise collections of residues that are often physically close to one another ( right panel of Figure 3 ) . Different blocks are often associated with different functions . In CheY , the cyan block includes highly conserved amino acids ( D12 , D57 , and K109 ) involved in phosphorylation and regulation of this receiver domain [42] , [43] . The magenta block contains residues E89 and Y106 , which play critical roles in conformational switching through CheZ-mediated dephosphorylation [44] and binding to the flagellar motor switch , FliM [45] , [46] . These two blocks are also related to functional regions observed in a previous study of internal dynamics with NMR [47] . In the αL I-domain , the blocks of coupled residues divide into three groups , which roughly map out a connection path between helix α7 ( cyan ) and the ICAM-1 binding site ( residues within the magenta block such as D127 & L205 ) . The yellow block that connects these regions includes residues from the β6-α7 loop and the hydrophobic pocket proposed to be responsible for the ratchet-like conformational switching [48] . In Ras , the magenta and yellow colored blocks contain residues in the helical-loop segment known as switch II [49] , which is directly involved in conformational switching between the active and inactive states . The cyan colored block contains contact pairs within the hydrophobic core that is highly conserved among Ras family proteins . This block is comprised of a set of coupled pairs that span the core β-sheet , connecting one side of the protein to the other .
Using the Rosetta rebuild and refinement sampling methods , the bound states of three allosteric proteins were observed to relax to the lower energy unbound states . Accurate prediction of the unbound state is facilitated by focusing sampling on the loops and secondary structure regions that differ between states . The Rosetta energy function is able to identify the correct structure; the need to focus the rebuilding protocol on regions known to differ is consistent with previous observations that conformational sampling is the primary limiting factor in high-resolution prediction . Nevertheless , our successes provide evidence of useful progress toward predicting conformational changes in allosteric proteins when only the bound structure is available . These successes are indicated by a decrease in the overall Cα-rmsd between the low-energy model and the alternative state , as well as a substantial improvement in the Cα-rmsd between the low-energy model and the alternative state for the subregions that differ most between states ( Table S1 ) . The sampling strategy failed to explore conformational space near the alternative state in proteins with large conformational changes that involved the hinge motion of multiple helices . The Rosetta energy function is insufficiently accurate to identify the correct structure for proteins with subtle loop changes where the energy difference between states is likely quite small , or those with electrostatic interactions with divalent cations . These challenges emphasize the need for improvements in both the Rosetta energy function and sampling strategies for exploring conformational space . However , since predicting the unknown conformation of an alternative state remains an unsolved problem , even partial success in this direction is encouraging and suggests that this approach warrants further development . We calculated the mean differences between pairwise interactions and side chain χ1 angles in ensembles of low-energy models near each state . These calculations provide a way to screen in silico a large number of conformational differences to identify a smaller set of promising residues to target for further experimental investigation . As indicated by random mutagenesis and mutations found in clinical samples , the state-dependent residues are enriched in amino acids known to control function ( Table 2 ) . The positive correlation between predictions and experiments suggests that ensembles could be used to predict state-dependent residues to mutate in order to alter the regulation of conformational switching . For example , it may be possible to change the overall activity but not the specificity of a protein by mutating state-dependent residues that are not in either effector or active sites , but rather in the pathway between them . The state dependent contact pairs group into clusters ( blocks ) that are often nearby on the three-dimensional structure and correlated with specific functions . These clusters of residue pairs tend to switch together in conformations spread throughout the energy landscape between the starting and alternative states . Each switching group maintains a weak association to other blocks of residue pairs , and these blocks form a weakly coupled system that could pass information between more distance regions of a protein . We propose a new “block” model ( Figure 4C ) for allosteric transitions that is intermediate between a concerted model , where all structural changes are tightly coupled and conformational switching is completely cooperative ( Figure 4A ) , and a sequential or domino model , where binding of a molecule at one site causes a sequential propagation of changes across the protein in a defined pathway ( Figure 4B ) . This suggestion is conceptually similar to the previous suggestion , based on dynamics simulations , that protein conformational changes [20] , including those the occur due to ligand binding [21] , can occur via a pathway that involves multiple basins . Because the two methods have been applied to different proteins , and because the data that suggests multiple intermediates is of a different nature , however , it is difficult to compare the details of the proposed intermediates . The high-energy states of all three models in Figure 4 are not readily observed experimentally . However , our model suggests that stabilizing the energetically coupled residues in one conformational state would lower the energy of that intermediate state to the point where it might be observed . The block model is physically plausible in that sets of residues that pack together would be expected to be highly correlated and switch states cooperatively , while more weakly coupled to residue clusters at distant sites . Allostery in this model is a result of the ( weak ) coupling between clusters of tightly interacting residues: a switch in state at a first cluster alters the energetic balance between alternative states at other clusters . Our approach differs in both methodology and conclusions from previous computational methods of studying allostery [11]–[13] , [22] , [23] , [50] . It is particularly instructive to compare our approach to previous work using all-atom molecular dynamics . A clear disadvantage of our method is that since we do not simulate dynamics , we can obtain no explicit information about trajectories , dynamics , or kinetics . We cannot observe pathways directly . On the other hand , our approach has two clear advantages . First , each data point is from a completely independent Monte Carlo Minimization simulation , hence observed correlations between contacts and other properties cannot be attributed to lack of independence in sampling ( as might be the case for different snapshots from a long MD trajectory ) . Second , each data point represents a relatively deep local minimum ( the lowest energy point found in the MCM simulation ) , and hence associations between residues may be stronger than in higher energy states—the higher the energy , the larger the noise due to energy fluctuations . Our approach focuses on the energetic coupling between interactions in allosteric transitions rather than the dynamic coupling .
To test whether it is possible to predict a ligand-induced conformational change in allosteric and non-allosteric proteins , we selected a set of 8 pairs of ligand bound and ligand free protein structures from the Protein Data Bank [51] ( Table 1 ) . Coordinates for the starting structure of the αL I-domain were modified according to [52] . The selection criteria were the availability of structures of ligand bound and ligand free forms , a significant structural rearrangement ( Cα-Cα differences >3 . 5 Å ) between the two forms , and size less than 200 amino acids to ensure the tractability of the search problem . All crystal structures had a resolution ≤ 2 . 5 Å , and with the exception of three bound structures ( PDB ID: 1b0o , 1f4v , 1d5w ) the structures were ≤ 2 . 0 Å resolution . Test cases were grouped into categories based on their conformational change and their structural classification ( all-α , all-β , mixed α/β or α+β ) [53] . These categories allowed us to evaluate the method's ability to predict both localized and allosteric conformational changes with high-resolution accuracy , as well as to consider how a protein's fold affected the conformational sampling and prediction accuracy . Starting models were created from the crystal structures by fixing the bond lengths and angles at chemically ideal values , and representing all atoms explicitly using internal coordinates ( φ , ψ , ω , χ1 , χ2 , χ3 , & χ4 ) . Following idealization , all models were minimized as a function of all backbone and side chain angles using the Davidon-Fletcher-Powell ( DFP ) algorithm [54] . The structure prediction protocol is based on the “rebuild and refinement” method that is outlined in detail elsewhere [30] . Briefly , the overall approach consisted of three parts , ( 1 ) generating structural diversity , ( 2 ) optimizing the side chain position for every residue , and ( 3 ) minimizing all atoms in the protein . In the rebuild step , structural diversity was created by replacing backbone torsion angles of the loops with one or three or nine consecutive residues “fragments” from non-homologous structures in the Protein Data Bank . Initially , all loop regions were remodeled . Based on insufficient sampling of the conformational space near the alternative structure , we then chose to rebuild continuous sequences of 4 or more residues where the pairwise Cα-Cα difference was greater than 1 Å ( >1 . 5 Å for Troponin C and S100A6 ) . These chosen regions were randomly selected during a simulation to be remodeled using the fragment insertion protocol as described in [55] . Briefly , a chain break ( “cut” ) was made to the remodeled segment at a randomly chosen position within the region . Randomly chosen nine-residue , three-residue , or one-residue fragments were inserted into randomly chosen positions in the region being rebuilt , and the Metropolis Monte Carlo criterion was used to accept or reject the newly inserted fragment . To maintain the connectivity of the protein chain , cyclic coordinate descent [56] was used to close the chain break at a stochastically selected position of the region rebuilt . In the refinement protocol , all of the backbone and side chain atoms in the protein are explicitly represented . The entire protein is allowed to move through a series of steps that introduce a random perturbation to the backbone atoms , and then optimize the backbone and side chain coordinates for the new backbone position ( see [30] for a detailed description of the types of random perturbations and the move sequences ) . Optimal side chain conformations for each residue were selected from the Dunbrack rotamer library [57] . After the backbone perturbation and side chain optimization , the energy of the entire structure was minimized as a function of all backbone and side chain dihedral angles using the DFP algorithm . The new angles were accepted or rejected using the standard Metropolis criterion between the energy of the minimized structure and the initial conformation prior to the random perturbation . This entire cycle of rebuild and refinement was repeated ∼100 , 000× , generating ∼100 , 000 low-energy conformations of each protein in the test set , and exploring a broad set of local minima within the energy landscape that are both near and far from the starting conformation . The top 20 low-energy models were selected from the set of ∼100 , 000 simulations and clustered based on a structural similarity using an algorithm that has been described previously [58] . Briefly , pairwise Cα-rmsd comparisons were made between all 20 models using a threshold of 1 . 0 Å to define neighboring structures . The structure with the largest number of neighbors within this threshold was considered to be the center of the first , largest cluster . This cluster center and its neighbors were removed from the population and the pairwise comparison was repeated until all structures in the set were examined . The lowest-energy structure in the cluster with the largest number of neighbors was selected for comparison to the starting and alternative crystal structures . The crystal structure was taken as the starting template for creating an ensemble of near-native models . Bond lengths and angles were fixed at ideal values and each structure was minimized . Following idealization and minimization , all proteins within the test set were subjected to the Monte Carlo plus minimization ( MCM ) protocol to generate 500 models in the vicinity of the crystal conformation . The MCM strategy uses the all-atom , high-resolution refinement protocol that has been described previously [29] , [59] . Briefly , the MCM strategy consists of small , random perturbations to the backbone torsion angles , optimization of the side-chain rotamer conformations for the new backbone angles , and minimization of the backbone and side chain degrees of freedom using the DFP algorithm . The pairwise interaction energy ( prE ) was computed from a subset of terms in the Rosetta energy function including the Lennard-Jones attractive and repulsive , hydrogen bonding , solvation , and a statistical term ( “pair” ) that approximates electrostatics and disulfide bonds , . Mean prE differences greater than 1 Rosetta energy unit between the ensembles of 500 near-native models were considered to be state dependent . The χ1 side-chain angle ( dihedral rotation about Cα–Cβ bond ) was computed for all residues except alanine and glycine . Mean χ1 differences [60] greater than 46° [61] between the ensembles of 500 near-native models were considered to be state dependent . State-dependent predictions were compared against residues that have been experimentally found to alter protein function by random mutagenesis , or mutations found in clinical samples . Function-altering mutations identified by site-directed mutagenesis studies were excluded since they are designed to target regions believed by the researchers to be important , which would cause an undesirable bias for our purpose . The fraction of residues involved in either pairwise interactions or side chain differences that are known to alter protein function was computed for the whole protein ( ftot ) , the differences in the crystal structures ( fxtal ) , and the state-dependent residues in the ensembles ( fens ) . The ratio of fractions ( fxtal/ftot and fens/ftot ) was calculated to determine the enrichment of function-altering residues present in the computed differences versus the whole protein . The pairwise interaction energy ( as described above ) was computed for all residue pairs in both states of CheY , the αL I-domain , and Ras . Pairwise coupling was evaluated by examining the pairs that changed contact between states . These changes were considered to be binary and involved going from interacting ( prE <–1 . 25 Rosetta energy units ) to non-interacting ( prE >−0 . 5 Rosetta energy units ) . Calculations were performed on all models from the two sets generated by starting from the bound and unbound states and running the “rebuild and refinement” protocol to explore the neighboring energy landscape . Associations between pairwise interactions were computed from the φ coefficient , where . χ2 is the chi-square statistic for testing independence ( , where O and E are the observed and expected frequency ) and N is the number of observations . Associations were clustered using the complete-linkage , hierarchical clustering algorithm implemented in the R statistical package ( http://www . r-project . org/ ) . All plots were made with gnuplot ( http://www . gnuplot . info/ ) or the R statistical package ( http://www . r-project . org/ ) . Images of protein structures were generated using PyMOL [62] . The Rosetta source code is available without charge for academic users from http://depts . washington . edu/ventures/UW_Technology/Express_Licenses/rosetta . php | A common means of biological regulation is allostery , in which an effector molecule binds to one site on a protein and induces a conformational change which changes activity at a distant active site . Frequently high resolution structures are determined for one state of an allosteric protein but not the other . To probe the allosteric conformational changes in such cases , we describe a computational method for predicting the structure of one allosteric state of a protein starting with knowledge of another . Our method also provides a detailed map of the free energy landscape traversed in an allosteric transition and reveals the coupling between interacting residue pairs that underlies the transition . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"computational",
"biology/protein",
"structure",
"prediction",
"biophysics/biomacromolecule-ligand",
"interactions"
] | 2009 | Computation of Conformational Coupling in Allosteric Proteins |
Both theory and experiments have demonstrated that sex can facilitate adaptation , potentially yielding a group-level advantage to sex . However , it is unclear whether this process can help solve the more difficult problem of the maintenance of sex within populations . Using experimental populations of the facultatively sexual rotifer Brachionus calyciflorus , we show that rates of sex evolve to higher levels during adaptation but then decline as fitness plateaus . To assess the fitness consequences of genetic mixing , we directly compare the fitnesses of sexually and asexually derived genotypes that naturally occur in our experimental populations . Sexually derived genotypes are more fit than asexually derived genotypes when adaptive pressures are strong , but this pattern reverses as the pace of adaptation slows , matching the pattern of evolutionary change in the rate of sex . These fitness assays test the net effect of sex but cannot be used to disentangle whether selection on sex arises because highly sexual lineages become associated with different allele combinations or with different allele frequencies than less sexual lineages ( i . e . , “short-” or “long-term” effects , respectively ) . We infer which of these mechanisms provides an advantage to sex by performing additional manipulations to obtain fitness distributions of sexual and asexual progeny arrays from unbiased parents ( rather than from naturally occurring , and thereby evolutionarily biased , parents ) . We find evidence that sex breaks down adaptive gene combinations , resulting in lower average fitness of sexual progeny ( i . e . , a short-term disadvantage to sex ) . As predicted by theory , the advantage to sex arises because sexually derived progeny are more variable in fitness , allowing for faster adaptation . This “long-term advantage” builds over multiple generations , eventually resulting in higher fitness of sexual types .
The pervasiveness of sex , given its varied and potentially large costs , is highly perplexing [1]–[10] . Numerous hypotheses have been proposed and sophisticated theoretical analyses have helped to define the conditions under which particular hypotheses may apply [3] , [11]–[14] . Despite the importance of this problem , rarely have the hypotheses been tested by examining how key factors affect the evolution of sex [15]–[18] . Over a century ago , Weismann [19] , [20] argued that sex might be beneficial because it helps generate the variation necessary for adaptation . While intuitively appealing , the idea is not necessarily correct as sex will increase the variance in fitness only if there is a preponderance of “negative genetic associations” such that good alleles are often found in genomes with bad alleles . It was later realized that such negative associations may develop under certain forms of nonlinear selection ( as occurs when approaching an adaptive optimum [21]–[25] ) or , perhaps more importantly , due to an interaction between directional selection and drift , known as the Hill-Robertson effect [6] , [26] . For these more sophisticated reasons , Weismann's original conjecture is thought to be valid and is considered by many as the leading explanation for the evolutionary function of sex [27] . Rigorous theory shows that sex can facilitate adaptation [21] , [24]–[26] , [28] , but the conditions under which this will translate into a net selective advantage for sex itself are more limited [21] , [24] , [29]–[34] , especially given the infamous costs of sex [1] , [3] . Indeed , a number of studies have demonstrated that sexual populations adapt faster than asexual populations [7]–[10] , [35] , [36] . Such studies imply a population- or group-level advantage to sex , though none of these studies directly competed sexual and asexual populations against one another during adaptation . Consequently , it is impossible to know whether any benefit to sex with respect to adaptation would have been outweighed by its immediate costs . More importantly , group-level advantages to sex cannot be used as evidence for the maintenance of sex within populations , as emphasized by John Maynard Smith [1] and George Williams [37] . Better support for adaptation providing a “gene-level” advantage to sex comes from survey studies showing that recombination tends to increase as an incidental by-product of directional selection on other traits [38] , [39] . However , the evolution of recombination is not the same as the evolution of sex . The intrinsic costs of sex and recombination differ , and even ignoring these costs , theory shows that selection on recombination is often not an accurate predictor of selection on sex because of segregation effects [29] , 40 , 41 . More evidence for adaptation favouring genetic shuffling comes from a recent study in C . elegans showing adaptation favours outcrossing over self-fertilization [42] . Though a related phenomenon , this is not direct evidence for the role of adaptation in maintaining sex . The contrast between selfing and outcrossing is not the same as the contrast between asex and sex because different types of genetic associations are involved . Further , the intrinsic costs of sex ( relative to asexuality ) differ from the intrinsic costs of outcrossing ( relative to selfing ) . Despite these important differences , the recombination and outcrossing studies offer indirect evidence that adaptation can select for sex . However , direct experimental evidence for adaptation favouring sex is lacking . Beyond the crucial step of empirically demonstrating the requirements necessary to cause an evolutionary increase in sex , a more thorough understanding requires identifying the population genetic mechanisms that drive the evolution of sex . A general theoretical framework divides the total selection on sex into components arising from “short-term” and “long-term” effects [30] ( see [41] , [43] for further discussion of these terms ) . The “short-term” effect of sex refers to the immediate fitness consequences of rearranging gene combinations . Sex does not directly change allele frequencies , but it does re-distribute alleles ( i . e . , breaks down genetic disequilibria ) . Whenever alleles interact to affect fitness ( i . e . , if there is dominance or epistasis ) , altering gene combinations will change fitness . For this reason , the mean fitness of sexual-derived progeny can differ from that of asexually derived progeny coming from the same set of parental genotypes . The short-term effect of sex results from alleles that promote sex being associated with different gene combinations than the alleles that promote asexual reproduction [10]–[12] . Regardless of whether there are gene interactions or not , the redistribution of alleles through sex can result in the variance of sexually derived offspring being different ( higher or lower ) than that of asexually derived offspring . If the sexually derived subpopulation has more variance than the asexually derived subpopulation , then the former will better respond to subsequent selection . Though sex does not immediately affect allele frequencies , it alters the genetic variance , which allows subsequent selection to cause allele frequencies to diverge between more versus less sexual lineages . The “long-term” effect of sex refers to selection on sex that results from genes that promote sex becoming associated with a different frequency of fitness-affecting alleles [10]–[12] . It is worth noting that the label “long-term” effect is somewhat misleading as long-term effects can arise over a single complete generation involving both reproduction and selection . While long-term effects can build in strength over multiple generations , it is not necessary to have hundreds of generations for this form of selection to alter the evolution of sex . There is a myriad of hypotheses for the evolutionary maintenance of sex , but they can all be interpreted as providing an advantage to sex through either short- or long-term effects [11] , [12] . Despite the importance of these general mechanisms to our understanding of selection on sex and the potential to study these effects by examining the effect of sex on the mean and variance in fitness , no empirical study has clearly linked the evolution of sex to either of these mechanisms . Here we examine the Weismann hypothesis by evaluating whether sex is favoured during adaptation to a novel environment . We do this by ( i ) examining whether sex increases in frequency during adaptation and ( ii ) measuring the difference in fitness between naturally occurring sexual and asexual genotypes at various points during the course of adaptation . Finally , we test whether the advantage to sex arises from a short- or long-term effect by examining the effects of sex on the mean and variance in fitness at several points over the course of adaptation . This allows us to test the prediction that sex is favoured through a long-term advantage [24] , [27] , [31] , [32] . To test Weismann's hypothesis at the within-population level , we used replicated experimental populations of the haplodiploid monogonont rotifer Brachionus calyciflorus . These rotifers are facultatively sexual , reproducing amictically at low densities but changing to mictic ( sexual ) reproduction in response to a chemical stimulus indicative of high density [44] . When stimulated , the amictic mothers produce daughters that develop into mictic females . Unfertilized mictic females produce haploid eggs that develop into males , and if young mictic females mate , her haploid eggs are fertilized and develop into resting eggs . Amictic females hatch from resting eggs when stimulated by environmental cues . Previous work with rotifers from this source population reveals there is substantial genetic variation in the strength of the stimulus needed to induce sex , thus allowing for the evolution of rates of sex [17] . Amcitic eggs develop within 1 d and the time the females start producing their first offspring is less than 24 h after hatching . Fertilized mictic eggs ( resting eggs ) from this population hatch spontaneously at a high rate under typical lab conditions ( between 1 and 5 d after they are produced; see Material and Methods and [45] ) . From these observations , we approximate the mean time to complete an asexual generation to be ∼1 . 5 d . The “sexual cycle” takes ∼6 d but involves two generations ( ∼1 . 5 for the production of mictic females and then ∼4 . 5 d for sexually derived offspring to hatch and mature ) . Given that the overall rate of sexual reproduction is low , the average generation time is expected to be closer to 1 . 5 d than 4 . 5 d . All of our replicate populations descended from a common natural source . However , 10 replicates came from subpopulations more recently adapted in the lab to one environment ( “Environment A” ) , whereas 10 other replicates came from subpopulations more recently adapted to another ( “Environment B” ) . The two environments differ in their algal food source and NaCl concentration . For our main experiment , 10 replicates ( 5 from each environment ) serve as control ( non-adapting ) populations and are maintained under the environmental conditions to which they had already adapted . The remaining 10 populations are transitioned to the alternative environment ( 5 replicates A→B; 5 replicates B→A ) ; we refer to these as “adapting” populations . This reciprocal experimental design offers the opportunity to infer the role of adaptation per se , rather than a particular environment , in affecting the evolution of sex . Population sizes are relatively large throughout the experiment ( N≅3 , 500–7 , 500 ) .
Over the course of 70 d ( ca . 45 asexual generations ) of evolution , there is clear evidence of adaptation in the populations that experience an environmental change . Population densities , which initially plummet during the transition to the alternate environment , increase to stable levels characteristic of well-adapted populations ( Figure 1 ) . Moreover , estimates of individual fitness show similar increases over time ( Figure 2 ) . In contrast , control ( non-adapting ) populations remain stable both in density and in fitness assay measures over this period . As predicted by the Weismann hypothesis , rates of sex increase during the period of rapid adaptation . Later , sex declines as adaptation slows , presumably reflecting the intrinsic costs of sex outweighing the diminishing benefits of sex as the opportunity for adaptation declines . These temporal changes in sex are evident in two separate measures of sex . First , we use the fraction of fertilized mictic eggs ( out of all eggs ) as an in situ measure of sexual investment ( fertilized mictic eggs are visibly distinct from other eggs ) . There is an obvious increase in the investment in sexual eggs during the period of rapid adaptation , followed by a decrease ( see Figure 1 legend for statistics ) . This pattern cannot be explained by density effects directly triggering sex as the observed changes in sex go in the opposite direction from the well-known pattern for this species in which high density induces sex [44] . In contrast to the adapting populations , the percentage of fertilized mictic eggs in the control populations shows little change . Our second measure is based on a controlled assay of the propensity for sex . Each week , 42 rotifers are isolated from each population and maintained individually under standardized conditions for three clonal generations . Third generation individuals are exposed to a specified concentration of a sex-inducing stimulus . We determine the fraction of individuals that are induced into sexual reproduction by this cue . In the adapting populations , we observe a significant increase in the propensity for sex during the early phases of adaptation , followed by a subsequent decline ( Figure 3 , see legend for statistics ) . In contrast , the propensity for sex declines monotonically in the control populations . On day 37 , a second set of 10 adapting populations ( five for each environment ) was initiated from the control populations . We refer to these as the “Set 2 adapting populations . ” Our data on this second set are less detailed and over a shorter period , but these populations also show a similar increase in sex . Several lines of evidence indicate that the changes in the propensity for sex ( Figure 3 ) are not due to a plastic response stimulated by moving into a new environment . First , the assays are always performed in the third generation after isolation into standardized conditions so these changes cannot be due to the immediate shock of changing environments . Second , the Day 0 data represent assays on third generation clonal descendants of rotifers that have just transitioned to the alternative environment . As there is no difference in sex between control and adapting populations at this initial time point , it is clear that sex is not a stress-induced response resulting from a mismatch between genotype and environment . An unlikely third possibility is that the stress of a novel environment accumulates over multiple generations to induce the delayed rise in sex observed in Figure 3 . As described in Figure S1 , we tested this by transferring rotifers to the alternative environment and propagating them clonally for an extended period as individual lineages to prevent changes due to selection . Compared to rotifers maintained in the original environments , there was no change in the propensity for sex either in the short term or after a 16-generation delay ( which corresponds to the same time period as the rise in sex observed in Figure 3 ) . The results of our in situ measure of “investment in sex” ( Figure 1 ) and our well-controlled “propensity for sex” assay ( Figure 3 ) are reasonably congruent for the adapting populations , but there is a puzzling inconsistency with respect to the controls . In the control populations , the “propensity for sex” declines monotonically , whereas the “investment in sex” is low and relatively constant . Because the strength of the sex-stimulating cue used in the “propensity for sex” assay is much stronger than the expected strength of the cue experienced in situ in the control populations based on their densities , we do not expect to see the same magnitude of change in the two types of measures . Nonetheless , some corresponding decline in the “investment in sex” measure is expected but not observed . As the data are somewhat noisy , it is conceivable that we simply lack the statistical power to detect a decline . In this system , as in most others , the products of sexual reproduction are not phenotypically identical to those of asexual reproduction ( i . e . , fertilized mictic eggs are different than amictic eggs ) . Consequently , it is a concern whether changes in sex are actually due to selection for sex rather than a by-product of selection for some correlated feature . However , this alternative interpretation is inconsistent with our results . The parallel responses of adapting populations in both environments , as well as the pattern of temporal change within environments ( increases during adaptation followed by decreases as fitness plateaus ) , indicate that neither environment favours resting ( fertilized mictic ) eggs per se . Nonetheless , it would be more compelling to show differences in fitness between sexual and asexual genotypes to provide direct evidence of selection on the genetic consequences of sex . For this purpose , we sample fertilized mictic and amictic eggs weekly from each population . Rotifers are propagated individually before we measure lifetime reproduction for multiple clones of the third generation of each genotype , allowing us to compare recently created sexual and asexual genotypes that all develop from the same type of egg . The results , representing fitness measures on ∼22 , 000 individuals , are presented in Figure 2 . In the control populations from both environments , genotypes derived from sexual reproduction are much less fit than those from asexual reproduction . In contrast , when populations transition to a new environment , we initially find no difference in fitness between sexually and asexually derived genotypes . As adaptation proceeds , sexually derived genotypes become significantly more fit than asexually derived genotypes ( days 21–35 for A→B; days 21–49 for B→A ) . As populations approach their new fitness equilibrium , the pattern reverses again and the sexual load characteristic of well-adapted populations begins to re-emerge ( days 42–70 for A→B; days 63–70 for B→A; see Table S1 for statistical comparisons between sexuals and asexuals at each time point ) . The assays described above reflect differences in fitness between naturally occurring sexually and asexually produced offspring . The genotypes isolated for this assay are appropriately biased in that sexual genotypes will tend to descend from lineages with more sex in their history than asexual genotypes . This difference between the genealogical histories of naturally occurring fertilized mictic and amictic eggs is what yields a measure of the net effect of sex , but this also precludes a more detailed understanding of the population genetic mechanisms responsible . We cannot tell whether an observed advantage of sex results from the immediate benefit of genetic mixing ( “short-term advantage” ) or the accrued benefit of past selection on genetic variation released by previous bouts of sex ( “long-term advantage” ) . To differentiate between short- and long-term effects as mechanisms driving the evolution of sex , it is necessary to examine how sex affects the fitness of progeny from an unbiased set of parents . By comparing sexually and asexually derived offspring from random sets of parents , we can determine how sex affects the distribution of offspring fitness values without the confounding effects of past selection associated with sexually inclined lineages . By exposing a random sample of rotifers from each population to an extremely strong sex stimulus that induces sex at a very high rate across a wide array of genotypes [17] , we obtain sexually derived offspring from a largely unbiased sample of parents . We obtain asexually derived offspring from random samples of rotifers from each population kept at low densities . Eggs are isolated and maintained individually under standardized conditions for multiple clonal generations before replicate measures of lifetime reproduction are made for each genotype . ( This procedure is illustrated in Figure S2 , where it is contrasted with the assay procedure for measuring fitness from naturally occurring sexual and asexual genotypes . ) We perform this type of assay for the first set of adapting populations at two time points , sampling parents on day 33 ( shortly after the propensity for sex has peaked ) and day 67 ( when adaptation is near complete and the propensity for sex is in decline ) . For the second set of adapting populations , we sample parents somewhat earlier during the course of adaptation ( 16 and 30 d after their initiation , corresponding to days 53 and 67 on Figures 2 and 3 ) . We have analogous data for control populations for each of these time points . The distributions of sexually and asexually derived offspring fitnesses are shown in Figures S3 and 4; ratios comparing key properties of these distributions are shown in Figure 4 . Sexually produced offspring have lower mean fitness than asexually produced offspring ( t = −18 . 9 , df = 16 , p = 2 . 3×10−12 for adapting populations; t = −62 . 8 , df = 26 , p<2×10−16 for control populations; Day 67 data from the first set of adapting populations are not used in these comparisons as fitness has plateaued before this point ) . The lower average fitness of sexually produced offspring is predicted whenever non-additive gene action ( dominance and/or epstasis ) plays an important role in shaping patterns of genetic associations ( disequilibria ) [3] , [11] . Bad combinations of alleles that have been eliminated by past selection can be recreated by sex , reducing mean fitness , a phenomenon that can be thought of as a “sexual load” and is sometimes called “genetic slippage” [46] , [47] . Although the distributions of sexually derived offspring fitness have lower averages than the corresponding distributions for asexuals , the variances for sexuals are higher ( t = 17 . 0 , df = 16 , P = 1 . 1×10−11 for adapting populations; t = 6 . 1 , df = 26 , P = 1 . 8×10−6 for control populations ) . This increased variance associated with sex reflects the existence of negative genetic associations likely generated either by epistasis or Hill-Robertson effects [6] , [26] , [32] . Sex and recombination are expected to dissipate these disequilibria , resulting in an increase in genetic variance . The pattern of sex reducing the mean but increasing the variance , indicative of a short-term disadvantage but a long-term advantage to sex , is qualitatively similar in both adapting and control populations . Although the directions of the short- and long-term effects are the same between treatments , the relative magnitudes differ . In Environment A , sex reduces mean fitness by ∼30% in control populations but only by ∼20% in adapting populations ( before populations reach near complete adaptation ) . A similar effect occurs in Environment B , where sex reduces mean fitness by ∼45% in control populations but only by ∼20% in adapting populations . Thus , the short-term disadvantage of sex is ∼30%–50% smaller in adapting populations . This difference between adapting and control populations is supported by formal comparisons ( t = −3 . 8 , df = 16 , p = 0 . 001 for Environment A; t = −14 . 8 , df = 16 , p = 9 . 6×10−11 for Environment B ) . The long-term effect results from a difference in genotypic diversity in fitness created by sexual reproduction relative to that resulting from asexual reproduction . This is often discussed in terms of differences in variance . As described above , sexual genotypes are more variable in lifetime reproduction than asexuals in both adapting and control populations . The relative increase in variance due to sex is greater in adapting populations than in control populations ( t = −2 . 4 , df = 16 , p = 0 . 03 for Environment A; t = −5 . 4 , df = 16 , p = 5 . 9×10−5 for Environment B; Figure 4C , D ) . There are potential problems with using the variance as a measure of the long-term effect of sex when the mean fitness of sexually and asexually derived offspring differs . High variance of sexuals may result from the production of low fitness genotypes . The generation of such variants is not useful for adaptation and thus cannot contribute to a long-term advantage to sex . Rather , we are interested in whether sex tends to produce particularly good variants . For this purpose , we compare the average fitness of the top 10% of sexually and asexually produced genotypes ( Figure 4E , F ) . Sex generates significantly better genotypes in the top end of its fitness distribution than does asexual reproduction in adapting populations ( t = 3 . 06 , df = 16 , p = 0 . 007 ) , but the opposite is true in control populations ( t = −13 . 3 , df = 26 , p = 4 . 2×10−13 ) . Similar patterns are observed using the top 5% , 15% , or 25% ( see Figure S5 ) . Above , we have discussed fitness distributions for sexually and asexually derived offspring obtained in two different ways ( Figure S2 ) . First , we isolated naturally occurring fertilized mictic and amictic eggs , and thus , these two types of eggs came from lineages with different histories of sex ( Figure 2 ) . Second , we generated sexual and asexual offspring from random sets of parents ( Figures 4 , S3 , S4 , S5 ) . If we compare these assays at a similar time point during adaptation ( close to day 30 ) , there is a dramatic difference . In the first assay ( from non-random parents; Figure 2 ) , we find that sexually derived offspring have higher average fitness than asexually derived offspring . In the second assay ( random parents; Figure 4 ) , sexually derived offspring have lower average fitness . As discussed above , even though sex produces offspring that are less fit , on average , than asexually derived offspring , sex also generates some particularly high fitness genotypes . These genotypes contribute disproportionately to future generations , carrying alleles for sex with them . As a result of generating extreme variants and subsequent selection , alleles that increase sexual propensity become associated with alleles conferring adaptation . Consequently , naturally occurring sexuals eventually become more fit , on average , than asexuals because of this accrued benefit from past selection . The higher average fitness of sexuals observed in the first assay midway through the course of adaptation represents the long-term advantage realized . However , as time passes , beneficial alleles will eventually accumulate in less sexually inclined genotypes , and thus the advantage to sex will erode as populations approach an adaptive optimum and the influx of new beneficial alleles slows . The short-term disadvantages of sex , along with other costs of sex , can then drive an evolutionary decrease in sex .
Previous experiments have shown that sexual groups can adapt faster , thereby providing indirect evidence of a group-level advantage to sex ( at least in the absence of intrinsic costs ) [7]–[10] , [35] , [36] . For the first time , we demonstrate that the frequency of sex within a population rises over time during adaptation . These results are consistent with the idea that Weismann's hypothesis can provide an advantage to sex at the gene level that can be sufficiently strong to overwhelm the intrinsic costs of sex . Weismann's hypothesis and related theories [27] make a strong prediction that sex should be favoured during adaptation because of a long-term advantage , and we have found evidence supporting this mechanism . On the other hand , much of this body of theory [29] , [31]–[34] does not make a clear prediction with respect to short-term effects; when Hill-Robertson effects are responsible for negative disequilibria , short-term effects could be positive , negative , or zero , but models invoking non-linear selection to generate negative disequilibria predict negative short-term effects [24] . In our study short-term effects appear to be substantial . The reduction in negative short-term effects that seems to accompany the transition to a new environment ( possibly reflecting environment-specific epistasis ) is somewhat unexpected and reduces the threshold for a long-term advantage to create a net benefit to sex . Though our results provide direct support for the operation of the Weismann hypothesis , we have not shown quantitatively that the Weismann hypothesis alone can fully explain the observed evolution of sex . It is possible that other factors also contribute to these changes . Here we consider two alternatives , but our data do not provide strong support for either . In this system , as in most others , the products of sexual reproduction are not phenotypically identical to those of asexual reproduction ( i . e . , fertilized mictic eggs are different than amictic eggs ) . Consequently , some of the observed evolution of sex could be a by-product of selection for fertilized mictic eggs ( rather for genetic mixing ) . However , this alternative interpretation is inconsistent with our results . The parallel responses of adapting populations in both environments , as well as the pattern of temporal change within environments ( increases during adaptation followed by decreases as fitness plateaus ) , indicate that neither environment favours resting ( fertilized mictic ) eggs per se . Moreover , we have direct evidence that genotype , independent of egg type , is important during adaptation; naturally occurring , sexually derived genotypes are more fit than asexually derived genotypes even when both develop from the same egg type ( Figure 2 ) . A second factor of possible importance to our results is differential selection between the sexes [48]–[51] . In this system , ( sexual ) males are haploid , potentially allowing for more efficient selection on recessive beneficial alleles than can occur in the absence of sex . Under this hypothesis , we would expect that when we experimentally force individuals through the sexual cycle , the resulting offspring should , on average , be more fit than with asexual reproduction because of the extra sieve of haploid male selection that occurs incidentally during the process of creating sexual offspring . In fact , we observe the opposite; sexually derived genotypes from random sets of parents are less fit on average than asexually derived genotypes ( Figure 4 ) . This should not be taken as evidence that haploid selection has no effect at all , but rather it suggests that haploid selection does not play a strong role . The costs of sex are expected to be high in this system , and it is unclear whether the observed benefits can outweigh these costs . In this regard , it is worth considering three points . First , modifier alleles that increase the rate of sex by a small degree experience only a small fraction of the cost of sex [52] . Second , long-term benefits can be quite powerful , especially when the baseline rate of sex is quite low [30] , [52] , as it is in our system ( 5%–7% , Figure 1 ) . A modifier allele that slightly increases the rate of sex only suffers the cost of sex in those generations where it induces sex but enjoys the benefit of having created a good genotype for many generations . Third , the advantage gained by “high-sex” genotypes during adaptation is likely considerably larger than it appears . The observed advantage in fitness of naturally occurring , sexually derived genotypes over asexually derived genotypes during adaptation reaches 30%–50% ( Figure 2 ) , but this underestimates the difference in fitness between “high-sex” genotypes and “low-sex” genotypes . This is because the distinction between “high-sex” genotypes and “low-sex” genotypes with respect to degree of sex is quantitative; both types use both reproductive modes . Consequently , the naturally occurring fertilized mictic eggs will come from both “high-sex” and “low-sex” parental genotypes but be biased toward coming from the former . Conversely , the amictic eggs will come from both “high-sex” and “low-sex” parents but be biased toward the latter . Thus , the difference in fitness between genotypes isolated from fertilized mictic eggs versus those isolated from amictic eggs will clearly underestimate the true difference in fitness between “high-sex” and “low-sex” genotypes . The two environments used here were used in a previous study of the evolution of sex . The main result of that study was that higher rates of sex were maintained when populations experienced spatial heterogeneity in selection [17] . However , that experiment also provided a hint of the Weismann effect as even the spatially homogenous ( control ) populations showed an initial increase in sex followed by a decline on a time scale similar to that observed here . Because both environments were novel compared to the source population of rotifers , it is likely that the initial increase was due to an advantage to sex during adaptation to those environments . Though adaptation itself was not measured in that study , those results are consistent with what we have reported there . Despite its importance to theory , the effect of sex on the distribution of offspring fitness has been measured in only a handful of taxa [45] , [53]–[57] . In several of those cases [54] , [55] , [57] , sex has been observed to reduce the mean but increase the variance , suggesting that long-term advantages to sex may be reasonably common but in none of those previous cases were evolutionary changes in the rate of sex measured . As seen here , short-term disadvantages coupled with long-term advantages can occur in cases where sex increases ( adapting populations ) as well as in cases where sex continuously declines ( controls ) . However , we found substantial differences in the magnitudes of these effects between adapting and control populations . Moreover , the direction of the “long-term effect , ” rather than just the magnitude , differs between adapting and control treatments if one considers the top 10% rather than the variance ( the use of the latter is based on a weak selection approximation [30] ) . While our experiment is unique in being able to link a change in sex to short- and long-term effects , a number of details remain unknown . A long-term advantage is expected to exist when sex dissipates negative genetic associations . Are negative genetic associations built by non-linear selection [21] , [24] , [25] or Hill-Robertson effects [6] , [26] , [32] ? Similarly , we do not know whether dominance or epistasis is responsible for the immediate consequences of sex ( short-term effects ) . Such information will be important to help understand the relative importance of segregation and recombination in driving the evolution of sex . For sex to have any effect genetically , there must be genetic variation within populations . Even in well-adapted populations , we see clear evidence of genetic variance; when sex is imposed on random samples of parents , there is a dramatic decline in fitness . What sort of variation is responsible for this effect ? One simple explanation is that recessive deleterious alleles hitchhike to high frequency in a heterozygous state and can persist as long as populations reproduce asexually much of the time so that deleterious homozygotes are rarely produced . A second explanation is that multiple high-fitness co-adapted genotypes are maintained by some form of balancing selection such as frequency-dependent selection . When it occurs , sex and recombination breaks down these co-adapted genotypes , resulting in low fitness genotypes . Unlike the first explanation , this alternative can apply to both haploid and diploid systems and so has been invoked to account for sex-induced reductions in fitness in studies on haploid Chlamydomonas [8] , [54] , [55] , [58] . Though our experiment is consistent with the main tenets of the Weismann hypothesis , it also demonstrates a well-known weakness of this idea . The advantage to sex observed here is brief on an evolutionary time scale . Perhaps if adaptive optima are continually shifting , selection for sex could be maintained indefinitely [24] . Do selective pressures in nature change sufficiently frequently to explain the observed levels of sex ? This is an empirical issue requiring data from the field . Lab-based studies such as the one reported here are necessary to directly evaluate the potential of hypotheses and to test their underlying mechanisms . However , such studies alone cannot prove any hypothesis as the explanation for the ubiquity of sex in nature . Attempts to study the evolution of sex in the field [15] , [18] , [53] , [59] will be needed to evaluate the importance of results from theory and lab experimentation .
Replicate experimental populations were either maintained in the same environment to which they had previously adapted ( non-adapting control populations; n = 5 per environment ) or were transitioned to the other environment 10 d after the start of the experiment—that is , either from Environment A to B , or from B to A ( adapting populations; n = 5 per environment ) . The transition occurred by substituting the other algae source during the regular food replacement schedule ( see above ) . About 95% of the algae was replaced after 1 wk . Ten additional adapting populations ( n = 5 per environment ) were started at day 37 of the experiment . To create these populations , the 10% extracted media of the control populations on day 36 were pooled with others from the same environment and the following day were distributed among five new populations for each adapting population . The remaining volume was filled with fresh medium and the respective algae solution . Sexual reproduction in Brachionus species is density dependent and stimulated by a chemical signal that is produced by the rotifers [61] . The propensity for sex was measured weekly and followed the protocol in [17] . Briefly , we isolated 42 asexual individuals from each population and individuals were transferred to single wells with 10 ml of food containing medium , so that each rotifer received the same food from which they were isolated . Individual rotifers were maintained under these conditions for two generations and one neonate of the third generation after isolation was individually transferred to a single cell of a 96-well plate with conditioned medium [17] containing the same food source from which they were isolated . The initial female was removed after they produced the first offspring and the offspring was scored as amictic or mictic by the type of offspring they produced . Sexual females produced only males ( haploid ) because they were unmated in the assay . Ten fertilized mictic ( resting ) eggs and 10 amictic eggs were isolated weekly from each population and transferred to a single well of a 24-well plate for hatching ( Figure S2 ) . The two types of eggs can be distinguished by their morphology: amictic eggs are completely filled and have a pale gray colour , while resting eggs are only partially filled and have a much darker coloration . Rotifer females from amictic eggs hatched within 1 d after isolation , and females from resting eggs hatched within 1 to 5 d . To avoid differences that could occur because sexually derived genotypes develop from resting eggs whereas asexually derived genotypes develop from amicitc eggs , we maintained each genotype by clonal reproduction for two generations prior to fitness measurements ( in the same food environment from which they were isolated ) . The first five offspring from the third generation ( asexual ) after isolation were used to measure lifetime reproduction ( five individuals per genotype ) . Each individual was placed in an individual well , and each day , the number of offspring was recorded and the female was transferred to a new well with fresh medium and food until the female died . Lifetime reproduction was used as a measure for fitness . Spontaneously occurring fertilized mictic eggs are expected to originate from a non-random subsample of the population . To examine the effects of sex on a more random sample of genotypes , we transferred 5% of the populations to a new flask , added additional food , and allowed the population to grow to high densities ( Figure S2 ) , inducing almost the entire population to switch to sexual reproduction ( density >30 females per ml; all genotypes are expected to switch to sexual reproduction at this density; cf . [17] , Figure S2 ) . Another 5% were transferred to a flask containing a large volume of medium and food , and these subpopulations were kept at low densities to ensure only asexual reproduction ( less than one female per ml ) . After 7 d , 20 resting eggs were isolated from the high-density subpopulations and transferred individually to single wells for hatching and fitness assays as described above . Similarly , 20 amictic eggs were transferred from the low-density subpopulations . This procedure was applied to samples collected on Days 33 and 67 for the first set of adapting populations and on Days 53 and 67 for the second set of adapting populations . For each of these time points ( Days 33 , 53 , and 67 ) , similar data were collected from the non-adapting control populations . Multivariate statistical analyses were done in the R statistical environment [62] . Treatment ( Control A , Control B , Adapting B→A , Adapting A→B ) specific models ( generalized mixed models GLMM using the lmer4 package [63] ) were used to test for differences in the percentage of fertilized mictic eggs ( Figure 1 ) and propensity to reproduce sexually ( Figure 3 ) with time as a fixed effect and replicate population as a random effect ( using binomial error structure ) . To test for the increase and decrease in sex in the adapting populations , quadratic and linear models were compared . The effect of sex on the distribution of genotype fitnesses was examined as follows . All analyses were performed on genotypic mean values ( from five clonal replicates per genotype ) . Mean fitness of sexually and asexually derived rotifers hatched from naturally occurring eggs isolated directly from the experimental populations ( Figure 2 ) were compared using environment and time-point-specific generalized mixed models ( GLMM ) with reproduction mode ( sexually or asexually ) as fixed and replicate population nested in reproduction mode as random effect . To examine the effects of sex on a more random sample of genotypes , the distributions of sexually and asexually derived offspring were compared with respect to mean , variance , and mean of the top 10% . In each case , the data were analyzed with a linear model on the difference between sexuals and asexuals , using population as the unit of replication . To evaluate the effect of sex within treatments , we examined the significance of the intercept in separate analyses for adapting and control populations ( variables were coded such that the intercept reflects the average effect across environments and time ) . For adapting populations , only Day 33 data for Set 1 were used as fitness had plateaued before Day 67 ( Figure 2 ) . For Set 2 , we used the average values from Days 53 and 67 for each population ( these represent days 16 and 30 of adaptation for Set 2 ) . We obtained qualitatively similar results , using a total evidence approach by combining p values [64] from individual paired t tests ( sex versus asex ) for each set in each environment . To directly compare the effects of sex between adapting and control populations , we analyzed the difference in log of fitness between sexuals and asexuals in a linear model including both adapting and control treatments . Variance was calculated as the variance among genotypic means . | For well over a century , biologists have wondered why sex is such a common mode of reproduction , given the immediate 2-fold fitness cost entailed by the reduced number of offspring per parent . The most classic explanation is that sex is favoured because it helps to generate the variation necessary for adaptation . While theoretical models and indirect lines of evidence support this idea , there are no direct experimental data and it is far from obvious whether any such advantage could balance the considerable costs of sex . Using experimental populations of a facultatively sexual species of rotifer , we demonstrate that rates of sex evolutionarily increase as populations adapt to novel environments . We show that sex creates a diverse array of genotypes , including many that are quite unfit but also others that are very fit in the new environment . Though the average fitness of these sexually derived offspring is lower than that of asexuals , those well-adapted genotypes generated by sex contribute disproportionately to future generations , causing the genetic propensity for sex to ultimately increase . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"evolutionary",
"ecology",
"adaptation",
"population",
"genetics",
"biology",
"evolutionary",
"biology",
"evolutionary",
"processes"
] | 2012 | The Evolution of Sex Is Favoured During Adaptation to New Environments |
Hsp90 is a molecular chaperone essential for protein folding and activation in normal homeostasis and stress response . ATP binding and hydrolysis facilitate Hsp90 conformational changes required for client activation . Hsp90 plays an important role in disease states , particularly in cancer , where chaperoning of the mutated and overexpressed oncoproteins is important for function . Recent studies have illuminated mechanisms related to the chaperone function . However , an atomic resolution view of Hsp90 conformational dynamics , determined by the presence of different binding partners , is critical to define communication pathways between remote residues in different domains intimately affecting the chaperone cycle . Here , we present a computational analysis of signal propagation and long-range communication pathways in Hsp90 . We carried out molecular dynamics simulations of the full-length Hsp90 dimer , combined with essential dynamics , correlation analysis , and a signal propagation model . All-atom MD simulations with timescales of 70 ns have been performed for complexes with the natural substrates ATP and ADP and for the unliganded dimer . We elucidate the mechanisms of signal propagation and determine “hot spots” involved in interdomain communication pathways from the nucleotide-binding site to the C-terminal domain interface . A comprehensive computational analysis of the Hsp90 communication pathways and dynamics at atomic resolution has revealed the role of the nucleotide in effecting conformational changes , elucidating the mechanisms of signal propagation . Functionally important residues and secondary structure elements emerge as effective mediators of communication between the nucleotide-binding site and the C-terminal interface . Furthermore , we show that specific interdomain signal propagation pathways may be activated as a function of the ligand . Our results support a “conformational selection model” of the Hsp90 mechanism , whereby the protein may exist in a dynamic equilibrium between different conformational states available on the energy landscape and binding of a specific partner can bias the equilibrium toward functionally relevant complexes .
Heat Shock Protein 90 ( Hsp90 ) is an essential ATPase directed molecular chaperone required for folding quality control , maturation and trafficking of client proteins [1]–[4] . Hsp90 represents a fundamental hub in protein interaction networks [5] , [6] , with key roles in many cellular functions . Hsp90 oversees the correct maturation , activation and trafficking among specialized cellular compartments [7] of a wide range of client proteins [4] , [7] , [8] . The functions of clients range from signal transduction to regulatory mechanisms and immune response [3] . Client proteins typically include numerous kinases , transcription factors and other proteins that serve as nodal points in integrating cellular responses to multiple signals [3] . Given its role at the intersection of fundamental cellular pathways , it is becoming increasingly clear that Hsp90 deregulation can be associated with many pathologies ranging from cancer to protein folding disorders and neurological diseases [9] , [10] . Because of this role in disease development , pharmacological suppression of Hsp90 activity has become an area of very intense research , in molecular oncology in particular . Targeted suppression of Hsp90 ATPase activity with a small molecule inhibitor , the benzoquinone ansamycin antibiotic 17-allylamino-17-demethoxygeldanamycin ( 17-AAG ) , and some of its derivatives [11] , [12] , has shown promising anticancer activity in preclinical models and has recently completed safety evaluation in humans [13] . Further clinical trials have also been initiated with other small molecules also used in drug combinations in various cancer types [13] . Hsp90 operates as a dimer in a complex cycle driven by ATP binding and hydrolysis and by ATP/ADP exchange . Initial structural efforts concentrated on isolated , individual domains of human [14]–[16]or yeast Hsp90 [3] , [4] , [17]–[21] , the ER homologue Grp94 [22] , [23] or the Escherichia coli homologue , HtpG [20] , [24] . The crystal structures of larger constructs have also been reported [20] , [25] . The first X-ray crystal structures of full-length Hsp90 from yeast bound to the ATP mimic AMPPNP revealed a homodimeric structure in which the individual protomers have a twisted parallel arrangement [26] . Each protomer , in turn , is characterized by a modular architecture with three well-defined domains: an N-terminal regulatory Domain ( NTD ) , responsible for ATP binding , a Middle Domain ( M-domain ) , which completes the ATPase site necessary for ATP hydrolysis and binds client proteins , and a C-terminal dimerization Domain ( CTD ) which is required for dimerization [26] . The same global topology is shared by the ATP-bound states of the E . coli homolog HtpG [25] and by the Endoplasmatic Reticulum ( ER ) paralog Grp94 [27] . Interestingly , crystal structures of the full-length constructs for Htpg or Grp94 in complex with either ADP or in the apo state showed substantially different conformations . The HtpG apo state adopted an open structure in which each of the three domains exposed hydrophobic surface area , while in the ADP-bound form these hydrophobic surfaces clustered to form a more compact state [25] . Structural and biochemical studies of the solution state of Hsp90 and its complexes using small angle X-ray scattering ( SAXS ) [28] have provided the first experimental evidence of a highly dynamic and stochastic nature of Hsp90 , whereby the equilibrium between different conformational states of the molecular chaperone can be readily shifted to recruit a Hsp90 conformation that is suitable for efficient Cdc37 co-chaperone recognition . More recent solution structure data obtained using SAXS , single particle cryo-electron microscopy and modeling approaches showed that the apo-Hsp90 dimer ( from both prokaryotic and eukaryotic sources ) may be in equilibrium among different open , extended states , still preserving the constitutive dimerization provided by the CTDs , and that nucleotide binding may shift this equilibrium towards compact conformations [29]–[31] . In particular , SAXS data have revealed that the ADP-bound compact state of HtpG can be in equilibrium with an extended state , which could be significantly populated in the absence of crystal packing effects [30] . In contrast , crystal structures of AMPPNP and ADP-bound forms of the ER-paralog Grp94 showed that there is relatively little difference in conformation between the two nucleotide bound states in the crystal [27] , representing extended structures . Recent studies based on mutation analysis , cross-linking and electron microscopy [32] , [33] suggested that different , compact states can be accessed by Grp94 in the presence of ATP . These studies have indicated that upon binding to a specific partner , functional states of Hsp90 can be recruited using the intrinsic conformational flexibility of Hsp90 . Although the exact mechanism of coupling between ATP-binding/hydrolysis and client protein folding is still unclear , the combination of X-ray structural observations and biochemical data supports a picture in which the chaperone undergoes conformational rearrangements bringing the two NTDs in close association in the ATP-bound state , but not in the ADP-bound or apo states . This defines a conformational cycle that involves constitutive dimerization through the CTDs and transient , ATP-dependent dimerization of the NTDs in a “molecular clamp” mechanism . In terms of intrinsic protein dynamics , the mechanism of conformational coupling to the ATPase cycle involves a “tense” , structurally rigid conformational state of Hsp90 upon ATP binding , whereas a subsequent hydrolysis to ADP leads to a more “relaxed” , structurally flexible state of Hsp90 [3] , [25] , [26] . Finally , in the nucleotide-free form , the dimer moves to an “open” state . The crystal structures of the full-length dimer also highlight the remarkable flexibility of the ATP-lid , a segment composed of two helices and the intervening loop located immediately adjacent to the ATP binding site [26] . The lid is displaced from its position in the isolated Hsp90 NTD structure and folds over the nucleotide pocket to interact with the bound ATP yielding the ‘closed’ conformation indicating its possible importance in the progression of the chaperone cycle . These studies are reminiscent of the results from an H/D exchange mass spectrometry investigations on the human Hsp90 in solution , which showed that the co-chaperone and inhibitor binding to the NTD can induce conformational changes at the Hsp90 domain-domain interfaces [34] . Moreover , Frey and coworkers [35] have shown that kinetic and equilibrium binding constants depend on the intrinsic conformational equilibrium of the Hsp90 obtained from different species , reflecting differential affinity and reactivity towards ATP . The kinetic analysis of the ATPase cycle has suggested that during the ATPase cycle Grp94 may be predominantly in the open state ( 97% ) . In contrast in the yeast Hsp90 the open state is only populated to 20% and a closed structure is observed in the presence of nucleotides [26] . Hence , conformational transitions during the ATPase cycle are structurally similar for different Hsp90 proteins , while the energetic balance between individual steps may be species-dependent , which is manifested in differences in the binding kinetics [35] . Overall , the solution data have suggested that the molecular mechanism of the Hsp90 chaperone cycle can be more adequately described as a stochastic process , in which ATP binding can shift the intrinsic conformational equilibrium of Hsp90 between the open apo state , the ADP-bound compact and the ATP-bound , closed protein state seen in different crystal structures . The most recent structural studies of the apo and nucleotide-bound conformations of the E . coli , yeast , and human Hsp90 homologs have further supported the existence of a universal three-state conformational cycle for Hsp90 , consisting of open-apo , ATP-closed and ADP-compact nucleotide-stabilized states , whereby the intrinsic conformational equilibrium between these states can be highly-species dependent [33] . According to these results , the evolutionary pressure may act through thermodynamic stabilization of the functionally relevant Hsp90 conformations recruited from the conformational equilibrium , to ensure the adequate response to the presence of organism-specific co-chaperones and protein clients . Importantly , ATP or ADP binding can shift the conformational equilibrium far away from the apo state for E . coli and yeast Hsp90 , whereas the conformational equilibrium for human Hsp90 is largely dominated by the open form , even in the presence of the nucleotide binding . Strikingly , this study has shown that nucleotide binding provides only small stabilization energy , thereby biasing , rather than determining , the occupancy of different conformational states existing in a dynamic equilibrium . Overall , the intrinsic conformational flexibility of Hsp90 is critical to the molecular chaperon cycle , including structural adaptation to diversity of co-chaperones and client proteins [21] . Different steps in the cycle are accompanied by binding to different co-chaperone proteins with specific functions . The Hop co-chaperone , for instance , arrests ATP hydrolysis and binds simultaneously to the Hsp70 molecular chaperone , coupling the two systems . The Hop binding to Hsp90 involves interactions at both M-domains and CTD domains [36] , [37] stabilizing a conformation that is incompetent for ATP hydrolysis and N-terminal dimerization [38] . In contrast , the stress-regulated co-chaperone Aha1 substantially increases ATPase rates increasing Hsp90 chaperone activities [39] . In the case of binding to other co-chaperones , Cpr6 and Sba1 , it was shown that ATP-binding and hydrolysis is required to ensure productive complex formation: interestingly , Sba1 binds to the NTD while Cpr6 binds to the CTD [40] . These observations suggest a role for the nucleotide in selecting and stabilizing different conformations of Hsp90 , related to specific different functions in the chaperone cycle [33] . These crystallographic , cryo-EM , SAXS and three-dimensional single-particle reconstruction studies , applied to the isolated Hsp90 domains and full Hsp90 dimer in different species , have provided a wealth of novel insights into the molecular mechanism and function of Hsp90 . However , there are still a number of important unresolved problems concerning the atomic resolution understanding of the interplay between ligand binding and the global functional motions of the molecular chaperone . We have recently performed computational studies of the Hsp90 conformational dynamics and analyzed at atomic resolution the effects of ligand binding on the energy landscape of the Hsp90 NTD by all-atom MD simulations . MD simulations of Hsp90 NTD have been carried out for the apo protein and Hsp90 complexes with its natural ligands ATP , ADP , small molecule inhibitors , and peptides [41] . These simulations have clarified the role of ATP-lid dynamics , differences in local conformational changes and global flexibility , as well as the functional interplay between protein rigidity and entropy of collective motions depending on the interacting binding partners . We have found that the energy landscape of the apo Hsp90 NTD may be populated by structurally different conformational states , featuring local conformational switching of the “ATP-lid” which is accessible on longer time scales . The results of this study have suggested a plausible molecular model for understanding the mechanisms of modulation of molecular chaperone activities by binding partners . According to this model , structural plasticity of the Hsp90 NTD can be exploited by the molecular chaperone machinery to modulate enhanced structural rigidity during ATP binding and increased protein flexibility as a consequence of the inhibitor binding . [41] . The molecular basis of signal propagation mechanisms and inter-domain communication pathways in the Hsp90 as a function of binding ligands cannot be inferred directly from crystallographic studies . As a result , computational approaches are instrumental in revealing the atomic details of inter-domain communication pathways between the nucleotide binding site and distant CTD , which may be involved in governing the chaperone equilibrium between major conformational states . In this work , we have embarked on a comprehensive computational analysis of Hsp90 dynamics and binding which provides important insights into our understanding of the Hsp90 molecular mechanisms and function at atomic resolution . We describe large-scale MD simulations to study the conformational motions and inter-domain communication pathways of the full-length yeast Hsp90 in three different complexes: with ATP , with ADP and in the apo form . In support of the experimental hypotheses , our results provide atomic models of a cross-talk between N- and C-terminal binding sites that may induce an allosteric regulation of the complex molecular chaperone machinery . These results of our study suggest that the low-resolution features of communication pathways in the Hsp90 complexes may be determined by the inherent topological architecture of the chaperone , yet specific signal communication pathways are likely to be selected and activated based on the nature of the binding partner .
To further characterize differences in the dynamic properties of the molecular chaperone , the global flexibility parameters were calculated for respective domains of the Hsp90 dimer , namely for the two NTDs , the two M-domains , and the two CTDs ( Table 1 ) . The global flexibility parameter is defined as the sum of the average fluctuations of Cα atoms during the MD trajectory . This measure of flexibility may be considered as a semi-quantitative measure of the differences in the underlying dynamics of studied Hsp90 complexes . Interestingly , the global flexibility parameters of the NTDs in the full-length Hsp90 complexes reflect a similar trend which was previously found for the isolated Hsp90 domains , namely in the presence of ATP the Hsp90 NTD domains are more rigid than in the presence of ADP or in the unbound apo Hsp90 form [41] . Hence , this analysis may capture global flexibility differences induced by the ligands in different environments: an isolated Hsp90 NTD in solution and Hsp90 NTD as an integral part of the molecular chaperone structure . Interestingly , the global flexibility parameters of the M-domains and CTD domains are also affected by the ligands . The M-domains are more rigid in the ATP-bound complex compared to the ADP-bound case , whereas for the CTDs the situation is reversed . In the presence of ATP , the CTDs show an increase in local fluctuations with respect to the simulation with ADP . The CTD interface helices ( helices 4 and 5 ) in the ATP-complex simulation are in fact involved in hinging the rotations of the two protomers on the path to the closed conformation ( see next subparagraph on ED analysis ) . The apo system shows an intermediate behavior at the NTD and a larger flexibility at the M-domains and CTD domains . Essential Dynamics ( ED ) [44] was then used to further investigate the influence of different ligands on the dominant conformational modes of the chaperone . ED identifies functionally relevant displacements of groups of residues and emphasizes the amplitude and direction of dominant protein motions by projecting them on a subset of the principal eigenvalues and eigenvectors of the residue pair covariance matrix calculated from MD . Using this approach , we have identified the protein regions which are involved in large-scale conformational changes . In the presence of ATP , the most relevant motion involves the ATP lid at the NTD and , at the opposite end of the protein , the CTDs ( Figures 9 and 10 ) . In both protomers , the NTDs undergo a concerted rotation and compaction motion , minimizing the distance between their respective hydrophobic interfaces ( Figures 10 and 11 and movies of the trajectory projections in Videos S1 and S2 ) . The CTDs in turn rotate around the dimer interface axis . The C-terminal dimerization site is made of the two terminal helices , helix 4 and helix 5 , from each protomer anchored to each other at the interface . Helix 4 is involved in the rotation of the external part of the CTD , while helix 5 acts as the hinge around which the rotation takes place . Globally , the dynamics of ATP-bound Hsp90 involves a concerted twisting and closing motion consistent with the molecular clamp model . In the presence of ADP , the projection of the MD trajectory onto the first eigenvector ( Figures 9 and 11 and movies of the trajectory projections in Videos S1 and S2 ) shows that the main motions involve the NTD and the M-domain of Hsp90 . The role of the C-terminal dimerization site is to provide the contacts necessary for the stability of the dimer either in the final open or in extended dimer conformation . At the N-terminal end , the rotation of both monomers is coupled to an increase of the inter-domain distance , as observed in the previous section . This motion propagates to the M-domains , which rigidly move , increasing their mutual distance and relative inclination . As an effect , the onset of the opening motion of the clamp is observed . Finally , for the apo protein , an opening motion from the closed conformation is observed , and is characterized by intra-domain oscillations that involve the whole chaperone , without any particular dynamic signature in terms of preferential directions of motion . The ED analysis reflects the onset of different slow functional modes , indicating that the main features of the global motions are determined by the identity of the ligands at the binding site . Despite long simulation time scales , a direct observation of conformational transitions in the Hsp90 dimer as a function of binding partner is not yet computationally feasible . However , it is important to emphasize that functional analysis of slow modes and global motions for all the three studied Hsp90 systems is fully consistent with experimental structural and dynamical characteristics of Hsp90 . For instance , our analysis shows that ATP can induce a closing motion stabilizing N-terminal dimerization , while ADP favors a transition towards a more open , relaxed state [26] , [29] , [30] , [33] . The essential dynamics spaces ( slow modes ) have been shown to be exceptionally robust when calculated over trajectories of different duration , reflecting the self-similarity of the structure of the various free energy minima available to a certain system [45] . The typical amplitudes of the motions projected along the slowest modes , on the contrary , depend on the number of visited minima [45] . As a consequence , we speculate that the preferred directions of the slow modes depending on the nucleotide identity capture the fundamental properties of the biologically relevant functional modes . The possibility to extend the simulations to longer timescales could eventually allow the analysis of large-scale conformational transitions . The results of our work have shown that the conformational dynamics of Hsp90 domains and relative stability of different conformational states can be modulated upon binding to different ligands . Importantly , we have also observed that the effect of ATP and ADP binding at the N-terminal binding site can propagate beyond the immediate binding partner and affect even remote regions of the Hsp90 dimer at a distance of more than 80 Å away from the binding interface: specifically , the fluctuations at the CTD ( Figures 10 and 11 ) and the global motion of the M-domain depend on the nature of the ligand . The chain connectivity and organization of secondary structure elements have been identified as the main factors in determining equilibrium dynamics and regulating the mechanical tasks necessary to carry out biological functions [46] . In this context , it is of prime interest to investigate which regions and secondary structure elements of the chaperone are mostly responsible for transmitting signals coded by ATP or ADP from the N-terminal binding site to the C-terminal binding site , and whether the inter-domain communication pathways and respective signal propagation mechanisms can be modulated by the presence of the binding partner . In order to elucidate the mechanisms of signal propagation from the nucleotide binding site , we have extended and adapted a recent approach proposed by Bahar and coworkers [47] , [48] to the analysis of all-atom MD simulation trajectories . The analysis of signal propagation , which was developed based on elastic network models [47] , [48] , describes signal transduction events in proteins as directly related to the fluctuation dynamics of atoms , defining the communication propensity of a pair of residues as a function of the fluctuations of inter-residue distance . In the present study , we investigate the presence of long-range communication propensities between NTD residues close to the nucleotide binding site and remote regions of the protein , such as the CTD interface . The communication propensity ( CP ) is calculated for any pair of residues during the trajectory . It is worth noting that CP describes a communication time , therefore low CP values are related to efficiently communicating residues . The average CP value for consecutive amino acids along the sequence , calculated considering for each residue i the neighbors comprised between i−4 and i+4 , is 0 . 025 . The average CP value for residues distant more than 40 Å is 0 . 12 . In the ADP and ATP simulations , around 1 percent of residue pairs have CP<0 . 025 even if they are at distances larger than 40 Å , while the percentage decreases to 0 . 5 percent in the unbound complex ( Figure S4 ) . Therefore , in the presence of ligands , a number of very distant residues may have a high communication propensity despite their physical separation and we set CP = 0 . 025 as a convenient threshold for discriminating fast communications at long distance . In order to investigate this aspect more thoroughly , for each complex ( ATP-bound , ADP-bound and apo ) a set of histograms was created , in order to scan communication efficiencies at increasing distances ( Figure 12 ) . Each bin refers to a residue and gives the fraction of residues that have high communication efficiency with it ( CP<0 . 025 ) at distances larger than an increasing cutoff of 40 Å , 60Å and 80 Å respectively . Residues corresponding to histogram peaks define regions that are specifically involved in efficient long-range communications . The histogram at 40 Å indicates that residues active in long range signaling belong to NTD , at the M-domain and CTD of the dimer . According to the experimental data , a number of experimentally studied mutations , known to determine temperature sensitive phenotypes [49] , and therefore perturbing functionally relevant regions , involve residues which in the present analysis are related to peaks of efficient signal communication ( Figures 12 and 13 ) . For instance , the T101I mutant has normal ATP affinity but reduced ATPase activity , whereas the mutant T22I displays enhanced ATPase activity and AMP-PNP-dependent NTD dimerization [50] . Also mutation of several M-domain residues ( Arg-376 , Gln-380 in yeast Hsp90 ) has been shown to affect ATPase activity [51] whereas mutants A587T and T101I of yeast Hsp90 render cells much more sensitive to Hsp90 inhibitor drugs [52] . Sites T22 , A41 , G81 , G170 , E381 and A587 correspond to peaks in both the ATP and ADP bound cases ( Figures 12 and 13 ) . Two of them , namely T22 and G81 , as well as residues adjacent to T101 , correspond to N-Domain long range communication sites , whose signaling properties persist at very long distances , as discussed in the following . Upon increase of the residue-residue distance in the CP scanning histograms , some peaks become progressively smaller or disappear , since the fraction of effectively coordinated residues decreases at longer physical distances . On the other hand , since the total number of possible pairs also decreases with increasing distance , for some residues the fraction of efficient communications may grow at longer distances , and those residues we define to be strongly active in long range signaling . The important finding from our analysis is that in the absence of any ligands , all peaks decrease at increasing distances , while in the presence of both ATP and ADP a number of peaks grow when distance increases . The most efficient communications at very long distance ( over 80 Å ) involve a subset of specific residues in all complexes , while for the apo form of Hsp90 only a small fraction of these residues is active . Moreover , we observe that in the presence of ATP the long-range communication from the binding site is mainly directed to specific residues at the CTD interface , while ADP activates communications between the NTD binding region and the CTD region surrounding the interface ( Figure 14 ) . In simulations with the ATP-bound Hsp90 complex , the NTD residues 81–95 and 121–140 ( Hsp90 residues numbering as in the pdb entry 2CG9 ) show a high long-range signaling propensity with segments 574–580 and with the two C-terminal interface helices , made of residues 645–654 ( helix 4 ) and 661–671 ( helix 5 ) respectively . In simulations with the ADP-bound Hsp90 complex , the NTD residues 75–83 and 91–96 communicate with the C-terminal regions comprising residues 566–578 , 611–625 and again the helical segment 645–654 ( helix 4 ) ( Figure 14 ) . In agreement with our previous results , this analysis suggests that ATP and ADP may have a different effect on the Hsp90 conformational dynamics , which is manifested in the activation of differential signal propagation pathways . However , both ATP and ADP may activate the inter-domain communication between the NTD and the β-sheets adjacent to the C-terminal interface . The A577C mutation in this β-sheet ( Figure 14 ) coupled with successive S-nitrosilation , was experimentally shown to affect the ATPase activity of the chaperone ( Marco Retzlaff and Johannes Buchner , personal communication ) . It is important to emphasize that the role of specific point mutations Y24F and A577C , affecting the inter-domain signal propagation , was independently validated by different experimental groups ( Len Neckers , Marco Retzlaff and Johannes Buchner and personal communication ) . Inhibition of the ATPase activity of Hsp90 by either mutation or small molecule inhibitors results in the degradation of client proteins in vivo , demonstrating the central importance of ATPase activity to the function of the chaperone [53] . The evidence of an efficient molecular communication between NTD residues near the binding site and CTD regions at the interface is of special importance given the rapidly growing interest in developing novel and specific Hsp90 inhibitors inhibition targeting allosteric Hsp90 regions , including CTD . It has been shown that the binding of coumarin-type antibiotics , such as novobiocin , at the carboxyl terminus antagonizes geldanamycin binding at the NTD [54] , blocks the binding of immunophilin co-chaperones in vitro [55] and markedly reduces the cellular levels of several oncogenic proteins kinases in vivo [56] . Also , inhibitor binding to the NTD site affects the binding of co-chaperones at the C-terminal site whereas the presence of co-chaperones there impacts negatively on the N-terminal binding of nucleotides or inhibitors [54]–[56] . The protein segment found to be involved in the binding of C-terminal inhibitors has sequence YETALLSSGFSLED in human Hsp90 [54] , [56] and corresponds to residues 647 to 660 in yeast Hsp90 . This segment is located at the interface helices ( helix 4 and subsequent loop connecting to helix 5 ) that we define to be active in communicating with the N-Domain binding site in the ATP simulation ( Figure 14 ) . Hence , the inter-domain communication from the NTD to the C-terminal interface is visible only in the presence of ATP ( Figure 14 ) . Our data therefore support , at atomic-resolution , the picture of a cross-talk between N- and C-terminal binding sites inducing allosteric regulation , as hypothesized by experimental observations . The method of CP analysis is based on the hypothesis that signal propagation in complex molecular systems can inflicted by correlated fluctuations determined by the intrinsic network topology . The ordered fluctuations are coupled to the global dynamics sampled under equilibrium conditions , thus controlling allosteric effects . In the original formulation , based on a coarse-grained representation of the protein , the Gaussian Network Model ( GNM ) [47] , [48] , provides an elegant , yet largely qualitative characterization of signal transduction pathways . Our study is an extension of this approach to the analysis of long MD simulations and allows to include all-atom representations and the detailed energetics of the system , that more adequately reflect the effect of subtle , yet important chemical modifications . These results suggest that the coarse-grained features of communication patterns may be dependent on the inherent topological architecture and packing of the protein [45]–[48] , [57]–[59] , but also that specific pathways can be selected and activated based on the binding partner . We speculate that structural architecture of the molecular chaperone and the intrinsic dynamic equilibrium between major conformational states can define the topology and energetics of signal communication pathways , which may be modulated by the binding partner . The correlation analysis based on all-atom MD simulations was compared with the results of the elastic network ( EN ) and GNM approaches . These coarse-grained models are widely used for elucidating the collective dynamics of proteins and exploring their relevance to biological function [60] , [61] . The basic ingredient in the GNM models is the topology of the inter-residue contacts in the native structure , which appears to be the major determinant of equilibrium dynamics [60] , [61] . We have found that the correlation matrix obtained from the GNM analysis applied to the full-length dimer of Hsp90 , in the absence of ligands , is consistent with that obtained from MD simulations . Namely , it reveals a positive correlation between the NTD and CTD within each protomer , which may be considered as an intrinsic feature of the three-dimensional structure of the system . Within the GNM model , the inter-protomer correlation between the two M-domains is not seen , which is in agreement with the results obtained from the all MD simulations of the apo Hsp90 . Importantly , we have confirmed that signal propagation pathways emerging from the GNM model [47] are similar to the ones inferred from the all-atom description of the system ( Figures S5 and S6 ) . In this context , it should be noted that GNM methods are limited by the lack of information on residue specificities ( effects of mutations ) , atomic details ( chemical modifications ) or side chain motions . ATP and ADP are very similar molecules and the simplification of their representation in terms of interaction centers , consistent with the GNM approach used here , would not allow to explicitly distinguish between different complexes , starting from the same chaperone structure . In this context , all-atom MD simulations allow to differentiate the effects of inter-domain correlations resulting from subtle chemical differences between ATP and ADP . For a further comparative analysis of all-atom and GNM representations of the system , we have selected the representative Hsp90 conformations from the most populated structural clusters obtained from ATP-bound and ADP-bound simulations . These protein conformations are then used in the GNM model to elucidate the main chain collective motions . Interestingly , the GNM analysis of the representative ATP-bound structure produces slow modes reflecting a closing motion involving the NTD and CTD . In contrast , a coarse-grained analysis of the ADP-bound representative conformation encodes for an opening motion of the two NTDs ( data not shown ) . Hence , these complementary models have recapitulated major features of the Hsp90 conformational cycle consistent with the “molecular clamp” mechanism . A combination of all-atom MD and GNM approaches may be a promising approach for exploring large conformational changes in complex molecular systems [62] . The nucleotide-dependent collective motions observed in our study recapitulate the essential atomic-level determinants of the molecular-clamp mechanism: ATP-binding brings the molecular chaperone into a closed , “tense” state characterized by an extended network of coherently moving residues , a compaction of the hydrophobic inter-protomer interfaces at the NTD and M-domain and determines a principal collective motional mode compatible with the closure of the clamp . ATP hydrolysis or ADP binding induces a high degree of anti-correlation in the inter-protomer motions , and the principal collective mode leads to a transition towards the “open” state of the molecular clamp . The two NTDs move apart and the NTD and M-Domain hydrophobic interfaces begin to be disrupted , the number of contacts decreases and their interaction is looser . This picture is consistent with a conformational change in the direction of the extended ( or completely open ) structure of the dimer . The collective motions of the apo Hsp90 structure do not reveal any specific signature , suggesting that its dynamics may be readily biased towards different regions of the conformational space upon ligand binding . Within the limitations of all-atom MD simulations , the results of our analysis suggest possible links between local interactions and short timescale fluctuations with slower biologically relevant functional motions . The results of our simulations refer to timescales of tens of nanoseconds and report on the microscopic behavior of the chaperone complexed to different ligands , while functional motions may take place on timescales that are orders of magnitude higher . It is worth noting at this point that in order to reach the states competent for client binding , folding or release , Hsp90 structures have to undergo extensive reorganization in an efficient fashion . One possible way to achieve this goal would be to sample pre-organized states and motions that are dictated by the local organization of interactions and by the local correlations/fluctuations described above and modulated by different nucleotides . This mechanism would define a hierarchy of possible dynamic substates that the chaperone can search in a more efficient way than by random sampling of all possible conformations . The concept of hierarchy of substates has already been investigated previously [63]–[65] , together with the idea that preferred relatively small , local fluctuations in enzyme systems lead to states resembling the catalytically active conformations . In this framework , Henzler-Wildman et al . in a fundamental study on the enzyme Adenylate Kinase ( Adk ) [66] , correlated smaller amplitude , faster motions ( ps to ns ) at the atomic scale with the large amplitude , slow motions on the micro- to millisecond scales related to enzymatic reactivity . The results of our simulations suggest that global rearrangements of Hsp90 , occurring on longer timescales , can be facilitated by nucleotide-dependent microscopic motions and correlated fluctuations occurring on the shorter timescales that are now accessible to all-atom MD simulations . The “conformational selection model” [67] appears to be the most suitable model to explain the functional dynamics of the full-length Hsp90 dimer . The protein fluctuates at equilibrium among different dynamic states available on the energy landscape , and binding of a specific partner may shift the equilibrium towards thermodynamically most stable complexes . Local microscopic fluctuations accessible in the MD timescale facilitate the large-scale , slower global motions that regulate the chaperone cycle [68] . Different parts of the N-Domain are involved in sensing the presence of a certain specific ligand and switching to the specific phase of the chaperone cycles . The formation of this complex is sensed by the ATP-lid , and the interfaces of the N- and C-Domains , that couple nucleotide binding and hydrolysis to global dynamics and hydrophobic surface remodeling . These considerations are consistent with the energy landscape theory , whose application is extended here to macromolecular machines , for which coupling between dynamics and binding is often accompanied by conformational transitions associated with the biological functions of proteins and as such are intimately connected to the underlying energy landscape . These results also recapitulate the notion that binding of different ligands modulates the degree of local energetic frustration of the Hsp90 interface regions that are important for function [69] , i . e . the hydrophobic inter-protomer interface between the N-Domains and M-Domain , favoring or disfavoring the formation of stable inter-protomer interactions . The structural plasticity observed in our study may favor adaptation of the chaperone to bind with sufficient affinity the wide range of protein clients with different sequences and conformations . These findings are in accordance with the recent results advocating a potential universal role of conformational selection mechanisms in a variety of biological systems [67] , [70]–[72] . In this study , we have presented a comprehensive molecular analysis of signal propagation mechanisms and long-range communication pathways in the molecular chaperone Hsp90 . The analysis is carried out based on large-scale all-atom MD simulations of Hsp90 molecular chaperone full-length dimer in explicit water . We have supplemented these simulations with a battery of computational modeling and analysis tools , including essential dynamics , correlated motions analysis and signal propagation modeling . The results of our simulations shed light on functionally relevant motions of the full-length Hsp90 , recapitulating fundamental features of the chaperone conformational cycle consistent with the “molecular clamp” mechanism . We have elucidated the mechanisms of signal propagation and discovered functionally important residues and secondary structure elements that emerge as hot spot mediators of communication between the nucleotide-binding site and the C-terminal interface . Furthermore , we have fond that inter-domain signal propagation pathways may be activated as a function of the binding partner . In particular , ATP binding may switch on a rapid communication between the ATP-lid , the nucleotide binding site , the catalytic loop of the M-Domain and the two helices defining the interface between two CTDs from the two different monomers . In contrast , the presence of ADP defines a different signal transduction path involving the main helix of the NTD , and the external helices of the CTD . We have provided a comprehensive computational analysis of the Hsp90 communication pathways and conformational dynamics at atomic resolution , revealing the role of the nucleotide in effecting conformational changes . The results of this study support a “conformational selection model” of the Hsp90 binding mechanism , whereby the protein may exist in a dynamic equilibrium between different conformational states available on the energy landscape , and binding of a specific partner can bias the equilibrium towards functionally relevant complexes .
All-atom MD simulations in explicit water have been independently carried out on a long simulation time scale of 70 ns for Yeast Hsp90 dimer complexed with the natural substrates ATP and ADP and for the unbound dimer ( apo form ) . The crystal structure ( pdb entry 2CG9 ) [26] containing Yeast Hsp90 dimer bound to ATP and complexed with co-chaperone p23/Sba1 has been used as starting conformation , upon removal of the co-chaperone . The crystal structure employed as a starting point for the simulations was obtained by the deletion of the long disordered loop connecting the NTD and the M-domain and by introducing the Ala107Asn point mutation , that was used to stabilize ATP-binding and the closed structure in order to obtain more favorable crystallization conditions [26] . These two factors were shown not to have an impact on the general functional properties of the chaperone [26] . However , the charged loop connecting the NTD and the M-domain , which is not present in the crystal structure , is replaced by a model linker made of 10 Gly residues and modeled by the ModLoop server . Also disordered loops in the CTD are modeled with ModLoop , but conserving the original sequence [73] . The p23/Sba1 co-chaperone atoms were removed from the starting structures used in the simulations . The crystal structure also included the p23/Sba1 co-chaperone protein [26] , which may also contribute to the stabilization of the closed Hsp90 structure . In simulations , the p23/Sba1 co-chaperone atoms were completely removed from the starting structures . The crystal structure ( pdb entry 2CG9 ) does not contain Mg2+ ions , most likely due to the resolution at which the crystals were obtained , which is too low to exactly localize the ion . Mg2+ is required for activity in many ATPases and is also present in the 1Y4S structure of the HtpG , where it coordinates ADP . Mg2+ ions were not included in our simulations . This choice was due to the absence of the coordinates for the ion in the X-ray data from the pdb due to low resolution , and to uncertainties related in placing the ion in the active site either by hand or by using automated docking programs . Furthermore , it was shown previously that the introduction of a small number of ions can lead to very severe sampling and equilibration problems suggesting that in practical calculations convergence can best be achieved by incorporating either no counter ions or by simulating at high ionic strength to ensure sufficient sampling of the ion distribution [74] . Finally , in runs on the isolated NTD domains used for the study reported in [41] we noticed no substantial differences on the global dynamic properties , at least in the time scales simulated , in the case of presence or absence of Mg2+ . The apo and ADP simulations have been carried out respectively by removing the ligand and by replacing ATP with ADP , as in [41] . The tetrahedral solvation box contains around 57000 particles . All simulations and the analysis of the trajectories have been performed using the GROMACS software package [75] using the GROMOS96 force field [76] and the SPC water model [77] . Each Hsp90 dimer system simulated in this study was first energy relaxed with 2000 steps of steepest descent energy minimization followed by another 2000 steps of conjugate gradient energy minimization . The energy minimization was used to remove possible bad contacts from the initial structures . For each of the simulations , the system was equilibrated by 50 ps of MD runs with position restraints on the protein and ligand to allow relaxation of the solvent molecules . These first equilibration runs were followed by other 50 ps runs without position restraints on the solute . The first 5 ns of each trajectory were not used in the subsequent analysis in order to minimize convergence artifacts . Equilibration of the trajectories was checked by monitoring the equilibration of quantities such as the RMSD with respect to the initial structure , internal protein energy , fluctuations calculated on different time-intervals . The electrostatic term was described by using the particle mesh Ewald algorithm . The LINCS [78] algorithm was used to constrain all bond lengths . For the water molecules the SETTLE algorithm [79] was used . A dielectric permittivity , ϵ = 1 , and a time step of 2 fs were used . All atoms were given an initial velocity obtained from a Maxwellian distribution at the desired initial temperature of 300 K . The density of the system was adjusted performing the first equilibration runs at NPT condition by weak coupling to a bath of constant pressure ( P0 = 1 bar , coupling time τP = 0 . 5 ps ) [80] . In all simulations the temperature was maintained close to the intended values by weak coupling to an external temperature bath [80] with a coupling constant of 0 . 1 ps . The proteins and the rest of the system were coupled separately to the temperature bath . The structural cluster analysis was carried out using the method described by Daura and coworkers with a cutoff of 0 . 25 nm [81] . Principal Component Analysis or Essential Dynamics ( ED ) reduces the dimensionality of the covariance matrix by diagonalization [44] . This method describes global protein motions that are represented by the matrix eigenvectors and eigenvalues . Principal eigenvectors are in general associated with large-scale movements , also termed slow modes , which are responsible for protein functions . The covariance matrix for each of the simulated Hsp90 dimer systems was built by averaging motions of Cα atoms deviating from the mean structure , with the latter calculated over the trajectory excluding the first 5 ns needed for equilibration . Ligands are not included in the calculation . Translational and rotational degrees of freedom are eliminated and the average atomic coordinates , xi , ave , i = 1 , … , 3N , are calculated along the MD trajectory ( 20 ) . The essential directions of correlated motions during dynamics are then calculated by diagonalizing the 3N×3N covariance matrix Cij . The MD trajectory can be projected onto the main essential direction , corresponding to the largest eigenvector , in order to visualize the extreme structures and the major fluctuations of the correlated motions . The correlation matrix Corrij is a N×N array , whose i-j entry summarizes the correlation between the motion of atom i and of atom j , is obtained from the reduction and normalization of the covariance matrix . We have extended and adapted a recent approach proposed by Bahar and coworkers to the analysis of all-atom MD simulation trajectories . The analysis of signal propagation , which was developed based on elastic network models [47] , defines signal transduction events in proteins as directly related to the fluctuation dynamics of atoms , defining the communication propensities ( CP ) of a pair of residues as a function of the fluctuations of interresidue distances . Residues whose Cα-Cα distance fluctuates with a relatively small intensity during the trajectory are supposed to communicate more efficiently than residues whose distance fluctuations are large . In the former case , a perturbation at the one site , affecting the Cα position , is likely to be visible ( reflected ) at the second site , while in the latter case the communication is less efficient , due to the intrinsic amplitude of the distance fluctuations . The Communication Propensity ( CP ) of any two residues is defined as the mean-square fluctuation of the interresidue distance defining as distance between the Cα atoms of residue i and residue j , respectively:By projecting these quantities on the 3D structures of the protein bound to different ligands , it will be possible to identify possible differences in the inter-domain and inter-protomer long-range redistributions of interactions . We employed the implementation of the GNM approach developed by Micheletti and coworkers [82] to identify collective conformational changes and large-scale movements for the apo-protein , and compare the main slow modes to those identified by the all-atom MD simuation model . This model implements a Cα-based amino acid representation of the protein . The energy function used to simulate the thermal equilibrium fluctuations of the protein around the reference conformation is obtained by introducing the following penalties for displacing two Cα's , i and j from their reference positions , ri0 and rj0 to generic ones ri and rj . where r0ij is the distance vector , is the distance vector change , ν and μ and run over the three Cartesian components and k is a parameter controlling the strength of the quadratic coupling . The form of the energy function in equation 2 is common in most ENM or GNM approaches . Exploring the dynamics of the protein complexes in this simplified theoretical framework allows calculating the degree of correlation of the displacement from the equilibrium position of pairs of residues . Communication Propensities in this framework are calculated according to [47] . | Dynamic processes underlie the functions of all proteins . Hence , to understand , control , and design protein functions in the cell , we need to unravel the basic principles of protein dynamics . This is fundamental in studying the mechanisms of a specific class of proteins known as molecular chaperones , which oversee the correct conformational maturation of other proteins . In particular , molecular chaperones of the stress response machinery have become the focus of intense research , because their upregulation is responsible for the ability of tumor cells to cope with unfavorable environments . This is largely centered on the expression and function of the molecular chaperone Hsp90 , which has provided an attractive target for therapeutic intervention in cancer . Experiments have shown that the chaperone functions through a nucleotide-directed conformational cycle . Here , we show that it is possible to identify the effects of nucleotide-related chemical differences on functionally relevant motions at the atomic level of resolution . The protein may fluctuate at equilibrium among different available dynamic states , and binding of a specific partner may shift the equilibrium toward the thermodynamically most stable complexes . These results provide us with important mechanistic insight for the identification of new regulatory sites and the design of possible new drugs . | [
"Abstract",
"Introduction",
"Results/Discussion",
"Methods"
] | [
"biophysics/protein",
"folding",
"biophysics/theory",
"and",
"simulation",
"pharmacology/drug",
"development",
"biophysics/biomacromolecule-ligand",
"interactions",
"computational",
"biology",
"computational",
"biology/molecular",
"dynamics"
] | 2009 | Modeling Signal Propagation Mechanisms and Ligand-Based Conformational Dynamics of the Hsp90 Molecular Chaperone Full-Length Dimer |
Inosine monophosphate dehydrogenase ( IMPDH ) catalyzes an essential step in the biosynthesis of guanine nucleotides . This reaction involves two different chemical transformations , an NAD-linked redox reaction and a hydrolase reaction , that utilize mutually exclusive protein conformations with distinct catalytic residues . How did Nature construct such a complicated catalyst ? Here we employ a “Wang-Landau” metadynamics algorithm in hybrid quantum mechanical/molecular mechanical ( QM/MM ) simulations to investigate the mechanism of the hydrolase reaction . These simulations show that the lowest energy pathway utilizes Arg418 as the base that activates water , in remarkable agreement with previous experiments . Surprisingly , the simulations also reveal a second pathway for water activation involving a proton relay from Thr321 to Glu431 . The energy barrier for the Thr321 pathway is similar to the barrier observed experimentally when Arg418 is removed by mutation . The Thr321 pathway dominates at low pH when Arg418 is protonated , which predicts that the substitution of Glu431 with Gln will shift the pH-rate profile to the right . This prediction is confirmed in subsequent experiments . Phylogenetic analysis suggests that the Thr321 pathway was present in the ancestral enzyme , but was lost when the eukaryotic lineage diverged . We propose that the primordial IMPDH utilized the Thr321 pathway exclusively , and that this mechanism became obsolete when the more sophisticated catalytic machinery of the Arg418 pathway was installed . Thus , our simulations provide an unanticipated window into the evolution of a complex enzyme .
Textbooks extol the extraordinary catalytic power and specificity of enzymes , yet the ability of many enzymes to promote several different chemical transformations is even more remarkable . In examples such as the polyketide synthases , the substrate is tethered to a flexible linker and swings gymnastically between separate active sites [1] . The evolutionary path to the assembly of such enzymes seems reasonably straightforward: gene duplication and recombination , followed by optimization of a promiscuous activity [2–6] . In contrast , enzymes such as IMP dehydrogenase ( IMPDH ) move around a stationary substrate , restructuring the active site to accommodate different transition states [7] . Such enzymes pose an evolutionary conundrum: it seems unlikely that Nature could simultaneously install multiple sets of catalytic machinery into the ancestral protein . IMPDH controls the entry of purines into the guanine nucleotide pool , which suggests that the origins of IMPDH are primordial , so the ancestral IMPDH probably utilized a simpler catalytic strategy . IMPDH catalyzes two very different chemical transformations: ( 1 ) a dehydrogenase reaction between IMP and NAD+ that produces a Cys319-linked intermediate E-XMP* and NADH , and ( 2 ) a hydrolysis reaction that releases XMP ( Figure 1A ) [7 , 8] . A mobile flap is open during the hydride transfer reaction , permitting the association of NAD+ . After NADH departs , this flap occupies the dinucleotide site , carrying Arg418 and Tyr419 into the active site and converting the enzyme into a hydrolase ( Figure 1B ) . Thus , the dehydrogenase and hydrolase reactions utilize mutually exclusive conformations of the active site . All enzymes that catalyze hydrolysis reactions have some strategy to activate water . This strategy has been difficult to recognize in IMPDH because the hydrolytic water interacts with three residues that are usually protonated at physiological pH: Thr321 , Arg418 , and Tyr419 ( Figure 1C ) [9] . The rate of the hydrolysis step decreases by a factor of 103 when Arg418 is substituted with Ala or Gln , whereas a decrease of approximately 20 is observed when Tyr419 is substituted with Phe [10 , 11] . Neither Arg418 nor Tyr419 is involved in the dehydrogenase reaction , as expected , given their position on the mobile flap . In contrast , Thr321 is found on the same loop as the catalytic Cys319 , and both the dehydrogenase and hydrolysis reactions are decreased by a factor of 20 when this residue is substituted [11] . These observations suggest that Arg418 is the most likely candidate for the role of general base in the IMPDH reaction [11 , 12] . We performed a series of hybrid quantum mechanical/molecular mechanical ( QM/MM ) simulations to further investigate the mechanism of the hydrolysis reaction of IMPDH . Surprisingly , these simulations find that IMPDH possesses two mechanisms to activate water: the Arg418 pathway as previously proposed , and a second pathway utilizing Thr321 . Phylogenetic analysis indicates that the Thr321 pathway was present in the ancestral enzyme . These observations suggest that the primordial IMPDH used the Thr321 pathway exclusively , and elimination of the Arg418 pathway by mutation of modern IMPDH creates an enzymatic atavist .
When Arg418 is deprotonated in the starting condition , the lowest energy pathway for the hydrolysis reaction involves the transfer of a proton to the neutral Arg418 ( the Arg418 pathway , Figure 2A–2C ) . Proton transfer is virtually complete at the transition state , and the developing hydroxide is stabilized by interactions with Tyr419 , Thr321 , and another water molecule . Importantly , a stable hydroxide intermediate is not observed; the developing hydroxide instantaneously reacts with E-XMP* . These results are in remarkable agreement with experimental observations: solvent isotope effects ( SIE ) demonstrate that proton transfer is rate limiting ( SIE = ∼2 [11] ) , and Bronsted analysis indicates that proton transfer is virtually complete in the transition state ( β = ∼1 [12] ) . However , the calculated energy barrier is only 8 . 0 kcal/mol , much less than the experimentally observed barrier of 16 kcal/mol [10] . This difference may reflect uncertainties in the calculation , but we believe this is unlikely . A more intriguing source of discrepancy arises from the starting condition of neutral Arg418; if only a small fraction of the enzyme exists in this state , the energy barrier will be correspondingly increased . Indeed , if the pKa of Arg418 is 12 . 5 , as for a typical Arg residue , the barrier would be increased by approximately 6 kcal/mol . The pKa of a Tyr residue is usually two units lower than an Arg , which suggests that a deprotonated Tyr419 might activate water while Arg418 remains protonated . Further simulations argue against such a mechanism; instead , the deprotonated , negatively charged Tyr419 interacts strongly with positively charged Arg418 and cannot interact with water . Therefore , Tyr419 is unlikely to play the role of general base in the wild-type enzyme . However , the situation changes when Arg418 is substituted with Gln: now the Tyr419 phenolate can accept a proton from water . The barrier is approximately 17 kcal/mol ( Figure 3 ) . Assuming a pKa of 10 , as is usual for a Tyr residue , then deprotonation of Tyr419 will further increase the barrier to 21–22 kcal/mol , which is very similar to the barrier observed in the reactions of the Arg418Gln and Arg418Ala variants ( ∼20 kcal/mol [10 , 11] ) . As above , the simulations suggest that proton transfer is rate limiting and essentially complete in the transition state . Whereas the landscape contains a shallow valley suggesting the presence of a hydroxide intermediate , the barriers are less than a kT , so the intermediate would not have a finite lifetime . This simulation is generally consistent with experiments , where SIEs of 3–5 are observed when Arg418 is substituted [11] . However , the magnitude of these SIEs is greater than expected if the transition state is indeed late as suggested by the simulations . Interestingly , no activity is observed in the Arg418Ala/Tyr419Phe double mutant , though this fact may equally well be attributed to the inability to form the closed conformation required for the hydrolysis reaction as to the loss of the general base catalyst [12] . Together , the simulations and experiments suggest Tyr419 may act as a surrogate general base in the absence of Arg418 . Similar surrogate residues have been invoked to explain residual activity in other enzyme systems [26] . In RNase T1 , His40 residue assumes the role the general base when Glu58 is substituted with Ala [27] . Similarly , in ketosteroid isomerase , Asp99 may catalyze proton transfers in the Asp38Ala variant [28] . Water or buffer molecules can also replace the function of missing catalytic residues [29 , 30] . These examples illustrate the resilience and plasticity of enzyme catalysis . Surprisingly , the simulations suggest a second pathway for water activation when the starting condition is protonated Arg418: Thr321 abstracts a proton from water while simultaneously transferring its own proton to Glu431 ( Figure 4 ) . As in the Arg418 pathway , the developing hydroxide is stabilized by Tyr419 and another water molecule , and the hydroxide attack occurs instantaneously; the protonated Arg418 also stabilizes the developing hydroxide by 1–2 kcal/mol . The calculated free energy barrier for the Thr321 pathway is approximately 20 kcal/mol . The proton transfers are simultaneous and rate limiting . When a simulation was performed with Glu431 treated as a molecular mechanical ( MM ) residue , which eliminates the possibility of proton transfer while maintaining electrostatic interactions , the energy barrier increases to at least 35 kcal/mol ( Figure S2 ) . Likewise , when Glu431 is substituted with Gln , the barrier increases to at least 38 kcal/mol . Therefore , the presence of Glu431 is essential for the operation of the Thr321 pathway . The simulations suggest that the Thr321 pathway is favored at low pH , whereas the Arg418 pathway becomes dominant at high pH , which predicts that the pH-rate profile will shift to the right when the Thr321 pathway is disrupted by the Glu431Gln mutation . This prediction was confirmed experimentally ( Figure 5 ) : the Glu431Gln mutation shifts the pKa from 7 . 2 ± 0 . 1 to 7 . 6 ± 0 . 1 , but has only a small effect on the pH-independent value of kcat ( kcat = 2 . 2 and 1 . 4 s−1 for wild type and Glu431Gln , respectively; these values are in good agreement with previous reports [31 , 32] ) . Assuming that the pKa shift is entirely attributable to the loss of the Thr321 pathway , the barrier for the Thr321 pathway is approximately 19 kcal/mol , as predicted by the simulations ( see Figure S3 ) . When Arg418 is substituted with Gln , the barrier for the Thr321 pathway is approximately 21 kcal/mol , which is similar to the barrier observed experimentally in the Arg418Ala and Arg418Gln variants [10 , 11] . Therefore , both the Thr321 pathway and the Tyr419 pathway can account for the residual activity of the Arg418Ala and Arg418Gln variants . However , since the Thr321 pathway involves the simultaneous transfer of two protons , this pathway can account for the large solvent isotope effects observed in the Arg418 variants ( SIE = 3–5 [10 , 11] ) . Therefore , we constructed the Arg418Gln/Glu431Gln variant , which should disrupt the Thr321 pathway but leave the Tyr419 pathway intact . The simulations predict that the activity of this variant should be approximately the same as the Arg418Gln , but that the solvent isotope effect should be reduced . These predictions were confirmed in subsequent experiments: ( 1 ) the value of kcat for Arg418Gln/Glu431Gln is decreased by 50% relative to that of Arg418Gln , as expected if the Thr321 pathway was lost ( 0 . 0020 ± 0 . 0002 s−1 and 0 . 0040 ± 0 . 0004 s−1 , respectively ) ; and ( 2 ) though the errors on the SIE are larger than ideal , nonetheless , a smaller SIE is observed in the reaction of Arg418Gln/Glu431Gln , consistent with the loss of the Thr321 pathway ( SIE = 2 . 1 ± 0 . 3 and 2 . 3 ± 0 . 4 for Arg418Gln/Glu431Gln in two independent determinations versus 2 . 9 ± 0 . 5 for Arg418Gln and 5 ± 2 for Arg418Ala [10 , 11] ) . These experiments confirm the operation of the Thr321 pathway in IMPDH . To the best of our knowledge , the presence of dual mechanisms for water activation in an enzyme active site is unprecedented . Why would an enzyme have two pathways to accomplish the same task ? We believe the Thr321 pathway may be vestige of evolution , and phylogenetic analysis is consistent with this hypothesis ( Figure 6; see Figure S4 for the complete phylogenetic tree ) . The closest relative of IMPDH is GMP reductase ( GMPR ) , which catalyzes the conversion of GMP to IMP and ammonia with concomitant oxidation of NADPH ( Figure 6 ) [33] . Cys319 , Thr321 , and Glu341 are also conserved in GMPR , which suggests that these residues were present in the IMPDH/GMPR ancestor . X-ray crystal structures show that the conserved Cys , Thr , and Glu display similar interactions in both GMPR and IMPDH ( Figure 6 ) , suggesting that these residues may have similar functions in both enzymes . To confirm that GMPR activity depends on the presence of Cys186 , Thr188 , and Glu289 , we tested the effect of mutations of these residues on the activity of Escherichia coli GMPR in a complementation assay ( Figure 7 ) . E . coli H1173 requires both adenosine and guanosine for growth due to mutations in both purA and guaC [34] ( Figure 7 ) . Growth on guanosine alone is restored with plasmid pGS682 , which carries the wild-type E . coli guaC gene [35] . However , mutations in Cys186 , Thr188 , and Glu289 clearly compromise the ability of pGS682 to restore GMPR activity , demonstrating that selective pressure exists to conserve these residues . Although the mechanism of the GMPR reaction has not been characterized , some clear parallels can be drawn with the IMPDH reaction , and E-XMP* may well be an intermediate . Importantly , if E-XMP* forms as proposed , the active site must be constructed to prevent the hydrolysis reaction . Kinetic and structural experiments clearly indicate that the reaction only proceeds when NADPH is bound in the active site and can block the access of water [33 , 36 , 37] . Moreover , GMPR does not contain the Arg418-Tyr419 dyad , and the flap is truncated relative to the corresponding region of IMPDH , as expected , given that the hydrolysis of E-XMP* must be avoided during the GMPR reaction . Therefore , the Arg418-Tyr419 dyad could have been installed as IMPDH optimized . Alternatively , the dyad may have been present in the ancestral IMPDH/GMPR , but was subsequently remodeled in the GMPR lineage; since the flap binds in the same site as NAD+ , this scenario suggests that the ancestral IMPDH/GMPR was a hydrolase . While we cannot rule out the latter scenario , we note that IMPDH is a member of the FMN oxidoreductase superfamily of ( β/α ) 8 barrel proteins ( unfortunately , none of these proteins is sufficiently similar to permit rooting of the tree ) [38–40] . Therefore , it seems more likely that the ancestral enzyme was a promiscuous dehydrogenase , and the flap carrying the hydrolase activity was the later addition . In contrast , the Thr321 pathway was likely present in the ancestral IMPDH/GMPR . All IMPDHs and GMPRs contain Thr321 ( Figures 6 and S4 , and Text S1 ) . As noted above , Thr321 also plays a role in the dehydrogenase reaction of IMPDH [11] , which suggests that Thr321 , like Cys319 , was inherited from the ancestral redox enzyme . Glu431 is conserved among GMPRs , suggesting that the Thr321 pathway has a crucial function in this reaction , perhaps operating in the reverse to protonate the ammonia leaving group . Curiously , although Glu431 is highly conserved among IMPDHs , it is substituted with Gln in the eukaryotic branch as well as in a few prokaryotic IMPDHs . We suggest that the ancestral IMPDH/GMPR utilized the Thr321 pathway exclusively , but this pathway became expendable once the Arg418 pathway was established . Phylogenetic analysis is consistent with this view: maximum likelihood analysis indicates that the ancestral enzyme almost certainly contained Glu at position 431 ( probability = 0 . 87 ) [41] . Why then is Glu431 conserved in the majority of prokaryotic IMPDHs ? The presence of the Thr pathway increases turnover , which may be important in maintaining the high concentration of guanine nucleotides required to support the rapid proliferation of prokaryotes . More intriguingly , Glu431 provides 5–10-fold resistance to mycophenolic acid , a natural product that specifically inhibits IMPDH [32] . Approximately 5% of microorganisms contain some mechanism to modify mycophenolic acid , which suggests that this compound is reasonably prevalent in the environment [42] . Indeed , the extraordinary divergence of the adenosine subsite of IMPDH may be a response to the assault of natural product inhibitors such as mycophenolic acid and mizoribine [43] . This divergence occurs despite the multiple functional constraints imposed by interactions with both NAD+/NADH and the flap . The presence of the Thr pathway could facilitate this adaptation , making the evolutionary challenge of the IMPDH reaction much less formidable .
Plasmid pGS682 , a pUC plasmid carrying the 1 . 4-kb guaC insert from pGS89 [35] , was a generous gift from Simon Andrews ( University of Sheffield ) . E . coli strain H1173 was obtained from the E . coli Genetic Stock Center ( Yale University ) . Atoms within a radius of 22 Å around the reaction center were treated as the dynamic region; this region was propagated with regular Newtonian dynamics by applying leapfrog integrator and 1-fs time step . The atoms in the layer between the radii of 22 Å and 25 Å were treated as the buffer region; the heavy atoms in this region were harmonically restrained with the force constants scaled linearly with the distance from the sphere center . The force constants around the boundary of the 25 Å sphere were set as implied by the B factors of the crystal structure . In the buffer region , Langevin dynamics were applied with the friction coefficients also linearly scaled with the distance from the sphere center . The friction coefficients around the boundary 25 Å sphere were set as 60 . CHARMM 22 force fields [22] were utilized as the molecular mechanical potentials in these simulations ( colored in blue in Figures 2–4 ) and SCCDFTB ( self-consistent charge density-functional tight-binding ) method was applied as the quantum mechanical potential on the atoms involved in the chemical reactions ( colored in red in Figures 2–4 ) . For the nonbonded interactions , an extended electrostatic treatment was applied with the electrostatic interactions within 12 Å described by group-based coulombic interactions . IMP , acetylpyridine adenine dinucleotide ( APAD+ ) , Tris , and MES were purchased from Sigma . DTT was purchased from Research Organics . Wild-type and Glu431Gln IMPDH from T . foetus were expressed in E . coli and purified as described previously [10 , 32] . All assays were performed at 25 °C . The release of NADH is partially rate limiting [11 , 31] . Therefore , to ensure that hydrolysis is completely rate limiting , these experiments used APAD+ [31] . Pre-steady-state experiments were performed to demonstrate that hydride transfer and APADH are rapid over the entire pH range ( [11] and unpublished data ) . Standard IMPDH assays contained saturating concentrations of IMP ( 2 mM ) and varying concentrations of APAD+ in 100 mM KCl , 1 mM DTT , and 50 mM of the appropriate buffer ( MES for pH 5 . 0–7 . 0 , and Tris-HCl for pH 7 . 3–9 . 3 ) . Activity was measured by monitoring the absorbance of APADH at 363 nm on a Hitachi U-2000 UV-visible spectrophotometer . Steady-state parameters with respect to APAD+ were derived at saturating IMP concentrations by plotting the initial velocity against APAD+ concentration and fitting to an equation describing uncompetitive substrate inhibition using SigmaPlot ( SPSS ) : where ( kcat ) app are the apparent values at each pH , ( kcat ) indep are the pH-independent values , and Ka is the acid dissociation constant for the most acidic ionization . IMPDH/GMPR amino acid sequences ( IMPDH IPR005990 , GMPR1 IPR005993 , and GMPR2 IPR005994 ) were retrieved from the InterPro database ( http://www . ebi . ac . uk/interpro/ ) . Additionally , BLAST [44] searches with the T . foetus IMPDH ( P50097 ) and human GMPR1 ( P36959 ) amino acid sequences were performed . Sequences from the BLAST search that were already part of the InterPro dataset were removed , and an initial multiple sequence alignment was performed with MUSCLE [45] . A neighbor joining tree ( unpublished data ) was constructed in PAUP* 4 . 0b10 [46] , and 95 sequences were selected for a Bayesian phylogenetic analysis . The sequences of this subset were realigned with Espresso [47 , 48] . A Bayesian phylogenetic analysis was performed with the parallel version of MrBayes 3 . 1 . 2 [49 , 50] . Amino acid substitution rates and state frequencies were fixed to the WAG parameters [51] . A uniform ( 0 . 0 , 200 . 0 ) prior was assumed for the shape parameter of the gamma distribution of substitution rates [52] , an unconstrained exponential prior with rate 10 . 0 for branch lengths , and all labeled topologies were a priori equally probable . Two independent MCMC analyses were run , each with one cold chain and three heated chains , with the incremental heating schema implemented in MrBayes ( λ=0 . 2 ) . Convergence was assumed after the topology samples from the two cold chains had reached an average standard deviation of split frequencies of less than 0 . 01 ( after 1 , 610 , 000 generations ) . Accession numbers , detailed results , and the full tree are found in Text S1 . E . coli strain H1173 ( F- , guaC23 , tonA2 , proA35 , lacY1 , tsx-70 , supE44 ? , gal-6 , l- , trp-45 , tyrA2 , rpsL125 , malA1 ( lR ) , xyl-7 , mtl-2 , thi-1 , purH57 ) contains mutations in purH and guaC , and therefore requires both adenosine and guanosine for growth . Bacteria were transformed with pGS682 carrying either the wild-type guaC gene or variants containing C186A , T188A , and E289Q mutations . Cultures were grown overnight in LB or LB/ampicillin and 5 μl of 1/20 serial dilutions were plated on M9 minimal media containing 0 . 5% casamino acids , 100 μg/ml l-tryptophan , 0 . 1% thiamin , 50 μg/ml guanosine , and/or 50 μg/ml adenosine . | Many enzymes have the remarkable ability to catalyze several different chemical transformations . For example , IMP dehydrogenase catalyzes both an NAD-linked redox reaction and a hydrolase reaction . These reactions utilize distinct catalytic residues and protein conformations . How did Nature construct such a complicated catalyst ? While using computational methods to investigate the mechanism of the hydrolase reaction , we have discovered that IMP dehydrogenase contains two sets of catalytic residues to activate water . Importantly , the simulations are in good agreement with previous experimental observations and are further validated by subsequent experiments . Phylogenetic analysis suggests that the simpler , less efficient catalytic machinery was present in the ancestral enzyme , but was lost when the eukaryotic lineage diverged . We propose that the primordial IMP dehydrogenase utilized the less efficient machinery exclusively , and that this mechanism became obsolete when the more sophisticated catalytic machinery evolved . The presence of the less efficient machinery could facilitate adaptation , making the evolutionary challenge of the IMPDH reaction much less formidable . Thus our simulations provide an unanticipated window into the evolution of a complex enzyme . | [
"Abstract",
"Introduction",
"Results",
"and",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"biochemistry",
"computational",
"biology"
] | 2008 | An Enzymatic Atavist Revealed in Dual Pathways for Water Activation |
Dengue virus ( DENV ) circulates in human and sylvatic cycles . Sylvatic strains are both ecologically and evolutionarily distinct from endemic viruses . Although sylvatic dengue cycles occur in West African countries and Malaysia , only a few cases of mild human disease caused by sylvatic strains and one single case of dengue hemorrhagic fever in Malaysia have been reported . Here we report a case of dengue hemorrhagic fever ( DHF ) with thrombocytopenia ( 13000/µl ) , a raised hematocrit ( 32% above baseline ) and mucosal bleeding in a 27-year-old male returning to Spain in November 2009 after visiting his home country Guinea Bissau . Sylvatic DENV-2 West African lineage was isolated from blood and sera . This is the first case of DHF associated with sylvatic DENV-2 in Africa and the second case worldwide of DHF caused by a sylvatic strain .
Dengue viruses ( DENV ) are the most widely distributed arboviruses in the world . The four distinct serotypes belong to the family Flaviviridae and are some of the most important vector-borne pathogens of humans [1] . DENV circulation occurs in two cycles: an endemic/epidemic cycle between humans and peridomestic mosquitoes , Aedes aegypti and Ae . albopictus , and a sylvatic enzootic cycle between non-human primates and several arboreal Aedes species . In Asia , circulation of sylvatic DENV-1 , -2 and -4 has been detected in Ae . niveus mosquitoes and/or sentinel monkeys [2] . Although sylvatic DENV-3 strains have not been isolated to date , they are believed to circulate in Malaysia based on seroconversion of sentinel monkeys [3] , [4] . In Africa , only sylvatic DENV-2 has been reported as associated with Ae . luteocephalus , Ae . taylori , Ae . furcifer and other forest gallery mosquitoes [5] . Sylvatic DENV strains are ecologically and evolutionary different from endemic strains [6] and the Asian ( Malaysia ) and West African lineages of sylvatic DENV-2 are clearly divergent [7] . Since the 1970s in Africa , sylvatic DENV-2 has been detected primarily in mosquitoes in south eastern Senegal ( Kedougou region ) , but also in Guinea , Ivory Coast and Burkina Faso [5] , [7] , [8] , [9] with sporadic isolations from humans and monkeys [10] , [11] , [12] , [13] . Epizootic DENV-2 was first described in the region of Kedougou in Senegal in 1974 , and successively in 1980-82 , 1989-90 and 1999-2000 , with a periodicity of approximately 8 to 9 years [5] . The first isolation of sylvatic DENV-2 from a human was reported in 1970 in Bandia , Senegal ( sixty kilometers east of Dakar ) from a young girl [14] , [15] . Subsequently , DENV-2 was isolated in 1983 from a French expatriate in south western Senegal in the Casamance region [13] bordering the Gambia and Guinea Bissau ( Figure 1 ) and in 1990 in south eastern Senegal [10] . The first documented outbreak of sylvatic DENV in humans was identified in a retrospective study on samples collected in Ibadan , Nigeria from August 1964 to December 1968 [16] . Thirty-two strains of DENV were recovered of which 14 were DENV-2 [12] . Three complete genomes from samples taken in 1966 were obtained and analyzed and found to be West African sylvatic DENV-2 [16] . These findings suggested that DENV-2 was enzootic in West Africa at least 40 years ago with a limited spillover into humans . More recently , in Malaysia in 2005 , a DENV-1 detected from a febrile patient clustered with the ancestral sylvatic DENV-1 isolated from a sentinel monkey in 1972 ( strain P72_1244 ) [17] . However , recent evidence based on complete genome phylogenetic analysis suggests this strain falls into the human dengue diversity of endemic DENV-1 [18] . The epidemic and pathogenic potential of sylvatic DENV strains remains controversial . Some authors argue that sylvatic viruses do not pose a threat to public health [19] . In 2008 , two events were linked to sylvatic dengue virus activity in humans in West Africa and Asia . Two imported dengue fever cases were reported in France in travelers returning from West Africa , Mali and Senegal [20]; concurrently a dengue fever outbreak occurred in a town 450 km from Bamako , Mali [21] . Genetic analysis of isolates from both sites revealed 99 . 6% sequence identity; phylogenetic analysis revealed that all sequences clustered with the West African sylvatic DENV-2 [21] . Almost simultaneously , the first documented case of dengue hemorrhagic fever ( DHF ) involving sylvatic DENV was observed in Malaysia [22] . A student returning from a trip to peninsular Malaysia was diagnosed with grade II DHF associated with sylvatic DENV-2 . This 2008 Malaysian strain sequence [22] clustered with the sequences of Asian sylvatic dengue strains recovered from monkeys in the 1970s . Here , we report the first African case of DHF associated with sylvatic DENV-2 of a West African lineage .
A previously healthy young male native to Guinea Bissau , but living in Spain , was admitted to Emergency Services at the Infanta Cristina Hospital , Madrid , Spain . The patient had returned to Madrid after a 3-month trip to his home country . When admitted , the patient presented with fever , headache , conjunctival hyperemia , and musculoskeletal pain . He did not have nausea , vomiting , voiding symptoms , or skin lesions . Cardiopulmonary examination yielded no findings . The patient was normotensive with good baseline oxygen saturation . He had not received any type of prophylaxis for malaria or other agents before the trip . During hospitalization , blood samples were obtained for virological and serological testing at the National Centre of Microbiology ( CNM ) , Instituto de Salud Carlos III . Dengue case classification and management was performed according to WHO criteria [23] and subsequently adjusted to the new WHO/TDR guidelines [24] . The institutional review board at the Instituto Carlos III approved this study under the PI08/0834 identification . Through this protocol , the patient's identity was un-linked from the sample , test result , and Genbank accession number . EEB-17 is an arbitrary identifier assigned to the viral culture sample used to sequence the genome and thus cannot be linked with the patient's identity . Whole blood and serum obtained at 5 and 6 days post onset respectively were processed for virological and serological tests ( Table 1 ) . Viral isolation was attempted from serum and blood samples inoculated into C6/36 mosquito cells . Five days post infection; cells were harvested after observation of evident cytopathic effect . This viral isolate was assigned the strain name EEB-17 . DENV-2 was identified by indirect immunofluorescence assay using the commercial monoclonal antibody 8702 ( clone 3H5-1 ) , ( Chemicon , Temecula , CA , USA ) . Viral supernatant was inactivated in lysis buffer ( AVL , Qiagen , Valencia , CA , USA ) for nucleic acid extraction and PCR amplification . IgM and IgG antibodies were determined by capture and indirect-ELISA , respectively ( Panbio , QLD , Australia ) . As previously described [25] , IgM reactivity was confirmed by the performance of a background assay ( e . g . parallel assessment with presence or absence of antigen ) . NS1 antigenemia was determined by using PLATELIA dengue NS1 capture Antigen ELISA ( BioRad , Marnes la Coquette , France ) . Finally , to differentiate primary from secondary infection an avidity IgG test was done [25] . Sera , blood and viral supernatants , inactivated with AVL/carrier RNA solution Buffer ( Qiagen ) , were RNA extracted with the QIACube ( Qiagen ) extractor . RT-PCR and product sequencing were performed as previously described [26] . A different RT-PCR technique that amplifies 1019 bp of the NS5 flaviviral gene [27] was used as a confirmatory test in the supernatants of the viral isolates . RT-PCR products were detected by gel electrophoresis and purified by the QIAquick PCR purification kit ( Qiagen ) and sequenced with ABI prism Big Dye terminator cycle sequencer v3 . 1 ready reaction ( Applied Biosystem , Foster City , CA , USA ) and analyzed on the ABI prism 377 DNA Analyzer ( Applied Biosystem ) . To sequence the genome , primers were designed to produce amplicons of 1000 bps with 500 bp overlaps based on Dengue virus type 2 isolate Dak Ar D75505 , complete genome ( EF457904 ) , which was , based on similarity analysis of the consensus NS5 PCR amplicon , the closest relative to our strain ( Table S1 ) . A total of 31 amplicons , approximately 900 bps in length were used to produce 62 chromatograms . This provided greater than four times coverage of the complete genome with 400–450 bp overlap amongst one another . The accession number of the complete genome of EEB-17 is JF260983 , which is 10 , 176 bps long and includes the complete open reading frame of the polyprotein . Conventional PCRs were performed with HotStar polymerase ( Qiagen ) on PTC-200 thermocyclers ( Bio-Rad , Hercules , CA , USA ) : an enzyme activation step of 5 min at 95 °C was followed by 45 cycles of denaturation at 95°C for 1 min , annealing at 55°C for 1 min , and extension at 72°C for 1 min . Amplification products were size fractionated by electrophoresis in 1% agarose gels , purified ( MinElute , Qiagen ) , and directly sequenced in both directions with ABI PRISM Big Dye Terminator 1 . 1 Cycle Sequencing kits on ABI PRISM 3700 DNA Analyzers ( Applied Biosystems ) . To determine the evolutionary history of isolate EEB-17 , we performed a phylogenetic analysis on all available complete genome sequences of DENV-2 ( total alignment length of 10 , 173 nt; 906 sequences , Oct 15 , 2010 ) . First , we performed a clustering analysis to group available sequences at 97; 95 and 90% similarity using the CD-HIT clustering program at http://weizhong-lab . ucsd . edu/cdhit_suite/cgi-bin/index . cgi . This step allowed us to exclude similar sequences , identify clusters corresponding with known lineages and identify the more divergent members of each cluster for proper phylogenetic analysis . After identification of clusters in correspondence with the accepted groupings of DENV-2 genotypes , we identified representative sequences from each separate cluster to perform deep phylogenetic analysis . Specifically , we inferred a Maximum Clade Credibility ( MCC ) tree using the Bayesian Markov Chain Monte Carlo ( MCMC ) method available in the Beast package [28] , thereby incorporating information on virus sampling time . This analysis utilized a strict molecular clock and a GTR+Γ model of nucleotide substitution for each codon position , although very similar results were obtained using other methods ( data available upon request ) . The analysis used a Bayesian skyline model as a coalescent prior as was recently reported in a similar study [22] . All chains were run until convergence for all parameters with 10% removed as burn-in . The analysis also allowed us to estimate divergence times for each node in the DENV-2 genotypes . The degree of uncertainty in each parameter estimate is provided by the 95% highest posterior density ( HPD ) values , while posterior probability values provide an assessment of the degree of support for each node on the tree .
On November 9 , 2009 , a 27-year-old man travelling from Guinea Bissau was admitted to the Hospital Infanta Cristina in Parla , Madrid , after returning from his home country . He had spent 3 months in the city of Canchungo , in the northwestern coast of Guinea Bissau . The patient had initiated his return trip from Canchungo to Dakar , Senegal on October 28 , by car , through the Casamance region in south-western Senegal ( Figure 1 ) . He stayed in Dakar for approximately a week before travelling by air to Spain . On November 6 , the patient arrived in Madrid . He reported feeling ill during the flight . At admission , he presented with fever ( 38 . 6 °C ) , conjunctival hyperemia , headache and musculoskeletal pain but no nausea or vomiting , voiding symptoms , or skin lesions . During his intake history , the patient reported insect bites during his recent travel to Dakar . Laboratory investigation demonstrated leukopenia with marked left shift , and significant thrombocytopenia ( 13000/µl ) . No alteration of clotting time was observed . Renal function was normal . Liver function impairment was demonstrated by elevated lactate dehydrogenase ( LDH ) , alanine transminase ( ALT ) and aspartate transaminase ( AST ) . Normal values were obtained for total bilirubin ( TBIL ) and direct bilirubin , gamma glutamyl transpeptidase ( GGT ) , amylase and lipase demonstrating normal cholestasis data ( Table 1 ) . Radiological studies of the chest and abdomen were normal . After admission to internal medicine service for fever and thrombocytopenia , the patient experienced epigastric pain , without hemodynamic compromise . Follow-up analytical testing showed a progressive decline in platelet numbers , elevated liver enzymes and hemoglobin with normal bilirubin . Malaria , parasites in blood , feces and urine , as well as HIV , HAV , HCV and HBV infections were ruled out . Considering the epidemiological context and data concerning the fever , thrombocytopenia , abdominal pain , elevation of liver enzymes and negative results for malaria , dengue virus infection was suspected . One day after admission , samples were taken and sent to the CNM for dengue PCR and the next day for serology . During an abdominal ultrasound no evidence of injury to the liver or spleen was observed . Gallbladder and bile ducts were also normal . On the other hand , the presence of free liquid in the sub-hepatic area and pouch of Douglas were remarkable . During hospitalization , the patient also experienced episodes of self-limited epistaxis , bleeding from venipuncture sites , episodes of hemoptoic sputum without respiratory or hemodynamic impact , and increased hyperemia with self-limited subconjunctival hemorrhage . Furthermore , the hemoconcentration observed on the 5th day after onset was very high having risen more than 32% above baseline . While admitted , the patient was only treated with supportive care ( hydration and non-inflammatory analgesia ) , without requiring platelet transfusion . After four days hospitalized , the number of platelets , liver function and general clinical symptoms started to show improvement; no new episodes of bleeding or abdominal pain were observed . Eight days after admission the patient was discharged . According to the original WHO definition of DHF [23] , the case met the four criteria for classification as grade II DHF ( fever , thrombocytopenia , hemorrhage and plasma leakage ) . However , according to the newly proposed WHO definition [24] , the case would be classified as “dengue with alarm signs . ” By day 22 , the patient was clinically asymptomatic . All analytical parameters were normal with a slight increase of AST and ALT ( Table 1 ) . DENV-2 was found in blood and sera in the samples taken 4 and 5 days post onset of fever by routine consensus PCR over the E/NS1 junction [26] . PCR amplification in the NS5 was also performed to confirm the initial results ( PCR E/NS1 and IFI ) . BLAST analysis of the sequence obtained from the PCR products gave a similarity index of 97% with sylvatic West African lineage DENV-2 ( Accession number EF457904 ) . Both IgM and IgG were positive in the sera taken at day 6 after onset . The quantitative results obtained ( 40 . 9 for IgM and 58 . 8 for IgG ) suggested a secondary infection . The avidity assay characterized the IgG response as high avidity . NS1 antigen was detected in the sample at high concentration . Phylogenetic analysis from the E/NS1 and NS5 regions revealed that the sequences fell into DENV 2 sylvatic genotype group , within the African isolates ( data not shown ) . Complete genome phylogenetic analysis confirmed that EEB-17 fell within the West African sylvatic lineage of DENV-2 and was most closely related to isolates DakAr141069 and DakAr141070 [7] ( Figure 2 ) . Both isolates were detected in 1999 in Ae . luteocephalus mosquitoes from Kedougou region ( South-eastern Senegal ) [5] . Interestingly , sylvatic strains of African origin showed a clustering by year of isolation . This phylogenetic history was supported by high posterior probability values ( 1 . 0 ) . Molecular clock analysis suggests that the African sylvatic lineage shares a common ancestor approximately 103 years ago ( 95% HPD of 66–153 years ) . Thus , DENV-2 has likely been circulating in non-human primates in Africa for at least this long . Our analysis also predicts the Time to the Most Recent Common Ancestor ( TMRCA ) for each of the human genotypes demonstrating that the divergence of the Asian and African lineages of sylvatic DENV-2 is comparable to the observed difference among human lineages . The mean evolutionary rates estimated ranged from 3 . 7×10−4 substitutions per site per year ( 95% HPD , 1 . 8×10−4 to 5 . 9×10−4 ) .
Although sylvatic dengue cycles occur in West African countries and Malaysia , only a few cases of human disease caused by sylvatic strains have been reported [5] . Given that infection with sylvatic DENV may result in clinical presentation indistinguishable from that presented due to infection with strains from the human transmission cycle , it is possible that cases of sylvatic dengue are underreported . Nevertheless , after 30 years of silence , sylvatic dengue has re-emerged causing severe disease in a young man [22] . Although silent transmission during those 30 years is possible ( human or sylvatic ) , no reports suggest it has occurred . We describe here an imported DHF case caused by a sylvatic strain in a healthy man returning to Madrid from Guinea Bissau through Senegal . This is the first report of DHF caused by a sylvatic DENV-2 West African lineage virus . The patient had dengue fever with alarming symptoms , such as haemorrhagic manifestations ( mucosal bleeding , hematocrit 32% above baseline ) , thrombocytopenia and fluid accumulation in the abdominal cavity associated with abdominal pain . Under the old WHO definition , the patient classifies as grade II DHF with a risk of developing into grade III [23] . As indicated by the IgG avidity test , the patient showed high avidity antibodies suggesting a secondary flavivirus infection . Secondary infection with sylvatic DENV may have resulted after a primary infection with another dengue serotype , considered a risk factor for development of DHF[29]; after a primary infection with another flavivirus , since such viruses circulate throughout Guinea Bissau and neighbouring nations [30]; or after the development of immunity resulted by YFV vaccination . The IgM/IgG ratio or IgG avidity index has limitation to differentiate true dengue secondary infection in individuals with previous immunity against other flavivirus[25] . Evidence of previous infection with another DENV serotype might explain disease severity . Lack of evidence would suggest severe disease in a primary case of dengue caused by a sylvatic strain . Unfortunately , we cannot demonstrate previous infection with another DENV serotype . DENV is endemic in West Africa; and the predominant serotype with sustained circulation in the region is DENV-2 from sylvatic or epidemic lineage [31] . DENV-3 was recently detected in Ivory Cost [32] and it continues to expand throughout the region [20] , [33] . In Dakar , Senegal , DENV-3 was detected in 2009 [20] . Based on the incubation period of dengue and previous knowledge describing the zone of circulation of sylvatic DENV-2 strains in Senegal , we posit that the infection was acquired during the overland trip from Canchungo , Guinea-Bissau to Dakar , Senegal , via Casamance , a region of documented sylvatic DENV activity [13] . The following points were considered to build that hypothesis: i ) sylvatic DENV-2 has been frequently isolated in Senegal [5]; ii ) our isolate EEB-17 is closely related to isolates obtained from Ae . luteocephalus captured in Senegal in the year 1999; iii ) absence of sylvatic dengue circulation in Guinea-Bissau; iv ) During October 2009 , DENV-3 circulation was identified in Dakar [20] , [34] , the same time this DENV-2 case was identified . Moreover , outbreaks in Dakar and in the Kedougou region were reported [20] . In the Kedougou region , a sylvatic amplification cycle was detected with the isolation of sylvatic strains during surveillance activities by Institute Pasteur of Dakar ( Dr . Amadou Sall , personal communication ) ; and vi ) Circulation of sylvatic DENV-2 would follow the same pattern of periodicity of the epizootics in the area with an interval cycle of approximately 8–9 years [5] . In addition , our complete genome phylogenetic analysis revealed that EEB-17 belongs to the West African lineage , along with isolates post-1980 . According to the molecular clock , the African sylvatic lineage has shared a common ancestor approximately 103 years ago . These values are in agreement with previous reports [7] , [22] . The mean evolutionary rates estimated were in complete concordance with the rates obtained by Vasilakis et al [7] . Introduction of DENV in Europe was observed in the summer of 2010 . Autochthonous circulation of DENV was detected in France [35] and Croatia [36] . Mosquito surveillance in the countries on the Mediterranean basin demonstrates the presence of Ae . albopictus , and highlights the risk of initiation of a local transmission cycle in the presence of high vector densities . In that context , DENV strains could be introduced in naïve countries with the potential to develop limited or even extensive outbreaks , such as occurred with the introduction of Chikungunya virus in Italy in 2007 [37] . In summary , we described here the first case of DHF grade II caused by a sylvatic DENV-2 belonging to a West African lineage virus . The findings suggest that sylvatic strains of DENV might have a greater pathogenic potential than previously thought [19] . | Dengue viruses are mosquito borne pathogens that cause up to 100 million infections annually . Dengue viruses circulate in either humans or non-human primates in two separate cycles: human and sylvatic . All four different serotypes infect humans causing frequent epidemics and severe disease , including hemorrhagic fever . Until a few years ago , viruses circulating in the sylvatic cycle had been associated with asymptomatic to mild disease only in humans . For example , retrospective analysis of human serum samples collected in Nigeria from the 1960s proved that sylvatic dengue was able to cause an outbreak of dengue fever in human population . Recently a case of DHF in Asia was linked to sylvatic DENV-2 ( strain DKD811 ) infection . Here , we report the first case of DHF associated with sylvatic DENV-2 infection in Africa caused by a complete different lineage of the virus . All these events underscore the need to be vigilant about the re-emergence of DHF from sylvatic cycles . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"infectious",
"diseases",
"emerging",
"infectious",
"diseases",
"virology",
"emerging",
"viral",
"diseases",
"dengue",
"fever",
"neglected",
"tropical",
"diseases",
"travel-associated",
"diseases",
"biology",
"microbiology"
] | 2011 | First Report of Sylvatic DENV-2-Associated Dengue Hemorrhagic Fever in West Africa |
In reward learning , the integration of NMDA-dependent calcium and dopamine by striatal projection neurons leads to potentiation of corticostriatal synapses through CaMKII/PP1 signaling . In order to elicit the CaMKII/PP1-dependent response , the calcium and dopamine inputs should arrive in temporal proximity and must follow a specific ( dopamine after calcium ) order . However , little is known about the cellular mechanism which enforces these temporal constraints on the signal integration . In this computational study , we propose that these temporal requirements emerge as a result of the coordinated signaling via two striatal phosphoproteins , DARPP-32 and ARPP-21 . Specifically , DARPP-32-mediated signaling could implement an input-interval dependent gating function , via transient PP1 inhibition , thus enforcing the requirement for temporal proximity . Furthermore , ARPP-21 signaling could impose the additional input-order requirement of calcium and dopamine , due to its Ca2+/calmodulin sequestering property when dopamine arrives first . This highlights the possible role of phosphoproteins in the temporal aspects of striatal signal transduction .
Reinforcement learning plays an important role in building the learned-behavioral repertoire of an organism . It operates by updating the salience of an environmental cue or a cue-response association which has elicited a reward in the past . Basal ganglia are critical for reward learning and the input nucleus , striatum , is the locus of integration for the environmental and the reinforcement signals [1] . The environmental stimuli are largely conveyed to the striatum by the cortical glutamatergic afferents which converge onto the striatal medium-sized spiny neurons ( MSNs ) . The incoming glutamatergic activity leads to an influx of calcium ions through N-methyl-D-aspartate receptor ( NMDAR ) in the postsynaptic MSNs [2–4] . On the other hand , the reinforcement signal is encoded by the dopaminergic inputs from the mid-brain which activates the dopamine D1 receptors ( D1R ) in one of the MSN populations [5 , 6] . The postsynaptic integration of NMDAR-mediated calcium and dopamine-dependent D1R signaling leads to the potentiation of corticostriatal synapses on D1R-expressing MSNs , thus resulting in reward-learning [7–9] . Ca2+-Calmodulin-dependent kinase II ( CaMKII ) and Protein-Phosphatase 1 ( PP1 ) signaling plays an important role in this process [9] . There are several substrate proteins , like receptor subunits and translational regulators , which are phosphorylated by CaMKII and dephosphorylated by PP1 , and this in turn could influence synaptic strength [10 , 11] . However , in order to elicit the CaMKII/PP1-dependent downstream response in MSNs , the calcium and dopamine inputs should fulfill two temporal requirements [9] . Specifically , the calcium and dopamine inputs should be in temporal proximity ( input-interval constraint ) and the dopamine should follow , and not precede , the calcium input ( input-order constraint ) . Thus , only those calcium and dopamine signals which adhere to these constraints are effective to produce corticostriatal potentiation [9] . Despite the physiological significance of these temporal constraints , the molecular mechanism underlying their emergence is not clear . In this computational study , we investigated the integration of calcium and dopamine signals by D1R-expressing MSNs to explain the emergence of the aforementioned temporal constraints , using quantitative kinetic modeling . Our results suggest that DARPP-32 ( Dopamine and cAMP-regulated Phosphoprotein 32kDa ) could play an important role in the dopamine-dependent gating of the calcium signaling , which aligns with previous experimental observations [9 , 12] . DARPP-32 is believed to be an important integrator of calcium and dopamine signaling in striatum [13] . According to our simulations , the transient nature of DARPP-32 signaling could be responsible for the emergence of the requirement regarding temporal proximity of calcium and dopamine signals in the integration process . However , it appears that this DARPP-32 mediated signaling alone could not distinguish the temporal order of calcium and dopamine signals . Therefore , it may not be sufficient to explain the emergence of the input-order constraint . We propose that another striatally-enriched phosphoprotein , ARPP-21 ( cAMP-Regulated Phosphoprotein 21kDa ) , has the potential to introduce the input-order dependency into the integration process . ARPP-21 , upon dopamine-dependent phosphorylation , has the ability to bind with Ca2+/calmodulin , thus affecting the calcium signaling [14 , 15] . In our signaling model , ARPP-21 imposes the input-order constraint by implementing an input-order dependent threshold-like function for CaMKII activation . Thus , our results predict an important mechanistic role for ARPP-21 , whose physiological relevance remains elusive , in the context of striatal reward learning . Moreover , in the case of a multi-trial scenario , an inter-trial refractoriness could also emerge due to ARPP-21 signaling . Thus , this study puts forth a novel , but readily testable , mechanism which could explain various aspects of the striatal calcium-dopamine integration . In general , it also highlights the possible role of regulatory-phosphoproteins in shaping the temporal aspects of subcellular signal integration . Such a phosphoprotein dependent mechanism could represent a more generic signaling motif for different brain regions where DARPP-32 and ARPP-21 are expressed and input timing is crucial .
It is known that the integration of transient calcium and dopamine signals leads to synaptic potentiation of the corticostriatal synapses and this process is mediated by CaMKII activity [8 , 9 , 17] . CaMKII activation leads to the phosphorylation of its substrate but this could be counteracted by PP1 [9] . Thus , we read out the phosphorylation level of a generic CaMKII/PP1 substrate to understand the calcium-dopamine integration . Several quantitative molecular phenotypes have been used to constrain the model . Similar to the previous modeling studies , here also the term “phenotype” has been used to refer a set of observables in a specific experimental setting [28 , 50] . These target phenotypes , most of which are MSN specific , have been collected from published literature and are listed in Table 1 . The phenotypes which we used are either one-time measurement for a specific stimulus ( “SliceDA” , “SliceCa” , “SlicePDE10Inhibition”; refer Table 1 ) , time-series measurement ( “ReceptorGProtein” , “SlicePKADynamics” , “PDEKinetics” , “CaMKIIDynamics”; refer Table 1 ) or dose response ( “DoseResponse”; refer Table 1 ) . The comparison between the model output and the experimental data for the one-time measurements has been done using fold change of the respective marker . The time-series data have been compared between the model and experiments by parameterizing them into simple functions , like monoexponentials ( “ReceptorGProtein” , “PDEKinetics” , “CaMKIIDynamics” ) and difference of two exponentials ( “SlicePKADynamics” ) . The dose response data has been parameterized with Hill equation for the purpose of comparison between the model and experiment . Apart from comparing the stimulated state of various effectors , the model has also been constrained by the known basal level of various effectors ( “Basal”; refer Table 1 ) . The total amounts of various species have also been taken from published literature wherever possible . In the case of the model without ARPP-21 ( Fig 1A ) , the ARPP-21 related phenotypic variables ( in “Basal” and “SliceDA” ) have been ignored . Figs 1B and 2B show the comparison between the experimental and simulated values of phenotypic variables for the signaling models without and with ARPP-21 . The differences are minor . The current biochemical-reaction model considers two input signals: ( 1 ) dopamine and ( 2 ) calcium . The dopamine input represents increased extracellular dopamine concentration due to the burst activity of dopaminergic neurons in response to an unexpected reward [5] . Such a burst could lead to a dopamine peak of around 1 . 5μM amplitude [56] . We use the same amplitude in our model . The basal concentration of dopamine is assumed to be 20nM in this study . On the other hand , the calcium input in the model represents an increase in the intracellular calcium concentration in response to 10 glutamate-triggered calcium pulses at 10 Hz . These calcium input parameters are in accordance with the previously published experiments where the temporal requirement of calcium and dopamine integration have been observed [9] . The raise and decay time-constants of individual calcium pulses are obtained from the observed kinetics of intracellular NMDAR calcium [57] . The amplitude is set such that the maximum intracellular calcium concentration reaches around 5μM . The basal concentration of the intracellular calcium is assumed to be 60nM in this study .
To investigate the integration of dopamine and calcium inputs by the subcellular signaling processes in the D1R expressing MSNs , we developed a kinetic model of the standard striatal signaling network , Fig 1A [28 , 50 , 58–60] . This kinetic model is comprised of a dopamine/D1R/AC5/cAMP/PKA/DARPP-32 and a calcium/calmodulin/CaMKII signaling axes ( refer “Materials and Methods” ) , and it has been constrained to closely match various known experimental observations , Fig 1B ( refer “Target Molecular Phenotypes” in “Materials and Methods” ) . In order to understand the downstream effect of calcium-dopamine integration on the CaMKII/PP1 signaling , we read out the phosphorylation level of a generic CaMKII/PP1 substrate while simulating the signaling network . This substrate is phosphorylated by CaMKII and dephosphorylated by PP1 , Fig 1A . We first looked at how dopamine could affect the calcium triggered CaMKII signaling . To this end , we measured the difference between the CaMKII/PP1 substrate phosphorylation produced by a transient calcium input ( elevation in intracellular calcium concentration; refer “Materials and Methods” ) with and without an accompanying transient dopamine input ( elevation in dopamine concentration; refer “Materials and Methods” ) . The dopamine input , in this case , follows the calcium input after 1s . The simulation results indicate that the level of substrate phosphorylation is significantly higher when the calcium transient is paired with a dopamine transient compared to the response produced by the calcium alone , Fig 1C . This suggests that a transient dopamine input could gate the calcium triggered CaMKII/PP1-dependent signaling . Since the substrate phosphorylation depends on the coordinated activity of CaMKII and PP1 , we looked at how the input signals affect the level of active CaMKII and PP1 to get further insights . The simulations suggest that calcium leads to CaMKII activation in both the cases with and without dopamine , though there are slight differences , Fig 1D ( lower panel ) . On the other hand , the level of active PP1 , unlike CaMKII , shows a larger difference between the two cases , Fig 1D ( upper panel ) . The level of active PP1 remains high throughout the simulation without much difference when the model is presented only with the calcium input , Fig 1D . However , the active level of PP1 is transiently reduced when a dopamine input is also presented to the model along with the calcium input , Fig 1D . Under the basal condition , the substrate is under dephosphorylation pressure due to the high PP1 level . This is in line with recent experimental observations [9] . Our simulations suggest that even though a calcium input alone could lead to CaMKII activation thereby increasing the phosphorylation of the substrate , this is not sufficient enough to counteract the dephosphorylation pressure of PP1 . The inhibition of PP1 , produced by a concomitant dopamine input , could transiently relieve the dephosphorylation pressure which in turn leads to a significant increase in the resulting substrate phosphorylation and a slight increase in CaMKII activation ( Fig 1C and 1D ) . Dopamine dependent PKA activation leads to the phosphorylation of Thr-34 residue of DARPP-32 and this turns DARPP-32 into a potent inhibitor of PP1 [61] . Thus , to verify whether DARPP-32 is responsible for the dopamine-dependent increase in substrate phosphorylation , we simulated the knocked down of the Thr-34 phosphorylation of DARPP-32 ( T34A ) in our model by removing the respective reaction . A comparison of the substrate response between the default ( wildtype; WT ) and the mutant ( T34A ) models indeed suggests that the dopamine dependent gating is mediated by the phosphorylation of Thr-34 , Fig 1E . In the T34A mutant case , there is no difference between the substrate response produced by calcium alone and calcium accompanied by a dopamine input , Fig 1E . As expected , there is no difference in either active CaMKII or PP1 between with and without dopamine in the case of T34A mutant , Fig 1F . Using the aforementioned signaling network , we then investigated how the substrate phosphorylation depends on the temporal relation between calcium and dopamine inputs . The interval between the inputs is referred as Δt , such that Δt = ( tdopamine—tcalcium ) , where tcalcium and tdopamine are the start time of calcium and dopamine transients , respectively . A small absolute value of Δt indicates that the calcium and dopamine inputs are close in time . A positive value of Δt means that the calcium input is followed by the dopamine input and vice-versa . This relation between the substrate response and Δt could shed light on whether the input-interval and input-order constraints exist in this signaling network . If the substrate response is higher for smaller absolute values of Δt compared to large Δt values then this suggests the existence of an input-interval constraint . In other words , calcium and dopamine should be close in time for effective downstream activation . Similarly , if substrate response is higher for positive values of Δt compared to negative values then this indicates the existence of input-order constraint , i . e . the calcium input should be followed by the dopamine input for effective downstream response . Our simulations with different Δt values suggest that there appears to be a relation between the substrate phosphorylation and Δt , Fig 1G . For smaller absolute values of Δt the substrate phosphorylation is generally higher , Fig 1G and 1H . As the absolute value of Δt increases the downstream response declines , Fig 1G . Thus , these results suggest that the temporal proximity between the calcium and dopamine inputs is important for an effective dopamine-dependent gating thereby indicating the existence of an input-interval constraint in this signaling network . However , there is a slight asymmetry in the gating function around Δt = 0 , e . g . the amplitude of the phosphorylated substrate for Δt = 1s is slightly higher that the amplitude at Δt = -1s , Fig 1H . For longer values of positive Δt ( e . g . Δt = 4s ) , the timecourse of substrate phosphorylation appears to be biphasic , Fig 1H . The first phase corresponds to the start of phosphorylation process under the dephosphorylation pressure of PP1 and the second phase represents the facilitation of the substrate phosphorylation due to the dopamine dependent inhibition of PP1 . Since the signaling network built in this study is quite detailed and contains several parameters , we looked at the robustness of our result towards uncertainty in the parameter-values . In order to test the robustness of the relation between Δt and downstream substrate phosphorylation , we made a 2-fold increase or decrease in the individual parameter value , one at a time , and then simulated the perturbed model . The distribution of the results arising from parameter perturbations suggests that the Δt-dependence in this signaling network is quite robust , Fig 1I . We then investigated the reason for the emergence of the relation between Δt and downstream substrate phosphorylation . Since the effective substrate phosphorylation depends on CaMKII and PP1 , we looked at the behavior of CaMKII and PP1 in order to identify the reason for the existence of the input-interval constraint , Fig 1J . Similar to the above result ( Fig 1D ) , there is CaMKII activation and DARPP-32-mediated PP1 inhibition in response to calcium and dopamine inputs , respectively , Fig 1J ( Δt = 1s ) . This CaMKII activation and PP1 inhibition could be seen for different values of Δt ( 1s , 4s , -1s , -4s ) , Fig 1J . However , the coincidence between the time-window of CaMKII activation and PP1 inhibition is considerably different for various Δt , Fig 1J . This overlap between the CaMKII activation and the PP1 inhibition decreases with an increase in the interval between the calcium and dopamine inputs thus reducing the downstream effect , Fig 1J ( Δt = 4s , -4s ) . This difference in the temporal overlap between the CaMKII and PP1 response leads to the Δt-dependence of the substrate phosphorylation response . As mentioned previously , the PP1 inhibition is produced by dopamine dependent DARPP-32 phosphorylation at Thr-34 . DARPP-32 is also dephosphorylated by PP2B in response to calcium . Thus , it may be possible that different Δt values may lead to different level of DARPP-32 phosphorylation thereby affecting the PP1 inhibition . Thus , we also looked at how the level of phosphorylated DARPP-32 changes for different values of Δt . The simulations indicate that DARPP-32 phosphorylation is not very sensitive to various Δt values around zero , Fig 1K . For negative values of Δt , there is a visible calcium dependent DARPP-32 dephosphorylation . However , this effect is quite small relative to the overall response . Thus , the Δt-dependence of the downstream substrate response in our simulations is largely produced by the difference in the temporal overlap between the transient CaMKII and DARPP-32 mediated PP1 response rather than any regulation of the relative strength of DARPP-32 phosphorylation . The response in this DARPP-32 mediated signaling network depends mostly on the magnitude rather than the sign of Δt , i . e . there is a substantial amount of downstream response even for negative values of Δt , Fig 1G . Thus , the DARPP-32 signaling makes no significant distinction between whether dopamine input precedes or follows the calcium input as long as they are temporally close . In other words , this signaling network could account for the input-interval constraint but not the input-order constraint of the striatal calcium-dopamine integration . Since the DARPP-32-containing signaling network was not sufficient to explain the input-order constraint , we considered an important addition to it . Apart from DARPP-32 , the striatal MSNs also express significant amounts of ARPP-21 [55] . As mentioned above , ARPP-21 has the potential to act as a point of cross-talk between the calcium and dopamine signaling . It could compete with Ca2+/calmodulin dependent proteins for the shared calmodulin resource pool [15] upon dopamine dependent phosphorylation [54] . Thus , we included ARPP-21 into the signaling network to test whether it has the ability to introduce the input-order constraint , Fig 2A ( refer “Materials and Methods” ) . This updated model has also been constrained to closely match various known experimental observations including the known striatal ARPP-21 data , Fig 2B ( refer “Target Molecular Phenotypes” in “Materials and Methods” ) . We then used this updated model to explore the relation between Δt and the substrate phosphorylation . Fig 2C shows the relation between Δt and the substrate phosphorylation for this updated signaling model . The addition of ARPP-21 significantly altered the relation between Δt and the substrate phosphorylation compared to the “without ARPP-21” network ( compare Fig 2C with Fig 1G ) . For the updated signaling network , positive and preferably smaller values of Δt produced higher substrate phosphorylation , Fig 2C . Fig 2D shows the traces of substrate phosphorylation produced for different Δt . Unlike the previous case , the downstream response is highly sensitive to the sign of Δt . If dopamine comes before calcium ( Δt < −0 . 5s ) then the downstream response is significantly lower that the corresponding positive Δt response , Fig 2C . Thus , the addition of ARPP-21 confers the calcium-dopamine integration process with the previously unexplained input-order constraint . We assessed the robustness of the updated signaling network by changing the parameter values , one at a time , by 2-folds . The distribution of the results arising from parameter perturbations suggests that the Δt-dependence in this updated signaling network is also quite robust , Fig 2E . The simulation of Ser-55 mutant of ARPP-21 ( A21S55A ) shows that the phosphorylation of this residue is specifically important for the emergence of the input-order constraint , Fig 2F . In the previous section , we presented that DARPP-32 mediated signaling promotes the CaMKII/PP1 signaling and explain the input-interval constraint on the calcium-dopamine integration . In the current section , we illustrated that including ARPP-21 into the network could lead to the appearance of the input-order constraint in addition to the input-interval requirement . When taken together , the emergence of both input-interval and input-order constraints could be explained by the coordinated activity of DARPP-32 and ARPP-21 . The role of DARPP-32 and ARPP-21 could be further clarified by simulating the signaling network with loss-of-function mutation in DARPP-32 and ARPP-21 . A comparison between simulated wild-type ( WT ) and Ser-55 mutant of ARPP-21 ( A21S55A ) shows that loss-of-function mutation in ARPP-21 abolishes the input-order interval by significantly affecting the response in the negative Δt region , Fig 2F . On the other hand , the difference between the response of WT and Thr-34 mutated DARPP-32 ( D32T34A ) suggests that DARPP-32 is responsible for the overall dopamine dependent gating of the CaMKII/PP1 dependent substrate phosphorylation , Fig 2G . There is no significant substrate response in the case of DARPP-32 mutation , Fig 2G . As shown above , the addition of ARPP-21 into the signaling network could introduce the input-order constraint by significantly reducing the downstream response for the negative Δt regime , Fig 2C and 2D ( compare with Fig 1G ) . Here , we illustrated the mechanism with which the updated model with ARPP-21 enforces this input-order constraint . We looked at how active CaMKII and PP1 levels are affected for different Δt values to understand the relation between substrate phosphorylation and Δt for this updated signaling network , Fig 3A . The CaMKII and PP1 activity patterns for positive Δt values ( Δt = 1s and Δt = 4s ) are not significantly different between the updated network ( Fig 3A ( upper panel ) ) and the network without ARPP-21 ( Fig 1J ( upper panel ) ) . However , there is a significant reduction in the CaMKII activation for negative Δt values in the updated signaling network , Fig 3A ( lower panel; compare with Fig 1J ( lower panel ) ) . There appears to be an input-order dependent threshold-like relation between the amplitude of active CaMKII and Δt , Fig 3B . Specifically , low CaMKII activation level for negative Δt and a sudden increase as the Δt becomes positive . This sigmoid-dependence of CaMKII activation on Δt is mediated by ARPP-21 as shown by the difference between with and without ARPP-21 case , Fig 3B . We then investigated how the ARPP-21 could reduce the CaMKII activation in the negative Δt regime , i . e . only when dopamine input precedes the calcium input . It is known that dopamine leads to the phosphorylation of Ser-55 residue on ARPP-21 and this phosphorylation turns ARPP-21 into an excellent binding partner of Ca2+/calmodulin [15] . This could suggest a possibility of Ca2+/calmodulin sequestration by ARPP-21 thus reducing the CaMKII activation . To ascertain this possibility we first considered the case of Δt = -2s as an example , Fig 3C1 . In this case , the dopamine input precedes the calcium input by 2s . The dopamine input leads to the phosphorylation of ARPP-21 and by the time the calcium input arrives there is already a significant level of phosphorylated ARPP-21 ( pARPP-21 ) , Fig 3C1 ( left panel ) . As expected the calcium transient leads to an increase in the level of Ca2+/calmodulin . A significant part of the Ca2+/calmodulin binds to the pARPP-21 produced by the preceding dopamine , Fig 3C1 ( left panel ) . This effectively reduces the amount of Ca2+/calmodulin available for CaMKII activation , Fig 3C1 ( right panel ) . This reduction in the available Ca2+/calmodulin leads to a lower CaMKII activation . Thus , the sequestration of Ca2+/calmodulin by pARPP-21 , thereby reducing its availability , indeed explains the low CaMKII activation in the negative Δt regime . As the interval between dopamine and calcium decreases in the negative Δt region , e . g . Δt = -1s , the sequestration of Ca2+/calmodulin by pARPP-21 is also reduced , Fig 3C2 ( left panel ) . This in turn leads to more Ca2+/calmodulin available for CaMKII activation , Fig 3C2 ( right panel ) . This is due to the lower level of pARPP-21 at the time of the calcium input; compare between Δt = -2s and Δt = -1s , Fig 3C1 and 3C2 . On the other hand , the situation is quite different for a positive Δt value , Δt = 1s . In this case , there is no significant level of pARPP-21 at the time of calcium incidence because the dopamine input has not yet arrived to phosphorylate ARPP-21 , Fig 3C3 ( left panel ) . Thus , the sequestration of Ca2+/calmodulin by ARPP-21 is negligible , Fig 3C3 ( left panel ) , and the Ca2+/calmodulin available for CaMKII activation is not reduced , Fig 3C3 ( right panel ) . Therefore , the CaMKII activation is affected by the phosphorylation of ARPP-21 only if the value of Δt is negative , i . e . when dopamine input precedes the calcium input . The level of ARPP-21 phosphorylation itself does not appear to be dependent on the value of Δt , Fig 3D . The level of phosphorylated ARPP-21 starts to increase 1s after the arrival of dopamine input and reaching its maximum in ~4s . Thus , the downstream CaMKII activation due to any calcium input arriving after the dopamine input with a delay of around 1s or greater , i . e . Δt < = -1 , is significantly dampened . As highlighted in the previous section , the dopamine-dependent phosphorylation of ARPP-21 results in the dampening of a subsequent calcium-triggered CaMKII activation . In a two-trial scenario , this could mean that the ARPP-21 which is phosphorylated in the first trial may affect the CaMKII activation in the second trial . To test this cross-talk between successive trials we considered the calcium-dopamine integration in a two-trial scenario . The two trials are separated in time by an inter-trial interval ( ITI ) . Each of the trials is represented by a pair of calcium and dopamine inputs with Δt = 1s . We first tested a scenario with an ITI = 10s . In this scenario , the substrate response produced by the first trial was similar to the response observed in a single trial case ( as in the previous sections ) , Fig 4A ( first peak ) , but the response produced by the second trial was significantly reduced , Fig 4A ( second peak ) . The response reduction in the second trial was abolished in the case of the simulated ARPP-21 Ser-55 mutation ( A21S55A ) , Fig 4A . This suggests that ARPP-21 phosphorylation could indeed affect the response produced in a subsequent trial . Fig 4B shows the phosphorylation of ARPP-21 in response to a single dopamine input response . There is a transient elevation in the level of ARPP-21 and then it returns to the basal level in 30s . Thus , phosphorylation dependent inhibitory effect of ARPP-21 on the second trial may not be present if ITI is longer that 30s . To test this we simulated the two trial scenario with an ITI = 30s . As expected , there was no significant difference between the responses of the two trials in this case , Fig 4C . Thus , it appears that having ARPP-21 in the signaling network could impose an inter-trial refractoriness for the Ca2+/calmodulin-dependent downstream response in an ITI dependent fashion . We define a metric called refractoriness ( REF ) to quantify the ITI dependent effect of the first trial on the response of the second trial . REF is defined as follows: REF ( t ) = 1− ( Activation Area in second trial ) ITI = t ( Activation Area in second trial ) ITI = 100 where ( Activated Area in second trial ) ITI = t is the area under the curve for substrate phosphorylation in the second trial which is separated by the first trial with an ITI = t seconds . ( Activated Area in second trial ) ITI = 100 is the activation area for second trial with an ITI = 100s . The trials with ITI = 100s are considered as well separated trials because with this ITI the second trial response is not affected by the first trial . Thus , if REF = 1 , then the response in the second trial is fully eliminated due to the inhibitory effect of the first trial and if REF = 0 , then there is no effect of the first trial on the response of the second trial . On the other hand , a negative value for REF indicates a potentiating effect of the first trial on the second trial . The simulations suggest that the REF decreases as a function of ITI , i . e . REF is higher for lower ITI values and it decreases as ITI increases Fig 4D . Thus , for some range of ITI the two trails do not act independent of each other but as the ITI increases each trial becomes more decoupled . Simulation of the ARPP-21 mutant , A21S55A , suggests that the relation between REF and ITI in this model appears due to the phosphorylation of ARPP-21 , Fig 4D . It could be possible that in a more complex scenario ( e . g . 10 trials in a row ) the downstream response may integrate over the trials thereby overcoming the refractoriness imposed by pARPP-21 . To test this possibility , we implemented a 10 trial scenario with ITI = 10s . Similar to the two trial scenario , each trial consists of 10 calcium pulses at 10 Hz followed by a dopamine input with Δt = 1s . The simulation for this 10 trial scenario suggests that there appears to be a small build of substrate response over the trials , Fig 4E . However , this integration over the trials is relatively small . A comparison between the wild-type and A21S55A scenario suggests that a strong ARPP-21 dependent refractoriness exists in this massed scenario , as well . Amplitude of the striatal dopamine represents the reward prediction error and it could vary depending on the difference between the expected and the acquired reward [62] . In a similar fashion , the cortical activity depends on the state of the organism [63] which in turn may affect the striatal calcium signal . If the temporal constraints of the input signal integration are of generic nature then they might be relatively robust to such input variability . Thus , we looked at the effects of input variability on the calcium-dopamine integration time window . In our model the strength of effective dopamine input is represented by its amplitude whereas the effective amplitude of the calcium input is controlled by its frequency ( refer “Materials and Methods” ) . Our simulations with varying amplitude of dopamine input indicate that there is no significant downstream response for low dopamine amplitudes , Fig 5A , as these dopamine levels are suggested to be not sufficient for significant PKA activation and DARPP-32 phosphorylation [28] . As the dopamine amplitude increases , the downstream response also increases . However , the relation between Δt and the downstream response is preserved for those cases which produced a response , Fig 5A . Similarly , the temporal constraints are preserved against the variation in the calcium input frequency as well , Fig 5B . For higher calcium frequencies , causing increased calcium levels , the input-interval constraint appears to be slightly more stringent as the range of Δt which could produce the downstream response is relatively narrower . However , in both the cases the input-order constraint of the integration process remains robust despite the changes in input strength . This relative robustness of the temporal constraints against variations in input signal properties suggests that these integration rules could be generic and not specific to certain input-signal parameters . The quantitative model built in this study is quite detailed and contains several parameters . Many of the parameters which might be important are not directly measured for the system of interest . They are rather optimized to fit experimental observations/phenotypes which emerge from the interactions of multiple parameters . However , it is important that the biological phenomena and predictions in the study should be robust with respect to changes in parameters around the selected point in the parameter space . Here , the parameters of interest are the kinetic rate constants of individual reactions and the starting amount of various conserved species . As mentioned above , the model output appears to be quite robust with respect to different parameter perturbation , Fig 2E . However , there could be certain parameters to which the output might be particularly sensitive . To identify these top sensitive parameters we perturbed all parameters one-at-a-time by ±20% of their original value . We then looked at the change in the model output produced by each of these perturbations . Fig 6A shows the distribution of substrate phosphorylation for all perturbations at different Δt values . Most of the perturbations produced a fractional response change in the range of [-0 . 01 , +0 . 01] ( between -1% and +1% ) for all Δt , Fig 6B . However , there were some perturbations which produced higher fractional response changes , Fig 6B . Since the signaling branches responsible for the model behavior in the positive and negative Δt regions are different , the sensitive parameters for these two Δt regions could also be different . Thus , we identified the sensitive parameters separately for the two Δt regions . The response change due to each parameter averaged separately for positive ( μ[response change ( Δt ≥ 0 ) ] ) and negative ( μ[response change ( Δt < 0 ) ] ) values of Δt have been used to identify the effect of individual parameters in the respective Δt regions . The top fifteen sensitive parameters affecting the positive and negative Δt regions are shown in Fig 6C and 6D , respectively , along with the fractional response change produced due their perturbation ( both +20% and -20% ) . The top sensitive-parameters list for the positive Δt region includes several DARPP-32 related parameters , Fig 6C . Specifically , the kinetic parameters governing the DARPP-32 phosphorylation at Thr-34 by PKA , dephosphorylation of DARPP-32 Thr-34 by PP2B and the rate constant for the DARPP-32 mediated PP1 inhibition . The emergence of these parameters as sensitive aligns with the aforementioned critical requirement of fast DARPP-32 signaling . This could also suggest that the onset and decay kinetics of DARPP-32 phosphorylation along with the kinetics of DARPP-32 mediated PP1 inhibition is carefully regulated by these neurons . However , total amount of DARPP-32 does not significantly affect the output for positive Δt . This is because MSNs are known to express a significantly high amount ( ~50μM ) of DARPP-32 . Therefore , the total amount of DARPP-32 is not a limiting factor . Apart from these DARPP-32 related parameters the sensitive parameters for the positive Δt also included kinetic parameters for the activation of PP2B by calmodulin ( CaM ) . Since , PP2B is responsible for the dephosphorylation of DARPP-32 at Thr-34 these PP2B activation parameters indirectly affect the state of DARPP-32 phosphorylation . This is true for the total amount of PKA and CDK5 also which are involved in the distribution of the phosphorylated states of DARPP-32 . In the case of negative Δt , the top sensitive parameters include the total amount of ARPP-21 and the kinetic parameter for ARPP-21 phosphorylation by PKA , Fig 6D . This aligns with the proposed ARPP-21 mediated effect on the negative Δt region according to our simulation results . The negative Δt also appears to be sensitive to DARPP-32 related parameters . A perturbation in total amount of DARPP-32 has an effect on the negative Δt region . This is because the estimated amount of DARPP-32 ( 50μM ) is significantly higher than the total amount of ARPP-21 ( 20μM ) in MSNs and both of these proteins are phosphorylated by PKA . The active PKA is a limited resource shared by DARPP-32 and ARPP-21 thus a change in the total amount DARPP-32 may affect the ARPP-21 mediated signaling . Additionally , the total amounts of CaMKII and PP2B also appear in the list of sensitive parameters in this case . This becomes clear by considering that both CaMKII and PP2B are also the downstream targets of calmodulin similar to ARPP-21 and changes in their total amounts could affect the level of binding between phospho-ARPP-21 and Ca2+/calmodulin . It is also interesting to note that the total amount of calmodulin appears to be a sensitive parameter for both positive and negative Δt region . This could be because calmodulin has several interactions with the DARPP-32 , ARPP-21 and CaMKII mediated signaling . The combined effects of calmodulin on the DARPP-32 and CaMKII significantly affect the gating potential of DARPP-32 signaling . Moreover , the core ARPP-21 mediated signaling relies on the calmodulin sequestration/competition thus adding further to the sensitivity of the model/predictions towards calmodulin .
The subcellular integration of the NMDA-dependent calcium and dopamine by D1R-expressing MSNs is believed to be an important biological process for reward learning . In this kinetic modeling study , we propose a novel mechanism which could explain the emergence of the observed input-interval and input-order constraints on the integration of calcium and dopamine . These temporal constraints could emerge as a result of the coordinated signaling through two striatally enriched phosphoproteins , namely DARPP-32 and ARPP-21 , Fig 7 . Both of these phosphoproteins are localized significantly in the dendritic compartments of MSNs and this co-localization could indicate a possible functional cooperation [64] . The major hypotheses , observations and testable-predictions made by this study are listed in Table 2 . In our simulations , dopamine acts as a gate for the calcium-triggered CaMKII/PP1 signaling and this aligns with known observations [8 , 9 , 17] . This gating is mediated by the phosphorylation of DARPP-32 in response to dopamine-triggered cAMP/PKA signaling . DARPP-32 mainly acts by inhibiting PP1 . If PP1 is not inhibited then it is sufficiently strong to counteract the effect of CaMKII . Thus , the coincidence of a calcium-triggered CaMKII activation and dopamine-dependent PP1 inhibition leads to a significant substrate phosphorylation . Such cAMP/PKA dependent effect on CaMKII signaling is not just specific to striatum . A similar phenomenon has been observed in the hippocampus as well , where it is mediated by Inhibitor-1 , a homologue of DARPP-32 [65 , 66] . The transient nature of both CaMKII activation and PP1 inhibition limits the possible time-interval ( Δt ) between the respective calcium and dopamine inputs , which could be synergistically integrated . Moreover , the PP1 inhibition kinetics should be fast enough to match the temporal characteristics of CaMKII activation . In other words , the PP1 inhibition should occur before the transient CaMKII activation fades away if these two were to interact . This requires particularly fast kinetics for the cAMP/PKA signaling axis which is already known to be the case in striatal neurons [40] . Recent observations in the dendritic compartments of these neurons indicate that DARPP-32 mediated effect is manifested within a few seconds , thus supporting the requirement for fast DARPP-32-dependent PP1 inhibition [9] . Even though the transient DARPP-32-dependent effect could implement the observed input-interval constraint , according to our results , DARPP-32 signaling alone may not distinguish between the order of calcium and dopamine incidence . Moreover , the molecular identity of the substance which could make the striatal CaMKII/PP1 signaling sensitive to the temporal order of calcium and dopamine is not clear . However , as presented above , ARPP-21 has the potential to implement an input-order dependent filtering mechanism by acting as a competitor to CaMKII . This could enable the striatal signaling machinery to distinguish between the incidence order of calcium and dopamine . When ARPP-21 is phosphorylated by a preceding dopamine it turns into a potent sequestering agent for subsequent calcium-triggered Ca2+/calmodulin , thus reducing the available Ca2+/calmodulin for CaMKII activation . Similar to the DARPP-32 mediated phenomenon , this ARPP-21 mediated effect also demands fast phosphorylation kinetics . As aforementioned , the ARPP-21 mediated mechanism rests on the idea that ARPP-21 , upon phosphorylation , competes with CaMKII for the available Ca2+/calmodulin . This competition stipulates that the amount of calmodulin be lower than the collective amount of its possible binding partners , i . e . Ca2+/calmodulin is a limiting resource . If Ca2+/calmodulin is not a limited resource then pARPP-21 and CaMKII do not have to compete with each other , thereby reducing the inhibitory effect of ARPP-21 on CaMKII activation . This in turn will eliminate the input-order sensitivity of the calcium-dopamine integration . Even though the absolute amount of calmodulin could be high in many cell types , various studies suggest that free calmodulin is indeed a limiting resource [67–69] . The same has been inferred for neurons as well [70] . Moreover , such a limit on available Ca2+/calmodulin seems to exist also in MSNs [15] thus supporting this necessary requirement for the proposed mechanism . In addition to CaMKII , Ca2+/calmodulin is also responsible for the activation of PP2B which dephosphorylates DARPP-32 Thr34 . However , unlike CaMKII , pARPP-21 does not have any significant effect on the level of PP2B activation , for transient input signals , in our signaling model because the affinity of Ca2+/calmodulin towards PP2B is assumed to be very high ( Kd in subnanomolar range; [71] ) compared to the affinity of Ca2+/calmodulin towards pARPP-21 ( Kd in tens of nanomolar ) . Thus , the current study does not predict any significant ARPP-21 mediated effect on DARPP-32 dephosphorylation within the assumed parameter space for transient input signals . However , this does not rule out such a regulation . Rather this could be due to the insufficiency of data to tune this aspect of the current signaling model . Even though the role of striatal ARPP-21 is not clear , it seems to be involved in addiction and motivational behavior [54 , 72] . The current study predicts a clear physiological role for ARPP-21 in the context of reward learning . We also predict that the downregulation of ARPP-21 could significantly hinder the ability of MSNs to identify the temporal order of calcium and dopamine signals . On a behavioral level , this implies an increased appearance of association between a stimulus and a reinforcement which preceded the stimulus . This kind of backward association tends to eliminate the predictability of a conditioned stimulus . Thus , we also predict that an ARPP-21 downregulated phenotype could have an aberrant causality perception , more specifically deficits in associating a reward with the optimal cue in a complex environment . Apart from imposing the input-order constraint on the calcium-dopamine integration , our simulations also suggest that ARPP-21 could impose an inter-trial refractoriness as a function of the ITI in a multi-trial conditioning scenario . This could imply that an MSN , or its dendritic segment , which is potentiated in the recent past would resist a further potentiation within a refractory period . This could mean that multiple trials with shorter ITIs may not be completely independent to each other . As the ITI increases the trials start to become decoupled . Such , an inverse relation between conditioned response and ITI is well known for classical conditioning [73 , 74] . Although there may be several factors responsible for such a relation , ARPP-21 could be one of the molecular players involved in this . This study highlights the possible role of phosphoproteins in controlling the temporal aspects of striatal signal transduction . Furthermore , we also attempt to illustrate the idea that different aspects of a single phenomenon ( temporal constraint in this case ) could be distributed among parallel signaling modules ( DARPP-32 and ARPP-21 ) and do not necessarily have to be implemented by a single signaling node . One such single-signaling node which is believed to be responsible for similar temporal constraints in other systems , like Aplysia serotonin response , is Ca2+/calmodulin stimulated Adenylyl cyclase , which exhibits calcium-dependent priming on the G-protein activation [75 , 76] . A similar Ca2+/calmodulin stimulated Adenylyl cyclase , Adenylyl cyclase type I ( AC1 ) , has also been suggested to be involved in the emergence of striatal input-interval and input-order constraints [9] . AC1 is known to be synergistically activated by Gs ( Golf ) G-protein and Ca2+/calmodulin and is known to act as a coincidence detector for these two signals [77] . Even though the activation pattern of AC1 is not clear for transient inputs , there could exist kinetic parameter sets for which AC1 may enforce the temporal constraints ( see Fig A a , b in S1 Text ) . However , the expression of AC1 is low in the striatum of adult organisms even though it may exist in the early stages of development [78 , 79] . During postnatal development most of the AC1 is replaced by AC5 , and AC5 is the predominant functional enzyme in adult organisms [31 , 79 , 80] . AC5 , which we use in our model , is strongly activated by the Golf and does not display any calcium-dependent synergy [81] . It could be a possible scenario that lower amounts of AC1 may still be expressed in adult MSNs making the overall AC population a mixture of AC5 and AC1 . In such a mixed AC population , which contains both AC5 and AC1 , the overall behavior may be dominated by the AC with the higher fraction ( see Fig A c in S1 Text ) . For example , the AC1 dependent temporal constraints may emerge , irrespective of ARPP-21 , if the fraction of AC1 is high in the total AC ( see Fig A c in S1 Text ) . However , for a scenario consistent with striatal data , i . e . high AC5 and low AC1 the contribution of AC1 is overshadowed by the bulk response of AC5 and the overall response appears to be similar to the system containing only AC5 ( see Fig A c in S1 Text ) . Thus , in striatum , the bulk AC5 dependent cAMP response may significantly dampen any AC1 dependent component of temporal constraints due to the high amount of AC5 . On the other hand , the DARPP-32/ARPP-12 dependent mechanism proposed here is quite generalizable to various systems consisting of different AC isoforms . Even though we focused on the potential role of DARPP-32 and ARPP-21 , they might just be part of a much wider signaling mechanism to produce the temporal constraints on striatal calcium-dopamine integration . For example , the activation of other signaling processes , like PKA , have also been observed to display the input-interval and input-order constraints [9] . Thus , further exploration of the wider signaling network involved in this process is of interest . For example , it is known that several other crosstalk points exist between calcium and dopamine signaling axis at various downstream levels of the postsynaptic signaling [50 , 82–84] . Some of them might play a role in further fine-tuning of the temporal constraints . It is interesting to note that there are brain regions other than the striatum , e . g . the amygdaloid complex and the bed nucleus of the stria terminalis ( BNST ) , where the co-expression of DARPP-32 and ARPP-21 is known to exist [64 , 85–87] . Both amygdala and BNST are critical for fear conditioning and they play a role in the association between cue/context to a stress-inducing unconditioned response [88] . Thus , the combined DARPP-32/ARPP-21 signaling may play similar roles to the ones discussed in this study for these brain regions . Thus , the phosphoprotein-dependent mechanism proposed in this study could represent a more general brain-wide signaling motif responsible for the implementation of temporal constraints on subcellular signal integration of environmental cues and reinforcement . | A response towards an environmental stimulus could be reinforced if it elicits a reward . On the subcellular level , the environmental stimulus and the reward signal lead to a transient increase in striatal calcium- and dopamine-signaling , respectively . The integration of calcium and dopamine signals , which is important for reward-learning , could elicit a downstream response only if they are close in time and arrive in correct order ( first calcium and then dopamine ) . This study proposes that the requirement for the input signals to be temporally close and in correct order could emerge due to the coordinated signaling via two striatal phosphoproteins , DARPP-32 and ARPP-21 . The DARPP-32 signaling implements an input-interval dependent gating function and ARPP-21 implements an input-order dependent threshold-like function . Thus , a molecular mechanism has been presented here which could explain the emergence of important temporal aspects of subcellular signal integration in reward-learning . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"phosphorylation",
"medicine",
"and",
"health",
"sciences",
"neurochemistry",
"chemical",
"compounds",
"signaling",
"networks",
"brain",
"neuroscience",
"organic",
"compounds",
"adenylyl",
"cyclase",
"signaling",
"cascade",
"hormones",
"network",
"analysis",
"amines",
"neurotransmitters",
"catecholamines",
"calcium",
"signaling",
"dopamine",
"computer",
"and",
"information",
"sciences",
"animal",
"cells",
"proteins",
"chemistry",
"neostriatum",
"biochemistry",
"signal",
"transduction",
"cellular",
"neuroscience",
"organic",
"chemistry",
"cell",
"biology",
"post-translational",
"modification",
"anatomy",
"neurons",
"biogenic",
"amines",
"biology",
"and",
"life",
"sciences",
"phosphoproteins",
"physical",
"sciences",
"cellular",
"types",
"cell",
"signaling",
"signaling",
"cascades"
] | 2016 | Role of DARPP-32 and ARPP-21 in the Emergence of Temporal Constraints on Striatal Calcium and Dopamine Integration |
Indonesia reports the second highest dengue disease burden in the world; these data are from passive surveillance reports and are likely to be significant underestimates . Age-stratified seroprevalence data are relatively unbiased indicators of past exposure and allow understanding of transmission dynamics . To better understand dengue infection history and associated risk factors in Indonesia , a representative population-based cross-sectional dengue seroprevalence study was conducted in 1–18-year-old urban children . From October to November 2014 , 3 , 210 children were enrolled from 30 geographically dispersed clusters . Serum samples were tested for anti-dengue IgG antibodies by indirect ELISA . A questionnaire investigated associations between dengue serologic status and household socio-demographic and behavioural factors . Overall , 3 , 194 samples were tested , giving an adjusted national seroprevalence in this urban population of 69 . 4% [95% CI: 64 . 4–74 . 3] ( 33 . 8% [95% CI: 26 . 4–41 . 2] in the 1–4-year-olds , 65 . 4% [95% CI: 69 . 1–71 . 7] in the 5–9-year-olds , 83 . 1% [95% CI: 77 . 1–89 . 0] in the 10–14-year-olds , and 89 . 0% [95% CI: 83 . 9–94 . 1] in the 15–18-year–olds ) . The median age of seroconversion estimated through a linear model was 4 . 8 years . Using a catalytic model and considering a constant force of infection we estimated 13 . 1% of children experience a primary infection per year . Through a hierarchical logistic multivariate model , the subject’s age group ( 1–4 vs 5–9 OR = 4 . 25; 1–4 vs . 10–14 OR = 12 . 60; and 1–4 vs 15–18 OR = 21 . 87; p<0 . 0001 ) and the number of cases diagnosed in the household since the subject was born ( p = 0 . 0004 ) remained associated with dengue serological status . This is the first dengue seroprevalence study in Indonesia that is targeting a representative sample of the urban paediatric population . This study revealed that more than 80% of children aged 10 years or over have experienced dengue infection at least once . Prospective incidence studies would likely reveal dengue burdens far in excess of reported incidence rates .
Dengue is an arbovirus transmitted to humans via the bites of infected Aedes mosquitoes . It is the most rapidly spreading mosquito-borne viral disease with a global incidence that has increased 30-fold over the last 50 years [1] . While reliable burden estimates remain elusive , two studies have estimated the global symptomatic disease burden to be 96 million and 58 . 4 million cases/year , with 70–80% of cases occurring in the Asia-Pacific region [2 , 3] . Traditionally an urban disease , dengue disease is increasingly reported in rural areas and its geographic range has expanded to more than 125 tropical countries [1] . There is no specific antiviral treatment; clinical management is focused on careful fluid management and detection of early warning signs of severe disease . Historically , prevention measures have focused on vector control , education and behavioural changes to reduce interactions between humans and vector mosquitoes [4 , 5] . Improved clinical management and public awareness have contributed to declining case fatality rates to below 1% in most countries [1] . While this represents important progress , overall dengue incidence continues to rise and fatalities remain unacceptably high , suggesting that traditional control approaches are not sufficient . Vector control measures are important yet operationally challenging , of variable effectiveness and costly to sustain [6] . Routine vaccination is becoming a reality: several dengue vaccines are at different stages of clinical development [7] and a chimeric tetravalent vaccine from Sanofi Pasteur is being licensed in an increasing number of countries in Latin America and Asia [7 , 8] . In this new era of dengue as a vaccine-preventable disease , an accurate understanding of disease burden and transmission patterns will be essential to inform vaccine policy decisions . Dengue is hyper-endemic with frequent epidemic cycles in Indonesia . The disease is most common in urban areas and in recent years has reportedly spread to smaller , more rural villages . Reported incidence remains highest in children 1–15 years of age , but since the 1980s incidence in persons over 15 years of age has gradually increased [9 , 10] . Reporting of dengue haemorrhagic fever ( DHF ) is mandatory in Indonesia and the country typically reports the highest number of cases in the WHO Southeast Asia Region [1] . Between 2001 and 2011 , there was an average of 94 , 564 reported cases and between 472 and 1 , 446 reported deaths per year [1 , 11] . Dengue disease reporting is acknowledged by Indonesian experts to be incomplete and to vary widely between provinces , with reported incidence rates ranging from 2 . 2 to 168 . 5 cases per 100 , 000 inhabitants in 2013 [12] . An improved understanding of dengue epidemiology , burden and its dynamic characteristics are important for public health planning . Seroprevalence studies in healthy volunteers provide information on infection history in the population , from which inferences about disease burden may be drawn . Since age reflects duration of exposure , age-stratified data provide insights into transmission dynamics [13–17] . There is a lack of dengue seroepidemiological data from Indonesia and no previous study has used a population representative sample of urban Indonesian children [18–20] . This is a particularly important gap as it will provide information on whether the variations in reported incidence from different Indonesian provinces are reflective of underlying transmission dynamics or to the result of the reporting or surveillance practices employed . We conducted a seroprevalence study in urban-dwelling Indonesian children to improve understanding of dengue epidemiology and infection risk factors and inform future dengue vaccine policy decisions .
The protocol was reviewed and ethical approval was obtained from the Health Research Ethics Committee of Faculty of Medicine of University of Indonesia . Indonesia is the largest country in Southeast Asia , with an area of 1 . 91 million km2 . The country has a population of 252 . 2 million living on five main islands and four archipelagos ( >17 , 000 islands ) administratively divided into 34 provinces [21] . In 2014/2015 , approximately 60% of Indonesians were living on the island of Java and 53 . 3% lived in urban areas [21 , 22] . Indonesia is divided into five administrative levels: provinces ( n = 34 ) , regencies ( n = 416 ) , cities ( n = 98 ) , subdistricts ( n = 7 , 024 ) , and villages ( n = 81 , 626 ) . Villages are considered either as rural ( desa ) or urban ( kelurahan ) based on population density , percentage of agricultural household and number of urban facilities such as schools and hospitals [21 , 23] . A population-based cross-sectional study design was adapted from the World Health Organization ( WHO ) Expanded Program on Immunization ( EPI ) cluster survey method . This approach considers 30 clusters as an adequate number for their means to be normally distributed , thus permitting statistical theory based on the normal distribution to be used to analyse the data [24 , 25] . Based on the probability proportional to population size , 30 urban subdistricts were selected using demographic data from 2009 or 2010 , provided by the Sub-Directorate of Statistical Services and Promotion , Statistics Indonesia . The geographical coordinates of Indonesian administrative units were retrieved from the Global Rural-Urban Mapping Project , maintained by the Socioeconomic Data and Applications Center [26] . Provinces were listed based on their mean geographical coordinates from West to East ( Fig 1 ) and the cumulative urban population of their subdistricts was calculated using 2010 population data . To ensure the population of clusters was sufficient to enrol the desired sample , a minimum population of 1 , 000 persons per subdistrict was defined and any smaller subdistricts were removed from the list . The first cluster was selected by generating a random number between 1 and 1/30th of the total urban population , using Epi Info Version 7 , and selecting the first subdistrict for which the cumulative population was superior or equal to this random number . Subsequent clusters were selected by adding 1/30th of the urban population to the random number and selecting the first corresponding subdistrict for which cumulative population was higher or equal so that: Clusteri cumulative population ≥ random number + i*1/30 of urban population The 30 subdistricts selected by this method are listed in Appendix 1 . Each subdistrict in Indonesia contains one main health centre ( puskesmas kecamatan ) whose catchment area was the site of the study . Households in the five neighbourhood associations located closest to the health centre ( each comprising 30–50 households , giving a total of 150–250 households ) were eligible to participate in the study . Household visits were conducted , inviting one child from each household to participate , until the sample size was reached . A table indicating the required number of children from each of four age groups was provided to the health centre study teams . If a household had only one eligible child , the child was invited . When a household had several eligible children , a child in the age group with the fewest children already participating was selected . Towards the end of the survey , survey teams were allocated a specific number of subjects in each age group to recruit to avoid over-sampling . If the parents refused the participation of the selected child , the household was not included . This process was continued until the desired sample size was achieved in each of the 30 clusters . The sample size was calculated using EpiInfo Version 7 to estimate seroprevalence in each of four age groups ( 1–4 , 5–9 , 10–14 and 15–18 years old ) with 95% confidence , a margin error of 5% and accounting for clustering with a design effect of 2 . The expected national seroprevalence , based on Indonesian expert opinion and published regional data [14 , 19 , 27 , 28] , was 25% in the 1–4-year-old group , 45% in the 5–9-year-old group , 55% in the 10–14-year-old group and 65% in the 15–18-year-old group . To account for incomplete data , a 10% contingency was applied . The total sample size was 3 , 210 children , 660 from the 1–4-year-old group ( 22 per cluster ) , 870 from the 5–9-year-old group ( 29 per cluster ) , 870 from the 10–14-year-old group ( 29 per cluster ) and 810 from the 15–18-year-old group ( 27 per cluster ) . In total , 107 children were enrolled in each cluster . The study was presented to families during monthly neighbourhood association meetings . After household visits , eligible subjects were invited to the healthcare centre for enrolment and blood sampling if they were healthy , 1–18 years of age on inclusion day , and had lived in the location for at least 1 year . An informed consent form was signed by a parent or legal guardian , and by the subject if aged 13–18 years . Subjects aged 8–12 years provided signed assent . A questionnaire was administered to collect information on demographics , knowledge of dengue symptoms and transmission , vector control practice , and medical history in the household . For each subject , 2mL of venous blood was drawn into plain vacutainer tubes . After centrifugation , serum aliquots were frozen at -20°C before refrigerated transport by courier to a central laboratory for analysis . Each specimen was tested for dengue IgG antibodies by ELISA using the commercial Panbio Dengue IgG Indirect ELISA kit ( sensitivity = 96 . 3%; specificity = 91 . 4–100% according to manufacturer’s instructions; Panbio , Alere , Australia ) [29] . Samples were considered positive for previous dengue infection according to the standard protocols of the manufacturer ( Panbio units <9 is negative; 9–11 is equivocal; and >11 is positive ) . All analyses were run using SAS 9 . 4 .
From a total of 6 , 299 Indonesian subdistricts , 2 , 823 with urban population were identified , 2 , 756 of which had an urban population >1 , 000 and were thus used for sampling . A map of the 30 selected clusters is presented in Fig 1 . From 30 October 2014 to 27 November 2014 , a total of 3 , 210 subjects were enrolled in the study; 39 subjects ( 1 . 2% ) were excluded due to at least one criteria of eligibility not being fulfilled and four subjects ( 0 . 1% ) due to missing or incomplete data ( demographic or serologic status result ) . A total of 3 , 194 subjects ( 98 . 7% ) were included in the analyses ( Fig 2 ) ; there were 107 subjects per site with the exception of four sites with 106 subjects , three sites with 105 subjects and one site with 101 subjects . There were 672 subjects in the 1–4-year-old age group , 861 subjects in the 5–9-year-old age group , 886 in the 10–14-year-old age group and 775 in the 15–18-year-old age group . Among them , 47 . 8% were male and the mean age was 9 . 7 years . The age-specific seroprevalence ranged from 26 . 4% ( 95% CI: 15 . 8–37 . 1 ) in those aged 1-year-old to 95 . 3% ( 95% CI: 89 . 8–100 ) in the 18-year-old subjects ( Fig 3 ) . The median age at seroconversion was 4 . 8 years . The overall nationwide seroprevalence was 69 . 4% , with a minimum of 34 . 6% and a maximum of 87 . 9% observed per site , and the seroprevalence per age group was 33 . 8% in the 1–4year-old group , 65 . 4% in the 5–9-year-old group , 83 . 1% in the 10–14-year-old group and 89 . 0% in the 15–18-year-old group ( Table 1 ) . In the final data set , the level of non-response ( “no data” ) varied from 0 . 4 to 14 . 0% ( Table 1 ) . Subjects were familiar with dengue disease , with 92% having heard about dengue and 91 . 4% able to cite at least one symptom . Control practices reported included use of repellent cream or mosquito spray ( 43 . 8% ) , elimination of mosquito breeding sites by covering water containers ( 59 . 0% ) and eliminating stagnant water around the home ( 85 . 1% ) . Most subjects ( 75 . 3% ) reported they had never been diagnosed with dengue . Age and gender were associated with dengue serological status , with seroprevalences increasing with age ( p<0 . 0001 ) and values of 71 . 1% ( 95% CI: 65 . 9–76 . 3 ) in females versus 67 . 4% ( 95% CI: 62 . 4–72 . 5 ) in males ( p = 0 . 018 ) ( Table 1 ) . After univariate analysis , the type of household ( p = 0 . 08 ) , the level of education of the parents/guardians ( p<0 . 0001 ) , the number of persons living in the household ( p<0 . 0001 ) , knowledge about dengue symptoms ( p = 0 . 14 ) , sleeping under an untreated bed net ( p = 0 . 10 ) , the number of dengue cases identified since the subject was born ( p<0 . 0001 ) , and a previous clinical diagnosis of dengue for the subject ( p<0 . 0001 ) were also associated with dengue serological status . In the multivariate model ( Table 2 ) , two variables remained associated with the dengue serologic status , the subject age group ( 1–4 vs 5–9 OR = 4 . 25; 1–4 vs . 10–14 OR = 12 . 60; and 1–4 vs 15–18 OR = 21 . 87; p<0 . 0001 ) and the number of cases diagnosed in the household since the subject was born ( p = 0 . 0004 ) . The constant force of infection model was valid and estimated a force of primary infection of 13 . 1% per year in dengue-naïve children . As a result of the goodness of fit statistic being close to 0 . 05 , a model of varying force of infection ( age groups of one year ) was run to examine the homogeneity of the force of primary infection estimates per age group . As suggested by the first model , there was no clear trend in changes in force of infection with age; the estimates were overlapping , ranging from 10 . 2% to 18 . 5% per year . The highest force of primary infection was observed in the 1-year-old age group ( Table 3 ) .
This is the first dengue antibody seroprevalence study conducted in a representative population of urban dwelling Indonesian children . The findings benefit from a cluster sampling design with probability proportional to size method , and sensitive and specific dengue diagnostic assays performed in the same laboratory . This study found that 69 . 4% of children had been previously infected with dengue virus , more than 80% of children aged 10 years or over , indicating that the disease burden is extremely high . A seroprevalence study conducted in 1995 in healthy children in Yogyakarta , Indonesia , using the plaque reduction neutralization test to determine previous exposure , reported the presence of neutralizing antibodies in 56 . 2% of 4–9-year-old children , ranging from 37 . 2% in 4-year-old subjects to 69 . 7% in those 9 years of age . These are slightly lower than the rates observed in our study ( Fig 3 and Table 1 ) and may be reflective of increasing dengue endemicity in the intervening decades , or geographic variability [19] . Our results also show higher levels of dengue virus exposure than those reported in other dengue endemic countries such as Sri Lanka ( Colombo , 2008 , 52 . 0% in those <12 years of age , and median age of seroconversion of 4 . 7 years ) [13 , 35] , and Vietnam ( Binh Thuan , 2003 , 65 . 7% in 7–13 year olds ) [14] . This elevated dengue exposure risk was also observed during a 2011 dengue vaccine trial in 5 Asian countries , where baseline dengue seroprevalence was highest in Indonesian children [36] . Our constant force of infection model estimated a 13 . 1% annual rate of primary infection among 1–18-year-old children , while the variable model estimated a force of infection that varied from 10 . 2% to 18 . 5% . These estimates are similar to those reported in Sri Lanka in 2008 ( 14 . 1% in those aged <12 years ) and Southern Vietnam in 2003 ( 11 . 7% in 7–13-year-old children ) [13 , 14] . Despite these similarities between Vietnam and Indonesia in terms of transmission dynamics , the reported incidence of disease in Vietnam is more than twice that in Indonesia . [37] . A number of hypotheses could explain this difference in findings: most likely , it is reflective of Indonesia’s specific case definition for reported dengue disease ( only DHF is reported ) , but underlying virological , genetic or epidemiological differences could play a role . From the constant force of primary infection model , it can be assumed that the average rate of primary infection was not highly variable over the past 18 years . Additional analysis may be needed to better understand infection risk over time . The recently observed increase in age distribution of reported cases may have been driven by more variable virologic , demographic , reporting or other determinants of disease [10] . A similar phenomenon was illustrated by a study conducted in Thailand showing that the upward shift in dengue case age was associated with demographic changes [38] . It can be assumed that dengue awareness , through social mobilization and education campaigns , begun in the 1970s , and the increasing public health importance associated with high media coverage , has steadily increased [39] . Knowledge of dengue transmission and symptoms was high within the study subjects; 92% of households had heard about dengue before our study and were able to cite at least one of the disease symptoms , and more than 80% knew that dengue virus is transmitted by diurnal mosquito bites . In term of exposure , household practices were focused on destroying mosquito breeding sites rather than personal protection . The level of exposure to the virus , however , is strong evidence that these reported behaviours are inadequate to protect against infection and additional prevention and control measures are urgently required . In the multivariate model , only subject age group and the number of dengue cases that occurred in the household were associated with seropositive status . Some of the parameters significantly associated with dengue seropositivity in univariate models were also implicated in other dengue studies conducted in Latin America and Asia . For example , parental level of education and dengue illness history in the household have been associated with dengue seropositivity [17] . Other parameters , such as household size , exhibit an association inverse to that previously reported in the literature [40] . This is most likely explained by confounding effects from known risk factors such as age or unknown , socio-demographic drivers of exposure risk . The lack of significant associations between socio-demographic and behavioural factors with serological status provides evidence that essentially everyone is at risk of infection; that knowledge of prevention and control at the individual/household level is not protective against infection; and that additional measures to prevent transmission are required . The retrospective nature of our questionnaire limits the robustness of our results; recall bias may have been an issue . A recent expansion in dengue virus transmission from urban to peri-urban and rural areas has been described [15] and the identification of provinces or areas of high transmission risk is a focus of prevention and control planning . This study showed a high level of exposure across urban Indonesia and , while we excluded rural areas from this study for operational reasons , it is likely that nearby peri-urban populations may have experienced similar high levels of exposure [40] . Another possible limitation is that cross-reaction between flaviviruses has been documented and the risk of false positives cannot be excluded . We consider this risk as low , because reports of other viruses such as Japanese encephalitis and Zika , in Indonesia , are rare . This study was not designed to make national-level infection or disease burden estimates but the observation that 13 . 1% of children suffer a primary infection per year translates into many millions of infections per year . Adults are presumably infected with a similar frequency . A proportion of these infections will be secondary , predisposing to symptomatic and severe disease . While a modelling approach would be required to quantify this burden , these data are strongly suggestive that dengue infections result in a significant burden of symptomatic and severe disease in urban Indonesia . | Indonesia reported to the WHO the world’s second highest average number of dengue cases and the highest in Asia from 2004 to 2010 . These passive surveillance reports vary widely within the country and are likely to be a severe under-estimation of the full disease burden as frequently only dengue haemorrhagic fever is captured . Understanding the intensity of dengue virus transmission and associated risk factors nationwide is necessary to guide and prioritize appropriate prevention and control measures against dengue disease , especially considering the availability of the first dengue vaccine and recent recommendations for its use in areas of high endemicity , as measured by seroprevalence and other indicators . Age-stratified seroprevalence data provide robust estimates of past exposure and can inform on transmission intensity . Therefore , we conducted a seroprevalence study of anti-dengue IgG antibodies in a representative sample of urban-dwelling Indonesian children . We found an overall dengue seroprevalence of 69 . 4% with half of the children having been infected at least once by the age of 5 years . Age of the subject and the number of dengue cases diagnosed in the household were associated with serostatus . These results confirm the high dengue disease burden in Indonesia and the urgency of implementation of effective prevention and control measures . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"dengue",
"virus",
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"demography",
"pathogens",
"geographical",
"locations",
"tropical",
"diseases",
"microbiology",
"indonesia",
"viruses",
"age",
"groups",
"rna",
"viruses",
"neglected",
"tropical",
"diseases",
"infectious",
"disease",
"control",
"public",
"and",
"occupational",
"health",
"infectious",
"diseases",
"serology",
"medical",
"microbiology",
"dengue",
"fever",
"microbial",
"pathogens",
"people",
"and",
"places",
"asia",
"flaviviruses",
"oceania",
"viral",
"pathogens",
"population",
"groupings",
"biology",
"and",
"life",
"sciences",
"viral",
"diseases",
"organisms"
] | 2017 | Dengue seroprevalence and force of primary infection in a representative population of urban dwelling Indonesian children |
Mutations in ribosomal protein ( RP ) genes can result in the loss of erythrocyte progenitor cells and cause severe anemia . This is seen in patients with Diamond-Blackfan anemia ( DBA ) , a pure red cell aplasia and bone marrow failure syndrome that is almost exclusively linked to RP gene haploinsufficiency . While the mechanisms underlying the cytopenia phenotype of patients with these mutations are not completely understood , it is believed that stabilization of the p53 tumor suppressor protein may induce apoptosis in the progenitor cells . In stark contrast , tumor cells from zebrafish with RP gene haploinsufficiency are unable to stabilize p53 even when exposed to acute DNA damage despite transcribing wild type p53 normally . In this work we demonstrate that p53 has a limited role in eliciting the anemia phenotype of zebrafish models of DBA . In fact , we find that RP-deficient embryos exhibit the same normal p53 transcription , absence of p53 protein , and impaired p53 response to DNA damage as RP haploinsufficient tumor cells . Recently we reported that RP mutations suppress activity of the AKT pathway , and we show here that this suppression results in proteasomal degradation of p53 . By re-activating the AKT pathway or by inhibiting GSK-3 , a downstream modifier that normally represses AKT signaling , we are able to restore the stabilization of p53 . Our work indicates that the anemia phenotype of zebrafish models of DBA is dependent on factors other than p53 , and may hold clinical significance for both DBA and the increasing number of cancers revealing spontaneous mutations in RP genes .
The stabilization of the p53 tumor suppressor is a pivotal event in the programmed cell death response . Levels of p53 protein are normally kept very low through its physical association with the MDM2 protein , an E3 ubiquitin ligase that constitutively ubiquitinates p53 and targets it for proteasomal degradation [1] . Many kinds of cellular stress , including DNA damage and oncogene presence , activate different signaling pathways that result in the dissociation of p53 and MDM2 . p53 then stabilizes and translocates to the nucleus where it targets genes that arrest the cell cycle and turn on DNA repair , or genes that induce apoptotic cell death if the damage is deemed irreparable [2] . The stabilization of p53 has been reported to trigger human bone marrow failures such as dyskeratosis congenita and Fanconi anemia [3 , 4] . While Fanconi anemia is predominantly linked to mutations in DNA repair enzymes , several genes found mutated in dyskeratosis congenita patients have a known role in the rRNA maturation steps of early ribosome biogenesis . The mutation of these latter genes in zebrafish stabilizes p53 , as does the mutation of several other genes important for the processing of rRNA [5–7] . In human bone marrow failures syndromes linked to RP haploinsufficiency such as Diamond-Blackfan anemia ( DBA ) and 5q-myelodysplastic syndrome , the loss of hematopoietic progenitor CD34+ cells by p53-induced apoptosis is believed by some to be the major cause of cytopenia [8] . However , the contribution of p53-induced apoptosis specifically to the cytopenia phenotype remains controversial . Recent studies demonstrated that patient CD34+ hematopoietic progenitor cells carrying mutations in the most commonly mutated gene linked to DBA ( RPS19 ) do not reveal any hallmarks of apoptosis as they are induced to differentiate into erythrocytes [9] . This work also showed that while the co-depletion of p53 with RPL11 in CD34+ cells reduced some apoptotic effects , it did not restore the proliferation capacity lost upon RPL11 depletion alone . Therefore the contribution of p53 stabilization to the loss of erythrocytes in DBA patients is possibly less significant than previously thought . In addition to being important for erythrocyte production , there also exist several reports indicating a role for RPs as tumor suppressors . Human patients with 5q-MDS or DBA are more predisposed to developing malignancies , both acute myeloid leukemia ( AML ) and solid tumors , respectively [10 , 11] . Importantly , the advent of exome sequencing has unveiled a surprising number of somatic RP gene mutations in an array of human cancers . These recent exome sequencing reports have identified mutations in RPL5 and RPL10 in T-cell acute lymphoblastic leukemia , mutations in RPL5 in gliomas , and mutations in RPL22 in human endometrioid endometrial cancer and colorectal cancer [12–14] . Embryonic zebrafish mutants and morphants are popular models of DBA and RP loss . In mutant lines where RP genes are disrupted by murine virus integrations , the homozygous embryos reveal a progressive reduction of the RP over the first 3 days post fertilization ( dpf ) coupled to a failure of hemoglobin-expressing cells to develop [15 , 16] . In contrast , haploinsufficient RP embryos reveal no cytopenia or any other conspicuous phenotype except for a mild growth defect that does not affect their development into adulthood [17] . Once reaching adulthood however , many of the haploinsufficient mutant lines reveal the formation of malignant peripheral nerve sheath tumors ( MPNSTs ) , a tumor type rarely observed in laboratory strains of zebrafish [18] . Interestingly , this exact tumor type arises with the same frequency in zebrafish carrying homozygous mutations of p53 in a highly conserved residue within one of the DNA-binding domains ( M214K ) [19] . Further study into this revealed that while wild type p53 mRNA was normally transcribed in the RP-mutant tumor cells , p53 protein was unable to be detected [20] . This observation was despite the application of ionizing radiation and/or proteasome inhibition with MG132 , two conditions that were found to stabilize p53 in zebrafish tumor cells carrying wild type RP genes . Actinomycin D is a drug that disrupts ribosome biogenesis by inhibiting rRNA polymerase and results in the stabilization of p53 [21] . It was recently reported that this induction of p53 requires the activity of the survival factor AKT/PKB [22] , a kinase that responds to the stimulation of many growth factor receptors , including insulin receptors [23] . While activated AKT has many functions , one major event of insulin stimulation directly downstream of AKT is the phosphorylation and inhibition of glycogen synthase kinase-3 ( GSK-3 ) [24] . Under normal conditions GSK-3 phosphorylates and activates MDM2 in a way that promotes p53 degradation [25] . However upon DNA damage by ionizing radiation , the phosphorylation of AKT results in the inactivation of GSK-3 [26] . This AKT-mediated inactivation of GSK-3 in response to ionizing radiation begins with the DNA-dependent protein kinase ( DNA-PK ) , a protein that recognizes the double-stranded breaks on DNA and signals through a cascade that ultimately results in p53 stabilization and programmed cell death [27] . Thus one mechanism of the p53 stabilization in response to ionizing radiation is through the activation of DNA-PK and AKT inhibiting the downstream activity of GSK-3 and MDM2 . We previously reported that the loss of RP genes in mammalian cells and in zebrafish embryos results in a loss of AKT activity that could be overcome by the addition of insulin [16] . This observation led us to consider the possibility that the AKT pathway was involved in the regulation of p53 in RP mutant cells , and that the impairment of the DNA damage pathway through AKT may have a role in the predisposition to malignancy caused by RP gene mutations .
The RP mutant zebrafish lines we used for this study were generated by viral insertions in the introns of RP genes , two of which ( rpS7 and rpL11 ) have homologues found mutated in DBA patients [28 , 29] . We find that at 2 dpf these embryos display normal expression of the scl transcription factor required for the genesis of hematopoietic stem cells in both primitive and definitive hematopoiesis , a result that was also observed in zebrafish embryos with deletions in the rpS19 gene ( S1 Fig ) [30–33] . However , RP mutants reveal a marked decrease in the expression of the βE1-globin gene , which ( at this stage of development ) normally becomes up regulated as cells commit to the erythrocyte lineage ( S1 Fig ) [34] . Zebrafish embryos carry maternal stores of RPs such that these mutants reveal a progressive loss of RP expression . The rpS7 mutants express about 50% of wild type rpS7 levels at 1 dpf while this percent reduction is not observed in rpL11 mutants until 3 dpf [16 , 35] . This likely explains why despite the fact that embryos from both mutant lines reveal hematopoietic and developmental phenotypes , these are more severe in the rpS7 mutants [16 , 35] . Because these phenotypes at 1 dpf are much more pronounced in rpS7 embryos , we selected them for the following analysis of apoptosis . S2 Fig provides images of the mutants compared to wild types and shows the morphological features that we use to initially select the mutants , such as the smaller head and eye , inflated hindbrain , and the dent in the mid-hind brain barrier . We measured overall levels of cell death in developing embryos carrying the rpS7 mutation with acridine orange ( AO ) , a stain commonly used to detect cells undergoing apoptosis in zebrafish embryos [36] . Fig 1A and 1B show that at 1 dpf , rpS7 mutant embryos reveal a significantly larger number of apoptotic cells compared to wild type controls . Closer visualization of an rpS7 mutant in Fig 1C reveals that these AO-stained cells are found clustered in the brain area and evenly distributed on the entire surface of the tail . The injection of a translation-blocking morpholino that we have previously demonstrated is specific to zebrafish p53 ( p53 MO ) but not a missense control morpholino ( mis MO ) , completely rescued the number of cells undergoing apoptosis in the rpS7 mutants , reinforcing a role for p53 in cell death as a result of RP loss ( Fig 1A and 1B ) [37] . The p53 MO also results in the rescue of morphological phenotypes commonly observed in ribosome biogenesis mutants such as the inflation of the hindbrain vesicle and pericardial edemas , a rescue effect that has been previously demonstrated in other zebrafish models of RP loss ( S2 Fig ) [15 , 38] . To determine if the p53 MO was sufficient to rescue the defective hematopoiesis of the mutants we used o-dianisidine , which stains hemoglobin-expressing cells evident at 2 dpf . Staining with o-dianisidine revealed that despite the rescue of apoptotic cells by p53 depletion observed at 1 dpf in the rpS7 mutants , the p53 MO was not able to rescue hematopoietic development to any appreciable degree ( Fig 1D ) . The levels of caspase-induced apoptosis in human CD34+ cells with RP loss vary depending on the RP [9] . To determine if the loss of RPs triggers caspase-induced apoptosis in zebrafish embryos we measured both the basal levels of activated caspase 2 or 3/7 and their levels in response to ionizing radiation , comparing wild type embryos with those carrying mutations in rpS7 and rpL11 , as well as rpS3 and rpL36 genes . Fig 2A and 2B indicate that in 2 dpf embryos the caspase 2 and 3/7 basal levels in the mutants is equivalent to wild types , and the caspase response to DNA damage is severely impaired in all the mutants . In fact , the RP mutant lines show the same suppressed caspase 2 and 3/7 response to ionizing radiation as the homozygous p53M214K/M214K mutant line ( Fig 2A and 2B ) , which is severely impaired in its ability to induce apoptosis [19] . To determine if other hallmarks of apoptosis such as DNA fragmentation are present in RP mutants we performed TUNEL assays on rpS7 embryos at 2 dpf . Fig 2C and 2D show that while there is no appreciable difference in the levels of TUNEL-positive cells between the rpS7 mutant and wild type embryos , the exposure of mutants to ionizing radiation results in a similar significant increase of TUNEL-positive cells as observed in the wild types . Closer visualization of these DNA-damaged embryos reveals a localization of TUNEL-positive cells in the brain area similar to what is seen with the AO staining , however the localization of TUNEL-positive cells in the tail region is found much more restricted to the dorsal area above the notochord ( Fig 2E ) . This may be due to the limits of penetration of the AO stain , or may suggest that the cells in this dorsal area are especially sensitive to DNA damage-induced apoptosis , as are the rapidly proliferating cells in the brain . The increased transcription of the zebrafish p53 gene and its p53Δ113 isoform ( a target gene of stabilized p53 ) has been described in several models of RP loss and likely reflects the early response of p53 that triggers an up regulation of its own transcription and the transcription of p53Δ113 [15 , 35 , 38–41] . In line with these results , we found using real-time quantitative PCR analysis with primers that amplify both full-length p53 and p53Δ113 that p53 mRNA levels were significantly higher in rpS7 and rpL11 mutants at 1 and 2 dpf both in the presence and absence of ionizing radiation compared to untreated wild type embryos ( Fig 3A and 3B ) . Semi-quantitative PCR analysis of p53 mRNA levels in several other RP-mutant embryos ( rpS3a , rpL23a , and rpL36 ) at 2 dpf similarly revealed equivalent levels of p53 transcription in the mutants compared to wild types ( S3 Fig ) . However , when we performed western blotting analysis using a zebrafish p53-specific antibody , we were unable to detect any appreciable amount of p53 protein in the rpS7 or rpL11 mutants in either the presence or absence of ionizing radiation at either 1 or 2 dpf ( Fig 3C ) . This is the case in all the mutant RP lines we tested including in rpS3a , rpL23a , and rpL36 ( S3 Fig ) . We often observe what may be a p53-specific isoform such as p53Δ113 on the western blots , but this may also be a p53 degradation product and in this work we cannot be certain of its exact identity . Taken together , the results suggest that although the p53 response to the RP mutation on a transcriptional level may function normally , an additional level of p53 post-translational regulation exists in the presence of RP mutations that serves to reduce p53 protein . We recently demonstrated that AKT phosphorylation activity is impaired in zebrafish embryos carrying mutations in rpS7 , rpL11 , and rpS3a [16] . Since the loss of AKT activity would theoretically lead to an increase of p53 proteasomal degradation due to an increase of GSK-3 phosphorylation of MDM2 , we used the proteasome inhibitor MG132 to determine if we could restore p53 expression . Fig 4A illustrates that MG132 is able to stabilize p53 in wild type embryo cells as expected , and moreover it is able to stabilize p53 in the rpS7 mutants . This supports the notion that the loss of p53 in the RP mutants is the result of excessive proteasomal degradation . We therefore hypothesized that stimulating AKT in the presence of ionizing radiation may rescue the p53 response to DNA damage . Fig 4B ( lanes 1–4 ) shows p53 stabilization approximately 6 hours after exposure of wild type embryos to ionizing radiation when the embryos are treated with the anti-oxidant Trolox , insulin , or both . Fig 4B ( lanes 5–8 ) illustrates that the addition of insulin , but not Trolox , immediately following the ionizing radiation results in a rescue of the stabilization of p53 in rpS7 mutants . These data suggest that overcoming the RP mutation-induced inhibition of AKT with insulin , which would stimulate AKT more directly than Trolox , is able to rescue the p53 stabilization response to ionizing radiation . A diagram illustrating how the impairment of the AKT pathway can lead to p53 protein degradation through GSK-3 is shown in Fig 4C . We hypothesized that the impairment of AKT activity observed in cells with RP mutations could cause constitutive activation of GSK-3 , phosphorylation of MDM2 , and ultimately resulting in the constitutive degradation of p53 . Therefore we reasoned that the inhibition of GSK-3 with lithium chloride ( LiCl ) in cells with RP mutations could restore p53 stabilization . Fig 5A and 5B show a partial rescue of p53 stabilization in rpL11 and rpS7 mutant embryos exposed to ionizing radiation followed by 6 hours of LiCl treatment . Finally , p53 expression is completely restored in rpS7 haploinsufficient MPNST tumor cells when the cells are plated overnight with different dosages of LiCl ( Fig 5C ) . This figure also shows LiCl induces the expression of what may be different isoforms of p53 such as p53Δ113 resulting from restored transcriptional activity of p53 in the tumor cells , for LiCl treatment of MPNST tumor cells expressing the transcriptionally impaired p53M214K/M214K mutant does not result in the appearance of same bands ( Fig 5D ) . However we are still unable to state for certain what the exact identity of these bands are , although they appear to be specific to p53 activation . These data strongly implicate the impairment of AKT , in particular the loss of AKT inhibition of GSK-3 , as a driving force behind the high levels of constitutive degradation of p53 observed in tumor cells with RP gene haploinsufficiency .
While stabilization of the p53 tumor suppressor has long been held the culprit for the cytopenia phenotype observed in human diseases linked to RP gene mutations , it has been a decade-long mystery as to why no p53 protein is detectable in RP haploinsufficient tumor cells . In the context of the RP mutant zebrafish embryos , we show that a similar loss of p53 protein is evident as the embryos age to 2 dpf , the maternal stores of RPs are depleted , and the RP deficiency resulting from the mutation becomes more severe . We do not believe that this loss of p53 in RP mutants is simply due to the inability of the cells to make protein per se , for in other zebrafish models of ribosome biogenesis deficiency such as nucleostemin , gnl2 , and nop10 mutants we are able to visualize robust p53 stabilization by western blot analysis even in the absence of ionizing radiation up until 4 dpf [5 , 6] . Interestingly , the nop10 , gnl2 , and nucleostemin genes all code for proteins with important roles in early ribosome biogenesis and the processing of rRNA . While it has been shown that there are indeed detectable defects in rRNA processing in DBA patient CD34+ cells with RP mutations , these are by and large more subtle than the defects we observe upon the loss of nop10 , gnl2 , or nucleostemin [5 , 6 , 42] . This suggests that , as with actinomycin D , defective rRNA processing is a critical mediator of p53 stabilization and may in fact be a causative agent in the cytopenia phenotype of human diseases such as dyskeratosis congenita . However , our data suggest the molecular pathology underlying the anemia in diseases linked to RP mutations is likely to implicate mechanisms that go beyond stabilization of the p53 protein . The fact that others and we observe no difference in early hematopoietic stem cell expression in RP mutant zebrafish embryos coupled to the anemia failing to be rescued upon p53 loss suggests that the anemia phenotype cannot be explained by the organism’s general p53 response to the ribosome biogenesis defects induced by RP loss [30] . Clearly the embryos younger than 1 dpf are experiencing some p53-induced apoptosis otherwise we would see no AO staining rescue upon injection of the p53 MO and no partial rescue of the morphological phenotypes . We therefore propose that p53 has such an early effect in the developing RP mutant embryo that by the time of 1 dpf ( approximately 30 hours post fertilization is when the experiments would begin ) all that we are able to observe by laboratory techniques are the apoptotic cells in the wake of a brief p53 activation . The remaining live cells at 1 dpf appear devoid of p53 protein , and if the apoptotic cells do still carry p53 protein we cannot detect it by western blotting . We were unable to detect p53 by western blotting in any embryo younger than 1 dpf , but this is likely a technical issue reflecting the abundance of yolk protein still present at this early stage . The apoptotic cells rescued by p53 loss that we are able to observe at 1 dpf also do not appear to be important for red cell development , as we detect no rescue of the anemia phenotype upon p53 loss . The results of TUNEL assay suggests that some DNA fragmentation upon acute DNA damage is still possible in the 2 dpf RP mutant cells despite no evidence of p53 stabilization or caspase activation . Other studies have reported similar findings of caspase-independent DNA fragmentation in cancer epithelial cells in response to diverse toxins , and the authors of this work suggest that the DNA damage induced by these toxins may lead to noncaspase-mediated proteolytic activation of DNases [43] . Interestingly , studies in Drosophila have shown that ionizing radiation leads to cells acquiring RP haploinsufficiency , and that these cells undergo apoptosis that is also p53-independent [44] . Taken all together , our work supports a model where the impairment of mature red blood cell formation in organisms with RP deficiency is due to cell loss that is not dependent on either p53 stabilization or caspase activation . However , it should be pointed out here that the work in this study is entirely based on zebrafish models of DBA , which carry homozygous RP mutations ( as noted in the introduction , the haploinsufficient mutants have no embryonic phenotype except a slight growth defect ) . Therefore it may be that p53 has a more prominent role in cells with true RP haploinsufficiency , and this remains an issue that must be kept in mind when interpreting data generated from zebrafish models of DBA . While other studies have suggested that the p53 MO or p53M214K/M214K mutant background rescues both general apoptosis and the number of hemoglobin-expressing cells in embryos deficient for RPs [15 , 38] , we believe our genetic models are more consistent than using morphants and that our approach to quantifying the mutant phenotypes are both more robust ( we use sample sizes N > 100 ) and reliable ( we genotype the embryos after the blind scoring to resolutely identify the homozygotes ) . Other studies support us by demonstrating that p53 independent pathways are contributing to the anemia phenotype of RP-mutant zebrafish embryos , and that the loss of p53 rescues the morphological abnormalities but not the anemia phenotype of embryos with reduced RP expression [45–47] . It remains an open question as to what the major pathways beyond p53 and caspase activation contribute to the death of RP deficient cells and the anemia phenotype in zebrafish models of DBA . Interestingly , a recent study has shown that the mutation of rpS19 in zebrafish embryo erythrocytes specifically reduces the translation , but not the transcription of the hemoglobin gene hbbe3 [30] . We also recently reported that the up regulation of autophagy , the cellular process of self-digestion that is tightly regulated during hematopoiesis , is observed in both zebrafish models of DBA and in human DBA cells with RP haploinsufficiency [16] . It may therefore be that RP loss in maturing erythrocytes derails their proper differentiation by failing to translate critical mRNAs , or by the constitutive activation of autophagy resulting in the erythrocyte progenitor cells essentially eating themselves before they are able to fully differentiate . The former possibility is supported by several findings suggesting that L-leucine , an amino acid that increases translation by activating the mTOR pathway , is able to partially rescue the anemia phenotype of zebrafish RP morphants and increases the number of erythroid cells in red cell culture assays where CD34+ cells are infected with shRNAs against RPS19 or RPS14 [47 , 48] . The latter possibility is enticing for one would not expect constitutively up regulated autophagy to elicit an apoptotic response . Or the cells may be undergoing an as of yet unknown p53- and caspase-independent mechanism of cell death that awaits identification . We had previously reported that the addition of MG132 to the MPNST cells with RP gene haploinsufficiency was not able to restore p53 stabilization [20] . At the time this led us to believe that the block in p53 expression was at the level of protein synthesis , since MG132 should be able to stabilize any protein undergoing ubiquitination and proteasomal degradation . However our present study suggests otherwise , that in fact the MG132 treatment is not sufficient to restore p53 stabilization in RP mutant MPNST cells . There may be additional factors at the tumor level contributing to the degradation of p53 that are as of yet not known , or it may be that the robustness of the degradation in the tumors is much stronger than in the embryos . While our present study is not able to delineate between these possibilities , it may be that more powerful proteasome inhibitors are capable of restoring p53 to the same levels as we observe with the LiCl treatment of tumor cells , and these studies will be of great future interest . It should also be pointed out here that given the very wide range of effects that LiCl has on many cellular signaling pathways , there may be other effects beyond the AKT pathway that are contributing to the restoration of p53 that we observe . Very recent work suggests that a mechanism for malignant transformation in RP-haploinsufficient cells involves cells acquiring oncogenic mutations that allow for the bypass of a 60S ribosomal subunit integrity checkpoint [49] . We propose that these mutations ( which have yet to be identified ) coupled to excessive degradation of wild type p53 would be sufficient to promote malignant transformation . Interestingly , the T-ALL cells with RPL5 and RPL10 mutations , similar to the RP-haploinsufficient zebrafish tumor cells , the p53 gene remains wild type ( personal communication with Dr . De Keersmaecker ) . We thus posit that cells with RP deficiencies are able to exert selective pressure to overcome the p53 activity not just by acquiring p53 gene mutations but also by suppressing AKT activity . DBA is a rare disease with a prevalence that is approximately 7 in 1 million live births [50] . In addition , since the majority of DBA patients die early in life from bone marrow failure or from complications stemming from chronic blood transfusions , the number of DBA patients who ultimately experience malignant transformation is very low ( it would be next to impossible for us to obtain DBA patient-derived tumor tissue for p53 analysis ) . In the absence of a mammalian RP mutant cancer model , the zebrafish RP mutants to date remain the best possible option for molecular studies of RP mutation-driven tumors . That said , we believe our results may still have some important implications for human leukemia . For example , acute myeloid leukemia ( AML ) has very low rates of p53 genetic inactivation ( similar to the RP mutant MPNSTs ) compared to very high rates in many other tumor types [51 , 52] . Inhibition of GSK-3 with small molecules in AML cells leads to increased differentiation , impaired growth and proliferation , and the induction of apoptosis [53] . Interestingly , the immunomodulating agent lenalidomide , used to treat multiple myeloma , results in the phosphorylation of GSK-3 at the same serine residues that inhibit GSK-3 phosphorylation of MDM2 [54] . This drug has also been reported in 5q-MDS patients to result in increased survival and a reduced risk of transformation to AML [55] , raising the possibility that one mechanism of action of lenalidomide is to inhibit GSK-3 phosphorylation of MDM2 and restore a normal p53 response to the acquisition of oncogenic mutations . In sum , we suggest the new mechanisms driving p53 loss that we report here may be useful pathways to target in some cancers .
Zebrafish mutants were created and maintained as described [19 , 28 , 56] . Animal experiments were conducted in accordance with the Dutch guidelines for the care and use of laboratory animals , with the approval of the Animal Experimentation Committee ( Dier Experimenten Commissie ) of the Royal Netherlands Academy of Arts and Sciences ( Koninklijke Nederlandse Akademie van Wetenschappen [KNAW] ( Protocol # 08 . 2001 ) . Performed as previously described using probes against scl [31] and βE1-globin [57] . At least three independent stains were performed with clutches > 60 embryos . Clutches were stained simultaneously and mutants confirmed afterwards using Sanger sequencing , as previously described [58] . Embryos were sorted by gross phenotype and photographed with a Leica MZ FLIII microscope . At least three independent clutches were analyzed and the expected Mendelian ratio of mutants was always observed . Mutants were confirmed afterwards using Sanger sequencing , as previously described [58] . Embryos were microinjected with ~1ng MO at the one-cell stage . The p53 MO ( Gene Tools , Inc . Philomath , OR , USA ) is 5’-GCGCCATTGCTTTGCAAGAATTG-3’ and has been previously described [59] . The missense MO sequence is 5'-CATGTTCAACTATGTGTTAGCTTCA-3' ( Gene Tools , Inc . Philomath , OR , USA ) Live 1 dpf embryos were incubated in E3-embryo medium + 10μg/mL AO stain ( Sigma ) in the dark for 30 min . Photos were obtained using a Leica MZ FLIII microscope and cells were blindly counted within a defined area that included the tail starting at the dorsal end of the yolk extension using Image J v1 . 44 software . At least 8 animals per condition were used for counting . Embryos were genotyped afterwards using Sanger sequencing , as previously described [58] . At least 100 embryos per clutch of a mating between two heterozygous hi1034b fish were injected at the one-cell stage with MOs as described above . At 2 dpf they were stained with o-dianisidine ( Sigma ) as previously described [16] . Scoring of the phenotype severity was done blindly . Embryos were genotyped afterwards using Sanger sequencing as previously described [58] . 2 dpf embryos were untreated or exposed to 25 Gy of ionizing radiation . Six hours after irradiation TUNEL assays were performed as previously described and TUNEL-positive cells counted as for AO staining [6] . Embryos at 2 dpf were either untreated or subjected to 25 Gy ionizing radiation , and mechanically lysed 6 hours later with a P200 pipette ( each sample = 3 embryos per well of a 96-well plate ) using 100μL of the western blotting lysis buffer described above . The Caspase-Glo® 2 or 3/7 assay ( Promega , Madison , WI , USA ) was then performed per the manufacturer’s instructions . 5 embryos per sample were added to 100μL Trizol ( Life Technologies , Carlsbad , CA , USA ) , RNA was isolated and used to make cDNA with the iScript cDNA Synthesis Kit ( Bio-Rad , Hercules , CA , USA ) per the manufacturer’s instructions . Primers used were as follows: p53 forward 5’- GCTTGTCACAGGGGTCATTT-3’ , p53 reverse 5’-ACAAAGGTCCCAGTGGAGTG-3’ , GAPDH forward 5’-GGATCTGACAGTCCGTCTTGAGAA-3’ , GAPDH reverse 5’- CCATTGAAGTCAGTGGACACAACC , actin forward 5’-GCCCATCTATGAGGGTTACG-3’ , actin reverse 5’-GCAAGATTCCATACCCAGGA-3’ . Quantitative PCR ( qPCR ) was performed using iQ SYBR Green Super Mix and a MyiQ Single-Color PCR thermal cycler ( Biorad , Hercules , CA , USA ) . Each experiment was performed in biological triplicate . p53 mRNA expression in mutants relative to wild types was normalized to GAPDH and calculated according to the Cτ method . Semi-quantitative analysis was performed with a standard PCR method using either p53 or actin primers , the products run on a 1% agarose gel . Statistics were performed with a Student’s t-test . Embryos were subject to 25 Gy ionizing radiation and then lysed 6 hours later . 350nM insulin , 10mM Trolox or 100μM LiCl ( all from Sigma ) was added to the E3 media immediately following the ionizing radiation . For the MG132 ( Sigma ) experiment , embryo cells were dissociated mechanically with a 200μL pipette tip and added to DMEM media +10% FCS with or without 20μM MG132 for 6 hours at 28°C before lysing the cells . The zebrafish specific p53 antibody and western blotting of zebrafish embryos and MPNST cells has been previously described [20] . For MPNST cells the tumor was dissected , cells mechanically dissociated with a P200 pipette , and plated in a 6 well dish with the indicated concentration of LiCl at 28°C in DMEM + 10% fetal calf serum ( Life Technologies , Carlsbad , CA , USA ) media overnight . 50μg of total protein from the MPNST cells was used for each sample , measured by Bradford assay ( BioRad , Hercules , CA , USA ) . Other antibodies used at 1:1000 dilution included anti-phospho-AKT ( Cell Signaling #9275 ) and anti-actin ( Santa Cruz Biotech , #sc-1616 , Santa Cruz , CA , USA ) . Antibodies used at 1:5000 included donkey anti-goat IgG-HRP ( Santa Cruz Biotech , #sc-2020 , Santa Cruz , CA , USA ) and sheep anti-mouse IgG-HRP ( GE Healthcare , #NA931 , Little Chalfont , Buckinghamshire , United Kingdom ) . Quantifications of western blots were performed using Image J v1 . 44 software . Unless otherwise stated , all embryos subject to western blotting were 2 dpf . | The p53 tumor suppressor is the most commonly mutated gene in human cancers . However , cancer cells exploit multiple mechanisms to silence the p53 pathway in addition to inactivation of the p53 gene . We previously reported that one of these mechanisms is found in tumor cells with ribosomal protein ( RP ) gene mutations . These cells transcribe wild type p53 mRNA yet do not stabilize p53 protein when exposed to DNA damaging agents . In this work we demonstrate that this loss of p53 protein is due to its constitutive degradation . This degradation is due to impairment of the AKT pathway , which normal signals for p53 to stabilize when the DNA is damaged . By re-activating the AKT pathway in RP-mutant cells we are able to restore p53 stabilization and activity , which may hold clinical significance for cancer treatment . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2015 | Ribosomal Protein Mutations Result in Constitutive p53 Protein Degradation through Impairment of the AKT Pathway |
Biological systems produce phenotypes that appear to be robust to perturbation by mutations and environmental variation . Prior studies identified genes that , when impaired , reveal previously cryptic genetic variation . This result is typically interpreted as evidence that the disrupted gene normally increases robustness to mutations , as such robustness would allow cryptic variants to accumulate . However , revelation of cryptic genetic variation is not necessarily evidence that a mutationally robust state has been made less robust . Demonstrating a difference in robustness requires comparing the ability of each state ( with the gene perturbed or intact ) to suppress the effects of new mutations . Previous studies used strains in which the existing genetic variation had been filtered by selection . Here , we use mutation accumulation ( MA ) lines that have experienced minimal selection , to test the ability of histone H2A . Z ( HTZ1 ) to increase robustness to mutations in the yeast Saccharomyces cerevisiae . HTZ1 , a regulator of chromatin structure and gene expression , represents a class of genes implicated in mutational robustness . It had previously been shown to increase robustness of yeast cell morphology to fluctuations in the external or internal microenvironment . We measured morphological variation within and among 79 MA lines with and without HTZ1 . Analysis of within-line variation confirms that HTZ1 increases microenvironmental robustness . Analysis of between-line variation shows the morphological effects of eliminating HTZ1 to be highly dependent on the line , which implies that HTZ1 interacts with mutations that have accumulated in the lines . However , lines without HTZ1 are , as a group , not more phenotypically diverse than lines with HTZ1 present . The presence of HTZ1 , therefore , does not confer greater robustness to mutations than its absence . Our results provide experimental evidence that revelation of cryptic genetic variation cannot be assumed to be caused by loss of robustness , and therefore force reevaluation of prior claims based on that assumption .
Biological systems produce phenotypes that appear to be robust to genetic and non-genetic sources of variation [1] , [2] . It has been proposed that understanding robustness is crucial for understanding healthy and diseased states [3]–[5] , as well as the potential for populations to adapt to evolutionary pressures [6] , [7] . Advancing this understanding will require much greater knowledge of the mechanisms by which robustness is achieved [2] . A particularly important gap in our understanding is that no specific gene product has been shown to confer robustness against naturally occurring mutations , over and above some baseline level of robustness that would exist in the absence of the gene product [8] . It might come as a surprise to some people that this lacuna exists . After all , there is a long history of studies showing that various perturbations , including loss or impairment of specific gene products , reveal previously hidden ( “cryptic” ) genetic variation [2] , [8]–[16] . The earliest study was by Waddington , who observed a crossveinless wing phenotype in Drosophila melanogaster only after heat stress and only in some individuals [11] . The basis of the crossveinless phenotype was genetic , as it could be selected for , and lines were established in which the wing phenotype was highly penetrant even without exposure to heat stress [11] . The most prominent recent example of such an experiment involves the molecular chaperone Hsp90 , the impairment of which reveals cryptic variation in several evolutionarily distant species [10] . In flies , impairment of HSP90 reveals phenotypic variation in several traits , and , as in the Waddington experiments , this variation is heritable [12] . Although recent work has shown that severe Hsp90 impairment induces mutations via mobilization of transposable elements [17] , new mutations do not explain all revealed variation [18] , [19] . Other recent work has expanded the scope of studies of cryptic genetic variation to other model organisms , including the nematode worm Caenorhabditis elegans [20] , [21] , the budding yeast Saccharomyces cerevisiae [22]–[24] and the bacterium Escherichia coli [25] . Such studies have also expanded to non-model species , including tobacco hornworms [26] , dung flies [27] and a beetle-associated nematode [28] . The list of genes in D . melanogaster whose impairment can reveal cryptic variation is poised to expand as well: a genetic screen using deficiency chromosomes recently showed that there are at least 10 regions of the D . melanogaster genome containing a gene that reveals cryptic variation in wing morphology when hemizygous [29] . The decades of studies revealing cryptic variation have been taken as evidence that the wild-type state is more robust to mutations than the perturbed state , and that the perturbed gene products normally contribute to this robustness . However , this logic is flawed [8] . To appreciate the logical flaw , consider the following example . A gene X has two alleles , X1 and X2 , that confer equal robustness against mutations entering a population . That is , the distribution of mutational effects , including the proportion of mutations with no phenotypic effect ( neutral mutations ) , is identical in the X1 and X2 genetic backgrounds . In this example , the only difference between X1 and X2 is in which mutations they make neutral ( Figure 1 ) . A population fixed for X1 will accumulate the mutations that are neutral in the context of X1 . Replacing X1 with X2 in members of the population will reveal cryptic genetic variation — the subset of mutations that are neutral in the X1 genetic background but non-neutral in the X2 genetic background . Likewise , a population fixed for X2 will accumulate the mutations that are neutral in the context of X2 , a subset of which are non-neutral in the context of X1 . Replacing X2 with X1 will therefore also reveal cryptic genetic variation . It is thus clear that the revelation of cryptic genetic variation is not sufficient to indicate a decrease in robustness . Indeed , cryptic genetic variation can be revealed even by a perturbation that makes the system more , not less , robust . The revelation of cryptic genetic variation by a perturbation merely indicates that some mutations are conditionally neutral; it does not indicate whether the perturbation is more likely to convert a neutral allele into one with a phenotypic effect than to do the reverse [8] . To ascertain the relative amount of mutational robustness conferred by alternative alleles , it is necessary to measure , in the context of each allele , the phenotypic effects of a large sample of spontaneous mutations . As has been noted [8] , an approximation of this experiment has been conducted in D . melanogaster [30] and E . coli [31] , [32] . However , the mutations assayed in those studies had survived the filter of natural selection , and so a major assumption had to be made that the lines had not experienced selection toward a common optimum [8] . The most appropriate sample of mutations must include those mutations that would otherwise be purged by selection in the presence of one or the other allele [8] . One way to construct the appropriate sample is to allow mutations to accumulate in independent lines by serial passaging through bottlenecks . The bottlenecks keep the effective population size low and therefore minimize the power of natural selection to purge deleterious alleles . Then a corresponding set of lines in which one allele is replaced with the other allele can be constructed . The difference in phenotypic variation between the two sets of lines — or , in the parlance of quantitative genetics , the difference in their mutational variances [33]–[35] — indicates the relative robustness conferred by the two alleles . Here , we perform this test using strains of S . cerevisiae . Yeast , due to their rapid growth rate , ease of handling , and routine genetic techniques , are highly amenable to large-scale experiments such as this one . To avoid the problem of using organisms maintained under selection when testing the extent of mutational robustness conferred by a candidate gene , we use yeast lines generated in a mutation accumulation ( MA ) experiment [36] , [37] . In this MA experiment , 151 replicate lines were founded from a single diploid ancestral strain , and cultured independently for approximately 2062 generations with single-individual bottlenecks at approximately 20-generation intervals [37] . The use of a diploid ancestor promoted genome stability [38] and , because propagation was asexual , completely shielded recessive deleterious mutations from selection . Based on estimates from a different MA experiment [39] , each line should contain approximately 8 single-nucleotide changes ( point mutations ) per haploid genome . These strains also likely acquired mutations in repetitive sequences . Again based on prior estimates [39] , the expected number of microsatellite mutations per haploid genome per line is approximately 4 and the expected number of mutations in short homopolymer runs per haploid genome per line is approximately 638 . We chose to test if the presence of a particular chromatin regulator , HTZ1 , confers greater robustness to new mutations than its absence . HTZ1 encodes a histone variant , H2A . Z , that can take the place of histone H2A in nucleosomes . Nucleosomes containing HTZ1 are usually found in the promoter regions of repressed or stress-responsive genes , and HTZ1 is necessary for these genes' full activation [40] . Despite its widespread effects on gene regulation , HTZ1 is not required for viability and indeed its deletion has only a modest effect on growth , making it a convenient choice for genetic analysis of robustness . Moreover , HTZ1 was identified in a systematic screen for yeast genes that contribute to robustness to microenvironmental sources of variation , including the immediate external environment as well as internal stochastic processes [41] . That screen identified a few hundred genes that , when absent , significantly increased variance of many cell-shape traits [41] . Genes involved in chromosome organization were over-represented among the significant genes , and HTZ1 was one of the most highly significant [41] . Although only microenvironmental sources of variation were present in that study , as cells were genetically identical and cultured together , several lines of argument have led many to predict that genes contributing to robustness to one source of variation will increase robustness to other sources as well [2] , [42] , [43] . The prediction is especially strong for chromatin regulators , which have been found in other studies to affect levels of phenotypic variation due to genetic and systematic environmental differences [44]–[46] . A recent study found that chromatin regulators suppress gene expression differences between S . cerevisiae and its close relative S . paradoxus [44] . Orthologous genes show similar expression profiles in wild-type strains of each species . However , the expression profiles became more dissimilar when any one of eight chromatin regulators , including HTZ1 , was deleted [44] . A study that investigated chemical-protein relationships in S . cerevisiae found that although chromatin regulators did not directly interact with many chemicals , their presence increased resistance to many chemicals [45] . In Drosophila , impairment of a chromatin-regulator network causes developing flies to be more sensitive to temperature variation [46] . We knocked out HTZ1 in 79 MA lines . We first converted the existing diploid MA lines [37] to haploids to study the effects of accumulated mutations without the complication of dominance . We did this conversion before deleting HTZ1 in each line so that the same sample of mutations would be assayed in the presence of HTZ1 as in the absence of HTZ1 . That is , we created 79 strain pairs , with the members of each pair having identical genotype except at the HTZ1 locus . We measured phenotypic variation in cell morphology in each strain , using an established assay [47] that we adapted for higher throughput . In this assay , cells are fixed and stained with fluorescent markers of the cell surface and the nucleus , then imaged . Based on these markers , CalMorph image-analysis software automatically measures 187 cell shape parameters , such as cell diameter and budding angle [47] . We present an analysis of phenotypic variation within and among HTZ1+ and HTZ1− lines as a test of the contribution of wild-type HTZ1 function to robustness .
The main goal of this study was to test whether the chromatin protein HTZ1 increases robustness of morphological phenotypes to new mutations , by collecting morphological data on 79 pairs of HTZ1+ and HTZ1− MA lines . As described in Materials and Methods , morphological measurements of individual cells were obtained by adapting an established method of automated image analysis of fluorescence micrographs [47] . HTZ1 was chosen as a candidate mutational-robustness factor in part because it had previously been found to confer robustness to microenvironmental variation , in that morphological variation increased among genetically identical cells when HTZ1 was deleted [41] . Before addressing our main goal , we therefore first sought to confirm this previous finding , by asking if HTZ1 deletion increases within-line variation of morphological phenotypes . Because the morphological assay consists of partially redundant phenotypes , we used principal component analysis ( PCA ) to identify orthogonal linear combinations of the phenotypes to use for downstream quantifications of morphological variation . PCA was performed separately for the three cell types ( unbudded , small-budded and large-budded ) , because each type has its own suite of phenotype measurements . Only principal components that explained more variance than the random expectation were used in the analysis ( see Materials and Methods ) . This reduced the dimensionality of the data to six significant principal components for unbudded cells , 10 for small-budded cells and 17 for large-budded cells ( Figure S1 ) . We estimated variance parameters by fitting a linear model . A standard approach to doing so would be to use maximum-likelihood based methods to fit mixed models in which genotype is a fixed effect and MA line is a random effect . However , we chose instead to estimate variance components using a Bayesian approach based on Markov chain Monte Carlo ( MCMC ) sampling ( see Materials and Methods ) . The MCMC approach has the advantages of: 1 ) high flexibility in modeling different within-line and between-line variances for each HTZ1 genotype , and 2 ) straightforward assessment of the precision of variance estimates by constructing credible intervals from the posterior distributions of the parameters . We compared estimates of within-line variance for HTZ1+ and HTZ1− lines to determine the effect of HTZ1 on microenvironmental robustness . As shown in Table 1 , for each of the 33 principal components , the within-line variance is greater in HTZ1− lines than in HTZ1+ lines . The differences are substantial: in only three cases is there overlap of the estimates' 95% highest posterior density ( HPD ) intervals ( credible intervals that are akin to confidence intervals but computed as the shortest intervals containing 95% of the posterior-distribution density ) . Likewise , in only these three cases did the 95% credible interval for the difference between the HTZ1− and HTZ1+ within-line variances include 0 or negative values . To confirm that this result was not due to any unknown bias of the MCMC approach , we also compared within-line variances using a model-independent approach . We compared median-corrected median absolute deviations ( a robust measure of within-line spread ) between HTZ1+ and HTZ1− lines and found , as expected , lower within-line spread for HTZ1+ than for HTZ1− ( see Text S1 , Figure S2 ) . These results confirm that HTZ1 mutation increases within-line variation , as shown in our previous study [41] . That is , the results confirm that HTZ1+ increases robustness to microenvironmental sources of variation , and suggest this ability is not dependent on the line background . We next asked if HTZ1 affects between-line morphological variation and consequently affects robustness to new mutations . Figure 2 shows the HTZ1+ and HTZ1− line means for three principal components from each cell type ( see Figure S3 for the remaining principal components from each cell class ) . MCMC-based estimates of the between-line variances of HTZ1+ and HTZ1− lines and 95% credible intervals are shown in Table 2 . For nine of 33 principal components , the 95% credible intervals for the difference between the HTZ1+ and HTZ1− between-line variances do not include 0 . In four of these cases , the HTZ1− lines have higher between-line variance , whereas in the other five the HTZ1+ lines have higher between-line variance . As above for the within-line variance comparison , we used a model-independent test to corroborate the MCMC-based analysis . Using Levene's test for differences in between-line variance yielded qualitatively similar results ( see Text S1 , Table S1 ) . Taken together , these results demonstrate that HTZ1 does not systematically affect between-line variance , especially considering that the principal components that showed a significant difference did not consistently show an effect in the same direction . The absence of a systematic effect on between-line variance supports the scenario diagrammed in Figure 1 , where neither HTZ1+ nor HTZ1− contributes more to genetic robustness than the other . Instead , each HTZ1 allele interacts epistatically with a certain subset of accumulated mutations to produce the range of morphological variation seen in this experiment . The subsets for the two alleles may overlap only partially , but their sizes , as measured by the alleles' effects on morphology , are similar . Because principal components represent combinations of morphological trait values , as opposed to an individual trait such as cell circumference or budding angle , relating a principal component to a biologically meaningful phenotype can be difficult . However , inspection of individual cells from line pairs with divergent mean principal component values indicates that this difference accurately reflects underlying differences in cell morphology . For example , consistent with which original phenotypes load heavily onto each principal component , principal component 4 for unbudded cells appears to correspond to how elongated a cell is , whereas principal component 1 for large-budded cells appears to correspond to cell size ( Figure 3 ) . In principle , the absence of consistent , major differences in between-line variance could be caused either by the lack of any effect of HTZ1 genotype on line means or by a significant genotype-by-line interaction effect that takes the form of line crossing rather than line spreading . Note that “line” here refers not to MA line , but to a line as a geometric object connecting the means of the two HTZ1 genotypes of the same MA line in a plot such as Figure 2 . That is , line crossing refers to the change in rank order of line means , and line spreading refers to the change in dispersion of line means . Figure 2 and Figure S3 appear to indicate a large extent of line crossing . To measure variance due to genotype-by-line interaction , and to partition this interaction into components representing line crossing and line spreading , we used the variance components estimated by MCMC . The genotype-by-line interaction variance , Vg × l , is estimated as [48] , [49]:where VHTZ1+ is the HTZ1+ between-line variance , VHTZ1− is the HTZ1− between-line variance and CovHTZ1+ , HTZ1− is the genetic covariance between HTZ1+ and HTZ1− . A test of the significance of the genotype-by-line interaction is not possible with the MCMC approach , because: 1 ) the use of an information criterion that penalizes additional parameters , akin to the Akaike information criterion or Bayesian information criterion , is not well established for model selection in this context; and 2 ) variances are constrained to be positive and therefore credible intervals will not overlap 0 even for negligible variances . Nevertheless , an indication that the interaction variance is substantial is that it is similar in magnitude to the magnitudes of the between-line variances ( compare Table 2 to Table 3 , which shows the interaction-variance estimates and their 95% credible intervals ) . An equivalent way of saying this is that the genetic correlation is far from unity [48] , as is indeed the case for the correlation between HTZ1+ and HTZ1− lines for each principal component . It is possible to perform a significance test for nested models fit by maximum-likelihood approaches . We did this , applying a likelihood-ratio test to models with and without an interaction term , and found that models containing the interaction term fit the data better than models without , for all principal components ( see Text S1 , Table S1 ) . A genotype-by-line interaction term can be partitioned into terms representing the spreading of line means and the crossing of line means [30] , [48] , [49] , as described in Materials and Methods . The percentages of the interaction variance explained by line crossing and spreading , along with credible intervals , are reported in Table 4 . For each principal component , the vast majority of the genotype-by-line interaction is indeed explained by line crossing , rather than the spreading of line means ( median percentage of interaction explained by line crossing = 99 . 9%; Table 4 , column 3 ) . We obtained similar results when using variance estimates from models fit by restricted maximum likelihood ( see Text S1 , Figure S4 , Table S1 ) . Note that , for the principal component with the highest estimate of the spreading component ( small-budded principal component 4 , estimated percentage spreading = 12 . 46% ) , the HTZ1+ between-line variance is higher than the HTZ1− between-line variance . That is , the spreading is in the direction of the wild type rather than the mutant . To confirm that our conclusions were not dependent on the method of dimensional reduction , we repeated our interaction analysis using partitioning around medoids ( PAM ) instead of PCA , as we had done previously to reduce phenotypic redundancies [41] . The number of clusters used in a PAM analysis is often chosen to maximize average silhouette width [50] . Alternatively , the number of significant principal components ( determined as described above ) can be taken as an appropriate number of clusters . For our data , the number of significant principal components is smaller than the number of clusters with the highest mean silhouette width . However , the mean silhouette width corresponding to this smaller number of clusters is very similar to the maximum mean silhouette width , suggesting that adding more clusters than the number corresponding to the number of significant principal components does not improve the clustering much ( Figure S5 ) . We therefore used the smaller number of clusters . In all respects , our main findings were not altered when repeating analysis with PAM instead of PCA . HTZ1− lines have greater within-line variance for each medoid , and the differences are substantial ( Table S2 ) . In only two cases did the 95% credible interval of the difference between within-line variances overlap 0 . Using medoids instead of principal components resulted in 11 of 33 medoids showing evidence of a difference in between-line variance , in the form of the 95% credible interval of the difference in between-line variance not overlapping 0 . For six of these 11 , the HTZ1+ lines showed higher between-line variance , and for the other five the HTZ1− lines showed higher between-line variance ( Table S3 ) . The magnitude of the genotype-by-line interaction variance was similar to the magnitude of the between-line variance , as in the PCA-based analysis . In addition , the model-selection analysis of models fit by maximum likelihood showed that for 32 of 33 medoids the best-fit model includes an interaction term ( Text S1 , Table S4 ) . As noted in Materials and Methods , the MA lines and their HTZ1− derivatives are expected to produce red colonies due to the presence of an ade2-101 mutation , yet a subset of lines displayed white sectors or colonies , suggesting an epigenetic switch was at play . We therefore repeated analyses with the restricted set of line pairs that showed stable-red inheritance . The restricted analysis yields the same general conclusions as the analysis with the full set of strains . For each principal component , HTZ1− lines have a greater within-line variance . For only two principal components did the 95% credible interval of the difference between within-line variances overlap 0 . In addition , for only four principal components did the credible intervals of the difference in between-line variances not overlap 0 . For three of these four , the HTZ1− lines had greater between-line variance and in one the HTZ1+ lines did . For each principal component , the magnitude of the between-line variance and genotype-by-line interaction variance is similar . The model-selection analysis showed that for all principal components , the best-fit model included an interaction term . One potential , related concern about our experimental approach is that estimates of line means and between-line variances might be sensitive to the number of lines assayed . However , it is unlikely that sampling any more lines would change our results . Estimates of the mean and standard deviation of line means for various principal components were recalculated with a line sample size ranging from 5 to 79 pairs . As shown in Figure S6 for the example of principal component 1 for unbudded cells , these estimates do not change greatly even in the range where the number of lines is approximately half of the number we used , and they show extremely small differences as the number of lines approaches 79 . This observation suggests that sampling more MA lines would not change the results of this experiment , and that the amount of morphological variation caused by accumulated mutations has been adequately measured . An alternative ( although clearly not independent ) way of framing the question of whether HTZ1 increases robustness to mutations is to compare the mutational variance , VM , estimated from the HTZ1+ MA lines to that estimated from the HTZ1− MA lines . If HTZ1+ were to contribute to greater robustness to mutations , then the VM should be higher for the HTZ1− lines than the HTZ1+ lines . The magnitude of VM ( scaled by the environmental variance , VE ) is also of interest , as it relates to the mutational target size for the trait of interest and the neutral expectation for segregating variation [33] , [34] . VM for cell morphology , which had not been previously estimated , was estimated for lines with and without HTZ1 ( see Materials and Methods ) . The VM estimates and their 95% credible intervals are given in Table 5 for HTZ1− and HTZ1+ lines for each principal component . The average VM/VE of HTZ1− principal components is 5 . 1×10−5 and of HTZ1+ principal components is 7 . 8×10−5 . Plots of VM and VE estimates for each principal component in HTZ1− versus HTZ1+ lines are shown in Figure 4 , and capture our main conclusions: HTZ1 mutation increases environmental variance but does not increase mutational variance . Our VM/VE estimates are lower than those of previous studies measuring a variety of phenotypes in a variety of organisms , which tended to report VM/VE values between 10−4 and 5×10−2 [34] . Future experiments can address why this discrepancy exists . Our results might reflect a genuinely restricted range of mutational variance for this suite of traits in this organism ( compared to other traits in multicellular organisms in particular ) . Alternatively , it must be considered that VM/VE might increase with more precise measurements of VE , which is necessarily a combination of actual biological variation within lines and variation in measurement . Decreased measurement variance could be achieved , for example , by more precisely staging cells ( reducing the variance in cell-cycle stage at which cells are measured ) .
Previous studies have demonstrated the release of cryptic genetic variation after a genetic or environmental perturbation [2] , [9]–[16] , [20]–[29] . This release is often conflated with a breakdown in mutational robustness [8] . However , the release of cryptic genetic variation is not a reliable indicator of mutational robustness when the genetic backgrounds that are studied have been subject to artificial or natural selection , as was the case in all prior studies [8] . Our study compares , for the first time , the relative mutational robustness conferred by two alleles , in a panel of genetic backgrounds that had accumulated naturally occurring mutations with minimal selection . We find that MA lines that are HTZ1+ are not more robust to new mutations than lines that are HTZ1− , as the two genotypes display similar extents of morphological variation across lines . Nevertheless , we find strong evidence of epistasis between HTZ1 and new mutations , manifest as a significant interaction between HTZ1 genotype and line . We also find strong evidence corroborating our previous finding [41] that HTZ1 deletion increases within-line variation . Taken together , our results indicate that wild-type HTZ1 function increases robustness to microenvironmental variation but not to mutations . Theoretical studies have tended to predict congruence between robustness mechanisms , or in other words that mechanisms contributing to robustness to one source of variation will contribute to robustness to other sources [2] , [42] , [43] . The results presented here do not support this conclusion , at least with regard to HTZ1 . Additional doubt has been cast on the congruence hypothesis by studies in Drosophila [15] and E . coli [51] . Future studies will be required to test whether the congruence hypothesis also does not hold for other genes or whether HTZ1 and these other cases are aberrations . Our results highlight the importance of using MA lines for tests of mutational robustness . The lines used in this study have accumulated extensive genetic variation affecting cell morphology ( Figure 2 and 3 ) . It is reasonable to assume that had the HTZ1+ lines been exposed to stabilizing selection , then the phenotypic and genetic variation among HTZ1+ lines would have been reduced . However , replacing HTZ1+ with HTZ1− in this scenario would still likely reveal extensive phenotypic variation , because the mutations allowed to accumulate in the presence of HTZ1+ can have very different effects in its absence . We hypothesize that the findings of greater expression divergence between S . cerevisiae and S . paradoxus upon deletion of chromatin regulators ( including HTZ1 ) [44] are due to the effect of selection , and that analysis of expression variation in wild-type and mutant MA lines would not reveal a suppressive effect of HTZ1 on expression change . Our results are reminiscent of work on Hsp90 in yeast , even though that work was conducted with strains that had been subject to natural selection [23] , [52] . Specifically , Hsp90 impairment in S . cerevisiae has been associated not only with increased between-strain variation for some traits but also with suppression of variation for other traits . For example , wild-type HSP90 function is necessary for some drug-resistance mutations to have their effects [52] . In a larger survey [23] , 102 genetically divergent yeast strains were analyzed for their ability to grow in a variety of conditions , with wild-type and reduced levels of HSP90 . A QTL analysis of HSP90-dependent traits showed that 44 HSP90-dependent growth QTLs were present at wild-type HSP90 levels and 63 HSP90-dependent growth QTLs were present at reduced HSP90 levels [23] . These findings have led to the description of Hsp90 as both a “capacitor” of phenotypic variation ( suppressing the effects of genetic variants unless impaired ) and a “potentiator” of phenotypic variation ( permitting the effects of genetic variants unless impaired ) [12] , [23] , [53] . Our work suggests that what is important about highly pleiotropic factors such as HSP90 and HTZ1 is not that they reduce robustness to mutations when impaired ( which appears not to be true at least for HTZ1 ) but that they interact epistatically with mutations . That is , the effects of perturbing such a factor will be context- and phenotype-dependent , as will the factor's apparent role as capacitor or potentiator . The present study adds to growing empirical support for the notion that pleiotropy and epistasis are widespread [54] . Understanding the evolutionary roles of HSP90 , HTZ1 and other factors with large potential effects on phenotypic variation will require both more experimental analysis and better theoretical models of complex traits [54] . In particular , the evolutionary role of cryptic genetic variation remains poorly understood . Although cryptic genetic variation has historically been viewed as a product of mutational robustness , we lend empirical support to the argument [8] that mutational robustness is a side question . We suggest that there should be more focus on the cryptic variation itself and the mechanisms that reveal it , rather than on the putative cause of its existence .
Diploid yeast MA lines were provided by David Hall [36] , [37] . In brief , the lines originated from a haploid strain ( a spore from a DBY4974/DBY4975 diploid ) with genotype ade2-101 , lys2-801 , his3-Δ200 , leu2-3 . 112 , ura3-52 , Gal+ , ho [36] . This strain was made diploid using an HO-expressing plasmid , creating a diploid line homozygous at each locus except the mating-type locus , that served as the ancestor for all the MA lines [36] . Yeast with an ade2 mutation build up a red metabolite during respiration , which acts as a visual marker that cells are not “petite” or respiration-deficient [55] . Cells lacking the red pigment were not passaged during the MA experiment to avoid accumulating mutations affecting mitochondrial function [36] . To produce the haploid MA lines for our study , we sporulated the diploid MA lines using a standard protocol [56] . For each MA line , a single non-petite spore of mating type a was chosen at random to be the representative HTZ1+ haploid of the line for the remainder of the experiment . The HTZ1 coding sequence was completely eliminated from representative haploid lines by homologous recombination of a linear fragment containing 471 base pairs of homology to the sequence immediately upstream of the HTZ1 coding sequence , 477 base pairs of homology to the sequence immediately downstream of the HTZ1 coding sequence , and , in between these , the URA3 selectable marker , using standard techniques [57] . PCR analysis of transformants capable of growth on medium lacking uracil confirmed the correct incorporation of the linear fragment and the complete absence of the HTZ1 coding sequence . A single confirmed transformant from each line was chosen as the representative HTZ1− line for morphological analysis . Ultimately , 79 line pairs were used for our experiments ( listed in Table S5 ) . Although the MA lines have the ade2-101 mutation , which causes colonies to be red , we noted the appearance of white or sectored colonies while propagating some of the lines . The loss of red pigmentation can indicate the presence of a petite mutation , a mutation that restores ADE2 function , or an epigenetic mechanism affecting the adenine pathway . The frequency of white-to-red switches and red-to-white switches , as well as sizes of the sectors or colonies , suggested an epigenetic factor was responsible . We hypothesize that the epigenetic factor is [PSI+] , a prion form of the translation-termination factor SUP35 , because [PSI+] causes increased read-through of stop codons [58] and ade2-101 is a premature stop-codon mutation [59] . Further analysis of this hypothesis will be presented elsewhere . For the purposes of the experiments presented here , we performed analyses two ways: by ignoring the sectoring behavior and by restricting our attention to only the 43 lines that stably maintained red color . These analyses yielded substantially similar conclusions so for simplicity we report the analysis for the full set of lines throughout the paper , while noting in some important places the results for the restricted analysis as well . The identities of the 43 stable-red lines are indicated in Table S5 . Cell morphology was measured in many cells from each pair of MA lines ( HTZ1+ and HTZ1− ) using a microscopy-based phenotyping assay [47] , adapted for a 96-well plate format rather than for individual glass slides . The original assay used three fluorescent signals to measure cell morphology . However , only two , ConcanavalinA-FITC , which stains the cell surface , and DAPI , which stains the nucleus , were used in this study . A previous study indicated that the third signal , rhodamine-phalloidin , did not add significant information to the analysis of morphological variance [41] , and was consequently excluded from this analysis . Cells were grown to mid-log phase in 96-well culture plates . Cells were then fixed with 4% para-formaldehyde , stained with 250 ug/mL ConcanavalinA-FITC , and plated into a 96-well glass-bottom microscope plate in mounting medium containing an anti-fade agent ( Vectashield or a 5 µM p-phenylenediamine/glycerol solution ) and 70 ng/mL DAPI . Plates were imaged using an inverted fluorescent compound microscope ( Nikon TE-2000 ) . Seventy-five random non-overlapping images were acquired per well with a Qimaging FireWire camera and NIS Elements software . Because CalMorph processes jpeg image files of particular pixel dimensions , one raw 16-bit 1392×1040 tiff image was split in to two 696×520 , 8-bit jpeg images . A minor background correction using the imadjust function in matlab was done to avoid discrepancies in staining quality . This adjustment did not significantly alter raw data values ( data not shown ) . Images were then analyzed using CalMorph software , developed specifically for this assay [47] . Raw cell measurements from CalMorph were analyzed using the R programming environment . CalMorph splits cells into three types: cells with no bud , a small bud , and a large bud . Each type has a unique set of morphological phenotypes and is therefore analyzed separately throughout this study . To allow for comparisons between CalMorph phenotypes of cells of the same type , raw values were transformed using the Box-Cox method [60] from the car package in R , and standardized to have a mean of zero and a standard deviation of one . The Box-Cox transformation method uses maximum likelihood to determine which among a family of power transformations produces a phenotypic distribution that approximates normality best [60] . Previous studies using CalMorph data have used this transformation [47] , [61]–[63] . If a phenotype did not show an approximately normal distribution on a qq-plot after this procedure , it was excluded ( the phenotypes used in the analysis are shown in Table S6 ) . Only cells without missing values for each remaining phenotype were included in the final analysis . The phenotypes measured by CalMorph are not completely independent [41] , so principal component analysis ( PCA ) was used to eliminate redundant measures . PCA was performed on real and randomly permuted data ( where raw values for a given trait were reassigned to a randomly chosen cell ) . Principal components capturing greater variance than the random expectation were used in downstream analyses . Performing PCA separately on HTZ1+ or HTZ1− cells yielded principal components with highly correlated loadings . To confirm that our choice of PCA for eliminating redundancy did not unexpectedly bias results , we performed alternative analyses using partitioning around medoids ( PAM ) , a variant of k-means clustering , as well . PAM clusters phenotypes by similarity and designates a “medoid , ” the phenotype that is most representative of the other phenotypes in the cluster . PAM was used previously to eliminate redundancies in CalMorph-generated data [41] . Normalized and standardized data were clustered using the pam function of the cluster package in R . For each cell type , the number of significant principal components was used to determine the number of clusters used in the final analysis . Variance-component estimates from linear models were obtained using Markov chain Monte Carlo sampling , with the R package MCMCglmm [64] . Highest posterior density interval estimates were obtained from the MCMC samples ( or functions of these samples ) using the HPDinterval function in the coda package [64] . The linear model specified two within-strain variances , one each for HTZ1+ and HTZ1− , using the “idh” variance structure for the residual variance . Likewise , it specified two between-strain variances , and also a genetic covariance , using the “us” variance structure . Inverse-Wishart priors were used . Parameter-expanded priors , which are suggested to have better properties when variance components are close to 0 , were also tried , with negligible difference in results . Markov chains were run with a burn-in period of 6000 , and samples were stored at intervals of 15 iterations for 30000 total iterations . Chain lengths were kept relatively short because of the large number of models that were run , so it was important to verify that the chains were well mixed . We did so by examining the autocorrelation in parameter estimates between successive stored samples , which was very close to 0 after the burn-in period . We also ran longer chains for a small subset of models , and found negligible effects on the parameter estimates . To test for an interaction between line and genotype , linear mixed models were fitted using the lmer function from the lme4 package in R . Genotype ( HTZ1+ or HTZ1− ) was modeled as a fixed effect , whereas MA line and a line-by-genotype interaction term were modeled as random effects . Note that lmer assumes a single within-line variance and a single between-line variance . The genotype-by-line interaction variance was partitioned into components that correspond to line crossing and line spreading by this formula: ( 1 ) where sdHTZ1+ and sdHTZ1− are the HTZ1+ and HTZ1− between-line standard deviations , respectively . This formula is identical to that given for Vg×l previously [30] , except rHTZ1+ , HTZ1− , a term representing the correlation of line means , is replaced by [65] . The crossing of line means is represented by the first bracketed term in Equation 1 , and the spreading of line means is represented by the second bracketed term in Equation 1 . See Text S1 for estimation of the crossing and spreading terms when models were fit by restricted maximum likelihood . Estimates of mutational variance were made using the equation [66]:VL is the between-line variance , for one or the other HTZ1 genotype , estimated by MCMC , discussed above . t , the number of generations , was estimated as 2062 [37] . Note that this VM is the mutational variance of the haploid lines in which we actually measured phenotypes , not of the diploid lines from which they derive . At least 150 cells were measured of each cell type from each line . This number was chosen based on a sub-sampling analysis , where an increasing number of cells was randomly drawn ( without replacement ) and used to calculate a mean and standard deviation value . Each of these estimates tended to converge when sampling more than 150 cells , suggesting this number is sufficient to adequately estimate the line means and within-line variances for a given morphological trait ( see Figure S7 for an example trait ) . | Natural populations typically harbor much genetic variation . Some of this variation is cryptic — it does not affect observed traits except if the organism is exposed to a major environmental or genetic perturbation . One often-proposed explanation for the revelation of cryptic genetic variation is that the perturbation has made the organisms less robust to mutations , thereby revealing effects of previously neutral mutations that natural selection had allowed to accumulate . Such effects would be dependent on genetic background , as particular cryptic variants will be present in some individuals and not others . We show that a perturbation of chromatin architecture caused by mutation of the budding-yeast gene HTZ1 , encoding histone H2A . Z , does alter phenotypes in a genetic background-dependent manner but does not reduce mutational robustness . Using a large set of yeast lines that accumulated mutations with minimal natural selection , we compared variation in cell morphology with and without HTZ1 . The effect of eliminating HTZ1 was highly line dependent , suggesting that HTZ1 interacts extensively with genetic variation in the lines . However , HTZ1+ lines span a range of phenotypes similar to that of the corresponding HTZ1− lines . Our results therefore call into question prior studies linking revelation of cryptic genetic variation with reduced mutational robustness . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"mutation",
"model",
"organisms",
"molecular",
"cell",
"biology",
"genetic",
"polymorphism",
"chromosome",
"biology",
"population",
"genetics",
"yeast",
"and",
"fungal",
"models",
"biology",
"evolutionary",
"theory",
"evolutionary",
"biology",
"saccharomyces",
"cerevisiae",
"evolutionary",
"genetics",
"chromatin"
] | 2013 | Histone Variant HTZ1 Shows Extensive Epistasis with, but Does Not Increase Robustness to, New Mutations |
In bistable vision , subjective perception wavers between two interpretations of a constant ambiguous stimulus . This dissociation between conscious perception and sensory stimulation has motivated various empirical studies on the neural correlates of bistable perception , but the neurocomputational mechanism behind endogenous perceptual transitions has remained elusive . Here , we recurred to a generic Bayesian framework of predictive coding and devised a model that casts endogenous perceptual transitions as a consequence of prediction errors emerging from residual evidence for the suppressed percept . Data simulations revealed close similarities between the model’s predictions and key temporal characteristics of perceptual bistability , indicating that the model was able to reproduce bistable perception . Fitting the predictive coding model to behavioural data from an fMRI-experiment on bistable perception , we found a correlation across participants between the model parameter encoding perceptual stabilization and the behaviourally measured frequency of perceptual transitions , corroborating that the model successfully accounted for participants’ perception . Formal model comparison with established models of bistable perception based on mutual inhibition and adaptation , noise or a combination of adaptation and noise was used for the validation of the predictive coding model against the established models . Most importantly , model-based analyses of the fMRI data revealed that prediction error time-courses derived from the predictive coding model correlated with neural signal time-courses in bilateral inferior frontal gyri and anterior insulae . Voxel-wise model selection indicated a superiority of the predictive coding model over conventional analysis approaches in explaining neural activity in these frontal areas , suggesting that frontal cortex encodes prediction errors that mediate endogenous perceptual transitions in bistable perception . Taken together , our current work provides a theoretical framework that allows for the analysis of behavioural and neural data using a predictive coding perspective on bistable perception . In this , our approach posits a crucial role of prediction error signalling for the resolution of perceptual ambiguities .
During bistable perception , observers experience fluctuations between two mutually exclusive interpretations of a constant ambiguous input . Remarkably , percepts evoked by ambiguous stimuli usually closely resemble the experience of unambiguous objects and thus illustrate the constructive nature of perception . However , the mechanisms driving transitions in bistable perception remain poorly understood . Previous neuroimaging work [4 , 5 , 6 , 7 , 8 , 9 , 10] has sought to distill the neural processes underlying bistable perception by recurring to a ‘replay’ condition , in which physical stimulus changes mimic the perceptual alternations induced by ambiguous stimuli . This approach revealed a right-lateralized assembly of fronto-parietal areas whose activity is specifically enhanced during endogenously evoked transitions ( ambiguity ) as compared to exogenously evoked transitions ( replay ) [4 , 5 , 7 , 9] . However , the functional role of fronto-parietal areas in bistable perception is a matter of ongoing debate . According to one view , transitions in bistable vision are primarily a result of adaptation and inhibition within visual cortex , while switch-related activations in fronto-parietal areas reflect a mere ‘feedforward’ consequence of neural events at sensory processing levels [6 , 10] . Another view proposes that fronto-parietal areas may be involved in stabilizing and destabilizing perception , thus causally contributing to perceptual switching via ‘feedback’ mechanisms [4 , 5 , 11 , 7] . Here , we sought to resolve this debate by using model-based fMRI to empirically test a theoretical model that has the potential to integrate these two seemingly contradictory views of perceptual bistability . From a theoretical perspective , endogenous transitions might be explained by framing perception as an inferential process generating and testing hypotheses about the most likely causes of sensory stimulation [12 , 13 , 14] . Such processes can be elegantly implemented by hierarchical predictive coding [15 , 16 , 17] . Here , ‘predictions’ encoded at higher levels are compared against ‘sensory input’ represented at lower levels , while a mismatch between the two elicits a prediction error , updating higher-level predictions [15] . Such belief-updating schemes can be translated onto Bayes’ rule , where prior distributions ( ‘predictions’ ) are combined with likelihood distributions ( ’sensory input’ ) into posterior distributions in a sequential manner [16 , 18] . Here , we tested whether this framework provides a mechanistic explanation for perceptual transitions and related neural activity during bistable perception . We devised a computational model that formalizes perceptual decisions ( i . e . , decisions that define the content of conscious perception , as indicated by participants’ response ) to be performed on the basis of posterior probability distributions . This model is a modification of an approach introduced by [19] , who propose that perceptual time-courses during bistable perception result from samples drawn subsequently from a posterior distribution . The authors implement a memory decay favoring recent over older samples as well as stationary prior capturing the effect of context on bistable perception . Our model , in turn , posits that the shape of the posterior distribution changes dynamically over time in response to prediction errors emerging from the currently suppressed interpretation of the ambiguous input . Importantly , this model has the potential to integrate feedforward and feedback mechanisms in bistable perception: The prediction errors arising from sensory processing levels may be propagated up to higher-level brain areas in a feedforward fashion . The registration of prediction errors in higher-level brain areas leads to an updating of predictions that may in turn drive perceptual switching through a feedback mechanism . To test this hypothesis , we began with data simulations to establish that our model’s predictions match the key characteristics of perceptual bistability . We proceeded by fitting our model to behavioural data from a fMRI experiment on bistable perception [7] . In this experiment , participants viewed a Lissajous figure [42] rotating either clockwise ( as viewed from above , i . e . movement of the front surface to the left ) or counter-clockwise ( vice versa ) and indicated their current perception via button-presses . Participants were presented with alternating blocks of ambiguous and disambiguated Lissajous figures: In the ambiguous condition , we presented bistable Lissajous figures which elicited spontaneous ( endogenous ) alternations in perception . In the disambiguated ( ’replay’ ) condition , we mimicked the endogenous perceptual time-course by introducing exogenous perceptual switches . Ambiguous and disambiguated stimuli were constructed by presenting two Lissajous figures separately to the two eyes: In the ambiguous condition , both eyes received identical stimulation . In the replay condition , the two Lissajous figures were slightly phase-shifted against each other , biasing perception in the direction of the phase shift . Having inverted our predictive coding approach based on behavioural data from this experiment , we investigated whether our model accurately explains individual perceptual time-courses during ambiguous and replay stimulation . In a supplementary analysis ( see S2 Text ) , we furthermore compared our model to three established models of bistable perception: Firstly , we tested an oscillator model [1] , which is based on mutual inhibition between to competing neural populations coding for the alternative perceptual outcomes during bistable perception . Here , the currently dominant population suppresses activity in the alternative population . However , due to adaptation in the dominant population , this relation reverses over time , leading to regular oscillations in perception . Secondly , we constructed a noise-driven attractor model of bistable perception [2] . In this framework , internal and external sources of noise trigger transitions between two stable states in an attractor network , representing the two perceptual interpretations associated with a bistable stimulus . Thirdly , we tested an intermediate model [3] , which contains both adaptive processes and noise . We validated our approach against these models by the use of Bayesian Model Comparison [20] . We then conducted a model-based fMRI-analysis [21] based on the predictive coding model to test whether prediction errors account for transition-related neural activity during bistability . Additionally , we compared the model-based fMRI analysis with conventional fMRI analyses using a Posterior Probability Map ( PPM ) approach [22] .
Our Bayesian modelling approach draws on the view that perception is an inferential process in which perceptual decisions are based on posterior distributions [13] . According to Bayes’ rule , the posterior combines information in the current sensory data ( likelihood ) with information from previous visual experience ( prior ) in a probabilistically optimal manner . Crucially , this posterior at a given moment becomes a prior for the current perceptual decision , which entails a prediction error signal that influences on the prior at the next moment . Hence , the posterior not only provides the basis for current perception , but also shapes future perception . In line with previous theorizations [12] , we reasoned that the ambiguous likelihood provides equally strong sensory evidence for two different percepts . We further hypothesized that the current percept establishes an implicit prior belief about similar percepts in the future , thereby contributing to stability of visual perception . The application of Bayes’ rules combines the likelihood for ambiguous stimuli with the stability prior into a posterior that represents stronger evidence for the dominant percept , but still contains residual evidence for the suppressed percept . While the stronger evidence for the dominant percept will again favor this percept for the upcoming perceptual decision , the residual evidence for the suppressed percept is equivalent to a prediction error that leads to an update of the stability prior . Over time , the stability prior is weakened and the posterior shifts towards the suppressed percept , paralleled by an escalating prediction error . When the residual evidence for the suppressed percept equals the evidence for the dominant percept , the prediction error reaches a maximum and a perceptual transition is most likely to occur . Once such a transition has occurred , the process starts over again , minimizing the current prediction error . Please note that our approach was influenced by the work of [19] , who argue that bistable perception is a product of Bayesian decision making in ambiguous sensory environments . They study the effects of viewpoint context on perception of the Necker Cube and propose that bistable perception arises from sampling a bimodal posterior distribution . Here , the sample with the highest ‚weight’ determines the content of conscious perception . Key elements of their model are ( 1 ) , a stationary prior , whose precision reflects interindividual differences in the effects of viewpoint context on perception of the Necker Cube and ( 2 ) , a memory decay that discounts the weight associated with a sample drawn from the posterior distribution by its age and influences on the length of individual phase durations . In contrast to [19] , our model does not assume a specific memory decay process , but controls the length of phase durations by means of the dynamically updated stability prior . In analogy to the stationary viewpoint prior in [19] , our model captures the influence of additional sensory evidence on perceptual decisions using a ‚stereodisparity’ distribution , whose precision determines the effectiveness of disambiguation . Please refer to to the mathematical appendix ( see S1 Text ) for a complete description , to Fig 1 for a step-by-step illustration of our approach and to Table 1 for a summary of model parameters and quantities . For computational expediency , we assume Gaussian probability distributions defined by mean and variance ( or inverse precision ) . To test whether our model is able to reproduce the temporal dynamics of bistable perception , we used it to generate perceptual time-courses from some ambiguous visual input such as the Lissajous figure . We assumed a sampling rate of 0 . 33 Hz , which was chosen to be close to the average overlap frequency in the behavioural experiment ( see below ) , and simulated for a total of 6 * 105 seconds . To model the ambiguous visual input , the impact of the stereodisparity weight was suppressed by setting μstereo = 0 . 5 and πstereo = 0 . We further assumed fixed values for the precision πinit , which was set to 3 . 5 to match the posterior parameter value from our behavioural modelling ( see Modelling analysis of behavioural data ) . To examine whether our prediction error model might account for bistable perception and associated neural activity in human observers , we used data from an fMRI experiment applying the Lissajous figure . Results from conventional analyses but not from behavioural modelling or model-based fMRI ( see below ) have been reported previously [7] . We recorded BOLD images by T2-weighted gradient-echo echo-planar imaging ( FOV 192 , 33 slices , TR 2000 ms , TE 30 ms , flip angle 78° , voxel size 3 x 3 x 3 mm , interslice gap 10 percent ) on a 3T MRI scanner ( Tim Trio , Siemens ) . The number of volumes amounted to 402 ( 0 . 15 Hz and 0 . 2 Hz ) or 415 ( 0 . 12 Hz ) volumes , respectively . We used a T1-weighted MPRAGE sequence ( FOV 256 , 160 slices , TR 1900 ms , TE 2 . 52 ms , flip angle 9° , voxel size 1 x 1 x 1 mm ) to acquire anatomical images . Image preprocessing ( standard realignment , coregistration , normalization to MNI stereotactic space using unified segmentation , spatial smoothing with 8 mm full-width at half-maximum isotropic Gaussian kernel ) was carried out with SPM8 ( http://www . fil . ion . ucl . ac . uk/spm/software/spm8 ) . To probe whether our predictive coding model might explain perceptual time-courses during bistable perception in human observers , we fitted our model to the behavioural data collected during the fMRI experiment . We optimized our model for the prediction of perceptual outcomes , i . e . on the perception of clockwise or counter-clockwise rotation as indicated by the individual participants . To this end , participants’ responses were aligned to the overlapping stimulus configurations of the Lissajous figure ( ’overlaps’ ) . This refers to timepoints during presentation when fore- and background of the stimulus cannot be discerned ( i . e . depth-symmetry ) [25 , 26] . Depending on the rotational speed of the stimulus and the associated ‘overlap’ frequency , sampling rates varied across participants between 0 . 24 Hz and 0 . 40 Hz ( see above ) . We first constructed models incorporating all combinations of the likelihood weight ‘stereodisparity’ and prior ‘perceptual stability’ , yielding a total of 4 behavioural models ( behavioural model 1: no stereodisparity , no perceptual stability; behavioural model 2: no stereodisparity , perceptual stability; behavioural model 3: stereodisparity , no perceptual stability; behavioural model 4: stereodisparity , perceptual stability ) to be compared . The respective precision of these distributions was optimized for the prediction of perceptual outcomes based on posterior distributions using a free energy minimization approach [27] . This method minimises the surprise about the individual participants’ data , thereby maximising log-model evidence . For model inversion , precisions were modelled as log-normal distributions . πinit and πstereo were either estimated as free parameters ( πinit: prior mean of log ( 3 ) and prior variance of 5; πstereo: prior mean of log ( 5 ) and prior variance of 5 ) or fixed to zero ( thereby effectively removing the distribution from the model ) . We kept ζ , which represents the inverse decision temperature in the response model represented by Equation 11 ( see Mathematical Appendix , S1 Text ) , fixed to 1 , since we did not have a particular a-priori hypothesis regarding this parameter . Please note that when choosing ζ as a free parameter ( prior mean of log ( 1 ) , prior variance of 1 ) , results remained almost identical . Parameters were optimised using quasi-Newton Broyden-Fletcher-Goldfarb-Shanno minimisation as implemented in the HGF4 . 0 toolbox ( TAPAS toolbox , http://www . translationalneuromodeling . org/hgf-toolbox-v3-0/ ) . After identifying the optimal model using Random Effects Bayesian model selection [20] , as implemented in SPM12 ( http://www . fil . ion . ucl . ac . uk/spm/software/spm12/ ) , we analyzed its posterior parameters with regard to the respective precision of the prior distributions using classical frequentist statistics . Since parameters were estimated in log-space , we report the geometric mean ( i . e . the arithmetic mean in log-space ) . In a supplementary analysis ( see S2 Text ) , we further compared the explanatory power of our predictive coding model with established models of bistable perception . To this end , we implemented models of bistable perception belonging to three different classes ( [1] as an example of so-called oscillator models based on mutual inhibition and self-adaptation between two competing neuronal populations , [2] as a representative of noise-driven attractor models and [3] as an intermediate model ) , which can be fitted to experimental data . We conducted a Random Effects Bayesian Model Comparison [20] between the established models and our predictive coding model in order to probe the validity of our approach . To examine the neural correlates of prediction error time-courses from our model , we conducted model-based fMRI analyses [21] in SPM12 . We adopted a general-linear-model- ( GLM- ) approach , constructing a total of three models: The design matrix of the first GLM ( the ‘PE model’ ) represented prediction error trajectories timepoint by timepoint . To this end , the regressor ‘transitions’ and the regressor ‘overlaps’ were modelled as stick functions . Furthermore , we extracted the individual ‘Prediction Error’ time-course for every participant and run and used its absolute value as a parametric modulator for the regressor ‘overlaps’ . In order to enable a comparison to the conventional approach of analysing fMRI data on bistable perception , we constructed a second GLM that dissociated between transition-related activity specific to bistable perception and the replay condition [4 , 5 , 6 , 7 , 9 , 10] . In addition to the regressor ‘overlaps’ , the design matrix of this ‘Conventional model’ contained ambiguous and replay transitions represented by stick functions . To further investigate the specificity of the prediction error trajectories and their neural correlates , we constructed a third GLM that took into account the presence of ambiguity inherent to the bistable condition . The design matrix contained the regressors ‘transitions’ as well the regressor ‘overlaps’ modelled as stick functions . Here , however , we used a box-car function being 1 for ambiguous and 0 for ‘replay’ blocks as a parametric modulator of the regressor ‘overlaps’ . Hence , this ‘Block model’ only differs from the ‘PE model’ in the values of the parametric modulator and serves to investigate whether correlations with the prediction error ( which we assumed to be higher in the bistable condition ) merely correspond to ambiguity per se . All further analyses were conducted for all models in parallel: regressors were convolved with the canonical hemodynamic response function as implemented in SPM12 . We added six rigid-body realignment parameters as nuisance covariates and applied high-pass filtering at 1/128 Hz . In a first step , we tested which of the three models accounted best for the measured BOLD signal . Therefore , we conducted a voxel-wise model comparison of the ‘PE model’ with the ‘Conventional model’ and the ‘Block model’ , as described in [22] . In brief , this technique uses Bayesian statistics for the construction of ‘Posterior Probability Maps’ ( PPMs ) and ‘Exceedance Probability Maps’ ( EPMs ) , which enable the calculation of log-evidence maps for each participant and model separately . On a second level , these log-evidence maps can be combined , thereby enabling voxel-wise model inference at the group level . Using the ‘Bayesian 1st level’ procedure for model estimation , we constructed log-evidence maps for every participant and model separately and compared the ‘PE model’ to the other models on a group level using exceedance probabilities computed with Random Effects analyses . In a second step , we aimed to identify regions in which prediction error trajectories ( ‘PE model’ ) , ambiguity per se ( ‘Block model’ ) or ambiguous as compared to replay transitions ( ‘Conventional model’ ) were correlated with the recorded BOLD signals . To this end , we estimated single-participant statistical parametric maps , then created contrast images for the parametric regressor against baseline ( ‘PE model’ and ‘Block model’ ) or ambiguous against replay transitions ( ‘Conventional model’ ) . These were entered into voxel-wise one-sample t-tests at the group level . Voxels were considered statistically significant if they survived family-wise-error ( FWE ) correction for all voxels in the brain at p < 0 . 05 . Anatomic labeling of cluster peaks was performed using the SPM Anatomy Toolbox Version 1 . 7b [28] . In order to further visualize our results , we extracted eigenvariate time-courses ( without adjustment for effects of interest ) from spherical ROIs ( radius: 3 mm ) around peak voxels from clusters for the contrast ‘Prediction Error vs baseline’ ( thresholded at p < 0 . 05 ) corresponding to left IFG ( peak voxel: [-54 2 22] ) , right IFG ( peak voxel: [51 8 10] ) , left insula ( peak voxel: [-30 20 10] ) and right insula ( peak voxel: [33 23 7] ) . These time-courses were extracted for ambiguous stimulation only . The time-courses for all perceptual phases were aligned with the respect to the end of the perceptual phase and averaged within and across observers .
To test whether our predictive coding model was able to reproduce perceptual switching in bistable perception , we used the model to generate perceptual time-courses during simulated viewing of an ambiguous stimulus . The distribution of perceptual phase durations followed a sharp rise and slow fall ( Fig 2 ) typical for bistable stimuli [29 , 30] . Mean and median simulated phase durations were 10 . 40 and 10 . 00 seconds , closely matching the results from behavioural analysis ( see Modelling analysis of behavioural data ) . As illustrated by exemplary time-courses of model parameters , the prediction error PE ( Fig 2A ) increases over time while one percept is dominant and is reduced once a new percept is adopted , reflecting the accumulation of evidence from the suppressed percept . The variance ( 1/πstability ) of the prior ‘perceptual stability’ ( Fig 2C ) increases over a perceptual phase as a function of the prediction error . In line with the hypothesized role of prediction errors in driving perceptual transitions , the prediction error PE and , hence , the variance 1/πstability are maximal when the posterior P ( θ > 0 . 5 ) relaxes to 0 . 5 ( Fig 2B ) , thereby increasing the probability of a new perceptual transition . To investigate whether our model is able to explain the dynamics of perceptual bistability in human observers , we fitted our model to behavioural data collected from 20 healthy participants during an fMRI experiment , in which participants viewed ambiguous and unambiguous ( replay ) versions of a rotating Lissajous stimulus . As reported previously , perceptual transitions occurred on average every 9 . 3 seconds in the ambiguity condition and neither block-by-block ratings nor debriefing after the experiment revealed differences in perceived appearance between the ambiguity and the replay condition [7] . We first performed a model comparison with other models that lacked the key conceptual elements of our model . By eliminating either the likelihood weight ‘stereodisparity’ or the prior ‘perceptual stability’ or both from the model , we constructed three additional models which we compared to our model using Random Effects Bayesian Model Selection . Our model ( i . e . behavioural model 4 ) was identified as a clear winning model with a protected exceedance probability of 99 . 96% , demonstrating that the incorporation of both the likelihood weight ‘stereodisparity’ and the prior ‘perceptual stability’ best explained participants’ perception . From this model , we extracted the parameters for πinit and πstereo and averaged across runs and participants ( Fig 3A ) . We predicted average prediction errors to be lower in replay as compared to the ambiguous condition , since the presented stereodisparity reduces the ambiguity left in the experimental display , and hence , the residual evidence for suppressed percept . Consistently , mean prediction errors were significantly higher in the ambiguous condition than in the replay condition ( 0 . 36 +/- 0 . 03 vs . 0 . 26 +/- 0 . 02 , mean +/- s . e . m . , p < 10−6 , t19 = 7 . 06 , two-sample t-test , Fig 3B ) , providing support for a correct implementation of our predictive coding model . Given that πinit describes the strength of the initial stabilization after a switch in perception , we expected this parameter to be related to the frequency of perceptual transitions . In line with this , model parameter estimates πinit were negatively correlated with perceptual transition frequencies across participants ( ρ = −0 . 88 , p < 10−7 , Pearson correlation , Fig 3C ) , providing a sanity check for model fit . Notably , this correlation was also significant when we correlated model parameter estimates for πinit averaged over run 1 and 2 with perceptual transition frequencies from run 3 ( ρ = −0 . 76 , p = 10−4 , Fig 3D ) , corroborating that our model successfully accounted for observers’ perception evoked by an ambiguous stimulus . We furthermore validated our approach by comparing our predictive coding model to established models of bistable perception from three different classes: oscillator models [1] , attractor models [2] and intermediate models [3] ( see Supplementary Methods in S2 Text ) . Data simulations indicated that all established models , similar to our predictive coding model , were able to produce spontaneous transitions in perception and a typical gamma-like distribution of perceptual phase durations ( see Supplementary Results and Fig . A-C in S2 Text ) . Fitting of the behavioural data further showed that both the oscillator and the intermediate , similar to our predictive coding model , adequately accounted for the observers’ perceptual decisions during bistable perception ( see Supplementary Results and Fig . D-I in S2 Text ) . In order to validate our approach , we conducted a Bayesian Model Comparison , which showed that our predictive coding model compared to these established models was best in explaining the behavioural data collected during this experiment ( see Fig . J in S2 Text ) . Please note that we did not carry out these analysis to demonstrate a superiority of our approach over these earlier models , which were initially conceived mainly for binocular rivalry and not for the prediction of behavioural responses during presentation of the Lissajous figure ( a specific type of structure-from-motion stimulus ) . On the contrary , we aimed at probing the validity of our approach and tried to ascertain that the predictive coding approach was at least equivalent to existing models of bistable perception . One central aim of our study was to gain mechanistic insight into the neural processes underlying transition-related activity during bistable perception . We therefore performed both a model-based fMRI analysis suitable to identify the neural correlates of modelled prediction errors ( ‘PE model’ ) , and , for the purpose of comparison , a conventional analysis ( ‘Conventional model’ ) dissociating between ambiguous and replay transitions as well as a ‘Block model’ accounting for effects of ambiguity per se . To test the validity of these models , we first searched for voxels that were more active during visual stimulation as compared to baseline ( ‘overlaps vs . baseline’ ) . For the ‘PE model’ , this analysis revealed significant clusters ( p < 0 . 05 , FWE-corrected across the whole brain ) bilaterally in middle occipital cortex ( right: [39 -9 1] , T = 10 . 21; left: [-30 -94 1] , T = 13 . 30 ) , in V5/hMT+ ( right: [45 -70 1] , T = 11 . 61; left: [-45 -73 4] , T = 14 . 09 ) , as well as in superior parietal cortex ( right: [27 -49 58] , T = 10 . 26; left: [-36 -46 -61] , T = 8 . 62 ) . The same analyses for the ‘Conventional model’ and the ‘Block model’ yielded virtually identical results ( see Tables 2–4 ) , confirming the comparability between all three models . We then investigated which voxels were more active during perceptual transitions as compared to baseline ( ‘transitions vs . baseline’ , Fig 4A ) : For the ‘PE model’ , we found significant activations of motor-related areas in left precentral gyrus ( [-36 -16 67] , T = 12 . 23 ) extending to left postcentral gyrus ( [-63 -19 25] , T = 8 . 62 ) as well as significant clusters in regions previously associated with transition-related activity during bistable perception: right inferior frontal gyrus ( [54 17 13] , T = 7 . 96 ) , right inferior parietal lobulus ( 54 -37 52 , T = 9 . 32 ) and right middle frontal gyrus ( [39 44 31] , T = 7 . 57 ) . Additional clusters were located in bilateral posterior-medial frontal gyrus ( right: [6 2 67] , T = 9 . 50; left: [-6 2 55] , T = 12 . 63 ) . Again , repeating this analysis for the ‘Block model’ and the ‘Conventional model’ yielded qualitatively very similar results as in the ‘PE model’ ( see Tables 5–7 ) , thereby providing further evidence for the validity and comparability of all three models . To formally test whether the modelled prediction error explains the BOLD signal better than the conventional comparison of ambiguous with replay perceptual switches ( ‘Conventional model’ ) , or the mere ambiguity of the visual display ( the ‘Block model’ ) , we performed a PPM analysis [22] to compute voxel-wise exceedance probability maps for the ‘PE model’ against the ‘Conventional model’ and the ‘Block model’ ( Fig 4C ) . We restricted this analysis to areas of the fronto-parietal cortex , which be delineated by intersecting the statistical-parametric maps for ‘transitions vs . baseline’ thresholded at p < 0 . 05 FWE for all three models considered . Remarkably , when applying a conservative threshold of an exceedance probability of γ = 99% and a cluster size of n > 10 voxels , we found clusters in right insula ( [39 26 -2] ) and right inferior frontal gyrus ( [51 14 1] ) to show strong evidence for the ‘PE model’ as compared to the ‘Block model’ and the ‘Conventional model’ . Additional clusters were located in right posterior medial frontal gyrus ( [6 5 49] ) as well as left precentral gyrus ( [-36 -16 52] ) . Conversely , for the exceedance probability map of the ‘Conventional model’ compared against ‘Block model’ and ‘PE model’ , no voxels survived the conservative threshold used in the main experiment . For the exceedance probability map of the ‘Block model’ compared against the ‘Conventional model’ and ‘PE model’ , we found clusters in bilateral inferior parietal lobule at an exceedance probability of 99% and a cluster size > 10 . For our central analysis aimed at identifying the neural correlates of modelled prediction errors , we searched for voxels in which BOLD activity was related to the parametric modulator of the ‘PE model’ that encoded prediction error trajectories from our Bayesian model of bistable perception ( Fig 4B ) . We found significant clusters ( p < 0 . 05 , FWE-corrected across the whole brain ) in bilateral insulae ( right: [33 23 7] , T = 7 . 24; left: [-30 20 10] , T = 7 . 88 ) and bilateral inferior frontal gyri ( right: [51 8 10] , T = 6 . 89; left: [- 54 2 22] , T = 6 . 67 ) . These regions are located in close anatomical proximity to frontal regions previously suggested to mediate perceptual transitions in bistable perception [4 , 5 , 7] . In order to further visualize the correlation between modelled prediction error and BOLD activity , we extracted eigenvariate time-courses from right insula , left insula , right IFG as well as left IFG and averaged across perceptual phase durations and observers . As expected , these time-courses showed a gradual increase towards a transition in perception ( Fig 5 ) , nicely mirroring the build-up of prediction error during a perceptual phase .
In this work , we present a Bayesian predictive coding model for bistable perception , which rests on the basic assumption that prediction errors are elicited by the unexplained alternative interpretation of an ambiguous stimulus and represent the driving force behind perceptual transitions during bistable perception . We found that this model is able to reproduce key temporal characteristics of human bistable perception and that it explains observers’ behaviour during a bistable perception experiment . Our central finding shows that modelled prediction errors correlate with BOLD activity in bilateral insulae and bilateral inferior frontal gyrus . Remarkably , our PPM analysis revealed that modelled prediction errors best accounted for BOLD activity as compared to mere occurrence of endogenous perceptual transitions or ambiguity of the visual display in these frontal regions . Hence , our current results suggest that prediction errors might provide the mechanistic basis for perceptual switching in bistable perception and offer a novel interpretation of frontal activity in bilateral insulae as well as the right inferior frontal gyrus during bistable perception . The functional significance of enhanced frontal brain activity for transitions during bistability as compared to an unambiguous control condition is a matter of ongoing debate: Some authors proposed that non-sensory higher-level brain regions are actively implicated in resolving the perceptual conflict during bistable perception , thus mediating perceptual transitions [4 , 31 , 5 , 11 , 7] . Others have argued that perceptual conflicts are resolved primarily in sensory brain areas and that activity in frontal and parietal regions reflects the registration and/or report of perceptual transitions , rather than their cause [6 , 8 , 10] . For a detailed discussion of this debate , see “Brascamp , Sterzer , Blake and Knapen , Multistable perception and the role of frontoparietal cortex in perceptual inference , Annual Review of Psychology , in press . ” Here , we provide further evidence for an active implication of frontal regions in bistable perception by functionally relating these regions to a prediction error signal . Hence , our work is in line with hybrid models that suggest bistable perception to arise from an interplay between lower-level sensory and higher-level non-sensory areas [32 , 12 , 11] . In this context , it might be speculated that prediction errors are computed in frontal regions based on feedforward signals from visual and parietal cortex; and that these prediction errors , in turn , modulate activity in visual cortex via feedback signals . In addition to the prediction error , the stability prior represents an essential element of our predictive coding model of bistable perception , since its initial precision determines the frequency of perceptual transitions . The notion of such a stability prior is supported by experimental work on serial dependence in visual perception: In an orientation-judgement task , [33] showed that perceived orientation was biased by recently observed stimuli and reasoned that the visual system might use past experiences as predictors of present perceptual decisions , thereby incorporating representations of the continuity of the visual environment . Corroborating these results in a fMRI experiment , [34] found that orientation signals in early visual cortex were biased towards previous perceptual decisions . At this point , however , we can only speculate about the neural correlates of the stability prior from our model: In recent work on the role of parietal cortex in bistable perception , [35] and [9] have proposed a functional segregation of the superior parietal lobulus ( SPL ) , which they deduced from differential effects of grey matter volume on perceptual dominance durations and analyses of effective connectivity on the basis of fMRI . By interpreting their results in a Bayesian framework , the authors argued that posterior SPL might represent a prediction error , while the anterior SPL would entertain a perceptual prediction . A key advantage of our predictive coding model of bistable perception is that it allows us to treat ambiguous and replay stimulation within the same framework . By formalizing the disambiguating factor as a weight on a bimodal likelihood distribution , such models can be used to investigate perceptual transitions under varying degrees of ambiguity , thus dissolving the artificial dichotomy between the two conditions . Hence , such models provide a new perspective on how the brain might resolve perceptual conflicts despite the ambiguity inherent in every sensory signal and offer a generic tool for quantifying the contribution of different contextual factors on perceptual outcomes . The major strength of predictive coding models for bistable perception , however , lies in their ability to parsimoniously link different levels in the description of perceptual dynamics in ambiguous visual environments: On a computational level , prediction errors constitute the driving force behind perceptual transitions and are substantially reduced by additional sensory information ( such as stereodisparity ) during replay . On a neural level , casting frontal activity during rivalry in terms of prediction error signals nicely relates to increased transition-related activity [4 , 5 , 9] and connectivity [7] . On a theoretical level , viewing perceptual transitions as means of reducing prediction errors places bistable perception in the context of Bayesian theories of the brain [16 , 36 , 27 , 37] , and in particular the free-energy principle [13] . According to the latter , agents strive for a reduction of their model’s free energy , which translates onto a minimization of squared prediction errors in predictive coding schemes . When sensory information is constantly ambiguous , one possibility to reduce free energy is to update beliefs about the world , which ultimately corresponds to the adoption of a new percept . However , given that the Lissajous differs in some aspects from other types of bistable stimuli , one has to consider important limitations regarding the generalization of our findings: While being physically ambiguous for all angles of rotation , transitions almost exclusively occur at overlapping stimulus configurations , which is similar to the behaviour of some types of random dot kinematograms [26] or intermittent presentation of bistable stimuli [38] and accompanied by a reduced incidence of mixed percepts or incomplete transitions . Since these phenomena are present in many other forms of bistable perception and significantly affect frontoparietal activity during perceptual transitions [6] , our current imaging results can only be interpreted in relation to the specific stimulus used here . A similar limitation applies to the behavioural modelling presented in this manuscript: Previous work on computational modelling of bistable perception has focused on a variety of mechanisms at the heart of spontaneous perceptual transitions: Oscillator models have focused on mutual inhibition between two competing neuronal populations combined with slow adaptation of the currently dominant population [1] . [39] have studied the differential effects of short and long interruptions in intermittent bistable perception for binocular rivalry and structure-from-motion and presented a model based on adaptive processes , cross-inhibition and neural baseline levels . Importantly , this model also accounts for the possibility of voluntary control via attentional processes interacting with early processing stages . Alternative approaches view noise as the underlying cause of perceptual transitions [2] . Importantly , models belonging to this class have also taken account of the aforementioned mixed percepts and incomplete transitions during binocular rivalry [40] . Further models have related transitions in perception to a combination of adaptation and noise [3] . In this vein , [41] have argued for a neurodynamic mechanism at the bifurcation between adaptation- and noise-driven processes to be the basis for perceptual transitions during binocular rivalry . The majority of the models mentioned above has been developed for continuous presentation of binocular rivalry or ambiguous structure-from-motion , while [39] have also studied paradigms with intermittent presentation . As noted above , such stimuli differ significantly from the Lissajous figure used in our current study , which shares aspects with intermittent stimulation due to the existence of overlapping configurations facilitating transitions in perception . Hence , future theoretical and empirical work is needed to probe our modelling approach on paradigms such as binocular rivalry and ambiguous structure-from-motion for both continuous and intermittent presentation and to extend the predictive coding model in order to account for top-down attentional control as well as interactions at earlier processing stages . Taken together , our current work provides theoretical and empirical evidence across different levels for a driving role of prediction errors in bistable perception , thereby shedding new light on an ongoing debate about the neural mechanisms underlying bistable perception and , more generally , opening up a novel computational perspective on the mechanisms governing perceptual inference . | In bistable vision , perception spontaneously alternates between two different interpretations of a constant ambiguous stimulus . Here , we show that such spontaneous perceptual transitions can be parsimoniously described by a Bayesian predictive coding model . Using simulated , behavioural and fMRI data , we provide evidence that prediction errors stemming from the suppressed stimulus interpretation mediate perceptual transitions and correlate with neural activity in inferior frontal gyrus and insula . Our findings empirically corroborate theorizations on the relevance of prediction errors for spontaneous perceptual transitions and substantially contribute to a longstanding debate on the role of frontal activity in bistable vision . Therefore , our current work fundamentally advances our mechanistic understanding of perceptual inference in the human brain . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"and",
"health",
"sciences",
"diagnostic",
"radiology",
"functional",
"magnetic",
"resonance",
"imaging",
"brain",
"social",
"sciences",
"neuroscience",
"magnetic",
"resonance",
"imaging",
"perception",
"simulation",
"and",
"modeling",
"cognitive",
"psychology",
"brain",
"mapping",
"computational",
"neuroscience",
"vision",
"neuroimaging",
"coding",
"mechanisms",
"research",
"and",
"analysis",
"methods",
"imaging",
"techniques",
"behavior",
"psychology",
"radiology",
"and",
"imaging",
"parietal",
"lobe",
"diagnostic",
"medicine",
"anatomy",
"biology",
"and",
"life",
"sciences",
"sensory",
"perception",
"computational",
"biology",
"cognitive",
"science",
"cerebral",
"cortex"
] | 2017 | A predictive coding account of bistable perception - a model-based fMRI study |
DNA in eukaryotes is packaged into a chromatin complex , the most basic element of which is the nucleosome . The precise positioning of the nucleosome cores allows for selective access to the DNA , and the mechanisms that control this positioning are important pieces of the gene expression puzzle . We describe a large-scale nucleosome pattern that jointly characterizes the nucleosome core and the adjacent linkers and is predominantly characterized by long-range oscillations in the mono , di- and tri-nucleotide content of the DNA sequence , and we show that this pattern can be used to predict nucleosome positions in both Homo sapiens and Saccharomyces cerevisiae more accurately than previously published methods . Surprisingly , in both H . sapiens and S . cerevisiae , the most informative individual features are the mono-nucleotide patterns , although the inclusion of di- and tri-nucleotide features results in improved performance . Our approach combines a much longer pattern than has been previously used to predict nucleosome positioning from sequence—301 base pairs , centered at the position to be scored—with a novel discriminative classification approach that selectively weights the contributions from each of the input features . The resulting scores are relatively insensitive to local AT-content and can be used to accurately discriminate putative dyad positions from adjacent linker regions without requiring an additional dynamic programming step and without the attendant edge effects and assumptions about linker length modeling and overall nucleosome density . Our approach produces the best dyad-linker classification results published to date in H . sapiens , and outperforms two recently published models on a large set of S . cerevisiae nucleosome positions . Our results suggest that in both genomes , a comparable and relatively small fraction of nucleosomes are well-positioned and that these positions are predictable based on sequence alone . We believe that the bulk of the remaining nucleosomes follow a statistical positioning model .
DNA in eukaryotes is packaged with histone and other proteins into a chromatin complex . The most basic element of chromatin is the nucleosome , which consists of a core of eight histone proteins around which is wound approximately 147 bp of double-stranded DNA . The precise positioning of the nucleosome cores and the inter-nucleosomal linker regions allows for selective access to the DNA by the cellular machinery; understanding the mechanisms that control this positioning is therefore crucial to our understanding of gene regulation and expression . The recently published high-resolution maps of 20 histone methylations in H . sapiens CD4 T-cells [1] provided the first H . sapiens genome-wide experimental data from which nucleosome positions could be inferred . Barski et al . combined chromatin immunoprecipitation ( ChIP ) with direct high-throughput sequencing of the ChIP DNA samples in the new procedure known as ChIP-seq . To resolve the histone modification signals to individual nucleosomes , templates from purified CD4 T-cells were created by micrococcal nuclease ( MNase ) digestion of native chromatin , followed by a mononucleosome-length selection on a gel . The sequencing process resulted in roughly 185 million sequence tags which were unambiguously mapped to the H . sapiens genome . Zhang et al . developed and applied a computational approach for identifying positioned nucleosomes to this histone-methylation ChIP-seq data , and identified over 438 , 000 positioned nucleosomes [2] . A subsequent set of experiments by Schones et al . eliminated the ChIP step to produce genome-wide maps of nucleosome positions in both resting and activated H . sapiens CD4 T-cells [3] . These two experiments respectively resulted in 154 million and 142 million unambiguously mapped sequence tags . A similar genome-wide experiment , conducted in S . cerevisiae by Field et al . , produced 380 , 000 fully sequenced nucleosomes which were mapped to the S . cerevisiae genome with at least 95% identity [4] . In the past three years , at least eight significant papers have described nucleosome positioning models based on DNA sequence signals . A commonly cited nucleosome affinity feature is a 10 bp periodicity of certain dinucleotides , which was first described by Trifonov et al . in 1980 [5] and has since been confirmed in both synthetic [6] , [7] and natural sequences from a variety of organisms including chicken [8] , mouse [7] , S . cerevisiae [4] , [9] , [10] , worm [10] , [11] , and H . sapiens [12] . The periodic repetition of these sequence elements , with a period that matches the pitch of the DNA helix , is thought to encourage the large-scale bending of the DNA molecule necessary to form a nucleosome . As a result , several computational models have emphasized the presence of this dinucleotide periodicity within the nucleosome core [4] , [9] , [10] , [13] . However , based on a large dataset of S . cerevisiae nucleosomes , Mavrich et al . [14] observed that an enrichment of AA dinucleotides toward the 5′ end of the nucleosome was in fact a better descriptor of nucleosome positioning than the 10 bp periodicities of AA/TT . In contrast to the computational models derived from short sequences chosen for their high affinity to wrap around histones and form nucleosomes , models derived from larger nucleosome-occupancy datasets have frequently found that the strongest sequence signals are nucleosome-inhibiting rather than nucleosome-forming [4] , [14] , [15] . Discriminative models [16] , [17] as well as regression-based models [15] , [18] found that the most statistically significant features were more often exclusion signals rather than occupancy signals . In the approach described by Peckham et al . [16] , of the top 17 features only 5 are nucleosome occupancy signals . The same trend was observed by Yuan et al . [15] even though their statistical model explicitly sought to extract dinucleotide periodicities using wavelet analysis: out of the 17 selected features only 3 are positive for nucleosome occupancy , and none of the positive features were related to the 10 bp periodicity of any dinucleotide . Lee et al . [18] also concluded that nucleosome occupancy is probably more often directed by exclusion signals , and their Lasso-based model assigned the greatest significance to DNA structural features ( e . g . tilt and propeller twist ) . In this work we present a new approach to predicting nucleosome positioning directly from DNA sequence . Although our model also includes features describing dinucleotide and trinucleotide sequence patterns , it was originally inspired by our observation of a dramatic mono-nucleotide sequence pattern surrounding the nucleosome positions identified using the Nucleosome Positioning from Sequencing ( NPS ) algorithm [2] applied to the Barski at al . dataset . We subsequently obtained a nearly identical nucleosome pattern from the Schones dataset derived from resting H . sapiens CD4 T-cells [3] by using a version of the NPS software that we modified to estimate nucleosome dyad positions rather than nucleosome occupancy regions . An analysis of the distribution of start-to-start and start-to-end distances for the short-read sequencing tags ( as described in [11] , and shown in Figure S1 ) indicates that the Schones dataset has more consistent nucleosome-sized start-to-end distances than a combination of the 21 separate ChIP-seq experiments in the Barski dataset . We conclude that nucleosome dyad positions inferred from the Schones dataset have a smaller average error than those inferred from the Barski dataset , and therefore use the Schones data to evaluate the performance of our model . We show that in both H . sapiens and S . cerevisiae , the most informative individual features are the mono-nucleotide patterns , although the additional information provided by di- and tri-nucleotide features improves the performance of our sequence scoring method . Our method for computing the dyad score of a given DNA sequence position consists of two steps: first a set of patterns are correlated with the local DNA sequence , and second the resulting correlation values are weighted and summed to produce the final score . The two elements most responsible for our method's discriminative power are the length of the patterns used and the discriminative weights applied to the sequence features . We determined that the optimum pattern length is between 300 and 350 nucleotides—indicating that the DNA sequence pattern of a nucleosome includes not just the core region that is tightly wound around the histone proteins , but the adjacent linkers as well . Notably , this result was the same for both H . sapiens and S . cerevisiae . In the second step , the weights allow our method to selectively assign greater importance to the more informative features—e . g . , the trinucleotide AAA is given a higher weight than GTA . By examining the patterns associated with and the classification performance achieved by each of the mono , di- and tri-nucleotides , we may also be able to gain a deeper understanding of the forces that influence nucleosome positions within the chromatin structure , and to what extent these forces are consistent between H . sapiens and S . cerevisiae . Toward this end , we hypothesize that the close proximity of the two superhelical coils within each nucleosome and the structure of the 30 nm fiber also play a role in determining the DNA sequence preference of nucleosomes . The dyad scores produced by our method are relatively insensitive to local AT-content and can be used to accurately discriminate dyad positions from adjacent linker regions without requiring an additional dynamic programming step to capture the linker-nucleosome-linker pattern . Although not required , such a post-processing step can be easily applied to these scores in order to estimate the probability that a nucleosome is centered at any particular genomic position or that a particular nucleotide is “occupied” by a nucleosome , as has been done previously [4] , [19] . While such a post-processing step entails making assumptions regarding overall nucleosome density and the distribution of linker lengths , it can be used to find the most likely parse of a DNA sequence into nucleosomes and linkers , and to compute posterior probabilities of nucleosome occupancy at each position along the sequence . The most likely parse identifies nucleosome positions which can then be compared to experimentally estimated nucleosome positions , an evaluation method which has been used in , e . g . [9] , [16] . Posterior probabilities of nucleosome occupancy provide normalized scores which are more amenable to computing average landscapes of nucleosome occupancy surrounding genomic features such as transcription start sites . In this work we choose to evaluate our dyad-scoring method by testing how well the raw scores are able to discriminate dyad positions from adjacent linker regions , a similar but more stringent evaluation criterion than has been used previously [4] , [10] , [16] , [17] . We present an evaluation of our method on the Schones dataset derived from H . sapiens T-cells , as well as on the genome-wide S . cerevisiae data made available by Field et al . [4] . In addition , we compare our approach to two recently published methods [4] , [10] and show that our method is significantly better at discriminating dyad positions from adjacent linkers . Further , we compare the H . sapiens trained patterns to the S . cerevisiae trained patterns and find large-scale similarities despite the presence of 10 bp periodicities in the S . cerevisiae patterns and the striking lack thereof in the H . sapiens patterns . We also apply our method to the entire H . sapiens and S . cerevisiae genomes as well as to specific subsets of interest including transcription start sites , CTCF binding sites , and H . sapiens repetitive elements .
A recently published dataset [2] provides the largest collection to date of experimentally determined H . sapiens nucleosome positions . This set of 438 , 652 nucleosome positions , which we will refer to as the Zhang positions , was derived from the histone methylation ChIP-seq data from CD4 T-cells [1] using the NPS algorithm [2] . Each nucleosome position in the Zhang dataset is a short segment of DNA , specified by a pair of chromosome coordinates , and is annotated with a p-value and a list of histone marks . Estimating the nucleosome dyad position as the mid-point of each of the Zhang nucleosome regions , we extracted DNA sequence from the reference genome centered at each of these dyad positions , and we computed the mono-nucleotide position specific frequency matrix shown in Figure 1 ( top ) . Far from the nucleosome dyad ( as shown in Figure S2 ) , the background GC fraction is 0 . 46 , which is higher than the H . sapiens genome-wide average of 0 . 41 , and consistent with the known bias of the Barski et al . dataset toward GC-rich regions of the H . sapiens genome . In the nucleosome core , however , the average GC content is significantly higher than the average AT content . Within a narrow window around the dyad , a nucleosome-sized pattern is observed . The pattern is its own reverse-complement: the A and T traces mirror each other across the dyad , as do the C and G traces , and the pattern emerges even when only the reference strand is used for each nucleosome positions . ( Using both strands enforces this reverse-complement symmetry by construction . ) The reverse-complement symmetry is an expected consequence of the dyad symmetry of the nucleosome particle . However , the fact that each trace is not itself symmetrical around the dyad axis is intriguing and shows that there is a directionality to the nucleosome which obeys the antiparallel , complementary nature of the double-stranded helix: the highest local density of G's and the lowest local density of T's occur 40 nucleotides 5′ of the dyad , and the highest local density of C's and the lowest local density of A's occur 40 nucleotides 3′ of the dyad . The dominant hypothesis regarding DNA sequence preference of nucleosome formation is related to the curvature required to wrap the double helix tightly around the histone core [20] . However , as illustrated in Figure 2a , the curvature is relatively uniform throughout the nucleosome core [21] and therefore , while this hypothesis explains the frequently observed 10 bp periodicity , it does not explain asymmetric patterns such as those shown in Figure 1 , with extrema at some distance from the dyad . We propose that two other structural aspects of the chromatin may explain why the extremes of the nucleosome pattern ( local maxima for C and G , and local minima for A and T ) are centered approximately 40 bp on either side of the dyad rather than at the dyad itself . The first structural aspect that we will consider is the close proximity of the two superhelical coils within each nucleosome: DNA regions that are 80 bp apart are brought into close proximity [22] , as shown in Figure 2b , while the 10–20 bp immediately surrounding the dyad are not in similarly close proximity to another double helix , as shown in Figure 2c . Specifically , base pair is brought into close proximity with basepair , for ( with the dyad defined as position 0 ) . If this close proximity of the two double-helices has an effect on the nucleosome sequence preference , this effect would be observed most strongly 40 bp on either side of the dyad . The second structural aspect is related to the structure of the “30 nm fiber” . Although this structure is not yet well understood , all proposed structures are such that dyads face the center of the fiber while the DNA regions 20–60 bp on either side of the dyad form the exterior of the fiber [23]–[25] . We hypothesize that DNA regions on the outside of the 30 nm fiber may experience different selective pressures than regions on the inside of the fiber , and the result of this difference would be a nucleosome sequence pattern with extreme deviations centered approximately 40 bp on either side of the dyad . We note that while both of these hypotheses are consistent with the asymmetric patterns presented here , they would also be consistent with symmetric , M- or W-shaped patterns with local maxima or minima at 40 bp . In order to quantify the significance of the pattern shown in Figure 1 ( top ) , we consider each of the four mono-nucleotide traces separately . The most striking aspect of the pattern is the relatively large variation in the probability of each of the nucleotides , across a distance of less than 150 bp , averaged over more than 400 , 000 DNA segments . Based on a null model of a similarly constructed pattern using randomly sampled DNA segments , we estimate the probability that this observed variation could occur by chance to be . ( See Methods for details and Figure S3 for a plot of the null model distribution . ) In order to determine whether the observed pattern might be the result of an artifact in part of the dataset , we considered the possible impacts of varying AT-content and repetitive sequences . We found that each of the five patterns obtained after partitioning the data into quintiles according to AT-content were similar to the original pattern , disregarding vertical translations of the individual components reflecting increases in AT content and corresponding decreases in GC content ( Figure S4 ) . Partitioning the dataset into three subsets according to the distance to the nearest repeat also does not significantly alter the shape of the pattern ( Figure S5 ) . We noted earlier that the A and T traces mirror each other across the dyad , as do the C and G traces , and that intriguingly A and C mirror each other across a horizontal line of symmetry , as do the T and G traces . The first symmetry , of A/T and C/G across the dyad , is a natural consequence of the dyad symmetry of the nucleosome , while the second A/C and T/G symmetry is not . Although a similar downward trend 5′ to 3′ across the nucleosome dyad and a local minimum 3′ of the dyad in the AA dinucleotide frequency can be seen in the figures in an early paper by Ioshikhes et al . [26] , the trend was not explicitly noted . Instead the authors emphasized the asymmetry in the peaks of the dinucleotide patterns and found that the 10 bp periodicity exhibited by the AA and TT dinucleotides had opposite phase , in contrast to the same-phase periodic pattern described earlier by Satchwell et al . [8] and more recently by Segal et al . [9] . More recently , the enrichment of AA dinucleotides 5′ of the dyad and TT dinucleotides 3′ of the dyad has been described [14] , although no biological hypothesis for this directional preference has yet been suggested . We will refer to the pattern illustrated in Figure 1 as AGCT , based on the 5′ to 3′ ordering of the local maxima . This simple pattern is consistent with the common nucleosome model: higher AT content in the linker regions and higher GC content in the core . We hypothesized the existence of alternative forms of this pattern in which the ordering of the individual nucleotides is permuted while conforming to the common model—the other possible patterns would be ACGT , TGCA , and TCGA . To test this hypothesis , we created models of all four pattern variants and partitioned the input set of sequences according to which of the four patterns best matched each individual sequence ( if a particular DNA sequence did not correlate well with any pattern , it was assigned to the no-match partition ) . We found , as expected , that more of the input sequences correlated well with the AGCT pattern than with any other pattern ( 25% ) . However , our hypothesis was validated in that an even larger fraction ( 32% ) of the sequences correlated well with one of the other three patterns ( details in Supplement and Figure S6 ) . To verify that this nucleosome pattern is not an artifact of the Barski et al . dataset , we estimated nucleosome dyad positions from the tag coordinate files for resting CD4 T-cells published by Schones et al . [3] . The experimental procedure used by Schones et al . is very similar to the one used by Barski et al . , but without the ChIP step used to isolate specific histone modifications . We modified the NPS software [2] so that it would output nucleosome dyad positions rather than variable length nucleosome regions ( see Methods ) , and applied it to the Schones dataset . The result was a list of over 828 , 000 nucleosome dyad positions with NPS-assigned p-values . The mono-nucleotide patterns learned from this independently derived list of nucleosome positions are nearly identical to the corresponding patterns derived from the Zhang positions ( Pearson correlation 0 . 99 ) , as shown in Figure 1 . In addition , we computed patterns based on the top-50% and top-25% scoring dyad positions ( corresponding to p-value thresholds of and respectively ) , and found the correlations between the patterns derived from the full set and these subsets to also exceed 0 . 99 , indicating that the pattern is stable and can be learned from smaller datasets . A subset of the dinucleotide patterns are shown in Figure S7 . The patterns for AT , TA , GC , and CG are symmetric about the dyad , as expected , because each dinucleotide is its own reverse complement . What is intriguing , however , is that the standardized pattern for CG is nearly identical to that for GC—despite the dramatically different occurrence rates of these two dinucleotides . The standardized patterns for TA and AT are also nearly identical . We note that neither of these independently derived mono-nucleotide patterns show evidence of 10 bp periodicity , and this is also true of similarly computed dinucleotide patterns . There are two possible explanations for this lack of a periodic component in this pattern . First , the NPS software uses a bin size of 10 nucleotides in processing the short-read sequencing data and estimating the dyad positions , resulting in an average error of at least 5 nucleotides in each position estimate which would smooth out any 10 bp periodicity in the average pattern . Second , 10 bp periodicity of the AA dinucleotide has to date been observed only in small sets of H . sapiens nucleosomal sequences and is not observable on a genome-wide scale in H . sapiens , in sharp contrast to S . cerevisiae [12] , [27] . While removing the binning step in the NPS process may yield more accurate dyad positions , we caution that it may also amplify the impact of the MNase sequence specificity as is apparent in the higher-resolution yeast data set discussed below . Furthermore , 10 bp periodicity has been most apparent in alignments of nucleosomal sequences but has not been shown to be a significant factor in identifying and classifying such sequences . A recently published genome-wide experimental assay in S . cerevisiae produced a dataset of 380 , 000 fully-sequenced nucleosomal sequences [4] . This experiment was based on a sequencing technology capable of 200 bp reads , thereby eliminating the uncertainty inherent in the Barski and Schones datasets regarding the precise lengths of the MNase-cleaved DNA fragments . Estimating dyad positions from the genomic positions of these nucleosomal sequences can therefore be done using a simpler approach ( see Methods ) , which produced a total of 50 , 815 unique dyad positions . This is a significant fraction of the estimated 70 , 000 nucleosomes required by the entire 12 Mb genome . We divided this set of dyad positions according to the confidence associated with each position ( estimated as the number of locally overlapping reads ) , to produce successively smaller subsets of size 25384 , 12698 , 6355 , and 3180 respectively , with each dyad position in the smallest subset estimated from an average of 20 overlapping reads . The patterns derived from this set of S . cerevisiae dyad positions , as shown in Figure 3 , include two elements not present in the H . sapiens nucleosome patterns: a very strong artifact due to the MNase sequence bias at a distance of 80 bp on either side of the dyad ( corresponding to roughly half of the mean read length of 156 nucleotides ) , and evidence of 10 bp periodicity for certain dinucleotides . Figure 4 shows the summed patterns for A/T-only and C/G-only dinucleotides , illustrating the lack of apparent periodicity in H . sapiens as compared to S . cerevisiae , although we also note that the artifact in the S . cerevisiae patterns due to the MNase sequence-specificity has a significantly larger amplitude than the 10 bp periodicity . The common elements between S . cerevisiae and H . sapiens include the downward trend across the dyad of the A nucleotide in the 5′ to 3′ direction , the corresponding upward trend of the T nucleotide , and the local minima in the A- and T-patterns at bp and bp respectively . The standardized patterns for S . cerevisiae and H . sapiens for A , AA , and AAA are shown in Figure S8 . Aside from the 10 bp periodicity evident in the S . cerevisiae patterns , the overall shapes of these patterns are strikingly similar ( Pearson correlations are A:0 . 77 , AA:0 . 88 , AAA:0 . 91 ) , suggesting that perhaps it is this wider underlying oscillation , more than the 10 bp periodicity , which promotes nucleosome positioning across species . To avoid biases arising from the MNase sequence specificity , Field et al . restricted their model to the 127 positions centered at the dyad [4] . The presence of this artifact in the nucleosome patterns is also an indication that many of the estimated dyad positions are shifted by a few nucleotides from the true positions—more accurate dyad positions could potentially be estimated by inserting an alignment step in the pattern-estimation procedure similar to [26] , [28] . We instead eliminate the MNase artifact while still learning a large-scale pattern by linearly interpolating each pattern across a 30 bp width centered at 80 as shown in Figure 3 . Based upon the observed oscillatory pattern of nucleotide composition across the nucleosome , we present here a novel approach to predicting nucleosome positioning from DNA sequence alone . Previous methods have frequently taken a hypothesis-testing approach common in motif-finding algorithms in which a foreground or motif model score is compared to a background model score . In determining the nucleosome-formation potential of a given DNA sequence , a nucleosome model is used to compute a score under the nucleosome hypothesis , and a linker model is used to compute a score under the null ( linker ) hypothesis . The final score is typically either a ratio or a difference of these two scores . For a given input DNA sequence ( the length of which varies depending on the specific implementation but is generally 147 nucleotides or less ) , the basic question being asked is thus: which of these two models best represents this particular sequence ? When scoring the likelihood that a particular nucleotide is at the center of a nucleosome , we have found that using a wider sequence window asks the more appropriate question: do the 147 nucleotides centered at this position fit our model of the nucleosome core and do the adjacent regions fit our model of the linker region ? In previously published approaches , this alternating linker-nucleosome-linker model is captured by a subsequent dynamic-programming step ( e . g . [9] ) , but we will show that the predictive power of the model can be significantly improved by including this longer pattern directly into the initial scoring function . Based upon our observation that the overall shape of the mono-nucleotide pattern first derived using the Zhang positions was relatively insensitive to local AT-content , our initial insight was that the model should be insensitive to the average local sequence composition . This goal is consistent with the biological requirement for packaging DNA sequences with widely varying AT-content not only within any one genome but across the genomes of all eukaryotes [20] . We avoid inherent sequence composition bias by comparing the input DNA segment to the shape of the nucleosome pattern using a Pearson correlation which disregards vertical scaling or translation of the individual pattern components . Further , rather than framing the model in a probabilistic setting , we choose to take the more general approach of extracting an arbitrary number of informative , sequence-related features which are then individually weighted and combined to produce a final dyad score . The complete details of our algorithm are given in the Methods section , but we will outline the basic approach below . Given an input sequence , of ( odd ) length , we extract a number of descriptive features and compute the dyad score for the mid-point of sequence as the weighted sum of these features . The primary features in our model are correlation coefficients: each one represents the correlation between a previously learned pattern and the new input pattern for a given k-mer , of length . Based on our earlier observation that individual components within the pattern are occasionally reversed , each input pattern is compared to two versions of the trained pattern: and where is the m-pattern learned from the training set of nucleosome sequences aligned at the inferred dyads , and is simply the reflection of across the axis of symmetry at the dyad . We further add , as secondary features , the number of occurrences of each k-mer in the input sequence and its reverse-complement . The intuition here is that a correlation coefficient of 0 could be the result of sequence containing zero occurrences of , or it could be a true lack of correlation between two non-zero vectors . Likewise , a high correlation score may be more significant if it is based on a sequence with a high number of occurrences of m . In the final step of the training process , we train a binary classifier known as a linear support vector machine ( SVM ) [29] , [30] to discriminate between two sets of examples , each of which is described by a vector of the features defined above and is labeled either positive ( dyad ) or negative ( linker or non-dyad ) . The output of this training step is a set of feature-specific weights which , when applied to the set of training examples , optimizes the discrimination between the positive and negative examples . These weights can subsequently be used to compute a score ( the dyad score ) for any future test example . The sign of the score indicates which side of the decision boundary the test example falls on , and the magnitude of the score is an indication of the confidence of the classification . Although previously published models have commonly used long linkers or nucleosome-free regions as negatives in training and evaluation [4] , [15] , each nucleosome core is flanked by two linker regions , and we define a more stringent discrimination task by testing how well each dyad position in the test set can be distinguished from the corresponding set of adjacent linkers . In order to completely define the model described above , the width of the individual patterns , the k-mers of interest , and the distance to the linker positions to be used as negative training examples need to be specified . Note that these linker positions and the nucleotide sequence surrounding them are used only in training the SVM weights and do not affect the learning of the k-mer patterns . Further , there is no implicit relationship between the values of ( the pattern width ) and ( the distance between the dyad positions and the “negatives” examples ) . We evaluated the effect of these two parameters on the discrimination performance of our model and found that the width of the patterns ( ) has the most significant effect on the ultimate performance , as shown in Figure 5 . For shorter patterns ( ) the kmer-count features significantly improve the performance , while for longer patterns ( ) they provide relatively little improvement . The optimum pattern width varies somewhat across different datasets , but is generally between 301 and 351 nucleotides , i . e . extending 150–175 positions on either side of the dyad . We also examined the sensitivity of the discrimination performance to the distance between the dyad and non-dyad positions . When this distance is zero , it is of course impossible to discriminate between the two sets , and the resulting area under the ROC curve is . We would expect the performance to improve as the distance increases , up to a maximum value when the distance is roughly half the inter-dyad distance . Beyond this point , we would expect the performance to begin to get worse as the “negative” position approaches the neighboring dyad , and this intuition is borne out by the experimental results shown in Figure S9 . The peak performance occurs when the distance between the positive and negative examples is 110–120 nucleotides . Based on these analyses , our final model is defined with patterns of width 301 nucleotides , and the SVM is trained with negative examples at a distance of 110 nucleotides on either side of each dyad . We chose to limit our set of k-mers to those of length 1 , 2 , and 3 . Using longer k-mers would require exponentially more parameters , and we found that the improvement gained even by adding trinucleotides was relatively small . We evaluate our method using a cross-validation approach to ensure that the model is not over-fitting the data . For each chromosome , the training set contains all dyad positions not on chromosome , and the held-out test set consists only of those dyad positions on chromosome . The training of the model consists of three steps: first , a position-specific pattern of width is learned for each k-mer from the sequence centered at each in the training set . Second , features describing the local context of each position as well as are computed: these include correlation scores against each of the learned patterns and counts for each of the k-mers . Third , these feature vectors and labels are used to train a linear SVM . The evaluation on the held-out test set involves similarly computing features describing the DNA sequences centered at each and on chromosome , and computing scores for each by using the SVM weights learned during training . The ability of these scores to discriminate between the dyad and non-dyad positions in the test set are evaluated using standard ROC analysis . The datasets we used to train and test this model were described earlier and consisted of 800 , 000 H . sapiens dyad positions estimated from the Schones dataset , and 50 , 000 S . cerevisiae dyad positions estimated from the Field dataset . Although different methods were used to create each of these sets of dyad positions , both include experimentally-derived confidence scores . We used these confidence scores to further subdivide each dataset by repeatedly taking the top-scoring half , resulting in 3 H . sapiens sets ( all , top 1/2 , and top 1/4 ) and 5 S . cerevisiae sets ( all , top 1/2 , top 1/4 , top 1/8 , and top 1/16 ) . We trained our H . sapiens nucleosome model on the top 1/4 subset ( approximately 200 , 000 dyad positions ) , and we trained the S . cerevisiae nucleosome model on the top 1/2 set ( approximately 25 , 000 dyad positions ) . Figure 6 shows the composite results for our model , based on chromosome-by-chromosome cross-validated training and testing . The number of dyad positions on each chromosome and the per-chromosome area under the ROC curve for each dataset are provided in Tables S1 and S2 in the supplement . The area under the ROC curve is an indication of how well our model discriminates between dyad and linker positions , and the fact that the model performance improves for the highest-scoring subsets shows that , on average , the most-consistently positioned nucleosomes are also the ones given the highest scores by our model , while the adjacent dyads are simultaneously given lower scores . We also found that the difference between the cross-validated results shown here and those obtained when using all of the data for both training and testing were negligible , due to the large amount of training data and the relatively simple model being trained . We note that the performance for the top 1/4 H . sapiens dataset is nearly identical to the performance for the top 1/16 S . cerevisiae dataset , and the same is true for the top 1/2 H . sapiens dataset and the top 1/8 S . cerevisiae dataset . The top 1/16 S . cerevisiae dataset contains positions sampled on average every 4000 bp across the entire S . cerevisiae genome . Assuming an average nucleosome repeat length of 170 bp , this represents approximately 4% of all nucleosomes . The comparable H . sapiens dataset , based on the ROC curve , is the top 1/2 set which consists of positions sampled on average every 7500 bp across the H . sapiens autosomes . ( The X and Y chromosomes are significantly undersampled as compared to the autosomes , so we exclude them in this analysis . ) Considering the limits imposed by sequencing depth and the unique mapping of short sequence tags to the H . sapiens genome , it seems reasonable to suggest that perhaps half of the highly-positioned nucleosomes were missed in the genome-wide Schones experiment . With this assumption we estimate that approximately 4% of the nucleosomes in both H . sapiens and S . cerevisiae are positioned consistently enough across a population of cells to produce an area under the ROC curve of 0 . 91 , which corresponds in this case to a true positive rate of 73% at a false positive rate of 10% . Doubling the size of the set takes us to the next curve for each species: approximately 8% of nucleosomes are positioned consistently enough to produce an area under the ROC curve of 0 . 89 , which corresponds in this case to a true positive rate of 66% at a false positive rate of 10% . We compared the predictive performance of our model to the two recently published nucleosome prediction models described by Field et al . [4] and by Kaplan et al . [10] . These two previously published models are algorithmically very similar , the main distinction being that the Field model was trained on in vivo S . cerevisiae mono-nucleosomes , while the Kaplan model was trained using a genome-wide occupancy map of nucleosomes assembled in vitro on purified S . cerevisiae genomic DNA . For comparisons in H . sapiens , we downloaded the occupancy probabilities and raw binding scores from the Segal lab website , and for comparisons in S . cerevisiae , we downloaded the executable and obtained raw binding scores , start probabilities and occupancy probabilities for the entire S . cerevisiae genome . Our model is a purely local scoring function , requiring only 301 bp of sequence to make a prediction at a single point , and as such , is computationally most similar to the raw binding scores from these two models . However , the Field and Kaplan raw binding scores are sensitive to variations in the local AT-content and are not able to discriminate accurately between , for example , nucleosome dyad positions in high-AT regions and linker positions in low-AT regions . The dynamic programming stage of the Field and Kaplan models corrects for this sensitivity , and the resulting occupancy and start probabilities are better able to discriminate between dyads and linkers . The ROC curves for the Field and Kaplan models on representative datasets from S . cerevisiae and H . sapiens are shown in Figure 7 , together with the corresponding ROC curves for our model . In all binary classification tests presented here , the positive examples are the experimentally determined dyad positions , while the negative examples are the positions 110 bp to either side of each dyad . This definition of negatives examples ( i . e . linkers ) is different from that used in [4] in which linkers were defined as contiguous regions of length 50–500 bp not covered by any nucleosome . We chose to use a different definition for two reasons: first , long nucleosome-free regions may be a result of the experimental protocol [31] , [32] , and second , sequencing-depth limitations in H . sapiens genome-wide experiments mean that true negatives are far outweighed by false negatives . Furthermore , the extent to which an individual dyad position can be distinguished from its immediately adjacent linker regions is a direct indication of the apparent positioning stringency . The difference in discrimination performance between our model and the two previous models is more significant in the H . sapiens dataset . This is to be expected as the Field and Kaplan models were both trained on S . cerevisiae datasets . In the S . cerevisiae evaluation ( Figure 7a ) , the Field model outperforms the Kaplan model , which is also to be expected as it was trained on this test data ( although the classification task here is different than that shown in [4] because we have defined the negative class differently ) . In Figure 7a , at a false positive rate of 10% , our model has a true positive rate in S . cerevisiae of 64% compared to true positive rates of 55% , 51% and 22% for the three Field scores , and 46% , 38% , and 22% for the three Kaplan scores . For the H . sapiens evaluation shown in Figure 7b , the Kaplan model outperforms the Field model . At a false positive rate of 10% , our model has a true positive rate of 79% compared to true positive rates of 49% and 26% for the two Kaplan scores , and 41% and 17% for the two Field scores . We further analyzed the performance of our model by considering different subsets of the features as well as features associated with individual k-mers in order to determine which features are most informative , and to investigate whether this was consistent between H . sapiens and S . cerevisiae ( Figure 8 ) . Not surprisingly , in general the more features are used , the better the performance , but the best individual features are the mono-nucleotides—and this is true for both H . sapiens and S . cerevisiae . This result is rather surprising and indicates that these mono-nucleotide patterns are able to summarize the relevant information in longer homo-polymer stretches . The most informative di- and tri-nucleotides are AA/TT and AAA/TTT , reconfirming the importance of poly ( dA∶dT ) tracts in the organization of nucleosomes [33] . Nucleosomes are the basic repeat element of the first level of the chromatin structure , forming the “beads-on-a-string” fiber which in turn coils into a larger structure known as the 30 nm fiber . The average length of the linker DNA between adjacent nucleosomes defines the nucleosome repeat length which in turn affects the structure and size of the 30nm fiber [34] , [35] . Because our model is essentially a pattern-matching algorithm , we were interested in evaluating whether the nucleosome pattern described by our model appeared to repeat at regular intervals along the H . sapiens and S . cerevisiae genomes . It is possible to obtain an empirical distribution of distances between successive experimentally-determined S . cerevisiae dyad positions because this set constitutes a significant fraction of the total number of nucleosomes expected in S . cerevisiae . The resulting distribution is shown in Figure 9 and confirms the dominant nucleosome repeat length of 165 bp in S . cerevisiae [36] . The empirical distribution obtained from the far sparser set of experimentally-determined H . sapiens dyad positions is less reliable ( in the largest set of over 800 , 000 positions , over 80% of the positions are more than 500 bp away from the nearest upstream position ) , but it too shows a clear peak—in this case at 200 bp , although less than 2% of the dyad positions are between 190 and 210 bp from the nearest upstream position . We made genome-wide predictions for both H . sapiens and S . cerevisiae using our model , and found that the distribution of distances between successive predicted dyad positions ( local maxima with positive model scores ) , showed interesting trends . For S . cerevisiae , aside from an over-representation of dyads predicted to be close together , there is a single broad peak between 175 and 200 bp , while for H . sapiens there is a bimodal distribution with one peak at 175 , and one at 225 bp , as shown in Figure 9 . If the local maxima were randomly distributed , the distribution of distances from one to the next would follow a geometric probability distribution , monotonically decreasing for longer inter-dyad distances . We hypothesized that repetitive sequences may be responsible for a significant number of consistent inter-dyad distances in H . sapiens , and found that partitioning the predicted dyad positions according to the local repeat content showed that the predicted dyads in repetitive regions contribute to both local maxima in the distance distribution , while the predicted dyads in non-repetitive regions contribute mainly to the second peak in the distance distribution . The relationship between Alu repeats and nucleosome formation has been widely studied—specifically , Alu sequences have been shown to facilitate the formation of nucleosomes in vivo [37] , [38] , with one putative nucleosome centered over the RNA polymerase III promoter A box ( near the 5′ end of the Alu sequence ) , and a second putative nucleosome positioned over the right arm of the Alu element , flanked by two A-rich regions . Because the upstream and downstream sequence also affect our model predictions , we extracted DNA sequence surrounding 4930 separate AluSx sequences ( the most common type of Alu repeat ) , and found that our model predicts two strong dyad positions 170 nucleotides apart , as shown in Figure 10 . Our model's prediction of just two locally optimal dyad positions , which could be simultaneously occupied by a pair of adjacent nucleosomes , is in contrast with recently published predictions of several alternative nucleosome positions separated by multiples of 10 . 4 bases [39] . A nucleosome centered at position 40 would wrap the first 110 bp of the Alu sequence as well as approximately 30 upstream base pairs around the histone octamer , effectively blocking access to the two internal Pol III promoters and possibly an upstream enhancer , and rendering the Alu transcriptionally silent . We also analyzed all H . sapiens repetitive sequences in RepBase[40] ( considering only the consensus sequence and disregarding flanking regions from specific instances of the repeat in the H . sapiens genome ) and found that several longer repetitive elements resulted in predicted dyad positions at spacings between 165 and 185 bp , including endogenous retroviruses ( ERV1 , ERV2 and ERV3 ) , non-LTR retrotransposons ( L1 in particular ) , and DNA transposons ( such as the Mariner transposable element ) . Transcription start sites ( TSS ) and CTCF-binding sites have been shown to be strongly correlated with ordered arrays of nucleosomes based on numerous experimental assays . Using high-resolution tiling microarrays to analyze nucleosomal DNA in S . cerevisiae , Lee et al . [18] identified a general pattern of nucleosome occupancy anchored at the TSS . The average nucleosome occupancy signal , aligned at the TSS and averaged over all genes , shows a nucleosome free region centered approximately 30 bp upstream of the TSS , flanked by the −1 nucleosome centered 170 bp upstream of the TSS and the +1 nucleosome centered 100 bp downstream . The strict positioning of nucleosomes further upstream and downstream from the TSS decays gradually , although more slowly in the transcribed region ( downstream ) . A statistical packing model of nucleosome positioning was proposed by Mavrich et al . [14] whereby the genomic sequence specifies the locations of the +1 and −1 nucleosomes , and these strictly positioned nucleosomes maintain a relatively large nucleosome-free region over the TSS , acting as barriers against which adjacent nucleosomes are packed . Using the Barski ChIP-seq dataset , Zhang et al . showed similar results in H . sapiens , with the +1 nucleosome again the most strongly positioned , just downstream of the TSS [2] . Using the same approach , they observed 3-4 well positioned nucleosomes on either side of CTCF binding sites , while the binding site itself showed strong depletion . Further studies of nucleosome organization surrounding CTCF binding sites have suggested that binding of CTCF provides an anchor point for positioning nucleosomes [41] , [42] , while unbound sites are occluded and rendered inaccessible by the presence of a nucleosome [42] . We note , however , that recent studies have shown that regions that have been previously described as ‘nucleosome-free’ are in fact frequently occupied by nucleosomes that are unusually unstable under the conditions normally used in sample preparation [31] , [32] . In order to evaluate to what extent our model replicates these experimental results for transcription start sites , we applied our scoring method to a set of S . cerevisiae DNA segments aligned at 5 , 015 high-confidence TSS and clustered according to promoter nucleosome signatures as in [18] ( Figure S10 ) , and to a set of H . sapiens DNA segments aligned at the 32 , 079 TSS from the DBTSS database of H . sapiens transcriptional start sites [43] ( Figure S11a ) . The S . cerevisiae TSS predictions agree with the experimental results in the cluster-dependent strength of the nucleosome-free region upstream of the TSS , and indicate that the sequences associated with clusters identified experimentally also result in different average model predictions . The H . sapiens TSS predictions are dominated by the +1 nucleosome dyad peak 65 bp downstream of the TSS , with two additional peaks clearly visible further downstream . Unlike the S . cerevisiae TSS predictions , which show evidence of a larger than average linker between the +1 nucleosome at 60 bp downstream of the TSS and the −1 nucleosome at 190 bp upstream , the H . sapiens TSS predictions suggest the presence of a weakly positioned “0” nucleosome 105 bp upstream of the TSS , between the −1 nucleosome ( at approximately −270 bp ) and the +1 nucleosome ( at approximately +65 bp ) . We caution , however , that computing average profiles by averaging these locally-computed scores over a set of aligned DNA sequences has inherent drawbacks in that the resulting average will be dominated by the sequences with the highest-scoring dyad locations ( and the lowest-scoring linker locations ) as well as by the relative positions of these peaks ( and troughs ) . Posterior probabilities of nucleosome occupancy , obtained by post-processing these local scores using a dynamic programming approach may result in average profiles that more closely reproduce experimental results . Indeed , such average profiles , representing thousands of TSS , whether based on predictions or on experimental data , fail to convey the substantial variation that exists in the position of the +1 nucleosome . In order to confirm that our model accurately reproduces this variation , we formed subsets of the DBTSS sites according to the position of the “+1” nucleosome in the Zhang dataset ( if any ) , using 30 bp windows spanning the region from 40 bp upstream of the TSS to 200 bp downstream , and computed average predictions for each of these subsets . We found that the average predicted position of the +1 nucleosome for each of these subsets correlates extremely well ( R = 0 . 99 ) with the average experimentally-inferred position . ( See Figure S11b for three representative subsets . ) We similarly aligned and analyzed a pair of CTCF-binding site datasets [41]: a set of 6000 occupied binding sites and a set of 6000 unoccupied sites ( Figure S12 ) . The average model scores near CTCF binding sites indicate that the binding site itself is a favorable nucleosome dyad position , in agreement with the experimental observation that unbound sites are occluded by nucleosomes [42] . These predictions are similar to those in [15] , although our method produces a significantly narrower peak at the binding site . On either side of the CTCF binding site , at 300 bp are two weak peaks in the average dyad score , but the regular pattern of nucleosomes on either side of occupied CTCF sites that has been observed experimentally [41] , [42] is not matched by the predictions , suggesting that the positioning of nucleosomes near occupied CTCF sites is driven primarily by statistical packing against a barrier .
The nucleosome DNA pattern that we have described here is largely consistent with previously described nucleosome positioning signals . However , our finding that the two mono-nucleotide patterns ( A/T and G/C ) are individually more predictive than any single dinucleotide or trinucleotide pattern represents a significant departure from the widely held belief that dinucleotide periodicities and poly-A/T tracts are the strongest nucleosome positioning elements . Our model agrees with the hypothesis that periodicity seen in the average profile of a set of nucleosomal sequences reflects an alignment imposed by the structural organization of the nucleosome core particle , rather than periodicity in individual sequences [27] . Elevated GC-content is widely known to be a key feature of nucleosomal sequences and was previously found to be one of the strongest individual predictors of nucleosome occupancy in S . cerevisiae [16] , [18] . GC-rich dinucleotides have also been associated with reduced DNA deformation energy which would facilitate their integration into the core of the nucleosome [44] , [45] . The downward trend 5′ to 3′ across the nucleosome core of the AA dinucleotide frequency has been previously observed [14] , [26] , but the significance of this asymmetry in localizing nucleosomes has not been emphasized . The symmetry of the nucleosome around the dyad axis and the reverse-complementarity of the two strands of the double helix require that the A and T patterns form a mirror-image pair , and likewise for G and C . If each individual pattern was symmetric around the dyad axis , then only two distinct patterns would exist: one for A/T and one for G/C . Furthermore , because p ( A+T ) and p ( G+C ) must sum to unity , these two patterns would be perfectly negatively correlated , and from an information-theoretic point of view the second pattern would provide no additional information not already available in the first . Because each individual pattern is not symmetric around the dyad axis , the four mono-nucleotide patterns combine to provide more information regarding the locally optimal dyad position . We verified this by mapping the DNA sequences down to a two-letter alphabet and training the model as before , and found that the discrimination performance as measured by the area under the ROC curve was significantly reduced . Our model combines the features that promote nucleosome occupancy as well as those that enforce exclusion into a set of k-mer specific patterns . The pattern-correlation method that we use is normalized to remove sequence composition biases , as is also done in the nucleosome-core portions of the Field and Kaplan models [4] , [10] . However , the widening of the model to include the adjacent linkers in each pattern similarly removes composition bias from the scoring of the linkers , resulting in an overall description of the nucleosome that is insensitive to large-scale variations in AT content , an insensitivity which naturally reflects the pervasive presence of nucleosomes in all genomic regions . Our approach also combines elements of previous probabilistic models [4] , [9] , [10] , [13] , with a discriminative approach [16] , [17] . This weighted combination of features allows us to simultaneously make use of mono , di- and tri-nucleotide patterns which provide complementary information . Ioshikhes et al . [13] modeled only the distribution of AA and TT dinucleotides , effectively giving zero weight to all other dinucleotides . Our approach is a generalization of this idea , and we confirm that , of the 10 unique dinucleotides , AA/TT is the most predictive of nucleosome position , while AC/GT and GA/TC are the least predictive . Examining distances between predicted dyad positions on a genome-wide scale , we find evidence for two classes of preferred nucleosome repeat lengths in H . sapiens—one near 175 bp and the other near 225 bp . In S . cerevisiae , a similar analysis produces a broad peak between 175 and 200 bp . This predicted distribution in S . cerevisiae implies longer linkers than the experimentally inferred distribution , a bias toward longer nucleosome repeat lengths which may be caused by the length of our pattern . Although the pattern length of 301 bp was chosen to optimize performance on our datasets of nucleosome positions , the original experiments themselves and the post-processing of the data to obtain estimated dyad positions may produce an ascertainment bias that favors not only highly-positioned nucleosomes but also those flanked by longer linkers . In the H . sapiens genome , the shorter class of linkers are associated with repetitive elements while the longer class of linkers are associated with both repetitive and non-repetitive elements . We hypothesize that , by preferring two different classes of linker lengths , the repetitive elements promote the formation of the two distinct classes of 30 nm chromatin fiber described by Robinson et al . [24] . The processing of the high-throughput sequencing data ensures that only the most stringently positioned nucleosomes will result in high-confidence dyad positions . The largest S . cerevisiae dataset we considered contained approximately 50 , 000 nucleosome positions , or nearly 70% of the expected total number of nucleosomes within the S . cerevisiae genome . By contrast , the largest H . sapiens dataset we considered contained only 5% of the 15 , 000 , 000 nucleosomes we estimate would be required by a single copy of the H . sapiens genome . Our model achieved similar performance on two pairs of datasets: the 3 , 000 S . cerevisiae nucleosomes and 400 , 000 H . sapiens nucleosomes , and the 6 , 000 S . cerevisiae nucleosomes and 800 , 000 H . sapiens nucleosomes . For the smaller pair of datasets , we report a true positive rate of 74% at a false positive rate of 10% , and for the larger datasets , we report a true positive rate of 65% at a false positive rate of 10% . If we assume that roughly half of the well-positioned nucleosomes in H . sapiens were missed through a combination of issues due to short-read sequence mappability and sequencing-depth limitations , then these two pairs of datasets represent 4% and 8% respectively of the entire set of nucleosome positions for these two genomes . This implies that , in both genomes , a comparable and relatively small fraction of nucleosomes are well-positioned and that these positions are predictable based on sequence alone . We believe that the bulk of the remaining nucleosomes follow a statistical positioning model [14] . Our results lead us to a middle ground between , on the one hand , the idea that nucleosome positions in vivo are determined primarily by DNA sequence [9] , [10] , and , on the other , the idea that intrinsic histone-DNA interactions play no part in creating the in vivo pattern [46] . Nucleosome occupancy models in which short linkers are preferred [19] may predict certain nucleosomes to be well-positioned not as a result of a strong local sequence signal , but rather as a direct result of a nearby nucleosome that is itself positioned by a particularly strong sequence signal , the effects of which ripple outwards in the chromatin structure . Knowing which nucleosomes are strongly positioned due to local sequence signals and which ones are merely “packed” against a barrier would further our understanding of the organization of the chromatin . Our estimates that relatively small fractions of nucleosomes are strongly-positioned based on local sequence alone may seem surprising in light of some earlier claims that 50% or more of nucleosome positions could be accurately predicted based on sequence alone [9] , [26] . However , these earlier claims were based on very small sets of well-positioned nucleosomes ( a few hundred as opposed to tens or hundreds of thousands ) , or on criteria which could be satisfied for 32–45% of nucleosome positions by chance . We have defined a more stringent classification task and have tried to assess the fraction of nucleosome positions that are strongly influenced by local sequence features . Further avenues for research to improve this model include discriminatively combining patterns of different lengths or different horizontal scales to capture the variation in linker lengths , as well as investigating the possibility that different types of nucleosomes may be associated with different DNA sequence patterns—for example , a difference in the GC profile of H2A . Z nucleosomes has been recently described [27] . Another interesting direction is to use these dyad scores as well as the experimentally estimated nucleosome positions to train a dynamic Bayesian network which could then be used to make nucleosome-positioning and occupancy predictions . In addition , these predictions could be constrained by the experimental evidence and used to fill in gaps in the data .
The dyad positions in H . sapiens were estimated using a modified version of the NPS ( Nucleosome Position from Sequencing ) software [2] . The original implementation combines offset tags from each strand into a smoothed nucleosome occupancy trace . A p-value threshold is applied to this trace , and the end-points of the regions that exceed the threshold ( with boundaries on the minimum and maximum region extents ) are called positioned nucleosome regions . Our initial pattern was obtained using the Zhang nucleosome positions by assuming that the dyad was at the mid-point of each of these nucleosome regions . The modified NPS software finds the local maximum within each region that crosses the threshold and calls that the dyad position . To estimate dyad positions from the Field dataset of mapped reads [4] , each mapped read was represented on the genomic axis by a triangle of height 1 , and base given by the length of the actual read , and these overlapping triangles were summed to produce a “dyad” trace . All local maxima within local windows of length 141 nucleotides were called dyad positions . Given a set of nucleosome dyad positions and a k-mer of length , we compute the m-pattern in four steps as follows . First , extract a DNA segment of width ( where is odd ) centered at each dyad position from the reference genome . Second , for and its reverse complement , convert the DNA segment into a numerical representation in which all positions are zeros except each position of an exact sub-string match to is set to the value . If is a mono-nucleotide , then , and the numerical representation is a simple bit vector of 1's and 0's . For longer it is possible to have overlapping matches ( e . g . the dinucleotide AA occurs four times with overlap in the segment 5-mer AAAAA ) , and in such cases the values are summed . The sum of the values in the resulting numerical representation is equal to the number of ( possibly overlapping ) occurrences of in the input DNA segment and its reverse complement . Third , average all numerical representations to obtain the average pattern , and finally standardize this pattern such that its mean is equal to zero and its variance is equal to 1 . This procedure ( excluding the standardization step ) is expressed in the following equation:where is the position relative to the dyad and ranges from to , , , indicates the substring in from position through position inclusive , and is the indicator function which is equal to if the argument is TRUE , and otherwise . This procedure will produce mirror-image patterns for k-mers that are reverse-complements of one another . For example , the pattern for the mono-nucleotide A , and pattern for the mono-nucleotide T , will be related as follows: , for as defined above . Dinucleotides , such as TA or GC , which are their own reverse-complements will result in symmetric patterns for which . The 3D visualizations of the nucleosome core particle shown in Figure 2 were created using PyMol [47] and PolyView-3D [48] and PDB [49] structure 1KX5 [50] . Our null model for the variation expected by chance of each nucleotide across a distance of 151 bp was derived empirically from 800 , 000 random sets of DNA sequence fragments . Each random set is equal in size ( ) to the Zhang positions set: for each nucleosome dyad position , we choose a random position within 1000 bp ( in either the 3′ or the 5′ direction ) . We then extract a set of DNA sequences of length 151 bp centered at each of the random positions , and construct the position specific frequency matrix ( PSFM ) as described earlier . We search for the absolute maximum and minimum values for each nucleotide across the 151 bp PSFM , and compute the difference , . For each of the four traces in the observed pattern shown in Figure 1 ( top ) , . Our empirical null model is shown in Figure S3 . The right tail of the empirical null model falls off proportional to based on which we estimate the probability of observing a by chance to be . Given the pre-computed pattern for the k-mer and a new input DNA sequence of length , we start by translating both and its reverse-complement into the numerical representations and according to m:Both and are then standardized to have mean zero and variance one , and the sum of the dot-products between each of these vectors and the pre-computed pattern vector is our correlation coefficient:This correlation coefficient represents how well the input DNA sequence pattern matches the patterns for both the k-mer and its reverse complement . Given a set of positive and negative examples described by feature vectors of length , a support vector machine ( SVM ) learns an optimal discriminant function defined by a weight vector such that the dot-product will “best” separate the positive examples ( with ) from the negative examples ( with ) . If the positive examples cannot be separated from the negative examples by a hyperplane in the high-dimensional feature space , will define the hyperplane that minimizes the misclassification costs [30] . ROC curves provide a means of evaluating a binary classifier . A set of positive examples and a set of negative examples are assigned scores , and the examples are then ordered from highest score down to lowest . The ROC curve is generated by varying a threshold from the minimum score up to the maximum score , and at each step computing the true positive rate ( the fraction of positives scoring above the threshold ) and the false positive rate ( the fraction of negatives scoring below the threshold ) . A perfect classifier will result in a line from ( 0 , 0 ) up to ( 0 , 1 ) and across to ( 1 , 1 ) , and an area “under” the curve ( AUC ) of 1 . 0 , while a random classifier will result in a diagonal line from ( 0 , 0 ) to ( 1 , 1 ) , and an AUC of 0 . 5 . For our classification tasks , the negative examples were defined to be at a distance of 110 bp on either side of each positive example , resulting in twice as many negative examples as positive examples . When evaluating the Field and Kaplan models , start probabilities were converted to dyad probabilities by shifting the predictions by 73 bp . The raw binding scores were locally averaged over a window of width 9 , and the start probabilities were locally averaged over a window of width 41—these window sizes were chosen to optimize the area under the ROC curve . Our model predictions were locally averaged over a window of width 11 . For the “25% repeat” and “75% repeat” curves in Figure 9 , dyads were partitioned into sets according to the fraction of bases annotated as repetitive by RepeatMasker [51] within a 200 bp window centered at the dyad . For the “25% repeat” curve , only those dyads with fewer than 50 out of 200 bp marked as repetitive were considered to compute the inter-dyad distance histogram . Similarly , for the “75% repeat” curve , only those dyads with more than 150 out of 200 bp marked as repetitive were considered . There are over 340 , 000 AluSx elements annotated by RepeatMasker [51] , with lengths generally between 291 and 313 . In order to perfectly align a large set of AluSx sequences with flanking regions , we used only the 4 , 930 AluSx sequences of length 313 with no insertions or deletions . We then extracted 901 bp of sequence surrounding the mid-point of each of these AluSx sequences and computed the dyad score along each of these sequences . The average of these dyad score traces is shown in Figure 10 . | DNA in eukaryotes is packaged into a chromatin complex , the most basic element of which is the nucleosome . The precise positioning of the nucleosome cores allows for selective access to the DNA , and the mechanisms that control this positioning are important pieces of the gene expression puzzle . In this work , we describe a large-scale DNA sequence pattern that jointly characterizes the sequence preferences of the nucleosome core and the adjacent linkers . We show that this pattern can be used to predict nucleosome positions in both H . sapiens and S . cerevisiae more accurately than previously published methods . The model is most accurate in predicting the most stably positioned nucleosomes , and describes a sequence composition pattern that determines a locally optimal dyad ( nucleosomal DNA mid-point ) position . In contrast to some previous models , this model is not based primarily on excluding poly-A/T sequences , nor does the model prefer 10 bp periodicity . Our results suggest that local sequence composition is one of many factors that direct the positioning of nucleosomes , while dynamic processes such as transcriptional elongation and the actions of chromatin remodeling complexes also play a significant role in the overall chromatin landscape . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"molecular",
"biology/bioinformatics",
"molecular",
"biology/chromatin",
"structure",
"computational",
"biology/genomics",
"computational",
"biology"
] | 2010 | Learning a Weighted Sequence Model of the Nucleosome Core and Linker Yields More Accurate Predictions in Saccharomyces cerevisiae and Homo sapiens |
Cholinergic neurons in the striatum are thought to play major regulatory functions in motor behaviour and reward . These neurons express two vesicular transporters that can load either acetylcholine or glutamate into synaptic vesicles . Consequently cholinergic neurons can release both neurotransmitters , making it difficult to discern their individual contributions for the regulation of striatal functions . Here we have dissected the specific roles of acetylcholine release for striatal-dependent behaviour in mice by selective elimination of the vesicular acetylcholine transporter ( VAChT ) from striatal cholinergic neurons . Analysis of several behavioural parameters indicates that elimination of VAChT had only marginal consequences in striatum-related tasks and did not affect spontaneous locomotion , cocaine-induced hyperactivity , or its reward properties . However , dopaminergic sensitivity of medium spiny neurons ( MSN ) and the behavioural outputs in response to direct dopaminergic agonists were enhanced , likely due to increased expression/function of dopamine receptors in the striatum . These observations indicate that previous functions attributed to striatal cholinergic neurons in spontaneous locomotor activity and in the rewarding responses to cocaine are mediated by glutamate and not by acetylcholine release . Our experiments demonstrate how one population of neurons can use two distinct neurotransmitters to differentially regulate a given circuitry . The data also raise the possibility of using VAChT as a target to boost dopaminergic function and decrease high striatal cholinergic activity , common neurochemical alterations in individuals affected with Parkinson's disease .
The striatum is the major input gateway to the basal ganglia . Striatal activity plays important roles in controlling motor functions and goal-directed and reward-related behaviours [1]–[4] . The striatum is the brain region mostly affected in motor diseases , such as Parkinson's disease ( PD ) , Huntington's disease , and dystonia [5] . Medium spiny GABAergic neurons ( MSN ) , activated by corticostriatal glutamatergic inputs , are the major output neurons for the striatum; these neurons are regulated extensively by the classical neurotransmitters dopamine and acetylcholine ( ACh ) [1] , [2] , [4] , [6] . These two neurotransmitters have reciprocal relationships , regulating each other's release at different levels , and they generally have opposing actions in the direct and indirect striatal pathways [1] , [5] , [7]–[9] . Regulation of MSNs by dopamine has received considerable attention , largely due to the well-known effects of reduced dopamine levels leading to motor symptoms in PD [10] and the role of dopamine in the effect of drugs of abuse [11] . In contrast to the widely known effects of dopamine in the striatum , we know considerably less about how ACh shapes striatal function . Cholinergic neurons form a small population of aspiny and large striatal interneurons that provide the sole source of cholinergic innervation to MSNs [12] , [13] . These neurons fire constantly and therefore ensure relatively high levels of extracellular ACh . To maintain high levels of transmitter release , cholinergic neurons transport ACh synthesized in the cytoplasm into synaptic vesicles , a process which requires the activity of the vesicular acetylcholine transporter ( SLC18A3 , VAChT [14] , [15] ) , the last cholinergic-specific step for ACh-mediated neurotransmission [16] . A variety of muscarinic receptors [17] , as well as nicotinic subtypes of receptors [18]–[21] , involved in controlling striatal function add complexity to unravelling the role of endogenous ACh in the striatum . To make matters more difficult , several central cholinergic neurons express both VAChT and distinct vesicular glutamate transporters ( VGLUTs ) and thus are able to store and release both ACh and glutamate [22] . Striatal cholinergic neurons express VGLUT3 [23]–[26] and simultaneously release glutamate and ACh [27] . It is unknown , however , if cholinergic neurons can use both neurotransmitters to regulate striatal function . Elimination of cholinergic neurons in the striatum , using ablation strategies , indicated that these neurons have a role in regulating spontaneous and cocaine-induced locomotor activity , as well as its rewarding properties [28]–[31] . These neurons have the capacity to release both ACh and glutamate; therefore , non-selective manipulations of striatal cholinergic neurons can affect both VAChT and VGLUT-mediated neurotransmission . Interestingly , mice null for VGLUT3 phenocopy many of the behavioural alterations found in mice that had their accumbens cholinergic neurons ablated [25] . However , because VGLUT3-null mice also presented a 40% decrease on acetylcholine release , it is difficult to discern the individual effects of these two neurotransmitters . Therefore , the specific roles of ACh for striatal function have not yet been addressed . To investigate the possibility that cholinergic neurons can use these two distinct neurotransmitters differentially to regulate striatal circuitry , we generated a novel mouse line in which we selectively eliminated ACh release by deleting the VAChT gene in the striatum . Our results reveal specific roles for ACh release in regulating dopamine receptor-mediated locomotor responses , but suggest that some of the previous functions attributed to these neurons are related to their ability to release glutamate .
To address specific roles of ACh release in striatal function we generated a VAChT floxed mouse line ( VAChTflox/flox , [32] ) , as constitutive VAChT knockout mice do not survive birth due to impaired breathing [16] . The addition of lox P sites did not change VAChT expression at the mRNA and protein levels when compared to wild-type control mice . VAChTflox/flox mice had normal levels of VAChT and other pre-synaptic cholinergic markers . In addition locomotor activity , grip-strength , and fatigue were identical in VAChTflox/flox mice and wild-type mice [32] . In order to selectively eliminate VAChT in the striatum , we used the D2-Cre bacterial artificial chromosome ( BAC ) transgenic mouse line generated by GENSAT [33] , which expresses the enzyme Cre recombinase under the control of regulatory elements of the D2 dopamine receptor ( D2R ) . Details related to this mouse line , including control experiments demonstrating that the expression of Cre has no effects on the parameters studied here , are presented in Experimental Procedures and Figure S6 . To test whether Cre was expressed in striatal cholinergic neurons , we crossed D2-Cre mice to Rosa26 reporter mice ( Rosa26-YFP mice ) , in which the Rosa26 locus expresses YFP once Cre-mediated recombination has occurred ( Figure 1a ) . We found that in D2-Cre;Rosa26-YFP mice almost 100% of striatal cholinergic neurons identified with an antibody against CHT1 also showed Cre-recombination ( YFP staining 98% co-localization , Table S1 ) . We did not detect co-localization of YFP in cholinergic neurons in the penduculopontine nucleus or in motoneurons in the brainstem ( Figure S1 and Table S1 ) . Partial localization of YFP in cholinergic neurons was detected in the basal forebrain , albeit to a much lower extent than in the striatum ( approx . 50% , Figure 1b and Table S1 ) . We therefore intercrossed D2-Cre mice to VAChTflox/flox mice to generate mice with selective elimination of VAChT in the striatum ( VAChTD2-Cre-flox/flox ) or control mice ( VAChTflox/flox ) . Genotyping for these lines is shown in Figure S2 . VAChTD2-Cre-flox/flox mice were born in the expected Mendellian ratio and did not present overt phenotypes . We found no gross morphological alterations in the striatum or other brain sections stained with hematoxylin/eosin in VAChTD2-Cre-flox/flox mice compared to control mice ( unpublished data ) . To assess the degree of Cre-mediated recombination we evaluated the expression of VAChT in the striatum of VAChTD2-Cre-flox/flox . As expected , based on the observations with the D2-Cre;Rosa26-YFP mice , both mRNA and protein levels for VAChT were almost abolished in the striatum of VAChTD2-Cre-flox/flox ( Figure 2a , d , g ) . In contrast , choline acetyltransferase ( ChAT ) and the high-affinity choline transporter ( CHT1 ) protein levels were not altered ( Figure 2e and f ) . There was no difference in VAChT protein expression levels in the hippocampus of VAChTD2-Cre-flox/flox mice when compared to controls ( Figure 2h and i ) . Accordingly , release of [3H]-ACh was abolished in striatal slices from VAChTD2-Cre-flox/flox mice depolarized with high KCl , whereas it was identical to controls in hippocampal slices ( Figure 3a and b ) . Acetylcholine can modulate glutamate release via pre-synaptic nicotinic receptors in projection glutamatergic nerve-terminals [34] . In addition , striatal cholinergic neurons can also release glutamate [27] . Therefore , we examined if there was any effect of VAChT elimination on glutamate release . Isolated nerve terminals were obtained from striatal tissue of VAChTD2-Cre-flox/flox and control mice and glutamate release was stimulated by KCl . We did not detect changes in glutamate release from isolated nerve terminals in VAChT-deficient mice compared to controls ( Figure 3c ) . It should be noted , however , that this method does not separate terminals containing VGLUT3 from nerve terminals containing other VGLUTs , and therefore only reflects global changes in glutamate release . Moreover , VGLUT3 mRNA expression by qPCR did not differ in VAChTD2-Cre-flox/flox mice compared to control mice ( Figure 3d ) . These results suggest that overall glutamate release is not grossly altered in these mice . Because we detected the presence of Cre-mediated recombination in motoneurons in the spinal cord ( Figure S1 ) , which could affect the behavioural performance in VAChTD2-Cre-flox/flox mice , we examined the cholinergic system in the spinal cord of VAChTD2-Cre-flox/flox mice . We did not find alterations in mRNA levels for VAChT in the spinal cord of VAChTD2-Cre-flox/flox mice ( Figure S3 ) . However , we detected an increase in ChAT mRNA and protein levels in the spinal cord . Surprisingly , there was also about a 50% decrease in VAChT protein levels . Previous experiments showed that up to a 50% decrease in the expression of VAChT in the spinal cord is well tolerated in mice and does not alter motor function [35] , [36] . In agreement with these previous results , VAChTD2-Cre-flox/flox mice showed no difference in grip-force strength ( Figure S4a , t ( 47 ) = 1 . 702 , p = 0 . 095 ) or fatigue ( detected by the Wire-hang task , Figure S4b , Mann-Whitney , T ( 13 ) = 49 , p = 0 . 710 ) . Interestingly , we also found that relative to controls , VAChTD2-Cre-flox/flox mice showed no deficit in motor performance or motor learning assessed using the rotarod test ( Figure S4c , Repeated Measures ANOVA reveal no difference between the two genotypes with respect to time to fall , F ( 1 , 261 ) = 0 . 0000409 , p = 0 . 995; both sets of mice improved their performance , F ( 9 , 261 ) = 41 . 614 , p<0 . 001; and there was no interaction between genotype and session , F ( 9 , 261 ) = 1 . 333 , p = 0 . 220 ) . These results show that despite a decrease in the levels of VAChT in the spinal cord there were no detectable changes in motor function . The rotarod experiments also suggest that VAChTD2-Cre-flox/flox mice are physically fit and that motor learning does not appear to depend on striatal cholinergic activity . Next , as a further control experiment , we determined if the elimination of ACh release in the striatum could interfere with cognitive performance that is believed to be generally independent of striatal function . We used object recognition memory , a task that is thought to be dependent on the hippocampus [37] , [38] and perirhinal cortex [39] , and has been previously shown to be sensitive to global decreases in VAChT levels [35] , [36] , [40] . In this test VAChTD2-Cre-flox/flox mice performed identically to controls , suggesting that important cognitive functions are preserved in this new mouse line ( Figure S4d , two-way ANOVA revealed no effect of genotype , F ( 1 , 16 ) = 0 . 651 , p = 0 . 431 , a significant effect for object , F ( 1 , 16 ) = 21 . 559 , p<0 . 001 and no Object × Genotype interaction , F ( 1 , 16 ) = 0 . 0185 , p = 0 . 893 ) . Previous experiments have shown that the density of cholinergic neurons in the accumbens , as well as expression of ChAT , is decreased in the post-mortem brain of schizophrenic individuals [41] , [42] . Moreover , partial ablation of cholinergic striatal neurons caused alterations in sensorimotor gating [43] . Therefore , we used habituation to acoustic startle and pre-pulse inhibition to assess sensorimotor gating , but found no effects of elimination of striatal VAChT on these parameters ( Figure S5 ) . These results show that decreased striatal ACh release does not cause sensorimotor gating dysfunctions in these animals and likely in schizophrenia as well . There are controversial views regarding the role of striatal cholinergic neurons in locomotion . Previous experiments in which cholinergic neurons in the nucleus accumbens were ablated indicated that loss of these neurons caused hyperlocomotion and increased sensitivity to the locomotor effects of cocaine [29]–[31] . However , more recent experiments using an optogenetics approach failed to detect an increased locomotor activity in mice in which striatal cholinergic neurons were acutely silenced [44] . In agreement with the latter , we found no differences in locomotor activity when we compared VAChTD2-Cre-flox/flox mice to controls ( Figure 4a ) . The dynamics of total horizontal activity ( Figure 4a and 4b , t ( 48 ) = 0 . 1464; p = 0 . 884 ) or counts of vertical activity ( unpublished data , t ( 24 ) = 1 . 027; p = 0 . 315 ) were essentially identical in the two strains . Importantly , in control experiments D2-Cre mice did not differ in locomotor activity from respective wild type mice ( Figure S6 ) . It has been shown that VGLUT3-null mice present hyperactivity , which was attributed to decreased ACh release from striatal cholinergic neurons due to decreased filling of synaptic vesicles with ACh [25] . Because these experiments with VGLUT3-null mice were performed in the initial hours of the dark cycle , we reproduced these conditions with a new cohort of our mice . The VAChTD2-Cre-flox/flox mice were no more active than their control counterparts during the first hours of the dark cycle ( Figure 4c and d , repeated measures ANOVA shows no main effect of genotype , F ( 1 , 1593 ) = 0 . 321 , p = 0 . 576 , significant effect of time , F ( 59 , 1593 ) = 14 . 411 , p<0 . 001 and no interaction Genotype × Time , F ( 59 , 1593 ) = 0 . 947 , p = 0 . 591; total activity was not different , Mann-Whitney , T ( 29 ) = 213 , p = 0 . 431 ) . Finally , we also tested inter-session habituation by investigating locomotor activity in 3 consecutive days in the open-field ( Figure 4e ) . We observed that both genotypes habituated similarly to the open-field . Repeated measures ANOVA confirmed that the general activity was the same for both genotypes ( genotype factor , F ( 1 , 58 ) = 0 . 932 , p = 0 . 342 ) . The activity decreased over the day ( day factor , F ( 2 , 58 ) = 10 . 244 , p<0 . 001 ) and both genotypes habituated to the environment at comparable rates ( interaction between genotype and day , F ( 2 , 58 ) = 1 . 506 , p = 0 . 230 ) . Evidently , deletion of VAChT in the striatum does not affect general spontaneous activity or compromise the capacity to habituate to a new environment . Previous experiments in mice in which cholinergic interneurons were ablated suggested that decreased ACh levels increase sensitivity of mice to the locomotor effects of cocaine [29]–[31] . However , these experiments did not separate the effects of VAChT and VGLUT3-mediated transmission . Interestingly , VGLUT3-null mice are also more sensitive to the locomotor effects of cocaine , a result that was attributed at least in part to a decrease in striatal ACh release [25] . Due to the surprising observations of normal locomotor activity in VAChT-deficient mice , we investigated the specific effects of the elimination of VAChT-mediated neurotransmission on the actions of cocaine . Administration of 5 , 20 , or 40 mg/kg of cocaine increased locomotor activity in VAChTflox/flox mice and VAChTD2-Cre-flox/flox mice ( Figure 5c , two-factor ANOVAs show a significant effect of genotype , F ( 1 , 51 ) = 6 . 531 , p = 0 . 014 , significant effect of treatment , F ( 3 , 51 ) = 15 . 611 , p<0 . 001 , and no Genotype × Treatment interaction , F ( 3 , 51 ) = 0 . 983 , p = 0 . 381 ) . There was no difference between the two genotypes in their ability to increase activity in response to cocaine-injected i . p . at 5 mg/kg dose ( Figure 5a , 5 mg/kg , repeated measures ANOVAs show no effect of genotype , F ( 1 , 322 ) = 0 . 201 , p = 0 . 661 , significant effect of time , F ( 23 , 322 ) = 12 . 820 , p<0 . 001 , and no Time × Genotype interaction , F ( 23 , 322 ) = 1 . 373 , p = 0 . 121 ) . Paradoxically , at 20 mg/kg VAChTD2-Cre-flox/flox mice showed a smaller effect of cocaine in locomotor activity than controls ( Figure 5b , 20 mg/kg , repeated measures ANOVA shows significant effect of genotype , F ( 1 , 480 ) = 11 . 345 , p<0 . 001 , significant effect of time , F ( 23 , 480 ) = 9 . 464 , p<0 . 001 , and no Time × Genotype interaction , F ( 23 , 480 ) = 0 . 945 , p = 0 . 537 ) . Analysis of total activity counts showed a clear effect of genotype ( Figure 5c , Mann-Whitney , T ( 23 ) = 166 . 000 , p<0 . 05 ) . At 40 mg/kg both genotypes showed similar responses ( Figure 5c , t ( 14 ) = 0 . 980 , p = 0 . 344 ) , suggesting that lack of striatal VAChT altered the response to 20 mg/kg of cocaine , but overall did not cause increased sensitivity to locomotor effects of cocaine . Cocaine increases firing of striatal cholinergic neurons [44] and the release of ACh in the striatum [45]–[47] . Previous experiments have suggested that striatal cholinergic neurons also play important roles in the rewarding effects of cocaine . Indeed , optogenetic silencing of striatal cholinergic neurons seemed to attenuate the response of cocaine in a conditioned-place preference ( CPP ) paradigm . Because these experiments did not separate the contribution of ACh from that of glutamate and to determine if there was a causal link between ACh release and expression of cocaine-induced CPP , we performed CPP experiments with VAChTD2-Cre-flox/flox mice . We were unable to obtain reliable CPP with either genotype at 5 mg/kg of cocaine ( unpublished data ) . In contrast , at 20 mg/kg we detected robust CPP in both genotypes ( Figure 6a , repeated measures ANOVAs show no effect of genotype , F ( 1 , 10 ) = 0 . 443 , p = 0 . 521 , significant effect of treatment , F ( 1 , 10 ) = 86 . 033 , p<0 . 001 , and no Genotype × Treatment interaction , F ( 1 , 10 ) = 0 . 0118 , p = 0 . 916 ) . In these experiments we used an extended protocol [48] with consecutive injections of cocaine in alternate days . We repeated the short protocol used before in the optogenetic experiments [44] with only one injection of cocaine ( 20 mg/kg ) , but we were unable to detect place preference in control or VAChTD2-Cre-flox/flox mice ( unpublished data ) . In addition , neither extinction of CPP nor relapse , measured as a reinstatement of CPP by a priming injection of cocaine after extinction , were altered in mice without striatal VAChT ( Figure 6b , repeated measures ANOVAs show no effect of genotype , F ( 1 , 7 ) = 0 . 00057 , p = 0 . 982 , significant effect of treatment , F ( 1 , 7 ) = 7 . 457 , p = 0 . 029 , and no Genotype × Treatment interaction , F ( 1 , 7 ) = 9 . 67×10−5 , p = 0 . 992 ) . Therefore , there was no difference in CPP response for the two genotypes . Behavioural sensitization protocols for cocaine likely reflect altered synaptic plasticity in response to the drug [49] , which manifests as an increase in the locomotor effects of cocaine . In a separate group of mice , we measured behavioural sensitization to 10 mg/kg of cocaine ( Figure 7 ) and found that repeated treatment with this dose of cocaine seems to cause slightly higher locomotor activity in VAChTD2-Cre-flox/flox mice , but the relative increase in behavioural sensitization was not different between genotypes ( Figure 7a , b , repeated measures ANOVAs show a significant effect of genotype , F ( 1 , 16 ) = 4 . 902 , p = 0 . 042 , significant effect of treatment , F ( 1 , 16 ) = 33 . 855 , p<0 . 001 , and no Genotype × Treatment interaction , F ( 1 , 16 ) = 0 . 496 , p = 0 . 491 ) . Thus , elimination of striatal ACh release caused a small change in the dose-response profile of cocaine-treated mice in intermediate doses: a slight increase in activity is observed at 10 mg/kg , whereas a decrease in locomotor response is observed at 20 mg/kg in mutant mice . The balance between acetylcholine-dopamine is important in a number of conditions , including PD; therefore we further investigated dopaminergic function in VAChTD2-Cre-flox/flox mice . For that , we first determined the concentration of dopamine and metabolites in the striatum of VAChTD2-Cre-flox/flox mice and compared these to control mice . In general there were no major changes in dopamine and metabolites in these mutant mice ( Table 1 ) . However , the ratio between dopamine and DOPAC as well as dopamine and HVA were significantly changed , showing that dopamine turnover is decreased by 25% ( p<0 . 001 ) , suggesting potential relatively minor alterations in dopamine dynamics or metabolism . To further assess dopaminergic function , we performed qPCR analysis for D1R and D2R expression in the striatum . We detected an increase in the expression of D1R and D2R mRNAs in the striatum of VAChTD2-Cre-flox/flox mice compared to control mice ( Figure 8a and b , D1R , t ( 12 ) = 2 . 756 , p<0 . 05 , D2R , t ( 14 ) = 2 . 300 , p<0 . 05 ) . In contrast , D2R mRNA expression in the midbrain was not altered ( Figure 8c ) . G-protein coupled receptors ( GPCRs ) can have agonist-independent effects; hence , altered expression of such receptors could modulate behaviour even in the absence of neurotransmitter release . We thus also investigated the expression of cholinergic receptors . Figure 8d–f indicates that expression of M1 and M2 muscarinc receptors ( mAChR1 and mAChR2 ) was unchanged , whereas M4 muscarinic receptors ( mAChR4 ) showed increased expression ( t ( 12 ) = 3 . 678 , p<0 . 05 ) . Homozygous mice expressing D2-BAC-GFP construct present some dopaminergic phenotypes [50] , however control experiments show that the heterozygous D2-Cre mice used here do not present any of the phenotypes associated with selective elimination of striatal VAChT ( Figure S6 ) . Because dopamine receptor expression was normal in D2-Cre mice , we conclude that these molecular alterations are due to the loss of ACh release . To confirm the increased alteration of dopamine receptors in the striatum of VAChTD2-Cre-flox/flox mice we initially performed Western blots . Unfortunately , we were unable to obtain a reliable D1 antibody that showed specific detection of D1R ( unpublished data ) . However , we obtained a D2 antibody that labelled only one major band with the correct molecular mass ( Figure 9a ) . Quantification of immunoblots confirmed increased expression of D2R ( Figure 9a , b , t ( 14 ) = −3 . 628 , p<0 . 01 ) . In order to provide an independent measure of D1R activity and test if D1-mediated responses would be altered in VAChT-eliminated mice , we used pharmacological magnetic resonance imaging ( phMRI ) [51] , [52] . phMRI is a variant of functional magnetic resonance imaging that indirectly detects neuronal activity using blood oxygenation level-dependent ( BOLD ) MRI signal changes [53] to detect functional effects of pharmacological agents in intact systems in vivo with high temporal and spatial resolution . A 9 . 4T anatomic MRI of the mouse brain ( Figure 9c ) was used to outline regions of interest in the striatum and cortex . The average difference in BOLD effect between striatum and cortex ( Figure 9d ) indicates that there is increased neuronal activation in the striatum in VAChTD2-Cre-flox/flox mice following injection of SFK 81297 ( 3 mg/kg , Figure 9d ) . The change in the BOLD response after administration of the selective D1R agonist SKF 81297 relative to baseline ( prior to injection ) was then compared between the two genotypes . Saline administration prior to SKF 81297 did not alter BOLD signal ( unpublished data ) . In contrast , injection of SKF 81297 lead to a slow increase in striatal BOLD response ( area under the curve ) in VAChTD2-Cre-flox/flox mice compared to control mice following injection of the D1R agonist ( Figure 9d and e , p<0 . 01 ) . To test if the increased expression/sensitivity of D1R and D2R has direct behavioural consequences , we investigated the effects of the selective dopaminergic agonists SKF 81297 ( D1R agonist ) and quinpirole ( D2R agonist ) on locomotor activity . VAChTD2-Cre-flox/flox mice had significantly higher locomotor responses to two doses of SKF 81297 ( Figure 10a and b , two-factor ANOVAs revealed significant effect of genotype , F ( 1 , 66 ) = 11 . 654 , p<0 . 01 , significant effect of drug concentration , F ( 3 , 66 ) = 34 . 476 , p<0 . 001 , and significant Drug Concentration × Genotype interaction , F ( 3 , 66 ) = 4 . 277 , p<0 . 01 , Tukey post hoc test showed significant differences with SKF 81297 doses of 3 mg/kg ( p<0 . 01 ) and 8 mg/kg ( p<0 . 001 ) ) . Moreover , VAChTD2-Cre-flox/flox mice also showed enhanced inhibition of locomotion in response to low doses of the D2R-selective agonist quinpirole ( Figure 10c and d , ANOVA showed significant effect of genotype , F ( 1 , 111 ) = 12 . 543 , p<0 . 001 , significant effect of drug , F ( 4 , 111 ) = 42 . 223 , p<0 . 001 , but the interaction was not significant , F ( 4 , 111 ) = 2 . 052 , p = 0 . 092 ) . Analysis of locomotion in response to the individual doses showed a significant difference for 0 . 005 and for 0 . 01 mg/kg quinpirole dose ( p<0 . 01 ) . Taken together , these data reveal important alterations in the expression and function of striatal dopamine receptors in VAChTD2-Cre-flox/flox mice .
Our studies in VAChTD2-cre-flox/flox mice indicated that elimination of ACh release in the striatum does not seem to play a major role in motor function and motor learning , at least for acrobatic motor skills in the rotarod test . This observation is also in agreement with previous experiments in striatal cholinergic neuron-ablated mice that presented no deficiency in rotarod performance [31] . However , we cannot completely exclude more subtle effects of ACh in fine motor tuning and motor tasks . For example , the chronic nature of elimination of ACh release in our experiments may lead to adaptations in motor behaviour . Future experiments using VAChTD2-Cre-flox/flox mice and more sophisticated motor behavioural tests may be necessary to pinpoint possible roles for striatal ACh in motor learning and performance . There are multiple lines of evidence that pharmacological modulation of cholinergic receptors regulates locomotor activity . It is known that muscarinic antagonists increase locomotor activity and M1 and M4 muscarinic receptor KO mice are hyperactive [54]–[57] . Moreover , we have recently observed that mice with a significant decrease in VAChT expression in the whole forebrain show hyperactivity [32] . The present work provides compelling evidence for more selective roles of the neurotransmitter ACh in the striatum , indicating that decreased striatal expression of VAChT does not cause overt motor consequences . These results may be of particular importance , since there have been reports that in Huntington's disease VAChT levels are decreased in the striatum [58] . Our data suggest , however , that this alteration is unlikely to contribute to gross motor symptoms observed in Huntington's disease . Cholinergic neurotransmission in brain regions other than the striatum may still play a role in control of locomotion . Previous attempts to assess the function of cholinergic neurons in the striatum were performed following the ablation of cholinergic neurons using immunotoxin-mediated cell targeting . Injection of toxin targeting transgenic cholinergic neuron in the accumbens led to an 80% decrease in ChAT-positive neurons [30] . Elimination of cholinergic neurons in the accumbens by this means inhibited certain forms of reward-related learning; however , it also induced hyperactivity and increased sensitivity to the locomotor and the rewarding effects of cocaine , including increased sensitivity in the CPP test to low doses of cocaine [28] , [29] , [31] . In contrast , recent experiments using an optogenetic approach to inactivate or activate cholinergic neurons in the accumbens found no effects of inactivation of these neurons on locomotor activity , albeit their silencing prevented the response to cocaine in a CPP test [44] . Thus , elimination of cholinergic neurons in the accumbens seemed to increase sensitivity to cocaine-induced CPP [29] , whereas optogenetics silencing of these neurons blocked cocaine-induced CPP [44] . The reason for the different outcome in these two experiments is not entirely clear at the moment , but could be related to the chronic versus acute nature of the manipulations . Although in our experiments we have targeted the whole striatum , rather than only the accumbens , we did not detect major alterations in cocaine-induced CPP , suggesting that the above effects obtained with neuronal ablation or by optogenetics manipulation may be linked not to loss of cholinergic transmission per se but rather to suppression of glutamate release from cholinergic neurons . While an optogenetic approach provides a novel paradigm to acutely activate or inactivate populations of neurons , it is unlikely that this method can separate VAChT from VGLUT3-dependent neurotransmission as selectively as that which can be achieved using VAChTD2-Cre-flox/flox mice . Interestingly , recent data have shown that cholinergic neurons in the habenula secrete both ACh and glutamate ( mediated by VGLUT1 ) , and release of either of these neurotransmitters appears to depend on the frequency of stimulation [22] . Basal forebrain neurons in culture release both ACh and glutamate [59] . Importantly , recent work shows that optogenetics stimulation of striatal cholinergic neurons can evoke synaptic glutamatergic neurotransmission onto MSNs , with predominant activity over NMDA receptors [27] . The co-release of glutamate with dopamine has also been described [60] , [61] , suggesting that interpretation of the roles of dopaminergic neurons will also need to take into account glutamate co-release . Therefore , the co-release of glutamate with classical neurotransmitters may be a more common mechanism than previously appreciated and may have a broad impact in circuitry control . However , we cannot discard the possibility that other neuromodulators released from cholinergic neurons , such as ATP or peptides , could also play a role as co-transmitters . The role of VGLUT3 in striatal function is far from being fully understood [62] . Interestingly , with respect to striatum-related behaviour , VGLUT3-null mice show hyperactivity and increased response to the locomotor effects of cocaine [25] . Therefore , mice lacking VGLUT3 show a phenotype that is remarkably similar to that of mice in which cholinergic neurons in the accumbens were targeted by an immunotoxin [29] , [30] . Experiments in VGLUT3-null mice suggested that the absence of VGLUT3 causes a decrease in striatal cholinergic tone . VGLUT3 is used by the striatal vesicles to facilitate VAChT-mediated ACh storage in synaptic vesicles [25] , [62] . However , measurements of ACh release in VGLUT3-null mice have indicated only a modest reduction , by 30% to 40% [25] , compared to almost 100% inhibition in VAChTD2-Cre-flox/flox mice . It is unlikely that 40% reduction in ACh release observed in VGLUT3-null mice can be responsible for the hyperactive phenotype . Indeed , independent mouse lines with a 50% decrease in VAChT expression , and concomitant reduction of ACh release [16] , [36] , [63] , did not present increased locomotor activity in the open field [16] , [35] . We conclude that the locomotor phenotypes observed previously in striatal cholinergic neuron-ablated mice [29] , [31] and in VGLUT3-null mice [25] are either a consequence of the disruption of VGLUT3-mediated neurotransmission or the combination of reducing both glutamatergic and cholinergic activity simultaneously from these neurons . Future experiments using VAChTD2-Cre-flox/flox mice , VGLUT3 floxed mice , and double knockouts will be necessary to provide an assessment of independent effects of VGLUT3-mediated neurotransmission in the striatum . Although we have focused on striatal-related behaviours , the extent by which alterations in VAChT expression in other brain regions in VAChTD2-cre-flox/flox mice may contribute to these phenotypes should also be taken into account . We did not detect Cre-expression in cholinergic neurons in the penduculopontine area , for example ( Figure S2 ) , which harbours groups of cholinergic neurons that project to the midbrain and thalamus and could influence striatal function . However , we cannot completely exclude the possibility that cholinergic neurons in other brain regions would not be targeted in our mouse line . At the same time , as the phenotypes described here seem to be mainly striatal specific and cholinergic interneurons provide the almost exclusive source of cholinergic tone in the striatum , it is unlikely that other groups of cholinergic neurons would have contributed to the observed behaviours . Elimination of cholinergic neurotransmission in the striatum did not cause hyperlocomotion , however the responses to direct activation of dopamine receptors were substantially increased . Both behavioural and phMRI analysis indicated an increased response to D1R agonist . Western blot analysis also showed selective increase of D2R expression in the striatum . Moreover , in addition to the increased D2R levels in the striatum , which likely reflect a combination of pre- and post-synaptic receptors , we also uncovered increased D2-like receptor pre-synaptic activity , revealed by the increased sensitivity of VAChTD2-Cre-flox/flox mice to low doses of quinpirole . Certainly , we cannot rule out that changes at the level of receptors play a more complex role in regulating locomotor activity in VAChTD2-Cre-flox/flox mice . Indeed , GPCRs may have agonist-independent activity [64] , [65] . The locomotor effects of cocaine seem to depend mainly on inhibition of the dopamine transporter [66] . However , acetylcholine can affect release of dopamine via distinct nicotinic receptors [19] , as well as regulate both dopamine release and activity of MSNs , via distinct muscarinic receptors [56] , [57] , [67] . The fact that both D1R and D2R had increased expression in the striatum would suggest that VAChTD2-Cre-flox/flox mice should be more responsive to dopamine and might present increased spontaneous locomotor activity or cocaine-induced locomotion or CPP . However , this was not the case . It is likely that cell-autonomous compensatory mechanisms related to disrupted cholinergic function significantly altered striatal circuitry , preventing such a simple relationship . For example , because M4 muscarinic receptors seem to specifically regulate D1R-mediated signalling [56] , [57] , [68] , it is possible that the increased expression of M4 receptors we detected in the striatum could counterbalance D1R-mediated responses in vivo , leading to unaltered locomotor activity . Moreover , because D2-like pre-synaptic receptors may be more active in VAChTD2-Cre-flox/flox mice , elimination of ACh release in the striatum may also affect pre-synaptic control of dopamine release . The slightly decreased turnover of dopamine in mice without striatal VAChT supports the notion of direct consequences of reduced cholinergic tone at the level of dopaminergic terminals . Thus , behavioural analysis of VAChTD2-Cre-flox/flox mice indicates that control of locomotor function and response to cocaine mediated by dopamine might become more complex in the absence of cholinergic tone . Future experiments will be needed to evaluate direct consequences of elimination of either acetylcholine or glutamate neurotransmission originating from striatal cholinergic neurons on dopamine transmission . The present data provide direct and indirect evidence that striatal cholinergic neurons can use two different neurotransmitters to regulate striatal function . Hence , re-evaluation of previously attributed functions of striatal cholinergic tone is warranted . The data indicate that VGLUT3-mediated glutamatergic neurotransmission originating from cholinergic neurons may have greater influence on striatal function than previously envisioned . The behavioural consequences of selective elimination of VAChT , and thus cholinergic transmission , in the striatum are remarkably minimal , at least for the locomotion control by the striatal complex . One intriguing phenotype uncovered in mutant mice is an increase in dopamine receptors' expression and function without major alterations in cocaine-induced behaviours . Our experiments provide evidence that targeting VAChT in the striatum can up-regulate dopamine receptors and thus could be used in conditions of dopamine deficiency and abnormally increased cholinergic activity , as found in individuals with PD .
The isolation of a VAChT genomic clone has been described previously [36] . The genomic clone was used to construct a gene-targeting vector in which we added LoxP sequences flanking the VAChT open reading frame and a TK-Neo cassette . Generation of VAChTflox/flox mice is described elsewhere [32] , and the construct is shown in Figure S2 . Briefly , after removal of the TK-Neo cassette , one LoxP sequence was present 260 bp upstream from the VAChT translational initiation codon , and a second LoxP sequence was located approximately 1 . 5 kb downstream from the VAChT stop codon and within the second ChAT intron . Note that this vector is distinct from that previously used for generation of VAChT KD mice [36] . D2-Cre mice ( Drd2 , Line ER44 ) were obtained from the GENSAT project via the mutant mouse regional resource centers . VAChTD2-Cre-flox/flox mice were generated by crossing VAChTflox/flox with the D2-Cre mouse line . We then inter-crossed VAChTD2-Cre-flox/wt to obtain VAChTD2-Cre-flox/flox mice . Because these mice were apparently normal and fertile , we bred VAChTD2-Cre-flox/flox mice and VAChTflox/flox to obtain all the mice used in the present study . These mice were backcrossed to C57BL/6J mice for five generations . Unless otherwise stated , all control mice used were VAChTflox/flox littermate mice without the Cre transgene . After the completion of this work we were made aware that the BAC used to generate D2-Cre mice carried an extra gene , ttc2 , and a recent report suggests that homozygous D2-GFP mice , generated using the same BAC construct , are hyperactive and show a number of dopamine-related phenotypes [50] . However , as these authors point out , their experiments cannot discern if the phenotypes uncovered are due to the BAC positioning insertion or to the extra copy of ttc2 . We confirmed that the D2-Cre indeed have increased expression of the TTC2 mRNA ( unpublished data ) . However , heterozygous D2-Cre mice showed no locomotor phenotype . Moreover , these mice showed normal levels of D1R , D2R , and M4-muscarinic receptors ( Figure S6 ) . Hence , neither the phenotypes nor the molecular changes observed in VAChTD2-Cre-flox/flox mice are due to the BAC transgene . Rosa26-YFP mice ( B6 . 129X1-Gt ( ROSA ) 26Sortm1 ( EYFP ) Cos/J , stock number 006148 ) were obtained from Jackson Laboratories . Animals were housed in groups of three to four mice per cage without environment enrichment in a temperature-controlled room with 12-h light–12-h dark cycles , and food and water were provided ad libitum . Mouse stocks were SPF , however experimental subjects were kept in a conventional mouse facility . All studies were conducted in accordance with the NIH and the Canadian Council of Animal Care ( CCAC ) guidelines for the care and use of animals with approved animal protocol from the Institutional Animal Care and Use Committees at the University of Western Ontario ( protocol number 2008–089 ) . Only male mice were used for the behavioural studies , and they were at least 12 weeks old . Mice were randomly assigned to distinct experimental groups . Only mice used for evaluation of spontaneous locomotor behaviour were used in other tasks . For the immunofluorescence experiments we followed a protocol previously described [35] , [69] . For mRNA analyses tissue samples were frozen in a mixture of dry ice/ethanol and kept at −80°C until used as described [70] . Immunoblotting was performed as described elsewhere [36] , [71] , [72] . Slices were obtained from the striatum and hippocampus of control and test mice , labelled with [3H]methyl-choline , and the release of labelled ACh was determined essentially as described [73] except that 33 mM KCl was used as a depolarizing stimuli . Striatal synaptosomes were prepared by the method of [74] , [75] as previously described [76] . Glutamate release was followed continuously using a fluorimetric method [77] exactly as previously described [78] . All behavioural experiments were performed between 9 a . m . and 4 p . m . in the light cycle , essentially as previously described [16] , [35] except the spontaneous activity of the first hours of the dark cycle was done from 7 p . m . to 10 p . m . The dissected brain tissues were homogenized in 0 . 2 M perchloric acid with 100 µM EDTA-2Na . Samples were spun in a microcentrifuge at 12 , 000 rpm for 15 min at 4°C . Samples of the supernatant were then analyzed for norepinephrine ( NE ) , dopamine ( DA ) , and its two metabolites , 3 , 4-dihydroxyphenylacetic acid ( DOPAC ) and homovanillic acid ( HVA ) by NoAb BioDiscoveries ( Mississauga , ON ) . The HPLC used was an Eicom EP-700 with electrochemical detection ( Eicom ECD-700 ) . To elute catecholamines from the reverse phase column ( 3 . 0×100 mm SC-3ODS column , Eicom ) , a mobile phase consisting of 0 . 1 M citric acetate buffer pH 3 . 5 with 5 mg/ml EDTA-2Na , 220 mg/L sodium octane sulfonate , and 22% methanol was used . These experiments have been described in [79] . Briefly , animals were acclimatized 3–5 times to the startle boxes ( Med Associates ) . Habituation of startle was measured using 30 startle pulses ( 20 ms , white noise , 115 dB on a 65 bd white noise background ) with an inter-trial interval of 20 s . Subsequently , prepulse inhibition was measured by displaying 50 startle stimuli with either no prepulse ( pulse alone ) , a 75 dB ( 4 ms white noise ) prepulse preceding the pulse by either 30 ms or 100 ms , or a 85 db prepulse ( 30 ms or 100 ms interval ) . Each of the five trial types were displayed 10 times in a pseudorandomized order . PPI is expressed as the average startle response to the respective prepulse trials in relation to the pulse alone trials . A Grip Strength Meter from Columbus Instruments ( Columbus , OH ) was used to measure forelimb grip strength essentially as described [36] . For the wire-hang test each mouse was placed on a metal wire-grid , which was slowly inverted and suspended 40 cm above a piece of foam as previously described [36] . The time it took for each mouse to fall from the cage top was recorded with a 60 s cut-off . The rotarod task followed a previously described protocol [36] . The CPP protocol was modified from [80] . Briefly , CPP was performed in a three chamber apparatus containing two large compartments with differences in visual and tactile cues , separated by a neutral area . In day 1 ( habituation ) , mice were placed in the central compartment and allowed free access to the entire apparatus for 30 min . The time spent in each compartment was recorded . On days 2–7 ( the conditioning phase ) , mice received alternating injections of cocaine or vehicle and were immediately confined into one of the two large compartments for 30 min . A combination of unbiased and biased allocation was used . On day 8 ( test day ) mice were once again allowed free access to all three compartments for 30 min , and the time spent in each compartment was recorded . For the CPP extinction and reinstatement , the protocol previously described [80] was followed . Behavioural sensitization was performed as described [57] . The general procedure was previously described , but for analysis we used Anymaze [35] . Mice ( VAChTD2-cre/flox/flox , N = 5; control , N = 4 ) were anesthetized with 4% isofluorane and maintained at 1 . 5% isoflurane during the MRI scanning . Two intraperitoneal ( I . P . ) catheters ( 26 gauge , Abbotcath ) were used for injection of saline and SFK . The catheters were secured in place with subcutaneous sutures . Catheters were connected to polyethelyne tubing ( PE50 , VWR , Canada ) and to syringes containing saline and SFK for remote injection during imaging . Mice were placed in a custom built frame designed to secure the skull and minimize respiration induced movement during image acquisition . Mice were imaged on a 9 . 4 Tesla small animal MRI scanner ( Agilent , Palo Alto , CA ) equipped with a two-channel surface coil ( diameter = 2 cm ) . A fast low angle shot ( FLASH ) pulse sequence was used to acquire anatomical images ( field of view = 19 . 2×19 . 2 mm2 , matrix = 128×128 , repetition time = 50 ms , echo time = 11 ms , flip angle = 11° , and 10 averages ) . Respiratory gated lower resolution FLASH images were also acquired for pharmacological imaging ( field of view = 19 . 2×19 . 2 mm2 , data matrix = 64×64 , repetition time = 15 ms , echo time = 7 ms , flip angle = 11° , and 1 average ) to measure blood oxygen level–dependent ( BOLD ) signal changes . Seven contiguous axial slices ( 500 μm thick ) covered the brain . Each animal received two injections: first , an injection of 0 . 5 ml physiological saline ( 0 . 9% ) administered over a 30 s period ( control ) , and second , SFK 81297 ( 3 mg/kg ) , diluted in 0 . 5 ml physiological saline , also administered over a 30 s period ( drug ) . For the control experiment , images were acquired for 8 min prior to saline injection and then for 20–50 min following injection . For the drug experiment , images were acquired for 8 min prior to drug injection and then for 80–180 min after injection . Throughout the imaging session , body temperature and respiration rate was monitored every 10 min using the MR-Compatible , Model 1025 monitoring system ( Small Animal Instruments Inc . , Stony Brook , NY ) . Temperature was maintained at 37 . 5°C using a warm air blower , and respiration rate ranged from 45–66 ( mean 54 BPM ) . Following imaging , mice were euthanized by cervical dislocation while still under isoflurane anaesthesia . To limit the influence of global motion on the functional result , the signal intensity difference between striatum and cortex was used to examine the effect of SFK 81297 on the striatum as a function of time . A single slice transecting the striatum was chosen for analysis in each animal ( Figure 9c ) . BOLD signal change was expressed as the percentage change relative to the average baseline signal ( first 50 images ) prior to drug injection . Data are expressed as mean ± SEM . Sigmastat 3 . 1 software was used for statistical analysis . Comparison between two experimental groups was done by Student's t test or Mann-Whitney Rank Sum Test when the data did not follow a normal distribution . When several experimental groups were analyzed , we used two-way analysis of variance ( ANOVA ) . For locomotion experiments we used ANOVA with repeated measures , and when appropriate , a Tukey post hoc comparison test was used . For pharmacological MRI , the area under the curve of the signal time course was compared between VAChTD2-cre/flox/flox mice and control mice using a Student's t test . | The neurotransmitters dopamine and acetylcholine play opposite roles in the striatum ( a brain region involved in motor control and reward-related behaviour ) , and their balance is thought to be critical for striatal function . Acetylcholine in the striatum has been linked to a number of functions , including control of locomotor activity and response to drugs of abuse . However , striatal cholinergic interneurons can also release glutamate ( in addition to acetylcholine ) and it is presently unclear how these two neurotransmitters regulate striatal-dependent behaviour . Previous work has attempted to resolve this issue by ablating cholinergic neurons in the striatum , but this causes loss of both cholinergic and glutamatergic neurotransmission . In this study , we created a novel genetic mouse model which allowed us to selectively interfere with secretion of acetylcholine in the striatum , while leaving total striatal glutamate release intact . In these mice , we observed minimally altered behavioural responses to cocaine , suggesting that striatal glutamate , rather than acetylcholine , is critical for cocaine-induced behavioural manifestations . Furthermore , elimination of striatal acetylcholine release affects how striatal output neurons respond to dopamine , by up-regulating dopaminergic receptors and changing behavioural responses to dopaminergic agonists . Our experiments highlight a previously unappreciated physiological role of cholinergic-glutamatergic co-transmission and demonstrate how a population of neurons can use two distinct neurotransmitters to differentially regulate behaviour . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"medicine",
"neuropsychiatric",
"disorders",
"neurochemistry",
"mental",
"health",
"anatomy",
"and",
"physiology",
"neuroscience",
"motor",
"systems",
"neurotransmitters",
"neurological",
"system",
"neuroimaging",
"nervous",
"system",
"physiology",
"biology",
"psychiatry",
"neuromodulation",
"central",
"nervous",
"system",
"synapses",
"physiology",
"animal",
"cognition",
"substance",
"abuse"
] | 2011 | Elimination of the Vesicular Acetylcholine Transporter in the Striatum Reveals Regulation of Behaviour by Cholinergic-Glutamatergic Co-Transmission |
The evolutionarily conserved Hippo ( Hpo ) signaling pathway plays a pivotal role in organ size control by balancing cell proliferation and cell death . Here , we reported the identification of Par-1 as a regulator of the Hpo signaling pathway using a gain-of-function EP screen in Drosophila melanogaster . Overexpression of Par-1 elevated Yorkie activity , resulting in increased Hpo target gene expression and tissue overgrowth , while loss of Par-1 diminished Hpo target gene expression and reduced organ size . We demonstrated that par-1 functioned downstream of fat and expanded and upstream of hpo and salvador ( sav ) . In addition , we also found that Par-1 physically interacted with Hpo and Sav and regulated the phosphorylation of Hpo at Ser30 to restrict its activity . Par-1 also inhibited the association of Hpo and Sav , resulting in Sav dephosphorylation and destabilization . Furthermore , we provided evidence that Par-1-induced Hpo regulation is conserved in mammalian cells . Taken together , our findings identified Par-1 as a novel component of the Hpo signaling network .
The control of organ size , which requires the delicate coordination of cell growth , cell proliferation , and cell death , is a fascinating biological process . The identification of the Hippo ( Hpo ) signaling pathway has shed some light on this biological phenomenon . The Hpo pathway has emerged as an evolutionarily conserved pathway that controls organ size during animal development . It regulates tissue growth by balancing cell proliferation and apoptosis and has also been implicated in stem cell maintenance , tissue homeostasis , and repair [1]–[4] . In addition , it has been reported to play a role in cell contact-dependent growth inhibition [5] . Accumulating evidence has suggested that mutations and malfunctions of the components of the Hpo pathway result in a wide range of human cancers and diseases [6] . The Hpo pathway can be divided into three parts: upstream regulatory inputs , core kinase cassette , and downstream transcriptional output [2] . Core to the Hpo pathway is the kinase cascade , which acts sequentially to inhibit the nuclear translocation and activity of the growth-promoting transcriptional coactivator Yorkie ( Yki ) or Yap/TAZ ( mammalian homologue of Drosophila Yki ) [7] . The core kinase cascade of the Hpo pathway consists of four tumor suppressor proteins , including two kinases , the serine/threonine Ste20-like kinase Hpo or its mammalian homologues MST1/2 [8]–[12] , and the nuclear Dbf-2-related ( NDR ) family kinase Warts ( Wts ) or its mammalian homologues LATS1/2 [13] , [14] , and the scaffold proteins of the kinases , Salvador ( Sav ) [15] , [16] and Mob as tumor suppressors ( Mats ) [17] . Hpo phosphorylates and activates Wts via the formation of a complex with Sav [15] , [18] , [19] . Wts functions in a complex with Mats to restrict the nuclear translocation of Yki by phosphorylating Yki at multiple sites [7] , [17] , [19] , [20] . In the absence of suppression from the Hpo pathway , Yki associates with transcription factors , primarily Scalloped ( Sd ) [20]–[22] and other factors , including Homothorax [23] , Teashirt , and Mad [24] , in the nucleus to promote proliferation and to inhibit apoptosis by inducing the expression of target genes , such as bantam , cyclinE , and diap1 [25] . Comparing the clear linear relationship between the core kinase cassette and the downstream transcriptional output , this pathway is regulated by multiple upstream regulatory branches , such as the Merlin-Expanded ( Ex ) -Kibra complex [26]–[29] , Fat ( ft ) and Dachsous [30]–[33] , Crumbs and the Lgl-Scrib-Dlg complex [34] , [35] , and , the most recently identified , Echinoid and Tao-1 [36]–[38] . Several recent proteome-wide phosphorylation studies , which have uncovered a large number of previously unknown phosphorylation events in Hpo signaling [39] , [40] , have indicated the involvement of a large number of unknown participants in the Hpo pathway . To identify novel pathway modulators , we performed a gain-of-function EP screen and identified Par-1 as a novel Hpo pathway regulator . Par-1 is a multifunctional serine/threonine kinase containing an N-terminal conserved catalytic domain , a ubiquitin-associated ( UBA ) domain adjacent to the catalytic domain , and a kinase associated domain 1 ( KA1 domain ) within the last 40 amino-acids [41] . Par-1 plays a major role in anterior/posterior ( A/P ) axis formation and germline determinant polarization , and it also regulates diverse cellular processes , including microtubule dynamics and neuronal polarity [42]–[47] . Although Par-1 is involved in multicellular processes , little is known regarding the function of Par-1 in disease and tumor formation . Hyper-phosphorylation of the microtubule-associated protein Tau by microtubule affinity regulating kinase , the homolog of Drosophila Par-1 ( MARK ) [48] , which is activated by upstream kinases , such as LKB1 [49] and Tao-1 [50] , results in microtubule depolymerization and abnormal aggregation of Tau in Alzheimer disease . Additionally , MARK4 has been reported to be involved in hepatocellular carcinogenesis and gliomagenesis [51] , [52] . In this study , we identified Par-1 as a negative regulator of the Hpo kinase complex . We found that overexpression of Par-1 drove tissue overgrowth and upregulated the expression of Hpo pathway-responsive genes . Moreover , knockdown of Par-1 blocked tissue growth and downregulated the expression of Hpo pathway-responsive genes . We demonstrated that par-1 functioned downstream of ex and ft but upstream of hpo and sav . We also provided evidence that Par-1 associated with the Hpo-Sav complex and regulated the phosphorylation of Hpo at Ser30 to regulate Hpo activity . Furthermore , we found that Par-1 promoted the dissociation of Sav from the Hpo-Sav complex , eventually resulting in Sav dephosphorylation and destabilization . Thus , these results identified Par-1 as a novel regulator of the Hpo signaling pathway and supported a model by which Par-1 regulates Hpo phosphorylation and Hpo-Sav association to control organ growth .
To identify novel candidates of the Hpo pathway , we performed an overexpression screen in which flies carrying GMR-Gal4 and UAS-Yki ( referred to as GMR-Yki ) were crossed with a collection of EP lines . Overexpression of UAS-Yki posterior to the morphogenetic furrow ( MF ) under the control of the GMR-Gal4 driver ( GMR-Yki ) resulted in enlarged eyes ( compare Figure 1A′ with 1A ) , providing a sensitive background for a genetic modifier screen [53] . Each EP line was crossed with GMR-Yki flies , and the F1 progeny was screened for an increase in eye size . From more than 10 , 000 EP lines , we screened numerous lines that enhanced the overgrowth phenotype induced by Yki overexpression . We then analyzed the UAS element insertion sites of these lines and found that the insertion sites of L[484] , L[507] , and F[727] were all within the 5′ UTR region of the par-1 gene ( Figure 1C ) . Although these three EP lines did not display an overgrowth phenotype when driven by GMR-Gal4 in Drosophila eyes ( compare Figure 1B with 1A , and unpublished data ) , these lines dramatically enhanced the GMR-Yki-induced overgrowth phenotype ( compare Figure 1B′ with 1A′ , and unpublished data ) . In addition , the expression of these lines driven under the wing-specific Gal4 driver MS1096 produced enlarged adult wings ( Figure 1D–1D″ ) , indicating that the candidate genes expressed in these lines may play a role in organ size control . To determine whether the UAS element of these lines regulated par-1 gene expression , real-time PCR analysis was performed for the L[484] line . The mRNA level of par-1 was significantly upregulated when the L[484] line was crossed with MS1096 , whereas the mRNA level of genes located proximal to par-1 , mei-W68 , and hpo , demonstrated a slight or no change ( Figure 1E ) , suggesting that ectopic Par-1 expression could be responsible for the tissue overgrowth phenotype that we observed . To determine whether overexpression of L[484] promoted tissue growth via Hpo signaling , the L[484] line was expressed under the control of the hh-Gal4 driver , which drives gene expression in the posterior compartment ( P-compartment ) . As shown in Figure 1G–1G′ , ex-lacZ ( EX-Z ) , an enhancer trap for ex [27] , was increased in the P-compartment of the wing imaginal disc , suggesting an inhibition of Hpo signaling . Furthermore , the Hpo downstream marker diap1-lacZ was also upregulated in the flip-out clones expressing L[484] ( Figure 1H–1H″ ) . Briefly , these observations suggested that the expression of the L[484] line promoted tissue growth via Hpo signaling by controlling the expression of Par-1 . To verify the functional relationship between Par-1 and the Hpo pathway , a dual luciferase assay , which reflected Sd-Yki transcriptional activity [20] , was performed . As shown in Figure 2A , in S2 cells , coexpression of Yki and Sd activated the luciferase reporter gene ( 3×Sd2-Luc ) , which was greatly promoted by Par-1 , indicating that Par-1 enhanced the activity of the Sd-Yki transcriptional complex in vitro . To further determine the functional relationship between Par-1 and the Hpo pathway in vivo , Myc tagged Par-1 transgenic flies were generated . Consistent with the results in Figure 1A–1B′ , overexpression of two copies of UAS-Myc-Par-1 , using the GMR-Gal4 driver ( referred to as GMR/2*Myc-Par-1 ) , resulted in rough eyes without a discernible overgrowth ( compare Figure 2B′ with 2B ) , while coexpression of UAS-Myc-Par-1 with GMR-Yki enhanced the overgrowth phenotype caused by GMR-Yki ( compare Figure 2C′ with 2C ) . Although GMR/2*Myc-Par-1 did not induce a discernible overgrowth phenotype in the eyes ( Figure 2B′ ) , the expression of two copies of UAS-Myc-Par-1 , using the MS1096 driver ( referred to as MS1096/2*Myc-Par-1 ) , resulted in enlarged wings and caused a wing bending-down phenotype , which indicated an expansion of the wing ( Figure 2D′″ and compare Figure 2D′ with 2D and 2E′ with 2E ) . We also found that the relative P-compartment area of the wings expressing UAS-Par-1 , using the hh-Gal4 driver , was increased ( Figure S1A–S1B ) . We then examined whether overexpression of Par-1 affected the expression of Hpo pathway-responsive genes . We found that flip-out clones expressing UAS-Myc-Par-1 in the wing imaginal discs showed upregulated expression of EX-Z ( Figure 2F–2F′ ) , diap1-lacZ ( Figure 2H–2H′ ) and diap1-GFP3 . 5 ( a diap1 enhancer element reporter [20] , Figure S1C–S1C″ ) , suggesting compromised Hpo signaling activity . Taken together , these results suggested that overexpression of Par-1 promoted tissue overgrowth by inhibiting Hpo pathway activity . Considering that the Ser/Thr kinase activity of Par-1 was important for its function in polarity regulation , we speculated that the function of Par-1 in Hpo signaling might also be dependent on its kinase activity . To examine this hypothesis , we first constructed a kinase-dead form of Par-1 ( Par-1-KD ) , which contained the T408A and S412A mutations . Par-1 , containing these two mutations , was thought to be a kinase inactive mutant because the activation loop of the catalytic domain was disrupted [54] . This was confirmed by an in vitro kinase assay in which the kinase activity of Par-1-KD was completely abolished ( Figure S1D ) . Unlike the phenotype observed with the expression of two copies of the Par-1 transgenes , the expression of two copies of the Par-1-KD transgenes had no obvious effect on either eye growth or wing growth and did not dramatically enhance the overgrowth phenotype induced by Yki overexpression ( compare Figure 2B″ , 2C″ with 2B–2B′ , 2C–2C′ and 2D″ , 2E″ with 2D–2D′ , 2E–2E′ ) . Importantly , to exclude the possibility that the functional variation between Par-1 and Par-1-KD was due to a low expression level of Par-1-KD , we compared the overexpressed Par-1 and Par-1-KD protein levels in both the eye and wing imaginal discs using direct Western blot analysis . We found that the overexpression level of Par-1-KD was higher compared to Par-1 , verifying that the functional variation did not result from a low Par-1-KD expression level ( Figure S1E–S1F ) . Furthermore , Par-1-KD failed to elevate Ex and Diap1 expression in flip-out clones ( Figure 2G–2G′ and 2I–2I′ ) . Taken together , these observations demonstrated that overexpression of Par-1 promoted tissue overgrowth by promoting the activity of the Sd-Yki complex and upregulating the expression of Hpo pathway-responsive genes in a kinase-dependent fashion . To determine whether Par-1 is necessary for normal growth , we examined the effect of the loss-of-function of Par-1 on Hpo signaling . By expressing UAS-Par-1-RNAi under the control of eyeless-Gal4 ( ey-Gal4 ) or MS1096 , adult eye/wing sizes were reduced ( Figure 3A–3B″ ) , suggesting that Par-1 ( activity ) was required for normal eye and wing development . Par-1 RNAi efficiency was also confirmed by in vivo staining , in which Par-1-RNAi transgenes were expressed under the control of hh-Gal4 . As shown in Figure S2B–S2B′ , endogenous Par-1 protein levels were efficiently knocked down by expressing Par-1-RNAi in the P-compartment . In addition , shrinkage of the P-compartment was also observed ( Figure S2B–S2B′ ) . To eliminate the concern regarding Par-1 RNAi off-target effects , a second line of Par-1 RNAi ( Par-1-RNAi-2 ) , which targets a different region of Par-1 , was generated . Par-1-RNAi-2 also efficiently knocked down endogenous Par-1 expression ( Figure S2C–S2C′ ) and restricted wing growth when expressed by MS1096 ( Figure S2D–S2D′ ) . Furthermore , the expression of Par-1-RNAi by GMR-Gal4 resulted in the detection of caspase-3 in its active ( cleaved ) form ( Figure 3D–3D′ ) , indicating a role for Par-1 in restricting tissue growth by inducing apoptotic cell death . We further tested whether knockdown of Par-1 resulted in downregulation of Hpo pathway-responsive genes . Expression of either UAS-Par-1-RNAi or UAS-Par-1-RNAi-2 by hh-Gal4 resulted in diminished levels of EX-Z , DIAP1 , and diap1-GFP3 . 5 and a reduced P-compartment size ( Figures 3E–3E′ , 3H–3H′ , and S2E–S2E′ , S2F–S2F″ ) . Consistent with these results , a bantam sensor ( mic32-GFP ) signal was upregulated in Par-1-RNAi flip-out clones ( Figure 3F–3F′ ) or wing discs expressing Par-1-RNAi-2 by hh-Gal4 ( Figure S2G–S2G″ ) , suggesting a restriction of microRNA bantam expression by knockdown of Par-1 . Thus , this evidence suggested that the inactivation of Par-1 resulted in abnormal growth by antagonizing the expression of Hpo-responsive genes . To further strengthen this conclusion , the expression of Hpo-responsive genes was examined in par-1w3 mosaic clones . As shown in Figure 3I–3I″ and 3J–3J″ , in par-1 null clones , the diap1 transcriptional level was reduced ( Figure 3I–3I″ ) , and bantam-lacZ was decreased ( Figure 3J–3J″ ) . Importantly , the size of the par-1 null clones was significantly reduced compared to their twin spots ( Figure 3C–3C′ ) , indicating a proliferation disadvantage for par-1 null clones . Taken together , these observations demonstrated that par-1 was essential for normal growth and that perturbation of Par-1 expression resulted in growth suppression and apoptosis by stimulating the Hpo pathway . Given the findings presented in the previous section , we next determined the functional relationship between Par-1 and Hpo pathway components by identifying the genetic interactions of Par-1 in the Hpo pathway . We first examined whether the function of Par-1 was dependent on the activity of the Sd-Yki transcriptional complex because this complex was the main downstream effector of the Hpo pathway . Although expression of two copies of UAS-Myc-Par-1 in flip-out clones increased diap1-lacZ ( Figure S3A–S3A′ ) , no increase in diap1-lacZ was detected when UAS-Yki-RNAi was coexpressed ( Figure S3B–S3B′ ) . Coexpression of UAS-Sd-RNAi suppressed the overgrowth phenotype induced by Par-1 overexpression in Drosophila wings ( Figure S3C–S3C′ ) . In addition , the elevated levels of diap1 transcription caused by Par-1 overexpression were reverted by coexpression of Sd RNAi ( Figure S3D–S3F′ ) . Furthermore , ectopic Yki expression reverted downregulated DIAP1 levels and the shrunken P-compartment phenotype induced by the expression of Par-1 RNAi ( Figure 4A–4B′ ) . These results indicated that par-1 functioned upstream of the Sd-Yki transcription complex in the Hpo pathway . To strengthen this conclusion , the levels of phosphorylated Yki , which reflected Hpo/Wts activity , were examined . As expected , Par-1 , but not Par-1-KD , reduced phosphorylated Yki levels ( Figure 4C ) . In addition , Par-1 also inhibited the Hpo/Wts signaling-induced Yki mobility shift ( Figure 5I , lanes 2–5 ) . These findings suggested that Par-1 functioned upstream of yki to affect the activity of the Sd-Yki transcription complex . We next examined the genetic epistasis between Par-1 and the upstream components of the Hpo pathway . We found that elevated DIAP1 levels and the enlarged P-compartment size ( Figure 4D–4D′ ) , resulting from ex RNAi expression by hh-Gal4 , were suppressed by coexpression of the Par-1 RNAi transgene ( Figure 4E–4E′ ) , suggesting that Par-1 functioned downstream of ex . In addition , coexpression of Par-1 RNAi suppressed ex RNAi-induced wing overgrowth ( Figure S3G–S3G′″ ) . Furthermore , Par-1 RNAi also suppressed Ft RNAi-induced wing overgrowth ( Figure S3H–S3H′″ ) . These observations supported the notion that par-1 functioned downstream of or in parallel to ex and ft . We then determined whether Par-1 regulated the Hpo pathway via Wts , which phosphorylates Yki at Ser168 to retain Yki in the cytoplasm [19] . By generating Par-1 RNAi clones in eye discs using the mosaic analysis with a repressible cell marker ( MARCM ) technique , we found that the size of Par-1 RNAi clones was extremely small compared to that of the control clones ( compare Figure 4F′ with 4F ) , indicating adverse development of the Par-1 RNAi clones . Strikingly , we found that knockout of wts rescued the adverse developmental phenotypes of the Par-1 RNAi clones and that the size of the wts mutant clones expressing Par-1 RNAi was more comparable to that of the wts clones , rather than that of Par-1 RNAi clones ( Figure 4G and compare Figure 4F″ with 4F′″ ) . These findings indicated that par-1 functioned upstream of wts . Because Wts was activated and phosphorylated by upstream components of the Hpo pathway , we then tested whether Par-1 regulated Wts phosphorylation . Par-1 , but not Par-1-KD , reduced the mobility shift of Wts phosphorylation in the presence of Hpo/Sav/Merlin/Tao-1 ( Figure 4H ) , suggesting that par-1 functioned upstream of wts and regulated Wts phosphorylation in a kinase-dependent manner . We then determined whether Par-1-regulated Hpo signaling was dependent on the activity of Hpo , which restricted tissue growth by phosphorylating Wts . By generating hpo mutant clones in Drosophila compound eyes using the MARCM system , we found that the ablation of Hpo resulted in tumor outgrowth ( compare Figure 4I′ with Figure 4I ) . Furthermore , we found that Par-1 RNAi was incapable of reverting the growth advantage of hpo null clones ( compare Figure 4I″ with Figure 4I′ ) , indicating that Par-1 functioned upstream of hpo to regulate Hpo signaling . Given that Par-1 functions upstream of hpo ( Figure 4 ) , we speculated that Par-1 modulated the function of the Hpo-Sav complex . To test our hypothesis , we first examined whether Par-1 bound to the Hpo-Sav complex using a co-immunoprecipitation assay . As expected , full-length Par-1 interacted with both Hpo and Sav , although these associations were weak ( Figure 5A–5B ) . In addition , we also found that the HA-tagged N-terminal fragment of Par-1 ( Par-1-N , Figure S4A ) had a strong association with both Hpo and Sav , whereas there was only a weak interaction between the Par-1 C-terminal fragment ( Par-1-C , Figure S4A ) and Hpo/Sav had been detected ( Figure 5A–5B ) . Importantly , the interaction between Par-1 and Hpo/Sav was specific because neither the full-length nor the N-terminal fragment of Par-1 co-immunoprecipitated with Merlin ( unpublished data ) , Wts , or Mats ( Figure S4B ) . On the basis of the previous results that Par-1 functioned in a kinase-dependent manner in Hpo signaling , we performed a phosphorylation shift experiment using a phos-tag gel to determine whether Par-1 affected the phosphorylation of Hpo and Sav in vitro . The Phos-Tag is a phosphate binding compound that , when incorporated into polyacrylamide gels , results in an exaggerated mobility shift for phosphorylated proteins , which is dependent upon the degree of phosphorylation [18] , [55] . We observed that Flag-Hpo cotransfected with Par-1 , but not Par-1-KD , in S2 cells that exhibited a mobility shift ( Figure S5A ) . However , such a mobility shift was not detected for Sav when it was cotransfected with either Par-1 or Par-1-KD ( Figure 6B , compare lanes 3–4 with lane 1 ) . To remove the effect of Hpo auto-phosphorylation , a kinase-dead form of Hpo ( Hpo-KD ) was also tested in the phosphorylation shift experiment . As shown in Figure 5C , Par-1 also induced Hpo-KD to generate a phosphorylation mobility shift . These results suggested that Par-1 specifically induced Hpo phosphorylation in S2 cells . To identify which sites on Hpo were affected by Par-1 , we cotransfected Flag tagged Hpo with HA tagged Par-1 in S2 cells . Flag-Hpo protein was then immunoprecipitated and separated using SDS-PAGE and harvested for a semi-quantitative mass spectrometric ( MS ) analysis ( details are described in the Materials and Methods ) . According to the MS results , we identified four potential phosphorylation sites that were affected by Par-1 expression , S30 , S66 , Y365 , and T615 . We then generated different Hpo variants by individually mutating these candidate sites and then examined how these mutations affected Par-1-induced Hpo phosphorylation . Interestingly , only Hpo ( S30A ) failed to generate mobility shifts in a direct Western blot analysis compared to wild-type Hpo protein ( Figure S5C ) , while other Hpo mutations , T615A , S66A , and Y365E , did not affect the mobility shift upon Par-1 induction ( Figure S5B–S5C ) . Moreover , we also found that Hpo ( S30A ) -KD could not be shifted by Par-1 ( Figure 5D ) . These results suggested that Par-1 may regulate the phosphorylation of Hpo at Ser30 . To verify our prediction , an antibody that specifically recognized phosphorylated Hpo at Ser30 was generated . As shown in Figures 5E and S5D , Par-1 but not Par-1-KD was able to induce Hpo phosphorylation at Ser30 . To validate the phospho-Hpo S30 antibody , a phosphatase treatment was applied . Upon phosphatase treatment , the p-Hpo ( S30 ) band was undetectable ( Figure 5E , compare lane 5 with lane 2 ) . To further test the specificity of this antibody , we transfected Hpo ( S30 ) mutants with Par-1 . As shown in Figure 5F and in Figure S5E , in the presence of Par-1 , Hpo ( S30A ) could not be detected by the p-Hpo ( Ser 30 ) antibody , whereas wild-type Hpo ( S30 ) was detected . These findings indicated that Par-1 regulated the phosphorylation of Hpo at Ser30 . In recent proteome-wide phosphorylation studies using Drosophila embryos [40] , it was suggested that Hpo was phosphorylated at Ser30 in vivo , indicating an important role for the Ser30 site in the regulation of Hpo activity . To determine the biological significance of Hpo phosphorylation at Ser30 induced by Par-1 , we first detected whether Ser30 phosphorylation state affects Hpo phosphorylation at Thr195 , which was important for Hpo activation . As shown in Figure 5G–5H , Par-1 , but not Par-1-KD , significantly inhibited Hpo phosphorylation levels at Thr195 , whereas this inhibitory effect was abolished when the Ser30 site was mutated . More importantly , phosphorylation at Thr195 was slightly elevated when Ser30 was mutated into an alanine ( Figure 5G , compare lane 4 with lane 1 , and Figure 5H ) . These findings suggested that Par-1 regulated Hpo activity via antagonizing phosphorylation at the Thr195 site by regulating Ser30 phosphorylation . It has been reported that the Hpo Thr195 site is not only auto-phosphorylated but also phosphorylated by Tao-1 [37] , [38] , which is a partner of Par-1 in the regulation of microtubule dynamics . Thus , we asked whether Par-1-induced phosphorylation at Ser30 also affected Tao-1-mediated phosphorylation at Thr195 . As shown in Figure S5F , consistent with the results shown in Figure 5G–5H , Par-1 suppressed Tao-1-mediated phosphorylation at Thr195 . The antagonistic effect of Par-1 and Tao-1 on Hpo phosphorylation at Thr195 motivated the examination of the interrelationship of Par-1 and Tao-1 in the Hpo pathway . We found that Tao-1 disrupted Par-1-induced a phosphorylation mobility shift of Hpo-KD ( Figure S5F ) , suggesting that the function of Par-1 in the Hpo pathway was modulated by upstream signaling . Because Hpo ( S30A ) demonstrated a higher activity compared to the Hpo wild type , we examined whether Hpo ( S30A ) exerted its inhibitory effect on Yki . We found that Hpo ( S30A ) resulted in a dramatic Yki mobility shift , whereas the Hpo wild type resulted in a moderate phosphorylation of Yki ( Figure 5I , compare lane 2 with lane 6 ) . Furthermore , we found that co-transfection of the Hpo ( S30A ) mutant blocked Par-1-induced Yki dephosphorylation ( Figure 5I , compare lane 3 with lane 7 ) , which further confirmed our conclusion that Par-1 modulated Hpo activity by regulating Hpo phosphorylation at Ser30 . To further investigate the role of Par-1-induced Hpo phosphorylation at Ser30 in Hpo activity regulation , we generated transgenic flies of attB-UAS-Hpo variants at the 75B1 attP locus , which ensured equal expression of different forms of the Hpo mutant . Because the flies were grown at room temperature , overexpression of Hpo under the control of GMR-Gal4 , Ci-Gal4 , and hh-Gal4 resulted in adult lethality; therefore , we selected weak drivers , such as C765 , which induced gene expression moderately in the Drosophila wing to compare the activity of Hpo variants . Consistent with our in vitro studies , Hpo ( S30A ) mutants exhibited much smaller wings compared to wild-type Hpo flies ( Figure 5J–5K ) , indicating that Hpo ( S30A ) variants demonstrated a higher activity than wild-type Hpo in vivo . To further strengthen this conclusion , we used an inducible Ci-Gal4 , which was controlled by a temperature-sensitive Gal80 , to drive Hpo expression in order to exclude the early lethality . Because some progeny were viable when Hpo expression was induced after pupa formation , we then compared the activity of Hpo variants by measuring the wing size of the survivors and calculated the mortality rate . We found that survivors with Hpo expression maintained a relatively normal wing size ( Figure S5G–S5H ) , while survivors with Hpo ( S30A ) expression demonstrated much smaller wings compared to controls ( Figure S5G–S5H ) . Meanwhile , Hpo ( S30A ) flies exhibited a higher mortality rate compared to wild-type Hpo flies ( Figure S5I ) . These findings suggested that the activity of Hpo ( S30A ) mutants was higher compared to wild-type Hpo . Taken together , on the basis of the above biochemical and in vivo evidence , we speculated that Par-1 inhibited Hpo activity via the regulation of Hpo phosphorylation at the Ser30 site . We observed that Par-1 blocked Hpo-induced Sav stabilization in a kinase-dependent manner ( Figure S5C , and Figure 6A , compare lane 3 with lane 2 ) . Thus , we examined whether Par-1 induced Sav destabilization , which was dependent on the phosphorylation of the Hpo Ser30 site . Interestingly , we found that the stabilization of Sav by Hpo was not affected by Par-1-induced Hpo phosphorylation at Ser30 because Hpo ( S30A ) mutants were still able to stabilize Sav , and this stabilization could be reversed by Par-1 ( Figure S6A ) . In addition to the change in Sav protein levels , Par-1 also decreased the phosphorylation status of Sav , which was induced by Hpo in a kinase-dependent manner ( Figure 6B , compare lanes 5–6 with lane 2 ) . Given that full-length Par-1 weakly interacted with both Hpo and Sav ( Figure 5A–5B ) , we then determined whether Par-1 affected the association between Hpo and Sav . As shown in Figure 6C , Par-1 , but not Par-1-KD , impaired the association of endogenous Hpo and transfected Sav , suggesting that the interaction between Hpo and Sav was disrupted by Par-1 overexpression . To mimic the disruption of the Hpo-Sav complex , we ablated sav using the MARCM system . We found that the growth defect induced by Par-1 RNAi was incapable of inhibiting sav mutant-induced clone growth ( Figure S6B–S6C ) and adult eye overgrowth ( Figure 6D-–6D″ ) , suggesting that Par-1 functioned upstream of sav . These observations supported the notion that Par-1 kinase activity was important to restrain the function of the Hpo-Sav complex . On the basis of the evidence provided above , we propose a model of how Par-1 restricts the activity of Hpo pathway ( Figure 6E ) . Par-1 regulates Hpo phosphorylation at Ser30 to modulate the Hpo kinase activity; simultaneously , Par-1 also promotes the dissociation of Sav from Hpo , resulting in the dephosphorylation and destabilization of Sav , thereby repressing the function of the Hpo-Sav complex . An evolutionally conserved function of Par-1 in regulating microtubule dynamics has been reported [41] . To determine whether the function of Par-1 on the Hpo pathway is conserved , the human homologue of Par-1 , MARK1 and MARK4 were cloned . The Gal4-tead4 reporter [56] was used to examine the effect of MARK1 and MARK4 on the mammalian Hpo pathway . As expected , both MARK1 and MARK4 , but not the kinase-dead form of MARK4 , activated the YAP transcription co-activator activity ( Figures 7A and S7A ) . Since MARK4 activated YAP more than MARK1 , we investigated whether MARK4 affected YAP phosphorylation . Indeed , the phosphorylation levels of YAP were significantly decreased upon MARK4 overexpression ( Figure 7B ) , indicating that MARK inhibited Hpo signaling in mammals . Finally , we investigated whether MARK also resulted in MST ( human homologue of Hpo ) phosphorylation . We found that both MARK4 and MARK1 induced mobility shifts of MST2 ( Figures 7C and S7B ) . Taken together , these findings suggested that the inhibitory function of Par-1 on Hpo signaling was conserved from Drosophila to mammals . Our study suggested that Par-1 may have a procarcinogenic role because its hyperactivation in Drosophila is sufficient to induce tissue overgrowth , and in mammals , Par-1 is sufficient to activate YAP . Interestingly , MARK4 has been suggested to play a role in hepatocellular carcinogenesis and gliomagenesis [51] , [52] . However , whether MARK1 is also involved in carcinogenesis is largely unknown . To further elucidate this question , we characterized the mutation and expression of MARK1 in different cancer samples using the COSMIC and GEO databases . Although no mutation or deletion of MARK1 has been reported in the tumors that were surveyed , the transcription levels of MARK1 showed significant upregulation in squamous lung cancer samples and during the progression of prostate cancer ( Figure 7D–7E ) . Taken together , these findings suggested that Par-1 was a potential oncogene and that its regulatory role in Hpo signaling could be conserved .
The Hpo signaling pathway has emerged as a conserved pathway that controls tissue growth and balances tissue homeostasis via the regulation of the downstream Sd-Yki transcription complex . Despite the importance of this pathway in development and carcinogenesis [2] , 6 , many unknown regulators of the Hpo pathway remain to be identified . Here , we identified Par-1 as one such Hpo pathway regulator via a genetic overexpression screen using Drosophila EP lines . In this study , we demonstrated that Par-1 was essential for the restriction of Hpo signaling . We also demonstrated that overexpression of Par-1 promoted tissue growth via the inhibition of the Hpo pathway , whereas loss of Par-1 promoted Hpo signaling to suppress growth and induce apoptosis . Using the Drosophila eye and wing imaginal discs as well as cultured cells , we provide the first genetic and biochemical evidence for a function of Par-1 in the Hpo pathway . Although the conserved function of Hpo has been well studied , the regulatory mechanism of its kinase activity is still largely obscure . Currently , the regulatory mechanism of Hpo kinase activity was believed to mainly be dependent on autophosphorylation by altering the phosphorylation status of the Thr195 site [37] , [53] , [57] . However , whether the uncharacterized phosphorylation events of Hpo , which have been identified in several recent proteome-wide phosphorylation studies [39] , [40] , contributed to the regulation of Hpo activity is still unknown . By studying the mechanism underlying Par-1 function in Hpo signaling , we demonstrated that Par-1 induced Hpo phosphorylation at Ser30 and this lead to the regulation of Hpo kinase activity . Although we have extensively studied how Par-1 regulates the Hpo pathway in this study , several unresolved questions remain . The interaction between Par-1 and Hpo/Sav may be tightly regulated because full-length Par-1 only weakly interacted with Hpo/Sav , unlike the interaction with the N-terminal fragment of Par-1 ( Figure 5A–5B ) . However , the triggering signal for Par-1 to interact with Hpo/Sav is still unknown . It has been reported that Par-1 was activated by Tao-1 and LKB1 [49] , [50] . In this study , we established that Par-1 antagonized Tao-1 in Hpo signaling , and interestingly , in Drosophila , the antagonistic relationship between Par-1 and Tao-1 in microtubule regulation has been previously reported [57]–[59] . Thus , it is unlikely that Tao-1 functions as the trigger . We then investigated whether LKB1 functioned as an activator of Par-1 in Hpo signaling by expressing the LKB1 transgene in different organs . Unlike Par-1 , ectopic LKB1 expression limited both wing and eye growth ( unpublished data ) , indicating that LKB1 was also not the trigger . We have shown that Par-1 and Tao-1 exhibited opposing effects on Hpo signaling ( Figure S5F ) . Given that Tao-1 and Par-1 were partners that regulated microtubule dynamics via the phosphorylation of Tau [48] , Tau may have a function in Hpo signaling . To investigate this hypothesis , we employed genetic and biochemical studies and found that Tau RNAi failed to suppress the expression of Hpo pathway-responsive genes ( Figure S8A–S8B ) . In addition , Tau did not trigger Hpo phosphorylation and Sav dissociation in vitro ( Figure S8C–S8D ) , indicating that Par-1 regulated Hpo signaling independent of Tau . Interestingly , it has been previously suggested that Par-1 did not regulate Tau activity in Drosophila [46] , indicating an evolutionary difference between Par-1 and Tau-1 function . We have provided evidence that Par-1 regulated Hpo signaling via the phosphorylation of Hpo or the destruction of the Hpo/Sav complex . Because Par-1 is a well-known polarity regulator and polarity components , such as Crumb and Lgl , have been shown to be involved in the Hpo signaling pathway [35] , it is possible that Par-1 may regulate Hpo signaling via a polarity complex , or its activity might be regulated via a polarity complex . Indeed , the localization of Crumb and Patj were affected by Par-1 expression ( unpublished data ) . Thus , further studies on polarity complexes and Hpo signaling will help elucidate this problem .
Par-1 fragments were amplified from the BDGP DGC Clone ( number RE47050 ) using the PCR . The sequence of full-length Par-1 used in this study was the same as Par-1-N1S ( accession number NP_001014542 ) . which has been previously described [43] . The Par-1 kinase-dead mutant was generated by converting the conserved Ser412 and Thr408 into alanine at the activation loop of the Par-1 catalytic domain . All of the primers used in this study are available upon request . The EP library for the screen was a gift from Jianming Chen ( Third Institute of Oceanography , State Oceanic Administration , China ) . Par-1w3 is a null allele from the St Johnston lab , and FRT/FLP-mediated mitotic recombination was used to generate mutant clones , as previously described [20] . The genotypes used to generate the clones were the following: FRTG13 Par-1w3/eyflp; FRTG13 ubi-GFP . Par-1 RNAi fly was purchased from the Bloomington Drosophila Stock Center ( stock number 32410 ) . To generate the Par-1 RNAi-2 transgenic fly , an artificial microRNA method was adopted , which was reported to efficiently silence gene expression [60] . Briefly , we designed two hairpin oligos that were targeted to the 1227–1247 base pair and the 1551–1571 base pair regions of Par-1-N1S ( accession number NM_001014542 ) . Next , these two hairpin oligos were ligated to create a tandem hairpin RNA . The procedure for this complete construction can be found in the supplementary information of Wang et al . 's study [61] . The following transgenes were used in this study: bantam-lacZ ( a gift from Wei Du , The University of Chicago ) , UAS-ex-RNAi ( V22994 , VDRC ) , UAS-fat-RNAi ( V9396 , VDRC ) , and UAS-Tau-RNAi ( V25024 , VDRC ) . Other stocks included: bantam sensor mic32-GFP , ex-lacZ , diap1-GFP3 . 5 , hh-Gal4 , GMR-Gal4 , MS1096 , act > CD2 > Gal4 , eyless-Gal4 , diap1-lacZ , UAS-Yki , UAS-Yki-RNAi , UAS-Sd-RNAi , hpoBF33 , wtslatsX1 , and SavSH13 , which have all been previously described [20] , [53] . The generation of the transgenes at the attP locus has been previously described [20] . Unless otherwise indicated , all of the flies were cultured at 25°C . S2 cells were maintained in Drosophila Schneider's Medium ( Invitrogen ) supplemented with 10% heat-inactivated fetal bovine serum , 100 U/ml of penicillin , and 100 mg/ml of streptomycin . The cells were incubated at 25°C in a humidified air atmosphere . Plasmid transfection was performed using Lipofectamine ( Invitrogen ) , according to the manufacturer's instructions . For all of the transfection experiments , a ubiquitin-Gal4 construct was co-transfected with the pUAST expression vectors . The procedures for the immunoprecipitation , Western blotting , and immunostaining analyses were previously described [20] , [53] . The following antibodies were used in the immunoprecipitation or Western blot analyses: rabbit anti-Hpo antibody [53] , rabbit anti-Phospho Hpo ( Thr195 ) antibody ( Cell Signaling Technology ) , mouse anti-Flag antibody ( Sigma ) , mouse anti-Myc antibody ( Santa Cruz ) , mouse anti-V5 antibody ( Invitrogen ) , and mouse anti-GFP/CFP antibody ( Santa Cruz ) . The antibodies used in the immunostaining experiments were the following: rabbit anti-Par-1 antibody ( a gift from the Montell lab , Lerner Research Institute ) , mouse anti-DIAP1 ( a gift from Bruce A . Hay , California Institute of Technology ) , rat anti-cubitus interruptus ( Ci ) antibody ( Developmental Studies Hybridoma Bank , DSHB ) , rabbit anti-lacZ antibody ( Invitrogen ) , mouse anti-CD2 antibody ( Invitrogen ) , and rabbit anti-cleaved caspase-3 antibody ( Cell Signaling Technology ) . The rabbit anti-Phospho Hpo ( Ser30 ) antibody was generated by Abgent . For the luciferase reporter assay , the 3×Sd2-Luc reporter has been previously described [20] . The luciferase assay was performed using the Dual Luciferase Assay System ( Promega ) . For the phosphorylation mobility shift assays , Phos-Tag AAL-107 ( FMS Laboratory ) was introduced to enlarge the mobility shift . The operating procedure was performed according to the manufacturer's instructions . For all of the mobility shift assays , the protein samples were processed using an SDS-PAGE gel under a low voltage . According to the molecular weight of the protein , a 6% or 8% resolving gel was used for the Wts and the Sav and Hpo mobility shift assays , respectively . Immunoprecipitated cell lysates or purified protein were directly incubated in 20–40 µl of kinase assay buffer ( 250 mM HEPES , pH 7 . 4 , 0 . 2 mM EDTA , 1% glycerol , 150 mM NaCl , and 10 mM MgCl2 ) . The reaction were initiated by the addition of an ATP mixture ( 2 µl 1 mM ATP , 0 . 2 µl [γ-32P] ATP [10 mCi/ml] ) and then incubated at 30°C for 30 min . The reactions were terminated by the addition of an SDS sample buffer . Next , the samples were boiled for 5 min at 100°C followed by SDS-PAGE and autoradiography . S2 cells were transfected with Flag-Hpo or cotransfected with Flag-Hpo and Par-1 . 48 h after transfection , the cells were harvested and then lysed . SDS-PAGE and Colloidal Blue staining ( Invitrogen , LC6025 ) were then performed on the protein samples . The Hpo protein was cut from the gel and sent to the Protein Center , SIBCB for mass spectrometric analysis . A detailed procedure of the mass spectrometric analysis may be obtained from the Protein Center , SIBCB . The candidate sites were identified by the increased phosphorylation abundance in the cotransfected Flag-Hpo and Par-1 cells versus Flag-Hpo transfected cells . All of the data in this study were expressed as the mean ± standard error of the mean ( SEM ) and were analyzed using Student's t test by R 2 . 9 . 0 . The results were considered statistically significant if p<0 . 05 . | An organism's organ size is determined by cell number , the size of each cell , and the distance between cells . All of these factors are controlled by the coordination of different cell signaling pathways and other mechanisms . The Hippo signaling pathway controls organ size by restricting the number of cells that make up the organ . Malfunction of this pathway leads to abnormal overgrowth , and is involved in a large number of human diseases and cancers . We identify here a component of the Hippo pathway , Par-1 , which controls tissue growth by negatively regulating the Hippo pathway . We show that overexpression or depletion of Par-1 influences tissue growth in fruit flies via Hippo signaling . Then , by genetic and biochemical experiments , we show that Par-1 interacts with Hippo , regulating the Hippo Ser30 phosphorylation status to alter Hippo activity . In addition , we found that Par-1 regulates Hippo signaling via inhibition of the Hippo-Salvador association in a kinase-dependent fashion . We predict that Par-1 is a potential oncogene and that its regulatory role in Hippo signaling could be conserved . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"developmental",
"biology",
"biology"
] | 2013 | Par-1 Regulates Tissue Growth by Influencing Hippo Phosphorylation Status and Hippo-Salvador Association |
An Argonaute homolog and a functional Dicer have been identified in the ancient eukaryote Giardia lamblia , which apparently lacks the ability to perform RNA interference ( RNAi ) . The Giardia Argonaute plays an essential role in growth and is capable of binding specifically to the m7G-cap , suggesting a potential involvement in microRNA ( miRNA ) -mediated translational repression . To test such a possibility , small RNAs were isolated from Giardia trophozoites , cloned , and sequenced . A 26-nucleotide ( nt ) small RNA ( miR2 ) was identified as a product of Dicer-processed snoRNA GlsR17 and localized to the cytoplasm by fluorescence in situ hybridization , whereas GlsR17 was found primarily in the nucleolus of only one of the two nuclei in Giardia . Three other small RNAs were also identified as products of snoRNAs , suggesting that the latter could be novel precursors of miRNAs in Giardia . Putative miR2 target sites were identified at the 3′-untranslated regions ( UTR ) of 22 variant surface protein mRNAs using the miRanda program . In vivo expression of Renilla luciferase mRNA containing six identical miR2 target sites in the 3′-UTR was reduced by 40% when co-transfected with synthetic miR2 , while the level of luciferase mRNA remained unaffected . Thus , miR2 likely affects translation but not mRNA stability . This repression , however , was not observed when Argonaute was knocked down in Giardia using a ribozyme-antisense RNA . Instead , an enhancement of luciferase expression was observed , suggesting a loss of endogenous miR2-mediated repression when this protein is depleted . Additionally , the level of miR2 was significantly reduced when Dicer was knocked down . In all , the evidence indicates the presence of a snoRNA-derived miRNA-mediated translational repression in Giardia .
The role of small non-coding RNAs in gene regulation has been extensively studied in recent years [1] . MicroRNAs ( miRNA ) are a major class of small RNAs that are involved in gene regulation via a translational repression mechanism . They play important roles in regulation of genes involved in development [2] , cell differentiation [3] , and cell maintenance [4] . In higher eukaryotes , maturation of miRNAs from the initial RNA Polymerase II transcripts requires the actions of several proteins . Drosha , a nuclear endoribonuclease III , is known to cleave the primary-miRNAs to produce pre-miRNAs [5] . Exportin5 is responsible for exporting the pre-miRNAs out of the nucleus [6] , [7] . Dicer , a cytoplamic endoribonuclease III , cleaves the pre-miRNAs to produce mature miRNAs [8] . Argonaute , which is a major component in the RNA-induced silencing complex ( RISC ) , binds to the mature miRNA [9] , [10] . An imperfect complementation between the miRNA incorporated into the RISC complex and its target site located at the 3′-untranslated region ( UTR ) of mRNA results in translational repression [11] , [12] . One possible mechanism of repression involves binding of the Argonaute in the RISC complex to the 7-methylguanosine ( m7G ) cap of the mRNA resulting in inhibition of translation initiation [13] . Giardia lamblia is a unicellular and binucleated protozoan responsible for giardiasis in humans [14] . Phylogenetic analysis has classified it as one of the earliest branching eukaryotes with many primitive features [14] . In higher eukaryotes , gene expression is highly regulated both transcriptionally and translationally . In Giardia , however , few consensus promoters have been identified and an 8 bp AT-rich region was sufficient to initiate transcription [15] . Additionally , Giardia mRNAs have exceedingly short 3′ and 5′-UTRs , thus greatly reducing the availability of regulatory sites for translational regulation . For instance , ribosomal scanning , an essential mechanism for translation initiation in higher eukaryotes and yeast , is absent from Giardia [16] . Therefore , Giardia represents a unique model for studying the evolution of eukaryotic translational regulation . No RNA interference ( RNAi ) has been identified in Giardia in spite of repeated trials by several laboratories in the past ( data unpublished ) . An analysis of the Giardia genome showed no homolog of Drosha or Exportin5 . It , however , identified a Dicer ( XP 001705536 ) and an Argonaute homolog ( XP 001707926 ) . Giardia Dicer is the only Dicer protein whose three-dimensional structure has been resolved by X-ray crystallography [17] . It was shown to cleave double-stranded RNA ( dsRNA ) in vitro and support RNAi in a Schizosaccharomyces pombe Dicer deletion mutant [17] . Giardia Dicer is thus likely a bona fide functional Dicer . Though the function of Giardia Argonaute-like protein ( GlAgo ) remains largely unexplored , antisense-ribozyme-mediated knockdown of this protein inhibited cell growth and purified recombinant GlAgo can bind specifically to m7G-cap-Sepharose ( see below ) . These data raised the possibility that the Argonaute and Dicer in Giardia may be involved in miRNA-mediated translational repression . Abundant antisense RNAs , up to 20% of the total mRNAs , and small RNAs have been identified in Giardia [18] , [19] . About 20 snoRNAs have also been identified in Giardia [20] . Previous efforts at cloning small RNAs from Giardia have resulted in identifying small RNAs homologous to the telomeric retroposons , which were postulated to function in silencing retroposons [19] . In our current study , we isolated , cloned and sequenced small RNAs from Giardia and identified some of the known snoRNA sequences among them . One of the small RNAs , miR2 , was identified as a Dicer-digested product from GlsR17 , previously identified as a box C/D snoRNA in Giardia [20] . Putative target sites for miR2 were identified at the 3′-UTRs of many variant surface protein ( VSP ) mRNAs . Expression of a reporter mRNA carrying these putative target sites was specifically inhibited by miR2 without affecting the mRNA level . Subsequent analysis also indicated the dependence of this inhibition on the presence of Argonaute , thus verifying the ability of a snoRNA derived miRNA to function in miRNA-mediated translational repression in Giardia .
To find out if GlAgo plays a critical role in the proliferation of Giardia , mRNA encoding this protein was knocked down by expressing an antisense-ribozyme RNA in giardiavirus-infected Giardia trophozoites [21] . Quantitative RT-PCR indicated that the mRNA was reduced by 50% in the transfected cells ( data not shown ) . Growth of the knockdown cells was inhibited , reaching only 58% of the wild type level after 4 days of cultivation ( Figure S1A ) , suggesting that GlAgo plays an important role in Giardia growth . Recent studies of human Argonaute have identified the presence of a cap-binding motif in the protein [13] , which enables it to compete with eIF4E for binding to the m7G cap in mRNA that may explain a part of the mechanism involving Argonaute in miRNA-mediated translational repression [13] . To test whether this cap-binding motif is also present in GlAgo , m7G-Sepharose was incubated with recombinant GlAgo purified from E . coli . The majority of the GlAgo was found in the flow-through suggesting an excessive loading of the recombinant protein to the beads . After extensive washing with the binding buffer and buffer containing 0 . 1 mM GTP to remove non-specific binding , the protein was specifically eluted off the Sepharose beads with m7GpppG ( Figure S2 ) , suggesting that GlAgo binds to the m7G cap in a highly specific manner , which has been demonstrated to be the cap of Giardia mRNA [22] . Therefore , GlAgo may function in miRNA-mediated translational repression in Giardia by competing with eIF4E for binding to the mRNA m7G-cap [13] . To verify if miRNA-mediated translational repression is functional in Giardia , total RNA was isolated from cultures of Giardia WB trophozoites and size fractionated for small RNAs of <40 nucleotides ( nts ) . Isolated small RNAs were cloned using RNA linkers that require the presence of a 5′-phosphate in the small RNA to be ligated by T4 RNA ligase [23] . Therefore , all the cloned small RNAs should contain a 5′ phosphate , which is a known characteristic of Dicer processing product [23] . A library of 101 clones with unique sequences ranging in sizes between 20 and 34 nts was created with the following distributions: 3% 20 nts , 11% 21 nts , 12% 22 nts , 8% 23 nts , 6% 24 nts , 7% 25 nts , 14% 26 nts , 6% 27 nts , 5% 28 nts , 5% 29 nts , 9% 30 nts , 9% 31 nts , 5% 32 nts , and 1% 34 nts . Interestingly , the size distribution results in a peak at 26 nts , which is the expected size of Giardia Dicer cleaved products based on the crystal structure [17] . Of these sequences , 11 were found to be identical to other clones in the library but with longer nucleotide extensions at the 3′ ends . The shorter RNAs were presumed to be degradation products and were discarded . Potential origins of the small RNAs were identified by BLAST search analysis of the Giardia genome database [24] . The results showed that 81 of the 101 clones were fragments of ribosomal RNA or tRNA and were also discarded . Of the remaining clones , 15 were fragments of open reading frames , 4 were fragments of 3 different box C/D snoRNAs [20] , and 1 was a fragment of retrotransposon GilT . Of the four small RNAs from snoRNAs , one is a 24 nt fragment ( miR4 ) from GlsR1 ( 85 nts ) , two are 21 and 26 nt 3′-fragments ( miR1 and miR3 ) of GlsR16 ( 77 nts ) and one is a 26 nt 3′-fragment ( miR2 ) from GlsR17 ( 144 nts ) . These non-coding RNAs have been previously described as snoRNA based on the presence of box C/D motifs [20] ( see Discussion ) . GlsR1 is predicted to function in 2′-O-ribose methylation based on its sequence complementary to that of Giardia 16S ribosomal RNA as well as its homology to yeast snR70 [20] . SnoRNAs GlsR17 and GlsR16 , however , did not contain antisense sequences complementary to ribosomal RNA and were considered to be “orphan” snoRNAs . Therefore , miR2 ( 5′ CAG CCU AAU CAC CGC CCC UAU AGU CC 3′ ) from snoRNA GlsR17 and miR1 ( 5′ CAA CGC ATC ACC GCT CTG ACC 3′ ) and miR3 ( 5′ GCA GAC AAC GCA TCA CCG CTC TGA CC 3′ ) from snoRNA GlsR16 were of particular interest in terms of their potential function in Giardia . Box C/D snoRNAs typically form a hairpin structure with the 5′ and 3′ ends forming part of the stem . Therefore , it is not surprising that MFOLD analysis of full-length snoRNA GlsR16 and the 64 nt 3′-portion of GlsR17 resulted in the formation of thermodynamically favorable hairpin loop structures with the corresponding miRNAs localized to one of the two arms at the 3′-end ( Figure 1 ) [25] . Interestingly , MFOLD analysis of the full length GlsR17 resulted in a double stem loop structure similar to that observed for box H/ACA snoRNAs with miR2 located in the second hairpin structure ( Figure S3 ) . The relatively small sizes and the hairpin loop structures of these snoRNAs qualify them as suitable substrates for Dicer action . To ascertain that miR2 and miR3 are not random degradation products isolated during the initial cloning of small RNAs , their presence in the RNA from Giardia was monitored . The presence of GlsR16 and GlsR17 in total Giardia RNA was repeatedly demonstrated by Northern analysis as anticipated ( data not shown ) . But miR2 and miR3 remained undetectable on these blots presumably due to their relatively low levels . A splinted ligation analysis was then performed on the size-fractionated small RNA samples ( <40 nts ) because of its relatively high sensitivity based on direct labeling of the specific small RNA of interest through a 3′-end specific ligation [26] . The results from this analysis confirmed the presence of miR2 in the original small RNA sample ( Figure 2 ) . The single labeled band with the anticipated size indicates that miR2 is a product of precise processing of GlsR17 in Giardia and not the result of nonspecific degradation . Splinted ligation analysis of miR3 also showed a single specific band with the anticipated size of a 26 nt RNA but not a 21 nt RNA ( data not shown ) . Thus , miR3 is also a likely natural product from GlsR16 in Giardia whereas miR1 is more likely a nonspecific degradation product . A similar analysis of the small RNA derived from retrotransposon GilT failed to produce a specifically labeled band ( data not shown ) . These data indicate that miR2 and miR3 are naturally processed small RNAs from their respective snoRNAs while the retrotransposon GilT small RNA is probably a degradation product . Identification of putative miR2 target sites in the Giardia genome was performed using the miRanda program [27] . Since most known miRNA target sites have been localized to the 3′- UTRs of mRNAs [28] , we focused our target identification to the 3′-UTRs of Giardia open reading frames ( ORF ) . To limit the number of possible candidate target sites , segments of 250 nts ( 50 nts upstream and 200 nts downstream from the stop codon ) were extracted from each of the 9 , 649 ORFs in the Giardia genome database [24] and used to identify possible target sites for miR2 using miRanda ( score threshold = 120 , energy threshold = 20 kcal/mol , and scaling factor = 4 ) . Identification was based on localization of the putative binding site ( between 50 nts upstream and 50 nts downstream from the stop codon ) and the presence of a seed sequence ( 5 consecutive nts , without G:U base pairings , within 3 nts of the 5′ end of the miRNA ) . Of the 9 , 649 ORFs in the Giardia database [24] , 296 ORFs contained potential target sites for miR2 based on the criteria mentioned above . Among these ORFs , 191 were hypothetical proteins and 105 were annotated proteins . Of the latter , 22 were variant-specific surface proteins ( VSP ) and 7 were trophozoite cysteine-rich surface antigens . VSPs are known to have a highly conserved 3′ end [29] . An alignment of 10 randomly chosen VSP sequences indicated that the last 100 nts at the 3′-end are highly conserved ( data not shown ) . The predicted miR2 target site is 30 nts long with a 20 nt segment within the coding region and 10 nts in the 3′-UTR with the required perfect complementation between the seed sequence of miR2 and the target site located in the 3′-UTR . This remarkable conservation of a 3′-UTR sequence among the 22 VSP transcripts could mean that their expression is subject to a common mechanism of regulation mediated by miR2 . To test the potential consequence of in vivo interactions between miR2 and the putative target sites identified in the 3′-UTR of VSP transcripts , six copies of the putative binding sites from VSP AS12 were added to the 3′-UTR of Renilla luciferase gene in a plasmid construct ( Figure 3A ) . Capped mRNA of this chimeric gene was transcribed in vitro ( RL-TS ) and electroporated into Giardia WB trophozoites together with chemically synthesized miR2 . After incubation at 37°C for 5 hrs , the transformed Giardia cells were lysed and assayed for luciferase activity . Expression of RL-TS , when introduced into Giardia alone , was set at 100% . The inclusion of 0 . 5 µg of miR2 reduced the luciferase activity by 23% ( Figure 3B ) , while 1 µg of miR2 decreased luciferase activity by 35% ( Figure 3B ) . Little additional inhibition was observed when the concentration of miR2 was increased to 2 µg ( 40% ) ( Figure 3B ) . Thus , exogenously introduced miR2 does repress the expression of RL-TS in Giardia in a concentration-dependent but saturable manner . Quantitative RT-PCR estimation of the RL-TS mRNA levels in transfected Giardia with or without the presence of exogenously introduced miR2 resulted in two Ct values of 20 . 4±0 . 95 and 21 . 2±0 . 8 , respectively ( Figure 3C ) . This lack of apparent difference in mRNA levels indicates that the miR2 repression of luciferase expression is not due to enhanced mRNA degradation . To confirm that the reduced luciferase activity is attributed to a specific interaction between miR2 and the putative target sites in the 3′-UTR of RL-TS , capped luciferase mRNA without the target sites ( RL ) was transfected into Giardia cells with or without 1 µg exogenous miR2 . The luciferase activity increased in both cases to ∼115% of the control value ( Figure 3D ) , suggesting that the repressive effect of miR2 requires the presence of target sites in the mRNA . It also suggests that the absence of target sites in the mRNA may also relieve the latter from repression by endogenous miR2 so that the expression of RL goes ∼15% beyond that of RL-TS ( Figure 3D ) . To further test the specificity of the interaction between miR2 and its target sites , Giardia cells were transfected with a mutant miR2 , miR2neg ( 5′ p-CAG GGA UUA CAC CGC CCC UAU AGU CC 3′ ) , in which the five underlined nucleotides of the “seed” sequence in miR2 were mutated to the complementary sequence . The miR2neg is not expected to interact with the target site . Luciferase activity of the RL-TS mRNA was not affected when it was introduced into Giardia with 1 µg of miR2neg . This suggests that the interaction between miR2 and its target sites is sequence specific and essential for the repression of luciferase activity ( Figure 3D ) . snoRNAs are essential for the maturation of ribosomal RNA and are localized to the nucleolus . miRNAs , however , are localized to the cytoplasm to regulate gene expression . To verify the localizations of GlsR17 and miR2 , fluorescence in situ hybridization ( FISH ) was performed . A 26 nt RNA probe directed against the 5′ end of GlsR17 was 5′-end labeled with FAM and another 26 nt probe complementary to miR2 at the 3′-end of GlsR17 was 5′ end labeled with Cy3 . Hybridization of these probes to Giardia WB trophozoites resulted in an exclusive localization of GlsR17 to the nucleus with a specific focus in the putative nucleolus ( Figure 4 ) . This outcome is in good agreement with our previous finding that the trimethyl cap , known to cap the snoRNAs in Giardia [30] , was specifically localized to the nucleolus-like organelle in Giardia nucleus [30] . The probe for the 3′-end of GlsR17 , where miR2 is positioned , stained the nucleolus like the GlsR17 5′ probe , but the majority of the stain was found in the cytoplasm ( Figure 4 ) , indicating that miR2 , the presumed product from GlsR17 by Dicer digestion , is localized primarily to the cytoplasm . Thus , in the apparent absence of Drosha and Exportin5 , the 144 nt snoRNA GlsR17 could be transported into the cytoplasm and trimmed to the mature 26 nt miR2 by Dicer in Giardia . A surprising and unexpected result from the FISH experiments was that the majority of nucleolar GlsR17 was found in only one of the two Giardia nuclei ( Figure 4 ) . While many previous studies indicate that the two nuclei in Giardia are virtually identical in many aspects [31] , [32] ( see Discussion ) , our current serendipitous finding may indicate functional differences related to snoRNAs between the two nuclei . Dicer is responsible for the final maturation of miRNAs . We reasoned that if miR2 and miR3 were miRNAs involved in translational repression in Giardia , prior Dicer processing of their precursors would be required for their maturation . Therefore , a reduction in the Dicer level in Giardia would decrease the levels of both miR2 and miR3 . For miR2 , Dicer depletion would also relieve the repression of RL-TS expression by the endogenous miR2 . Introduction of a specific antisense-hammerhead ribozyme RNA into giardiavirus-infected Giardia trophozoites was used to knock down Giardia Dicer mRNA [21] . This resulted in a 60% knockdown of Dicer mRNA as measured by semi-quantitative RT-PCR and an ∼47% decrease in the growth of Giardia after 4 days , indicating a role of Dicer in Giardia proliferation ( Figure S1C and S1D ) . Splinted ligation analysis of size fractionated small RNAs ( <40 nts ) from Dicer knockdown cells showed a drastic decrease in the levels of both miR2 and miR3 ( Figure 5A and 5B ) , indicating that Giardia Dicer is required to produce mature miR2 and miR3 . To ascertain that other RNA species are not nonspecifically affected by the Dicer knockdown , size-fractionated RNAs ( <200 nts ) from the Dicer knockdown and the control cells were compared ( Figure 5C ) . There is little difference among the RNAs with >100 nts between the two cell lines . But a band of ∼85 nts appears enhanced and a smeared band close to the 26 nt region seems diminished in the Dicer knockdown cells , which could represent certain miRNA precursors and total miRNAs , respectively . Dicer knockdown thus does not appear to affect RNA level nonspecifically . Expression of capped RL-TS mRNA transfected into Dicer knockdown cells by electroporation showed a 15% increase from the wild type control ( data not shown ) . This suggests that the lowered miR2 concentration , resulting from the Dicer knockdown , relieves the previously observed endogenous translational repression of RL-TS ( Figure 3D ) . Domain analysis of GlAgo identified a PIWI domain but not a PAZ domain [33] , [34] . Sequence alignments between GlAgo and the Argonautes of various origins , however , was able to identify both a PIWI and a PAZ domain [35] . Since the PAZ domain is known to interact with the small RNAs [36] , [37] , its apparent absence from GlAgo domain analysis raised the question whether GlAgo could bind to miR2 . Recombinant His-tagged GlAgo was expressed and purified from transformed Escherichia coli . Gel shift analysis , using radiolabeled miR2 and recombinant GlAgo , indicated a slower moving radiolabeled band suggesting the formation of a GlAgo-miR2 complex ( Figure S4 , lanes 2 and 3 ) . This binding of miR2 to GlAgo could be efficiently competed off by unlabeled miR2 ( Figure S4 , lane 5 ) . Yeast RNA ( Figure S4 , lane 4 ) was also able to compete off miR2 , indicating a lack of RNA sequence specificity for the binding . This result agrees with the previous observations on binding properties of Argonautes , which indicated that only the presence of a 5′ phosphate and a 3′ hydroxyl in an RNA molecule are required for binding but not sequence specificity [36]–[38] . Thus , GlAgo is apparently capable of binding small RNAs in the same way as the other Argonautes . To confirm that the repression of luciferase expression by miR2 in Giardia requires GlAgo , the GlAgo knockdown cells developed previously were transfected with the RL-TS transcript along with or without synthetic miR2 . The results indicated that , in the absence of 50% of the GlAgo mRNA , expression of the luciferase activity exceeds that of the wild type control by about 20% ( Figure 6 ) . This enhanced activity was not affected by introducing exogenous miR2 into the cells , indicating that , by knocking down GlAgo , the repression of luciferase expression by both endogenous and exogenous miR2 was largely abolished . Thus , GlAgo is clearly playing an essential role in the miR2-mediated repression of luciferase expression in Giardia .
In eukaryotes , snoRNAs are generally regarded as being responsible for the maturation and modification of ribosomal RNA . Two families of snoRNAs have been identified based on sequence conservation , box C/D and box H/ACA . Box C/D snoRNAs contain conserved Box C ( UGAUGA ) and Box D ( CUGA ) elements near the 5′ and 3′ ends , respectively as well as internal copies of these elements termed Box C′ and Box D′ [45] , [46] . An interaction between the 5′ and 3′ termini allows the formation of a stem bringing the Box C and Box D elements together to form a hairpin structure . Box C/D snoRNAs serve as the guide for 2′-O-ribose methylation of ribosomal RNA . Box H/ACA snoRNAs have secondary structures consisting of two hairpins separated by a hinge region and a short tail . The box H ( ANANNA ) element is located in the hinge region while an ACA element is located in the tail , 3 nucleotides from the 3′ end [45] , [46] . Box H/ACA snoRNAs guide the pseudouridylation of ribosomal RNA . Interestingly , snoRNAs containing either the box C/D or box H/ACA motif but without an rRNA antisense region to guide modification of rRNA have been also identified [46] . The role of these “orphan” snoRNAs remains unclear . Dicer could theoretically trim the general hairpin structures of snoRNAs , which range from 60 to 160 nts in length in Giardia , without prior processing to produce potentially functional miRNAs . In Giardia , small nuclear RNAs ( snRNAs ) were originally identified with antibodies directed against the trimethyl cap [47] . Subsequently , cloning and bioinformatics analysis identified approximately 20 characterized snoRNAs and 60 putative snoRNAs [20] , [48] . GlsR16 and GlsR17 were among the 20 characterized snoRNAs and categorized in the orphan group . The 60 putative snoRNAs have not yet been tested experimentally for their real functions in Giardia , though some of them were postulated to modify rRNA due to the presence of anti-sense sequences that target rRNA . It is possible that some or all of them may function as precursors of miRNAs for specific translational repression . This is not to say that the snoRNAs are the only miRNA precursors in Giardia . Other non-coding RNAs , such as the abundant antisense RNAs [18] , could also be a potential source for miRNA . However , in the apparent absence of Drosha/Pasha and Exportin5 , one has to question the ability of Giardia Dicer to digest relatively large RNA hairpins . Giardiavirus , a dsRNA virus with a genome size of 6 , 277 bps , is known to multiply vigorously in the cytoplasm of infected Giardia trophozoites [49] , [50] . This fact appears to rule out the possibility that Giardia Dicer could digest relatively long dsRNA . The snoRNAs identified in Giardia have lengths ranging from 60 to 160 nts [48] , which appear to be the right size of RNA to fold into hairpin structures suitable for digestion by Dicer . Thus , the mature snoRNA could be a Dicer substrate without additional processing . So , it is not entirely unlikely that snoRNAs and perhaps some other small RNAs may constitute the reservoir for miRNA in Giardia . They could be exported from the nucleus by some yet unidentified means and digested by the Dicer for miRNA . Future investigations will provide answer to these intriguing possibilities . In light of Giardia's minimalistic nature , it is reasonable to hypothesize that Giardia utilizes a single RNA processing mechanism for the maturation of both snoRNAs and miRNA precursors with characteristics similar to snoRNA . In human cells , snoRNAs are transcribed by RNA pol II [51] . Maturation of snoRNAs from the primary transcript involves processing of the 3′ end and hypermethylation of the m7G cap to a trimethyl cap ( m2 , 2 , 7G ) [51] . Recently , studies have shown that snoRNAs have a cytoplamic component during biogenesis . snoRNAs injected into the cytoplasm of Xenopus oocytes are imported into the nucleus , indicating a mechanism for nuclear transport [52] . Additionally , snoRNAs have been shown to associate with the nuclear export complex [53] , [54] and reach their maturation in the cytoplasm [55] . Although this mechanism has only been shown for snoRNAs known to be involved in rRNA maturation , it is possible that “orphan” snoRNAs may also mature utilizing the same mechanism . The outcome from our FISH experiments indicated the localization of GlsR17 primarily in the nucleolus and miR2 in the cytoplasm ( Figure 4 ) . A faint signal of GlsR17 was , however , also detectable in the cytoplasm . It is unclear if this signal is due to the presence of GlsR17 in the cytoplasm or background . It is possible that Giardia snoRNAs have a cytoplasmic component during maturation . After processing of the pre-snoRNA in the cytoplasm , the mature snoRNA may either be rapidly imported into the nucleus or processed by Dicer . Therefore , only small undetectable amounts of snoRNA may remain in the cytoplasm . The export of snoRNAs could provide an ideal method for bypassing the requirements for Drosha/Pahsa in producing pre-miRNAs . Although the proposed mechanism for production of pre-miRNAs is unique in Giardia , the remaining aspects of the miRNA pathway appear similar to that found in higher eukaryotes . These similarities include the requirement of Dicer for the production of mature miRNAs and an essential role of Argonaute in translational repression . In the absence of Dicer , mature miR2 and miR3 were virtually abolished . Additionally , both miR2 and miR3 are 26 nts in size , reflecting the expected size of the Giardia Dicer cleavage products based on its crystal structure [17] . GlAgo plays an essential role in miR2-mediated translational repression . The apparent formation of a GlAgo-miR2 complex in vitro and the specific binding of GlAgo to m7G-Sepharose suggests that it functions like a bona fide Argonaute by becoming incorporated into RISC with miRNA and competing for the cap with eIF4E [13] , which results in the inhibition of translation initiation . When GlAgo was partially depleted from Giardia , translation of reporter mRNA RL-TS was increased by ∼20% with or without the exogenously introduced miR2 . This indicates an inability of endogenous as well as the exogenously introduced miR2 to repress the translation of RL-TS when GlAgo was knocked down . Thus , the cytoplasmic component of the miRNA pathway in Giardia functions similarly to those seen in other eukaryotes . A third of the mRNAs of annotated proteins identified to carry a potential target site for miR2 encode either VSPs or trophozoite cysteine-rich surface antigens . Giardia contains approximately 235–275 VSP genes in clusters of two to nine in a head to tail orientation [14] . Only a single VSP , however , is expressed on the cell surface at any given time [29] . Expression of multiple VSPs is only observed during VSP switching or excystation [29] , [56] . Recently , Kulakova et al . showed that the activation of VSP expression is regulated by acetylation of the histone upstream of the gene [57] . It is , however , still unclear how VSP exchange occurs . During VSP exchange , expression of the currently expressed VSP must be repressed and the expression of a new VSP has to be up-regulated . In this genetic turmoil , it is possible that multiple VSP mRNAs are transcribed . The role of miR2 may be to differentially limit the translation of a particular family of 22 VSP mRNAs . It is possible that the 235–275 VSP transcripts can be classified into several families each possessing a similar 3′-UTR targeted by a specific miRNA . When the level in one of these miRNAs is lowered , a specific family of VSP will be over expressed , though it is still not clear how only one of the family members will eventually be expressed on the cell surface . The 22 identified VSP genes carry a similar miR2 target site , but they are not identical . Differences in the free energy of binding between miR2 and individual VSP target sites could differentially regulate the expression of each VSP so that only one of them eventually becomes expressed on the cell surface when the level of miR2 is lowered to a certain specific range . This may contribute to the pathogenicity of Giardia by limiting the number of antigens exposed to the immune system at a given time and thereby increasing the number of VSPs novel to the host . As a consequence , Giardia may be able to evade the host immune response . The precise role of miR2 in regulating expression of the 22 VSP genes needs to be investigated further . There has been an accumulation of experimental data over the past years indicating that the two nuclei in Giardia are virtually identical . Kabnick K . S . and Peattie D . A . showed that both nuclei contain equivalent rDNA based on in situ hybridization and that both nuclei are transcriptionally active based on the incorporation of [3H] uridine [31] . Yu et al . used FISH and demonstrated that each of the two nuclei has a complete copy of the genome and are partitioned equally during cytokinesis [32] . Therefore , the localization of GlsR17 to the nucleoli of a single nucleus was surprising . Localization of the other identified snoRNAs will determine if this is a unique occurrence or a general phenomenon in Giardia . Differences in the number of chromosomes [58] as well as discrepancies in the number of nuclear pores [59] between the two nuclei have recently been described in Giardia . Since the number of nuclear pores has generally been correlated to transcription activity in other eukaryotes [59] , it suggests that one of the two nuclei may have higher transcriptional activities than the other . It is possible that one of the two nuclei may be solely responsible for production and export of some or all of the snoRNAs . A fascinating aspect may concern the apparent exclusive import of the snoRNAs back into the same original nucleus and the mechanism dictating it . Clarifications of this unusual finding will have to wait for further investigation . Giardia is a deeply branched eukaryote showing a combination of prokaryotic and eukaryotic features [14] . The presence of the miRNA pathway in this organism suggests that miRNA-mediated repression of translational initiation could be an ancient mechanism of gene regulation and that the snoRNAs may represent the original miRNA precursors . During evolution , snoRNAs , while maintaining their original function in ribosomal RNA maturation , may have become involved in gene regulation . These small non-coding RNAs would make ideal substrates for a primative RNase III enzyme before the evolution of new enzymes for the production of miRNAs from other much larger precursors . Since snoRNAs have been routinely discarded from the libraries of potential miRNA precursors in the past , it is not inconceivable that some of the snoRNAs in higher eukaryotes may still assume the role of miRNA precursors today , and could be readily tested .
The Giardia Argonaute was PCR amplified from Giardia genomic DNA using primers Ago full U-2 ( 5′ GAG CCC GGG TCA CTA GTG CCA TGG TAG CAG ATG TTG TCA C 3′ ) and Ago full L-2 ( 5′ GAG CTC GAG GCG GCC GCC TAG TGG TGG TGG TGG TGG TGT ATG AAG AAT GGT CTG TAC T 3′ ) . The amplified product was cloned into the pGEM-T Easy vector ( Promega ) , sequenced , and subcloned into the pET28b ( Novagen ) expression vector using NcoI/XhoI . The pET28b GlAgo was electroporated into BL21 ( DE3 ) cells containing the pG-KJE8 plasmid ( Takara ) , which provides over-expression of chaperone proteins . An overnight culture was diluted 1∶100 into fresh LB media containing 10 ng/ml tetracycline and 4 mg/ml arabinose and incubated at 37°C for 1 hour to allow for the expression of chaperone proteins . Expression of GlAgo was induced with 0 . 1 mM IPTG and incubated at room temperature for 5 hours . Pelleted cells were lysed with BugBuster ( Novagene ) and bound to Ni-NTA beads in the presence of 40 mM imidazole , 5 mM ATP , and 10 mM MgCl2 at 4°C overnight . Beads were washed with 15 ml of wash buffer ( 20 mM sodium phosphate , pH 60; 500 mM NaCl ) containing 60 mM imidazole and the protein was eluted in 4 ml of wash buffer containing 150 mM imidazole , concentrated and transferred to storage buffer . Small RNAs were cloned following the protocol from the Ambros lab ( http://banjo . darthouth . edu/lab/MicroRNAs/Ambros_microRNAcloning . htm ) . In short , Giardia cells were lysed and small RNAs were enriched using the mirVana kit from Ambion . The purified small RNAs were further size fractionated using the Ambion FlashPAGE fractionator to select small RNA less then 40 nts . The small RNAs were then linked to a 3′ linker ( AMP-5′p-5′p/CTG TAG GCA CCA TCA AT di-deoxyC- 3′ ) and size fractionated on a 15% urea-polyacrylamide gel . Those with attached 3′ linkers ( ∼40 nt ) were electroeluted from the gel and ethanol precipitated . The 3′ linked small RNAs were then ligated to a 5′ linker ( 5′- ATC GTrA rGrGrC rArCrC rUrGrA rArA –3′; “r” denotes RNA ) . This step is designed to select for small RNAs with a 5′ monophosphate , as would be expected from Dicer processing . The reaction was cleaned by phenol/chloroform extraction followed by ethanol precipitation . The cleaned product was used in an RT-PCR reaction to make cDNA , which was gel purified , digested with BanI and ethanol precipitated . The cleaned product was used in a ligation reaction to concatamerize the PCR product together . The reaction was run on an agarose gel and fragments between 600–1000 bp were isolated . The isolated fragments were re-amplified using PCR and cloned into pGEM-T Easy using the pGEM-T Easy kit ( Promega ) . Colonies containing insert were sequenced . The Hammerhead ribozyme was incorporated into the GlAgo antisense RNA using recombinant PCR . Briefly , PCR1 containing the 5′ complementary sequence and the ribozyme was amplified using Ago HHR PCR1 F ( 5′ CGC GCG CTC GAG CTC CCA GAT TGA CCT GGG ATC 3′ ) and Ago HHR PCR1 R ( 5′ CTG CCC CTG AAC TAT AGA GTG CTG ATG AGT CCG TGA GGA CGA AAC TCT GAA AAC CTT TCC GTT G 3′; ribozyme underlined ) . PCR2 containing the 3′ complementary sequence was amplified using Ago HHR PCR2 F ( 5′ CAC TCT ATA GTT CAG GGG CAG 3′ ) and Ago HHR PCR2 R ( 5′ GCG CGC GAG CTC CTT CAA TGG TAA CTA TAC GAG 3′ ) . Purified PCR1 and PCR2 were then used as the template for the recombinant PCR using Ago HHR PCR1 F and Ago HHR PCR2 R . The recombinant PCR was cloned into pGEM-T Easy and sequenced . The Hammerhead ribozyme surrounded by 500 nt of complementary GlAgo sequence was then cloned into pC631pac using the XhoI restriction site . This clone was linearized with NruI and used for in vitro transcription . The transcribed RNA was transfected into giardiavirus-infected Giardia . Puromycin ( 100 µg/ml ) was used to select for expression of the ribozyme . A similar approach was used to incorporate a Hammerhead ribozyme into the Dicer antisense . PCR3 was obtained using primers Xho-Dicer 5′ ( 5′ GCC TCG AGT TTA GTA GGA ATG CAT GCT TTG G 3′ ) and Dicer RZ2 ( 5′ GGG TAG AAT CGA TCC CAA GAA CCT GAT GAG TCC GTG AGG ACG AAA CAT AAA GAG ACC AGC 3′; ribozyme underlined ) . PCR4 was obtained using primers Xho-Dicer 3′ ( 5′ GCC TCG AGG GAT ATT ACA CTA CGC ATC AGC 3′ ) and Dicer RZ1 ( 5′ GCT GGT CTC TTT ATG TTT CGT CCT CAC GGA CTC ATC AGG TTC TTG GGA TCG ATT CTA CCC 3′; ribozyme underlined ) . Purified PCR3 and PCR4 were used as templates for recombinant PCR using Xho-Dicer 5′ and Xho-Dicer 3′ . After cloning into pGEM-T Easy and sequencing , the PCR product was cloned into the pC631neo vector using XhoI and the in vitro transcript transfected into giardiavirus-infected Giardia as described above . Neomycin ( 800 µg/ml ) was used to select for expression of the ribozyme . Giardia WB trophozoites were grown in modified TYI-S-33 media to a density of 107 per ml , washed twice in phosphate buffered saline ( PBS ) , once in electroporation buffer ( 10 mM K2HPO4–KH2PO4 ( pH 7 . 6 ) , 25 mM HEPES ( free acid ) , 120 mM KCl , 0 . 15 mM CaCl2 , 2 mM EGTA , 5 mM MgCl2 , 2 mM ATP , 4 mM Glutathione ) , and finally resuspended in electroporation buffer . RL-TS mRNA ( 3 . 5 µg ) , yeast tRNA ( 125 µg ) , and , if needed , 1 µg of 5′-phosphate-miR2 RNA ( miR2 ) ( IDT ) were added to the cell suspension , incubated on ice for 10 minutes and then subjected to electroporation using a Bio-Rad Gene Pulser Xcell ( Voltage: 450 V , Capacitance: 500 µF , Resistance: ∞ ) . Cells were then incubated on ice for 10 minutes and added to pre-warmed culture medium . The transfected cells were incubated at 37°C for 5 hours , pelleted , washed once in PBS , and lysed using the Renilla luciferase assay kit ( Promega ) . The lysate was centrifuged at 12 , 000 g for 2 min to remove cellular debris . The cleared lysate was used to test for Renilla luciferase activity . The protein concentration of the cleared lysate was measured by the Bradford method ( Bio-Rad ) and used to normalize the luciferase activity . Total RNA was isolated from Giardia using Trizol ( Invitrogen ) while RNA <200 nts was isolated using the mirVana kit ( Ambion ) . Further fractionation of total RNA to <40 nts was accomplished by standard denaturing PAGE and electroelution or by using the Ambion FlashPAGE followed by ethanol precipitation . Splinted ligation was preformed as previously described [26] . Size fractionated RNA was incubated with 100 pmoles of the “bridge” oligo B1 ( 5′ C3 spacer-GAA TGT CAT AAG CGG GAC TAT AGG GGC GGT GAT TAG GCT G–C3 spacer 3′ ) containing the miR2 binding site ( underlined ) and 100 fmoles of the “linker” oligo L1 ( 5′ CGC TTA TGA CAT TCddC 3′ ) with 20 mM Tris-HCl ( pH 8 . 0 ) and 75 mM KCl . The reaction mixture was incubated at 95°C for 1 min , 65°C for 2 min and 37°C for 10 min . Finally , 1X T4 DNA ligase buffer and 10 U of T4 DNA ligase ( NEB ) was added to the reaction and incubated at 30°C for 1 hour . The ligase was heat inactivated by incubation at 75°C for 15 min . The reaction mixture was loaded on to a pre-run 15% denaturing urea-polyacrylamide gel and quantified using a PhosphorImager . Giardia WB trophozoites were harvested by placing culture tubes on ice for 15 minutes and centrifuging to pellet the cells . The cells were suspended in 1 ml of modified TYI-S-33 culture medium , placed on cover slips pretreated with 0 . 1% poly-L-lysine , and incubated at 37°C for 30 minutes to allow the trophozoites to adhere . They were then fixed in 4% paraformaldehyde for 30 minutes at room temperature and washed with PBS and 2X SSC ( 300 mM NaCl , 30 mM sodium citrate ) for 5 minutes each . The cells were permeabilized with 0 . 5% Triton X-100 for 5 minutes at room temperature and dehydrated in 70% ethanol followed by 100% ethanol for 5 minutes each . The dehydrated cells were denatured with 2X SSC in 70% formamide for 2 minutes at 70°C and dehydrated again with cold 70% and 100% ethanol . Salmon DNA ( 10 µg ) , yeast tRNA ( 25 µg ) , and 100 pmoles of the GlsR17 RNA 5′-UTR probe ( 5′ 6-carboxyfluorescein ( FAM ) –CCC GGA UCC UCA CCA CGA GUA AAC CC 3′ ) and miR2 RNA 3′-UTR probe ( 5′ Cy3–GGA CUA UAG GGG CGG UGA UUA GGC UG 3′ ) ( IDT ) were added to 100 µl of 100% formamide , heated at 75°C for 10 minutes and placed immediately on ice . The probe was mixed with an equal volume of hybridization buffer ( 4X SSC , 20% dextran sulfate , and 4 mg/ml BSA ) , added to the cover slip and incubated overnight at 37°C . The cover slips were washed 3 times with 0 . 1% SSC in 50% formamide at 50°C for 5 minutes each to remove unhybridized probe , placed facedown on clean glass slides with 1 drop of Vectashield ( Vector Labs ) mounting media with DAPI ( 4′ , 6 diamidino-2-phenylindole ) and sealed with paraffin wax . Cells were examined using a Nikon TE2000E motorized inverted microscope equipped with 60X bright-field and epifluorescence optics . Images were acquired with the NIS-Elements Advanced Research software ( Nikon ) and analyzed with ImageJ ( http://rsbweb . nih . gov/ij/index . html ) . | Gene regulation in Giardia lamblia , a primitive parasitic protozoan responsible for the diarrheal disease giardiasis , is poorly understood . There is no consensus promoter sequence . A simple eight–base pair AT-rich region is sufficient to initiate gene transcription in this organism . Thus , the main control of gene expression may occur after the stage of transcription . The presence of Dicer and Argonaute homologs in Giardia suggested that microRNA ( miRNA ) -mediated translational repression could be one mechanism of gene regulation . In this work , we characterized the presence of the miRNA pathway in Giardia as well as identified the novel use of small nucleolar RNA ( snoRNA ) as miRNA precursors . Potential target sites for one small RNA ( miR2 ) were identified with the miRanda program . In vivo reporter assays confirmed the specific interaction between the target sites and miR2 . A ribozyme-mediated reduction of Dicer and Argonaute in Giardia showed that the former is required for miR2 production whereas the latter functions in mediating the inhibition of reporter expression , which agrees with the roles of these two proteins . This is the first evidence of miRNA-mediated gene regulation in Giardia and the first demonstration of the use of snoRNAs as miRNA precursors . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"molecular",
"biology/translation",
"mechanisms",
"molecular",
"biology/translational",
"regulation",
"microbiology/parasitology"
] | 2008 | snoRNA, a Novel Precursor of microRNA in Giardia lamblia |
The Chikungunya virus infection zones have now quickly spread from Africa to parts of Asia , North America and Europe . Originally thought to trigger a disease of only mild symptoms , recently Chikungunya virus caused large-scale fatalities and widespread economic loss that was linked to recent virus genetic mutation and evolution . Due to the paucity of information on Chikungunya immunological progression , we investigated the serum levels of 13 cytokines/chemokines during the acute phase of Chikungunya disease and 6- and 12-month post-infection follow-up from patients of the Italian outbreak . We found that CXCL9/MIG , CCL2/MCP-1 , IL-6 and CXCL10/IP-10 were significantly raised in the acute phase compared to follow-up samples . Furthermore , IL-1β , TNF-α , Il-12 , IL-10 , IFN-γ and IL-5 had low initial acute phase levels that significantly increased at later time points . Analysis of symptom severity showed association with CXCL9/MIG , CXCL10/IP-10 and IgG levels . These data give insight into Chikungunya disease establishment and subsequent convalescence , which is imperative to the treatment and containment of this quickly evolving and frequently re-emerging disease .
The Chikungunya virus ( CHIKV ) , an arthropod-borne virus ( arbovirus ) , is a single-stranded positive-sense RNA virus with three genotypes . The virus is of the Alphavirus genus in the Togaviridae family [1] , [2] . CHIKV has been shown to infect and be transmitted by Ae . aegyptii and Ae . albopictus mosquitoes . It was identified in East Africa in the early 1950s and since then has caused epidemics in continental Africa , the Indian Ocean region , and countries of Southeast Asia such as India , where since 2006 suspected cases have been estimated to be 1 . 39 million , and Singapore [3]–[6] . The only reported outbreak outside these areas was in Italy in the Emilia Romagna region in 2007 . Small non-epidemic imported cases have been reported in other regions such as North America , France and Japan , which were caused by travelers returning from affected areas [7]–[9] . The epidemic occurring on La Reunion Island in the Indian Ocean remains the most devastating of all CHIKV outbreaks where over one-third of the population was affected [10] . During this outbreak , the CHIKV acquired a genetic mutation allowing the new vector Ae . albopictus mosquito to carry the virus where previously CHIKV only circulated in Ae . aegyptii mosquitoes [10] , [11] . The Ae . albopictus differs in susceptibility to various genetically different isolates of the virus compared to the Ae . aegyptii [12] . CHIKV is now of global health concern since expansion of mosquito vectors has created potential for the Chikungunya virus to spread to temperate areas as Ae . albopitcus inhabits regions in North America and Europe [2] , [13] . CHIKV infection is clinically characterized by the sudden appearance of high fever , rash , headache , nausea , vomiting , myalgia and arthalgia or severe joint pain . Severe joint pain is the defining symptom of CHIKV disease [11] . The word Chikungunya originated from the Tanzanian and Mozambique region of Africa meaning that which bends up . A bent posture is often taken by those in severe joint pain after being infected with CHIKV . CHIKV symptoms start 4 to 7 days after exposure and most resolve within the acute phase of the disease . Although the acute phase lasts approximately 2 weeks , joint pain can persist for months or years following initial infection [1] , [14] , [15] . Minimal research has been done investigating the immune response following CHIKV infection . Currently , there is no CHIKV specific therapeutic available . The Italian outbreak of CHIKV spread through communities surrounding the city of Ravenna during August to October 2007 and also involved the major Italian city of Bologna [15] , [16] . A recorded 254 people were identified to be infected through the Ae . albopictus mosquito which has inhabited the Emilia Romagna region since 1990 [14] , [17] , [18] . The virus brought to the Emilia Romagna region by a traveler returning from a CHIKV affected country was of the Central/East African genotype and matched most closely ( 100% amino acid identity ) with the IND-06 virus isolated from the Reunion Island outbreak [14] , [17] . The amino acid identity confirmed that this virus included a substitution mutation in the E1 envelope protein ( E1-A226V ) [19] which is important for viral entry into host cells . This mutation was acquired during the 2005-2006 Indian Ocean CHIKV outbreak and increased the virus's infectivity to the Ae . albopictus mosquito [20] . Cytokines are important immune mediators that conduct immune responses . Recently , cytokine profiles have been investigated in CHIKV infected humans by two groups [21] , [22] . Ng and colleagues established cytokine profiles from 10 CHIKV patients that were infected during the Singapore 2007 CHIKV outbreak [22] . Although this study reported that IL-1β , IL-6 and RANTES were correlated with severe acute phase CHIKV disease , cytokine profiles were not determined for the progression and convalescence of the disease . Here we investigated cytokine profiles during the acute phase and 6- and 12-month follow-up of CHIKV infected patients of the Italian 2007 outbreak . Since CHIKV disease can have severe acute phase symptoms and be followed by persistent symptoms in the convalescence phase it was important to investigate the immune response responsible for these maladies . Furthermore , the Italian CHIKV included the A226V mutation and the Singapore virus did not . Furthermore , we analysed the relationship between cytokine levels and patient severity , and IgG levels linking high CXCL9 , CXCL10 and IgG levels with disease severity . Therefore , the results presented here are virus specific and reflect previously unreported cytokine profiles which may be important for the development of future therapeutics for CHIKV outbreaks .
Patients all gave written consent to the participation in scientific studies . Permission to perform scientific studies was given by Comitato Etico di Area Vasta Romagna Et IRSTof the Servizio Sanitario Regionale Emilia-Romagna , Italy . Since the immune response during CHIKV disease has not been extensively investigated , our objectives were to create a clear clinical picture of CHIKV disease at the acute phase and during convalescence at 6- and 12-month follow-up by cytokine profiling . To achieve this objective , we investigated the cytokine profiles from patients at the acute phase and at 6- and 12-month follow-up . Included patients were from the region of Emilia-Romagna in north-east Italy suspected to be infected with CHIKV since they showed symptoms such as myalgia , severe back and joint pain , headache , and skin rash . Collaboration with the regional microbiology reference laboratory of Bologna University and the Department of Infectious and Parasitic Diseases of the National Institute of Health in Rome was initiated and identified the patients as having CHIKV . The clinical criteria was described as acute onset of fever ( >38 . 5°C ) and severe arthralgia not explained by other medical conditions . CHIKV infection was confirmed by one or more of the acute phase tests: virus isolation , reverse transcriptase-PCR ( RT-PCR ) positive for CHIKV nsp1 gene , seroconversion to virus-specific serum antibodies collected at least 1 to 3 weeks apart , or presence of virus-specific IgM antibodies in a single serum sample collected [15] . Acute CHIKV patient samples were determined to be in the viral stage if the sample was PCR positive for CHIKV ( 7 patients ) , in the IgM antibody initiation stage if the sample was PCR negative , IgM positive , IgG negative ( 6 patients ) or in the seroconversion stage ( 22 patients ) if the sample was PCR negative , IgM positive and IgG positive . The samples were considered to be high IgG if the IgG level was greater than 6400 ( 6 months ) ( 31 patients high out of 50 ) or greater than 3200 ( 12 months ) ( 20 patients out of 50 ) . IgG levels below or equal to these thresholds were considered low IgG . CHIKV patients were considered to be non-symptomatic ( 15 patients out of 50 at the 6-month follow-up; 34 patients out of 50 at the 12 month follow-up ) , have mild symptoms ( 21 patients out of 50 a the 6 month follow-up; 14 patients out of 50 at the 12 month follow-up ) , or have severe symptoms ( 14 patients out of 50 at the 6 month follow-up; 2 patients out of 50 at the 12 month follow-up ) based on their responses to a questionnaire at the time of sampling which was based on: articular pain , muscle pain , mono-arthritis , oligo-arthritis , symmetric polyarthritis , asymmetric polyarthritis , tenosynovitis , arthralgia and fibromyalgia . Control samples were collected from 10 healthy volunteers screened for symptoms of viral infection . Blood samples were collected from consenting CHIKV positive patients at the time of diagnosis . Viral infection was determined as described above . Two follow-up samples were then collected from each patient at the 6-month evaluation and the 12-month evaluation . After sampling , serum was extracted and immediately frozen at −80°C until serum analysis . Serum samples were analyzed for cytokine levels using BD™ Cytometric Bead Array ( CBA ) Human Chemokine Kit , Human Inflammatory Cytokine Kit , and Human Th1/Th2 Cytokine Kit ( BD Biosciences ) according to the manufacturer's instructions for a total of 13 cytokines . Capture Beads were added to the serum sample followed by the PE detection reagent . The samples were then incubated for 3 hours at room temperature and washed with the assay Wash Buffer and resuspended again in Wash Buffer for analysis on the Flow Cytometer . CBAs were then run on a BD FACSCalibur Flow Cytometer . CBA data was analysed using SPSS statistical software . Box-whisker plots were created from the CBA FACS raw data . Six-month and 12-month samples were compared to the acute samples using the non-parametric two-tailed Wilcoxon signed-rank test for related samples to determine statistical significance . Each acute phase , 6-month and 12-month cytokine sample sets were statistically compared to healthy control CBA data using the non-parametric two-tailed Mann Whitney Test for unrelated samples .
CHIKV causes a disease of crippling joint pain that has affected most of Asia and has demonstrated the capability to spread to non-tropical areas such as Europe and parts of North America [1] . Cytokines are inflammatory mediators and their balance is often associated with inflammatory disease [23] . Previously , the cytokine profiles of acute phase CHIKV patients have been examined [22] . Here we profiled cytokine levels in acute phase and 6- and 12-month follow-up CHIKV patient serum samples to determine a cytokine signature that may correlate with acute symptoms , following persistent joint pain and/or disease resolution . Blood samples were collected from 50 patients suffering from CHIKV infections during the 2007 Italian outbreak . Serum separated from whole blood was analyzed by cytokine bead analysis ( CBA ) for 13 cytokines with the intention of determining a cytokine profile during CHIKV acute phase and disease convalescence . Three cytokine profiles emerged from our data: decreasing , increasing and no-trend . The first trend showed cytokine levels significantly higher in the acute samples compared to the follow-up time points revealing a decreasing pattern as patients left the acute phase . CXCL9/MIG ( CXCL9 ) , IL-6 , CCL2/MCP-1 ( CXCL2 ) and CXCL10/IP-10 ( CXCL10 ) cytokines had significantly decreased at both 6-month and 12-month follow-ups ( Figure 1 ) . Interestingly , some patients had extremely high levels in the acute phase; CXCL9 and CXCL10 levels decreased 1000 fold to 10 , 000 fold during convalescence . Furthermore , the decreasing trends for IL-6 , CXCL9 , and CXCL10 reached similar levels as those of the community control levels ( shown by dotted line ) . CCL2 levels decreased significantly lower than the control levels by 12 months . Taken together , this data demonstrated that CXCL9 , IL-6 , CCL2 and CXCL10 were initially increased with acute CHIKV infection and decreased over time . The second cytokine trend that emerged described cytokines that significantly increased following the acute phase . Cytokine profiles that were markedly lower in the acute phase and subsequently increased included IL-1β , TNF-α , IL-12 , IL-5 , IL-10 and IFN-γ ( Figure 2 ) . The cytokine increase was more gradual than the previous decreasing trend , where fold changes were generally closer to 2 . Both the 6- and 12-month follow-up were statistically increased compared to acute values for IL-5 levels . IL-1β , TNF-α , IL-12 , IL-10 and IFN-γ had significantly increased by 12 months . Even though the average for these cytokines had also risen by 6 months it was not significant . Furthermore , the increasing trends for TNF-α , IL-5 , and IL-10 reached similar levels to the community control levels ( shown by dotted line ) and IFN-γ reached significantly higher than controls at 12 months . Interestingly , although IL-1β and IL-12 increased through the observed time , these cytokines stayed significantly lower than those of the controls . This data showed that cytokines IL-1β , TNF-α , IL-12 , IL-5 , IL-10 and IFN-γ increased in the convalescence phase of CHIKV infection . No significant change was seen for IL-2 , IL-4 , and IL-8 from the acute phase to the 12-month follow-up ( Figure 3 ) . Interestingly , IL-2 reached similar levels to those of the controls where IL-8 and IL-4 remained significantly raised and lowered , respectively . Previously , we have shown that the acute phase of West Nile Virus ( WNV ) can be described in 3 stages [24] . Since the samples from the CHIKV patients were taken at various stages of the acute phase , we went on to determine if there were cytokines marking the viral ( V ) , IgM antibody initiation ( AI ) or seroconversion ( SC ) stage . Acute CHIKV patient samples were determined to be PCR positive for CHIKV ( viral stage ) , IgM positive , IgG negative ( IgM antibody initiation stage ) or IgM positive , IgG postive ( seroconversion stage ) . Samples were put into Viral stage ( V ) , Antibody Initiation stage ( AI ) or Seroconversion stage ( SC ) according to the presence of CHIKV , IgM and IgG antibodies . Cytokine Bead Array analysis of the serum samples showed a significant decrease in CXCL10 and IL-10 from the viral stage to the seroconversion stage of the acute phase ( Figure 4 ) . The median of CXCL10 in the viral stage was approximately 7000 pg/ml and dropped to less than 1000 pg/ml after seroconversion . Interestingly , the IL-10 median decreased by 3 fold from the viral stage to the seroconversion stage . Next we sought to determine if the high levels of IgG in the follow-up phases were also significantly associated with cytokine levels compared to the cytokine levels of patients with low levels of IgG . The patients were put into an IgG high group ( H ) or an IgG low group ( L ) and the levels of each cytokine were statistically compared for each group using the Mann-Whitney U Test . In the 6-month follow-up phase CXCL9 , CXCL10 and Il-6 were found to be statistically different between the high IgG group and the low IgG group ( Figure 5A ) . High levels of all 3 cytokines were associated with high levels of IgG antibodies . IL-10 is also shown for comparison since it was statistically significant during the acute phase breakdown and the 12-month follow-up . Interestingly , in the 12-month follow-up phase , CXCL9 was found to be statistically higher in the IgG high group where IL-10 was significantly lower in the high IgG group ( Figure 5B ) . CXCL10 is shown for comparison at 12 months although it was not significantly different between the IgG high and IgG low groups . In summary , the results suggested that CXCL9 , CXCL10 and Il-6 were associated with IgG levels in the 6-month follow-up phase and CXCL9 and IL-10 in the 12-month follow-up phase . IL-1β , IL-6 and RANTES were found to be associated with symptom severity of the Singapore 2007 CHIKV outbreak [22] . After determining the cytokine profiles of our Italian 2007 CHIKV patients during their disease resolution , we next sought to determine the association between symptom severity and cytokine levels . Patients were determined to be non-symptomatic ( N ) , to have mild symptoms ( M ) or to have severe symptoms ( S ) depending on their responses to a questionnaire . The cytokine levels were then grouped by symptom level and a Mann-Whitney U test was used to determine significance among the severity groups for each cytokine . CXCL10 and CXCL9 were found to be significantly increased in the patients with mild and severe symptoms at 6 months following initial infection compared to the patients reporting no symptoms ( Figure 6A ) . A 2 fold difference was seen between the medians of the non-symptomatic and severe patients for CXCL10 and a 5 fold difference for CXCL9 . No statistical difference was seen for any of the 13 cytokines profiled for the 12-month follow-up . CXCL10 and CXCL9 at 12 months are shown for comparison ( Figure 6B ) . Moreover , we analyzed the IgG levels at the 6-month follow-up in patients with no symptoms , mild symptoms , and severe symptoms . The results showed IgG levels were statistically increased with symptom severity ( Figure 6C ) . Taken together , these results suggested CXCL10 , CXCL9 and IgG to be possible markers of CHIKV severity during early phases of disease resolution .
CHIKV disease is a self-limiting disease caused by an alphavirus of the Togaviridae family . Although historically the virus only caused a disease of mild symptoms , a recent outbreak on La Reunion Island caused significant mortality due to genetic alterations [1] , [20] , [22] , [25] , [26] . Here we have investigated the immune response of an Italian population infected with the Indian Ocean genotype of CHIKV and have generated a cytokine signature for the initial infection to convalescence phase of CHIKV disease , the signature of viral and antibody production phases , and identified cytokines raised in patients with severe symptoms . We found that initial infection and the subsequent convalescence were described by a set of decreasing and increasing cytokines . Furthermore , we have shown that CXCL10 and IL-10 levels were associated with the viral stage of the acute phase and CXCL10 and CXCL9 with high IgG levels of the 6-month follow-up . As well , CXCL10 and CXCL9 were markers of symptom severity . Importantly , these identified signatures depict the immunological programs and may be key to the development of therapeutics for the frequently re-emerging CHIKV disease . Our analysis has indentified 2 cytokine profiles that followed disease onset and continued with disease progression/convalescence . We found CXCL9 , CXCL10 , CCL2 and IL-6 levels were high in the acute phase and decreased as patients convalesced . Conversely , the trend for TNF-α , IL-1β , IL-2 , IL-5 and IL-12 were low initially and increased as patients began to recover from acute illness . High levels of CXCL9 , CXCL10 , CCL2 , and IL-6 in the acute phase possibly indicated an inflammatory program initiating adaptive T-cell immunity [27] . CXCL9 and CXCL10 are both chemokines induced by IFN-γ and are part of the chemokine program that regulates the migration of monocytes/macrophages , memory T cells and NK cells and are associated with the polarization of T cells [27] . IL-6 , a pleiotropic cytokine , has a destructive role in rheumatoid arthritis ( RA ) , contributing to joint inflammation as well as joint pain [28] , [29] . The increased IL-6 levels in the acute phase of our study may be the initiating factor of the severe joint pain symptoms reported in CHIKV patients which mimics RA . Furthermore , IL-6 is important during acute phases of the disease by acting as an important immune mediator of fever activating muscle metabolism to increase core body temperature [28] . Since fever is a common symptom of acute CHIKV disease , it is highly probable that the high IL-6 levels were contributing to the acute fever and the IL-6 decreasing trend followed patient core body temperature as it returned to resting temperature in the follow-up . A second host immune response profile was characterized by TNF-α , IL-1β , Il-10 , Il-12 , IFN-γ and IL-5 , which increased from the acute phase into convalescence . Interestingly , TNF-α and IL-1β , which are known to co-induce the other's expression , are both involved in chronic inflammatory diseases such as RA , chronic hepatitis B and C infection [23] , [29] , [30] . Importantly , TNF-α and IL-1β are main contributors to joint pain , which is the major symptom of RA . An internal balance of TNF-α or IL-1β levels is imperative as mis-regulation of either has been shown to be a major proponent of chronic diseases ( RA ) . Our data indicated that these cytokines increased significantly during patient convalescence above those of the control group , but were not statistically changed in patients reporting mild or severe symptoms . These data may imply that the increased levels of TNF-α and IL-1β in the convalescence phase were not a major contributor of chronic damage causing persistent severe joint pain symptoms during the Italian outbreak even though these cytokines have previously been found to play a destructive role in chronic inflammatory diseases . Furthermore , TNF-α immunomodulators have previously been used as a common treatment for RA and IL-1β immunomodulators are effective with other chronic diseases such as systemic-onset juvenile idiopathic arthritis and in adult onset Still's diseases [29] , [31] . Taken together , these findings suggest TNF-α and IL-1β therapies would not be effective controlling the prolonged symptoms of CHIKV disease since the raised levels during convalescence were not associated with patient severity . Previously , cytokine profiles have been analyzed from patients during an Asian outbreak of CHIKV [22] . Ng and colleagues , investigated 30 cytokines and growth factors from 10 acute phase CHIKV patients , determined that the levels of 8 cytokines , 2 chemokines and 3 growth factors were significantly raised in patients compared to those of the control group . In accordance with their data , IL-6 , CXCL9 and CXCL10 were also increased in the acute phase of our patients as compared to control . Since the results from the previous study did not follow the patients during convalescence , our study added significant insight to the progression of CHIKV disease . We found that these three cytokine levels decreased as the patients convalesced as discussed above . Conversely , we found CCL2 also to be increased in the acute phase compared to that of the control group , which was unchanged in the Ng study . Furthermore , Ng et al . found IL-5 and IL-10 were significantly increased in the acute phase where our data indicated that IL-5 and IL-10 were initially low and below control levels and increased following the acute phase . These discrepancies can possibly be explained by the patient populations: the Ng study patients and our patients were from significantly varied genetic backgrounds ( Asian and European , respectively ) . Therefore , differences in immune response may reflect variations in immunological genetic programs . As well , the virus that caused the two outbreaks also differed genetically . Even though the virus that caused the Singapore outbreak was the same genotype as the Italian virus , the virus that infected the Italian patients had the A226V mutation in the E1 gene which was acquired on La Reunion Island [19] . Although the impact of this mutation has been shown to increase vector infectivity [20] , the mutation has not been investigated on the human or mammalian immune response . Previously , we have identified 3 stages of the acute phase of WNV; a viral stage , antibody initiation stage , and seroconversion stage [24] . As the viral load decreased in the WNV patients , IgM antibody levels were initiated and followed by IgM conversion to IgG , thereby marking the 3 stages of the acute phase . From this work we were able to identify the stages of the CHIKV patients and compare their respective serum cytokine levels . We found CXCL10 and IL-10 levels decreased as patients progressed through the viral stage to seroconversion . Since CXCL10 is often correlated with viral load , this observation was not surprising [32] . High IL-10 levels in the viral stage may act in an effort to control the IFN-γ program [33] shown by high CXCL10 levels . Furthermore , plasma levels of CCL2 , IL-6 as well as CXCL10 have all been correlated with viral loads in virus infected individuals [24] , [34] , [35] . It is possible that the decrease of CCL2 and IL-6 we observed subsequent to the acute phase [14] correlated with viral clearance , although not with antibody levels . Analysis of the cytokine/antibody response was carried on to the 6- and 12-month follow-up where we grouped the patients on their IgG levels . CXCL9 , CXCL10 and IL-6 were raised in the patients with increased IgG levels at 6 months and CXCL9 at the 12-month follow-up . These proinflammatory cytokine associations with high IgG levels may represent the persistence of an active immune system . CXCL9 and CXCL10 as well as high IgG levels were found to be biomarkers of severe CHIKV symptoms . Previously , CXCL10 has been associated with severe viral disease supporting a role for CXCL10 in severe CHIKV disease [34]–[37] . These studies did not find an association with CXCL9 and severity as seen in our CHIKV patients . Our findings suggest high CXCL10 and CXCL9 associated with severity to be a unique signature of CHIKV . Interestingly , not only are CXCL10 and CXCL9 expressed in RA and other arthritis related disease patients [38]–[41] , but they have been shown to be biomarkers for RA symptoms , implying a similar mechanism for joint destruction in CHIKV disease [42] . Moreover , CXCL9 and CXCL10 may be contributors of persistent immune activation in CHIKV disease leading to chronic symptoms , which implies cytokine immunomodulation may significantly improve patient treatment and recovery . Importantly , CXCL10 also has prognostic value in the treatment of viral hepatitis where CXCL10 levels follow disease recovery [43] supporting our finding and proposes a prognostic role for CXCL10 in CHIKV . Furthermore , our data puts forth CXCL10 and CXCL9 as possible drug targets for treatment of CHIKV symptoms in the convalescence phase due to the association with severity; however , further investigation is needed on CXCL10 and CXCL9 efficacy . In addition , not only are the identified cytokines useful as possible drug targets but the cytokine signatures described can also be applied when testing newly developed CHIKV therapeutics . As CHIKV therapeutics are evaluated in the CHIKV disease model , cytokine profiles can be used as an output for determining the efficacy of the novel therapeutics . The synovial mast cell remains an important component during RA joint destruction by the exocytosis of intracellular granules containing inflammatory mediators . Currently , mast cell activation through FcγRs by high levels of circulating IgG antibodies is hypothesized to contribute to the pathological destruction of synovium in RA [44]–[46] . In addition , antibody immune complex formation within the joint stabilizing inflammatory mediators , such as chemokines and complement proteins , is another possible facet of pathogenesis during RA [47] , [48] . We found high concentrations of IgG to be associated with symptom severity in CHIKV patients . Similar IgG mediated mechanisms leading to synovium destruction and severe pain experienced by CHIKV patients are possible . Taken together with the roles of IgG in RA , our findings support the need for further investigation into the contribution of IgG levels and mast cells to CHIKV symptoms . We have found that the cytokines investigated had one of two immunologically important profiles during CHIKV disease onset and convalescence; furthermore , CXCL10 and CXCL9 were makers of disease severity . By identifying the immune profiles , we have created a clearer clinical picture of CHIKV disease . Importantly , further investigation is needed to correlate these profiles with disease onset and progression to use the cytokine profiles as biomarkers for severity and symptom persistence . The data presented here suggest that with further investigation , immunomodulators may significantly enhance patient recovery . | Chikungunya virus ( CHIKV ) is transmitted by mosquitoes and causes a human disease clinically characterized by sudden appearance of high fever , rash , headache , nausea , and severe joint pain ( the defining symptom ) . Chikungunya was identified in Africa and the word Chikungunya means that which bends up , describing the bent posture of CHIKV patients while in severe pain . CHIKV , a current problem in Africa , Indian Ocean region , and Southeast Asia , is now spreading to temperate regions of North America , France and Italy . Presently , the immune response for CHIKV infection remains largely uninvestigated and no treatment is available . We investigated cytokine profiles at diagnosis and follow-up of CHIKV infected patients during the Italian 2007 outbreak and associated cytokine levels with antibody level and symptom severity . Cytokines , important immune mediators , are often drug targets . Since CHIKV symptoms can persist for months or years following infection it is important to investigate possible drug targets to alleviate discomfort . We found cytokine profiles that describe the initial infection and recovery phase . We determined the cytokines CXCL9/MIG and CXCL10/IP-10 as well as antibody levels were associated with symptom severity . These results reflect previously unreported cytokine profiles which may be important for the development of future therapeutics for CHIKV outbreaks . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"infectious",
"diseases",
"public",
"health",
"and",
"epidemiology",
"diagnostic",
"medicine",
"clinical",
"immunology",
"drugs",
"and",
"devices",
"endocrinology",
"global",
"health",
"immunology",
"biology",
"microbiology",
"molecular",
"cell",
"biology",
"diabetes",
"and",
"endocrinology",
"public",
"health"
] | 2011 | Inflammatory Cytokine Expression Is Associated with Chikungunya Virus Resolution and Symptom Severity |
Within the last 10 years Zika virus ( ZIKV ) has caused unprecedented epidemics of human disease in the nations and territories of the western Pacific and South America , and continues to escalate in both endemic and non-endemic regions . We evaluated the vector competence of Australian mosquitoes for ZIKV to assess their potential role in virus transmission . Mosquitoes were exposed to infectious blood meals containing the prototype African ZIKV strain . After 14 days incubation at 28°C and high relative humidity , infection , dissemination and transmission rates were assessed . Infection in Culex annulirostris and Cx . sitiens could not be detected . 8% of Cx . quinquefasciatus were infected , but the virus did not disseminate in this species . Despite having infection rates > 50% , Aedes notoscriptus and Ae . vigilax did not transmit ZIKV . In contrast , Ae . aegypti had infection and transmission rates of 57% and 27% , respectively . In susceptibility trials , the virus dose required to infect 50% ( ID50 ) of Ae . aegypti was106 . 4 tissue culture infectious dose50 ( TCID50 ) /mL . Additionally , a threshold viral load within the mosquito of at least 105 . 1 TCID50 equivalents/mL had to be reached before virus transmission occurred . We confirmed Ae . aegypti to be the most likely mosquito vector of ZIKV in Australia , although the restricted distribution of this species will limit the receptive zone to northern Queensland where this species occurs . Importantly , the role in ZIKV transmission of Culex and other Aedes spp . tested will be negligible . Despite being the implicated vector , the relatively high ID50 and need for a high titer disseminated infection in Ae . aegypti suggest that high mosquito population densities will be required to facilitate epidemic ZIKV transmission among the currently immunologically naïve human population in Australia .
Zika virus ( ZIKV ) was first isolated from a rhesus monkey in the Zika Forest of Uganda in 1947 during studies investigating the ecology of yellow fever virus [1] . It was subsequently isolated from Aedes africanus , indicating a mosquito-borne transmission cycle . It was not until almost 20 years later that human clinical disease attributed to ZIKV infection was recognized [2] . ZIKV circulates in a sylvatic transmission cycle between non-human primates and mosquitoes . Humans are only incidentally infected in this sylvatic cycle , but are the primary amplifying hosts during epidemics [3 , 4] . Although 80% of infections remain asymptomatic , clinical disease caused by ZIKV is typical of many mosquito-borne viruses , and is characterized by fever , muscle and joint pain , headache , conjunctivitis , gastrointestinal manifestations and rash [2 , 5 , 6] . However , during its rapid expansion in the last 5 years , more severe disease manifestations have been recognized among those with Zika virus–like disease , most notably neurological symptoms , including Guillain-Barré syndrome , microcephaly and other central nervous system malformation in neonates [7 , 8] . Based on serological evidence , ZIKV infection of humans has historically been restricted to Africa and Asia [9] . In 2007 , an outbreak of ZIKV occurred on Yap Island in the western Pacific Ocean [5] , signaling the beginning of an unprecedented range expansion of this virus . Since 2013 , ZIKV has affected a number of countries and territories in the western Pacific , resulting in outbreaks of hundreds or thousands of suspected cases [10] . The 2013 epidemic in French Polynesia was the largest ever reported up to that point and resulted in an estimated 32 , 000 cases , representing 11 . 5% of the population [11] . In 2014 the virus was introduced into Brazil , from where it is believed to have subsequently spread to a number of countries in South America , causing over a million suspected cases [10 , 12] . Between 2007 and June 2 2016 , ZIKV was recognized in 63 countries or territories [13] . The risk of ZIKV spread to Australia is very high due to its close geographical proximity to the epidemic region in the western Pacific and high intake of travelers from this area . Indeed , between 2012 and June 3 2016 , 60 people infected with ZIKV have traveled to Australia [14] . Should a viremic patient be bitten by a local mosquito , or transmit the virus sexually [15] , there is potential for autochthonous transmission of the virus to occur , but this has not been reported to date . Clinical similarity of ZIKV disease with that caused by arboviruses , such as dengue ( DENV ) and chikungunya ( CHIKV ) , could delay ZIKV identification or detection of transmission , and there are currently no established vaccines or therapeutics . Therefore , it is very important to estimate the risk of ZIKV establishment in Australia for the purposes of heightened public awareness , and for adequate and appropriate response of public health authorities . In the current study we evaluated the ability of common Australian mosquito species to become infected with and transmit the African lineage of ZIKV . Although the Asian lineage is responsible for the recent activity in the Pacific and South America , we were not able to obtain an isolate to facilitate the assessment of the vector competence of local mosquito fauna for this ZIKV lineage in a manner that was timely for formulation of targeted control strategies . The sylvan vectors circulating ZIKV between primates in Africa are tree-hole inhabiting Aedes spp . , of which Ae . africanus is the most important [4] . The primary urban vectors are Ae . aegypti and Ae . albopictus [9] . Of the species implicated abroad , only Ae . aegypti occurs on the Australian mainland , albeit with a distribution restricted to northern Queensland [16] . Ae . albopictus is currently restricted to the Torres Strait islands off northern Australia [17] and was not tested in the current experiments . In terms of other potential vectors , Australia has a number of other species that could potentially transmit ZIKV . For instance , Ae . notoscriptus is a widespread urban species throughout Australia and was shown to be a competent laboratory vector of yellow fever virus [18] . Another potential species is Ae . vigilax , which has a widespread coastal distribution , is a notorious biter and a highly competent laboratory vector of CHIKV [19] . Finally , it has been postulated that Culex spp . , most notably Cx . quinquefasciatus , may play a role in transmission in South America [20] .
Mosquitoes were collected as eggs or adults from several locations in Queensland . Adult Ae . vigilax , Ae . procax , Cx . annulirostris and Cx . sitiens were collected using Centers for Disease Control light traps ( Model 512 , John Hock Co . , Gainesville , FL ) baited with CO2 ( 1kg dry ice ) from several suburbs in Brisbane , southeastern Queensland . Adult mosquitoes were transported to the laboratory and exposed to ZIKV within 5 h of collection . The Ae . aegypti used were F4 generation females from eggs collected using ovitraps from Townsville , northern Queensland in March 2015 . Eggs of Ae . notoscriptus and Cx . quinquefasciatus were collected using ovitraps and infusion buckets , respectively , from several suburbs in Brisbane . All larvae were reared at 26°C and 12:12 L:D . Both Aedes spp . were fed Hikari Cichlid Staple pellets ( Kyorin Co . Ltd , Himeji , Japan ) . First and second instar larvae of Cx . quinquefasciatus were fed a 1:1 mixture of brewer’s yeast ( Brewer’s Yeast , Healthy Life ) and fish flakes ( Wardley’s Tropical Fish Food Flakes , The Hartz Mountain Corporation , NJ ) , whilst third and fourth instars were fed Hikari Cichlid Staple pellets ( Kyorin co . Ltd , Himeji , Japan ) . Adults were held for 3–7 d at 26°C 12:12 L:D , and high relative humidity , and fed on 15% honey water ad libitum . Mosquitoes were starved for 24 h prior to virus exposure . The ZIKV strain ( MR 766 ) was the prototype strain isolated from a rhesus macaque monkey in the Zika Forest , Uganda in 1947 . The virus was sourced from the American Type Culture Collection ( Manassas , VA , USA ) . It had been passaged 146 times in adult mouse brain , once in suckling mouse brain and three times in in African green monkey kidney ( Vero ) cells . Mosquitoes were allowed to feed for 2 h on an infectious blood meal containing stock virus diluted in commercially available defibrinated sheep blood ( Institute of Medical and Veterinary Science , Adelaide , Australia ) . The blood meal was housed within a Hemotek feeding apparatus ( Discovery Workshops , Accrington , Lancashire , UK ) that was fitted with pig intestine as the membrane . Due to sufficient numbers and relatively high feeding rates , Ae . aegypti and Ae . notoscriptus were exposed to serial 10-fold dilutions of ZIKV , with 106 . 7± 0 . 2 tissue culture infectious dose50 ( TCID50 ) /mL the highest dose . The latter species was exposed to an additional blood feed at the highest dose . Due to limited numbers and low feeding rates , the other species were only exposed to blood meals containing a virus titer of 106 . 7± 0 . 2 TCID50/mL . Pre- and post- feeding samples of the blood/virus mixture were diluted 1:10 in growth media ( GM; Opti-MEM ( Gibco , Invitrogen Corporation , Grand Island , NY ) containing 3% foetal bovine serum ( FBS ) , antibiotics and antimycotics ) , and stored at -80°C . Immediately following virus exposure , mosquitoes were anesthetized with CO2 and engorged mosquitoes were placed in 900 ml gauze-covered containers . All mosquitoes were maintained on 15% honey water at 28°C , high relative humidity and 12L:12D light cycle within an environmental growth cabinet . Mosquitoes were processed at day 14 post-exposure . Due to sufficient numbers , Ae . aegypti were also processed at 5 , 7 and 10 d post exposure , whilst the additional Ae . notoscriptus were processed at day 7 post exposure . The ability for mosquitoes to become infected with and transmit ZIKV was assessed using a modified in vitro capillary tube technique [21] . Briefly , mosquitoes were anesthetized with CO2 , and the legs and wings removed . The saliva was collected by inserting the proboscis of the mosquito into a capillary tube containing GM with 20% FBS . After 30 min , the contents of the capillary tube were expelled into 500 μl of GM with 3% FBS . The legs+wings and bodies were placed in separate 2 mL tubes containing 1 mL of GM with 3% FBS and a single 5 mm stainless steel bead . Detection of virus in the legs+wings indicates that the mosquito has developed an infection whereby the virus has escaped the midgut and disseminated throughout the hemocoel , thus bypassing the midgut escape barrier [22] . All samples were stored at -80°C . The susceptibility of Ae . aegypti and Ae . notoscriptus to infection with ZIKV was calculated by probit analysis using SPSS release 16 . 0 . 0 . Log-log models were assessed using the Pearson χ2 goodness-of-fit statistic and susceptibility to infection was expressed as ID50 ± 95% confidence intervals ( CIs ) and defined as the virus dose per mL at which 50% of mosquitoes tested positive for ZIKV infection in the TaqMan RT-PCR . Infection , dissemination and transmission rates between Ae . aegypti and the other species were analyzed using Fisher’s Exact Tests with two-tailed P-values ( GraphPad Prism Version 6 ) . ZIKV TCID50/mL equivalents were tested for differences between species , and within Ae . aegypti using the Kruskal-Wallis test ( GraphPad Prism Version 6 ) .
To assess their susceptibility to infection , Ae . aegypti and Ae . notoscriptus were exposed to ZIKV doses ranging from 103 . 9 to 106 . 8 TCID50/mL and their bodies tested for infection 14 d post exposure ( Fig 1 ) . Both species were susceptible to infection , with ID50s of 106 . 4 ( 106 . 0 and 107 . 1 , 95% CL ) TCID50/mL ( χ2 = 2 . 49 , df = 2 , P = 0 . 288 ) and 106 . 6 ( 106 . 2 and 107 . 4 , 95% CL ) TCID50/mL ( χ2 = 8 . 49 , df = 2 , P = 0 . 654 ) for Ae . aegypti and Ae . notoscriptus , respectively . Five out of seven species , including all species of Aedes , were infected 14 days after being exposed to blood meals containing 106 . 7 ± 0 . 2 TCID50/mL of ZIKV ( Table 1 ) . Out of the 3 Culex spp . tested , Cx . annulirostris and Cx . sitiens were refractory to infection and only 2 out of 30 Cx . quinquefasciatus were infected but none developed a disseminated infection . On day 14 post exposure , infection rates in Ae . notoscriptus , Ae . procax and Ae . vigilax did not significantly differ ( P > 0 . 05 ) from Ae . aegypti . With the exception of one cohort of Ae . notoscriptus , dissemination rates in these three species did not significantly differ ( P > 0 . 05 ) from Ae . aegypti . However , Ae . aegypti was the only species that was able to transmit ZIKV , with transmission first observed at day 10 post exposure . ZIKV titers in the mosquito bodies did not significantly differ among the four Aedes spp . tested ( Fig 2 ) . In contrast there was a significant difference ( P = 0 . 0014 ) in the ZIKV titers in the legs+wings between these species . Importantly , transmission of ZIKV by Ae . aegypti on day 14 post exposure occurred only when the titer of ZIKV in the legs+wings was ≥ 105 . 6 TCID50 equivalents/mL , with all individuals at or above this threshold able to transmit ZIKV . For the Ae . aegypti , there was an increase in the estimated virus titer on the different days tested post exposure , although the differences were not significant ( P > 0 . 05; Fig 3A ) . Similarly , there was an increase in legs+wings titer over the various days , with the difference between days 5 and 14 being significant ( P < 0 . 05; Fig 3B ) . The three mosquitoes that transmitted ZIKV on day 10 post exposure possessed legs+wings titers that were lower than the 105 . 6 TCID50 equivalents/mL required for transmission on day 14 . The titer of the saliva expectorated did not significantly differ ( P > 0 . 05 ) between mosquitoes sampled on days 10 and 14 post exposure ( Fig 3C ) .
Commensurate with its role in transmission in Africa , the western Pacific and South America , we have demonstrated that a Townsville , Australia , population of Ae . aegypti is susceptible to infection and can transmit ZIKV . Transmission rates were higher than those recently reported for Brazilian and Senegalese strains of Ae . aegypti that were exposed to Asian and African lineages of ZIKV , respectively , but lower than those for a Singapore strain infected with MR 766 [26–28] . However , comparison of experimental vector competence outcomes between different laboratories should be viewed with caution due to differences in mosquito strain , virus strain , mosquito feeding method and virus assay used to analyze samples [29] . Regardless , we have confirmed that highly infected Ae . aegypti could be a potential source of transmission of ZIKV in Australia . Although Ae . aegypti was prevalent in eastern Australia during the first half of the 20th century , the distribution of this species is currently restricted to urban areas of northern Queensland [16] . Therefore , should the virus be introduced , this is the region that is most at risk of ZIKV transmission . However , if the distribution of Ae . aegypti expanded , or if Ae . albopictus , another potential vector [26] , invaded the Australian mainland , then the ZIKV receptive zone would need to be redefined . This highlights the importance of surveillance for these container-inhabiting species to ensure that any expansion of these species is recognized and elimination programs initiated . In addition to the intrinsic ability of Ae . aegypti to transmit ZIKV , this species exhibits a number of biological traits which dramatically elevate its role as a vector of this virus , as well as other viruses transmitted by this species . Jansen et al . [30] used a relatively simple vectorial capacity model [31] to assess the relative roles of Australian mosquitoes in the transmission of CHIKV . Vectorial capacity takes into account a number of factors , including mosquito density , host feeding patterns , survival , vector competence and the duration of the intrinsic incubation period of the virus . It was demonstrated that Ae . aegypti had the highest vectorial capacity even though other species , such as Ae . vigilax , had higher experimental transmission rates [19] and population densities compared to Ae . aegypti ( [32]; Queensland Health , state government data ) . The reason for this was that a high proportion of Ae . aegypti obtain their blood meals from humans [33 , 34] , and they exhibit multiple host blood feeding behavior [35] , whereby females probe/feed up to four times in a single gonotrophic cycle . This latter trait exposes them to more infected humans increasing the likelihood of consuming virus , as well as increasing the number of susceptible humans exposed to an infected mosquito . The majority of other mosquito species usually take only a single blood meal per gonotrophic cycle . To assess the potential risk for local ZIKV transmission in Australia , we exposed 6 other mosquito species to ZIKV and evaluated their ability to transmit the virus . The results demonstrated that the Culex spp . were either refractory to infection or exhibited a low infection rate but did not transmit ZIKV , suggesting that they should not be considered potential vectors of ZIKV in Australia . Despite being susceptible to infection , the inability of Ae . notoscriptus , Ae . vigilax and Ae . procax to transmit the virus , indicate that these species would be unlikely to play a role in ZIKV transmission . However , given the potential for intraspecies variation in vector competence of arboviruses [36 , 37] , it is important that the vector competence of other Australian populations of these species for ZIKV be assessed using the prototype strain , as well as the Asian lineage viral variant ( s ) currently circulating in the western Pacific and South America , and recently imported into Africa . The current study has highlighted aspects of the susceptibility , replication and transmission dynamics of ZIKV in the mosquito vector that may ultimately impact transmission cycles in the field . We demonstrated that the ID50 for Ae . aegypti was 106 . 4 TCID50/mL , which is considerably higher than that observed for other arboviruses in Australian populations of this species , including DENVs ( ≈ 105 . 5 TCID50/mL ) and CHIKV ( 104 . 9 TCID50/mL ) [19 , 38] . The relatively low infection and transmission rates reported for some populations of Ae . aegypti [26 , 27] would suggest a similarly high threshold for infection in this species . Paradoxically , the high ID50 required to infect Ae . aegypti in our study is potentially higher than the viremia values of 105 to 106 RNA copies per mL circulating in the blood of symptomatic patients during recent outbreaks in the western Pacific [39 , 40] . This may simply reflect the differences inherent to quantifying infectious virus and viral RNA or it may hint at factors other than low susceptibility to infection of the mosquito vector . Such factors include high mosquito population density and high survival rates of infected mosquitoes , coupled with a naïve human population suffering high viral loads , and alternative modes of transmission , such as sexual transmission , all of which may contribute to epidemic transmission of ZIKV . There are a number of intrinsic barriers that can influence the ability of an arbovirus to infect , disseminate within and be transmitted by mosquitoes [41] and the species tested in our study expressed one or more of these barriers . Based on their refractoriness to infection or low infection rate , it appears that the three species of Culex tested possess a midgut infection barrier . Similar to what has been observed previously with DENVs and CHIKV [19 , 42] , Ae . notoscriptus expresses a midgut escape barrier , as only 10% of mosquitoes tested at day 14 had a disseminated infection . With the exception of Ae . aegypti , the remaining Aedes spp . tested appeared to have a salivary infection/transmission barrier , as none of the individuals with a disseminated infection subsequently transmitted ZIKV . Interestingly , within the Ae . aegypti cohort , it was only those mosquitoes with a high disseminated infection titer of ≥ 105 . 1 TCID50/mL in the legs+wings that transmitted the virus . This suggests that a considerable quantity of virus is required to overcome the salivary gland barriers to transmission . Conversely , the other Aedes spp . had lower legs+wings titers of ≤ 104 . 8 TCID50/mL and this may explain why they did not transmit the virus . A correlation between the leg titer and the percentage of mosquitoes transmitting has previously been shown for dengue virus type 1 ( DENV-1 ) , whereby a threshold titer had to be reached before transmission occurred [43] . It was concluded that the high dissemination titers led to increased transmissibility by Ae . aegypti , which potentially resulted in a DENV-1 clade replacement event in Thailand . Future studies , using strains of ZIKV from the current epidemics and a considerably larger sample size , should examine whether a similar correlation exists between different viral lineages , proportion of each mosquito species infected , dissemination titers and the transmission rate . Importantly , this will reveal whether only a relatively small proportion of the mosquito population is likely to be contributing to the majority of virus transmission . In conclusion , Australia possesses the two key elements for ZIKV to occur: a competent ZIKV vector and importation of the virus by infected travelers . The Australian distribution of Ae . aegypti is currently restricted to north Queensland , so this is the region that is receptive to ZIKV . It is paramount that container mosquito surveillance is maintained and , when resources are available , enhanced to ensure that Ae . aegypti and Ae . albopictus do not expand their range . There is also a need to maintain comprehensive testing of travelers from epidemic regions to ensure that suspected cases are diagnosed , particularly those residing in the ZIKV receptive zone . Indeed , there have already been several cases notified in northern Queensland where Ae . aegypti occurs , although control strategies implemented routinely for DENVs [44] may have assisted in preventing local transmission of the virus to date . Unfortunately , tracking of ZIKV infected travelers is compounded by the high number of asymptomatic infections and local transmission may be occurring before ZIKV is tested for in human or mosquito populations . Ultimately , a combination of case recognition , specific laboratory diagnostics , virus surveillance in mosquito populations , vector surveillance to detect incursions of potential vectors into uninfested locations and targeted mosquito control strategies will reduce the risk of an explosive ZIKV outbreak occurring in Australia . | Zika virus was first isolated in Uganda in 1947 and exists in a transmission cycle between mosquitoes and non-human primates or humans . Whilst most clinical infections result in a self-limiting febrile illness , Zika virus has recently been linked to neurological syndromes , such as Guillain-Barré syndrome and congenital birth defects . Since 2007 , Zika virus has undergone a dramatic range expansion , causing epidemics in nations and territories of the western Pacific and South America . To assess the emergence and transmission risk of Zika virus emerging in Australia , we evaluated the ability of local mosquitoes to become infected with and transmit the prototype African Zika virus strain . In agreement with its substantiated role in Zika virus transmission overseas , Australian Aedes aegypti were shown to be competent vectors . Coupled with its anthropophilic feeding behavior , this species should be considered the primary potential Zika virus vector in Australia . Although other common Australian species , such as Ae . notoscriptus and Ae . vigilax , were readily infected , they did not transmit the virus . The species of Culex tested were either refractory to infection or had a low infection rate . We also demonstrated that the Zika virus dose necessary to infect Ae . aegypti was higher than virus levels reported in infected humans . Finally , a high threshold level of virus circulating through the mosquito body was required before Ae . aegypti transmitted the virus . These results suggest that an outbreak of Zika virus in Australia would require high mosquito population densities and a susceptible human population . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"body",
"fluids",
"pathology",
"and",
"laboratory",
"medicine",
"viral",
"transmission",
"and",
"infection",
"pathogens",
"geographical",
"locations",
"microbiology",
"australia",
"rna",
"extraction",
"saliva",
"animals",
"viral",
"vectors",
"viruses",
"rna",
"viruses",
"insect",
"vectors",
"extraction",
"techniques",
"research",
"and",
"analysis",
"methods",
"infectious",
"diseases",
"medical",
"microbiology",
"epidemiology",
"microbial",
"pathogens",
"arboviral",
"infections",
"disease",
"vectors",
"insects",
"hematology",
"arthropoda",
"people",
"and",
"places",
"mosquitoes",
"blood",
"anatomy",
"flaviviruses",
"oceania",
"viral",
"pathogens",
"physiology",
"virology",
"biology",
"and",
"life",
"sciences",
"viral",
"diseases",
"organisms",
"zika",
"virus"
] | 2016 | Assessment of Local Mosquito Species Incriminates Aedes aegypti as the Potential Vector of Zika Virus in Australia |
Previously , we discovered a conserved interaction between RB proteins and the Condensin II protein CAP-D3 that is important for ensuring uniform chromatin condensation during mitotic prophase . The Drosophila melanogaster homologs RBF1 and dCAP-D3 co-localize on non-dividing polytene chromatin , suggesting the existence of a shared , non-mitotic role for these two proteins . Here , we show that the absence of RBF1 and dCAP-D3 alters the expression of many of the same genes in larvae and adult flies . Strikingly , most of the genes affected by the loss of RBF1 and dCAP-D3 are not classic cell cycle genes but are developmentally regulated genes with tissue-specific functions and these genes tend to be located in gene clusters . Our data reveal that RBF1 and dCAP-D3 are needed in fat body cells to activate transcription of clusters of antimicrobial peptide ( AMP ) genes . AMPs are important for innate immunity , and loss of either dCAP-D3 or RBF1 regulation results in a decrease in the ability to clear bacteria . Interestingly , in the adult fat body , RBF1 and dCAP-D3 bind to regions flanking an AMP gene cluster both prior to and following bacterial infection . These results describe a novel , non-mitotic role for the RBF1 and dCAP-D3 proteins in activation of the Drosophila immune system and suggest dCAP-D3 has an important role at specific subsets of RBF1-dependent genes .
The RB family proteins ( pRB , p130 and p107 in humans; RBF1 and RBF2 in Drosophila ) co-ordinate changes in gene expression . Understanding the types of programs that these proteins regulate is important because of the unequivocal link between the inactivation of RB proteins and human cancer . Mutation of the retinoblastoma tumor susceptibility gene ( RB1 ) is the rate-limiting step in the genesis of retinoblastoma and over 90% of human tumors exhibit reduced pRB function [1] , [2] . RB family members are best-known for their roles in the regulation of E2F-dependent transcription . E2F-controlled genes are needed for cell proliferation and RB proteins suppress the expression of these targets during G0 and G1 of the cell cycle [3] . In addition , RB proteins are also important for the regulation of genes that are not involved in cell cycle progression . For example , osteoblast differentiation is modulated by pRB through its interaction with Runx2 [4]; in muscle cells , pRB promotes the expression of muscle-specific differentiation markers , enabling these cells to irreversibly exit the cell cycle [5]–[7]; in Drosophila , RBF1 cooperates with the Hippo pathway to maintain photoreceptor differentiation , independent of dE2F1 activity [8] . Such E2F-independent functions may help to explain why the inactivation of RB proteins can have very different consequences in different cellular contexts . However , many of the E2F-independent activities of RB proteins are not well-understood . At present , it is unclear if pRB has different activities in different cell types , or whether there is a yet-to-be discovered , general process that allows RB proteins to activate or repress the expression of variable sets of genes in different cell types . Recent studies have suggested that pRB family members may impact the organization of higher-order chromatin structures , in addition to their local effects on the promoters of individual genes [9] . Mutation of pRB causes defects in pericentric heterochromatin [10] and RBF1 is necessary for uniform chromatin condensation in proliferating tissues of Drosophila larvae [11] . Part of the explanation for these defects is that RBF1 and pRB promote the localization of the Condensin II complex protein , CAP-D3 to DNA both in Drosophila and human cells [11] . Depletion of pRB from human cells strongly reduces the level of CAP-D3 associated with centromeres during mitosis and causes centromere dysfunction [12] . Condensin complexes are necessary for the stable and uniform condensation of chromatin in early mitosis [13]–[16] . They are conserved from bacteria to humans with at least two types of Condensin complexes ( Condensin I and II ) present in higher eukaryotes . Both Condensin I and II complexes contain heterodimers of SMC4 and SMC2 proteins that form an ATPase which acts to constrain positive supercoils [17] , [18] . Each type of Condensin also contains three specific non-SMC proteins that , upon phosphorylation , stabilize the complex and promote ATPase activity [14] , [19] , [20] . The kleisin CAPH and two HEAT repeat containing subunits , CAP-G and CAP-D2 are components of Condensin I , while the kleisin CAP-H2 and two HEAT repeat containing subunits , CAP-G2 and CAP-D3 , are constituents of Condensin II . Given the well-established functions of Condensins during mitosis , and of RBF1 in G1 regulation , the convergence of these two proteins was unexpected . Nevertheless , mutant alleles in the non-SMC components of Condensin II suppress RBF1-induced phenotypes , and immunostaining experiments revealed that RBF1 displays an extensive co-localization with dCAP-D3 ( but not with dCAP-D2 ) on the polytene chromatin of Drosophila salivary glands [11] . This co-localization occurs in cells that will never divide , suggesting that Condensin II subunits and RBF1 co-operate in an unidentified process in non-mitotic cells . In various model organisms , the mutation of non-SMC Condensin subunits has been associated with changes in gene expression [21]–[24] raising the possibility that dCAP-D3 may affect some aspect of transcriptional regulation by RBF1 . However , the types of RBF1-regulated genes that might be affected by dCAP-D3 , the contexts in which this regulation becomes important , and the consequences of losing this regulation are all unknown . Here we identify sets of genes that are dependent on both rbf1 and dCap-D3 . The majority of genes that show altered expression in both rbf1 and dCap-D3 mutants ( larvae or adults ) are not genes involved in the cell cycle , DNA repair , proliferation , but are genes with cell type-specific functions and many are spaced within 10 kb of one another in “gene clusters” . To better understand this mode of regulation we have investigated the effects of RBF1 and dCAP-D3 on one of the most highly misregulated clusters which includes genes coding for antimicrobial peptides ( AMPs ) . AMPs are produced in many organs , and one of the major sites of production is in the fat body . Following production in the fat body , AMPs are subsequently dumped into the hemolymph where they act to destroy pathogens [25] . RBF1 and dCAP-D3 are required for the transcriptional activation of many AMPs in the adult fly . Analysis of one such gene cluster shows that RBF1 and dCAP-D3 bind directly to this region and that they bind , in the fat body , to sites flanking the locus . RBF1 and dCAP-D3 are both necessary in the fat body for maximal and sustained induction of AMPs following bacterial infection , and RBF1 and dCAP-D3 deficient flies have an impaired ability to respond efficiently to bacterial infection . These results identify dCAP-D3 as an important transcriptional regulator in the fly . Together , the findings suggest that RBF1 and dCAP-D3 regulate the expression of clusters of genes in post-mitotic cells , and this regulation has important consequences for the health of the organism .
Our previous data demonstrated that RBF1 co-localizes extensively with dCAP-D3 on polytene chromatin of non-dividing cells , leading us to hypothesize that the two proteins may co-operate to regulate transcription . To begin to test this idea , we first identified the stages of fly development where RBF1 and dCAP-D3 were most highly expressed . qRT-PCR using primers for dCap-D3 and rbf1 was performed on cDNA generated from various stages of the Drosophila life cycle ( Figure 1A ) . The results demonstrate that both genes are transcribed at the highest levels in late third instar larval and adult stages . Concordantly , immunostaining for dCAP-D3 and RBF1 in cryosections of the abdomens of wild type flies confirmed that both proteins are highly expressed in the adult and that they are both present in the nuclei of many cells in normal adult tissues ( Figure 1B ) . Preliminary experiments showed that dCAP-D3 levels could influence the expression of very few of the previously identified RBF1-dependent transcripts . To gain a more complete understanding of the abundance and characteristics of RBF1/dCAP-D3 shared transcriptional targets , we carried out a microarray analysis of the entire Drosophila melanogaster genome and compared gene expression profiles of wild type , dCap-D3 and rbf1 mutant flies , at both the third instar larval and adult stages ( Table S1 ) . Since the null mutants are lethal , females expressing a transheterozygous combination of null and hypomorphic alleles were used for these experiments . The mutant flies used for microarray analysis expressed about 15% of wild type levels of each gene as judged by qRT-PCR and western blot ( Figure S1 ) . The microarray results revealed an extensive and highly significant overlap between RBF1 and dCAP-D3 regulated gene sets in both adults and larvae ( Figure 2A and 2B ) . Shared target genes were evident in both upregulated and downregulated gene sets . Although some genes were mis-expressed in both larvae and adults , the majority of transcriptional changes were stage specific . The most highly significant p values for shared target gene sets were seen in upregulated larval genes ( genes repressed by RBF1 and dCAP-D3 in the larvae , p≤6 . 34E-130 ) and downregulated adult genes ( genes activated by RBF1 and dCAP-D3 in the adult , p≤9 . 88E-95 ) ( Figure 2B ) . This suggests that RBF1 and dCAP-D3 may cooperate to repress specific programs during one stage of development and activate other programs in a later , more differentiated stage . Interestingly , at both stages , the genes dependent on both RBF1 and dCAP-D3 represented 15–17% of the total number of genes dependent on RBF1 in a given developmental stage and 47–55% of the total number of genes dependent on dCAP-D3 in a given developmental stage . Thus RBF1 appears to be important at close to half of the transcriptional targets of dCAP-D3 . We noticed that the lists of RBF1/dCAP-D3 shared target genes had two general properties . First , these genes are almost completely different from the lists of E2F-regulated genes that have been reported previously [26] . As expected , many of the targets that were upregulated in rbf1 mutant larvae could be categorized as E2F target genes involved in DNA repair , DNA replication and continuation of the cell cycle ( comparison to microarray data from [26] and GO analyses of rbf1 mutant larvae- data not shown ) . However , few if any , of these cell cycle/proliferation related genes were altered in the dCap-D3 mutant flies ( Figure 2C ) suggesting that dCap-D3 regulates a different subset of RBF1 dependent targets . In fact , less than 6% of dCAP-D3/RBF1 shared target genes in larvae were found to be bound by dE2F1 in dE2F1 ChIP-chip experiments ( Korenjak et . al . , unpublished data ) . Unexpectedly , many of the known E2F target genes did not show a significant increase in expression in rbf1 mutant adults ( Figure 2C ) . This may reflect cell-type specific differences in the requirement for RBF1 . In support of this idea , qRT-PCR analysis of dissected tissues showed that few E2F-regulated genes were upregulated in ovaries of rbf1 mutants , but many did show a significant increase in the rest of the carcass ( Figure S2 ) . However , even in the tissues where these E2F-regulated proliferation genes did increase in expression levels in rbf1 mutant adults , these transcripts were not upregulated in tissues from dCap-D3 mutant flies ( Figure S2 ) . We infer that dCAP-D3 is not a key factor at most of the well-characterized E2F regulated genes in either larvae or adults . While unlikely , it is a formal possibility that the remaining amounts of dCAP-D3 protein present in the hypomorphic mutant flies might be sufficient for the regulation of E2F targets , but not for other target genes . Second , we noted that genes that are similarly dependent on RBF1 and dCAP-D3 tend to be clustered on the genome and are often positioned within 10 kb of one another ( Table 1 ) . To determine whether this was an unusual feature , we compared the frequency of RBF1/dCAP-D3shared target genes positioned within 10 kb of one another to hundreds of simulations of randomly chosen Drosophila genes ( Table 1 ) . The results showed that genes exhibiting increased expression in rbf1 and dCap-D3 mutant adults ( ie . genes that are apparently repressed both by RBF1 and dCAP-D3 are 25 times more likely to be clustered . Genes that were downregulated in rbf1 and dCap-D3 mutant adults ( i . e . genes apparently activated by both RBF1 and dCAP-D3 ) are 15 times more likely to be clustered . Clustering of shared target genes was also seen in the larvae , although the fold difference was greatly diminished ( 5 fold ) for the activated genes . Overall , the clustering effect was 3–7 fold more prevalent in dCAP-D3 regulated genes than in RBF-regulated genes . By way of comparison , RBF1/dCAP-D3 shared target genes in the larvae exhibited a much greater degree of clustering than the larval genes regulated by Hop or Nurf301 , two other well-known chromatin remodeling proteins shown to regulate clusters of genes [27] . A list of the actual groupings of clustered genes is presented in Table S2 . Although proliferation-related genes were missing , gene ontology ( GO ) classification of the RBF1/dCAP-D3 shared target genes revealed many significant categories in the lists of up- and downregulated genes in the adult , and for shared , repressed target genes in the larvae ( Figure 2D , complete list for downregulated adult genes in Table S3 ) . One of the most interesting GO categories represented in the adults were the defense response genes ( GO:0050830 ) . The fly relies on an innate immune system to defend against invading pathogens . This immune system is comprised of three major mechanisms: 1 ) phagocytosis , 2 ) induction of coagulation and melanization , and 3 ) production of Antimicrobial Peptides or AMPs . Phagocytosis is a conserved mechanism that is often the primary cellular defense used by many organisms to engulf and destroy pathogens . In Drosophila , circulating blood cells called hemocytes phagocytose bacteria , fungi , and parasitic wasp eggs [28] . RBF1 and dCAP-D3 mutant adult microarray data was analyzed for changes in levels of 19 different genes reported to be involved in phagocytosis in Drosophila ( Table S4 ) . Of these 19 genes , 2 genes demonstrated significant changes in transcript levels in adults . NimC1 , a gene expressed in plasmatocytes which make up 95% of Drosophila hemocytes , has been shown to be necessary for phagocytosis of bacteria [29] , and was significantly upregulated in RBF1 and dCAP-D3 mutant adults . Embryonic and larval hematopoiesis depends on a number of transcription factors including Gcm [30] , [31] . Gcm transcripts were demonstrated to be downregulated in both RBF1 and dCAP-D3 adults ( Table S4 ) . While adult hemocytes do display phagocytic properties , they do not differentiate into specialized cells upon immune challenge [32] , [33] , and it is therefore unlikely that misregulation of gcm in adults would affect phagocyte numbers . In response to septic injury , proteolytic cascades are triggered which lead to coagulation and melanization . Reactive oxygen species formed during these processes , as well as the actual deposition of melanin , are thought to be toxic to microorganisms [34] . After scanning the literature for genes involved in coagulation and melanization , and then analyzing RBF1 and dCAP-D3 mutant adult microarray data for changes in transcript levels of these genes , it was determined that only one of the reported genes , CG8193 was significantly increased in both RBF1 and dCAP-D3 mutant adults ( Table S5 ) . CG8193/PPO2 is thought to encode a phenol-oxidase constitutively expressed in crystal cells , a type of hemocyte cell involved in melanization [35] . However , overexpression of CG8193/PPO2 in cell lines or in flies does not induce constituitive melanization [36] , nor did we see any evidence of melanotic lesions in RBF1 or dCAP-D3 mutant adults . Several of the genes in the adult , downregulated GO category of “defense response to Gram positive bacteria” ( Figure 2D and Table S3 ) fall into a family of proteins known as Antimicrobial Peptides or AMPs . In fact , two of these genes , AttA and AttB , represented some of the most highly deregulated targets in the mutant adults . Upon closer inspection of the microarray data , it was revealed that many other AMP genes were also deregulated in dCAP-D3 and/or RBF1 mutant adults , however their p-values were just below the confidence level . In addition , many of the AMP genes are present in clusters and located immediately next to one another in the genome ( Figure 3A ) , making them an enticing group of genes for further study . To confirm that the transcription of AMPs was indeed dependent on both RBF1 and dCAP-D3 , qRT-PCR analysis was performed using cDNA generated from dCap-D3 or rbf1 transheterozygotes ( using whole female mutant flies whose ovaries had been dissected ) ( Figure 3B ) . Results showed that 17 of the 21 AMPs tested were downregulated in the Cap-D3 mutants and 10 of those genes were similarly dependent on RBF1 . qRT-PCR for AMPs performed on different allelic combinations of rbf1 and Cap-D3 mutants gave similar results ( Figure 3C ) . AMPs constitute one of the major defense mechanisms against bacterial and/or fungal infection in the fly [25] , [37] , [38] . They are produced in various adult tissues but one of the main organs responsible for their production is the fat body . Once produced in the fat body , AMPs are secreted into the hemolymph where they destroy or inhibit growth of pathogens [39] . We set out to test the hypothesis that RBF1 and dCAP-D3 regulate AMP genes in the adult fat body . First , we examined whether RBF1 and dCAP-D3 are expressed in this cell type . The yolk-GAL4 driver was used to express GFP in adult fat body cells , effectively marking this cell type in green . Combined immunostaining for RBF1 and dCAP-D3 localization in cryosections of adult wild type abdomens revealed a strong staining for both RBF1 and dCAP-D3 in the nuclei of adult fat body cells ( Figure 4A , yellow arrows ) . yolk-GAL4 has been characterized to drive expression in Drosophila at approximately 2–5 days post eclosure [40] , making it possible to drive expression of transgenes after the majority of fly development has occurred . The staining for RBF1 and dCAP-D3 in the adult abdomens was specific , as yolk-GAL4 driven expression of dsRNAs directed against RBF1 and dCAP-D3 specifically abrogated staining of their respective targets in fat body cells , without dramatically altering gross tissue morphology ( Figure S3 ) . Next , we measured the changes in expression of AMPs in animals where yolk-GAL4 driven expression of dsRNAs had reduced the expression of either RBF1 or dCAP-D3 in the fat body . qRT-PCR of cDNA from whole adult females showed a significant decrease in the expression of multiple AMP genes including diptericin , diptericin B and Cecropin A2 ( Figure 4B ) in the knockdown flies . Interestingly , the fold change in transcript levels for diptericin was comparable to the changes seen in dCap-D3 and rbf1 mutant animals . These results suggest that the yolk-GAL4-expressing cells are a primary site of constitutive diptericin expression in adult flies and that in these cells , RBF1 and dCAP-D3 are both needed to drive the basal expression levels of specific AMPs . AMP genes can be regulated by multiple transcription factors [41]–[44] . We sought to determine , therefore , whether transcriptional regulation of these genes by RBF1 and dCAP-D3 was direct . For our ChIP analysis we focused on diptericin and diptericin B; two AMP genes that are situated within 1200 bp of one another ( Figure 5B ) , that have well characterized promoters [45]–[47] , and whose basal expression was dependent on both RBF1 and dCAP-D3 in the fat body ( Figure 3B and Figure 4B ) . In addition , the basal transcript levels of at least one other gene in the region , CG43070 , was found to be significantly activated by both RBF1 and dCAP-D3 ( Figure S4 ) . To study the binding of RBF1 and dCAP-D3 at the diptericin locus in vivo , transgenic fly lines were created which expressed N-terminally FLAG-HA tagged dCAP-D3 or N-terminally FLAG-HA tagged RBF1 under the control of the UAS promoter . These lines were then crossed to yolk-GAL4/FM7 lines to create progeny in which the tagged protein was specifically expressed in the adult fat body . ChIP using FLAG antibody in FLAG-HA-dCAP-D3 expressing flies demonstrated that dCAP-D3 binds to two separate regions located approximately 3 kb upstream and 900 bp downstream of the diptericin locus ( Figure 5A and red bars in Figure 5B ) . Since diptericin is strongly induced in response to bacterial infection , we examined the effect of infection with S . aureus on the binding of dCAP-D3 to the diptericin locus . Strikingly , dCAP-D3 binding to the upstream site significantly increased after S . aureus infection ( compare red bars to yellow bars , Figure 5B ) . ChIP for FLAG in FLAG-HA-RBF1 expressing flies indicated that RBF1 binds to the identical upstream and downstream regions of the diptericin locus as dCAP-D3 ( red bars in Figure 5C ) . This binding was detected both before and after infection with S . aureus ( blue and yellow bars in Figure 5C ) , but unlike the results for dCAP-D3 binding , RBF1 binding was most significant prior to infection . ChIP for FLAG protein in flies expressing the FLAG-HA construct alone showed almost no signal at any of the primer sets used in these experiments ( Figure 5D ) . Taken together , ChIP results show that 1 ) RBF1 and dCAP-D3 can bind directly to an AMP gene cluster at identical binding sites , 2 ) that the binding sites flank the diptericin and diptericin B genes , and 3 ) dCAP-D3 binding increases when gene expression is induced in response to bacterial infection . For comparison , we also performed ChIP for dCAP-D3 on the CG5250 locus . CG5250 was the one previously identified direct target of RBF1 [26] that we found to be repressed by RBF1 and dCAP-D3 and to be consistently upregulated in all tissues of rbf1 and dCap-D3 mutant animals ( Figures S2 and S5A ) . ChIP using FLAG antibody in FLAG-HA-dCAP-D3 expressing flies demonstrated a small amount of binding in the open reading frame of CG5250 ( Figure S5B and S5C ) . This binding pattern obtained with the FLAG antibody closely resembled the ChIP signal found when a dCAP-D3 antibody was used to immunoprecipiate the endogenous dCAP-D3 protein expressed everywhere in the adult fly ( Figure S5D ) . The ability of RBF1 and dCAP-D3 to regulate basal levels of AMP transcription prompted the question of whether these proteins were also necessary for the regulation of AMP transcription in response to bacterial infection . cDNA was generated from female adult flies expressing dCAP-D3 or RBF1 dsRNAs specifically in the fat body , at various time-points post-infection with Staphylococcus aureus ( Figure 6 ) . dsRNAs have been used successfully in the past to decrease in vivo expression levels of proteins involved in innate immunity and to study their effects on responses to bacterial infection [48] . qRT-PCR for AMPs indicated two types of transcriptional defects in the RBF1 and dCAP-D3 deficient flies . In agreement with our earlier results , basal transcript levels of diptericin were reduced as a result of deficiency for either protein ( Figure 6B , inset boxes ) . Following infection , diptericin transcripts remained very low in the dCAP-D3 deficient tissue and induction was minimal and severely delayed in comparison to GFP dsRNA expressing “wild type” control flies . RBF1 deficiency , however , allowed normal induction of diptericin transcripts . Drosomycin is an AMP gene downstream of the Toll pathway , and it is strongly induced following infection with Gram positive bacteria or fungi [49] . qRT-PCR for levels of Drosomycin revealed a much different defect in expression . Neither dCAP-D3 nor RBF1 deficiency in the fat body had any effect on basal levels of Drosomycin , a result consistent with our microarray data from whole flies . However , both dCAP-D3 and RBF1 deficiency caused significant decreases in the maximal expression levels of drosomycin at 24 hours post-infection ( Figure 6A ) . Next , we tested whether the inefficient transcription of AMPs that results from decreased expression of RBF1 or dCAP-D3 has a significant effect on the ability of the fly to recover from exposure to pathogenic bacteria . Survival rates after infection with the Gram positive bacterium , Staphylococcus aureus ( Figure 7A , Figure 8A ) or with the Gram negative bacterium , Pseudomonas aeruginosa ( Figure 7B , Figure 8B ) were measured in five different genotypes: females expressing GFP dsRNAs under the control of the yolk-GAL4 driver ( “wild-type controls” , yolk-GAL4 driving expression of dCAP-D3 dsRNA in the fat body , yolk-GAL4 driving expression of RBF1 dsRNA in the fat body , and positive control females which were either mutant for the Eater protein or expressing dsRNAs against the IMD protein . IMD is a major mediator of innate immune signaling in Drosophila [50] . Eater is a known phagocyctic receptor necessary for the response to infection with Gram positive bacteria [51] . We did not include data on flies expressing Eater dsRNAs under the control of yolk-GAL4 , since Kocks et al [51] reported that Eater is not expressed in the fat body . In agreement with this , expression of Eater dsRNA in the fat body exhibited no changes in the ability of the fly to clear bacteria , while the Eater mutants described above showed a striking inability to phagocytize bacteria 5 hours following infection ( data not shown and Figure 7 ) . Following infection with S . aureus , both dCAP-D3 and RBF1 deficient flies were more susceptible to infection in comparison to flies expressing GFP dsRNAs ( Figure 7A and Figure 8A ) . dCAP-D3 deficient flies were also more susceptible to infection with Gram negative bacteria , but this was not the case for RBF1 deficient flies , as their survival rates were not significantly decreased ( Figure 7B and Figure 8B ) . These data demonstrate that acute knockdown of dCAP-D3 or RBF1 in the fat body of adult flies renders them more susceptible to bacterial infection , most likely due to inefficient transcription of AMP genes . Recently , a number of reports have identified genes whose mutation can reduce the ability of the fly to survive bacterial infection , without influencing the ability of the fly to clear bacteria [52]–[54] . These genes have been described as having affects not on the resistance mechanisms which exist in the fly , but on the tolerance mechanisms of the fly . Tolerance mechanisms limit the damage caused to the host by the infection , but do not actually limit the pathogen burden [55] . To determine whether loss of dCAP-D3 and/or RBF1 expression in the fat body did indeed result in diminished capacity of the fly to clear bacteria , we performed bacterial clearance assays and measured the number of bacteria present in the fly from 0–20 hours post-infection ( Figure 9 ) . Results showed that flies deficient for RBF1 or dCAP-D3 behave more like positive control flies deficient for IMD or Eater proteins , and exhibit significant increases in bacterial numbers at 15 hours post-infection with S . aureus . This suggests that RBF1 and dCAP-D3 most likely affect the resistance mechanisms ( i . e . AMP transcription ) , and not the tolerance mechanisms of the fly . Since the observed defects in survival rates and AMP induction were not as severe for RBF1 deficient flies in comparison to dCAP-D3 deficient flies , we wondered whether the other Drosophila RBF member , RBF2 , might compensate for loss of RBF1 activity . RBF2 has been shown to be upregulated upon depletion of RBF1 , and co-regulates many genes with RBF1 as a part of the dREAM complex [26] , [56] , [57] . To address this question , we tested survival times and AMP induction in flies deficient for RBF2 in the fat body ( yolk-GAL4<UAS-RBF2-dsRNA ) or a combination of both RBF1 and RBF2 in the fat body ( yolk-GAL4<UAS-RBF1 dsRNA , UAS-RBF2 dsRNA ) . The specific deficiencies in these flies were confirmed by qRT-PCR ( Figure S6 ) . qRT-PCR revealed that similar to the loss of RBF1 or dCAP-D3 , loss of RBF2 or RBF1/RBF2 resulted in decreased basal transcript levels of diptericin but not drosomycin ( Figure S7A and S7B , inset boxes ) . However , following infection with S . aureus , loss of RBF2 or RBF1/RBF2 did not cause decreased induction of either AMP transcript . In some cases , loss of both RBF1 and RBF2 actually resulted in an increase in diptericin transcription levels at 8 hours post infection . In response to infection with Gram positive bacteria ( Figure S8A ) or Gram negative bacteria ( Figure S8B ) , RBF2 deficient or RBF1/RBF2 deficient flies did not exhibit any changes in survival rates that were significantly different from wild type control flies . These results demonstrate that RBF2 does regulate basal AMP transcript levels , but does not compensate for RBF1 in induction of AMP transcription in Drosophila following infection . AMPs are conserved in many metazoans and play a very important role in fighting pathogens in barrier epithelial cells at mucosal surfaces [58] . pRB and CAP-D3 have been previously shown to interact physically and functionally in human cells [11] . Remarkably , and perhaps unexpectedly , the regulation of AMP genes by RB and CAP-D3 proteins may also be conserved in human cells . To determine whether pRB and CAP-D3 could regulate genes in human cells and , more specifically , whether the co-regulation of AMPs was conserved , siRNAs were used to decrease pRB and CAP-D3 expression in human Retinal Pigment Epithelial ( RPE-1 ) cells and in premonocytic U937 cells . ( Figure S9A and data not shown ) . qRT-PCR analyses of the levels of five different AMPs revealed that two AMPs ( DEFB-3 and DEFA-1 ) were expressed in RPE-1 cells and both genes were significantly downregulated following the depletion of either pRB or CAP-D3 ( Figure S9B ) . Interestingly , these genes are also located in a very large gene cluster , the Defensin locus , encompassing over 20 different AMPs . These data raise the possibility that the regulation of AMPs by CAP-D3 and pRB , and the ability of these proteins to regulate gene clusters , are properties that may be conserved in human cells .
In Drosophila , RB-family proteins are best known as transcriptional repressors of cell cycle and proliferation genes . Here we describe a different aspect of RB function and show that , together with the Condensin II protein dCAP-D3 , RBF1 functions to regulate the expression of a large number of genes during Drosophila development . A surprising characteristic of RBF1/dCAP-D3 regulated genes is that they do not seem to be the classically repressed genes with functions in cell cycle progression , DNA damage and DNA replication . Instead , many RBF1/dCAP-D3-dependent genes are classified as being involved in cell-type specific functions and include genes that are involved in enzymatic cascades , organ development and cell fate commitment . The idea that dCAP-D3 and RBF1 could cooperate to promote tissue development and differentiation is supported by the fact that both proteins are most highly expressed in the late stages of the fly life cycle , and accumulate at high levels in the nuclei of specific cell types in adult tissues . As an illustration of the cell-type specific nature of RBF1/dCAP-D3-regulation we show that dCAP-D3 and RBF1 are both required for the constituitive expression of a large set of AMP genes in fat body cells . The loss of this regulation compromises pathogen-induction of gene expression and has functional consequences for innate immunity . Interestingly , different sets of RBF1/dCAP-D3-dependent genes were evident in the gene expression profiles of mutant larvae and adults . Given this , and the fact that the gene ontology classification revealed multiple groups of genes , we suggest that the targets of RBF1/dCAP-D3-regulation do not represent a single transcriptional program , but diverse sets of cell-type specific programs that need to be activated ( or repressed ) in specific developmental contexts . The changes in gene expression seen in the mutant flies suggest that RBF1 has a significant impact on the expression of nearly half of the dCAP-D3-dependent genes . This fraction is consistent with our previous data showing partial overlap between RBF1 and dCAP-D3 banding patterns on polytene chromatin , and the finding that chromatin-association by dCAP-D3 is reduced , but not eliminated , in rbf1 mutant animals and RBF1-depeleted cells . Although we have previously shown that RBF1 and dCAP-D3 physically associate with one another [11] , and our current studies illustrate the fact that they each bind to similar sites at a direct target , the molecular events that mediate the co-operation between RBF1 and dCAP-D3 remain unknown . These results represent the first published ChIP data for the CAP-D3 protein in any organism . Although we have only examined a small number of targets it is interesting to note that the dCAP-D3 binding patterns are different for activated and repressed genes ( compare Figure 5 and Figure S5 ) . More specifically , dCAP-D3 binds to an area within the open reading frame of a gene which it represses ( Figure S5C and S5D ) . However , dCAP-D3 binds to regions which flank a cluster of genes that it activates ( Figure 5 ) . Whether or not this difference in binding is true for all dCAP-D3 regulated genes will require a more global analysis . Human Condensin non-SMC subunits are capable of forming subcomplexes in vitro that are separate from the SMC protein- containing holocomplex [59] , but currently , the extent to which dCAP-D3 relies on the other members of the Condensin II complex remains unclear . We note that fat body cells contain polytene chromatin . Condensin II subunits have been shown to play a role in the organization of polytene chromatin in Drosophila nurse cells [60] . Given that RB proteins physically interact with other members of the Condensin II complex [11] , it is possible that RBF1 and the entire Condensin II complex , including dCAP-D3 , may be especially important for the regulation of transcription on this type of chromatin template . A potentially significant insight is that the genes that are deregulated in both rbf1 and dCap-D3 mutants tend to be present in clusters located within 10 kb of one another . This clustering effect seems to be a more general feature of regulation by dCAP-D3 , which is enhanced by RBF1 , since clustering was far more prevalent in the list of dCAP-D3 target genes than in the list of RBF1 target genes . We chose to focus our studies on one of the most functionally related families of clustered target genes that were co-dependent on RBF1/dCAP-D3 for activation in the adult fly: the AMP family of genes . AMP loci represent 20% of the gene clusters regulated by RBF1 and dCAP-D3 in adults . ChIP analysis of one such region , a cluster of AMP genes at the diptericin locus , showed this locus to be directly regulated by RBF1 and dCAP-D3 in the fat body and revealed a pattern of RBF1 and dCAP-D3-binding that was very different from the binding sites typically mapped at E2F targets . Unlike the promoter-proximal binding sites typically mapped at E2F-regulated promoters , RBF1 and dCAP-D3 bound to two distant regions , one upstream of the promoter and one downstream of the diptericin B translation termination codon , a pattern that is suggestive of an insulator function . We hypothesize that RBF1 and dCAP-D3 act to keep the region surrounding AMP loci insulated from chromatin modifiers and accessible to transcription factors needed for basal levels of transcription . The modEncode database shows binding sites for multiple insulator proteins , as well as GATA factor binding sites , at these regions . GATA has been previously implicated in transcriptional regulation of AMPs in the fly [61] , and future studies of dCAP-D3 binding partners in Drosophila fat body tissue may uncover other essential activators . Additionally , the chromatin regulating complex , Cohesin , which exhibits an almost identical structure to Condensin [62]–[64] , has been shown to promote looping of chromatin and to bind proteins with insulator functions [65] , [66] . Therefore , it remains a possibility that Condensin II , dCAP-D3 may actually possess insulator function , itself . We would like to propose that dCAP-D3 may be functioning as an insulator protein , both insulating regions of DNA containing clusters of genes from the spread of histone marks and possibly looping these regions away from the rest of the body of chromatin . This would serve to keep the region in a “poised state” available for transcription factor binding following exposure to stimuli that would induce activation . In the case of AMP genes , which are made constituitively in specific organs at low levels [37] , [67] , [68] , dCAP-D3 would bind to regions flanking a cluster , and loop the cluster away from the body of chromatin . Upon systemic infection , these clusters would be more easily accessible to transcription factors like NF-κB . If dCAP-D3 is involved in looping of AMP clusters , then it may also regulate interchromosomal looping which could bring AMP clusters on different chromosomes closer together in 3D space , allowing for a faster and more coordinated activation of all AMPs . AMP expression is essential for the ability of the fly to recover from bacterial infection . Experiments with bacterial pathogens show that RBF1 and dCAP-D3 are both necessary for induction and maintenance of the AMP gene , drosomycin following infection , but only dCAP-D3 is necessary for the induction of the diptericin AMP gene . Similarly , survival curves indicate , that while dCAP-D3 deficient flies die more quickly in response to both Gram positive and Gram negative bacterial infection , RBF1 deficient flies only die faster in response to Gram positive bacterial infection . The differences seen between RBF1 and dCAP-D3 deficient flies in diptericin induction cannot be attributed to functional compensation by the other Drosophila RB protein family member , RBF2 , since results show that loss of RBF2 or both RBF2 and RBF1 do not decrease AMP levels following infection . Since results demonstrate that RBF1 binds most strongly to an AMP cluster prior to infection and regulates basal levels of almost all AMPs tested , we hypothesize that RBF1 ( and possibly RBF2 ) may be more important for cooperating with dCAP-D3 to regulate basal levels of AMPs . Reports have shown that basal expression levels of various AMPs are regulated in a gene- , sex- , and tissue-specific manner , and it is thought that constituitive AMP expression may help to maintain a proper balance of microbial flora and/or help to prevent the onset of infections [37] , [68] , [69] . In support of this idea , one study in Drosophila which characterized loss of function mutants for a gene called caspar , showed that caspar mutants increased constituitive transcript levels of diptericin but not transcript levels following infection . This correlated with increased resistance to septic infection with Gram negative bacteria [70] , proving that changes in basal levels of AMPs do have significant effects on the survival of infected flies . Additionally , disruption of Caudal expression , a protein which suppresses NF-κB mediated AMP expression following exposure to commensal bacteria , causes severe defects in the mutualistic interaction between gut and commensal bacteria [71] . It is therefore possible that RBF1 and dCAP-D3 may help to maintain the balance of microbial flora in specific organs of the adult fly and/or be involved in a surveillance-type mechanism to prevent the start of infection . RBF1 deficient flies also exhibit defects in Drosomycin induction following Gram positive bacterial infection . Mutation to Drosophila GNBP-1 , an immune recognition protein required to activate the Toll pathway in response to infection with Gram positive bacteria has been show to result in decreased Drosomycin induction and decreased survival rates , without affecting expression of Diptericin [72] , [73] . Therefore , it is possible that inefficient levels of Drosomycin , a major downstream effector of the Toll receptor pathway , combined with decreased basal transcription levels of a majority of the other AMPs , would cause RBF1 deficient flies to die faster following infection with Gram positive S . aureus but not Gram negative P . aeruginosa . Some dCAP-D3 remains localized to DNA in RBF1 deficient flies [11] and it is also possible that other proteins may help to promote the localization of dCAP-D3 to AMP gene clusters following infection . Given that dCAP-D3 regulates many AMPs including some that do not also depend on RBF1 for activation , and given that dCAP-D3 binding to an AMP locus increases with time after infection whereas RBF1 binding is at its highest levels at the start of infection , it may not be too surprising that dCAP-D3 showed a more pronounced biological role in pathogen assays involving two different species of bacteria . Remarkably , and perhaps unexpectedly , the levels of both RBF1 and dCAP-D3 impact the basal levels of human AMP transcripts , as well . This indicates that the mechanism of RBF1/dCAP-D3 regulation may not be unique to Drosophila . It is striking that many of the human AMP genes ( namely , the defensins ) are clustered together in a region that spans approximately 1 Mb of DNA . It seems telling that both the clustering of these genes , and a dependence on pRB and CAP-D3 , is apparently conserved from flies to humans . The fact that dCAP-D3 and RBF1 dependent activation of Drosomycin was necessary for resistance to Gram positive bacterial infection in flies suggests the same could also be true for the human orthologs in human cells . Human AMPs expressed by epithelial cells , phagocytes and neutrophils are an important component of the human innate immune system . Human AMPs are often downregulated by various microbial pathogenicity mechanisms upon infection [58] , [74]–[76] . They have also been reported to play roles in the suppression of various diseases and maladies including cancer and Inflammatory Bowel Disease [77] . We note that the chronic or acute loss of Rb expression from MEFs resulted in an unexplained decrease in the expression of a large number of genes that are involved in the innate immune system [78] . In humans , the bacterium , Shigella flexneri was recently shown to down regulate the host innate immune response by specifically binding to the LXCXE cleft of pRB , the same site that we had previously shown to be necessary for CAP-D3 binding [11] , [79] . An improved understanding of how RB and CAP-D3 regulate AMPs in human cells may provide insight into how these proteins are able to regulate clusters of genes , and may also open up new avenues for therapeutic targeting of infection and disease . Further studies of in differentiated human cells may identify additional sets of genes that are regulated by pRB and CAP-D3 .
W1118 flies were used as “wild type” controls for microarray experiments . Unless otherwise noted , the genotype of RBF mutants was a transheterozygous combination of rbf1Δ14/rbf1120a which was obtained by mating rbf1Δ14/FM7 , GFP virgins to rbf1120a/FM7 , GFP males at 18°C . Similarly , the genotype of CAP-D3 mutants was a transheterozygous combination of dCAP-D3Δ25/dCAP-D3c07081 which was obtained by mating dCAP-D3Δ25/CyO , GFP virgins to dCAP-D3c07081/CyO , GFP males at 23°C . yolk-GAL4/FM7c flies were a kind gift of M . Birnbaum and the timing of expression driven by yolk-GAL4 has been previously characterized in [40] . The RBF1 , dCAP-D3 , RBF2 , and IMD dsRNA expressing strains were obtained from the VDRC and their transformant IDs were 10696 , 29657 , 100635 , and 101834 respectively . UASt-FLAG-HA tagged strains were created by first amplifying the ORF from either the CAP-D3 RE18364 cDNA clone ( DGRC ) or the RBF1 LD02906 cDNA clone ( DGRC ) using Pfx polymerase ( Invitrogen ) . The pENTR/D-TOPO Cloning Kit ( Invitrogen ) was used to clone the ORF into a Gateway entry vector as described in the manufacturer's protocol and at http://www . ciwemb . edu/labs/murphy/Gateway%20vectors . html . The LR Clonase kit ( Invitrogen ) was then used to recombine the ORF into the pUASt-FHW vector ( DGRC ) described in detail at the website mentioned above . pUASt-FLAG-HA-RBF1 and pUASt-FLAG-HA-dCAP-D3 vectors were then injected into embryos to create transgenic fly lines expressing the tagged proteins . Mutant flies used as positive controls in infection experiments included the Imd1 strain which was a generous gift from L . Stuart and the Eater mutant strain [51] . All flies were maintained at 25°C and placed in vials containing standard dextrose medium . hTERT-RPE-1 cells were grown in Dulbecco'sModified Essential Medium ( DMEM ) supplemented with 10% fetal bovine serum ( FBS ) and 1% penicillin/streptomycin . RNAimax ( Invitrogen ) was used , according to manufacturer's protocol , to transfect non-targeting , RB , and CAP-D3 specific siRNAs ( described in [12] ) at final concentrations of 100 nM . Total RNA was harvested 48 hours post transfection and reverse transcribed into cDNA , as described below . TRIzol ( Invitrogen ) was used to harvest total RNA from whole flies/specific tissues according to the manufacturer's protocol . After RNA was purified using the Qiagen RNAeasy kit , the Taqman Reverse Transcription kit ( Applied Biosystems ) was used to reverse transcribe 1 . 5 µg of RNA into cDNA . qRT-PCR was performed using the Roche Lightcycler 480 to amplify 15 µL reactions containing . 5 µL of cDNA , . 5 µL of a 10 µM primer mix and 7 . 5 µL of SYBR Green Master Mix ( Roche ) . All qRT-PCR experiments were performed using three groups of 5 flies per genotype and three independent experiments were performed . Primer sequences are as follows: Rbf1qPCR F1-CTGCAGGGCTACGAGACGTAC , Rbf1qPCR R1 GTGTGCTGGTTCTTCGGCAGG , Rbf2qPCR R1-CTCCCAGTGCTTCTAGCACGC , Rbf2qPCR F1-CGTGAACGCCTTAGAGGTGCC , dCAP-D3 qPCR F3-CGTGCTGTTGCTTTACTTCGGCC , dCAP-D3 qPCR R3- GGCGCATGATGAAGAGCATATCCTG , AttAqPCR F1-GTGGTCCAGTCACAACTGGCG , AttAqPCR R1- CTTGGCATCCAGATTGTGTCTGCC , DroqPCR F1-CACCATCGTTTTCCTGCTGCTTGC , DroqPCR R1-GGTGATCCTCGATGGCCAGTG , AttBqPCR F1- CTCAAAGCGGTCCAGTCACAACTG , AttBqPCR R1- GAATAAATTGGCATGGGCCTCCTGC , Dro4qPCR F1- GTTTGCTCTCCTCGCTGTGGTG , Dro4qPCR R1-GCCCAGCAAGGACCACTGAATC , Dro3qPCR F1- GGCCAACACTGTTTTGGCACGTG , Dro3qPCR R1- GTCCCTCCTCAATGCAGAGACG , Dro2qPCR F1- GTTGTCCTGGCCGCCAATATGG , Dro2qPCR R1- GGACTGCAGTGGCCACTGATATG , DptBqPCR F1- GGACTGGCTTGTGCCTTCTCG , DptBqPCR R1- CAGGGGCACATCAAAATTGGGAGC , DrsqPCR F1-GTACTTGTTCGCCCTCTTCGCTG , DrsqPCR R1- CAGGTCTCGTTGTCCCAGACG , DptqPCR F1- GCTTATCCGATGCCCGACGAC , DptqPCR R1-GTGACCCTGGACTGCAAAGCC , DefqPCR F1- CAAACGCAGACGGCCTTGTCG , DefqPCR R1- AAGCGAGCCACATGCGACCTAC , Dro5qPCR F1- CAAGTTCCTGTACCTCTTCCTGGC , Dro5qPCR R1- CAGGGTCCTCCGTATCTTCCAG , Dro6 qPCR F1-CTTCGCACCAGCATTGCAGCC , Dro6qPCR R1- GAAGGTACAGACCTCCCTGTGC , Dro7qPCR F1- GGCTGCAGTGTCCACTGGTTC , Dro7qPCR R1- CACATGCCGACTGCCTTTCCG , MtkqPCR F1- GATTTTTCTGGCCCTGCTGGGTG , MtkqPCR R1- GGTTGGTTAGGATTGAAGGGCGAC , rp49qPCR F1- TACAGGCCCAAGATCGTGAAG , rp49qPCR R1- GACGCACTCTGTTGTCGATACC , CecCqPCR F1-CAATCGGAAGCCGGTTGGCTG , CecqPCR R1-GCGCAATTCCCAGTCCTTGAATGG , AndqPCR F1- CATTTTGGCCATCAGCGTGGGTC , AndqPCR R1- GGGCTTAGCAAAGCCAATTCCCAC , AttCqPCR F1- GTACTTGGCTCCCTTGCGGTG , AttCqPCR R1- CTTAGGTCCAATCGGGCATCGG , AttDqPCR F1- CCAAGGGAGTTTATGGAGCGGTC , AttDqPCR R1- GCTCTGGAAGAGATTGGCTTGGG , CecA1qPCR F1- CAATCGGAAGCTGGGTGGCTG , CecA1qPCR R1- GGCGGCTTGTTGAGCGATTCC , CecA2qPCR F1- GGACAATCGGAAGCTGGTTGGC , CecA2qPCR R1- GGCCTGTTGAGCGATTCCCAG , CecBqPCR F1- GATTCCGAGGACCTGGATTGAGG , CecBqPCR R1- GGCCATCAGCCTGGGAAACTC , tub84BqPCR F1- GGCAAGGAGATCGTCGATCTGG , tub84BqPCR R1- GACGCTCCATCAGCAGCGAG , hCAP-D3qPCR F1- TCCGGAAGCAGGCCCTCCAG , hCAP-D3qPCR R1- GGACCTGGCTGTCGTCCCCA , hRBqPCR F1- AGCTGTGGGACAGGGTTGTGTC , hRBqPCR R1- CAACCTCAAGAGCGCACGCC , eaterqPCR F1: CTCGTATCGGCTCAGATCTGCAC , eaterqPCR R1: CATCTGAGTGCGGAGCTCCTTAC , IMDqPCR F1- CGAATCCACTGGAGCAACAGCTG , IMDqPCR R1- GTTTCCACGCACTTGGGCGAG , hGAPDHqPCR F1- AGCCTCCCGCTTCGCTCTCT , hGAPDHqPCR R1- CCAGGCGCCCAATACGACCA , orc1qPCR F1- CATCATCCTCAAACACGCGCTGC , orc1qPCR R1- CCCTCGACGAGGCGTAAAAGC , cg5250qPCR F1- GACATTGCCGGAGGTGAAGAGC , cg5250qPCR R1- CTATTCGACTATGTGGTGGGCCTG , dupqPCR F1- GGGTGGCGGTATTTTTGTGGGAG , dupqPCR R1- CAACAGGAAACTCCGCGACGC , mus209qPCR F1- CTTGTCGAAGCCATCGGAACGC , mus209qPCR R1- GGGTCAAGCCACCATCCTGAAG , dnkqPCR F1- CCGCCCCAACCAACAAGAAGC , dnkqPCR R1- CCTCCAGCGTATTGTACATGCCC , RnrSqPCR F1- GAAGAAGGCAAGCACGTGCGAG , RnrSqPCR R1- CCAGTACCACGACATCTGGCAG , dnapoldeltaqPCR F1- CCATCGCCCATTAGCAGAGTCTG , dnapoldeltaqPCR R1- GGAACCTCCAATGGACATGCCAAG , mcm7qPCR F1- CATTGAGCACCGCCTGATGATGG , mcm7qPCR R1- GAGTGCGCCTTCTCTGTGGAC , mcm3qPCR F1- CGAGGTGATGGAACAGGGTCG , mcm3qPCR R1- GAAAGCAGCGAATCCTGCAGTCC , mcm2qPCR F1- GAGATCCCGCAGGACTTGTTGC , mcm2qPCR R1- CAAAAGACTCCTGTCGCAGCTGG , mcm5qPCR F1- CTGGTCTCACGGCTTCGGTTATG , mcm5qPCR R1- GCCACACGATCATCCTCTCGC , dnapolalpha50qPCR F1- CCTTCTACCGTTGGCTATCGTATGG , dnapolalpha50qPCR R1- CAGCTTGGGTATCAAAGCAGAGG , DEFA-1qPCR F1- TGCCCTCTCTGGTCACCCTGC , DEFA-1qPCR R1- GCCTGGAGTGGCTCAGCCTG , DEFB-3qPCR F1- GCGTGGGGTGAAGCCTAGCA , DEFB-3qPCR R1- AGCTGAGCACAGCACACCGG . The rabbit anti–dCAP-D3 YZ834 antibody was generated by Yenzyme Corporation . The antibody was purified using the BIO-RAD Affi-Gel 10 Gel according to manufacturer's protocol . Adult female flies were cryosectioned ( 10 µm ) and stained as previously described [80] . Primary antibodies included RBF1 ( DX2 ) , dCAP-D3 ( YZ384 ) , and anti-GFP ( Jackson Immunoresearch ) . Images were obtained using a Zeiss LSM510 Confocal microscope . Nimblegen microarray data were pretreated according to the manufacturer's recommendation and replicate probes were averaged . Affymetrix microarray data was downloaded from array express as raw . CEL files and normalized by robust multi array averaging ( RMA ) [RMS] before further analysis [81] . The entire set of microarray data can be found in Table S1 . Differentially expressed genes were identified using a linear model with a moderated T-test [82] . P values were corrected for multiple testing by calculating false discovery rates using the method of Benjamini and Hochberg [83] . Genes with a false discovery rate ( FDR ) <0 . 15 and a log2 fold change >0 . 1 were taken as significant . Gene ontology ( GO ) annotations were downloaded from FLYBASE [84] , and gene ontology terms overrepresented on the lists of differentially expressed genes were identified using a hypergeometric test . P-values from the hypergeometric test were corrected for multiple testing using the same method as for the individual genes and GO-categories with FDR<0 . 05 were taken as significant . Chromosomal positions of transcription start and stop sites for all genes on the chip were taken from FLYBASE . Genes were counted as clustered if they overlapped , or if the genes lay within 10 000 base pairs of each other . Overall chromosomal clustering for a list of genes was quantified as the number of genes that co-localize according to this criterion . Significance of co-localization was evaluated by comparing to lists of randomly selected genes from the same chip . S . aureus and P . aeruginosa bacteria were gifts from L . Stuart . S . aureus was grown in a shaking incubator at 37°C , in DIFCO Columbia broth ( BD Biosciences ) supplemented with 2% NaCl and P . aeruginosa was grown in a shaking incubator at 37°C in DIFCO Luria broth ( BD Biosciences ) . Bacteria were inoculated in 10 mL cultures grown overnight . 10∧4 bacterial cells were then inoculated into a new 10 mL culture and this was grown to an OD600 nm of 0 . 5 . These cultures were then centrifuged at 3000 rpm in a 1 . 5 mL eppendorf tube for 5 minutes at 4°C and subsequently washed twice with PBS . After a third centrifugation , PBS wash was removed from the pellet and 25 µL of new PBS was used to resuspend the pellet . Infections were performed as previously described [85] . Specifically , a . 25 mm diameter straight stainless steel needle and pin vise ( Ted Pella Inc , Redding , CA ) were used to infect adult flies . The needle was dipped into the resuspended bacterial pellet and used to prick the thorax of a CO2-anesthetized adult fly in a region just underneath where the wing connects to the thorax . Flies were then separated from the needle using a brush and put into fresh vials containing standard dextrose medium with no more than 10 flies per vial . 40 flies per IP were used in all ChIP experiments . Flies were homogenized with a KONTES pellet pestle grinder ( Kimble Chase ) in 1 mL of buffer A ( 60 mM KCl , 15 mM NaCl , 4 mM MgCl2 , 15 mM HEPES pH 7 . 6 , . 5% Triton X-100 , . 5 mM DTT , EDTA-free protease inhibitors cocktail ( Roche ) ) containing 1 . 8% formaldehyde . Homogenized flies were incubated at RT for 15 minutes , at which point glycine was added to a concentration of 225 mM . 2–4 mLs of homogenized flies were transferred to 15 mL conical tubes and centrifuged at 4°C for 5 min at 4000 g . Supernatant was discarded and pellets were washed with 3 mL of buffer A . Tubes were centrifuged as described above , supernatant was discarded , and pellets were washed with 3 mL of buffer B ( 140 mM NaCl , 15 mM HEPES pH 7 . 6 , 1 mM EDTA , . 5 mM EGTA , 1% Triton , . 5 mM DTT , . 1% sodium deoxycholate , EDTA free protease inhibitors cocktail ) . Tubes were centrifuged as described above , supernatant was discarded , and 500 µL of buffer B+1% SDS per IP was added to each tube . Tubes were rotated at 4°C for 20 min . Samples were then sonicated using the Branson sonifier at a setting of 3 , with 8 sonication intervals of 20 seconds interspersed by 10 second breaks . Tubes were centrifuged at 4°C for 5 min at 2000 RPM and 500 µL supernatant was used for each IP . 50 µL of Dynal Protein A beads ( Invitrogen ) per IP were prepared according to the manufacturer's recommendations . Beads were incubated with anti-FLAG M2 antibody ( Sigma ) or dCAP-D3 antibody ( YZ384 ) for 2 hours at RT with rotation . Beads were washed according to manufacturer's protocol and added to the diluted chromatin samples which were then incubated at 4°C overnight , with rotation . Samples were centrifuged at 3000 RPM , 4°C for 1 min and washed three times with buffer B+ . 05% SDS and once with TE . Bound protein was eluted by adding 125 µL of Buffer C ( 1%SDS , . 2% NaCl , TE ) to the beads for 30 min at 65°C . Samples were again centrifuged and eluates were harvested and incubated for 4 hours at 65°C to reverse crosslinks . Samples were digested with Proteinase K and RNase A ( Sigma ) , phenol-chloroform extracted , and ethanol precipitated . DNA pellets were dissolved in 105 µL of ddH2O and . 5 µL was used per qRT-PCR reaction . Flies were collected approximately 5–8 days after eclosure and were infected as described above . Following infection , each group of flies was placed in a new vial of food and monitored for the number of surviving flies at each timepoint . Three experiments were performed , with each experiment including 3 groups of 10 flies per genotype per timepoint . Survival statistics were calculated using a cox proportional hazard model , and hazard ratios with a two sided p-value less than 0 . 05 were taken as significant . Flies were anestitized by CO2 inhalation and infected as described above . Following infection , flies were dipped in 95% Ethanol , air dried , and placed into 1 . 5 mL Eppendorf tubes containing 500 µL of PBS . Flies were homogenized with a Kontes battery powered homogenizer and plastic pestle ( USA scientific ) . The tubes were centrifuged for 2 min at 3000 rpm . Various dilutions were plated onto Columbia CNA with 5% Sheep's Blood Agar ( Becton Dickinson and Company ) . This type of agar contains antibiotics to inhibit growth of organisms other than Staphylococcus aureus . | The retinoblastoma protein ( pRB ) is a tumor suppressor protein known for its ability to repress transcription of E2F-dependent genes and induce cell cycle arrest . We have previously shown that RB proteins in Drosophila and human cells interact with the Condensin II subunit , CAP-D3 , in an E2F-independent manner . Condensins promote condensation of chomosomes in mitosis . Our previous studies suggested that the Drosophila pRB and CAP-D3 homologs , RBF1 and dCAP-D3 , co-localize on DNA and may share a function in cells that never undergo mitosis . In this study , we show that one non-mitotic function shared between RBF1 and dCAP-D3 is the regulation of many non-cell-cycle-related , clustered , and cell-type-specific transcripts including a conserved family of genes that are important for the immune response in the fly . In fact , results show that normal levels of dCAP-D3 and RBF1 expression are necessary for the ability of the fly to clear infection with human bacterial pathogens . This work demonstrates that dCAP-D3 proteins can regulate a unique subset of RBF1-dependent transcripts in vivo and identifies a novel role for both RBF1 and dCAP-D3 protein in activation of innate immune genes , which may be conserved in human cells . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"immunity",
"gene",
"expression",
"genetics",
"molecular",
"genetics",
"biology",
"immunology",
"genetics",
"and",
"genomics"
] | 2012 | A Shared Role for RBF1 and dCAP-D3 in the Regulation of Transcription with Consequences for Innate Immunity |
No therapeutics or vaccines currently exist for human coronaviruses ( HCoVs ) . The Severe Acute Respiratory Syndrome-associated coronavirus ( SARS-CoV ) epidemic in 2002–2003 , and the recent emergence of Middle East Respiratory Syndrome coronavirus ( MERS-CoV ) in April 2012 , emphasize the high probability of future zoonotic HCoV emergence causing severe and lethal human disease . Additionally , the resistance of SARS-CoV to ribavirin ( RBV ) demonstrates the need to define new targets for inhibition of CoV replication . CoVs express a 3′-to-5′ exoribonuclease in nonstructural protein 14 ( nsp14-ExoN ) that is required for high-fidelity replication and is conserved across the CoV family . All genetic and biochemical data support the hypothesis that nsp14-ExoN has an RNA proofreading function . Thus , we hypothesized that ExoN is responsible for CoV resistance to RNA mutagens . We demonstrate that while wild-type ( ExoN+ ) CoVs were resistant to RBV and 5-fluorouracil ( 5-FU ) , CoVs lacking ExoN activity ( ExoN− ) were up to 300-fold more sensitive . While the primary antiviral activity of RBV against CoVs was not mutagenesis , ExoN− CoVs treated with 5-FU demonstrated both enhanced sensitivity during multi-cycle replication , as well as decreased specific infectivity , consistent with 5-FU functioning as a mutagen . Comparison of full-genome next-generation sequencing of 5-FU treated SARS-CoV populations revealed a 16-fold increase in the number of mutations within the ExoN− population as compared to ExoN+ . Ninety percent of these mutations represented A:G and U:C transitions , consistent with 5-FU incorporation during RNA synthesis . Together our results constitute direct evidence that CoV ExoN activity provides a critical proofreading function during virus replication . Furthermore , these studies identify ExoN as the first viral protein distinct from the RdRp that determines the sensitivity of RNA viruses to mutagens . Finally , our results show the importance of ExoN as a target for inhibition , and suggest that small-molecule inhibitors of ExoN activity could be potential pan-CoV therapeutics in combination with RBV or RNA mutagens .
The potential for CoVs to cause significant human disease is well demonstrated , with six known HCoVs—HKU1 , OC43 , NL63 , 229E , SARS-CoV and MERS-CoV—causing colds , pneumonia , systemic infection , and severe or lethal disease [1]–[5] . Four of these viruses have been identified in just the last 10 years , with two , SARS-CoV and MERS-CoV , causing lethal respiratory and systemic infection [1] , [3]–[6] . Studies over the past 10 years have expanded the known phylogenetic , geographic , and species diversity of CoVs , and support multiple emergence events of CoVs into humans from bats and other zoonotic pools [7]–[10] . The most recent evidence for CoV trans-species movement comes from the emergence of the novel MERS-CoV [1] , [11] , [12] . From April 2012 to June 2013 MERS-CoV has caused 72 laboratory confirmed cases and up to 50% mortality from severe respiratory and systemic disease in at least 8 countries , with evidence for human-to-human transmission [13] . MERS-CoV is most closely related to the bat CoVs HKU4 and HKU5 [11] , and the recently identified receptor dipeptidyl peptidase 4 ( DPP4 ) is present on both human and bat cells [14] , providing a compelling argument that zoonotic CoV infections resulting in severe human disease may be more frequent events than previously thought . Because of the lack of epidemiological data , it remains unknown whether multiple introductions from a zoonotic source or human transmission of a mild or asymptomatic disease is responsible for these continuing cases of sporadic severe infections . However , based on the high mortality rates associated with SARS-CoV and those reported for MERS-CoV [13] , this novel virus potentially represents a serious threat to global health for which no vaccines or therapeutics currently exist . CoVs contain the largest known RNA genomes ( 27–32 kb ) and encode an array of 16 viral replicase proteins , including a 3′-to-5′ exoribonuclease ( ExoN ) domain within nonstructural protein 14 ( nsp14 ) [2] , [15]–[17] . Similar to the proofreading subunit ( ε ) of E . coli DNA polymerase III , CoV nsp14-ExoN is a member of the DEDD superfamily of DNA and RNA exonucleases [15] , [18] . This superfamily contains four conserved D-E-D-D acidic residues that are required for enzymatic activity , and mutation of these critical residues within CoV ExoN ablates or significantly reduces ExoN activity [15] . Studies from our group have demonstrated that ExoN activity is essential for high-fidelity replication in both the model CoV murine hepatitis virus ( MHV ) and SARS-CoV [19] , [20] . Inactivation of ExoN activity due to alanine substitution of the first two active site residues results in 15- to 20-fold reduced replication fidelity in cell culture [19] , [20] and a 12-fold reduction during SARS-CoV infection in vivo [21] , associated with profound and stable attenuation of SARS-CoV virulence and replication . A recent study has shown that bacterially-expressed SARS-CoV nsp14-ExoN can remove mismatched nucleotides in vitro , and that ExoN activity is stimulated in vitro through interactions with the non-enzymatic CoV protein nsp10 [22] . Thus all bioinformatic , genetic and biochemical studies to date support the hypothesis that nsp14-ExoN is the first identified proofreading enzyme for an RNA virus and functions together with other CoV replicase proteins to perform the crucial role of maintaining CoV replication fidelity . Retrospective clinical studies during the SARS epidemic ultimately concluded that treatment with ribavirin ( RBV ) , an antiviral drug shown to be mutagenic for some RNA viruses [23] , [24] , was ineffective against SARS-CoV [25]–[28] . Because ExoN activity is required for CoV high-fidelity replication [19]–[21] , we sought to determine if ExoN was responsible for CoV resistance to RNA mutagens . Using the nucleoside analog RBV and the base analog 5-fluorouracil ( 5-FU; [29] ) we show that CoVs lacking ExoN activity ( ExoN− ) are up to 300-fold more sensitive to inhibition than wild-type CoVs ( ExoN+ ) . Additionally , using full-genome next-generation sequencing we show that ExoN− viruses accumulate 15- to 20-fold more A:G and U:C transitions , consistent with 5-FU incorporation during RNA synthesis . Ultimately our results suggest the exciting possibility that small-molecule inhibitors of ExoN activity could be potential pan-CoV therapeutics , especially when used in combination with RBV or RNA mutagens .
Murine astrocytoma delayed brain tumor cells ( DBT cells ) were grown at 37°C and maintained in DMEM ( Invitrogen ) containing 10% FBS , supplemented with penicillin , streptomycin , HEPES , and amphotericin B . VeroE6 ( Vero ) cells were grown at 37°C and maintained in MEM ( Invitrogen ) containing 10% FBS supplemented with penicillin , streptomycin , and amphotericin B . All work with MHV was performed using the reverse genetics infectious clone based on strain MHV-A59 [30] , and work with SARS-CoV was performed using the reverse genetics infectious clone based on the Urbani strain [31] . Viral studies using SARS-CoV were performed in Select Agent certified BSL-3 laboratories using protocols reviewed and approved by the Institutional Biosafety Committee of Vanderbilt University and the Centers for Disease Control for the safe study and maintenance of SARS-CoV . 5-fluorouracil ( 5-FU ) , ribavirin ( RBV ) , guanosine ( GUA ) and mycophenolic acid ( MPA ) were obtained from Sigma . 5-FU and RBV were made as 200 mM stock solutions , and were prepared in DMSO and sterile water , respectively . GUA and MPA were prepared in DMSO as 40 mM or 100 mM stocks , respectively . Low concentration ( µM ) working stocks were prepared as needed in sterile water prior to dilution in DMEM . Viability of DBT and Vero cells was assessed using CellTiter-Glo ( Promega ) in 96-well plate format according to manufacturer's instructions . DBT and Vero cells were seeded into opaque tissue culture grade 96-well plates , and DMEM containing RBV or 5-FU was added to each well to achieve the concentrations indicated . Water or DMSO vehicle controls were performed , in addition to a 20% ethanol control for cell death . The cells were then incubated at 37°C for either 12 or 24 h , and cell viability was determined using a Veritas Microplate Luminometer ( Promega ) . The resultant values were then normalized to untreated cells . Subconfluent monolayers of DBT cells in 6-well plates were pretreated for 30 min at 37°C with 1 mL of DMEM containing vehicle or the indicated concentration of RBV , 5-FU , MPA , or GUA . The drug was then removed and cells were infected with MHV-ExoN+ or ExoN− viruses at an MOI of 1 plaque forming units ( PFU ) /cell ( single-cycle ) or 0 . 01 ( multi-cycle ) for 30 min at 37°C . Virus was then removed and 1 mL of DMEM containing vehicle , RBV , 5-FU , MPA , or GUA was added to each well . Cells were then incubated at 37°C for either 12 ( single-cycle ) or 24 ( multi-cycle ) h . The supernatant was harvested and virus titer was determined by plaque assay on DBT cells . For SARS-CoV studies , subconfluent monolayers of Vero cells in T25 flasks were pretreated for 30 min at 37°C with DMEM containing vehicle , RBV , or 5-FU . The drug was removed and cells were infected with either SARS-ExoN+ or ExoN− viruses at an MOI of 0 . 1 PFU/cell ( single-cycle ) for 30 min . The virus was removed and DMEM containing vehicle , RBV , or 5-FU was added back . Cells were then incubated for 24 h , at which point the supernatant was harvested and virus titer was determined by plaque assay on Vero cells . All treated samples were normalized to the untreated vehicle control , and values were expressed as fold change from untreated virus titers . Viral RNA was harvested from infected cell monolayers using TRIzol reagent ( Invitrogen ) , and was reverse transcribed ( RT ) using SuperScript III ( Invitrogen ) . Random hexamers ( 1 µL of 50 µM stock ) and 1 µg of total RNA were incubated for 5 min at 70°C . The remaining reagents were then added according to the manufacturer's protocol , and the mixture was incubated at 50°C for 1 h and then at 85°C for 5 min . All RT reactions were performed in a final volume of 20 µL . Real-time qRT-PCR was performed on the RT product using the Applied Biosciences 7500 Real-Time PCR System with Power SYBR Green PCR Master Mix ( Life Technologies ) . Each reaction was performed in a total volume of 25 µL containing 12 . 5 µL of the Power SYBR Green PCR Master Mix , 125 ng each of the forward and reverse primers and 1 µL of the RT product which was diluted 1∶1000 . Viral genomic RNA was detected using primers ( forward: ACAGGGTGGAGTTCCCGTTA and reverse: ACGGAAGCACCACCATAAGA ) optimized to generate a ∼120 nt portion of ORF1a . These values were normalized using the 2−ΔΔCt method [32] to endogenous expression of the housekeeping gene glyceraldehyde-3- phosphate dehydrogenase ( GAPDH ) using primers ( forward: GGGTGTGAACCACGAGAAAT and reverse: CCTTCCACAATGCCAAAGTT ) optimized to yield a ∼120 nt portion of GAPDH [33] , [34] . Triplicate wells of each sample were analyzed , and averaged into one value representing a single replicate to minimize well-to-well variation . The cycle parameters were as follows: Stage 1 , ( 1 rep ) at 50°C for 2 min; Stage 2 , ( 1 rep ) 95°C for 10 min; Stage 3 , ( 40 reps ) at 95°C for 15 sec and 57°C for 1 min . One representative product from each treatment was verified by melting curve analysis and agarose gel electrophoresis . Viral RNA from SARS-ExoN+ or ExoN− infected Vero monolayers was harvested using TRIzol reagent , and was reverse transcribed ( RT ) using SuperScript III as described above except with 5 µL of random hexamers ( 50 µM stock ) , 5 µg of total RNA , and in a final volume of 100 µL for each reaction . Four microliters of RT product was then used to generate 12 overlapping ∼3 kb amplicons for each virus treated with either 0 or 400 µM 5-FU by PCR . The high-fidelity polymerase Easy A ( Agilent ) was used to ensure that errors were minimal during PCR . All primer sets generated single bands which were then purified using the Wizard SV Gel and PCR Clean-Up System ( Promega ) . Prior to sequencing , cDNA amplicons were fragmented ( Fragmentase , NEB ) , clustered , and sequenced with Illumina cBot and GAIIX technology as previously described [35] . Between 1 . 4×108 and 4 . 5×108 bases , comprised of ∼69-nt reads , were obtained per virus , and CASAVA 1 . 8 . 2 was used to demultiplex and create the fastq files . Low quality bases from the ends of each sequence read were then trimmed , using Phred scores as the guiding metric ( error probabilities higher than 0 . 001 ) , and sequences with less than 16 bases after trimming were discarded to reduce false alignment and subsequent false variant calls . The program fastq-clipper ( http://hannonlab . cshl . edu/fastx_toolkit/index . html ) was used for this quality filtering . The Burrows-Wheeler Alignment tool was then used to align reads to the SARS-CoV ExoN+ or ExoN− reference genomes with a maximum of two mismatches per read [36] . Base calling at each position was determined using SAMTOOLS [37] . After the pileup , an in-house script collected the data per-position . For each position throughout the viral genome , the bases and their qualities were gathered , each variant allele's rate was initially modified according to its covering read qualities based on a maximum likelihood estimation and test for significance using Wilks' theorem . Additionally , an allele confidence interval was calculated and output for each allele . Only alleles with statistically significant p<0 . 05 values were retained and considered to be true variants . Above 0 . 01% all variants were found to be statistically significant , while below 0 . 01% many variants could not be distinguished from background error . Thus , the background noise caused by sequencing error was determined to be 0 . 01% or less . Statistical tests were applied where noted within the figure legends and were determined using GraphPad Prism ( La Jolla , CA ) software . Statistical significance is denoted ( *P<0 . 05 , **P<0 . 01 , ***P<0 . 0001 ) and was determined using an unpaired , two-tailed Student's t test compared to either untreated samples or to the corresponding ExoN+ sample . For the cell viability studies , treated samples were compared to the DMEM sample containing DMSO .
Because RBV has been shown to be incorporated as ribavirin monophosphate ( RMP ) into viral RNA during replication [23] , [24] , [38]–[42] , the presence of a proofreading enzyme would be predicted to exclude and/or remove nucleotide misincorporation [43]–[47] . If ExoN is responsible for the resistance phenotype , viruses lacking ExoN activity ( ExoN− ) should demonstrate increased titer reduction following RBV treatment as compared to wild-type viruses containing ExoN activity ( ExoN+ ) . To test this hypothesis , we examined the sensitivity of MHV-ExoN+ and ExoN− viruses to RBV during single-cycle ( MOI = 1 PFU/cell ) replication in murine astrocytoma delayed brain tumor cells ( DBT cells ) . No toxicity was observed in DBT cells following treatment with up to 400 µM RBV ( Figure 1A ) . MHV-ExoN+ viruses were resistant to 10 µM RBV ( Figure 1B ) , while MHV-ExoN− virus titers decreased by ∼200-fold following treatment with 10 µM RBV . The capacity of 10 µM RBV to inhibit MHV-ExoN− replication is surprising because at least 10-fold higher concentrations of RBV are required to inhibit poliovirus and chikungunya viruses [48]–[50] . This observation could be due to the longer genomes of CoVs or to the mechanism ( s ) by which RBV inhibits CoV replication . If RBV is exerting antiviral activity primarily through mutagenesis following incorporation of RMP , MHV-ExoN− viruses should exhibit increased sensitivity during multi-cycle replication . To test this , we determined the sensitivity of MHV-ExoN+ and ExoN− viruses to RBV at a low multiplicity of infection ( MOI = 0 . 01 PFU/cell ) . Unexpectedly , multi-cycle replication of MHV-ExoN− viruses in the presence of RBV ( Figure 1B ) was indistinguishable from single-cycle replication . RBV has been reported to exert antiviral activity through numerous mechanisms [38] including disruption of viral RNA synthesis and inhibition of the cellular enzyme inosine monophosphate dehydrogenase ( IMPDH ) . To determine if RBV treatment was affecting CoV RNA synthesis , we performed two-step real-time quantitative reverse transcription PCR ( real-time qRT-PCR ) to determine viral genomic RNA levels in the presence or absence of RBV . Similar to Figure 1B , MHV-ExoN+ titers were unaffected , whereas there was a dose-dependent reduction in MHV-ExoN− titers following RBV treatment ( Figure 1C , filled bars ) . Corresponding dose-dependent reductions in MHV-ExoN− genomic RNA were observed ( Figure 1C , hatched bars ) following RBV treatment , demonstrating that treatment with 10 µM RBV decreased MHV-ExoN− RNA synthesis by nearly 100-fold during replication . Because RBV caused decreased RNA synthesis in MHV-ExoN− viruses , we calculated the relative specific infectivities of both viruses at each RBV concentration ( Table 1 ) . The relative specific infectivity of MHV-ExoN− viruses was decreased by 6- to 9-fold following treatment with RBV , while MHV-ExoN+ viruses were unaffected . In addition to decreasing viral RNA synthesis , RBV could be exerting antiviral activity against MHV-ExoN− through competitive inhibition of IMPDH by RMP [51] . To test this possible mechanism , we treated MHV-ExoN+ and MHV-ExoN− viruses with the specific IMPDH inhibitor mycophenolic acid ( MPA; [52]–[54] ) during both single- and multi-cycle replication . A concentration-dependent decrease in MHV-ExoN− virus titer was observed following MPA treatment during single-cycle replication ( Figure 1D ) . MHV-ExoN+ titers were reduced by less than 10-fold , consistent with what was observed following RBV treatment ( Figure 1B ) . Similar to RBV , increased sensitivity of MHV-ExoN− viruses to MPA was not observed during multi-cycle replication . If RBV is acting via IMDPH inhibition , addition of extracellular guanosine ( GUA ) should restore virus titers , as has been demonstrated previously for Dengue virus [55] . Addition of 100 µM GUA following RBV or MPA pretreatment and viral infection had no effect on MHV-ExoN+ viruses ( Figure 1E ) , but completely restored MHV-ExoN− titer even in the continued presence of 10 µM RBV or 1 µM MPA ( Figure 1F ) . These data indicate that the antiviral activity of RBV against MHV-ExoN− viruses is occurring at least in part through decreasing viral RNA synthesis and inhibition of IMPDH . Because our primary goal was to test the role of nsp14-ExoN in the prevention and/or removal of nucleotide misincorporation we did not further investigate how RBV was specifically inhibiting ExoN− viruses . However , these results do show that the presence of ExoN activity is capable of preventing RBV inhibition of CoV replication . We next examined the sensitivity of MHV-ExoN+ and ExoN− viruses to the pyrimidine base analog 5-FU , which has been shown to be mutagenic for many RNA viruses [29] , [56] . Treatment of DBT cells with up to 400 µM 5-FU did not result in any detectable cellular toxicity ( Figure 2A ) . Following treatment with up to 200 µM 5-FU ( Figure 2B ) during single-cycle infections , MHV-ExoN+ titers were inhibited less than 3-fold , while titers of MHV-ExoN− decreased ∼900 fold , representing a ∼300-fold increase in sensitivity as compared to MHV-ExoN+ . During multi-cycle replication , MHV-ExoN+ virus titers were reduced by less than 10-fold following 5-FU treatment , while MHV-ExoN− showed a ∼50 , 000-fold reduction in titer ( Figure 2B ) . Virus was undetectable by plaque assay at 5-FU concentrations above 80 µM . Analysis of viral RNA synthesis by two-step real-time qRT-PCR demonstrated that MHV-ExoN+ RNA levels were not reduced following 5-FU treatment , while 5-FU treatment resulted in minimal two-to-five fold decreases in MHV-ExoN− RNA ( Figure 2C ) . The specific infectivity of MHV-ExoN− was decreased by 14- and 128-fold following treatment with 100 µM and 200 µM 5-FU , respectively ( Table 1 ) . These results demonstrate that ExoN activity confers resistance to 5-FU , and support the hypothesis that 5-FU is driving increased genomic mutagenesis in MHV-ExoN− virus populations , leading to lethal mutagenesis and extinction . To determine whether SARS-CoV viruses lacking ExoN activity ( SARS-ExoN− ) also were inhibited by RBV and 5-FU , we infected Vero cells with either SARS-ExoN+ or ExoN− viruses in the presence or absence of RBV or 5-FU . Treatment of Vero cells with up to 400 µM RBV or 5-FU did not decrease cell viability by more than 20% ( Figure 3A ) . Recent reports have described the lack of RBV uptake by Vero cells due to the absence of specific equilibrative nucleoside transporters [57] , [58] . Additionally , previous studies have shown that RBV failed to inhibit SARS-CoV replication in Vero cells [59] . Consistent with those reports , in our experiments both SARS-ExoN+ and ExoN− viruses were unaffected by treatment with up to 400 µM RBV ( Figure 3B ) . We therefore performed subsequent experiments with 5-FU . SARS-ExoN+ titers were reduced 3- and 10-fold following treatment with 200 or 400 µM 5-FU , respectively ( Figure 3C ) . In contrast , SARS-ExoN− titers were reduced ∼300-fold by 200 µM 5-FU ( Figure 3C ) , similar to MHV-ExoN− viruses . At 400 µM 5-FU , SARS-ExoN− virus was inhibited 2 , 000-fold during a single replication cycle , representing a ∼160-fold increase in 5-FU sensitivity compared to SARS-ExoN+ viruses . Thus , our data indicate that increased sensitivity of CoVs to RNA mutagens in the absence of ExoN activity is conserved across diverse members of the CoV family . Of interest , our studies with SARS-ExoN+ also indicate that ExoN-mediated protection from nucleotide misincorporation can be overcome at higher concentrations of mutagen . Studies with the RNA viruses lymphocytic choriomeningitis virus ( LCMV ) , foot-and-mouth disease virus ( FMDV ) and vesicular stomatitis virus ( VSV ) have demonstrated that 5-FU is incorporated as 5-fluorouridine monophosphate ( FUMP ) into replicating viral RNA , thus increasing genomic mutations [60]–[62] . To determine whether 5-FU was causing increased mutagenesis in SARS-CoV populations , we performed full-genome NGS analysis of both virus populations replicating in the presence or absence of 5-FU . To analyze the entire spectrum of mutations arising during replication , we extracted total intracellular RNA from Vero cells infected with either SARS-ExoN+ or ExoN− viruses following treatment with either 0 µM or 400 µM 5-FU . We then generated 12 overlapping cDNA amplicons of approximately 3 kb in length for each sample . For each of the four samples , 1 . 4×108 to 4 . 5×108 bases were sequenced , corresponding to an average coverage depth of between 4 , 600 and 15 , 000 at each nucleotide position . We compared the statistically significant minority variants , defined as having a p-value of ≤0 . 05 following a multiple-testing correction ( Benjamini-Hochberg ) , between the untreated and 5-FU-treated SARS-ExoN+ and ExoN− populations . Following treatment with 400 µM 5-FU ( Figure 3D ) , there was an increase in mutations within the SARS-ExoN+ population from 11 to 259 ( 24-fold ) . In contrast , for SARS-ExoN− there were 3648 mutations present within the 5-FU-treated SARS-ExoN− population compared to the 99 mutations in the untreated population ( 40-fold increase ) . Most remarkably , this represented a 16-fold increase in the number of statistically significant minority variants between 5-FU treated ExoN+ and ExoN− SARS-CoV . Thus , these data support our hypothesis that 5-FU was increasing genomic mutations through incorporation of FUMP into viral genomes in the absence of ExoN activity . Incorporation of FUMP instead of uracil into replicating RNA allows FUMP to base pair with both guanosine and adenine [61] , [63] . This decreased specificity in base pairing has been shown in studies with LCMV and primarily results in A-to-G ( A:G ) and U-to-C ( U:C ) transitions [29] , [61] , [63] . To determine if FUMP was being incorporated at higher levels in the absence of ExoN-mediated proofreading , we analyzed the numbers and types of transitions and transversions occurring in each virus population ( Figure 4 ) . Transitions are indicated in grey boxes and transversions in white boxes , with the number for each shown . Transversions comprised the majority of variants for both untreated ExoN− and ExoN+ viruses . Treatment with 5-FU caused the number of U:C and A:G transitions to increase in both ExoN+ and ExoN− populations , from 2 to 197 for SARS-ExoN+ and from 16 to 3304 for SARS-ExoN− ( Figures 4A and B ) . This increase and bias toward U:C and A:G transitions is consistent with FUMP being incorporated into both minus- and plus-strand RNA [63] during both ExoN+ and ExoN− replication; however the absolute numbers were dramatically increased ( 16-fold ) during ExoN− replication compared to ExoN+ . In untreated cells , A:G and U:C transitions accounted for less than 25% of the total minority variants within each population ( Figure 4C ) . Following 5-FU treatment , A:G and U:C transitions accounted for 70–95% of the total minority variants within each population . To further examine the genomic distribution of these two transitions , we plotted the total number of A:G and U:C transitions occurring at a frequency of between 0 . 1% and 1% ( Figure 5 ) . Approximately 75% and 90% of the total minority variants occurring at a frequency between 0 . 1 and 1% following 5-FU treatment were due to A:G or U:C transitions ( Figure 5 ) , for the SARS-ExoN+ and ExoN− populations , respectively . In both populations , these mutations were distributed across the entire genome following treatment with 400 µM 5-FU . Thus our data provide direct evidence indicating that 5-FU drives increased genomic mutations within SARS-CoV in the absence of ExoN proofreading activity .
The antiviral nucleoside analog RBV is currently used to treat hepatitis C virus ( HCV; [73]–[75] ) , Lassa virus [76] and respiratory syncytial virus ( RSV ) infections [77] , [78] . The potential clinical use of RBV for CoV infections is complicated by the multiple mechanisms of action that have been reported [38] , and by the potential for disease exacerbation , as reported during the SARS-CoV epidemic [25]–[28] . Our data suggest that RBV primarily inhibits MHV-ExoN− virus replication through decreasing viral RNA synthesis and inhibition of IMPDH ( Figure 1 ) . Inhibition of IMPDH by RMP has been shown to decrease intracellular GTP pools [51] , thus altering the balance of nucleoside triphosphates ( NTPs ) within the cell . Decreased GTP levels could result in forced misincorporations due to NTP imbalances in the absence of ExoN activity [72] . However , the moderate 6- to 9-fold decreases in relative specific infectivity observed for MHV-ExoN− following RBV treatment ( Table 1 ) suggests that mutagenesis is not the primary mechanism by which RBV is exerting an antiviral effect . An additional possibility is that the antiviral activity of RBV against ExoN− viruses is unrelated to the putative proofreading function of this enzyme . Both biochemical and cell culture studies have demonstrated that loss of ExoN activity leads to impaired RNA synthesis [15] , [19] , [20] . Furthermore , in addition to ExoN activity , nsp14 contains N7-methyltransferase ( N7-MTase ) activity , a critical step in RNA capping [79] , [80] . A recent report has demonstrated that the ExoN and N7-MTase domains are structurally inseparable , and that residues within the ExoN domain are important for N7-MTase activity [81] . Thus , the increased sensitivity of MHV-ExoN− to RBV could result from the impairment of undefined functions of ExoN during CoV replication , particularly during RNA synthesis . The parallel use of ExoN+ and ExoN− viruses with RBV may allow us to define how RBV is exerting an antiviral effect against CoVs and the potentially novel mechanisms by which ExoN may act to counter that inhibition . Since the identification of nsp14-ExoN activity [15] and studies demonstrating the requirement for ExoN in high-fidelity replication [19]–[21] , mounting evidence points to a role for nsp14-ExoN in proofreading activity during RNA virus replication [22] . Here we used NGS to determine the number of mutations present in SARS-ExoN+ and ExoN− populations . The characteristic 5-FU-mediated transitions U:C and A:G comprised 90% of the total statistically significant minority variants within SARS-ExoN− population , and were present at levels 15- and 20-fold higher than those same transitions within the ExoN+ population ( Figure 4 ) . Overall , our data represent the first direct test of ExoN proofreading during SARS-CoV replication in the absence of ExoN . Furthermore , the sequencing depth attained using NGS shows that ExoN inactivation likely skews the spectrum of spontaneous mutations present within the untreated population ( Figure 4 ) . Such overrepresentation of specific mutations in the context of ExoN inactivation is similar to studies of S . cerevisiae DNA polymerases ε and δ containing mutations within their respective 3′-to-5′ DEDD exonucleases [82]–[86] . This altered distribution due to ExoN inactivation could have profound implications for CoV adaptation and evolution . Lethal mutagenesis occurs through the accumulation of mutations within the viral genome during replication , and ultimately results in virus extinction ( reviewed in [56] , [87] ) . While lethal mutagenesis has been studied extensively [87] , our work is the first to identify an RNA virus protein distinct from the RdRp that directly regulates the sensitivity of RNA viruses to genomic mutations resulting from mutagen incorporation . Currently , RBV is the only FDA-approved antiviral with demonstrated mutagenic activity . The first demonstration of RBV acting as a mutagen was performed using poliovirus [23] , [24] almost 30 years after the antiviral activity of RBV was described [88] . The nucleoside analog T-705 ( Favipiravir; [89] ) is currently in clinical development , and has been shown recently to drive lethal mutagenesis of influenza virus [90] . We have shown that ExoN+ viruses replicate well in the presence of RBV or 5-FU . However , we also have shown that ExoN− mutants of SARS-CoV and MHV have 15-to-20-fold decreased fidelity [19] , [20] , are attenuated , are subject to rapid loss of replication and clearance in vivo [21] , and are highly susceptible to low concentrations of RNA mutagens . An exciting possibility is that this conserved CoV proofreading enzyme could be targeted for inhibition , thus leading to the development of broadly useful CoV therapeutics . While ExoN inhibitors alone might be efficacious , combining an inhibitor of CoV fidelity with an RNA mutagen would magnify the intrinsic fidelity defect of ExoN inhibition and drive high-level mutagenesis . A potential advantage of such an approach would be to rapidly drive the virus to extinction , while limiting or blocking the capacity of the virus to overcome inhibition by reversion . ExoN− mutants of both MHV and SARS-CoV have shown no reversion over multiple passages in culture or during persistent infections in vivo [19]–[21] . Furthermore , we did not observe any primary reversions within the ExoN DEDD motif following 5-FU treatment . While mutations within the CoV RdRp could emerge during acute treatment , mutations within other RNA virus RdRps have demonstrated that the maximum tolerance for increased or decreased fidelity without loss of virus viability is between ∼3- to 6-fold [35] , [48] , [69] , [91] . In addition , our data demonstrate that ExoN− viruses are profoundly sensitive to inhibition by lower concentrations of mutagen , providing a possible improved therapeutic index and margin of safety for use . In summary , this study provides the most direct evidence to date that CoV ExoN provides a proofreading function during virus replication , and identifies ExoN as the critical determinant of CoV sensitivity to RNA mutagens . Because CoV replication fidelity is likely determined by the concerted effort of multiple virus proteins [19] , [20] , [22] , our data suggest the exciting possibility that significant attenuation of CoV fitness and pathogenesis could be achieved by targeting the conserved process of CoV replication fidelity . Ultimately , uncovering the mechanism of fidelity regulation and methodologies to disrupt this critical process will be vital to responding to both endemic and future emerging CoVs such as SARS-CoV and MERS-CoV . | RNA viruses have high mutation rates ( 10−3 to 10−5 mutations/nucleotide/round of replication ) , allowing for rapid viral adaptation in response to selective pressure . While RNA viruses have long been considered unable to correct mistakes during replication , CoVs such as SARS-CoV and the recently emerged MERS-CoV are important exceptions to this paradigm . All CoVs encode an exoribonuclease activity in nonstructural protein 14 ( nsp14-ExoN ) that is proposed to prevent and/or remove misincorporated nucleotides . Because of the demonstrated resistance of SARS-CoV to the antiviral drug ribavirin ( RBV ) , we hypothesized that ExoN is responsible for CoV resistance to RNA mutagens . Using RBV and the RNA mutagen 5-fluorouracil ( 5-FU ) , we show that CoVs lacking ExoN activity ( ExoN− ) are highly susceptible to RBV and 5-FU , in contrast to wild-type ( ExoN+ ) CoVs . The inhibitory activity of 5-FU against ExoN− viruses resulted specifically from 5-FU incorporation during viral RNA synthesis that lead to extensive mutagenesis within the viral population , and was associated with a profound decrease in virus specific infectivity . These results demonstrate the proofreading activity of ExoN during virus replication and suggest that inhibitors of ExoN activity could be broadly useful inhibitors of CoV replication in combination with RBV or RNA mutagens . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"mechanisms",
"of",
"resistance",
"and",
"susceptibility",
"virology",
"biology",
"microbiology"
] | 2013 | Coronaviruses Lacking Exoribonuclease Activity Are Susceptible to Lethal Mutagenesis: Evidence for Proofreading and Potential Therapeutics |
Numerous bacterial pathogens manipulate host cell processes to promote infection and ultimately cause disease through the action of proteins that they directly inject into host cells . Identification of the targets and molecular mechanisms of action used by these bacterial effector proteins is critical to understanding pathogenesis . We have developed a systems biological approach using the yeast Saccharomyces cerevisiae that can expedite the identification of cellular processes targeted by bacterial effector proteins . We systematically screened the viable yeast haploid deletion strain collection for mutants hypersensitive to expression of the Shigella type III effector OspF . Statistical data mining of the results identified several cellular processes , including cell wall biogenesis , which when impaired by a deletion caused yeast to be hypersensitive to OspF expression . Microarray experiments revealed that OspF expression resulted in reversed regulation of genes regulated by the yeast cell wall integrity pathway . The yeast cell wall integrity pathway is a highly conserved mitogen-activated protein kinase ( MAPK ) signaling pathway , normally activated in response to cell wall perturbations . Together these results led us to hypothesize and subsequently demonstrate that OspF inhibited both yeast and mammalian MAPK signaling cascades . Furthermore , inhibition of MAPK signaling by OspF is associated with attenuation of the host innate immune response to Shigella infection in a mouse model . These studies demonstrate how yeast systems biology can facilitate functional characterization of pathogenic bacterial effector proteins .
Bacterial pathogens have evolved numerous mechanisms to evade host cell defenses and promote infection . One strategy common to many pathogens is the manipulation of host cell processes by bacterial effector proteins that are directly delivered into host cells by specialized secretion systems [1] . Although these effector proteins are critical to pathogenesis , relatively few are well characterized . A number of effectors whose functions are understood manifest activity in Saccharomyces cerevisiae analogous to their activity in mammalian cells during infection , presumably because these proteins target fundamental cellular processes conserved among all eukaryotes . Consequently , we and others have exploited yeast as a model organism in which to identify and characterize bacterial effector proteins [2–5] . When expressed in yeast , bacterial effectors that target conserved eukaryotic cellular processes often inhibit growth [2 , 3 , 6] and/or exhibit conserved subcellular localization patterns [3 , 7] . Growth inhibition provides a measurable phenotype that can indicate the degree of disruption of cellular processes [8 , 9] . Because over 75% of the yeast genome is functionally annotated , systematic identification of genes or proteins that modulate a phenotype can identify pathways and processes involved in that phenotype [9–13] , which , in turn , can provide insights into its etiology . Thus , by systematically screening for yeast deletion strains hypersensitive to expression of a bacterial effector protein , cellular processes that buffer yeast against the toxicity of the effector can be identified . As proof of principal , we applied this approach to OspF , a Shigella effector protein . OspF is found in all pathogenic Shigella species [14] and is an established substrate of the Shigella flexneri type III secretion system [15] . At the start of this study , nothing was known about its molecular activity or role in pathogenesis . Having observed its mild toxicity to wild-type yeast under select conditions , we screened for yeast deletion strains hypersensitive to OspF expression . In parallel , we measured alterations in mRNA levels in wild-type yeast expressing OspF . The resulting complementary views of OspF activity led us to hypothesize , and subsequently demonstrate , that OspF inhibits both yeast and mammalian mitogen-activated protein kinase ( MAPK ) signaling cascades . Furthermore , this alteration in signaling is associated with attenuation of the host innate immune response to Shigella infection in the mouse lung model . These results are consistent with those of Aribe and colleagues who recently determined that OspF is a MAPK phosphatase [16] . Our study demonstrates how systems biological approaches using yeast can generate testable hypotheses regarding the roles of bacterial effector proteins in pathogenesis .
OspF is one of several Shigella effectors that inhibit growth when expressed in yeast ( Figure S1 and Protocol S1 ) . OspF also exhibits conserved localization to both the cytoplasm and nucleus when expressed de novo in yeast and mammalian cells ( Figure S2 ) . While OspF is toxic to yeast grown in synthetic liquid media , fusion of OspF to green fluorescent protein ( GFP ) partially relieves its toxicity , and growth in rich media almost completely abolishes its toxicity . We hypothesized that OspF activity is independent of the growth medium and that yeast compensatory mechanisms are merely enhanced in rich media . To identify genes that contribute to these mechanisms , we systematically screened the yeast haploid deletion strain collection for null alleles specifically impaired by expression of GFP-OspF in rich media ( Figure 1 , Protocols S2 and S3 ) . Each strain in the deletion collection contains a complete deletion of one nonessential open reading frame ( ORF ) . The deletion collection includes 4 , 773 strains covering 77% of the ∼6 , 200 annotated yeast ORFs . While most of these deletion strains grow as well as wild-type when cultured in rich media , growth of 98 of the deletion strains was severely and reproducibly impaired by expression of GFP-OspF but not of GFP alone ( Figure S3 ) . Fifteen of the deletion strains were excluded from further analyses . These included two strains deleted for dubious ORFs , three whose gene products are required for galactose metabolism ( our inducing condition ) , and ten whose gene products are ribosomal structural components . Thus , our result set contained 83 genes hypersensitive to OspF ( Table S1 ) . Our screen for null alleles hypersensitive to OspF is analogous to screens for synthetic lethal ( SL ) interactions . Rather than inferring functional relationships between genes by identifying pairs of mutant alleles that when combined result in inviability , we sought to infer functional relationships between yeast genes and OspF toxicity by identifying null alleles that exacerbate OspF toxicity . Null alleles that are hypersensitive to OspF expression effectively define SL interactions with OspF . No consensus exists , however , regarding proper interpretation of SL interactions in terms of pathways , and we faced a similar interpretive dilemma with the 83-gene result set of our hypersensitivity screen . Two interpretations motivated by the “circuit” concept of a pathway have currency . One explains SL interactions as resulting from cumulative insults to the same pathway ( a “serial” circuit ) that reduce its efficiency below some critical threshold . Alternatively , synthetic lethality may result from simultaneous insults to mutually compensating or “buffering” processes ( “parallel” circuits ) . Clearly , these interpretations depend on what constitutes a pathway , a concept which itself remains loosely defined . Nonetheless , these interpretations of SL interactions serve to delineate a range of possible explanations for hypersensitivity to OspF . Specifically , the result set of OspF-hypersensitive deletion strains might be enriched in proteins involved in cellular pathways directly insulted by OspF ( by analogy with the “serial” circuit interpretation ) . Alternatively , it might be enriched in pathways that mitigate the effects of OspF expression ( by analogy with the “parallel” circuit interpretation ) . In either case , it was essential to identify cellular processes represented in the result set . To identify pathways impaired among the 83 hypersensitive strains , we used the online data-mining tool FuncAssociate [17] to identify gene ontologies enriched among the deletion strains . Gene ontologies are semi-hierarchical classifications of genes based on the roles their proteins play in biological processes ( process ontologies ) , their cellular localizations ( component ontologies ) , and their biochemical activities ( molecular function ontologies ) . One of the major advantages of conducting genome-wide studies in yeast is the wealth of information available regarding individual yeast gene products . For example , over 75% of the yeast genome is functionally annotated to at least one type of gene ontology [18] . FuncAssociate revealed that 25 ontologies represented by the genes deleted in 79 hypersensitive strains were significantly overrepresented ( theoretical p-values < 0 . 001 ) ( Table S2 ) . We simplified these results by eliminating redundant and relatively broad ontologies such as “cellular process , ” “cytoplasm , ” and “cell organization and biogenesis” because these ontologies ( near the roots of the hierarchies ) are shared by hundreds or thousands of genes ( Table S2 ) . Additionally , in the six cases where two hierarchically related ontologies included the exact same set of genes in our results , the more general ( less descriptive ) ontology was discarded ( Table S2 ) . In order to visualize relationships among the remaining 16 ontologies , we performed two-dimensional clustering of the ontologies and the genes they subsume ( Figure 2 ) . Of the remaining 16 ontologies , nine were biological process ontologies . With respect to hierarchical relationships , these process ontologies fell into four groups rooted in “cell communication , ” “cell wall organization and biogenesis , ” “ubiquitin-dependent protein catabolism , ” and “cell division . ” For another perspective , we carried out a second type of enrichment analysis using existing SL interaction data . “Congruent” gene pairs have been defined as those sharing SL interaction partners [19] . Again by analogy , we defined a gene to be congruent to OspF expression if its set of SL interaction partners has statistically significant overlap with mutant alleles hypersensitive to OspF ( Figure S4 ) . Intuitively , if OspF is congruent to a gene , then expression of OspF effectively mimics mutation of that gene . We scanned the available database of 9 , 019 SL interactions encompassing 2 , 286 ORFs [18] for genes congruent to OspF . Almost all of the genes that are congruent to OspF encode proteins involved in either the cell wall integrity ( CWI ) pathway or chitin biosynthesis , both of which are related to cell wall biogenesis ( Table 1 ) [20] . The CWI pathway regulates cell wall biogenesis during the cell cycle as well as under conditions that perturb the cell wall . All of the proteins involved in chitin biosynthesis in Table 1 either directly encode or regulate the activity of chitin synthase III , a protein responsible for synthesis of the chitin ring at the bud neck during cell division and for chitin in the lateral cell wall . Although chitin normally accounts for only 1%–2% of the yeast cell wall , chitin can contribute up to 20% of the cell wall under times of stress [21] . Furthermore , with the exception of KRE1 , strains deleted for none of the genes congruent to OspF were hypersensitive to OspF , suggesting that these genes represented targeted rather than compensating pathways . Thus , the results of this second analysis raised the possibility that OspF acts to inhibit the CWI pathway and/or chitin biosynthesis . The implication of cell wall integrity in OspF toxicity by two independent analyses focused our attention on the yeast cell wall . Furthermore , the fact that the CWI pathway is a highly conserved MAPK signaling cascade made it a plausible target ( in yeast ) for an effector protein from a mammalian pathogen . Although this hypothesis resulted from data mining two different types of systematic data , functional annotation and genetic interaction data , it was based on one type of experimental data: identification of genes that modulate OspF toxicity . For a complementary systems view we also identified genes whose expression was modulated by OspF . We examined the transcriptional response of wild-type yeast to OspF expression using Affymetrix Yeast GeneChips . mRNA profiling studies conducted in triplicate identified only 19 genes regulated greater than 2-fold ( Table 2 ) 3 h after the induction of OspF expression . Strikingly , the second-most downregulated gene in all three independent mRNA profiling experiments encodes CWP1 , a major structural component of the yeast cell wall . Although loss of this protein does not result in measurable alterations in yeast growth under laboratory conditions , Δcwp1 strains are sensitive to agents that impair CWI , and this gene is normally upregulated in response to cell wall stress [22] . Furthermore , CWP1 is one of ∼20 proteins whose expression is regulated by the CWI pathway [23] . Activation of the CWI pathway results in phosphorylation of RLM1 , a transcription factor that induces expression of 19 genes including CWP1 and PRM5 and represses expression of five genes , including FIT2 [23] . The OspF-dependent reversed regulation of CWPI , PRM5 , and FIT2 suggested that OspF expression directly or indirectly inhibits RLM1-regulated transcription . Since we examined alterations in mRNA expression patterns among unsynchronized cells and the CWI pathway is only periodically induced during the cell cycle , it was perhaps not surprising that more significant alterations were not observed in additional RLM1-regulated genes . Indeed , when grown under normal laboratory conditions , activation of the CWI pathway is undetectable in asynchronous cells [22] . Nevertheless , although expression of the remaining RLM1-regulated genes was not altered greater than 2-fold , in general those genes normally induced by RLM1 were repressed in the presence of OspF , and those genes normally repressed were induced 3 h after the induction of expression of OspF ( Figure S5 ) . We next investigated whether OspF expression does indeed regulate expression of the CWI pathway by monitoring the effects of GFP-OspF expression on an RLM1 transcriptional reporter . This well-characterized reporter contains two RLM1 binding sites fused to the minimal CYC1 promoter driving lacZ [24] . Expression of this reporter gene is detectable in asynchronous yeast grown under standard laboratory conditions and responds appropriately to perturbations that activate the CWI pathway [3 , 10 , 25] . We observed that GFP-OspF inhibits basal levels of expression of this CWI pathway reporter and inhibits activation of the CWI pathway in response to heat shock ( Figure 3A ) . Thus , expression of OspF appears to inhibit activation of the CWI pathway . Many of the components of the yeast CWI pathway are highly conserved among eukaryotes ( Figure 3B ) . In fact , many bacterial effector proteins , including Shigella IpgB2 and Yersinia YopE , have been demonstrated to inhibit specific steps in this pathway in both yeast and mammalian cells [3 , 10 , 25] . IpgB2 is a Rho mimic that activates RHO signaling while YopE is a RhoGAP . In order to determine where OspF targets the CWI pathway in relationship to Rho1 , we first screened for alterations in the integrity or polarity of the yeast cytoskeleton in response to expression of OspF . No perturbations were observed ( unpublished data ) suggesting that OspF targets a component of the CWI signaling pathway downstream of Rho1 . To further localize the action of OspF , we took advantage of the cross-reactivity of the mammalian phospho-specific p42/44 MAPK antibodies and phosphorylated Slt2 . Since CWI pathway activation is barely detectable in wild-type yeast grown under standard laboratory conditions , we assayed for the ability of OspF to block activation of the MAPK pathway by either heat or hypoosmotic shock . OspF inhibited phosphorylation of SLT2 in response to both of these conditions , but did not alter total SLT2 levels ( Figure 3C , unpublished data ) . Thus , OspF targets a protein upstream of RLM1 and downstream of RHO1 in the CWI pathway . Several lines of evidence suggested that OspF might target additional cellular processes . For example , the growth inhibition of wild-type yeast due to expression of OspF cannot be explained by inhibition of just the CWI pathway since deletion strains that no longer express essential components of the CWI pathway are not inhibited for growth under the same conditions . Yeast encode five additional MAPK signaling cascades including the mating pathway ( Fus3 ) , the invasive growth pathway ( Kss1 ) , the hyperosmotic growth ( HOG ) pathway , the sporulation pathway ( Smk1 ) , and Mlp1 , a second MAPK implicated in the CWI pathway [26] . A strain deleted in all six MAPKs has been reported to be impaired for growth [27] . The severe growth inhibition of yeast observed in liquid media could be accounted for by OspF inhibiting multiple yeast MAPK signaling pathways . Indeed , we observed that OspF expression also impaired phosphorylation of three additional yeast MAPKs: Hog1 , Kss1 , and Fus3 ( Figure 3C ) . Thus , OspF appears to nonspecifically inhibit phosphorylation of yeast MAPKs . This general inhibition of yeast MAPK signaling pathways likely explains the OspF hypersensitivity of deletion strains not directly related to cell wall biogenesis . Because MAPK cascades are highly conserved among all eukaryotes , and OspF is a virulence protein from a human pathogen , it seemed likely that OspF would likewise target mammalian MAPK pathways . To test whether the presence of OspF altered MAPK signaling during infection , we compared MAPK phosphorylation patterns in mammalian cells infected with wild-type or ΔospF Shigella . As previously observed , infection with wild-type Shigella resulted in strong SAP/JNK phosphorylation [28] , weak ERK phosphorylation [29] , and no detectable p38 phosphorylation ( Figure 4A ) . In contrast , infection with ΔospF Shigella resulted in robust phosphorylation of all three MAP kinases , without alteration of the overall levels of MAP kinases in the cells . The ΔospF Shigella phenotype was complemented by expression of OspF from its endogenous promoter on a low copy-number plasmid ( Figure 4A ) . These observations suggested that the presence of OspF either inhibits ERK and p38 phosphorylation or the absence of OspF stimulates their phosphorylation . The observations in yeast suggested that the presence of OspF is sufficient to inhibit MAPK phosphorylation . Several experiments were conducted to test whether this is the case in mammalian cells . First , we compared the ability of wild-type and ΔospF Shigella to inhibit MAPK phosphorylation in response to the addition of exogenous agents that induce MAPK signaling . Wild-type , but not ΔospF Shigella , inhibit activation of ERK phosphorylation in response to epidermal growth factor and p38 phosphorylation in response to sorbitol ( Figure 4B ) . Second , we coinfected mammalian cells with wild-type and ΔospF Shigella at a ratio of 1:1 . The ΔospF Shigella strain demonstrated no defect in invasion alone or in competition with wild-type Shigella ( unpublished data ) . The presence of OspF in the mixed infection was sufficient to inhibit MAPK signaling pathways ( Figure 4C ) . Thus , OspF inhibits MAPK phosphorylation in both yeast and mammalian cells . While this inhibition is nonspecific in yeast , the signaling of the mammalian SAP/JNK pathway appeared to be unaffected by the presence of OspF . Inhibition of MAPK phosphorylation is a strategy used by other bacterial pathogens , including Yersinia species and Bacillus anthracis , to modulate the innate immune response . The Yersinia type III effector YopJ is an acetyltransferase that inhibits MAPKK phosphorylation by direct modification of residues that are normally phosphorylated [30] . B . anthracis lethal factor , a metalloprotease , cleaves the amino-termini of MAPKKs [31] . To determine whether Shigella also inhibits MAPK phosphorylation by modulation of MAPKKs , we monitored the phosphorylation state of the MAPKKs directly upstream of ERK and p38 . In both cases , infection with wild-type , but not ΔospF Shigella , resulted in the accumulation of phosphorylated MAPKKs ( Figure 5 ) to levels markedly greater than those observed in the absence of OspF . The presence of full-length phosphorylated MAPKKs in the absence of phosphorylated MAPKs suggests that OspF inhibits MAPK phosphorylation by a novel mechanism , by either blocking phosphorylation of MAPKs by activated MAPKKs or by dephosphorylating activated MAPKs . These results are consistent with recent observations that OspF is a MAPK phosphatase for ERK and p38 [16] . Mammalian MAPK signaling pathways regulate diverse cellular activities including cell proliferation , differentiation , motility , survival , and innate immunity . The innate immune response is activated when host cells recognize pathogen-associated molecular patterns that include microbial products like peptidoglycan and lipopolysaccharide . The pathogen-associated molecular patterns are recognized by pathogen recognition receptors , like the extracellular Toll-like receptors and the intracellular nucleotide-binding oligomerization domain proteins ( for review see [32] ) . For example , after Shigella invade host cells , mammalian NOD1 binds the bacterial peptidoglycan , resulting in activation of the SAP/JNK and ERK MAPK signaling pathways as well as NF-κB activation [28] . Many pathogens , including Shigella , modulate this host immune response presumably to promote their own survival . Specifically , the Shigella effectors OspG and IpaH9 . 8 downregulate the innate immune response by inhibition of NF-κB activation [33 , 34] . Therefore , we hypothesized that OspF inhibition of MAPK signaling might also serve to downregulate the mammalian immune response . We next investigated whether downregulation of MAPK signaling pathways by Shigella in cell culture reflected in vivo alterations in the innate immune response to infection . Although Shigella infections are normally restricted to the intestines in humans , the bacterium is unable to sustain an infection in the intestines of adult mice . However , the mouse lung infection is an established model for monitoring the immune response to Shigella infections [35] . After determining that ΔospF Shigella were not defective in invasion of host cells or in cell-to-cell spread ( unpublished data ) , we investigated the innate immune response to Shigella in the mouse lung infection model . As expected , 24 h after infection the lungs of mice infected with wild-type Shigella showed an inflammatory infiltrate dominated by polymorphic neutrophils ( Figure 6 ) . In contrast , infection with ΔospF Shigella triggered a markedly more aggressive immune response as manifested by the increase in polymorphic neutrophils in the lungs . Furthermore , the visible edema and hemorrhage resulted in considerable destruction of the lung architecture ( Figure 6 ) . These findings suggest that the presence of OspF attenuates the host innate immune response .
Although unicellular eukaryotes such as yeast cannot serve as models for the bacterial infection of host cells , the utility of S . cerevisiae for investigating the molecular activities of effector proteins is well established . Until now , little has been done to exploit the genetic tractability of yeast to determine cellular targets of effector proteins . In this study , we present the first use of yeast systems biology to identify a function for a poorly understood bacterial effector protein . By integrating complementary systems-level snapshots of the cellular perturbation caused by OspF , we were able to determine that OspF targets a well-characterized yeast-signaling pathway . These observations led us to hypothesize and subsequently demonstrate that OspF inhibits highly conserved MAPK signaling pathways including those that regulate cell wall biogenesis in yeast and the host innate immune response in mammals . These results are in agreement with the recent demonstration that OspF is a MAPK phosphatase specific for ERK and p38 [16] . Although , there is also evidence that OspF can activate ERK phosphorylation in polarized epithelial cells [36] . One of the hallmarks of Shigella infections is the dramatic inflammatory response characterized by the recruitment of polymorphic neutrophils to sites of infection . This response facilitates access of Shigella to the basolateral surface of epithelial cells where this intracellular pathogen mediates its own uptake into cells via a type III secretion system . Shigella trigger activation of the innate immune response , at least in part , through the action of pathogen-associated molecular patterns like LPS [37] and peptidoglycan [28] . Shigella also down regulate the innate immune response by the action of several proteins including OspF . The Shigella effector proteins , OspG and IpaH9 . 8 , inhibit NF-κB activity resulting in decreased production of proinflammatory cytokines [33 , 34] . In addition , the ShiA protein down regulates the innate T-cell response [38] . OspF-mediated downregulation of the innate immune response by inhibition of MAPK signaling complements the actions of these other Shigella proteins . These proteins presumably allow Shigella to fine-tune the host innate immune response over the course of infection . OspF homologs are found in pathogens of both plants and animals including Salmonella typhimurium ( SpvC ) and Pseudomonas syringae ( HopAI1 ) ( Figure S6 ) . HopAI1 was recently demonstrated to inhibit activation of the plant innate immune response by an unknown mechanism [39] . These observations suggest the existence of a new class of bacterial proteins , common to pathogens of both plants and animals that modulate the host innate immune response by inhibiting MAPK phosphorylation . Our study demonstrates how genome-wide yeast screens can help generate testable hypotheses about the roles of bacterial effector proteins in pathogenesis . We applied a standard technique of statistical data mining , enrichment analysis , to integrate our screen results with two of the many types of systems data available for S . cerevisiae , genetic interactions and gene ontology annotations . This implicated CWI in OspF toxicity . The juxtaposition of this observation with the apparent reverse regulation of the CWI pathway in microarray experiments provided the crucial insight . Our approach was necessarily heuristic . For example , the set of genes exhibiting differential expression in response to OspF was of a size that one might dismiss as uninformative if it were the only set of data available , but the implication of CWI involvement by the hypersensitivity screens made the presence of any CWI pathway–regulated genes among those differentially regulated conspicuous . Considerable evidence also motivated our focus on the cell wall . For example , at least five OspF hypersensitive deletion strains not accounted for in the enrichment analyses are impaired in protein mannosylation and glycosylphosphatidyl-inositol anchor biosynthesis . Both of these post-translational modifications are abundant among yeast cell wall proteins [40] . The corresponding ontologies were not statistically significant in our analyses , but their biological significance to cell wall biogenesis is well established . Similarly , six of the eight genes annotated to “ubiquitin-dependent protein catabolism” ( italicized in Figure 2 ) are involved in regulation of an alternative cell wall biogenesis pathway which only becomes essential when the normal cell wall biogenesis pathway is perturbed [41 , 42] . These and other observations highlight the limitations of statistical methods and the importance of complementing formal analysis with an in-depth literature review to accurately interpret screen results . The demonstrated inhibition of the CWI pathway by OspF accounts for much , but not all , of the observed hypersensitivity in deletion strains . The CWI pathway is constitutively active in at least three of the OspF hypersensitive cell wall biogenesis strains ( Δgas1 , Δsmi1 , and Δfks1 ) [22] . Presumably this activation is essential for their survival since these deletion strains are synthetically lethal in combination with mutations that impair the CWI pathway ( Δslt2 and Δbck1 ) [11] . However , other pathways and processes were also implicated to which no clear CWI connection exists . We assume that the multiplicity of process ontologies enriched in the hypersensitive strains reflects pathway dependencies centered on the cellular perturbation caused by OspF . For example , if a given pathway is required to mitigate toxicity of OspF , then any cellular process on which that pathway depends , will probably also be intolerant of impairment by gene deletion , though perhaps to a lesser degree . From this perspective , it is possible that OspF inhibition of cell wall biogenesis via the CWI pathway accounts for the hypersensitivity of cell division-associated null alleles since successful cell division depends strongly on CWI . Because yeast are under high osmotic pressure , budding and cytokinesis must be tightly coordinated with cell wall growth and remodeling to maintain CWI throughout the process . Defects in the cell wall can result in failure of cell division due to incomplete cytokinesis or cell lysis [43] . The congruence of chitin biosynthesis genes to OspF expression supports this speculation since chitin is a cell wall constituent involved in cell division [40] . However , concrete evidence explaining hypersensitivity of the cell division-associated null alleles in terms of OspF inhibition of the CWI pathway is lacking . The hypersensitivity of these null alleles may also be related to OspF inhibition of multiple MAPK cascades . The unanimity of our analyses in implicating CWI is likely due to the fact that this is the only MAPK cascade with demonstrated basal activity in laboratory conditions . Our study demonstrates how genome-wide yeast screens can help generate testable hypotheses about the roles of bacterial effector proteins in pathogenesis . The multiple perspectives gained from two genome-wide experiments ( as well as from multiple analyses of individual experiments ) allowed us to effectively “triangulate” the process being perturbed by OspF , as well as to identify the nature of the perturbation . Other choices of screens , e . g . , genome-wide identification of OspF binding targets using protein arrays or coimmunoprecipitation assays or identification of toxicity-suppressing conditions in synthetic liquid media ( in which OspF is more toxic to yeast ) , would likely have resulted in a different path to the same discovery . For example , a preliminary genome-wide screen for suppressors of OspF toxicity in liquid media suggested that Δsac7 suppresses OspF toxicity . SAC7 is a RhoGAP whose absence should lead to increased activation of Rho1 , so it makes sense , given what we establish in this paper , that it would suppress OspF toxicity . Genome-wide screens are potentially applicable to any bacterial effector that exhibits evidence of conserved activity in yeast such as toxicity or conserved localization . Hypersensitivity screens only make sense for effectors that exhibit mild or conditional toxicity , but other kinds of screens can be applied to extremely toxic effectors . For example , Alto and colleagues recently demonstrated how genome-wide suppressor screens of the yeast deletion strain collection can identify the cellular targets of effector proteins whose expression severely inhibits yeast growth [10] . They confirmed that Shigella IpgB2 mimics activated Rho1 in yeast when they observed that three deletion strains , all downstream components of the MAPK signaling cascade regulated by Rho1 , were resistant to IpgB2 toxicity . In this case , since expression of IpgB2 activates a nonessential signaling pathway , they were able to isolate suppressors that blocked persistent activation of the pathway . However , if effector proteins are toxic because they directly target an essential cellular process , it may not be possible to isolate suppressors by screening yeast strains that contain null alleles of nonessential genes . Alternatively , suppressors might be found by screening for yeast genes whose overexpression suppresses toxicity . A genome-wide library of such constructs is now available [44] . In summary , this study exemplifies how contemporary analytic tools and the simplest of eukaryotes , S . cerevisiae , can contribute to the study of bacterial effectors . Even when bacterial proteins target cellular processes like innate immunity , which are not conserved among all eukaryotes , if elements of these processes like MAPK signaling cascades are conserved , then screens in yeast can be helpful in elucidating a function for the effector proteins . This systems approach should be generally applicable to numerous microbial virulence proteins . Moreover , since this approach only requires bacterial DNA , it should be particularly valuable for pathogens difficult to genetically manipulate or dangerous to culture .
The gene encoding OspF was PCR-amplified from 2457T S . flexneri serotype 2a and subcloned into pFUS-GFP and pFUS-HIII1 [3] . Low-copy number versions of the GFP-OspF and GFP expression plasmids were constructed by homologous recombination in yeast to create pRS316-GAL10-GFP-OspF , pRS313-GAL10-GFP-OspF , pRS316-GAL10-GFP , and pRS313-GAL10-GFP . The nat1 gene , which confers resistance to nourseothricin , was cloned into the pPR316-based plasmids to create pRS316-GAL10-GFP-OspF-CN and pRS316-GAL10-GFP-CN . Genes encoding S . typhimurium SpvC and P . syringae HopAI1 were PCR-amplified from genomic DNA preparations ( E . Hohman and W . Songnuan , Massachusetts General Hospital ) and transferred into pBY011-D123 ( B . Bhullar , Harvard Institute of Proteomics ) , a yeast expression plasmid using Gateway technology ( Invitrogen , http://www . invitrogen . com ) . All infections were performed with S . flexneri 2457T serotype 2a . ΔospF Shigella was constructed using the λred recombinase-mediated recombination system [45] . A PCR fragment encompassing ospF plus 90 bp upstream and 300 bp downstream was cloned in pAM238 ( pSC101 ori ) to create pOspF , a complementing plasmid . All yeast transfers in 96- or 384-well arrays were conducted using a Biorobot 3000 robot ( Qiagen , http://www1 . qiagen . com ) outfitted with one of three floating pin tools described below ( V&P Scientific , Incorporated , http://www . vp-scientific . com ) . The entire yeast MATα haploid deletion strain collection ( 53 × 96-well plates/set ) ( Open Biosystems , http://www . openbiosystems . com ) was transformed twice with pRS316-GAL10-GFP-OspF-CN and twice with pRS316-GAL10-GFP-CN using our 96-well transformation protocol ( Protocol S2 ) . Transformants were first selected under noninducing conditions ( SC-URA 2% glucose ) on solid media . Each transformation was plated in quadruplicate to create a 384-well array . Colonies were transferred to a second noninducing plate to “normalize” the yeast in each spot , since each individual yeast transformation varied in efficiency . The yeast from this plate were transferred to rich- ( YEP ) inducing ( 2% galactose ) and noninducing ( 2% glucose ) media . In the case of inducing media , clonNAT ( 70 μg/ml ) was added to the media to ensure maintenance of the transformed plasmid . The plates were incubated at 30 °C for ∼48 h and then quantitated . Since the transformation efficiency was relatively poor , we assumed that each of the four transformation spots generated from each transformed strain was composed of independent transformations . Thus , each transformation resulted in four independent biological replicates . Each screen was repeated twice , thus we screened up to eight independent biological samples of each strain in the collection ( Protocol S3 , Figure S7 , Tables S3 and S4 ) . We extracted the SL interactions from the exhaustive table of protein interaction data from Saccharomyces Genome Database ( obtained by File Transfer Protocol , downloaded on 9 September 2006 ) . We filtered this data to include only those SL interactions qualified as “inviable . ” The bulk of this data comes from several large-scale screens , but over 1 , 000 small-scale screens also contribute to the total of 9 , 019 unique interactions between 2 , 286 genes . Only 1 , 041 of these interactions have been experimentally confirmed symmetric . For the purpose of our analysis we assumed all listed interactions were symmetric . Of the 83 null alleles hypersensitive to OspF , 58 were included among the 2 , 286 genes in the SL database . To assess the significance of overlap between these 58 and the prey sets of each of the 2 , 286 genes , we calculated the chance of each intersection occurring in randomly selected samples as follows: Let W be the set of all genes in the interaction database . Let H = { x ∈ W: null alleles of x are hypersensitive to OspF } . That is , H is the OspF “hit” set . For each gene g in W let P ( g ) = { x ∈ W/g : x is synthetically lethal with g } . P is the “prey” set of g . Then the probability of H∩P containing k or more genes is given by the hypergeometric distribution: where where C ( a , b ) is the binomial coefficient function . To minimize the false discovery rate we used an alpha value of 0 . 05/2286 ≈ 2 . 187 e-5 ( the Bonferroni correction ) . Three independent cultures of wild-type yeast , transformed with either a plasmid that conditionally expresses OspF ( without the GFP fusion ) from a low-copy number plasmid ( pBY011-D123-OspF ) or an empty vector control plasmid ( RS316 ) , were grown overnight in selective media supplemented with 2% raffinose ( raf ) . In the morning , each culture was diluted to an OD600 = 1 . 0 in fresh SC-URA 2% raf and incubated for 2 h at 30 °C . Galactose was then added to a final concentration of 2% to all cultures ( to induce expression of OspF ) . After 3 h , the yeast were pelleted and snap-frozen in liquid N2 . Total RNA was isolated using hot phenol followed by ethanol precipitation and then submitted to the Harvard Medical School Partners Healthcare Center for Genetics and Genomics ( HPCGG ) for further processing . HPCGG synthesize cDNA using the GeneChip Expression 3′-Amplification Reagents One-Cycle cDNA Synthesis Kit . They then preformed in vitro transcription ( IVT ) using the Affymetrix GeneChip Expression Amplification Reagents Kit and quantified the IVT samples with a Bio-Tek UV plate reader . Hybridization was carried out with Affymetrix yeast S98 chips according to the manual . Microarrays were scanned with a GeneChip 3000 7G Scanner controlled by the Affymetrix GCOS v1 . 3 operating system . We subsequently processed the raw CEL file output using the “affy” package [46] for the statistical program R [47] . Expression measurements were background corrected and normalized using the “rma” method . For each gene , an expression ratio was calculated using the ( arithmetic ) mean expression of the gene in each triplicate set ( control and inducing ) . Yeast transformed with GFP-OspF ( BYO11-D123-OspF ) or GFP ( pRS313-GAL10-GFP ) plus a RLM1-regulated LacZ reporter plasmid ( p1434 ) [24] were grown overnight in selective media supplemented with 2% raf at 30 °C . In the morning , cultures were back diluted to OD600 = 1 . 0 , grown at ∼23 °C , and allowed to recover for 2 h before the addition of 2% galactose . Half of the cultures were then transferred to 39 °C for 1 h , the other half kept at ∼23 °C . β-galactosidase assays were conducted as previously described [48] . Yeast transformed with pRS316-GAL10-GFP-OspF or pRS316 were grown overnight in selective media supplemented with 2% raf ( plus 1M sorbitol for hypoosmotic shock ) . In the morning , cultures were diluted ( OD600 = 1 . 0 ) in fresh media and incubated for 2 h at 30 °C ( 25 °C for the heat-shock experiment ) . OspF was induced by addition of 2% galactose . Cells were incubated for 2 h at 30 °C ( or 25 °C ) before introducing the stresses . Heat shock was performed as previously described [49] . Cells were shocked by dilution 1:1 with media pre-warmed to 55 °C followed by incubation at 39 °C for 30 min . The shock response was terminated by an additional 1:1 dilution with ice-cold stop mix . The mating and invasive growth MAPK pathways were induced by the addition of 200 nM α-factor for 15 min . The HOG pathway was induced by the addition of 400 mM NaCl for 5 min . Hypoosmotic shock was performed by diluting cells 1:10 in water . In all cases , yeast were pelleted and snap-frozen at the completion of the shock procedure . Protein was isolated from yeast and subjected to SDS-PAGE . Gels were blotted to nitrocellulose and probed with the indicated antibodies according to the manufacturer's directions . The phospho-p42/44 antibody recognizes yeast phosphorylated SLT2 , FUS3 , and KSS1 . The phospho-p38 antibody recognizes yeast phosphorylated HOG1 . The PSTAIRE antibody was purchased from Santa Cruz Biotechnology , Santa Cruz , California , United States . HeLa cells seeded in 6-well plates ( 2 ×105 cells/well ) were serum-starved overnight . HeLa cells were infected with Shigella in mid-exponential growth phase at an MOI of 10:1 . Once the bacteria were added , the plates were spun at 1 , 000 rpm for 5 min at RT . Each of the Shigella strains used carries the plasmid pIL22 that constitutively expresses the afimbrial adhesin from uropathogenic E . coli to synchronize infections [50] . The 6-well plates were subsequently incubated at 37 °C for 1 h . Cells were washed with ice-cold PBS plus 1 mM Na3VO4 and 10 mM NaF and then lysed with 300 μl RIPA buffer plus protease inhibitors . Equal volumes of samples were subjected to SDS-PAGE . Gels were blotted to nitrocellulose and probed with the indicated antibodies according to the manufacturer's recommendations ( Cell Signaling , http://www . cellsignal . com ) . C57BL/6 mice , aged 6–8 wk , were obtained from Jackson ImmunoResearch Laboratories and housed in specific pathogen-free animal facilities . The experimental procedures used in this study were approved by IACUC committee at HMS . For infection , mice were anesthetized by intramuscular injection of ketamine ( 12 mg/mL; Webster Veterinary Supply , Incorporated ) and xylazine ( 4 mg/mL; Webster Veterinary Supply , Incorporated ) in phosphate-buffered saline ( PBS ) . The inoculum used for each bacterial strain was ∼ 5 ×107 cfu re-suspended in PBS . Mice were inoculated intranasally in a single application of 20 μL with all mice receiving the same inoculum as determined by OD measurement and dilution plating of the inoculum . At the each time point following infection , mice were sacrificed and the lungs removed . Lungs were then fixed in 10% neutral buffered formalin and embedded in paraffin for hematoxylin and eosin staining . Images were obtained with 10× objective . | Many bacterial pathogens use specialized secretion systems to deliver effector proteins directly into host cells . The effector proteins mediate the subversion or inhibition of host cell processes to promote survival of the pathogens . Although these proteins are critical elements of pathogenesis , relatively few are well characterized . They often lack significant homology to proteins of known function , and they present special challenges , biological and practical , to study in vivo . For example , their functions often appear to be redundant or synergistic , and the organisms that produce them can be dangerous or difficult to culture , requiring special facilities . The yeast Saccharomyces cerevisiae has recently emerged as a model system to both identify and functionally characterize effector proteins . This work describes how genome-wide phenotypic screens and mRNA profiling of yeast expressing the Shigella effector OspF led to the discovery that OspF inhibits mitogen-activated protein kinase signaling in both yeast and mammalian cells . This inhibition of mitogen-activated protein kinase signaling is associated with attenuation of the host innate immune response . This study demonstrates how yeast functional genomic studies can contribute to the understanding of pathogenic effector proteins . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"biotechnology",
"infectious",
"diseases",
"cell",
"biology",
"mammals",
"microbiology",
"computational",
"biology",
"mus",
"(mouse)",
"molecular",
"biology",
"genetics",
"and",
"genomics",
"eubacteria",
"saccharomyces"
] | 2007 | Yeast Functional Genomic Screens Lead to Identification of a Role for a Bacterial Effector in Innate Immunity Regulation |
Non-Hodgkin lymphoma ( NHL ) represents a diverse group of hematological malignancies , of which follicular lymphoma ( FL ) is a prevalent subtype . A previous genome-wide association study has established a marker , rs10484561 in the human leukocyte antigen ( HLA ) class II region on 6p21 . 32 associated with increased FL risk . Here , in a three-stage genome-wide association study , starting with a genome-wide scan of 379 FL cases and 791 controls followed by validation in 1 , 049 cases and 5 , 790 controls , we identified a second independent FL–associated locus on 6p21 . 32 , rs2647012 ( ORcombined = 0 . 64 , Pcombined = 2×10−21 ) located 962 bp away from rs10484561 ( r2<0 . 1 in controls ) . After mutual adjustment , the associations at the two SNPs remained genome-wide significant ( rs2647012:ORadjusted = 0 . 70 , Padjusted = 4×10−12; rs10484561:ORadjusted = 1 . 64 , Padjusted = 5×10−15 ) . Haplotype and coalescence analyses indicated that rs2647012 arose on an evolutionarily distinct haplotype from that of rs10484561 and tags a novel allele with an opposite ( protective ) effect on FL risk . Moreover , in a follow-up analysis of the top 6 FL–associated SNPs in 4 , 449 cases of other NHL subtypes , rs10484561 was associated with risk of diffuse large B-cell lymphoma ( ORcombined = 1 . 36 , Pcombined = 1 . 4×10−7 ) . Our results reveal the presence of allelic heterogeneity within the HLA class II region influencing FL susceptibility and indicate a possible shared genetic etiology with diffuse large B-cell lymphoma . These findings suggest that the HLA class II region plays a complex yet important role in NHL .
Non-Hodgkin lymphoma ( NHL ) represents a diverse group of B- and T-cell malignancies of lymphatic origin . The most common subtypes are of B-cell origin and are further classified on the basis of their resemblance to normal stages of B-cell differentiation [1] . Epidemiological studies indicate that these may have different environmental and genetic risk factors , although some etiological factors may also be shared [2] . Familial studies provide substantial evidence for a genetic influence on susceptibility to the major mature B-cell neoplasms , including diffuse large B-cell lymphoma ( DLBCL ) , follicular lymphoma ( FL ) and chronic lymphocytic leukemia/small lymphocytic lymphoma ( CLL/SLL ) [3] , [4] . Recent genome-wide association studies ( GWAS ) of the FL subtype of NHL identified associations with two variants within the human leukocyte antigen ( HLA ) region , one at 6p21 . 33 ( rs6457327 ) [5] and the other at 6p21 . 32 ( rs10484561 ) [6] . Additional true associations , particularly in the HLA region , may have been missed because a limited number of samples were used in the initial genome-wide screens , and the selection of a few top single nucleotide polymorphisms ( SNPs ) for validation is further subject to chance . In this study , we conducted a larger independent genome-wide scan of FL using 379 cases and 791 controls from the Scandinavian Lymphoma Etiology ( SCALE ) study of Sweden and Denmark , which was used in the validation of the previous GWAS [6] . This scan was followed by two stages of validation in European-ancestry cases of FL and other common B-cell NHL subtypes and controls from the US , Canada and Australia ( Table 1 , Table S1 , Table S2 , Figure 1 ) .
In total , 298 , 168 SNPs were analyzed in Stage 1 ( λ = 1 . 028; λ1000 = 1 . 055 [7] ) , in which we observed suggestive associations ( adjusted trend P-value<10−5 ) at 4q32 . 3 , 6p21 . 32 and 10q25 . 3 ( Table S3 ) with the strongest at rs2647012 ( odds ratio ( OR ) = 0 . 58 , PPCAadjusted = 1 . 59x10−7 ) within the HLA class II region on 6p21 . 32 . Sixteen SNPs in close proximity to the HLA-DQ genes showed association with adjusted P-values<10−4 , including the previously reported rs10484561 ( Figure 2 , Table S4 ) [6] . The previously reported HLA class I associated SNP rs6457327 [5] was modestly associated with FL risk ( OR = 0 . 82 , P = 0 . 03 ) in Stage 1 , and was not in linkage disequilibrium ( LD; r2 = 0 ) with any of the top 100 SNPs . In Stage 2 , we carried out an in silico validation of the top 40 SNPs from Stage 1 ( Table S5 ) in 213 FL cases and 750 controls from the San Francisco Bay Area , USA ( Table 1 ) , the study that reported an association at 6p21 . 32 [6] . Among 38 out of 40 SNPs , seven showed association ( P<0 . 05 ) in Stage 2 ( Table S5 ) , six of which were located within the 6p21 . 32 region . We tested the independence of multiple association signals in 6p21 . 32 using a stepwise logistic regression analysis ( entering SNPs based on a criterion of likelihood ratio test p-value<0 . 05 ) and found that with rs2647012 ( the top SNP within the region ) forced in the model , only the addition of rs10484561 contributed significantly to the association with increased risk of FL . The OR for this SNP , adjusted for rs2647012 , was 1 . 43 , P = 0 . 006 ( Table S6 ) . After excluding previously identified and non-independent association signals , we selected rs2647012 , and an additional four top SNPs to be taken forward to a third stage ( Table S7 , S8 ) , wherein these were genotyped in 836 FL cases and 3202 controls from the Mayo Clinic ( US ) [8] , National Cancer Institute-Surveillance , Epidemiology and End Results ( NCI-SEER , US ) [9] , Yale University ( US ) [10] , New South Wales ( NSW , Australia ) [11] and British Columbia ( BC , Canada ) [12] studies . The association of rs2647012 with FL was validated , showing consistent associations with similar ORs ( no heterogeneity , P = 0 . 32 ) across all independent studies and reaching genome-wide significance in both the combined analysis of the validation samples ( P = 3×10−15 ) and the combined analysis of all three stages ( 1428 FL cases , 4743 controls; OR = 0 . 64 , P = 2×10−21 ) ( Table . 2 , Figure 3 ) . After adjustment for rs10484561 , the association at rs2647012 remained genome-wide significant with minimal change in magnitude ( ORadjusted = 0 . 70 , Padjusted = 4×10−12 ) . The LD between the two SNPs is low ( r2<0 . 1 in the SCALE controls and HapMap CEU [Utah residents with northern and western European ancestry] samples release27 ) . Taken together , our results suggest that the association at rs2647012 is independent from rs10484561 , and tags a different disease-predisposing variant . We also found suggestive evidence for an association at rs6536942 on 4q32 . 3 ( OR = 1 . 36 , P = 2×10−5 ) ( Table 2 , Figure S1A ) . To fine-map the association signals in the HLA class II region , we imputed 10 , 639 SNPs within 600 kb surrounding the top SNP rs2647012 using data from the 1000 Genomes ( 1000G , 60 CEU subjects , August 2009 ) and HapMap projects ( HapMapII release 22 , CEU ) in Stage 1 . Among the imputed SNPs , 258 SNPs located in a strong LD block of 236 kb ( r2>0 . 8 ) showed stronger evidence of association than all the genotyped SNPs within the region ( Figure S2 ) . Since a moderate discordance of reference genotypes was observed between 1000 G and HapMapII , we analyzed only SNPs showing a concordance of >95% in the two datasets and identified the strongest association at rs9378212 ( OR = 1 . 66 , P = 3 . 21×10−8 ) , located 219 kb upstream of rs2647012 ( r2 = 0 . 56 in controls ) . We subsequently confirmed the imputed genotypes by Taqman genotyping in 345 of the FL case subjects used in Stage 1 and found a 99 . 4% concordance with the imputed genotypes , demonstrating high confidence in the results of the imputation . Next , we performed a haplotype analysis using rs2647012 , rs10484561 and an additional 12 adjacent genotyped SNPs located within a block of minimal recombination . Out of the eight haplotypes identified , three were neutral ( OR = 0 . 9–1 . 1 ) , three increased risk ( ORs>1 . 2; strongest risk haplotype tagged by rs10484561 ) and two were protective ( OR≤0 . 8; both tagged by rs2647012 ) ( Table S9 ) , suggesting the presence of at least two susceptibility alleles within the region . Coalescence analysis of the eight haplotypes indicated that rs2647012 and rs10484561 arose on two distal branches of the ancestral recombination graph [13] ( Figure S3 ) , which was also supported by the analysis of median-joining network [14] using seven SNPs without any recombination ( Figure 4 ) . Further haplotype analysis of the seven genotyped SNPs ( Table S9 ) and the imputed SNP rs9378212 indicated that the two alleles of rs9378212 tag the two different evolutionary lineages ( Figure 4 ) , each harboring either rs2647012 or rs10484561 . Thus , the associations at the two SNPs are likely due to two distinct susceptibility variants , instead of a single risk allele , that arose independently on different haplotype backgrounds . The FL-associated SNP , rs10484561 , was previously found to tag the extended haplotype HLA-DQA1*0101-HLA-DQB1*0501-HLA-DRB1*0101 [6] . Here , to test whether any HLA class II alleles may also be responsible for the observed association at rs2647012 , we imputed known HLA tag SNPs [15] , [16] using data from the 1000G and HapMapII European datasets . We confirmed the association of the HLA-DRB1*0101-HLA-DQA1*0101-HLA-DQB1*0501 extended haplotype , tagged by rs10484561 . The association at rs2647012 remained significant after adjustment for these three HLA alleles ( OR = 0 . 64 , P = 8 . 11×10−6 ) , suggesting that these are not driving the association at rs2647012 . Furthermore , rs2647012 was not in strong LD ( r2<0 . 8 in HapMap CEU or SCALE controls ) with any other known HLA tags [15] , including those tagging FL-associated alleles previously reported [17] , [18] ( r2<0 . 39 with the six HLA-DRB1*13 tag SNPs [rs2395173 , rs2157051 , rs4434496 , rs6901541 , rs424232 , rs2050191] [17] and r2<0 . 25 with the three HLA-B*0801 and HLA-DRB*0301 tag SNPs [rs6457374 , rs2844535 , rs2040410] [15] ) . Of the other 17 HLA class II alleles ( ∼39% of all the class II alleles ) that could be imputed , none showed significant association or were found to be responsible for the association at rs2647012 ( Table S10 ) . Detailed HLA allelotyping on large numbers of cases and controls is needed to determine if particular HLA class II alleles are responsible for the observed association at rs2647012 . To assess whether the FL-associated SNPs may be involved in the development of other NHL subtypes , we genotyped the five SNPs selected for Stage 3 together with rs10484561 in a total of 1592 DLBCL , 1075 CLL/SLL , 336 marginal zone lymphoma ( MZL ) , 262 mantle cell lymphoma , 306 T-cell lymphoma and 878 rare or unspecified NHL cases and 5220 controls from the SCALE2 , SF2 , BC , Mayo , NCI-SEER , Yale and NSW studies ( Table 1 , Table S1 , Figure 1 ) . Among these SNPs , rs10484561 showed evidence of association with DLBCL ( OR = 1 . 36 , P = 1 . 41×10−7 ) ( Figure S1B ) and all NHL ( OR = 1 . 23 , P = 6 . 81×10−7 ) . ORs were consistent across the seven studies . There was also a suggestive association for rs2647012 with MZL ( OR = 1 . 32 , P = 6 . 34×10−4 ) ( Table . 3 ) , consistent across six studies . Finally , we investigated the possibility of additional susceptibility loci for FL outside of the HLA region by performing a joint analysis of the top 41 to 1000 variants of our scan and the previously published GWAS of follicular lymphoma [6] . From this combined analysis , we did not find any additional markers with a strong association ( P<10−6 ) with FL that were not in LD with our top 5 markers taken forward to stage 3 ( data not shown ) .
Through the identification of a second variant , rs2647012 , that is independent of the previously identified risk variant rs10484561 [6] within the 6p21 . 32 region , our findings substantiate a major link between HLA class II loci and genetic susceptibility to FL . In addition , our study revealed evidence that rs10484561 is associated with DLBCL risk suggesting some shared biological mechanisms of susceptibility between these two common NHL subtypes . The association of rs2647012 with FL risk was not detected in earlier GWAS studies [5] , [6] , and that of rs10484561 with DLBCL risk previously reported was only marginal [6] , perhaps because of the smaller sample sizes in Stage 1 . The number of FL cases scanned in this study was almost double compared to the previous individual GWAS [6] . HLA class II molecules are expressed in antigen presenting cells such as B-lymphocytes , and act to present exogenous antigens to CD4+ helper T-cells . Efficiency of antigen presentation may influence lymphomagenesis through effects on anti-tumor immunity or on immune response to infections that are directly or indirectly oncogenic ( e . g . , through viral genome insertion or nonspecific chronic antigenic stimulation ) [19] . Allelic variants in coding regions may affect the structure of the peptide binding groove of the class II molecules , leading to differences in the efficiency of oncogenic peptide binding or T-cell recognition . Coding sequence variation in the molecules encoded by the extended HLA-DRB1*0101-HLA-DQA1*0101-HLA-DQB1*0501 haplotype may be responsible for the association at rs10484561 [6] . Alternatively , variants in the regulatory sequences may influence the expression level of the HLA molecules and consequently the efficiency of antigen presentation . We note that rs2647012 is strongly associated with the average expression levels of HLA-DRB4 ( β = 0 . 78 , P = 3 . 4×10-22 ) and HLA-DQA1 ( β = -0 . 58 , P = 5 . 1×10−13 ) probes in Epstein-Barr virus-transfected lymphoblastoid cell lines ( mRNA by SNP browser ) [20] , and rs10484561 is also associated with the expression levels of HLA-DQA1 probes ( β = -0 . 884 , P = 1 . 6×10−10 ) . We speculate that this may be an alternative mechanism underlying the observed associations , especially at rs2647012 . Interestingly , SNPs within the same LD block harboring rs2647012 ( r2>0 . 7 in HapMap CEU ) have previously been associated with rheumatoid arthritis with the same direction of effect [21] . Since autoimmune disorders such as rheumatoid arthritis and Sjögren syndrome are associated with increased risk of NHL , in particular with DLBCL but also with FL [22] , our finding may suggest a molecular link between these diseases , although their associations within this region of high LD could also be due to different causal variants . Previously , large-scale candidate gene studies have pointed to susceptibility loci in the HLA class III region mainly between the TNF variant –308G->A ( rs1800629 ) and risk of DLBCL [23] , [24] . We provide novel evidence of association of DLBCL with an independent HLA marker in the class II region ( rs10484561; r2 = 0 ) , 1 . 1Mb away from rs1800629 , strongly suggesting that alleles in the HLA class II region may play an important role in the pathogenesis of this subtype as well . The weaker association of rs10484561 with DLBCL ( OR 1 . 36 ) than with FL ( OR 1 . 95 ) [6] could imply that the DLBCL-association is confined to a subset of DLBCL tumors with specific morphological or molecular features more closely related to FL , such as the germinal center-like B-cell phenotype [25] . However , the observed effects could also be due to modification of other concurrent DLBCL-specific susceptibility variants , or rs10484561 could tag a more strongly associated marker in this region of high LD . Moreover , we found suggestive evidence of association at rs6536942 on 4q32 . 3 , located within an intron of the tolloid-like 1 ( TLL1 ) gene , with FL risk . However , larger studies are needed to validate this finding . Although the strongest associations so far have been observed in the HLA region , and extended pooling of available scan data failed to identify additional loci outside of HLA , we expect that future larger meta-GWAS efforts will more robustly identify additional loci in other regions . In conclusion , our results strongly suggest that future genetic and functional work focused on the HLA class II region will provide important insight into the disease pathology of FL , DLBCL and other subtypes of NHL . In addition , further studies of this region and potential interaction with environmental factors in NHL risk , and of NHL prognosis are warranted .
The studies described in this manuscript have been approved by the ethics committee of the respective institutions: Karolinska Institutet ( Sweden ) , Scientific Ethics Committee system ( Denmark ) , University of California , Berkeley ( US ) , National Cancer Institute , National Institutes of Health ( US ) , Mayo Clinic ( US ) , University of British Columbia ( Canada ) , Yale University ( US ) , University of Sydney ( Australia ) . The SCALE study is a population-based study of the etiology of NHL carried out in all of Denmark and Sweden during 1999 to 2002 [26] . NHL subtype diagnoses were reviewed and reclassified according to the World Health Organization ( WHO ) classification [1] as previously described [26] . For this GWAS ( SCALE1 ) we used DNA from 400 cases with follicular lymphoma ( FL; 150 from Denmark and 250 from Sweden ) and from 150 Danish controls , individually matched to the Danish FL cases by sex and age at study inclusion . We also used material collected from 673 control subjects in a separate Swedish population-based case-control study of rheumatoid arthritis ( the Eira study ) [21] , [27] . The latter was conducted during 1996 to 2005 among residents 18 to 70 years of age in the southern and central parts of Sweden ( including 90% of Swedish residents ) . Hence , the population controls recruited in this study were considered to represent the same study population as the Swedish component of the SCALE study with regard to genetic variation . Genotyping completion rates were similar between cases and controls; out of 400 cases and 823 controls genotyped , 379 cases ( 95% ) and 791 controls ( 96% ) were included in the final analysis . Study subjects used in Stages 2 , 3 and validation in other NHL subtypes ( Table 1 , Table S1 , S2 ) have been previously described [6] , [8]–[12] , and details are available as supporting text ( Text S1 ) . For the SCALE2 NHL subtype validation study , we used the rest of the lymphoma cases with blood samples originally recruited in SCALE ( n = 1869 ) , Danish control subjects not included in the GWAS ( n = 556 ) , a second set of control subjects from the Eira study ( n = 742 ) and a third group of controls recruited in a national population-based case-control study of breast cancer , the Cancer and Hormones Replacement in Sweden ( CAHRES ) study [28] ( n = 720 ) . The control subjects from this study were randomly selected from the Swedish general population to match the expected age distribution of the participating breast cancer cases ( 50 to 74 years ) . Stage I genotyping of 317 , 503 single nucleotide polymorphisms ( SNPs ) was done on the HumanHap300 ( version 1 . 0 ) array . Validation genotyping was done using Sequenom iPlex; SNPs in the human leukocyte antigen ( HLA ) region that failed primer design for Sequenom assays were genotyped using Taqman ( Applied Biosystems ) . The scan included 317 , 503 SNPs from the HumanHap300 ( version 1 . 0 ) array . The datasets were filtered on the basis of SNP genotyping call rates ( ≥>95% completeness ) , sample completion rate ( ≥90% ) , minor allele frequency ( MAF; all subjects as well as cases and controls separately ≥0 . 03 ) and non-deviation from Hardy-Weinberg equilibrium ( HWE; p<10−6 ) . We also excluded SNPs with cluster plot problems , and those on the X and Y chromosomes . Study subjects with gender discrepancies and/or labelling errors were removed . We also removed individual samples with evidence of cryptic family relationships ( identified using the–genome command in PLINK ) . To detect outliers in terms of population stratification , we performed principal component ( PC ) analysis using the EIGENSTRAT software ( Figure S4 ) . A subset of linkage disequilibrium ( LD ) thinned SNPs was selected such that all pair-wise associations had r2<0 . 2 , and long-range regions of high LD , reported to potentially confound genome scans , were removed [29] . Twenty-five samples were removed as population outliers on the basis of their values on the first three PCs . To adjust for possible stratification in our association analyses we adjusted the regression analyses using the first three PCs; the number of PCs used for adjustment was determined by plotting the eigenvalues and locating the position of the “elbow” on the scree plot ( Figure S5 ) . Wald tests , treating minor allele counts as continuous covariates were used to test for association . The genomic inflation factor ( λ ) was calculated to be 1 . 0283 after adjusting for the first three PCs , suggesting the presence of minimal stratification . Quantile-quantile plots for the associations before and after adjustment are shown in Figure S6 . Finally , we assessed associations of age and sex with main genotypes among the control subjects to address the possibility of confounding by these factors ( Table S11 ) . As there was no evidence of associations of age or sex with genotypes among the controls , we did not adjust for them in the final main effects analyses of genotypes . In Stage 2 , similar quality control measures were applied as in Stage 1 , including genotyping call rate ≥95% , sample completion rate ≥90% , and MAF ≥0 . 05 . We tested each validation study for association using trend tests . For meta-analyses across studies and NHL subtypes , we used the Cochran-Mantel-Haenszel method to calculate the combined odds ratio and P-value , and χ2 tests for heterogeneity . Multivariate logistic regression was used to test for independence of SNP effects . For validation among other NHL subtypes , the control subjects were the same as those in Stages 2 and 3 for validation in FL for all studies except SCALE2 . Only European-ancestry subjects were included , and the possibility of population stratification affecting the results has been thoroughly explored and found to be low in earlier investigations in the same populations [6] , [8] . We used IMPUTEv1 for the imputation of SNPs from the 1000 Genomes pilot1 CEU data ( August 2009 release ) ; and the HapMap Phase II release 22 CEU data . We set a strict threshold for imputation , using only SNPs with confidence scores of ≥0 . 9 , call rates ≥90% , non-deviation from Hardy-Weinberg equilibrium P >0 . 001 and MAF >0 . 01 . The imputation was done on the Stage 1 samples separately for each of the two reference datasets and SNPs showing a discordance of >5% between the genotypes imputed with the two datasets were excluded from further analysis . The data were then merged using HapMap II as the master dataset to which additional imputed SNPs from the 1000 Genomes dataset were added . HLA alleles were imputed by identifying tag SNPs [15] from the genotyped and imputed SNP dataset . We used PLINK for haplotype imputation with the tag SNPs and downstream association analyses . Only haplotypes with call rates >90% , MAF>1% and probability thresholds >0 . 8 were analyzed . For coalescence analysis all 12 SNPs ( genotyped in this study and within a region of ∼177 Kb ) adjacent to the two SNPs associated with the FL risk were used to construct haplotypes . These were phased using the PHASE program [30] and tested for association using PLINK . The ancestral haplotype was constructed from the chimpanzee ( PanTro2 ) allele whenever possible , and otherwise from the macaque alleles . An ancestral recombination graph was constructed using the program Beagle [13] , [31] which allows recombination assuming an infinite site mutation model . After identifying the first recombination event the haplotype segment before the recombination spot was used to construct a median –joining network using the Network program [14] . The alleles of the imputed SNP rs9378212 were then phased on each haplotype segment using the PHASE program . The URLs for the data and analytic approaches presented herein are as follows: 1000 Genomes http://1000genomes . org HapMapII http://www . hapmap . org IMPUTEv1 https://mathgen . stats . ox . ac . uk/impute/impute_v1 . html mRNA by SNP browser http://www . sph . umich . edu/csg/liang/asthma/ R script for recombination plot http://www . broadinstitute . org/science/projects/diabetes-genetics-initiative/plotting-genome-wide-association-results | Earlier studies have established a marker rs10484561 , in the HLA class II region on 6p21 . 32 , associated with increased follicular lymphoma ( FL ) risk . Here , in a three-stage genome-wide association study of 1 , 428 FL cases and 6 , 581 controls , we identified a second independent FL–associated marker on 6p21 . 32 , rs2647012 , located 962 bp away from rs10484561 . The associations at two SNPs remained genome-wide significant after mutual adjustment . Haplotype and coalescence analyses indicated that rs2647012 arose on an evolutionarily distinct lineage from that of rs10484561 and tags a novel allele with an opposite , protective effect on FL risk . Moreover , in an analysis of the top 6 FL–associated SNPs in 4 , 449 cases of other NHL subtypes , rs10484561 was associated with risk of diffuse large B-cell lymphoma . Our results reveal the presence of allelic heterogeneity at 6p21 . 32 in FL risk and suggest a shared genetic etiology with the common diffuse large B-cell lymphoma subtype . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"hematology/lymphomas",
"and",
"chronic",
"lymphoblastic",
"leukemia",
"oncology",
"genetics",
"and",
"genomics/gene",
"discovery",
"genetics",
"and",
"genomics/gene",
"expression",
"genetics",
"and",
"genomics/complex",
"traits",
"genetics",
"and",
"genomics/genetics",
"of",
"disease",
"hematology",
"oncology/hematological",
"malignancies",
"oncology/myelomas",
"and",
"lymphoproliferative",
"diseases",
"genetics",
"and",
"genomics",
"genetics",
"and",
"genomics/medical",
"genetics",
"genetics",
"and",
"genomics/cancer",
"genetics"
] | 2011 | GWAS of Follicular Lymphoma Reveals Allelic Heterogeneity at 6p21.32
and Suggests Shared Genetic Susceptibility with Diffuse Large B-cell
Lymphoma |
Large-scale genome rearrangements have been observed in cells adapting to various selective conditions during laboratory evolution experiments . However , it remains unclear whether these types of mutations can be stably maintained in populations and how they impact the evolutionary trajectories . Here we show that chromosomal rearrangements contribute to extremely high copper tolerance in a set of natural yeast strains isolated from Evolution Canyon ( EC ) , Israel . The chromosomal rearrangements in EC strains result in segmental duplications in chromosomes 7 and 8 , which increase the copy number of genes involved in copper regulation , including the crucial transcriptional activator CUP2 and the metallothionein CUP1 . The copy number of CUP2 is correlated with the level of copper tolerance , indicating that increasing dosages of a single transcriptional activator by chromosomal rearrangements has a profound effect on a regulatory pathway . By gene expression analysis and functional assays , we identified three previously unknown downstream targets of CUP2: PHO84 , SCM4 , and CIN2 , all of which contributed to copper tolerance in EC strains . Finally , we conducted an evolution experiment to examine how cells maintained these changes in a fluctuating environment . Interestingly , the rearranged chromosomes were reverted back to the wild-type configuration at a high frequency and the recovered chromosome became fixed in less selective conditions . Our results suggest that transposon-mediated chromosomal rearrangements can be highly dynamic and can serve as a reversible mechanism during early stages of adaptive evolution .
Organisms have different ways to respond to environmental stresses and evolve corresponding adaptive functions [1] . At the genic level , adaptation can be achieved by subtle , small-scale nucleotide changes ( base insertions , deletions or substitutions ) that alter gene expression , protein structure or protein interactions . Alternatively , at the genomic level , large-scale genome rearrangements ( chromosome duplications , translocations and aneuploidy ) create copy number variations that may change gene dosage so as to shape adaptive evolution . Although a similar adaptive phenotype can be achieved by both mechanisms [2] , it is still unclear whether one type of mutations is specifically preferred under certain conditions , especially in natural populations . Unicellular organisms can quickly adapt to different environmental challenges in diverse niches . Comparing different populations of the same microbes that have adapted to distinct environments allows us to identify the underlying mechanisms of adaptive evolution [3] , [4] . Studies using the budding yeast Saccharomyces cerevisiae have revealed that both small- and large-scale adaptive changes have occurred in natural and laboratory yeast populations [5] . For example , in a natural yeast strain , a few point mutations in the transcriptional factors , IME1 , RME1 and RSF1 , were found to improve sporulation efficiency [6] . Another study in yeast isolated from sherry wines showed that this yeast strain carries two types of mutations in the gene encoding a cell surface glycoprotein . The mutations include a 111-bp deletion in the promoter region that increases its expression level and duplications of a tandem repeat in the coding region that enhance the protein's hydrophobicity [7] . Large-scale changes that involve chromosome duplication , translocation or aneuploidy have been observed in yeast populations during short-term evolution experiments [5] , [8]–[13] . Under glucose-limited conditions , evolved strains carry an amplified region that encodes a high-affinity hexose transporter [14] . Under sulfate-limited conditions , amplifications of a high-affinity sulfate transporter locus ( SUL1 ) were observed [15] . Beyond changes observed in experimental populations , a chromosome translocation resulting in overexpression of SSU1 , a gene encoding a sulfite efflux pump , was identified in a sulfate-tolerant wine strain [16] . In other yeast species , such as the clinical isolates of pathogenic Candida spp . , large-scale chromosomal rearrangements also play an important role in drug resistance . For example , aneuploidy and isochromosome formation increase the copy number and expression of critical genes for fluconazole resistance in Candida albicans [17] , [18] . Segmental duplications and new chromosome formation were found to be correlated with fluconazole tolerance in Candida glabrata [19] . These studies indicate that large-scale changes allow yeast to quickly adapt to different environments . Despite this wealth of experimental data , it is less clear how cells maintain these mutations over a long evolutionary timescale since large-scale rearrangements are often accompanied by extra costs . In sexual populations , large-scale rearrangements can also result in gamete lethality when they are heterozygous unless they localize near the telomeres and do not carry regions with essential genes [20] . The mutations that cause large-scale chromosomal rearrangements occur at a high frequency in yeast populations . In mutation-accumulation lines of haploid budding yeast , the estimated spontaneous mutation rate of large-scale changes was 4 . 8-fold higher than that of small-scale changes ( 0 . 019 and 0 . 004 per genome per cell division , respectively ) [21] . In another similar experiment in diploid yeast cells , it was shown that most structural variations occurred in the subtelomeric regions [22] . Like other types of mutations , most large-scale changes are probably deleterious and will quickly vanish from the population [23] . However , even in large evolving populations isochromosome formation and segmental duplication can be detected after as few as 5 or 100 generations , respectively [24] , indicating that large-scale mutations supply the population with genetic variation that could facilitate adaptation to novel environments . It has been suggested that Ty transposons may play an important role in the formation of large-scale chromosomal changes in yeast [25] . Although the yeast genome is relatively compact compared to other eukaryotic genomes , about 1–4% of the yeast genome is comprised of Ty sequences [26] . In addition , Ty sequences are often found in clusters [25] . Inverted arrays of transposon sequences can cause replication fork stalling that leads to chromosome breakage , especially when the replication machinery or checkpoints are compromised [27]–[29] . Those Ty-rich regions may constitute a preferred double-strand break site similar to the fragile sites observed in mammalian chromosomes [28] , [30] . Previous studies in budding yeast suggested that many observed chromosomal rearrangements might result from ectopic recombination between Ty sequences [14] , [30]–[33] . It is likely that Ty sequences often serve as initiation sites for generating chromosomal rearrangements . Our knowledge about natural adaptation of budding yeast is often complicated due to human interference in the natural history of yeast . Yeast strains collected from Evolution Canyon ( EC ) provide an excellent model for studying how yeast populations adapt to natural environments . EC is an east-west-oriented canyon at Lower Nahal Oren , Israel , that originated 3–5 million years ago and is believed to have experienced minimal human disturbance [34] . Its microclimates provide ideal conditions for diverse local adaptations of many organisms [35]–[37] . In previous work , we employed a panel of phenotypic assays to characterize 14 diploid yeast strains collected from different locations within EC . We observed that a specific group of EC yeast strains ( EC-C1 ) could tolerate a high concentration of cadmium . The cadmium-resistant phenotype was shown to be caused by an ancient allele of PCA1 ( PCA1-C1 ) , which encodes a metal efflux pump [38] . Here , we show that the same group of EC yeast strains was also highly resistant to another metal , copper . However , the copper-tolerant phenotype is not correlated with the PCA1-C1 mutant allele . Instead , the copper-tolerant phenotype mainly results from chromosomal rearrangements that increase the copy numbers of CUP1 and CUP2 , two major genes involved in copper regulation [39] , [40] . By analyzing the whole-genome expression pattern of cells carrying different copy numbers of CUP2 , we found three previously unidentified genes , PHO84 , SCM4 and CIN2 , whose expression was regulated by Cup2 dosage and contributed to copper tolerance . Finally , we observed that the chromosomal rearrangements in EC-C1 cells were highly reversible . When cells were growing in medium with 1 mM of copper sulfate , a wild type-like chromosome reappeared and was fixed in the population within 300 generations . These results suggest that large-scale chromosomal rearrangements provide not only a fast arising but also readily reversible source of variation during early stages of adaptive evolution .
Yeast strains collected from Evolution Canyon have been shown to adapt to various environmental stresses , such as oxidative stress , UV radiation , and high concentrations of cadmium [37] , [38] , [41] . In addition , most of the EC strains are heterothallic [42] . To further examine if EC strains have evolved other adaptive phenotypes , we tested the growth of EC diploid strains on several metal-containing plates . Interestingly , those cadmium-resistant strains ( EC-C1 strains , including EC9 , 10 , 35 , 36 , 39 , 40 , 57 and 58 ) could also tolerate high concentrations of copper sulfate ( Figure 1A ) . However , when we crossed the copper-tolerant haploids with a copper-sensitive strain and analyzed the meiotic products , we found that the copper-tolerant phenotype did not co-segregate with the PCA1-C1 mutation responsible for the cadmium resistance ( data not shown ) . In our previous study , we also showed that the PCA1-C1 allele did not increase the copper tolerance when it was put into a copper-sensitive strain [38] . Together , these results suggest that other genes are responsible for the tolerance to copper in the EC-C1 strains . In our previous study , we observed that the diploid S . cerevisiae strains isolated from Evolution Canyon comprised three major karyotypes ( with some minor deviations ) , including EC cluster 1 ( EC-C1 ) , EC cluster 2 ( EC-C2 ) and EC cluster 3 ( EC-C3 ) ( Figure 1B ) [38] . This karyotype clustering pattern is consistent with the results from the phylogenetic analyses [42] , [43] . Because all copper-tolerant strains belong to EC-C1 , it suggests that the metal-tolerant phenotypes had already evolved before the EC-C1 populations split . Therefore , we chose EC9 from EC-C1 as representative of this clade for subsequent genetic analyses . Laboratory evolution experiments have shown that chromosomal rearrangements can result in adaptive changes to gene copy number [14] , [15] , [44] . To further examine each individual chromosome , chromosomes separated by pulsed-field gel electrophoresis ( PFGE ) were subjected to Southern blotting using chromosome-specific DNA probes . The result showed that EC-C1 strains have high chromosome heterozygosity . They carry at least four heterozygous chromosome pairs ( chromosomes 5 , 6 , 8 and 14 ) , as revealed by length differences between homologous chromosomes . In addition , we observed several large chromosomal rearrangements in EC-C1 strains that had resulted in an elongated chromosome 10 , an elongated chromosome 8 of almost twice its original size , and a novel chromosome that was hybridized by probes from both chromosomes 7 and 8 ( Figure S1 ) . The fact that the latter two chromosomal rearrangement events that we observed both involved chromosome 8 prompted closer examination . The rearranged chromosomes were purified from PFGE gels and subjected to array-based comparative genomic hybridization ( aCGH ) using S . cerevisiae oligonucleotide microarrays . These experiments revealed that the aberrant 900-kb chromosome 8 is a fusion product of two chromosome 8 fragments ( between YHR015W to YHR210C and YHL008C to YHR219W ) and that the novel 650-kb chromosome is a fusion product of a small chromosome 7 fragment ( between YGL096W and YGL200C ) , a large chromosome 8 fragment ( between YHL050C and YHR145C ) and the telomere of chromosome 8 ( between YHR210C to YHR217C ) ( Figure 2 ) . We also conducted aCGH using genomic DNA isolated from EC-C1 diploid cells ( EC9 ) and haploid cells that carry both rearranged chromosomes ( EC9-7 in Figure 3B ) . The results confirmed that the copy numbers were indeed increased in the duplicated regions . To understand how novel chromosomes were formed , we fine-mapped the junction sites of the rearranged chromosomes . We designed primers near each possible breakpoint according to the aCGH data ( i . e . , regions close to YHR015W , YHR210C , YHL008C , YHR219W , YHL050C , YHR145C , YGL200C and YGL096W ) and used these primers to find out the junction site of two chromosomal fragments ( see Materials and Methods ) . As shown in Figure 2C and Figure S2 , the aberrant 900-kb chromosome 8 was formed by fusing the regions near YHR015W and YHL008C and the novel 650-kb chromosome was formed by fusing the regions near YHR145C and YGL096W . Interestingly , we found that the breakpoints were all flanked by Ty sequences ( next to YHR015W , YHL008C , YHR145C and YGL096W ) , indicating that transposable elements might be the mediator of these chromosomal rearrangements . Moreover , three out of the four flanking regions ( YGL096W , YHR015W and YHL008C ) contain multiple Ty long terminal repeats ( LTRs ) including at least one inverted LTR pair . It is possible that the double-strand break hotspots formed in these Ty arrays allow chromosomes to be rearranged at a high frequency . We also sequenced the breakpoints of the other EC-C1 strains . The result confirmed that the same chromosomal rearrangements exist in all EC-C1 strains ( Table S1 ) . To assess the contribution of the rearranged chromosomes to copper tolerance , we dissected tetrads ( meiotic products ) of EC9 and measured the copper resistance of individual haploid segregants ( spores ) . The spore viability of EC9 is about 60% due to the chromosomal rearrangement of EC-C1 strains . In eight four-viable-spore tetrads yielding 32 haploid segregants , all sixteen segregants containing both rearranged chromosomes showed higher copper tolerance than the other sixteen segregants containing only wild type copies of chromosomes ( Figure 3 and Figure S3 ) . However , we noticed that within these two groups of segregants , there were different levels of copper tolerance between individual clones , indicating that the copper-tolerant phenotype in EC-C1 strains was polygenic and genes on other chromosomes might also be involved in copper tolerance with minor effects . We measured the relative fitness of twenty EC9 haploid segregants in copper-containing medium ( Figure 3C ) . By comparing the fitness between stains carrying rearranged chromosomes and the wild type chromosome , we estimated that rearranged chromosomes are responsible for about 60% of the observed copper-tolerant phenotype . When the rearranged chromosomes were inspected , we observed that genes involved in response to copper ions ( CUP1 , CUP2 and COX23 ) were significantly enriched . CUP1 is a gene encoding a metallothionein and its expression level has been shown to play an important role in copper tolerance [39] . We measured the CUP1 gene copy number and expression level using quantitative PCR . The results showed that the CUP1 copy number and mRNA level in EC9 ( an EC-C1 strain ) were about 5–6-fold higher than expression in EC34 and EC63 ( EC-C2 and EC-C3 strains ) after cells were treated with CuSO4 ( Figure 4A and 4B ) . To confirm that the increased copies of CUP1 are important for copper tolerance in EC-C1 strains , we deleted eight copies of CUP1 in an EC9 haploid segregant ( EC9-7 in Figure 3B ) and measured their copper sensitivity . The results showed that cells with fewer copies of CUP1 were indeed less copper-tolerant ( Figure 4C ) . Cup2 is a copper-binding transcriptional factor that activates CUP1 expression [40] , [45] . The chromosome rearrangements increase the CUP2 copy number to three in EC-C1 cells . To determine whether the increased copy number of CUP2 contributes to copper tolerance in EC-C1 strains , we used EC9-7 to construct yeast strains carrying zero , one or two copies of CUP2 and tested their copper sensitivity . The sensitivity of cells to copper was negatively correlated with the copy number of CUP2 ( Figure 4D ) , indicating that an increase in the copy number of CUP2 is also important for copper tolerance . Our results indicate that amplification of both CUP1 and CUP2 genes are required for cells to achieve high copper tolerance . The dosage effect of CUP2 prompted us to investigate the downstream targets of the Cup2 transcription factor . In previous studies , Cup2 has been shown to regulate three metal-responsive genes , including two metallothionein genes , CUP1 and CRS5 , and the copper-zinc superoxide dismutase gene SOD1 [40] , [45]–[47] . To identify more candidate genes under the regulation of Cup2 , we performed a whole-genome expression pattern analysis of the EC9 haploid segregant carrying different copy numbers of CUP2 ( EC9-7 , EC9-7 cup2Δ and EC9-7 cup2Δ cup2Δ in Figure 4D ) . Cells were treated with 1 mM CuSO4 for 1 h , which resulted in no obvious effects on the growth of cup2 null mutant cells . Total RNA from treated samples was collected and analyzed using microarrays . In cup2 double deletion cells we found 39 genes with reduced ( 1 . 5-fold or more ) expression compared with wild type cells ( Table S2 ) . Among these genes , 18 showed a positive correlation between their expression levels and the copy number of CUP2 . To directly test the effect of these candidate genes on copper tolerance , we examined eight non-essential genes in the aforementioned group ( PDR5 , SNQ2 , CYB5 , SCM4 , STP4 , HPT1 , CIN2 and LIA1 ) and PHO84 , a gene whose expression showed more than a 3-fold difference between wild type and cup2 double deletion cells but did not correlate with the CUP2 copy number . We found that when individual genes were deleted , three mutant strains ( pho84Δ , scm4Δ and cin2Δ ) showed reduced tolerance to copper sulfate ( Figure 5A ) . To rule out the possibility that the mutant cells were sensitive to sulfates instead of copper , the deletion strains were also tested on plates containing copper chloride ( Figure S4 ) . Our result indicates that pho84Δ , scm4Δ and cin2Δ mutant cells were indeed sensitive to copper despite no reported linkage between these genes and copper tolerance . We further tested the effect of increased expression of these three genes by introducing a CEN plasmid carrying PHO84 , SCM4 or CIN2 into EC9-8 haploid cells and examining their copper tolerance . However , we were unable to detect obvious growth differences using spot assays ( data not shown ) . We also tried to measure cell fitness using a more sensitive competitive fitness assay , but it was unsuccessful as the EC cells tended to clump together in copper-containing medium . Our results suggest that PHO84 , SCM4 and CIN2 are involved in copper tolerance , but it is less clear how much the increased expression of these three genes contributes to the elevated copper tolerance in EC-C1 cells . To investigate the molecular mechanisms about how PHO84 , SCM4 and CIN2 affect copper tolerance , we tested whether expression of the CUP1 gene was affected by mutations in these three genes . Total RNA was isolated from cells treated with 1 mM CuSO4 and the expression level of CUP1 was measured using quantitative PCR . In pho84Δ cells , CUP1 expression was significantly reduced , suggesting that Pho84 may influence copper tolerance through a Cup1-dependent mechanism ( Figure 5B ) . In yeast , it has been shown that large-scale chromosomal rearrangements occur frequently and beneficial ones can become fixed rapidly in the population [14] , [21] , [24] . It is therefore of interest whether the observed rearranged chromosomes have existed in the EC strains for a long time or have been formed recently . We sequenced a 6628-bp fragment ( corresponding to positions 188 , 179 to 194 , 679 bp on chromosome 7 ) from the rearranged chromosome 7 and a 6602-bp fragment ( corresponding to positions 496 , 154 to 502 , 755 on chromosome 8 ) from the aberrant chromosome 8 and compared them with wild-type chromosomes ( see Materials and Methods for details ) . Both fragments had the same sequence as that on wild-type chromosomes . From the genome sequence divergence between two closely related yeast species , S . cerevisiae and S . paradoxus , we obtained an estimation that it may require about 0 . 25 million years to accumulate 1% of sequence divergence in intergenic regions [48] . If we applied this estimation to the case of the rearranged chromosomes in EC-C1 strains , it suggests that the rearranged chromosomes occurred in the last 3800 years assuming that there was no recombination between the rearranged and wild type chromosomes . Large-scale chromosomal rearrangements can spread through a population if they are beneficial . However , the rearranged chromosomes can be quite unstable when growth conditions change and selective pressure is relieved [24] , [49] . We set up a laboratory evolution experiment to investigate the stability of rearranged chromosomes . Five individual colonies derived from an EC9 haploid segregant carrying rearranged chromosomes ( EC9-7 in Figure 3 ) were used to set up 10 independent evolving lines , 5 with relaxed selection ( YPD containing 1 mM CuSO4 ) and 5 with strong selection ( YPD containing 5 mM CuSO4 ) . We regarded the medium containing 1 mM CuSO4 as a more relaxed growth condition because EC9 haploid segregants without rearranged chromosomes ( such as EC9-6 in Figure 3B ) could grow efficiently under such conditions . These cells were grown and diluted daily in fresh medium . After 400 generations , we observed that evolved cells with relaxed selection all exhibited improved growth on YPD or YPD with 1 mM CuSO4 , but decreased tolerance to high concentrations of copper ( Figure 6A ) . We examined the karyotype of evolved cells collected from generations 100 , 200 , 300 and 400 . In all five evolved cultures with relaxed selection , the rearranged chromosomes ( in the size range of 900 kb and 650 kb ) were replaced by a novel wild type-like chromosome 8 ( approximately 550 kb ) during the course of evolution ( Figure 6B ) . Since each evolved culture was initiated from an independent single colony , it is likely that the novel chromosome 8 repeatedly evolved at least five times in our evolution experiment . In addition , this novel chromosome 8 could be detected in the populations collected from generation 100 , suggesting that it already existed in the population at a very early stage of our experiment ( Figure 6B ) . In contrast , four out of five evolved cultures under strong selection pressure retained the rearranged chromosomes in the majority of populations even after 400 generations ( Figure S5A ) . We also set up evolution experiments using EC9 diploid cells . Similar results were observed except that it took a longer time to fix the wild type-like chromosome 8 in diploid populations ( Figure S5B ) . One possible explanation for the difference between haploid and diploid populations is that the cost of carrying extra chromosomal fragments is relatively lower in diploid cells [50] . Together , these results suggest that these chromosomal rearrangements are highly dynamic and reversible . To further investigate its structure , the novel 550-kb chromosome 8 of the evolved cells was purified from PFGE gels and subjected to array-based comparative genomic hybridization ( aCGH ) . The result showed that the novel chromosome 8 had almost the same gene content as the wild type chromosome 8 except for some telomeric genes; YHR217C , YHR218W and YHR219W in the right telomere and YHL044W , YHL045W , YHL048W , YHL049C and YHL050C in the left telomere were undetected in our assay . On the other hand , we detected signals of other telomeric regions , including YAR062W , YAR064W , YAR066W , YAR068W , YAR069C , YAR070C and YAR073W from the telomere of chromosome 1 and YFL065C , YFL066C and YFL067W from the telomere of chromosome 6 . Telomeric regions have been known to be very dynamic . It is possible that recombination between different telomeres occurred during our evolution experiment .
It has been observed that yeast can adapt to various nutrient-limited conditions [15] , [44] , [51] . Recent surveys on yeast strains collected from different continents also revealed that S . cerevisiae populations exhibit a high degree of phenotypic variance , suggesting that they have adapted to diverse ecological niches [26] , [52] . However , unlike experimentally evolved cells , adaptations in natural populations are more difficult to study . It has remained elusive whether the types of mutations commonly observed in laboratory adaptation are also involved in natural adaptation . Copper is an essential cofactor for many enzymes such as the cytochrome c oxidase in the respiratory chain . Nonetheless , an excess of copper is deleterious to cells [53] . The toxicity of copper may come from the generation of reactive oxygen species , the competition with other metals for their native binding sites , the alteration of protein conformations or interference with biochemical reactions [53] , [54] . Cells have evolved multiple mechanisms to regulate copper homeostasis including different metal transporters , sequestration factors , and detoxification enzymes [54] , [55] . The EC-C1 strains carry two rearranged chromosomes that significantly enrich the copy number of genes involved in copper regulation . Evolved phenotypes often arise from duplicated chromosomal fragments that contain the critical genes for adaptation in experimental yeast populations [14] , [15] , [44] . We speculate that the rearranged chromosomes in EC-C1 strains might result from selection for higher copper tolerance for the following reasons . First , among all the diploid yeast strains collected from Evolution Canyon , the EC-C1 strains constitute a major group ( 8/21 or 38% ) . In addition , the EC-C1 strains show very low levels of polymorphism in their microsatellite loci compared with other EC groups [43] , suggesting that EC-C1 strains carry some adaptive phenotypes allowing them to quickly spread in Evolution Canyon . Second , when the copper content in the soil samples collected from different sites of Evolution Canyon was measured using inductively coupled plasma-atomic emission spectroscopy ( see Materials and Methods ) , we found that the copper levels of most EC sites are above 30 ppm ( with an average of 38 ppm ) and in one area it even reaches 95 ppm , which are higher than the average copper content ( 20 ppm ) in soil [56] . Previous studies have suggested that increased copper levels in vineyard soil caused the wine yeast strains to evolve higher copper tolerance [57] , [58] . A similar adaptive process might also occur in the EC-C1 strains . Third , a previous study has shown that increased expression of CUP1 also enhanced cadmium resistance of cells . When we examined cadmium sensitivity of twenty EC9 haploid segregants , the cadmium-resistant phenotype was not co-segregated with rearranged chromosomes . This suggests that increased copies of CUP1 and CUP2 were not a result of selection for the pleiotropic effect on cadmium resistance ( Figure S6 ) . In our lab , we have examined the fitness of all EC strains under more than 30 different growth conditions ( including different temperatures , nutrient starvation , chemicals and metal ions ) . Only in two conditions , medium containing either high levels of copper or cadmium , EC-C1 strains showed higher growth rates ( [38] and our unpublished results ) . Nonetheless , we cannot completely rule out the possibility that chromosomal rearrangements in EC-C1 strains were caused by adaptive effects resulting from other amplified genes or other unknown pleiotropy of Cup1 and Cup2 . In the future , it will be interesting to compare the whole-genome gene expression patterns between haploid cells carrying rearranged or wild type chromosomes under different conditions . If differentially regulated genes are enriched in biological pathways other than copper tolerance , it may provide us a clue to further test other possible causes of chromosomal rearrangements in EC-C1 strains . Increased CUP1 copy number has been observed in copper-resistant strains isolated from laboratory evolution experiments , industry or natural habitats [59]–[61] . Nonetheless , EC-C1 diploid cells carry more than 20 copies of CUP1 that are much higher than the CUP1 amplification reported in previous cases ( ranging from 2 to 15 copies ) . In addition , we showed that increased CUP1 copy number alone was not enough to achieve the high copper tolerance observed in EC-C1 cells . Amplification of the copper-binding transcriptional factor CUP2 was also critical , suggesting that a more complex adaptive strategy has occurred in EC-C1 strains . Increasing the dosage of a transcriptional factor may influence the expression of its downstream target genes to different levels depending on its feedback regulation or other compensatory mechanisms . Although the CUP1 amplification clearly plays a major role in copper tolerance of EC-C1 cells , other downstream targets of Cup2 probably also contribute to the observed phenotype . By combining the whole-genome gene expression analysis of cells carrying different copy numbers of CUP2 and the functional assay , we identified and confirmed three previously unidentified genes , PHO84 , SCM4 and CIN2 , that were involved in copper tolerance . In two of them ( SCM4 and CIN2 ) we also observed a conserved Cup2-binding motif sequence in their promoters . CIN2 encodes a GTPase-activating protein involved in tubulin folding [62] , and cin2Δ mutant cells are also sensitive to another metal , arsenic , suggesting that Cin2 is involved in metal regulation [63] . SCM4 was previously identified as a suppressor of a cell cycle mutant of CDC4 [64] . Nonetheless , the Scm4 protein contains four transmembrane domains and localizes to the mitochondria , an organelle involved in many metal metabolic pathways . It will be interesting to determine whether Scm4 affects copper tolerance through its function in mitochondria . PHO84 encodes a high-affinity inorganic phosphate transporter that also functions in manganese homeostasis [65] , [66] . The enhanced tolerance of pho84Δ mutants to several metal ions ( including manganese , zinc , cobalt and copper ) has been attributed to defects in the uptake of metal ions [66] . However , we found that deletion of PHO84 in EC-C1 strains decreases tolerance to high concentrations of copper in a Cup1-dependent manner . This suggests that genetic background may have a strong influence on the regulatory network of metal metabolism . Large-scale chromosomal rearrangements can quickly change the expression level of multiple genes or even a whole pathway by changing the gene copy number . In addition , the spontaneous rate of chromosomal rearrangements is higher than the spontaneous rate of point mutations [21] . This class of mutations is most likely to be found at the early stage of adaptation since they allow a brute force change in phenotype by changing multiple genes in one step . Many such examples have been reported in short-term experimental evolution in S . cerevisiae and C . albicans [8]–[11] , [15] , [24] , [67] . But in addition to increasing the copy numbers of beneficial genes , rearrangement also increases copy numbers of other genes in the same chromosomal segments that may not be beneficial . It has been shown that when compared with euploid cells , aneuploid cells have higher fitness under certain conditions but have reduced fitness in general [50] , [68] . Thus , chromosomal rearrangement is unlikely to be an optimal form of mutation , but may allow a population to survive until either better mutations appear or until a population's environment becomes more permissive . In the EC-C1 strains , we examined two loci on the rearranged chromosomes 7 and 8 ( ∼7 kb/each ) and found that they had identical sequences as wild type chromosomes , supporting the idea that the rearranged chromosomes were recently generated rather than an ancient relic . A lingering question is then how cells adjust to the cost of these crude adaptive changes on longer evolutionary timescales , especially in a fluctuating environment . When EC-C1 cells carrying the rearranged chromosomes were propagated in medium containing 1 mM copper sulfate , a wild type-like chromosome 8 quickly became fixed in all five individual populations in as early as 200 generations . The repetitive appearance of this novel chromosome 8 suggests that some large-scale chromosomal rearrangements are highly dynamic and reversible . This result is in agreement with a previous study showing that large-scale inter- and intra-chromosomal duplications were intrinsically unstable when no selective advantages were provided by those duplications [49] . Previous studies in budding yeast suggested that clustered Ty sequences might serve as double-strand break hotspots to initiate ectopic recombination in the yeast genome [28] , [30] . This type of recombination allows cells to quickly adapt to stressful environments by duplicating chromosomal fragments that contain critical genes . Furthermore , when the stress is relieved or better mutations have evolved , the duplicated chromosomal fragments can revert back to the original configuration at a high frequency by another round of ectopic or homologous recombination . Such genome flexibility enables organisms to generate switch-like adaptive phenotypes . This would be especially valuable for sexual populations since large-scale chromosomal rearrangements often cause gamete lethality when they are heterozygous . This idea is indirectly supported by the observation that although large-scale chromosomal rearrangements occur frequently in laboratory evolution experiments or natural isolates , closely related yeast species such as S . cerevisiae and S . paradoxus still maintain colinear genomes [69] . Together with the facts that transposon expression is known to be activated under environmental stress and elevated transcription levels increase the rate of mitotic recombination [70]–[72] , the abovementioned Ty-mediated chromosomal rearrangements supply the population with an effective mechanism to quickly respond to environmental changes .
All EC diploid strains are Saccharomyces cerevisiae collected from an east-west facing canyon ( Evolution Canyon ) at Lower Nahal Oren , Israel [43] . In brief , EC33 , 34 , 35 and 36 were isolated from the south-facing slope ( SFS ) , EC9 , 10 , 39 and 45 from the valley bottom ( VB ) , and EC13 , 57 , 58 , 59 , 60 and 63 from the north-facing slope ( NFS ) . Substitutive and integrative transformations were carried out by the lithium acetate procedure [73] . Media , microbial and genetic techniques were performed as described [74] . A total of 1∼2×108 yeast cells were used for plug preparation . Cells were washed with 1 ml EDTA/Tris ( 50 mM EDTA , 10 mM Tris , pH 7 . 5 ) and transferred into EDTA/Tris with 0 . 13 mg/ml zymolyase ( Seikagaku America Inc . , St . Petersburg , FL ) . The cell mixtures were incubated for 30 s at 42°C and then embedded in low melting point agarose ( Sigma-Aldrich , St . Louis , MO ) . The agarose plugs were placed at 37°C overnight for zymolyase digestion . After digestion , the agarose plugs were placed in LET solution ( 0 . 5 M EDTA , 10 mM Tris , pH 7 . 5 ) containing 2 mg/ml protease K and 1% N-lauroylsarcosine at 50°C overnight . This step was repeated three times . The plugs were transferred to EDTA/Tris solution and dialyzed four times for 1 h at 37°C . Yeast chromosomes were separated on 0 . 7% agarose gels by pulsed field gel electrophoresis ( PFGE ) using a Rotaphor Type V apparatus ( Biometra , Göttingen , Germany ) . Electrophoresis was performed for 48 h at 13°C in 0 . 5× TBE buffer at a fixed voltage of 120 V and an angle of 115° with pulse time intervals of 30 s . After PFGE , the chromosomal DNA was depurinated and denatured by incubating the agarose gel in 0 . 25 N HCl and then in alkaline solution ( 0 . 5 M NaOH , 1 . 5% NaCl ) . The DNA was transferred to a charged nylon membrane , Immobilon-NY+ ( Millipore , Billerica , MA ) . DNA probes for each chromosome were obtained by PCR using the primers listed in Table S1 . The Digoxigenin-labeled DNA probes were prepared using the DNA labeling and detection kit ( Roche , Indianapolis , IN ) . Oligonucleotide arrays were produced at the Microarray Core , Institute of Molecular Biology , Academia Sinica , using an Omnigrid 100 arrayer ( Digilab , Holliston , MA ) and the Yeast Genome Array-Ready Oligo Set ( Version 1 . 1 , Operon , Huntsville , AL ) . The printing protocol can be found at the Institute of Molecular Biology Microarray Facility web site ( http://www . imb . sinica . edu . tw/mdarray/methods . html ) . Yeast genomic DNA was extracted using the Qiagen Genomic-Tip 100/G kit ( Qiagen , Valencia , CA ) . For individual-chromosome aCGH , DNA was excised from PFGE gel after EtBr staining and purified using the Geneaid Gel DNA Fragment Extraction kit ( Geneaid , Taiwan ) . The purified chromosomal DNA was further amplified using GenomePlex Whole Genome Amplification Kit ( Sigma-Aldrich , St . Louis , MO ) . Probe preparation and hybridization were performed as described [75] . The array data were analyzed using GeneSpring GX 7 . 3 . 1 ( Agilent , Santa Clara , CA ) . After copper treatment ( 1 mM CuSO4 ) for 2 h at 28°C , total RNA was isolated by Qiagen RNeasy Midi Kit ( Qiagen ) . First-strand cDNA was synthesized for 2 h at 37°C using the High Capacity cDNA Reverse Transcriptase Kit ( Applied Biosystems , Foster City , CA ) . A 20-fold dilution of the reaction products was then subjected to real-time quantitative PCR using gene-specific primers , SYBR Green PCR master mix and an ABI-7000 sequence detection system ( Applied Biosystems ) . Data were analyzed using the built-in analysis program . Fitness of individual strains was obtained by propagating replicate cultures in complete synthetic medium ( CSM ) with or without 1 . 5 mM CuSO4 in 96-well plates inside a temperature controlled , shaking plate reader Infinite F200 ( Tecan , Mannedorf , Switzerland ) . Growth rates were calculated as the maximum slope that could be derived from any continuous 2-hour period during the 20-hr assay . Four replicate cultures were used per strain . To determine competitive relative fitness , we measured the fitness of the experimental strains by competing them against a reference strain expressing PGK1::GFP in YPD media at 28°C . The testing cells and reference cells were inoculated in the YPD medium individually and acclimated for 24 h . The cells were subsequently diluted in fresh media and incubated for another 4 h . The reference and testing cells were then mixed ( 1∶1 ratio ) , diluted into fresh medium at a final cell concentration of 5×103 cells/ml , and allowed to compete for 17 h , which represents about 11 generations of growth . The ratio of the two competitors was quantified at the initial and final time points using a fluorescence activated cell sorter ( FACSCalibur , Becton Dickinson , Franklin Lakes , NJ ) . Four independent replicates for each fitness measurement were performed . We sequenced both wild type and rearranged chromosomes . To prevent cross-contamination between wild type and rearranged chromosomes , wild type chromosomes were purified from EC9-8 haploid cells that do not carry any rearranged chromosome , and rearranged chromosomes were purified from EC9-7 haploid cells that do not carry wild type chromosome 8 ( Figure 3 ) . Individual chromosomal DNA was purified from PFGE gels . Two regions on chromosome 7 ( from 188 , 179 to 194 , 679 bp ) and chromosome 8 ( from 496 , 154 to 502 , 755 bp ) were amplified by PCR using a set of primers . The PCR products were purified and then sequenced . The accession numbers for the sequences are JN835223 and JN835224 . To identify the junction sites of the rearranged chromosomes , we designed primers near each possible breakpoint ( YHR015W , YHR210C , YHL008C , YHR219W , YHL050C , YHR145C , YGL200C and YGL096W ) according to the aCGH data . For each rearranged chromosome , four different combinations of primer pairs were used to PCR the junction site . Only one pair of the primers could successfully amplify the junction site . The PCR products were purified and then sequenced . The accession numbers for the sequence are JX101633 and JX101634 . For the whole-genome gene expression analysis , log-phase cells were grown in YPD with 1 mM CuSO4 ( which does not affect the growth of cup2 null mutant cells ) for 1 h . Total RNA from the treated samples and the corresponding untreated control samples ( in YPD ) was isolated using the Qiagen RNeasy Midi Kit ( Qiagen ) . Probe preparation and hybridization were performed as described [75] . The array data were analyzed using GeneSpring GX 7 . 3 . 1 ( Agilent ) . We excluded the data with hybridization intensities lower than 500 as they were close to the background values ( ∼200 ) . The intensities of each array were normalized using a LOWESS function [76] . EC9-7 haploid cells were streaked out on a YPD plate to form single colonies . Five individual colonies ( A1–A5 ) were then used to initiate the evolution experiment . Cells were cultured in 3 ml YPD with 1 mM CuSO4 ( E1–E5 ) or YPD with 5 mM CuSO4 ( E6–E10 ) through a daily 1000-fold dilution ( about 10 generations ) . Once every five transfers population samples from each line were stored in 20% glycerol at −80°C for later analysis . For diploid evolution experiments , EC9 diploid cells were diluted and plated on an YPD plate to grow to single colonies . Five individual colonies ( AD1–AD5 ) were then used to initiate the evolution experiment ( ED1–ED5 ) with a protocol similar to that of haploid evolution experiments . Soil samples were collected at 7 locations of Evolution Canyon corresponding to the collection sites of the EC yeast strains ( three at the SFS , one at the VB , and three at the NFS ) . Copper contents in soil were measured by inductively coupled plasma-atomic emission spectroscopy ( ICP-AES ) using at least 200 g of individual samples . The array CGH data are available from the NCBI Gene Expression Omnibus ( GEO ) ( http://www . ncbi . nlm . nih . gov/geo/ ) under accession numbers GSE22431 , GSE38034 and GSE33652 . The expression data are available from GEO under accession number GSE31661 . | Large-scale chromosomal rearrangements are often associated with dramatic phenotypic changes such as cancer cell formation . It has been speculated that large-scale chromosomal rearrangements may play a crucial role at the early stages of adaptation , since they can quickly change the expression level of multiple genes or even a whole pathway by changing the gene copy number . Nonetheless , it remains unclear whether such mutations can be stably maintained in populations , especially in a fluctuating environment . Here we characterize an adaptive copper-tolerant phenotype in a wild yeast population . We discovered that the adaptive phenotype was contributed to by two large-scale chromosomal rearrangements , which increased the copy number of key components of copper regulation , including a crucial transcriptional activator , Cup2 . We further identified three previously unknown downstream targets of Cup2 that also contributed to copper tolerance . Finally , we conducted an evolution experiment to test the stability of the rearranged chromosomes under conditions of relaxed selection . We found that the rearranged chromosomes returned back to the original configuration at a high frequency , and the wild type-like chromosome became fixed in all the evolved cultures . Our results suggest that chromosomal rearrangements can provide a reversible mechanism for cells when adapting to a fluctuating environment . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"molecular",
"cell",
"biology",
"genetics",
"biology",
"genomics",
"evolutionary",
"biology",
"microbiology",
"genetics",
"and",
"genomics"
] | 2013 | Dynamic Large-Scale Chromosomal Rearrangements Fuel Rapid Adaptation in Yeast Populations |
Oligodendrocytes are the myelinating glia of the central nervous system and ensure rapid saltatory conduction . Shortage or loss of these cells leads to severe malfunctions as observed in human leukodystrophies and multiple sclerosis , and their replenishment by reprogramming or cell conversion strategies is an important research aim . Using a transgenic approach we increased levels of the transcription factor Sox10 throughout the mouse embryo and thereby prompted Fabp7-positive glial cells in dorsal root ganglia of the peripheral nervous system to convert into cells with oligodendrocyte characteristics including myelin gene expression . These rarely studied and poorly characterized satellite glia did not go through a classic oligodendrocyte precursor cell stage . Instead , Sox10 directly induced key elements of the regulatory network of differentiating oligodendrocytes , including Olig2 , Olig1 , Nkx2 . 2 and Myrf . An upstream enhancer mediated the direct induction of the Olig2 gene . Unlike Sox10 , Olig2 was not capable of generating oligodendrocyte-like cells in dorsal root ganglia . Our findings provide proof-of-concept that Sox10 can convert conducive cells into oligodendrocyte-like cells in vivo and delineates options for future therapeutic strategies .
Transcription factor-mediated reprogramming is currently the method of choice for the generation of induced pluripotent stem ( iPS ) cells [1] . It is also used to directly convert one cell type into another . Successful conversion depends on the choice of transcription factors , but is also influenced by the proteomic constitution of the targeted cell with some cells being more susceptible to acquiring a specific new identity than others [2] . Both reprogramming and conversion are usually performed in culture with low efficiencies and are rarely studied in vivo . Recently , murine fibroblasts have been converted into oligodendrocyte precursor cells ( OPC ) which in turn had the capacity to differentiate into myelinating oligodendrocytes when transplanted into the brain of a myelin-deficient mouse mutant [3 , 4] . This feat is important as generation of oligodendroglial cells from iPS cells is relatively inefficient and time-consuming [5] . Once optimized and adopted to human cells , it offers a potential source for cell replacement strategies in the various demyelinating and dysmyelinating diseases . The conversion to OPC was achieved by applying a cocktail of several transcription factors to fibroblasts . While one study settled on a set of eight transcription factors with the core group consisting of Sox10 , Olig2 and Nkx6 . 2 [3] , the other defined a three-factor mix of Sox10 , Olig2 and Zfp536 [4] . Sox10 and Olig2 thus seem to represent the minimal common denominator for the conversion process . The key role of Sox10 and Olig2 is not unexpected as previous studies had shown the exceptional importance of both transcription factors for oligodendroglial development and myelin formation during embryonic and postnatal development [6 , 7 , 8 , 9 , 10] . Olig2 is largely restricted to oligodendroglial cells . The few other Olig2-expressing cell populations ( i . e . neuroepithelial cells of the ventral ventricular zone , motoneuron precursors and a subset of astrocyte precursors ) are transient and restricted to the embryonic and early postnatal central nervous system ( CNS ) [11] . Sox10 , in contrast , additionally occurs in several other cell types outside the CNS which are mostly neural crest-derived , such as all glial cells of the peripheral nervous system ( PNS ) [12] . When tested as single factors for their ability to induce OPC features in fibroblasts , only Sox10 , but not Olig2 was found to exhibit some activity [4] . An independent study on cultured human neural progenitor cells recently confirmed Sox10 as the principle and rate-limiting determinant of myelinogenic fate [13] . This prompted us to postulate that it might be possible to induce oligodendrocyte properties in vivo in an especially conducive cell type with Sox10 alone . Indeed we found that its overexpression in already Sox10-positive satellite glia of PNS dorsal root ganglia ( DRG ) is sufficient to generate oligodendrocyte-like cells in vivo . The available evidence indicates that a key element in this conversion process is the activation of Olig2 as the second essential oligodendroglial identity factor mediated by a Sox10-responsive evolutionarily conserved enhancer of the Olig2 gene . Interestingly , analogous overexpression of Olig2 is not sufficient to convert satellite glia into oligodendrocyte-like cells . Our findings provide proof-of-concept that Sox10 can be used to convert a conducive cell type into oligodendrocyte-like cells in vivo and delineates options for future therapeutic strategies .
For targeted and strictly controlled Sox10 expression in vivo a transgene was generated in which rat Sox10 cDNA was placed under control of a bidirectional tetracycline-responsive promoter ( Fig . 1A ) . GFP expression from the same promoter and Sox10-tagging with an aminoterminal 9myc epitope were used for detection of transgene expression . Luciferase reporter gene assays in transiently transfected Neuro2A cells confirmed that the aminoterminal 9myc tag did not interfere with the ability of Sox10 to activate a series of its targets , including the promoters of the Mag ( myelin associated glycoprotein ) , Mbp ( myelin basic protein ) , Cx32 ( connexin 32 ) , Cx47 ( connexin 47 ) genes and the intronic oligodendrocyte enhancer of the Plp1 ( proteolipid protein 1 ) gene ( Fig . 1B ) . Of the founders obtained by pronucleus injection of this TetSox10 transgene one was expanded into a line . It contained less than 5 tandem copies of the transgene on the long arm of mouse chromosome 10 ( 10qD1 ) . In this study it was mostly combined with a Rosa26stopflox-tTA [14] and a Sox10::Cre [15] allele to direct transgene expression to all cells that normally express Sox10 during development or in the adult ( Fig . 1A ) . Brain extracts from Rosa26+/stopflox-tTA Sox10::Cre mice contained as much transgenic as wildtype Sox10 at embryonic day ( E ) 18 . 5 when they were hemizygous for TetSox10 , and approximately three times as much when homozygous ( i . e . 2TetSox10 ) ( Fig . 1C , D ) . Transgenic Sox10 expression corresponded to sites of endogenous Sox10 expression ( Fig . 1E , F ) . In spinal cord and other CNS areas , both endogenous and transgenic Sox10 were restricted to and present in the vast majority of Olig2-positive cells of the oligodendroglial lineage ( Fig . 1G , H ) . In the PNS , both DRG and nerves were labelled similarly by an anti-Sox10 antibody in the wildtype and an anti-myc-tag antibody in the transgenic animal ( Fig . 1E , F , I , J ) . However , while endogenous Sox10 was restricted to glial cells as previously shown [16] , transgenic Sox10 was additionally found in a subset of DRG neurons as a relic of their ontogenetic history ( Fig . 1I , J ) . DRG neurons stem from Sox10-positive neural crest precursor cells and therefore experience transient Sox10::Cre expression which triggers induction of the TetSox10 transgene . The continued presence of transgenic Sox10 in cell lineages that normally express the protein only transiently and the resulting developmental defects may be one reason for the very early postnatal death of Sox10::Cre induced , tTA expressing TetSox10 and 2TetSox10 mice . It had previously been shown that Sox10 deletion leads to cell loss and disorganization within the PNS , including a dramatic reduction of DRG size [16] . The CNS is less affected and mainly suffers from absent oligodendroglial differentiation and myelination [8 , 9] . When transgenic Sox10 is expressed homozygously on an otherwise Sox10-deficient background ( i . e . in 2TetSox10 under control of Sox10rtTA in Sox10rtTA/rtTA mice [17] following doxycycline treatment ) , these defects are rescued as indicated by a near normal DRG size and reappearance of myelin gene expression in the spinal cord of compound mutant embryos ( compare Fig . 1K-O to Fig . 1P-T and Fig . 1U-Y ) . This confirms functionality of the 9myc-tagged transgenic Sox10 in vivo . When analysing oligodendroglial development in late embryos that overexpress TetSox10 and 2TetSox10 under control of Sox10::Cre and Rosa26stopflox-tTA we observed an earlier appearance of myelin markers such as Plp1 and Mbp in spinal cord and other CNS regions . This may be indicative of a precocious oligodendrocyte differentiation . More intriguingly , Plp1 and Mbp expressing cells were also detected in substantial numbers in DRG of 2TetSox10 embryos at E18 . 5 , whereas they were rare in DRG of TetSox10 mice and absent from age-matched wildtype embryos ( Fig . 2B-D , F-H ) . Myelin gene expression in DRG of 2TetSox10 embryos went along with the selective presence of Myrf , Olig2 , Olig1 and Nkx2 . 2 ( Fig . 2E , I-L , P-R ) . These transcription factors are strongly associated with oligodendrocytes and oligodendroglial myelination , while absent from Schwann cells , the only cell type normally capable of myelination in the PNS . In contrast , markers of myelinating Schwann cells such as Oct6 and Krox20 transcription factors were not expressed in DRG and remained restricted to the peripheral nerves of 2TetSox10 embryos ( Fig . 2N , O , T , U ) . Dissociated cells from DRG of E14 . 5 2TetSox10 mouse embryos also gave rise to Mbp-positive cells with the typical morphology of myelinating oligodendrocytes at low frequency when co-cultured with rat DRG neurons under myelinating conditions ( Fig . 2W-Y ) . Such cells were not observed when dissociated DRG of E14 . 5 wildtype embryos were used instead ( Fig . 2V ) . We therefore conclude that the myelinating cells in DRG of 2TetSox10 mice closely resemble oligodendrocytes . As we cannot exclude the possibility that these cells retain differences to oligodendrocytes and as it is technically not feasible for us to collect enough of these cells for in-depth expression profiling and characterization , we will refer to them as oligodendrocyte-like cells . Despite the strong expression of oligodendrocyte lineage and differentiation markers , oligodendrocyte precursor cell ( OPC ) markers such as Sox9 , Pdgfra and NG2 were not detected in substantial levels in DRG of 2TetSox10 mice ( e . g . Fig . 2M , S ) . This leads to the conclusion that these oligodendrocyte-like cells have not gone through a classical OPC stage and may be derived from another cell source . A study of consecutive embryonic stages from E11 . 5 to E18 . 5 ( Fig . 3A-H ) revealed that ectopic Olig2 expressing cells in DRG of 2TetSox10 mice are not yet detectable at E11 . 5 , but are already present at E12 . 5 ( Fig . 3E , F ) approximately the same time when OPC start to be generated in the ventral ventricular zone and emigrate from the pMN domain into the marginal zone of the spinal cord ( Fig . 3A , B , E , F ) . This early appearance strongly argues for an origin of Olig2-positive cells in DRG of 2TetSox10 mice that is independent from OPC and outside the CNS . Considering that most of the PNS is neural crest-derived and that Wnt1::Cre is widely active throughout the early neural crest , we exchanged Sox10::Cre for this Cre driver to induce 2TetSox10 expression . Analysis of GFP autofluorescence as well as direct detection of transgenic Sox10 by anti-myc antibodies confirmed the widespread activation and expression of the transgenic construct throughout the embryonic PNS ( Fig . 3I , J ) . It went along with efficient generation of differentiating oligodendrocyte-like cells in DRG as evident from the induced expression of Olig2 , Nkx2 . 2 , Myrf , Plp1 and Mbp ( Fig . 3K-P ) . We therefore conclude that the oligodendrocyte-like cells stem from neural crest-derived cells of the PNS . Because boundary cap cells represent a versatile source for different neural crest-derived cell types in the PNS [18] and have been reported to give rise to oligodendrocytes after engraftment into the CNS [19] , we checked whether these cells were the source of Olig2-positive cells in the DRG . A Krox20::Cre driver in combination with Rosa26stopflox-tTA allows a restricted induction of 2TetSox10 in this transient cell population which is localized at the dorsal root entry zone during early embryonic times ( Fig . 4A , see arrows ) . However , such selective induction of transgenic Sox10 expression did not lead to the appearance of Olig2-positive cells in DRG ( Fig . 4E ) . Furthermore , Olig2-positive cells were never observed in substantial numbers in the dorsal root entry zone of mice in which 2TetSox10 expression was under control of Sox10::Cre at times when they were already numerous in the DRG ( compare Fig . 5A-G to Fig . 5H-M ) . This argues against a boundary cap derived-origin of the ectopic Olig2-positive cells . Brn4::Cre is active throughout the early CNS and in DRG neurons [20 , 21] . When this Cre line was used to activate 2TetSox10 expression ( Fig . 4B , C ) we again failed to observe any Olig2-positive cells in the DRG ( Fig . 4F , G ) . This finding not only provides additional evidence for an origin of the Olig2-positive cells outside the CNS , it also excludes DRG neurons as source . The latter finding is also supported by the fact that there was no co-labelling of Olig2-positive cells with NeuN or Islet1 as markers for PNS neurons in DRG of mice in which 2TetSox10 expression was under control of Sox10::Cre ( Fig . 5O-R , U-X ) . Instead , we observed a substantial overlap of Olig2 and Fabp7 staining in DRG of 2TetSox10 mice at E13 . 5 ( Fig . 5N ) . Considering that Fabp7 is the only reliable early marker for peripheral glia , we conclude that Olig2 induction occurs in glial cells of the DRG . Interestingly , Fabp7 co-staining was only observed in cells with weak , but not with strong Olig2 immunoreactivity arguing that co-expression is transient and restricted to the phase of Olig2 induction . Finally , we employed Dhh::Cre in combination with Rosa26stopflox-tTA to induce 2TetSox10 expression . Dhh::Cre is active in the Schwann cell lineage from the precursor stage onwards . Although we efficiently activated transgene expression in Schwann cells , for instance in spinal nerves in the immediate vicinity of the DRG ( Fig . 4D , see arrowheads ) , no Olig2-positive cells were generated in the DRG itself ( Fig . 4H ) . Considering ( i ) that the Olig2-positive cells are derived from PNS cells other than boundary cap cells , Schwann cells or DRG neurons and ( ii ) that they are glial in origin , satellite glia within the DRG remain as sole source . We thus conclude that overexpression of Sox10 in satellite glia leads to the generation of differentiating and myelinating oligodendrocyte-like cells . This conversion seems specific as we failed to obtain any evidence for a simultaneous generation of astrocytes or spinal cord neurons in DRG upon Sox10 overexpression in 2TetSox10 mice ( Fig . 5 , S , T , Y , Z ) It seemed reasonable to assume that one of the earliest events during the conversion of satellite glia into oligodendrocyte-like cells should be the Sox10-dependent activation of Olig2 as an essential determinant of oligodendroglial identity . During oligodendrocyte specification in the CNS , Olig2 is genetically upstream of Sox10 and appears to be a direct activator of Sox10 gene expression [22 , 23 , 24] . However , it has also been proposed that later on during oligodendrocyte development , Sox10 may in turn help to maintain Olig2 expression [25] . An increase of overall Sox10 levels in satellite glia upon transgene expression could thus be sufficient to activate Olig2 expression and thereby establish a key circuit of the oligodendrocyte regulatory network . To study this hypothesis , we searched for evolutionarily conserved non-coding regions ( ECR ) in the vicinity of the Olig2 gene . One such ECR was recently shown to be active in the early spinal cord , but was not analyzed at times relevant for oligodendrocyte development [26] . This 2 . 7 kb Olig2 ECR ( OLE ) is localized approximately 33 kb upstream of the transcriptional start of the mouse Olig2 gene ( Fig . 6A ) . It furthermore exhibited a robust response to the presence of Sox10 in transiently transfected Neuro2A cells and allowed a 25-fold Sox10-dependent activation of a luciferase reporter gene ( Fig . 6B ) . When split into a more distal ( OLEa ) and a more proximal part ( OLEb ) , OLEa retained Sox10 responsiveness and even elicited an increased activation of the luciferase reporter , whereas OLEb failed to do so arguing that OLEa may contain the core elements for Sox10 induction . In agreement , chromatin immunoprecipitation ( ChIP ) experiments on three-week old mouse brain and oligodendrocytes differentiated in culture for 6 days found a specific enrichment of OLEa in chromatin precipitated with α-Sox10 antibodies ( Fig . 6C , D ) arguing that the effect of Sox10 on OLEa is direct . Bioinformatic analysis of the OLEa sequence revealed the presence of 14 potential Sox binding sites , labelled I through XIV ( Fig . 6E and Fig . 7 ) . In electrophoretic mobility shift assays ( EMSA ) six sites were found to exhibit strong affinity for Sox10 . These were sites I , II , IV , IX , X and XIV ( Fig . 6F-P ) . Site V had a weaker affinity . Sites I and II were closely spaced and allowed binding of a Sox10 dimer . So did sites IX and X , whereas sites IV , V and XIV interacted with a Sox10 monomer ( Fig . 6F , H , I , M , P ) . Each of the sites was mutated in such a way that Sox10 binding was no longer possible ( Fig . 8A , C-R ) and mutations for the high-affinity sites were introduced into the context of OLEa . Luciferase reporter gene assays in transiently transfected Neuro2A cells showed that mutation of the dimer site IX/X had the largest impact on Sox10 responsiveness among the single site mutations and reduced activation rates from 59-fold to 15-fold ( Fig . 8B ) . The remaining activation rates were even further reduced when mutation of site IX/X was combined with additional mutations of the other sites such as site I/II and site IV . These in vitro studies therefore confirm that Sox10 binds and acts through multiple sites in OLEa . To analyse whether the identified ECR is active as an oligodendroglial enhancer in vivo , we used lacZ reporter gene constructs containing the 2 . 7kb OLE or its subfragment OLEa in front of a hsp68 minimal promoter and the reporter gene cassette to generate transgenic mice ( Fig . 9A ) . Five stably transmitting founders were obtained for the OLE-lacZ and the OLEa-lacZ reporter each ( Fig . 9B ) . Despite some variability among the established lines ( Fig . 9B ) , all exhibited staining in the spinal cord that was compatible with predominant expression in cells of the oligodendrocyte lineage ( Fig . 10A-D for OLE-lacZ and Fig . 10E-H for OLEa-lacZ ) . Outside the CNS , there was weak transgene expression in the DRG , sometimes accompanied by faint staining in cartilage or vasculature ( Fig . 9B ) . IHC at E18 . 5 confirmed the predominantly oligodendroglial expression of the transgene as the majority of β-galactosidase expressing cells were positive for Olig2 and Sox10 at perinatal times ( Fig . 10I , K ) . In contrast , only a small fraction of β-galactosidase expressing cells colabelled with NeuN as a neuronal marker ( ≤ 5% ) or glutamine synthetase and GFAP as astrocytic markers ( ≤ 20% ) ( Fig . 10M-P ) . The substantial overlap between β-galactosidase and Mbp furthermore argues that reporter gene expression is not restricted to OPC but also found in differentiating oligodendrocytes ( Fig . 10J , L ) . Nevertheless , there were some differences between OLE-lacZ and OLEa-lacZ lines ( Fig . 10A-H ) . The OLE-lacZ reporter was on average more widely expressed throughout the oligodendroglial population than the OLEa-lacZ reporter . In addition to this better coverage , only OLE , but not OLEa was strongly active in the pMN domain ( compare Fig . 10A , B to Fig . 10E , F ) . This confirms that OLEa contains the key regulatory elements for oligodendroglial activity , but may need to be modified by additional elements present in the larger OLE to faithfully recapitulate the complete developmental expression pattern of Olig2 . We also placed the OLE-lacZ and OLEa-lacZ transgenes on a background in which 2TetSox10 was expressed under Sox10::Cre-induced tTA control and investigated reporter gene activation in the DRG ( Fig . 10Q-T” ) . Both transgenic constructs were strongly activated in a subpopulation of cells within the DRG of 2TetSox10 mice ( Fig . 10R-R” , T-T” ) . In agreement with efficacy of transgene expression in the CNS , induction rates varied between the two transgenes and OLEa-lacZ transgenic animals reproducibly showed a lower amount of lacZ-expressing cells in their DRG than OLE-lacZ transgenic animals . Importantly , lacZ expression was restricted to a subset of the Olig2-expressing cells in the DRG ( Fig . 10R , T ) . This supports the notion that Olig2 expression in DRG of Sox10 overexpressing mice involves the identified Olig2 enhancer . Finally we asked whether the presence of Olig2 in DRG glia is sufficient to induce oligodendrocyte-like cells . For that purpose we exchanged the TetSox10 transgene by an analogously constructed TetOlig2 transgene [27] and probed the DRG of 2TetOlig2 mice at E18 . 5 for the expression of the myelin genes Mbp and Plp1 and Myrf as a marker for differentiating oligodendrocytes . Unlike 2TetSox10 mice , 2TetOlig2 mice were indistinguishable from the wildtype in that oligodendrocyte markers were not expressed ( compare Fig . 11C-E , G-I to Fig . 2C-E , G-I ) . Therefore Olig2 cannot convert satellite glia into oligodendrocyte-like cells .
Using a genetic strategy that allows transgene expression in all cells that normally express Sox10 , we have shown in this study that overexpression of Sox10 in DRG satellite glia is sufficient to directly convert these cells into cells that strongly resemble differentiating oligodendrocytes . Evidence that the reprogrammed cells are oligodendrocyte-like is manifold . These cells express the typical markers and regulatory network components of myelinating oligodendrocytes , including Olig2 , its relative Olig1 , Nkx2 . 2 and Myrf . Additionally we find expression of myelin genes such as Mbp and Plp1 , and when cultured with DRG neurons , some of these cells acquire the typical morphology of myelinating oligodendrocytes . The presence of Plp1 furthermore shows that the cells are not Schwann cells , which express Mpz instead . Similarly , characteristic regulatory network components of myelinating Schwann cells such as Oct6 and Krox20 were missing from these cells . Despite their clear oligodendrocyte character these cells were not of CNS origin as evidenced by the fact that CNS-specific overexpression of Sox10 failed to give rise to these cells . Their early appearance in DRG at E12 . 5 also argues against a CNS origin as it is difficult to imagine that the newly generated OPC could have migrated from the pMN domain all the way through the spinal cord parenchyma into the DRG during this extremely short time window . We also failed to detect substantial levels of Sox9 , Pdgfra and NG2 in reprogrammed cells at any time of their development arguing that these cells did not go through a classic OPC stage . The PNS origin of these oligodendrocyte-like cells was also supported by their appearance after neural crest-wide overexpression of Sox10 . Among neural crest-specific cellular sources for these oligodendrocyte-like cells within the PNS boundary cap cells and immature Schwann cells could as much be ruled out as DRG neurons . Instead , Fabp7-positive resident glia within the DRG were identified as the cells in which Olig2 induction occurred . Fabp7 is to date the only reliable marker for satellite glia during embryonic development [16] . We therefore conclude that the oligodendrocyte-like cells arise from satellite glia . However , we are aware that this assignment is based on a single marker and should be revisited once additional markers for embryonic satellite glia become available . Satellite glia represent a poorly characterized cell population . They are closely apposed to neuronal somata and appear to supply them with nutrients , neurotrophins and other essential molecules . Their intense communication with neurons and strong coupling by gap junctions has led to the assumption that they may be the PNS counterpart to CNS astrocytes [28] . Satellite glia furthermore appear to represent a persistent precursor cell population . They are slowly dividing in the adult and respond to noxious stimuli and inflammation by enhanced proliferation [29] . When taken from their normal environment and placed in culture they have been reported to display plasticity and give rise to various types of PNS and CNS glial cell types [30] . Our finding that satellite glia are prone to reprogramming may thus at least in part be attributable to their precursor cell characteristics and plasticity . To us , the frequency with which satellite glia are converted into oligodendrocyte-like cells upon Sox10 overexpression is particularly noteworthy . With standard in vitro conversion rates ( for review , see ref . 2 ) , we would have had little chance to observe this process in vivo . One reason for this phenomenon may actually be found in the microenvironment of satellite glia , including their close proximity to neurons which may supply instructive signals for oligodendrocyte development . However , it is probably also important that satellite glia already express some amount of endogenous Sox10 . Considering that Sox10 may function as a pre-patterning factor [31 , 32] , its presence may help to keep those chromatin regions in a poised state that need to be activated during the direct conversion of satellite glia into oligodendrocyte-like cells . It is this activity as pre-patterning factor that makes Sox10 especially suitable for reprogramming strategies . A valuable further property of Sox10 may be its capacity to induce many of the factors that it needs to cooperate with during oligodendroglial cell fate decisions and differentiation processes such as Nkx2 . 2 and Myrf [8 , 25] . Equally noteworthy is the fact that reprogramming is achieved by a change of dose rather than introduction of a novel factor . Sox10 amounts are tightly regulated and its functions are concentration-dependent during normal development in mouse and human [16 , 33 , 34 , 35] . One of the results of the increased Sox10 levels in satellite glia is the additional activation of Olig2 as a second essential factor for oligodendrogenesis and oligodendrocyte differentiation . This activation furthermore appears to be direct and mediated by an ECR in the distal upstream region of the Olig2 gene which is not only active in oligodendroglial cells , but also responds to the presence of Sox10 and is bound by this factor in vitro as well as in vivo . The multiplicity of Sox10 binding sites and the complicated structure of a core and adjacent accessory elements makes this ECR an ideal element for a Sox10 dosage-dependent enhancer that normally comes under Sox10 control in oligodendrocytes when amounts of this transcription factor increase with the onset of differentiation [9] , or artificially in satellite glia when Sox10 levels increase by overexpression . This ability of high levels of Sox10 to induce and maintain Olig2 expression is likely a central element in the conversion process as it establishes a key circuit in the corresponding regulatory network . However , our results also indicate that Olig2 induction is not sufficient to convert satellite glia into oligodendrocyte-like cells . This argues that additional Olig2-independent processes are set in motion by high Sox10 levels in satellite glia . The previously reported induction of Myrf expression may be one of them [8] . It is intriguing to assume that other Sox10 expressing neural crest-derived cells with precursor cell characteristics may similarly be convertible into oligodendrocyte-like cells . These include melanocyte stem cells and enteric glia [36] . Especially the latter are similar to DRG satellite glia in their close apposition and functional interaction with neurons , as well as in their maintenance of precursor cell characteristics and plasticity that allows enteric glia to respond to injury with increased proliferation and production of enteric neurons [37] . The presence of melanocyte stem cells and enteric glia in the adult and their relatively easy accessibility may make them amenable to isolation and Sox10-dependent conversion as a realistic source of oligodendrocytes for future applications like cell replacement strategies .
All Plasmids were generated by standard cloning procedures . Expression plasmids for 9myc-tagged Sox10 were based on pCMV5 and pBI-EGFP . Reporter plasmids for transgenic animals were generated by cloning the respective ECR fragments upstream of an Hsp68 minimal promoter followed by a lacZ cassette [38] . For luciferase assays the respective ECR fragments were cloned upstream of a β-globin minimal promoter followed by a luciferase cassette [22] , Sox binding sites were mutated using the QuikChange XL site-directed mutagenesis kit ( Stratagene ) . Mouse Neuro2a neuroblastoma cells and rat primary oligodendroglia were kept in culture as described [39] . Transient transfections of Neuro2a cells , luciferase assays and EMSA followed standard procedures [22] . ChIP was performed as reported [31] with the following modifications: Chromatin was prepared from primary oligodendrocytes kept under differentiating conditions for six days and from brain tissue of three week old mice . Fixation was with 1% formaldehyde in PBS . For precipitation of sheared chromatin , anti-Sox10 antiserum and corresponding preimmune serum were used in combination with protein G magnetic beads ( Cell Signaling Technology ) . A list of primers for cloning and detection of genomic fragments in PCR experiments , including their sequence and position is available upon request . All animal experiments were carried out with permission and in compliance with animal policies of the local authorities and governmental agencies . Mice transgenic for TetSox10 , OLE-lacZ or OLEa-lacZ were obtained by microinjecting the respective linearized DNA into male pronuclei of fertilized oocytes according to standard techniques . Mice transgenic for TetOlig2 have been described [27] . Expression of TetSox10 was achieved by combining one or two copies of the transgene with the Sox10rtTA allele and administration of doxycycline [17] . Alternatively , TetSox10 expression was induced by a combination of the Rosa26stopflox-tTA allele [14] and any of the following Cre alleles: Sox10::Cre [15] , Wnt1::Cre [40] , Krox20::Cre [41] , Brn4::Cre [20] or Dhh::Cre [42] . If not otherwise stated , analysed animals contained two copies of the TetSox10 , and one copy of the Rosa26stopflox-tTA and the Cre allele each . Expression of the TetOlig2 transgene was similarly achieved by combining two copies with one copy of the Rosa26stopflox-tTA and the Sox10::Cre allele . After genotyping , material from staged embryos was processed for X-Gal staining [9] , ISH with probes specific for Mbp , Plp1 and Myrf , or IHC using primary antibodies against Sox10 ( guinea pig antiserum in 1:1000 dilution ) [43] , Sox9 ( guinea pig antiserum in 1:500 dilution ) [44] , Glast ( guinea pig antiserum in 1:500 dilution , Millipore ) , Olig1 ( rabbit antiserum in 1:10000 dilution , Millipore ) , Olig2 ( rabbit antiserum in 1:1000 dilution , Millipore ) , Oct6 ( rabbit antiserum in 1:2000 dilution ) [31] , Fabp7 ( rabbit antiserum in 1:300 dilution , Millipore ) , Krox20 ( rabbit antiserum in 1:200 dilution , Covance ) , Mbp ( rabbit antiserum in 1:200 dilution , NeoMarkers ) , β-galactosidase ( rabbit antiserum in 1:500 dilution , ICN; goat antiserum in 1:500 dilution , Biotrend ) , myc-tag ( goat antiserum in 1:200 dilution , Abcam ) , Nkx2 . 2 ( mouse monoclonal in 1:5000 dilution , Developmental Studies Hybridoma Bank , University of Iowa ) , NeuN ( mouse monoclonal in 1:500 dilution , Millipore ) , Gfap ( mouse monoclonal in 1:100 dilution , Millipore ) , GlnS ( mouse monoclonal in 1:1000 dilution , BD Transduction Laboratories ) , Hb9 ( mouse monoclonal in 1:50 dilution , Developmental Studies Hybridoma Bank ) , Islet1 ( mouse monoclonal in 1:1000 dilution , Developmental Studies Hybridoma Bank ) , and GFP ( rat monoclonal in 1:2000 dilution , Nacalai Tesque ) . Olig1 and Nkx2 . 2 immunoreactivity were detected with the TSA Plus Cyanine 3 system ( PerkinElmer ) . Source and working concentration of fluorophore-labelled secondary antibodies were as described [10 , 21 , 39 , 45] . Nuclei were counterstained with Dapi . Coculture experiments were performed as described [46] with the difference that dissociated DRG cells from wildtype or 2TetSox10 mice were added instead of OPC to cultured rat DRG neurons . To this aim , DRG were dissected from E14 . 5 mouse embroys and dissociated with papain ( 4 U/ml ) , DnaseI ( 40 μg/ml ) and L-cysteine ( 240 μg/ml ) at 37°C for 60 min . The dissociated cells were added at a density of 150 , 000 per well in a 12-well plate and incubated under myelinating conditions for 4 weeks . After fixation , cells were stained with antibodies directed against Mbp ( rat monoclonal in 1:750 dilution , Serotec ) and Nf165 ( mouse monoclonal in 1:3000 dilution , Developmental Studies Hybridoma Bank ) . For western blots , brain extracts were prepared as described [47] , and proteins were detected with antibodies against Sox10 and Gapdh ( Santa Cruz Biotechnology ) . Mice experiments were in accord with animal welfare laws and approved by the responsible local committees and government bodies ( Regierung von Mittelfranken and Behörde für Gesundheit und Verbraucherschutz Hamburg ) . | Developmental or acquired defects of oligodendrocytes or their myelin sheaths impairs saltatory nerve conduction in the central nervous system and thus leads to severe neurological diseases . Strategies to regenerate or replace these cells require a deeper understanding of the regulatory processes that underlie their generation during development . Here we show in a Sox10 overexpressing mouse model that increase of the levels of a single transcription factor during embryogenesis efficiently converts the already Sox10 expressing satellite glial cells of the peripheral nervous system into oligodendrocyte-like cells by a mechanism that does not simply recapitulate developmental oligodendrogenesis but involves direct Sox10-dependent induction of the oligodendroglial differentiation network . Our study identifies mechanisms that may help to convert other cell types into oligodendrocytes and thus prove eventually useful for therapies of myelin diseases . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2015 | Elevated In Vivo Levels of a Single Transcription Factor Directly Convert Satellite Glia into Oligodendrocyte-like Cells |
Rare genetic variants , identified by in-detail resequencing of loci , may contribute to complex traits . We used the apolipoprotein A-I gene ( APOA1 ) , a major high-density lipoprotein ( HDL ) gene , and population-based resequencing to determine the spectrum of genetic variants , the phenotypic characteristics of these variants , and how these results compared with results based on resequencing only the extremes of the apolipoprotein A-I ( apoA-I ) distribution . First , we resequenced APOA1 in 10 , 330 population-based participants in the Copenhagen City Heart Study . The spectrum and distribution of genetic variants was determined as a function of the number of individuals resequenced . Second , apoA-I and HDL cholesterol phenotypes were determined for nonsynonymous ( NS ) and synonymous ( S ) variants and were validated in the Copenhagen General Population Study ( n = 45 , 239 ) . Third , observed phenotypes were compared with those predicted using an extreme phenotype approach based on the apoA-I distribution . Our results are as follows: First , population-based resequencing of APOA1 identified 40 variants of which only 7 ( 18% ) had minor allele frequencies >1% , and most were exceedingly rare . Second , 0 . 27% of individuals in the general population were heterozygous for NS variants which were associated with substantial reductions in apoA-I ( up to 39 mg/dL ) and/or HDL cholesterol ( up to 0 . 9 mmol/L ) and , surprisingly , 0 . 41% were heterozygous for variants predisposing to amyloidosis . NS variants associated with a hazard ratio of 1 . 72 ( 1 . 09–2 . 70 ) for myocardial infarction ( MI ) , largely driven by A164S , a variant not associated with apoA-I or HDL cholesterol levels . Third , using the extreme apoA-I phenotype approach , NS variants correctly predicted the apoA-I phenotype observed in the population-based resequencing . However , using the extreme approach , between 79% ( screening 0–1st percentile ) and 21% ( screening 0–20th percentile ) of all variants were not identified; among these were variants previously associated with amyloidosis . Population-based resequencing of APOA1 identified a majority of rare NS variants associated with reduced apoA-1 and HDL cholesterol levels and/or predisposing to amyloidosis . In addition , NS variants associated with increased risk of MI .
Genome-wide association studies have identified multiple loci associated with complex traits and diseases , but until now common genetic variants ( minor allele frequency >5% ) at these loci only explain small proportions of the heritability [1] , [2] . For example , the estimated heritability of high density lipoprotein ( HDL ) cholesterol in twin-studies is 50% [3] , but the common alleles together or in combination explain less than 5–10% of the variation in plasma levels of HDL cholesterol [4] . Rare genetic variants ( minor allele frequency <1% ) , which are identified by in-detail screening or resequencing of loci , may contribute to unravel this unexplained heritability [1] , [2] . Apolipoprotein A-I ( apoA-I ) is the major protein component of HDL in plasma , and is a cofactor for lecithin∶cholesterol acyltransferase ( LCAT ) , playing a key role in the so-called reverse cholesterol transport , i . e . the transport of cholesterol from peripheral tissues to the liver for excretion [3] . APOA1 ( MIM 107680 ) encodes a 267 amino acid prepropeptide , which is sequentially cleaved to yield the mature 243 amino acid protein . Mutations in apoA-I may associate with low levels of plasma HDL cholesterol and apoA-I due to defective LCAT activation or to amyloidosis , or to amyloidosis with only minor or no effects on apoA-I and HDL cholesterol levels [5]–[11] . However , at present we lack comprehensive information on the spectrum of genetic variants in this pleiotropic gene in the general population , on the phenotypic characteristics of such variants in individuals in the general population , and whether additional information is gained from resequencing a sample of the entire general population , rather than using an extreme phenotype approach , previously used by us and others [12]–[16] . In the first part of this study , the aim was to determine the spectrum and distribution of genetic variants in APOA1 using a population-based resequencing approach . In the second part of the study , the aim was to determine the association of nonsynonymous ( NS ) and synonymous ( S ) variants in APOA1 in the general population with plasma levels of apoA-I and HDL cholesterol , and with risk of myocardial infarction ( MI ) . In the third part of the study , we compared results using an extreme apoA-I phenotype approach with results from the population-based resequencing . For these purposes , we resequenced APOA1 in 10 , 330 participants in the Copenhagen City Heart Study ( CCHS ) , and used the Copenhagen General Population Study ( CGPS; n = 45 , 239 participants ) to validate phenotypic results .
Studies were approved by institutional review boards and Danish ethical committees ( Nos . KF-100 . 2039/91 , KF-01-144/01 , H-KF-01-144/01 ) and conducted according to the Declaration of Helsinki . Informed consent was obtained from all participants . All participants were white and of Danish descent . No participants appeared in more than one of the two studies , permitting independent confirmation of the findings in each group . In both studies of the general population , diagnoses of MI ( WHO International Classification of Diseases; ICD8:410 , ICD10:I21-I22 ) were collected from 1977 through May 10th 2011 , and verified by reviewing all hospital admissions and diagnoses entered in the National Danish Patient Registry , and all causes of death entered in the National Danish Causes of Death Registry . A diagnosis of MI followed the changing definitions over time [20] , [21] . As shown in flowchart ( Figure 1 ) . We screened the translated region of APOA1 in all 10 , 330 participants in the CCHS using four PCR fragments covering 119 bp upstream of exon 2 , exons 2–4 , and exon-intron boundaries ( APOA1 consensus sequence NC_000011 . 9 ) ( Table S1 ) . Mutational analysis was performed using a LightScanner ( Idaho Technology Inc . , Salt Lake City , UT , USA ) , followed by sequencing on an ABI 3730 DNA Analyzer ( Applied Biosystems Inc . , Foster City , CA , USA ) . NS and S variants identified in more than two individuals in the CCHS ( K12K , S25S , S36A , F71Y , K107del , L144R , A164S , A190A ) , were genotyped in the CGPS using an ABI PRISM 7900HT Sequence Detection System ( Applied Biosystems Inc . , Foster City , CA , USA ) and TaqMan-based assays . Colorimetric and turbidimetric assays were used to measure nonfasting plasma levels of apoA-I , HDL cholesterol , total cholesterol , triglycerides , and apoB ( Boehringer Mannheim , Mannheim , Germany and Konelab , Helsinki , Finland ) . LDL cholesterol was calculated using the Friedewald equation [22] when plasma triglycerides were ≤4 . 0 mmol/L , and otherwise measured directly ( Thermo Fisher Scientific , Waltham , MA , USA ) . Body mass index was measured weight ( kg ) divided by measured height squared ( m2 ) . Lipid-lowering therapy was self-reported . Physical inactivity , drinking , smoking , hypertension and diabetes were dichotomized and defined as physical inactivity ( less than 2–4 hours per week of light physical activity at leisure time ) , drinking ( more than 1 drink per week ) , current smoking , hypertension ( systolic blood pressure ≥140 mmHg or diastolic blood pressure ≥90 mmHg , and/or use of antihypertensive therapy ) , and diabetes ( self-reported disease , current use of anti-diabetic medication , and/or nonfasting plasma glucose >11 . 0 mmol/L ) . We used Stata/S . E . 10 . 1 . Two-sided p<0 . 05 was significant . χ2-tests evaluated Hardy-Weinberg equilibrium . To adjust for the effect of gender and age on absolute levels within studies , and for differences in absolute levels between studies , plasma levels of apoA-I and HDL cholesterol were converted to percentiles by gender and by age ( in 10-year age groups ) within each study , allowing for direct comparisons between percentiles for mutations both within and between studies ( CCHS and CGPS ) . Mean apoA-I and HDL cholesterol percentiles in individuals with a specific mutation were compared with the mean percentile ( = 50th percentile of the normalized distribution ) within the CCHS or CGPS as a whole using a z-test [23] . Mann-Whitney U-test and Fisher's exact test compared , respectively , continuous and categorical variables between heterozygotes for different mutations and noncarriers . Number of NS and S variants identified among individuals with extremely low or high plasma levels of apoA-I were compared using Fisher's exact test . Risk of MI for heterozygotes for all NS ( S36A , F71Y , K107del , L144R , A164S ) and for all S ( K12K , S25S , A190A ) mutations genotyped in both studies , was determined prospectively in the CCHS and CGPS combined , using Cox proportional hazards regression models with age as time scale and delayed entry ( left truncation ) in 1977 . Fify-one individuals with a previous MI were excluded from the risk analyses . Hazard ratios were adjusted for age and sex , or multifactorially for age , sex , diabetes , hypertension and smoking .
Resequencing APOA1 in the CCHS identified a total of 40 genetic variants ( Figure 2 and Table S2 ) . Only seven variants ( 18% ) had MAFs >1% ( all in Hardy-Weinberg equilibrium , P-values: 0 . 12 to 0 . 82 ) . Of these , none were in the coding region of APOA1 , that is in exons 1–4 coding for the 267 amino acid prepropeptide . The cumulative number of genetic variants in APOA1 as a function of the number of individuals resequenced showed that by resequencing fewer than 100 individuals all seven variants in APOA1 with a MAF >1% were identified ( Figure 2 ) . In contrast , the number of genetic variants identified with a 0 . 005%<MAF<0 . 1% increased almost logarithmically reaching a plateau around 5 , 000 individuals , while the number of singletons ( minor allele frequency = 0 . 005% ) increased almost linearly . In agreement with this , the number of different genetic variants identified in APOA1 as a function of the number of individuals with each variant , corresponding to the MAF , showed that seventeen of 40 variants ( 43% ) were exceedingly rare and identified in only one or two individuals in the CCHS ( Figure 3 ) .
Using population-based resequencing of APOA1 in 10 , 330 individuals allowed description of the spectrum and distribution of genetic variants in this gene . Our results showed that the vast majority of variants , including variants associated with apoA-I and HDL cholesterol phenotype , were individually rare , though collectively relatively common . These results are in complete agreement with results from two previous population-based resequencing studies of three other genes affecting , respectively , triglycerides and diabetes related traits [25] , [26] . Novel findings , compared to previous population-based screenings of apoA-I using isoelectric focusing [27] , [28] , include: First , the number of NS variants identified and the number of heterozygotes for these variants were , respectively , 5–6 fold and 10–20 fold increased . Second , we showed that 0 . 27% of the population were heterozygous for variants associated with substantial reductions in apoA-I and HDL cholesterol levels , and 0 . 41% were heterozygous for variants previously associated with amyloidosis , although none had been diagnosed with this disease . In addition , S variants , not identified by isoelectric focusing , also associated with reductions in apoA-I levels in the CCHS and CGPS combined ( n>55 , 000 ) . Third , heterozygosity for NS variants in APOA1 associated with a 2-fold increased risk of MI , largely driven by A164S , a variant not associated with apoA-I and HDL cholesterol levels . Finally , while these rare variants might have some effects on the extremes of the population distribution of apoA-I and HDL cholesterol and on levels in the individual , the contribution of both rare and common variants in APOA1 to the total variation in plasma levels of apoA-1 and HDL cholesterol were very modest , respectively , 0 . 03% and 0 . 3% , in agreement with the very large number of genes found to associate with apoA-I and/or HDL cholesterol levels in genomewide association studies [29] . The effect of common variants on plasma levels were 10-fold higher than for rare variants , suggesting that rare variants in this gene do not contribute in any major way to the missing heritability on a population level . An advantage of population-based resequencing is that the genetic variants identified can be tested against multiple phenotypes . This becomes especially important , if the gene under study has pleiotropic effects , i . e . affects multiple phenotypic traits , as is the case for APOA1: mutations in APOA1 have been associated with an inability to activate LCAT and with hereditary amyloidosis [5]–[7] . While mutations that poorly activate LCAT associate with low apoA-I and HDL cholesterol levels due to the rapid removal of lipid-poor discoidal HDL from the circulation [30] , mutations that cause amyloidosis may [5] , [6] or may not [9]–[11] associate with low apoA-I and/or HDL cholesterol , most likely depending on the severity of the mutation [8] . Of the seventeen nonsynonymous variants identified in the present study , only seven have been reported by others ( P4R , V11X , S36A , A37T , F71Y , K107del , L144R ) [11] , [13] , [24] , [27] , [31]–[41] . Of these variants , five ( P4R , V11X , S36A , K107del , L144R ) associated with low apoA-I and/or HDL cholesterol levels in the CCHS and in other studies [13] , [32]–[34] , [40] , [41] . L144R is unable to activate LCAT [13] , while S36A , F71Y , K107del , have been associated with amyloidosis [24] , [38] . A164S , a variant without any association with HDL cholesterol or apoA-I levels in the CCHS and CGPS , was associated with an increased risk of IHD , MI , and premature death , and with reduced survival after diagnosis of IHD in the CCHS , most likely due to an attenuated form of cardiac amyloidosis [13] . Thus , four variants in APOA1 identified in 42 individuals ( 0 . 41% of the population ) have either previously been associated with or suspected of causing amyloidosis . Only two of these variants , S36A and K107del , associated with low HDL cholesterol in the CCHS . This highlights two points: 1 ) The inherent weakness of the extreme approach when a gene has pleiotropic effects . Using this approach in the CCHS , both F71Y , a known amyloidosis mutation [24] , and A164S , a suspected amyloidosis mutation associated with increased risk of ischemic heart disease and early mortality [13] , would have been assumed to be nonfunctional; 2 ) That variants in APOA1 associated with amyloidosis are relatively common in the general population . Comparing the results from the population-based resequencing approach with the results obtained using only the extremes of the population distribution of apoA-I in the CCHS showed that NS variants were overrepresented in the lower percentiles of the apoA-I distribution , especially those predicted in silico to be more pathological , and correctly predicted association with low apoA-I and/or HDL cholesterol levels . Thus , 0 . 27% of the total population were heterozygous for one of nine different variants associated with substantial reductions in apoA-I and/or HDL cholesterol levels of up to , respectively , 39 mg/dL and 0 . 9 mmol/L . As previously shown in the CCHS [13] and validated in the present study including the CGPS , NS mutations in APOA1 may associate with increased risk of MI , without associating with reduced apoA-1 and HDL cholesterol levels . The most likley explanation for this is that these mutations represent less severe amyloidosis mutations manifesting clinically as MI , instead of the more severe restrictive cardiomyopathy [42] . Accordingly , we found that the combined NS mutations associated with an increased risk of MI , the main contributor to this increased risk was A164S , a presumed amyloidogenic mutation not associated with apoA-1 or HDL cholesterol levels in either the CCHS or the CGPS . Our study suggests that the extreme phenotype approach used in a number of previous studies [12]–[16] is a powerful analytical strategy to capture the effects of both common and rare genetic variants on a specific a priori specified complex trait , provided the gene in question does not have pleiotropic effects . However , the success of this strategy depends on using a gene-near phenotype , preferably the direct gene product , on the size of the underlying study , the cut-off point for the extreme percentiles screened , and the frequencies and effect sizes of the identified variants . In conclusion , population-based resequencing of APOA1 identified a majority of rare NS variants associated with reduced apoA-I and HDL cholesterol levels and/or predisposing to amyloidosis . In addition , NS variants associated with increased risk of MI . | Rare genetic variants , identified by in-detail resequencing of loci , may contribute to complex traits . We used the apolipoprotein A-I gene ( APOA1 ) , a major high-density lipoprotein ( HDL ) gene , and population-based resequencing to determine the spectrum of genetic variants , the phenotypic characteristics of these variants , and how these results compared with results based on resequencing only the extremes of the apolipoprotein A-I ( apoA-I ) distribution . By resequencing APOA1 in >10 , 000 Danes and genotyping an additional >45 , 000 , we show that population-based resequencing of APOA1 identifies a majority of rare genetic variants that together are relatively frequent: 0 . 27% of the population are heterozygous for nonsynonymous ( NS ) variants in APOA1 that associate with substantial reductions in apoA-I and HDL cholesterol , and 0 . 41% are heterozygous for variants predisposing to amyloidosis . NS variants associated with a hazard ratio of 1 . 72 ( 1 . 09–2 . 70 ) for myocardial infarction ( MI ) , largely driven by A164S , a variant not associated with apoA-I or HDL cholesterol levels . Resequencing only the extremes of the apoA-I distribution , between 79% and 21% of all variants are not identified; among these are variants previously associated with amyloidosis . These results provide direct evidence that rare NS variants in APOA1 contribute to low apoA-I and HDL cholesterol levels , to susceptibility to amyloidosis , and to risk of MI in the general population . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"quantitative",
"traits",
"lipids",
"proteins",
"lipoproteins",
"biology",
"biochemistry",
"phenotypes",
"heredity",
"genetics",
"apolipoprotein",
"genes",
"genetics",
"of",
"disease",
"genetics",
"and",
"genomics",
"complex",
"traits"
] | 2012 | Population-Based Resequencing of APOA1 in 10,330 Individuals: Spectrum of Genetic Variation, Phenotype, and Comparison with Extreme Phenotype Approach |
Debate over repealing the ivory trade ban dominates conferences of the Convention on International Trade in Endangered Species of Wild Fauna and Flora ( CITES ) . Resolving this controversy requires accurate estimates of elephant population trends and rates of illegal killing . Most African savannah elephant populations are well known; however , the status of forest elephants , perhaps a distinct species , in the vast Congo Basin is unclear . We assessed population status and incidence of poaching from line-transect and reconnaissance surveys conducted on foot in sites throughout the Congo Basin . Results indicate that the abundance and range of forest elephants are threatened from poaching that is most intense close to roads . The probability of elephant presence increased with distance to roads , whereas that of human signs declined . At all distances from roads , the probability of elephant occurrence was always higher inside , compared to outside , protected areas , whereas that of humans was always lower . Inside protected areas , forest elephant density was correlated with the size of remote forest core , but not with size of protected area . Forest elephants must be prioritised in elephant management planning at the continental scale .
Between 1970 and 1989 , half of Africa's elephants ( Loxodonta africana ) , perhaps 700 , 000 individuals , were killed , mostly to supply the international ivory trade [1] . This catastrophic decline prompted the Conference of the Parties ( CoP ) to the Convention on the International Trade in Endangered Species of Wild Flora and Fauna ( CITES ) to list African elephants on Appendix I of the convention , banning the international ivory trade . Today , opinions on the management of African elephants , including their international trade status , are polarized among range states , economists , and wildlife managers [2] . Southern African nations and wildlife managers argue that their ability to control poaching and manage elephants should be rewarded through the harvest and sale of their ivory stocks , thereby generating revenue for conservation programmes . A strong lobby headed by Kenya , the Central and West African nations , and conservationists in these regions maintain that re-opening the trade will increase the demand for ivory and stimulate the resumption of uncontrollable illegal killing of elephants throughout the continent . Among economists , conclusions are equivocal on whether resumption of the trade will have a positive or negative impact on elephant populations [3 , 4] . Central to an informed resolution of this debate is a clear understanding of the size and trends in elephant populations and rates of illegal killing for ivory across Africa . The status of savannah elephant ( L . africana africana ) populations in Eastern , Western , and Southern Africa are relatively well known , and most appear to be stable or increasing with generally low poaching rates [5] , though in Angola , Mozambique , and Zimbabwe , poaching for ivory may be on the increase [6] . The status of forest elephants ( L . africana cyclotis ) in the vast equatorial forest of Africa remains poorly known because methodological problems and severe logistical constraints have inhibited reliable population surveys and estimates of illegal killing [7] . In African savannahs , both elephant populations and illegal killing can be monitored through aerial surveys [8] , whereas an elephant massacre can remain undetected in the depths of the forest . The forest of Central Africa is of critical importance for elephants , comprising over 23% of the total continental elephant range , and the largest contiguous elephant habitat left on the continent [5] . In 1989 , following reconnaissance surveys on foot , the forest elephant population of the Congo Basin was estimated at 172 , 400 individuals , nearly one third of Africa's elephants at that time [9] . Poaching was rampant in some areas , notably the Democratic Republic of Congo [10] ( then Zaire ) , whereas Gabon's elephants were relatively unaffected [11] . Human activity , particularly road infrastructure , was found to be the major factor influencing the distribution of forest elephants [9 , 12 , 13] . Since 1989 , no further region-wide surveys have been conducted , despite dramatic increases in logging , road infrastructure development , growing human populations , and conflict [14–16] , accompanied by considerable development of the protected areas network and conservation funding [17] . Today , forest elephant population estimates are based on guesswork [5] , and inventory and monitoring must be improved for five main reasons: ( 1 ) forest elephants may still comprise a significant proportion of Africa's total elephant population [5]; ( 2 ) forest elephants are distinctive on morphological , ecological , behavioural , and genetic criteria , constituting at least a subspecies and possibly a distinct species of African elephant [18]; ( 3 ) Central Africa's forests are the source of much of the world's illicitly traded ivory [19]; ( 4 ) the trade status of ivory from Southern African elephants may have a serious impact on poaching levels in Central Africa due to changes in the dynamics of the international legal and illegal ivory trade [2]; and ( 5 ) logging and road development in the Congo Basin are increasing dramatically , which is opening up accessibility both to remaining elephant strongholds and to markets . During 2003–2005 , under the auspices of the Monitoring of the Illegal Killing of Elephants ( MIKE ) programme and the Projet Espèces Phares of the European Union , we collected data on the distribution , abundance , and illegal killing of forest elephants by means of systematic foot surveys on line transects and reconnaissance walks ( see Materials and Methods ) at six sites ( Figure 1 ) . These MIKE survey sites were centred on protected areas thought to contain nationally important forest elephant populations . We also collected complementary data in 1999 and 2000 on a single , continuous survey of over 2 , 000 km dubbed the “Megatransect” [20] , which ran through some of the most remote forest blocks in Africa ( Figure 1 ) . Our goals were to evaluate the conservation status of forest elephants , including population size , distribution , and levels of illegal killing in relation to human activity , isolation from roads , and the impact of protected areas .
Our results indicate that a combination of illegal killing and other human disturbance has had a profound impact on forest elephant abundance and distribution , including inside national parks ( NPs ) . The density of elephants in NPs surveyed varied over two orders of magnitude . In the Salonga NP , a remote United Nations Educational , Scientific , and Cultural Organization ( UNESCO ) World Heritage site , as few as 1 , 900 forest elephants remain at a mean density of 0 . 05 elephant km−2 . Salonga is the largest forested NP in Africa and the second largest on earth . In Nouabalé-Ndoki and Dzanga-Sangha NPs and their buffer zones ( Ndoki-Dzanga MIKE site ) , 3 , 900 elephants were estimated within a survey area of 10 , 375 km2 ( 0 . 4 elephant km−2 ) . Mean estimated forest elephant densities in the three NP sectors at this site were 0 . 66 , 0 . 65 , and 0 . 56 individuals km−2 for Nouabalé-Ndoki NP , Dzanga NP , and Ndoki NP respectively , compared with densities of 0 . 14 and 0 . 1 individuals km−2 in the peripheral zones of these NPs . In the 2 , 382-km2 Boumba Bek NP in southeast Cameroon , an estimated 318 elephants occurred ( 0 . 1 elephant km−2 ) . In the Bangassou Forest , one of only two regions in the Central African Republic ( CAR ) that still contain forest elephants , a formal estimate of elephant abundance was not made , but systematic observations along reconnaissance walks suggest that in the 12 , 000-km2 survey area , fewer than 1 , 000 forest elephants remain . In only two protected areas , Minkébé NP , northeast Gabon , and Odzala-Koukoua NP , northern Congo , did the mean estimated elephant density exceed 1 . 0 individual km−2 . Estimated population size was 22 , 000 individuals in the 7 , 592-km2 Minkébé NP ( 2 . 9 elephants km−2 ) and 14 , 000 in the 13 , 545-km2 Odzala-Koukoua NP ( 1 . 0 elephant km−2 ) . Poached elephant carcasses were found in all MIKE sites , even large , well-established NPs ( Table 1 ) . We found 53 confirmed elephant poaching camps and 41 elephant carcasses from 4 , 477 km of reconnaissance walks; we confirmed 27 carcasses as having been poached . Poached carcass encounter rate was highest in the Minkébé site , at 13 . 7 carcasses 1 , 000 km−1 , followed by Ndoki-Dzanga with 7 . 1 carcasses 1 , 000 km−1 . The tusks had been removed from all poached carcasses , though due to the level of decay , it was not possible to determine whether they had been poached primarily for ivory or for meat . Logistic regression [21] using the pooled elephant dung-count data from the Ndoki-Dzanga , Boumba Bek , Salonga , and Odzala-Koukoua surveys indicated a significant positive relationship between the probability of presence of elephants and increasing distance from the nearest major road ( Figure 2A ) . The data for Minkébé were omitted from this analysis because , unique to this site , forest elephant dung was recorded on all transects regardless of the distance from a road , and therefore the data were not informative for logistic regression . Model results were improved by including site as a factor covariate . The exceptions were Ndoki-Dzanga and Odzala that not only had the same slope , but also the same intercept term . Odzala-Koukoua and Ndoki-Dzanga consistently had the highest probability of elephant occurrence at all distances from the nearest road , with intermediate probability for Boumba Bek . Salonga , where elephant dung was recorded on just 22 out of 130 line transects , had the lowest probability of elephant occurrence ( see Figure 2A ) . Performing separate logistic regression analyses on each site's data confirmed the relationship between the probability of elephant occurrence and the distance from the nearest road , except for the Salonga site ( see Figure 3 ) , in which distance from the nearest road had no effect on the probability of elephant dung occurrence . Using the human-sign data pooled across the same MIKE survey sites , but this time including Minkébé , we found that the probability of human presence decreased with increasing distance from the nearest road , in contrast to the probability of elephant occurrence ( Figure 2B ) . However , the probability of human presence was not as dissimilar between the five sites as was the probability of elephant occurrence . In this case , Ndoki-Dzanga and Odzala were the most dissimilar , having the highest and lowest probability of human presence at all distances from the nearest road , respectively . Minkébé , Salonga , and Boumba Bek occupied the middle ground in terms of the probability of human presence and were not significantly dissimilar from one another . Like human sign , the encounter rate of poached elephant carcasses decreased with distance from the nearest road ( Spearman correlation coefficient ρ = −0 . 663 , n = 13 , p = 0 . 014 ) , and no poached carcasses were found beyond 45 km of a road . Generalized Additive Models [22] provide a flexible , non-parametric technique for modelling the extreme variation in the elephant dung counts . Conditioning on elephant presence , the results indicate a significant positive relationship between elephant density and distance from roads . However , including the site covariate in addition dramatically increased the deviance explained from 22 . 5% to 95 . 4% and reduced the Generalized Cross Validation ( GCV ) score [23] ( which is equivalent to Akaike's Information Criterion ) , from 14 . 734 to 6 . 742 . Figure 4 illustrates the estimated conditional dependence of elephant dung-pile numbers on distance from road . The significant difference between the MIKE sites highlighted by the site covariate indicates that there are site-specific ecological influences or additional local human pressures not captured by distance to the nearest major road . The scale of the Megatransect transcended site-level surveys and thus provided a useful extensive comparison to the more intensive , but localised , MIKE surveys . The Megatransect also traversed six protected areas , which allowed the effect of protected area status on forest elephants and human presence to be examined . Applying logistic regression [21] to the Megatransect data indicated a significant relationship between the probability of presence of elephants and the distance from the nearest road ( Figure 5A ) , consistent with the analysis of the MIKE dataset . Model results were not improved by including distance to the nearest protected area boundary as a covariate , but they were significantly improved by including a binary factor covariate describing whether or not the count data were collected within or outside of a protected area . Although the pattern of response of the probability of elephant occurrence to increasing distance from road is similar for within and outside of protected areas , protected areas consistently had the highest probability of elephant occurrence at all distances from the nearest road ( Figure 5A ) . Consistent with MIKE survey data , the probability of human presence on the Megatransect decreased significantly with increasing distance from the nearest road in contrast to the probability of elephant occurrence , and was consistently lower inside protected areas compared to outside for all distances from the nearest road ( Figure 5B ) . Generalized Additive Models [22] were applied to the elephant dung counts from the Megatransect while conditioning on elephant presence . The results indicate a significant relationship between elephant dung counts and both distance from roads and distance to protected areas . However , in contrast to the model fit to the MIKE data , this model is only able to explain 19 . 7% of the deviance . Figure 6 illustrates the estimated conditional dependence of elephant dung-pile numbers on distance from road ( Figure 6A ) and distance to protected areas ( Figure 6B ) that shows a positive relationship with increasing distance from roads and a negative relationship for increasing distance from protected areas .
Our surveys confirmed the observations of conservationists [24] that numbers and range of forest elephant populations are in decline and that they continue to be poached for ivory , and probably meat , including inside NPs . In common with previous work in the Congo Basin [13] , distance from the nearest road was a strong predictor of forest elephant abundance , human presence , and levels of poaching . Within the consistent pattern of increasing elephant abundance and decreasing human-sign frequency with increasing distance from roads , site-level differences were variable and informative . Minkébé was the only site in which elephant dung was recorded on all transects . For other sites , the probability of occurrence decreased in the order Odzala-Koukoua , Ndoki-Dzanga , Boumba Bek , and finally Salonga . Elephant density by NP decreased in the same order , which is consistent with the remoteness of sites from the nearest road ( Figure 7 ) . Total NP area was not correlated with elephant density; however , there was a significant positive correlation between the area of parks that was over 40 km from a road and mean elephant density ( ρ = 0 . 9 , n = 5 , p = 0 . 037 ) . Thus , although Salonga NP is close to three times bigger than any other park surveyed , it comprises two separate sectors with some 46% of the total surface area within 10 km of a road , and nowhere in the park is beyond 40 km from a road . By contrast , just 0 . 7% of the Minkébé NP is within 10 km of a road , and a full 59% is more than 40 km from a road . Only in Minkébé and Odzala-Koukoua NPs do areas exist that are more than 60 km from the nearest road . It is noteworthy that the road system of Salonga NP , which was well developed during colonial and immediately post-colonial times , has gradually fallen into disrepair , and today , the roads are used primarily as footpaths . In all other MIKE sites surveyed , the closest roads to the site are open to regular vehicular traffic , and many have been opened within only the last 10–20 y . Salonga has , therefore , a longer history of penetration by roads than other sites , which may be reflected , not only in the dearth of elephants , but the distribution of human signs , which were more likely to occur further from roads rather than closer to them . The long-term accessibility to the forest and heavy hunting in Salonga , including hunting for elephants [10] , appears to have extirpated wildlife close to roads , forcing hunters to become more active in the most-remote areas of the park . Several navigable rivers also run through Salonga NP , which provide access and may confound an effect of roads as a proxy for isolation . The trends observed in the other MIKE sites ( Figure 3 ) indicate that they have not yet reached such an advanced state of degradation as Salonga because strong relationships still exist between elephant abundance , human-sign frequency , and distance from the nearest road . Elephants still occur in moderate to high densities in remote areas , and at an exceptional density in Minkébé . However , it is clear that elephants are being concentrated into the most-remote sectors of all sites in a near-perfect juxtaposition with the distribution of human activity as exemplified by the simple interpolations of human-sign and elephant dung frequency from Ndoki-Dzanga ( Figure 8 ) . This startling image is reminiscent of Parker and Graham's description of savannah elephant distribution as the “negative” of human density [25] , which was identified as a major factor in the decline of the elephant in Eastern Africa . Without effective management intervention to reduce fragmentation of remote forests [26] , the human–elephant interface will move deeper into the forest , and elephants will continue to retreat into an increasingly less-remote core in the face of an advancing “human front . ” It is important to remember that the MIKE sites likely represent the “best-case” conservation status scenario because they were deliberately chosen from among the longest-established protected areas in some of the most-remote locations in Central Africa . Landscape-level conservation plans , which include conservation measures to reduce hunting and trafficking of bushmeat along roads , have been underway in Minkébé , Ndoki-Dzanga , Odzala-Koukoua , and Boumba Bek for at least a decade , and even Salonga has benefited from some conservation effort . Most of the remainder of the Congo Basin does not receive any tangible wildlife management , and the conservation status of forest elephants is probably considerably worse . A simple analysis of the degree of fragmentation caused by roads across the range of the forest elephant is revealing ( Figure 7 ) . In the 1 , 893 , 000 km2 of potentially available forest elephant habitat in the Congo Basin , some 1 , 229 , 173 km2 ( 64 . 9% ) is within 10 km of a road . Just 21 , 845 km2 is over 50 km from the nearest road in just three countries , Congo , Gabon , and the Democratic Republic of Congo . Only Congo has potential elephant habitat beyond 70 km from a road , in the vast Likouala swamps to the northeast of the country . The road shapefile ( Environmental Systems Research Institute [ESRI] ) used is also restricted to major roads and thoroughfares , since most logging roads are either not geo-referenced or not mapped . Therefore the true degree of fragmentation of Central Africa's forest is considerably worse than is depicted on this map . Figure 7 indicates that the current NP system in the Congo Basin does a reasonable job of capturing the most remote tracts of forest that remain ( with the exception of the Likouala swamps ) . Despite considerable budgetary increases in recent years , funding for NPs and conservation landscapes remains below that necessary for even minimal management [27 , 28] , and an appropriate question to ask is whether or not protected areas actually protect forest elephants . The Megatransect data suggest strongly that NPs and protected areas are making a positive contribution to conservation because at any given distance from the nearest road , protected areas have ( 1 ) lower incidence of human sign , and ( 2 ) higher incidence of forest elephant sign than non-protected forest , at least in Congo and Gabon . The situation in the rest of the protected areas system and the forest at large is likely to be considerably worse , particularly in areas of armed conflict , civil disorder , and deteriorating socio-economic conditions [29] . In the Ituri Forest of eastern Democratic Republic of Congo , for example , where some of the bloodiest fighting seen in recent decades has occurred , an estimated 17 , 000 kg of ivory was evacuated from a 25 , 000 km2 forest block in a 6 mo period during 2003 [30] . Given a mean estimated weight of ivory from African elephants of 6 . 8 kg [31] , this could represent some 2 , 500 elephants . There is no doubt that forest elephants are under threat from illegal killing across Central Africa's forests , and soon , the only elephants left to poach will be those that remain in the interior of a few remote , well-funded , and well-managed NPs in politically stable countries . In this paper , we have shown that even with a near-universal ban of the trade in ivory in place , forest elephant range and numbers are in serious decline . This is in contrast to much of the recent literature on “the African elephant” that indicates generally stable or increasing populations in Eastern and Southern Africa [32] , and in some cases , dramatic population growth and a “return of the giants” [33] . The decline of the ecologically , socially , morphologically , and genetically distinct forest elephant , ( perhaps a separate species [18] or , at the very least , an “evolutionary significant unit” [34] worthy of high conservation status ) has profound implications for the diversity and resilience of the African elephant . Given their vulnerability compared to savannah elephants , the wellbeing of forest elephants must be given priority when making decisions about elephant management on the continental scale . Key issues that fall into this category include the future of the ivory trade [35] and resource allocation for international support programmes , such as MIKE . Forest elephants will continue to decline unless four immediate actions are successfully implemented . First , a national- and regional-scale approach to road development planning and construction is necessary in which reduction of fragmentation of Africa's last forest elephant strongholds is a central component . Second , law enforcement , including aggressive anti-poaching , of remaining priority elephant populations in NPs must gain the financial and political commitment required to ensure management success . Third , the illegal trade in ivory must be brought under control in elephant-range states , transit countries , and destination nations . Forth , effective partnerships must be developed with private logging and mining companies to reduce their negative impacts in the peripheries of protected areas and stop encroachment into NPs .
Density estimates of forest elephants in MIKE survey sites were obtained from dung counts conducted on systematic line-transect distance sampling surveys [36] designed and analysed using the Distance 4 . 1 software package [37] . Distance sampling is a standard survey method for abundance estimation in both terrestrial and marine environments but , as far as we are aware , has never been used for ground-based surveys on foot on a scale approaching that of the present study , which comprised a total area of 60 , 895 km2 in some of the most remote and difficult terrain in forested Africa . Site boundaries were defined following discussions with the MIKE directorate , national wildlife directors , and site-based personnel , and were ultimately constrained by the total operating budget . Rivers , flooded forest , and swamps were excluded from site definitions because elephant dung piles cannot be surveyed in these habitats . An attempt was made to design site boundaries that captured the gradient of human impacts on elephants , balanced against the need for a reasonable level of precision within each survey stratum . “Reasonable” precision was defined as a 25% coefficient of variation ( CV ) for estimates of elephant dung density for each survey stratum . To improve precision , each MIKE site was stratified according to expected elephant dung-pile encounter rate ( n0/L0 ) based on either data from short pilot studies or from expert opinion in the case of the vast Salonga site , where a pilot study was prohibitively expensive . The effort in terms of total length of transect line required to attain the required precision was estimated according to the equation on page 242 of [36] using a value of three for the dispersion parameter b as recommended by Buckland et al . [36]: where CVt ( D^ ) denotes the target CV for the density estimate . Survey designs were completed using the “systematic segmented trackline sampling” option of Distance 4 . 1 , as systematic designs with a random start are more robust to variations in the distribution of the population being sampled in terms of estimator precision [38] . This is a survey design class that superimposes a systematic set of parallel tracklines onto the survey region with a random start , along which line-transect segments are evenly spaced , again with a random start , at intervals and lengths determined by the user . Spacing and length of line transects varied by stratum and site according to the required sampling intensity . To potentially improve precision , line transects were oriented at 90° to major river drainages to run parallel to possible gradients in elephant density . The start and end point of each line transect was uploaded to a Garmin 12XL GPS ( global positioning system; http://www . garmin . com ) to assist field navigation . If in the field , a line transect began in a swamp or river , it was displaced to the nearest location that could be found on terra firma . Similarly , when line transects traversed inundated areas , that portion of the transect was discarded , and an equivalent length was added to the end of the transect . Line transects were oriented using a sighting compass from the start point , and cut with a minimum of damage to the understorey . Observers walked slowly ( ca . 0 . 5–0 . 75 km hr−1 ) along the line transect , scanning the ground for elephant dung piles . Distance along transects was measured using a hip-chain and topofil to the nearest metre , and the distance of the centre of each dung pile to the centreline were measured to the nearest centimetre using a 10-m tape measure . Survey methods are described in detail in [39] . In the field , the end of one line transect and the beginning of another were connected by reconnaissance walks following a “path of least resistance” through the forest [40] . On reconnaissance walks , a general heading was maintained in the desired direction of travel , but researchers were free to deviate to avoid thickets and steep hills or to follow elephant trails , human trails , and even logging roads . On reconnaissance walks , a continuous GPS tracklog is maintained , with a fix taken every 10–15 s . Data collection included all elephant dung piles , human sign , and vegetation type , and data records were coded by time ( GMT ) . Data were later reconciled with GMT from the GPS tracklogs and thus geo-referenced and imported into ESRI ArcView 3 . 2 ( Redlands , California , United States ) . Such reconnaissance walks are particularly useful for assessing the intensity and types of hunting activity if signs of humans are followed when encountered . However data are biased and provide only a general overview of large mammal distributions and human activity in the forest . The Megatransect also used reconnaissance survey methods consistent with the MIKE methods . Elephant carcasses were defined as poached if evidence of hunting was obtained , which included gunshot holes in the carcass , removal of tusks , and meat on smoking racks . Elephant poaching camps were identified from other hunting camps by the presence of remains of elephant or very large meat-smoking racks . DISTANCE 4 . 1 software [37] was used to analyse the perpendicular distance data from the field measurements and to calculate the density of elephant dung piles by survey stratum and by individual line transect as described by Buckland et al . [36] . Different detection functions were fitted to the data sequentially using half-normal , uniform , and hazard rate key functions with cosine , hermite polynomial , and simple polynomial adjustment terms . The best model was selected on the basis of the lowest Akaike's Information Criterion score ( AIC ) [41] , and χ2 goodness-of-fit tests were used to examine the fit of the model to the data . On-site studies of elephant defecation and dung decay were not carried out due to the logistical and funding difficulties of working over such a large area , thus dung density was converted to elephant density using estimated conversion factors [42] of 19 defecations per day , and mean dung lifespan of 90 d for all sites . In preparation for the statistical modelling , the centroid of each transect and 5-km Megatransect segment was used to calculate the distance of each “sample unit” from the nearest road or protected area boundary using the ESRI ArcView 3 . 2 extension “Nearest Feature” [43] . A shapefile of Central African roads was obtained from Global Forest Watch ( World Resources Institute , Washington , D . C . , United States ) . The protected areas shapefile was provided by the Wildlife Conservation Society . Data from two MIKE sites , Dzanga-Sangha and Nouabalé-Ndoki , were pooled for analytical purposes since they are contiguous areas and therefore contained a single elephant population . Generalized Linear Models with a binary response and logistic transformation were used for the logistic regression analyses [21] . The Generalized Additive Models [22] fit to the dung-count data from the MIKE sites have the form where ni denotes the number of dung piles detected on the ith transect , li the length of the ith transect , and ॖ^ is a site-specific estimate of the effective strip half-width [36] calculated using the Distance 4 . 1 software [37] . The term 2li ॖ^ gives the area effectively surveyed on transect i . β0 is the intercept , and f ( zij ) is a smooth function of the jth covariate z associated with the ith transect . To deal with the over-dispersion in the data , a quasi-Poisson distribution was assumed . By including area effectively surveyed as an offset term in the model , dung density is , in effect , being modelled . The results are equivalent for elephant density if we assume constant conversion factors of 19 defecations per day and a mean dung lifespan of 90 d for all sites . The models were fit in R [44] using the mgcv package [45] . To avoid over-fitting , the degrees of freedom were restricted to two in the final model . The elephant dung-count data used in the analysis were over-dispersed in part due to the large number of zero counts . Some of these problems were eliminated by conditioning on elephant presence and only using non-zero counts for the analysis . In addition , using a quasi-Poisson model instead of a Poisson allowed for the modelling of over-dispersion by not assuming that the dispersion parameter is fixed at 1 . The standard diagnostic plots used in model selection and assessment of fit indicated that the model is consistently giving lower fitted values when these are compared to the response values . The extraordinarily high elephant dung counts for certain areas of Minkébé , and occasionally for Odzala and Ndoki-Dzanga , that are in stark contrast to the counts at other sites or transects within the same site contribute to this problem . The same methods were applied to the Megatransect data except that the offset term representing the area effectively surveyed term was omitted since this dataset does not permit the estimation of the effective strip half-width ॖ^ . Also , to avoid over-fitting , the degrees of freedom were restricted to 3 for both covariate terms in the final model for the Megatransect data . Spatial Analyst from ESRI was used to construct the images in Figure 7A and 7B , and the interpolations of human sign and elephant dung counts for Ndoki-Dzanga shown in Figure 8 were produced using the “Calculate Density” feature of the same extension . | Forest elephants , perhaps a distinct species of African elephant , occur in the forests of West and Central Africa . Compared to the more familiar savannah elephant of Eastern and Southern Africa , forest elephant biology and their conservation status are poorly known . To provide robust scientific data on the status and distribution of forest elephants to inform and guide conservation efforts , we conducted surveys on foot of forest elephant abundance and of illegal killing of elephants in important conservation sites throughout Central Africa . We covered a combined distance of over 8 , 000 km on reconnaissance walks , and we systematically surveyed a total area of some 60 , 000 km2 under the auspices of the Monitoring of the Illegal Killing of Elephants ( MIKE ) programme . Our results indicate that forest elephant numbers and range are severely threatened by hunting for ivory . Elephant abundance increased with increasing distance from the nearest road , and poaching pressure was most concentrated near roads . We found that protected areas have a positive impact on elephant abundance , probably because management interventions reduced poaching rates inside protected areas compared to non-protected forest . Law enforcement to bring the illegal ivory trade under control , and effective management and protection of large and remote national parks will be critical if forest elephants are to be successfully conserved . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"ecology",
"mammals"
] | 2007 | Forest Elephant Crisis in the Congo Basin |
Rabies continues to be a major public health problem in the Philippines , where 200–300 human cases were reported annually between 2001 and 2011 . Understanding the phylogeography of rabies viruses is important for establishing a more effective and feasible control strategy . We performed a molecular analysis of rabies viruses in the Philippines using rabied animal brain samples . The samples were collected from 11 of 17 regions , which covered three island groups ( Luzon , Visayas , and Mindanao ) . Partial nucleoprotein ( N ) gene sequencing was performed on 57 samples and complete glycoprotein ( G ) gene sequencing was performed on 235 samples collected between 2004 and 2010 . The Philippine strains of rabies viruses were included in a distinct phylogenetic cluster , previously named Asian 2b , which appeared to have diverged from the Chinese strain named Asian 2a . The Philippine strains were further divided into three major clades , which were found exclusively in different island groups: clades L , V , and M in Luzon , Visayas , and Mindanao , respectively . Clade L was subdivided into nine subclades ( L1–L9 ) and clade V was subdivided into two subclades ( V1 and V2 ) . With a few exceptions , most strains in each subclade were distributed in specific geographic areas . There were also four strains that were divided into two genogroups but were not classified into any of the three major clades , and all four strains were found in the island group of Luzon . We detected three major clades and two distinct genogroups of rabies viruses in the Philippines . Our data suggest that viruses of each clade and subclade evolved independently in each area without frequent introduction into other areas . An important implication of these data is that geographically targeted dog vaccination using the island group approach may effectively control rabies in the Philippines .
Rabies is a fatal viral disease that causes an estimated 55 , 000 human deaths globally ever year , of which 57% occur in Asia [1] . Although effective measures to control rabies , such as dog vaccination , are available , rabies still is a major public health problem in many countries such as the Philippines , where 200–300 human rabies cases were reported annually by the National Notifiable Diseases Surveillance System between 2001 and 2011 [2] . Human rabies cases were also exported from the Philippines to countries that had been declared rabies-free , including Japan in 2006 [3] , [4] and Finland in 2007 [5] . The National Rabies Control and Prevention Program of the Philippines is a joint effort by the Department of Agriculture , Department of Health , and other partners that aim to eliminate rabies from the country by the year 2020 [2] . Hence , there is an urgent need to establish an effective and feasible strategy for controlling rabies in the Philippines . The rabies virus of the Rhabdoviridae family is the major Lyssavirus responsible for majority of human and animal rabies cases . The rabies viral genome is a nonsegmented single-stranded negative-sense RNA of approximately 12 kb , which encodes a nucleoprotein ( N ) , a phosphoprotein ( P ) , a matrix protein ( M ) , a glycoprotein ( G ) , and a polymerase ( L ) [6] , [7] . The rabies virus N gene is most commonly used for diagnosis with the reverse transcription-polymerase chain reaction ( RT-PCR ) because it is the most conserved gene in lyssaviruses [8] . The N gene has also been used extensively in molecular epidemiological studies [7] , [9]–[16] and the G gene encodes a surface protein that is targeted by neutralizing antibodies . In addition , this surface protein is crucial to viral invasion of host cells because it attaches to the host receptors [17] . Many recent studies have utilized the rabies G gene for molecular epidemiological analyses [10] , [14]–[16] , [18] , [19] . Phylogenetic analysis of the partial N gene was reported previously for rabies viruses in the Philippines and indicated that these viruses formed a unique phylogenetic group that was further divided into two subgroups [9] . However , only a limited number of specimens were analyzed . Therefore , more detailed genetic information is needed to characterize circulating rabies viruses and to further resolve their transmission dynamics in the Philippines . Understanding the transmission dynamics and genetic diversity of rabies provides useful information for establishing a rabies control strategy [13] , [20] . The National Rabies Prevention and Control Program ( NRPCP ) of the Philippines is implementing a geographically targeted dog vaccination approach to eliminate rabies . As the Philippines is an island nation , this approach should be feasible and effective for eliminating rabies provided transmissions between islands are not frequent . Detailed molecular analyses of rabies viruses from different regions elucidates geographic compartmentalization , evolution , and transmission dynamics , which can be used to establish an effective rabies control strategy in the Philippines . In the present study , we analyzed the complete G gene ( 1572 nt ) of rabies-positive animal brain samples collected from different regions in the Philippines . We also performed partial N gene sequencing ( 1124 nt ) of the selected samples to compare the results with those of complete G gene sequencing .
Tissues from the hippocampus and medulla were collected from suspected rabid animals from different regions of the Philippines between 2004 and 2010 . These included samples retrieved from the sample bank of the National Rabies Reference Laboratory at the RITM , which were collected from 2004 to 2007 . The remaining samples were collected prospectively from 2008 to 2010 from the Regional Animal Disease Diagnostic Laboratories ( RADDLs ) of Regions 1 , 2 , 3 , 5 , 7 , and 10 , the Cordillera Autonomous Region ( CAR ) of the Department of Agriculture , and the Provincial Animal Disease Diagnostic Laboratory ( PADDL ) of Dumaguete City , Negros Oriental . The Philippines is an island nation comprising over 7 , 000 islands that are divided into three groups , Luzon , Visayas , and Mindanao . The country is governed as 17 regions and 80 provinces . Samples were collected from 11 regions , including Regions 1 , 2 , 3 , 4A , 4B , 5 , CAR , the National Capital Region ( NCR ) in Luzon , Region 7 in Visayas , and Regions 10 and 11 in Mindanao ( Figure S1 ) . A fluorescent antibody test ( FAT ) was performed at the RADDLs and PADDL to diagnose rabies infection , and positive samples were shipped to RITM for molecular analysis in accordance with the guidelines for shipping Category B infectious substances [21] . FATs were repeated at the RITM to confirm positive results . In brief , brain impression smears of the samples were fixed in cold acetone and then air-dried . Each smear was flooded with fluorescent isothiocyanate antirabies monoclonal globulin ( Fujirebio Diagnostics , Inc . , Malvern , PA , USA ) in 1% Evan's blue counterstain and were incubated in a humid chamber at 37°C for 30 min . The slides were washed twice with phosphate-buffered saline , rinsed with double-distilled water ( ddH2O ) , and air-dried . Buffered glycerol mounting medium was applied before viewing rabies virus antigens under an immunofluorescent microscope . Total RNA was extracted from animal brain tissues using a RNeasy Mini Kit ( Qiagen , Hilden , Germany ) . Approximately 30 mg of tissue was homogenized using a homogenization pestle , and RNA was purified according to the manufacturer's protocol for animal tissues . RNA was eluted in 50 µL of diethylpyrocarbonate-treated water and stored at −70°C until further processing . The rabies virus G gene was amplified using primers RV-7F and RV-9RPh to generate a 2 , 425-bp amplicon ( Table 1 ) . Primers p1 and 304 were used to amplify the rabies virus N gene , thereby generating a 1 , 511-bp amplicon ( Table 1 ) . Amplification was performed using Superscript III One-Step RT-PCR with Platinum Taq High Fidelity Polymerase ( Invitrogen , Carlsbad , CA , USA ) . The 25 µL RT-PCR reaction mixture contained 9 µL of ddH2O , 12 . 5 µL of 2× reaction mix , 0 . 25 µL of each primer at 50 µM , 0 . 5 µL of the enzyme mix , and 2 . 5 µL of RNA template . One step RT-PCR was performed using a TaKaRa PCR Thermal Cycler Dice ( TaKaRa Bio Inc . , Shiga , Japan ) under the following thermocycling conditions for the G gene: reverse transcription at 50°C for 30 min , denaturation at 95°C for 5 min , 25 cycles of 94°C for 1 min , 65°C for 1 min , and 68°C for 1 . 5 min , and a final extension at 68°C for 10 min . The following thermocycling conditions were used for the N gene: reverse transcription at 50°C for 30 min , denaturation at 95°C for 5 min , 25 cycles of 94°C for 1 min , 54°C for 1 min , and 68°C for 1 . 5 min , and a final extension at 68°C for 10 min . A SUPREC PCR Purification Kit ( TaKaRa Bio Inc . ) or a QIAquick PCR Purification Kit ( Qiagen ) was used to purify RT-PCR products prior to nucleotide sequencing . Cycle sequencing was performed using a BigDye Terminator v1 . 1 or v3 . 1 Cycle Sequencing Kit ( Applied BioSystems , Foster City , CA , USA ) in the TaKaRa PCR Thermal Cycler Dice . Sequencing reactions were purified using a BigDye XTerminator Purification Kit ( Applied BioSystems ) followed by loading into Genetic DNA Analyzers 310 , 3130 , or 3730xl ( Applied BioSystems ) . Bidirectional sequencing was performed using primers listed in Table 1 to resolve the complete G gene ( 1 , 572 nt ) and a partial region ( 1 , 124 nt ) of the N gene . Multiple sequence alignments and phylogenetic relationships were inferred using the maximum-likelihood method of the Molecular Evolutionary Genetics Analysis version 5 algorithm ( MEGA5; http://www . megasoftware . net/ ) with bootstrap probability calculated from 500 replicates . In this analysis , the general time reversible+γ model was selected as a substitution model based on AICc values ( Akaike Information Criterion , corrected ) using the model selection feature of the MEGA5 program . The cut-off value for the condensed tree was set at 80% and was used to define genogroups . Clades and subclades were defined when the genogroup comprised more than 4 strains . Genogroups consisting of fewer than 3 strains were not considered as clades and are indicated by the prefix “Gr . ” All reference sequences used for comparative analyses were obtained from the GenBank genetic sequence database ( http://www . ncbi . nlm . nih . gov/genbank/; Table 2 ) . Sequences described in this study were submitted to GenBank and corresponding accession numbers are listed in Figure S1 .
Of 317 brain tissue samples collected , 51 were excluded because of incomplete epidemiological information ( 5 samples ) , failure of PCR amplification ( 21 samples ) , or sequencing ( 25 samples ) . The complete G gene ( 1572 nt ) and parital G gene sequences were determined for 233 and 33 strains , respectively . A phylogenetic tree of partial G genes ( 698 nt ) from 266 strains was obtained ( Figure S2 ) , and because of low bootstrap values , only samples with complete G genes were included in the analysis . However , as a minor but unique branch was found in the phylogenetic tree of partial G genes ( 698 nt ) , two samples collected from Batangas and Mindoro Island were included despite incomplete sequences ( 1549 nt; Figure S2 ) . Therefore a total of 235 samples were analyzed ( Figure 1 ) . Of the 235 samples , 35 were collected from 2004 to 2007 and deposited in the sample bank , while 200 were collected prospectively from 2008 to 2010 . Among analyzed samples , 228 were collected from dogs , two from cats , one from a cow , and four from unknown host species ( Tables 3 and S1 ) . From 235 samples analyzed , phylogenetic trees were constructed using 57 strains of N ( 1124 nt ) and G ( 1572 nt ) genes with the maximum-likelihood method . To determine phylogenetic relationships , phylogenetic trees for the N and G genes were constructed to compare the Philippine strains with those from other Asian countries ( Figure 2A and 2B ) . All the Philippine strains were classified into one distinct cluster , which was divergent from those of other Asian countries . This cluster included only the Philippine strains and was previously named Asian 2b by Gong et al . [22] . In the phylogenetic tree , the cluster closest to the Philippine strains was Asian 2a , which comprised strains from China , as shown in previous studies [16] , [22] , [23] . The Philippine cluster was further divided into three major clades , namely L ( Luzon ) , V ( Visayas ) , and M ( Mindanao ) . However , two strains from southern Luzon and Mindoro Island , namely GrSL and GrMD , respectively , were not classified into any of the 3 major clades in either N or G phylogenetic trees ( Figure 2A and 2B ) . To investigate phylogenetic relationships between the Philippine strains in detail , we analyzed the full-length G gene sequence from 235 strains and the partial G gene ( 1549 nt ) from two strains . It was confirmed that there were three major clades and two distinct genogroups . Clade L and two genogroups were found in the island group of Luzon , whereas clades V and M were found in the specific geographic island groups Visayas and Mindanao , respectively . Interestingly , strains of the distinct genogroups GrSL and GrMD were found in Mindoro Island and none of the clade L strains were found on this island ( Figure 1 ) . Clade L was further divided into nine subclades , L1–L9 , and there were three distinct genogroups with less than four strains , which were named GrL1–GrL3 ( Figure 3 ) . L1 and GrL1 were found in the Pangasinan Province of the western part of central Luzon where L6 strains were also common , whereas L2 was found mainly in the eastern part of central Luzon ( Figure 3 ) and L3 was also mainly distributed in central Luzon . GrL2 consisted of only one strain that was found in Pampanga Province , and GrL3 consisted of two strains that were distributed in the provinces of Rizal and Cavite located around Laguna de Bay ( Figure 3 ) . L4 , L5 , and L7 were found in southwestern Luzon ( Region 5 ) with the exception of one strain each from L4 and L5 , which were located in NCR ( Figure 3 ) . L8 was further divided into L8a and L8b ( Figure 4 ) . L8a and L8b strains were found in the northeastern and northwestern parts of Luzon Island , respectively . L9 was divided into 4 subclades , L9a–L9d , and 2 genogroups , GrL9a–GrL9b ( Figure 5 ) . L9a , L9c , GrL9a , and GrL9b were found in central Luzon , especially in Pampanga Province , and L9b was found exclusively in Zambales Province . L9d was distributed in NCR and Region 4A , with the exception of two strains located in the provinces of Nueva Ecija and Pangasinan . Clade V ( Visayas ) was subdivided into the two subclades V1 and V2 . All V1 strains were found exclusively on Cebu Island , whereas V2 strains were found exclusively on Negros Island ( Figure 6 ) . Clade M ( Mindanao ) strains consisted of only one subclade ( M1 ) and one genogroup ( GrM1 ) , and all samples except 1 was from northern Mindanao . The single sample from Davao ( GrM1; southern Mindanao ) was distinct from all other strains from this island ( Figure 6 ) .
The N and G genes of rabies field strains obtained from 11 of 17 regions in the Philippines were analyzed . Our phylogenetic analysis of the N and G genes revealed that there were three major clades , namely L , V , and M ( Figure 2 ) . A previous study identified two distinct clades in the Philippines [9]; however , it was based on the analysis of only 59 strains and detailed geographic information was not provided . In this study , we analyzed more systematically collected strains from 11 different regions , including Luzon , Visayas , and Mindanao , and identified a previously unknown clade , clade V , from Visayas . The presence of two major clades from Luzon , Visayas , and Mindanao indicated that rabies viruses have evolved independently in these island groups mainly because they are physically separated by sea . Similar geographic compartmentalization of rabies viruses has been reported in Indonesia , which is also an island nation [11] . Four strains were not classified into any of the three major clades , including three in GrSL and another in GrMD ( Figure 1 ) . Because only a few samples were found containing these genogroups , evolutionary relationships with major clades could not be defined . GrSL was found both in Luzon and Mindoro Islands , which indicates that viral transmission occurred between these islands . A similar pattern was observed in L7 , which was found in Sorsogon in Luzon Island and Catanduanes Island . This suggests that even in areas separated by the sea , such as a narrow channel with frequent ferry traffic , complete geographic barriers for rabies transmission may not be established and that inter-island transmission is possible in some exceptional cases . It is possible that GrSL and GrMD formed distinct major clades , though this remains to be confirmed with analyses of more strains in these genogroups . There was also unique genogroup in clade L , i . e . GrL3 was diverted separately from other subclades in clade L with a bootstrap value of 42% ( data not shown ) . Interestingly , all the strains in these genogroups ( GrMD , GrSL , and GrL3 ) were found in southern Luzon and its neighboring island Mindoro . More samples from these areas should be analyzed to better define evolutionary relationships . A recent molecular evolutionary analysis indicated that rabies viruses in the Philippines ( Asian 2b ) diverged from viruses in China approximately 1813 ( from 1707 to 1821 ) [22] . Another molecular evolutionary analysis based on N gene sequences suggested that the time of their divergence may have ranged from 1825 to 1936 [23] . It has been indicated that rabies viruses were introduced into previously rabies-free regions by human-mediated animal movements [24] . The introduction of rabies viruses into the Philippines was also possibly associated with human migration from China to the Philippines . Further molecular evolutionary analyses of rabies viruses in the Philippines are required to determine precise evolutionary histories . In general , subclades of clade L were represented by viruses found in geographically distinct areas . Interestingly , L8a and L8b were clearly divided by a mountainous area ( Mountain Province and Ifugao Province; Figure 4 ) , indicating that the mountain range also had an important impact on rabies virus transmission . L9 was divided into four clades and two genogroups , indicating circulation within a small geographic area ( Figure 5 ) . In clade L9 , only L9d was found in NCR and Region 4A from 2004 to 2007 , whereas other genogroups were found in Region 3 during 2008 and 2009 ( Figure S3 ) . Thus , a possibility that subclustering is determined not only by geographical elements but also temporal elements cannot be ruled out and samples collected from NCR after 2008 are required to decipher these effects . The provinces of Pampanga and Nueva Ecija contained five subclades and branches each , and the Pangasinan Provinces and NCR contained four branches each ( Figures 3 and 5 ) . Careful monitoring of the circulating strains between these areas , in which multiple subclades were found may provide important information regarding temporal and geographic transmission patterns of viruses in the Philippines . In Region 5 of Luzon , the three subclades L4 , L5 , and L7 ( Figure 3 ) were present . All L4 and L5 strains were found in Region 5 , with the exception of one strain of each found in NCR ( Figure 3 ) . These long distance disseminations of viruses may have occurred through human-mediated dog movements [11] . However , it is unclear with the current data whether these originated in NCR or Region 5 . Similarly , all of the L6 strains were distributed in Regions 3 and 5 , with the exception of one strain found over the mountain range in the CAR . In Visayas , viruses from different subclades V1 and V2 exist on the Cebu and Negros islands , respectively ( Figure 6 ) . Although these are separated by a narrow strait , our molecular analysis indicated that there was no viral movement between the two islands . All of the samples from Mindanao Island , with the exception of a sample from Davao , were collected from the northern region , particularly around the city of Cagayan de Oro ( Figure 6 ) . More samples from other areas are required to elucidate subclades from Mindanao Island . Like other islands , Mindanao may have been the site of a unique viral evolution . The Philippines has a national goal of eliminating rabies by the year 2020 , and a campaign of dog vaccinations has been conducted by each local government . However , the program has not been systematically implemented and funding allocation depends on the priorities of each local government . Hence , there is an urgent need to develop a more effective and feasible strategy to control rabies . Being an island nation , the Philippines has an advantage as rabies has been successfully eliminated in other island nations such as the UK and Japan [25] , [26] . Our molecular analyses showed that most rabies virus strains in the Philippines are uniquely clustered , suggesting that introduction from other countries has not occurred recently . Understanding the transmission dynamics of the rabies viruses is necessary for the development of an effective and feasible elimination strategy , particularly in countries with limited resources . Because of financial and resource constraints , the NRPCP of the Philippines adopted a phasing approach to eliminate rabies by area rather than nation-wide implementation of extensive dog vaccinations . The effectiveness of this strategy relies on limited transmissions between areas . Our molecular analyses indicated that this strategy could be effective in controlling rabies in the Philippines because there was clear geographic clustering of clades and subclades . However , a synchronized campaign would be necessary if there is frequent inter-island traffic . The Republic Act No . 9482 , known as the Anti-Rabies Act of 2007 , was enacted to control and eliminate human and animal rabies . This mandated control over dog and cat movements during inter-island transport and involved valid vaccination certifications only allowing transport two weeks or more after vaccination or within 12 months of vaccination by a licensed veterinarian [27] . This policy must be strictly implemented to avoid inter-island virus transmission via animal transportation as observed for the GrSL strain between Luzon and Mindoro Islands and the L7 strain between Luzon and Catanduanes Islands . In some areas of Thailand , homogeneous viruses circulated without recent introductions , which can serve as initial control targets [13] . However , the introduction of new genotypes to other locations is likely because of human-mediated dog movements [13] . We found several instances where viral dispersal may have occurred with dog movements . However , the frequency of such dispersal appeared to be much lower than that in Thailand . A more detailed analysis of dog movements , particularly on Luzon Island , which is the largest island in the Philippines , should be conducted to evaluate the risk of geographic dispersal . There were several limitations to our study . Firstly , we analyzed strains from wide geographic areas , including 11 of the 17 regions of the Philippines; however , many areas remain unexamined . The number of samples analyzed per region also varied because some regional laboratories tested samples more actively than others . Further , more samples were tested in areas close to the regional laboratory . Therefore , samples from more remote areas were not included in the analysis . Lastly , due to the limited time interval of sample collection , we only analyzed geographic clustering of rabies virus in the Philippines and temporal analysis , including molecular evolutional analyses were not conducted in this study . Further studies are required to fully understand these viral evolutionary processes . Despite these limitations , our analysis provides valuable molecular phylogeny and spatial distribution data for rabies virus variants currently circulating in the Philippines . | Rabies continues to be a major public health problem in the Philippines . We conducted a molecular epidemiological study of rabies using the complete glycoprotein ( G ) gene from 235 animal brain samples collected in the Philippines between 2004 and 2010 . We identified three major clades and two distinct genogroups in the Philippines . The three major clades L , V , and M were found specifically in the Luzon , Visayas , and Mindanao island groups , respectively . Additionally , two minor genogroups were located in the Luzon island group . These data suggest that although human mediated transmission may have occurred , these virus clades evolved independently after a single introduction into each island group . All of the analyzed Philippine strains were clustered into Asian 2b , which diverged from the Chinese strain Asian 2a . No recent introduction of rabies into the Philippines from other countries was apparent . The elimination of rabies by 2020 is a national goal in the Philippines , necessitating urgent development of a more effective and feasible strategy for controlling rabies . Our findings indicate that a geographically targeted dog vaccination campaign may effectively control rabies in island nations such as the Philippines . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"infectious",
"diseases",
"rabies",
"public",
"health",
"and",
"epidemiology",
"molecular",
"epidemiology",
"epidemiology",
"neglected",
"tropical",
"diseases",
"public",
"health"
] | 2013 | Genetic Diversity and Geographic Distribution of Genetically Distinct Rabies Viruses in the Philippines |
Phages , like many parasites , tend to have small genomes and may encode autonomous functions or manipulate those of their hosts' . Recombination functions are essential for phage replication and diversification . They are also nearly ubiquitous in bacteria . The E . coli genome encodes many copies of an octamer ( Chi ) motif that upon recognition by RecBCD favors repair of double strand breaks by homologous recombination . This might allow self from non-self discrimination because RecBCD degrades DNA lacking Chi . Bacteriophage Lambda , an E . coli parasite , lacks Chi motifs , but escapes degradation by inhibiting RecBCD and encoding its own autonomous recombination machinery . We found that only half of 275 lambdoid genomes encode recombinases , the remaining relying on the host's machinery . Unexpectedly , we found that some lambdoid phages contain extremely high numbers of Chi motifs concentrated between the phage origin of replication and the packaging site . This suggests a tight association between replication , packaging and RecBCD-mediated recombination in these phages . Indeed , phages lacking recombinases strongly over-represent Chi motifs . Conversely , phages encoding recombinases and inhibiting host recombination machinery select for the absence of Chi motifs . Host and phage recombinases use different mechanisms and the latter are more tolerant to sequence divergence . Accordingly , we show that phages encoding their own recombination machinery have more mosaic genomes resulting from recent recombination events and have more diverse gene repertoires , i . e . larger pan genomes . We discuss the costs and benefits of superseding or manipulating host recombination functions and how this decision shapes phage genome structure and evolvability .
Genetic recombination plays key roles in biology . Recombinases are required for essential cellular functions such as repair of stalled or collapsed replication forks , DNA repair and chromosome segregation [1] , [2] . Recombination also drives genetic diversification and increases the efficiency of natural selection [3] , [4] . Inter-genomic recombination allows horizontal gene transfer between organisms and exchange of sequences between viruses infecting the same cell [5] . Illegitimate and homologous recombination events between bacterial viruses ( phages ) are frequent and result in strongly mosaic genomes , i . e . genomes with strong internal phylogenetic incongruences [6] , but the relative importance of each recombination mechanism remains unclear [7]–[9] . The group of lambdoid phages provides a striking example of this phenomenon . These temperate phages account for more than two thirds of E . coli prophages [10] , and are extremely diverse from the genetic , structural and physiological point of view . Nevertheless , they all have similar genetic organization and this allows the production of viable hybrids by inter-genomic recombination [11] , [12] . Lambdoid genomes are organized in relatively autonomous gene clusters with genes being encoded next to their interactants , i . e . genes encoding an interacting protein or the targeted DNA site [13] . Moreover , the organization of morphogenesis genes strikingly reflects the order of the proteins forming the virion structure , suggesting a direct link between gene order and function or structure within each module [13] . The extent and phylogenetic range of genetic exchange can be very large: lambdoids include phages with different virion structures such as Siphovirus Lambda , Podovirus P22 or Myovirus SfV , showing that recombination blurs the traditional taxonomy ( based on virion morphology ) . Nevertheless , two thirds of the lambdoid phages in E . coli are closely related to phage Lambda and display a Siphoviridae's virion structure ( Lambda-like elements ) [10] . Phages and bacteria are in constant evolutionary arms races [14] . Accordingly , bacterial outer membrane structures that are phage attachment sites evolve very fast because of the selective pressure imposed by phages [15] . Reciprocally , phage proteins involved in attachment to the host cell , such as tail-fiber proteins , evolve fast in response to these changes [16] . Recombination both in the bacteria and in the phage facilitates these diversifying selection processes , accelerating the rate of evolution [17] . Efficient encapsidation of phage Lambda requires concatemeric DNA ( reviewed in [18] ) . These concatemers can be produced by homologous recombination or rolling-circle ( sigma ) replication ( Figure 1 ) . However , rolling-circle replication is inhibited by the exonucleolytic activity of the host RecBCD enzyme from the major homologous recombination pathway [19] . Hence , the phage needs to either block this exonucleolytic activity or produce concatemers by homologous recombination . Phage Lambda encodes its own homologous recombination toolkit under the form of a 3-genes operon [20]: exo , bet and gam , that encode Redα , Redβ and Gam respectively . Redα is a double strand specific 5′ to 3′ exonuclease and Redβ is a recombinase of the Rad52 superfamily that mediates strand annealing and exchange reactions starting from DNA extremities . Redβ and RecA ( the host recombinase ) have different recombination mechanisms , substrates and rates [21] . The protein Gam inhibits the host RecBCD exonuclease activity thus allowing efficient rolling-circle replication [22] . Thus , Lambda blocks the host recombination , superseding it with its own encoded recombination machinery . Other phages use evolutionarily related ( e . g . Erf in P22 ) or unrelated recombinases ( Sak4 in HK620 , related to RecA ) as well as other inhibitors of the exonucleolytic activity of RecBCD ( e . g . Abc2 in P22 or gp5 . 9 in T7 ) [23] , [24] . Lambda and most of its mutants cannot produce concatemers from monomers using the host RecABCD pathway of homologous recombination because Gam inhibits RecBCD . When gam is experimentally inactivated , RecBCD prevents phage replication by degrading its genome . However , Lambda mutants that include a chromosomal sequence with the octamer Chi motif ( GCTGGTGG ) are viable [25] . This is because the destructive nuclease-helicase activity of RecBCD shifts to repair mode when it meets a Chi site by recruiting the RecA recombinase onto nascent Chi-containing ssDNA [26] . The single strand annealing protein RecA then promotes strand invasion and recombination . Chi sites are very abundant in E . coli , found in average every 5 kb , and much more frequently in the core genome than in recently acquired genes [27] , [28] . Chi sites are absent from the wild-type genome of Lambda and this prevents the use of RecBCD to produce phage concatemers . The high frequency of Chi in the E . coli genome and its rarity in Lambda and phage T4 led to the hypothesis that Chi is implicated in the discrimination between self and non-self and that the RecBCD-Chi system also functions to protect the genome from mobile genetic elements [29]–[31] . Phage fitness depends on its ability to control its host and on what it pays for that in terms of genome space and production costs [32] . Phages encoding their own recombination mechanisms gain an advantage by using proteins that co-evolved with the phage for a long period of time and are thus adapted to it in terms of processivity and tolerance to sequence divergence . However , the expression of recombination functions takes up resources . Encoding these functions also takes up genome space . Lambdoids rarely exceed 60 kb in size and most are between 40 kb and 50 kb [10] . This suggests the existence of an optimal size beyond which further accretion of genetic material lowers the phage fitness . Loss of the recombination module might facilitate acquisition of other functions with higher adaptive value in certain ecological contexts as long as recombination functions can be found in the host and manipulated by the phage . Increase in phage genome size might also be costly because of the replication cost and because such genomes require larger virions [33] . Phages that manipulate host recombination functions do not pay these additional costs , but they must use machineries adapted to their hosts . These proteins might not fit optimally the phage requirements and may have a cost in terms of host range . On the other hand , these mechanisms are well adapted to the host genetic background . Here , we study phage recombination functions to understand how the dilemma between encoding and manipulating them shapes phage evolution .
We analyzed recombination functions encoded by lambdoid phages . These phages account for the majority E . coli prophages [10] , and their recombination mechanisms have been thoroughly studied [18] . The classification of phages in the group of lambdoids is itself motivated by their ability to produce viable hybrids by recombination at high frequency . We identified Chi motifs in a set of 275 lambdoid phages of Escherichia and Salmonella ( see Materials and Methods ) . We computed the expected number of the 8-nucleotide Chi motifs using four different statistical models: accounting for the frequency of nucleotides , tri-nucleotides , penta-nucleotides and hepta-nucleotides ( see Materials and Methods ) . The different models gave concordant results ( Table S1 and S2 ) . We present the results for the tri-nucleotides model , which is the most adequate for the slightly degenerated Chi motif and the small genomes of phages ( see Materials and Methods ) . We computed the number of Chi motifs observed/expected ( O/E ) ratio separately for each phage genome . Surprisingly , we found that , as a whole , lambdoids have more Chi motifs than expected ( median O/E = 2 . 30 , p<0 . 0001 , Mann-Whitney test ) . In fact , most Lambda-like phages encode Chi motifs ( 85% ) , which are significantly more frequent in these phages than expected given sequence composition ( median O/E = 2 . 43 , p<0 . 0001 , Mann-Whitney test ) . These results show that Chi sites are far from rare in phage genomes . In fact , they are much more abundant than expected given genome size and composition . Phage genomes lacking recombinases require the host machinery to engage in homologous recombination . To test the hypothesis that this leads to selection for the presence of Chi sites to recruit RecBCD , we detected phage recombinases using protein clustering and profile-profile alignments ( see Materials and Methods ) . We identified a recombinase in 141 genomes of lambdoids , i . e . approximately half of our dataset ( Rec+ phages , 51% ) ( Figure 2A ) . Most of the identified recombinases ( 68% ) are from the Redβ family , the one encoded by Lambda ( Figure S1 ) . Phage genomes lacking recombinases ( Rec− phages ) display a significant over-representation of Chi sites ( median O/E = 3 . 12 , p<0 . 0001 , Mann-Whitney test ) . These results are well in agreement with our hypothesis that phages lacking recombination functions select for the presence of Chi sites to recruit the host recombination machinery . Phages encoding recombination functions but no RecBCD inhibitory functions could select for the presence of Chi motifs in their genomes to protect themselves from RecBCD exonucleolytic activity . To test this hypothesis , we searched for RecBCD inhibitors from the Gam and Abc2 families and identified 95 of these ( see Materials and Methods ) . We found no single phage lacking a recombinase and encoding a RecBCD inhibitor . Red−Gam+ Lambda mutants are viable [19] , showing that recombinases are not strictly required for phage replication when RecBCD is inhibited . On the other hand , RecBCD inactivation in the absence of phage recombinases has a very strong fitness cost in E . coli [34] . Cells where phages inhibit RecBCD without superseding it with their own recombinases lack tools to efficiently repair DNA double strand breaks . The fitness cost associated with this impairment might explain the lack of Rec−Inh+ phages in our dataset . We found 95 phage genomes encoding a recombinase and a recombination inhibitor ( Rec+Inh+ ) . Among Rec+ phages , Inh+ phages display a significant under-representation of Chi sites ( median O/E = 0 , p<0 . 0001 , Mann-Whitney test ) , whereas Inh− over-represent Chi motifs ( median O/E = 2 . 50 , p<0 . 0001 , same test ) ( Figure 2B and 2C ) . Importantly , while both Rec+Inh− and Rec− phages over-represent Chi , the latter show stronger over-representation ( p<0 . 03 , Wilcoxon test ) . Gam-like proteins inhibit RecBCD activity , whereas Abc2-like RecBCD inhibitors subvert RecBCD functions rendering them Chi-insensitive [35] . We tested if phages encoding Gam-like RecBCD inhibitors showed different degrees of avoidance of Chi motifs relative to those encoding Abc2-like RecBCD inhibitors . While there is a slightly stronger avoidance of Chi sites in Abc2 encoding phages ( p = 0 . 030 , Wilcoxon test ) , both Gam-like and Abc2-like RecBCD inhibitors are strongly associated with Chi motifs under-representation ( median O/E of 0 . 30 and 0 respectively , both p<0 . 0001 , Mann-Whitney tests ) . Hence , phages encoding recombinases but not RecBCD inhibitors have more Chi sites than expected , whereas phages with RecBCD inhibitors strongly avoid Chi sites . This suggests that Rec+Inh− phages select for the presence of Chi sites , whereas Rec+Inh+ phages select for the absence of Chi sites . Phage Lambda is thus a typical representative of the Rec+Inh+ class of phages . These results show a strong link between the ability of a phage to inhibit the exonucleolytic activity of RecBCD and the presence or absence of Chi motifs . We compared the frequency of Chi motifs in phages and their hosts . As observed previously [27] , [28] , Chi motifs are over-represented in the genomes of E . coli K12 and S . enterica Typhimurium ( O/E = 2 . 29 , p<0 . 0001 and O/E = 2 . 40 , p<0 . 0001 , Z score ) , and slightly more in the core genome of each species ( resp . O/E = 2 . 36 and 2 . 38 , both p<0 . 0001 , same test , see Table S3 for the different models ) . The density of Chi sites in Rec− phages is not significantly different from the host bacterial genome ( 0 . 2 Chi motifs/kb , p = 0 . 103 , Mann-Whitney test ) . However , given their composition , Chi motifs are more over-represented in these phages than in the core genome of E . coli ( p<0 . 0001 , Mann-Whitney test ) . The over-representation of Chi sites in Rec+Inh− phages is not significantly different from that of the core genome of E . coli ( p = 0 . 30 , same test , see Table S4 for the other models ) . These results suggest that phages lacking RecBCD inhibitors endure similar or even stronger selection for Chi motifs than their hosts . Some of the phages in our dataset were sequenced from virions whereas others were identified from bacterial chromosomes . We tested if inaccurate delimitation of the latter might have affected the number of Chi motifs found in our dataset . The median O/E number of Chi sites was not significantly different between Rec− phages and Rec− prophages ( resp . 4 . 76 and 3 . 08 , p = 0 . 45 , Wilcoxon test ) . This ratio was almost indistinguishable among Rec+Inh− phages and prophages ( resp . 2 . 42 and 2 . 52 , p = 0 . 58 , same test ) and among Rec+Inh+ phages and prophages ( both medians equal to 0 , p = 0 . 84 , same test ) . Thus , the trends we observe in the frequency of Chi motifs do not reflect biases associated with prophage detection . We also verified that Chi motifs in phages were not concentrated at the cargo region , typically at the edge of the element opposing the integrase [36] . Interestingly , we found that Chi motifs were concentrated far from this region and between the genes encoding the replication functions and the terminase , before the structural genes . In Lambda this corresponds to the region between the origin of replication ( in gene O ) and the cos site ( before the terminase gene Nu1 ) where DNA is cut during packaging ( Figure 3 ) . The distribution of Chi sites along the chromosomes of Rec+Inh− phages and Rec− phages is different ( p<0 . 0001 , Kolmogorov-Smirnov test ) . Chi motifs are more concentrated near the origin of replication of Rec− phages , and towards the cos site in Rec+Inh− phages . These results show that Chi over-representation in lambdoids cannot result from inaccuracies in the delimitation of prophages and suggests a tight association between recombination , replication and packaging in phages . Recombination between different phages leads to genetic mosaicism and increases the diversity of gene repertoires . Redβ catalyzes recombination at higher rates and is more tolerant to sequence divergence than RecA [8] . We thus hypothesized that phages encoding recombination functions have more diverse gene repertoires . We built the pan genomes ( i . e . the set of all different gene families ) of Rec+ and Rec− lambdoids ( see Materials and Methods ) . The pan genome of Rec+ phages is systematically ∼22% larger than the pan genome of Rec− phages for the same number of genomes ( Figure 4 ) . This effect could not be explained by genome size , which is indistinguishable between the two types of phages ( average of 45 kb , p = 0 . 85 , Wilcoxon test ) . Hence , the permissivity of phage recombinases might allow faster diversification of gene repertoires in phages encoding their own recombination functions . We then tested the hypothesis that these phages are also more mosaic , i . e . exchange homologous genes at higher rates . For this , we identified highly similar homologous genes present in highly dissimilar phage genomes ( see Materials and Methods ) . This is a conservative subset of the genes that have recently undergone recombination between distinct phages . We restricted the analysis to the 163 Lambda-like phages of E . coli since broader taxonomic groups share too few homologous proteins for reliable inference of distances between phages . We computed the distance matrices between homologous proteins ( d ) and between phages ( D ) and identified proteins for which d is small and D is large using a range of thresholds Td and TD ( see Materials and Methods ) . The results consistently show that genes with low d encoded in phages of high D are very significantly over-represented in Rec+ phages ( Figure 5 ) . Rec+ phages have up to 8 times more such genes than Rec− phages and this difference is particularly high for the most recent transfers ( corresponding to the lowest values of d ) . We tested if these results could be explained by the nature of the genes undergoing recombination . We analyzed the functional categories of the transferred genes ( Text S2 ) , and found no significant differences between them and the remaining genes ( p>0 . 1 , χ2 test ) . We conclude that the higher mosaicism of phages encoding recombinases is independent of its phage gene repertoire size or content .
In this work we studied the presence in phage genomes of genes and DNA motifs involved in homologous recombination . We showed that some phages encode a large number of Chi motifs and are thus able to manipulate RecBCD . This provides certain advantages . First , for similar genome size , and thus capsid volume , this allows the genome to encode other potentially adaptive functions . Second , Chi sites protect from the exonucleolytic activity of RecBCD and thus also from restriction-modification systems [37] . Third , RecABCD recombination is less frequent between very divergent sequences and could lead to fewer non-viable hybrid genomes . Finally , Chi motifs being important for genome maintenance , the presence of Chi in prophages might stabilize the element and lower its fitness cost for the host . Prophages make up to 35% of the pan genome of E . coli and we have shown that they encode motifs associated with their local context in the bacterial chromosome [10] . Hence , prophages with Chi motifs might integrate more seamlessly in the host chromosome . Some phages encode their own recombination machinery , inhibit the host's and avoid Chi motifs . Recombination autonomous to the host machinery also has some advantages . First , recombination machineries co-evolving with the phage should be better adapted to its specificities , e . g . in terms of recombination frequency , sequence composition or homology requirements . For example , RecT , a Redβ homolog from prophage Rac , shows preference for AT rich regions [38] , which are typical of phages . Second , reduced dependence on the host's machinery might broaden the range of possible hosts . Even if the composition of the machinery of homologous recombination is similar in most non-intracellular γ-Proteobacteria [39] , the Chi motifs of E . coli and Haemophilus influenzae show a number of differences [40] . Hence , phages relying on host recombination functions may be at a disadvantage in a new host encoding different Chi motifs . Third , Red recombination is more permissive to sequence divergence and this may enlarge the mutational landscape of the phage , accelerating its diversification . The dilemma of encoding or manipulating host recombination functions may also impact ecological interactions between mobile genetic elements . For example , the protein Old encoded by phage P2 targets Redβ [41] and the AbiK system of Lactococcus lactis plasmids targets different phage recombinase families [42] . On the other hand , encoding autonomous recombination functions may render the phage less susceptible to mobile elements that compete to manipulate host recombination . During co-infection , phages encoding RecBCD inhibitors might therefore have an important advantage over Chi-dependent phages by reducing the number of concatemeric chromosomes they can produce . The chromosomes of E . coli strains are packed with prophages , some of which contribute to important adaptive functions . Different temperate phages may recombine in the bacterial cell . These cells may thus work as ‘phage factories’ , releasing a wide variety of recombinant phages in the environment [43] . We have shown that phages carrying their own recombination functions have more mosaic genomes and larger pan genomes . The gene repertoires of bacteria are in constant genetic flux partly due to the action of phage transduction . For example , the recent epidemic of E . coli in Germany was the direct consequence of toxins encoded by prophages [44] . Adaptive associations between phage and bacteria can be very complex , e . g . a bacterial endosymbiont prophage protects aphids from parasitoid wasps [45] . As mentioned above , recombination is also important in the context of the ongoing arms races between phages and their hosts . Hence , the way phages recombine may impact their rates of diversification , but also those of their bacterial hosts . The absence of Chi in phage Lambda was instrumental to the discovery of the function of this motif [46] . It was also interpreted as lack of selection for the presence of Chi sites in phages carrying their own recombination systems [29] . Here , we showed that contrary to common belief Chi sites are very abundant in most phages . Yet , these results also put forward a puzzling observation . RecBCD inhibitors render Chi sites useless either by blocking the activity of the protein or by rendering it insensitive to Chi . Hence , phages encoding RecBCD inhibitors should have a number of Chi sites close to the random expectation given sequence length and composition . Surprisingly , we show that these phages strongly avoid Chi sites , i . e . they have fewer sites than expected . Chi is thus selected against in phages encoding RecBCD inhibitors and for in the other phages . This suggests that carrying simultaneously Chi sites and RecBCD inhibitors is deleterious for the phage . We have no good explanation for these intriguing results at the moment . One might speculate that Chi sites affect the efficiency of RecBCD inhibitors , but this is at odds with the observation that the E . coli chromosome is packed with Chi motifs . Chi avoidance might be related to the chromosomal context of the prophage and how it affects chromosome maintenance processes , e . g . selection for recombination outside the prophage element to avoid chromosomal rearrangements [47] . But this would suggest that Chi are deleterious to integrative elements , which seems at odds with the large number of Chi sites found in the majority of prophages . Understanding selection against Chi sites will require further experimental work . We showed that Chi sites in phages are concentrated between the origin of replication ( especially in Rec− phages ) and the packaging sites ( especially in Rec+Inh− phages ) . Naturally , the origin and cos ( or pac ) sites are unknown for the majority of phages and this result must be interpreted with care since it assumes that among lambdoids these positions are relatively unchanged . Nevertheless , the high density of Chi in the origin and packaging site regions , and the differences between the two regions in terms of phage recombination repertoires suggest some sort of selection for Chi sites in these locations . In fact , the very high frequency of Chi motifs in such a small region , up to three times the density in the E . coli core genome , might explain why this region is unusually variable among lambdoid genomes ( the nin region [7] , [48] ) . The association between replication and recombination is pervasive in cellular organisms [1] and phages lacking recombinases might thus select for Chi sites near the origin of replication to process stalled replication forks . In phages encoding a recombinase able to process stalled replication forks , Chi sites might be more important for protection of free DNA ends from degradation by RecBCD than for its recruitment for recombination , explaining the fewer Chi sites and their location close to the packaging site in these phages . Hence , the study of the roles of Chi sites in phages might enlighten further functional associations between recombination , phage replication and packaging . To check on the generality of our observations , we made some preliminary analyses of non-lambdoid E . coli phages in GenBank ( Table S5 and Text S3 ) . These analyses are hampered by the small dataset for each phage family and the lack of available information on the mechanisms of recombination in most genera . Yet , we could verify that phages requiring concatemers for packaging over-represent Chi motifs relative to phages able to encapsidate monomers ( p<0 . 0001 , Wilcoxon test ) . The two phage genera requiring concatemers for packaging and lacking recombinases ( T5-like and P1-like ) exhibit the strongest over-representation of Chi motifs ( Table S5 ) . The Chi abundance in P1-like phages shows that Chi sites can also be abundant in non-integrative temperate phages . T5 is a virulent phage showing that Chi over-representation is not limited to temperate phages . The reliable identification of presence or absence of specific RecBCD inhibitors is difficult in non-lambdoids because of the phage diversity and the tendency of RecBCD inhibitors to be small family-specific and fast-evolving proteins . Yet , these results suggest that Chi-dependent recombination might be widespread among phages packaging concatemeric DNA , for which recombination is important , even among virulent phages and non-integrative temperate phages . Dilemmas between manipulation and supersession of host functions are probably common in viruses . For example , some phages encode tRNAs to complement the host's repertoire [49] and some filamentous phages encode their own secretion apparatus whereas others manipulate their host's secretion systems [50] . In fact , pathogenic bacteria or protozoa manipulating host functions might also face similar trade-offs [51] . Understanding why different parasites evolved to manipulate host functions or to encode their own , can provide important clues on their mechanisms of virulence and , as we showed , of their evolvability .
The complete genomes of Escherichia ( 47 E . coli , 1 E . fergusonii ) and Salmonella ( 20 S . enterica and 1 S . bongori ) were downloaded from NCBI RefSeq ( ftp://ftp . ncbi . nih . gov/genomes/ ) . We analyzed a total of 275 phages including 38 lambdoid phages infecting enterobacteria ( downloaded from RefSeq ) and 237 long ( >30 kb ) non-redundant lambdoid prophages from the genomes of the abovementioned species identified in [10] with different mobile element detections [52]–[55] ( see also Text S1 and Table S6 ) . Among the 131 non-lambdoid genomes , 80 phage genomes of the Caudovirales order ( 69 virulent and 11 temperate ) were downloaded from RefSeq ( when classified in a genus defined by the ICTV ) . And 51 non-lambdoid prophages were identified in [10] with the same criteria ( >30 kb and non-redundant ) . The core genomes of E . coli and S . enterica were computed as described previously [10] . The pan genomes were computed from the 141 Rec+ lambdoid phages ( 9108 proteins ) , the 134 Rec− lambdoid phages ( 7554 proteins ) , and also the 163 Lambda-like phages of E . coli ( 9856 proteins ) . Homologous proteins were defined as pairs of proteins with more than 40% sequence similarity , computed using a Needleman-Wunsch end gap free alignment algorithm with the BLOSUM62 matrix , and with less than 50% of difference in length . Protein families were built from the pairwise analyses by transitivity , i . e . a protein is included in the family if it shares a relation of homology to a protein already in the family . The pan genome is the set of all different protein families . We excluded Genbank entries NC_004913 , NC_004914 and NC_003525 from this analysis because their annotations over-predict the number of genes ( nearly three times more genes per kilobase than the average lambdoid phage ) . We compared all lambdoid phage proteins to each other using blastp ( e-value<0 . 001 ) . The resulting blast bit score was used to cluster the proteins with MCL [56] . After testing the MCL inflation parameters in the range [1 . 2 to 5 . 0] , we used I = 3 . 0 because it was the smallest that produced protein clusters where all proteins of each cluster could be analyzed in a single multiple alignment . A total of 1812 protein clusters were obtained for the 16662 proteins analyzed . We aligned the proteins of each cluster with MUSCLE v3 . 6 [57] and built protein profiles with the HH-suite v2 . 0 . 9 [58] . The protein profiles of recombinases were initially found by comparison with published profiles [24] using HHsearch ( profile-profile comparison , p>95% in local and global alignments and >50% of profile coverage ) . We identified initially a subset of 14 protein clusters . To exclude helicases with ATPase domains from recombinases [24] we also made profile-profile comparisons with PFAM-A profiles ( downloaded the 11/25/2011 ) using HHsearch ( same parameters ) . We excluded the clusters matching PFAM-A profiles annotated as helicases ( e . g . DnaB , helicase-ATPase domain , DEAD/DEAH box helicase , PIF1-like helicase ) , producing a final set of 8 protein clusters of recombinases . This corresponds to 141 proteins found in 141 lambdoids . Our procedure was able to find all of the recombinases previously identified in lambdoid phages of enterobacteria [24] . We searched lambdoid phage genomes for hits of PFAM profiles of Gam ( PF06064 ) and Abc2 ( PF11043 ) proteins using HMMER v3 . 0 ( c-value<10−5 ) [59] . A total of 95 RecBCD inhibitors were detected: 56 Gam and 39 Abc2 proteins . The families of RecBCD inhibitors from T7 ( gp5 . 9 , NP_041987 ) , Enterococcus phage BC-611 ( ORF41 , BAM44931 ) , Clostridium phage phi8074-B1 ( phi8074-B1_00044 , AFC61976 ) and the DNA end protector from T4 ( gp2 , NP_049754 ) , have not been described among enterobacterial temperate phages . Indeed , we found no significant BLASTP hits ( at a threshold of e-value<0 . 001 ) to these proteins in our dataset of lambdoid phages . We used R'MES v3 . 1 . 0 to search the non-degenerated Chi motif 5′GCTGGTGG3′ and to compute significance of Z scores [60] . We computed the number of expected and observed Chi motifs accounting for the oligonucleotide composition separately for each genome . This was done to avoid putting together different phage genomes , which differ extensively in terms of nucleotide composition [61] . Four statistical models were analyzed for each genome . 1 ) The simplest model ( M0 ) accounts only for nucleotide composition . 2 ) The M2 model accounts for the composition in tri-nucleotides . 3 ) The M4 model accounts for the composition in penta-nucleotides . 4 ) The maximal model ( M6 ) accounts for the frequency of the maximal sub-strings of Chi motifs , i . e . hepta-nucleotides . The four models produced concordant statistics ( Table S1 ) . The M0 model is a poor predictor of random usage of large oligonucleotides because these are also affected by selection on other smaller oligonucleotides such as codons [62] . Phage genomes are small ( <50 kb on average ) and the Chi motif is slightly degenerated [63] . These two traits hinder the statistical power of the M6 and M4 models . Therefore we show in the text the results of the M2 model . The statistical significance of Chi sites over or under-representation in a given set of phages was computed using the Mann-Whitney test . Chi sites over-representation per genome was assessed by the Z score computed with R'MES . We computed all models under the compound Poisson approximation that is more adequate for low counts [60] . We initially aimed at using classical phylogenetic approaches to identify recombination events . Unfortunately , no proteins are ubiquitous to the whole set of 163 Lambda-like phages of E . coli . We therefore designed a method to find highly similar pairs of homologous proteins in two otherwise distantly related phages , which are likely the result of recent recombination events ( mosaic genes ) . This approach resembles closely that of [64] . First , we constructed the multiple alignment of each protein family of the pan genome of Lambda-like phages of E . coli with MUSCLE v3 . 6 [57] . Second , we extracted the informative positions in the alignments using BMGE with the BLOSUM30 matrix [65] . The 19 ( 4% ) protein families with trimmed alignments shorter than 50 sites were excluded due to the lack of phylogenetic signal . Third , we computed the protein distances ( di , jF ) of each pair of homologous proteins between two phages i and j in every protein family using TREE-PUZZLE v5 . 2 [66] . The distance matrix was computed using maximum likelihood under automatic estimation of the best substitution model and a Γ ( 8 ) correction for rate heterogeneity . Fourth , the distance matrix between phages Di , j was defined as the mean value of di , j for the orthologs shared by each pair of phages i and j . For each pair of phages , orthologous proteins were defined as unique reciprocal best hits with more than 40% similarity in amino acid sequence and less than 50% of difference in protein length . Finally , mosaic genes were identified as the ones encoding highly similar homologous proteins in highly dissimilar genomes for different thresholds Td and TD . More precisely , a pair of homologous genes between two phages i and j was regarded as mosaic if the encoded proteins were closey related ( dij<Td ) and the two phages were distantly related ( Di , j>TD ) . The different thresholds tested Td and TD showed qualitatively similar results . We did not analyze recombination events in genes encoding recombination functions , because they are absent from Rec− phages . We also ignored transposable elements , because they are self-mobilizable . | Bacterial viruses , called bacteriophages , are extremely abundant in the biosphere . They have key roles in the regulation of bacterial populations and in the diversification of bacterial genomes . Among these viruses , lambdoid phages are very abundant in enterobacteria and exchange genetic material very frequently . This latter process is thought to increase phage diversity and therefore facilitate adaptation to hosts . Recombination is also essential for the replication of many lambdoid phages . Lambdoids have been described to encode their own recombination genes and inhibit their hosts' . In this study , we show that lambdoids are split regarding their capacity to encode autonomous recombination functions and that this affects the abundance of recombination-related sequence motifs . Half of the phages encode an autonomous system and inhibit their hosts' . The trade-off between superseding and manipulating the hosts' recombination functions has important consequences . The phages encoding autonomous recombination functions have more diverse gene repertoires and recombine more frequently . Viruses , as many other parasites , have small genomes and depend on their hosts for several housekeeping functions . Hence , they often face trade-offs between supersession and manipulation of molecular machineries . Our results suggest these trade-offs may shape viral gene repertoires , their sequence composition and even influence their evolvability . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2013 | Manipulating or Superseding Host Recombination Functions: A Dilemma That Shapes Phage Evolvability |
Translocation of the Helicobacter pylori ( Hp ) cytotoxin-associated gene A ( CagA ) effector protein via the cag-Type IV Secretion System ( cag-T4SS ) into host cells is a hallmark of infection with Hp and a major risk factor for severe gastric diseases , including gastric cancer . To mediate the injection of CagA , Hp uses a membrane-embedded syringe-like molecular apparatus extended by an external pilus-like rod structure that binds host cell surface integrin heterodimers . It is still largely unclear how the interaction of the cag-T4SS finally mediates translocation of the CagA protein into the cell cytoplasm . Recently certain carcinoembryonic antigen-related cell adhesion molecules ( CEACAMs ) , acting as receptor for the Hp outer membrane adhesin HopQ , have been identified to be involved in the process of CagA host cell injection . Here , we applied the CRISPR/Cas9-knockout technology to generate defined human gastric AGS and KatoIII integrin knockout cell lines . Although confocal laser scanning microscopy revealed a co-localization of Hp and β1 integrin heterodimers on gastric epithelial cells , Hp infection studies using the quantitative and highly sensitive Hp β-lactamase reporter system clearly show that neither β1 integrin heterodimers ( α1β1 , α2β1 or α5β1 ) , nor any other αβ integrin heterodimers on the cell surface are essential for CagA translocation . In contrast , deletion of the HopQ adhesin in Hp , or the simultaneous knockout of the receptors CEACAM1 , CEACAM5 and CEACAM6 in KatoIII cells abolished CagA injection nearly completely , although bacterial binding was only reduced to 50% . These data provide genetic evidence that the cag-T4SS-mediated interaction of Hp with cell surface integrins on human gastric epithelial cells is not essential for CagA translocation , but interaction of Hp with CEACAM receptors is facilitating CagA translocation by the cag-T4SS of this important microbe .
Secretion systems of Gram-negative bacteria have evolved to mediate the passage of macromolecules across two or more cellular membranes , either into the extracellular space , or directly into selected host target cells [1] . A highly versatile group represents the bacterial Type IV secretion systems ( T4SS ) , which can transport diverse components in a contact-dependent manner , ranging from single proteins to protein-protein and protein-DNA complexes [2 , 3] . One of these bacteria is Helicobacter pylori ( Hp ) , which is recognized as one of the most prevalent bacterial pathogens worldwide and very efficiently utilizes the cytotoxin-associated gene ( cag ) type IV secretion system ( cag-T4SS ) as a major virulence determinant [4 , 5] . The effector protein CagA , together with a set of 27 proteins acting as structural and/or regulatory elements of the T4SS , are encoded on the cag pathogenicity island ( cagPAI ) , approximately 40 kb in size . Upon host cell contact the cag-T4SS forms needle-like surface appendages , the T4SS pili [6–8] , which are involved in the translocation of CagA from cell-adherent Hp across the bacterial and epithelial membranes into the host cell cytoplasm [8] . Our view on these fascinating nanomachines was extended recently by ultrastructural insights into the cag-T4SS-dependent membranous pilus-like appendages by in vivo electron cryotomography [9] . Injected CagA is tyrosine-phosphorylated on multiple Glu-Pro-Ile-Tyr-Ala ( EPIYA ) motifs in the C-terminal region , allowing its interaction with a set of cellular target proteins [10 , 11] . This results in dysregulation of the homeostatic signal transduction events in gastric epithelial cells , in loss of cell polarity , chronic inflammation and malignancy , qualifying CagA as a bacterial oncoprotein [12] . The cag-T4SS targets host cells via β1 integrin receptors [13 , 14] , and induces in these cells the production and secretion of proinflammatory cytokines and chemokines , such as interleukin-8 ( IL-8 ) [15] . The pilus-associated protein CagL has originally been reported to interact via an arginine-glycine-aspartate ( RGD ) motif with the α5β1 integrin heterodimer and thereby to activate Src and focal adhesion kinase , however , the requirement of the RGD motif for T4SS functionality was assessed differently [13 , 14] . Other cagPAI proteins , including CagY , CagI and CagA , have also been identified as interacting with the α5β1 integrin and an integrin binding domain for CagA was identified [16 , 17] . In addition to receptor binding by the cagT4SS itself , the outer membrane protein HopQ was identified to support CagA translocation by acting as a non-cagPAI-encoded cofactor of T4SS function [18] . Later on , HopQ was found to selectively bind a set of receptors from the carcinoembryonic antigen-related cell adhesion molecule family ( CEACAMs ) . CEACAM1 , CEACAM3 , CEACAM5 and CEACAM6 were identified as functional receptors for Hp via the outer membrane protein HopQ [19 , 20] . Hp-CEACAM binding not only plays a role for Hp adherence , but this interaction deeply contributes to the process of CagA translocation . Thus , the human embryonic kidney cell line ( HEK293 ) , which is devoid of CEACAM receptors on its surface , was resistant for CagA injection by Hp , but became readily susceptible upon functional expression of CEACAM1 or CEACAM5 on its surface [19 , 20] . The purpose of this study was to further dissect the role of integrin receptors versus the function of CEACAM receptors for the cagT4SS in the process of CagA translocation . Using the CRISPR/Cas9 system , we systematically generated single to multiple integrin knockout epithelial human cell lines ( AGS and KatoIII ) ending up with KatoIII cells without any integrin heterodimers on their surface . Unexpectedly , CagA translocation into these completely integrin-deficient cells was not significantly changed , suggesting that other integrin-independent Hp–host cell interactions must be important . In contrast , CRISPR/Cas9-mediated knockout of CEACAM receptors ( CEACAM1 , CEACAM5 and CEACAM6 simultaneously ) generated in KatoIII cells resulted in a strong reduction of CagA translocation capacity by Hp , suggesting that β-integrin receptors play a minor role in the T4SS-mediated CagA translocation , but the Hp-CEACAM interaction is of major importance .
The integrin receptor family is composed of 24 distinct integrin heterodimers , generated by different α and β subunits . Generally , integrin receptors follow a distinct tissue- and cell type-specific expression pattern in epithelial cells , leukocytes or platelets [21] . Thus , six β1 integrin heterodimers ( α1β1 , α2β1 , α3β1 , α5β1 , α6β1 and α9β1 ) , two αv integrins ( αvβ5 and αvβ6 ) and the integrin α6β4 are known to be epithelial-specific ( Fig 1A ) [21] . AGS and KatoIII cell lines are generally used as model systems for the evaluation of CagA translocation , since both cell lines were derived from human gastric epithelial cells . To get an overview of integrin expression on the surface of these cells , we stained them with different integrin-specific antibodies and determined the integrin expression profile by flow cytometry . AGS and KatoIII cells indeed produced β1 integrins ( including α1β1 , α2β1 , α3β1 , α5β1 , α6β1 and α9β1 ) , αv integrins ( αvβ5 and αvβ6 ) ( αvβ8 only by KatoIII ) and the β4 integrin ( α6β4 ) on their surface , however with varying expression levels ( Fig 1B and 1C ) . We planned to generate a β1 gene knockout in AGS cells that should lack surface expression of all potential β1 containing integrins , since the targeting of either subunit of a given integrin heterodimer should ultimately result in the depletion of the targeted integrin heterodimer [22 , 23] . In order to obtain integrin knockout cell lines without undesired off-target mutagenesis , the double nicking strategy was applied [24] . For design of paired short guide RNAs ( sgRNAs ) targeting the integrin β1 ( ITGB1 ) gene , the online CRISPR design tool ( http://tools . genome-engineering . org ) was used for optimal sgRNA analysis and identification ( See further details of the method in Experimental Procedures ) ( Fig 1D and S1 Table ) . For generation of a β1 integrin deficient AGS cell line , verified CRISPR constructs targeting exon 5 of the β1 integrin gene ( ITGB1 ) were transfected into AGS cells . Transfected cells went through a selection procedure to obtain knockout cell lines . Since CRISPR constructs contain the puromycin resistance gene , the transfected population was treated with puromycin to kill non-transfected cells . The surviving cells were stained with integrin β1 antibody for negative selection by FACS sorting . Finally , serial dilutions of the sorted negative populations resulted in stable cell lines , which could be verified as completely integrin β1-deficient by flow cytometry analysis ( Fig 2A ) . Furthermore , the complete absence of the gene product was verified by ( i ) demonstrating the disruption of the targeted gene sequence by PCR amplification and sequencing of the integrin β1 alleles ( S1A Fig ) and ( ii ) by immunoblotting of cell lysates using a β1 integrin-specific antibody ( S2A Fig ) . Next , the verified β1 integrin-deficient AGS cells were tested for CagA translocation capacity by Hp . Traditionally , CagA translocation is assessed by detecting tyrosine-phosphorylated EPIYA motifs as a phosphorylated CagA band via western blot . This can be used for quantification , but is not very sensitive and accurate . We have recently established a sensitive β-lactamase reporter system ( TEM-1 reporter assay ) to accurately determine Hp CagA translocation into host cells independently of its tyrosine phosphorylation and host cell kinase activity [25] . When applying the Hp strain P12[TEM-CagA] in the TEM-1 reporter assay , we surprisingly did not observe a significant difference in CagA translocation into AGS wild type versus β1 integrin-deficient cells ( Fig 3A ) . This observation suggested that the β1-integrin interaction was apparently not essential for the bacteria or the cag-T4SS to inject CagA . One possible explanation for this unexpected result might be that other integrin heterodimers ( αvβ5 , αvβ6 , or α6β4 ) , which are known to be expressed on AGS cells ( see Fig 1A ) , are able to functionally substitute β1 integrin heterodimers regarding CagA translocation . We therefore extended the CRISPR/Cas9-mediated knockout strategy to inactivate integrins αv and β4 separately , using the same procedure as for integrin β1 ( see S3 Fig and S4 Fig for design of integrin gene inactivation strategy ) . Furthermore , by targeting different combinations of two of the aforementioned genes in the same cell , different combinations of double mutants were obtained ( ΔITGAvB4 , ΔITGB1B4 ) , which did not produce the corresponding integrins on the cell surface , as determined by flow cytometry ( Fig 2B–2E ) . A concomitant knockout of all three integrin genes , which ultimately should result in cell lines devoid of all integrins on the AGS cell surface , could not be obtained in the AGS cell background , probably because they do not survive . The correct genetic inactivation of the integrin genes was verified by PCR amplification and sequencing of the corresponding αv- or β4-specific integrin alleles ( S1B and S1C Fig ) . The complete absence of the gene products was confirmed by immunoblotting with integrin αv- or β4-specific antibodies ( S2A Fig ) . We next asked whether obvious differences in the morphology , physiology or function of the integrin-knockout AGS derivatives are apparent compared to wild type AGS cells . A thorough study of the general cell morphology did not show any peculiarities . AGS AvB4 cells did only grow in tissue culture when collagen was added , which indicated that the integrin receptor-mediated binding to certain integrin ligands was absent . All wild type and knockout mutant cells showed the hummingbird phenotype ( S5 Fig ) . IL-8 induction in AGS wild type cells was slightly reduced in the P12ΔhopQ infecting strain , as compared to the P12 wild type ( wt ) and the complemented mutant strain ( S6 Fig ) . Interestingly the level of IL-8 induction was generally higher when integrin knockout cells were used as compared to AGS wild type cells , but the general pattern of reaction of the knockout versus the wild type cells was well conserved ( S6 Fig ) . This indicates that the main phenotypic characteristics of the knockout cells are still conserved in comparison to wild-type cells , arguing against unexpected compensatory mutations or significant alterations in signal transduction networks in the knockout cells . Interestingly , infection experiments based on the TEM-1 reporter assay showed no statistically significant difference in CagA translocation efficiency into AGS wild type versus single or multiple integrin αv- or β deficient cells ( Fig 3A ) . Similar results were obtained by the conventional tyrosine phosphorylation experiments upon infection of the mutant AGS epithelial cell lines ( Fig 3C ) . In conclusion , we demonstrate here that Hp is able to translocate its CagA protein into gastric epithelial AGS cells devoid of most integrin receptors on their surface , although a complete integrin-free state could not be obtained in the AGS cell background . In order to compare our data obtained from AGS gastric epithelial cells with another independent human gastric cell line we chose the KatoIII cells for integrin gene knockout experiments . The same strategy and knockout plasmids were applied . We finally obtained a total of seven stable integrin-deficient KatoIII cell lines , all of which could be verified to be completely devoid of their corresponding cell surface integrins , as determined by flow cytometry ( Fig 2F–2M ) . These included the single knockout cell lines ( ΔITGB1 , ΔITGAv , ΔITGB4 ) , the double knockout cells ( ΔITGB1B4 , ΔITGAvB4 , ΔITGAvB1 ) as well as a triple knockout cell line ( ΔITGB1AvB4 ) . The latter cell line indeed lacks all integrins we tested for by specific antibodies , as demonstrated by the absence of the αv and all individual β integrin subunits ( β1 –β8 ) on the cell surface ( see scheme Fig 1A and S7B Fig ) . Next , KatoIII wild type and the corresponding knockout cell lines were analyzed by immunoblotting with the corresponding anti-integrin antibodies to confirm the complete absence of the gene product ( S2B Fig ) . On the genetic level knockout mutations could be verified by sequencing of each gene ( S1 Fig ) . We then performed infection experiments to quantify CagA translocation for all seven different KatoIII integrin-knockout cell lines . Again , single or multiple integrin knockout cell lines did not show a significantly different CagA translocation efficiency as compared to wild type KatoIII cells ( Fig 3B ) . Using a plate reader assay for adherent AGS cells , or a flow cytometry approach for KatoIII suspension cells , we next quantified and compared CagA translocation of different Hp P12[TEM-CagA] strains . They comprised a hopQI gene deficient strain ( P12ΔhopQ[TEM-CagA] ) , a genetically complemented hopQI knockout strain ( P12ΔhopQ:hopQ[TEM-CagA] ) and a strain that served as a negative control for CagA translocation ( P12ΔcagT[TEM-CagA] ) . The outer membrane protein HopQ has been recently identified as a major Hp adhesin binding to host cell CEACAMs and was found to be a major contributing factor for CagA translocation [19 , 20] , whereas in AGS wild type and integrin knockout cells CagA translocation by a HopQ-deficient strain was generally reduced ( Fig 3A ) . In contrast to AGS cells , CagA translocation by the P12ΔhopQ[TEM-CagA] strain into wild type KatoIII , as well as into integrin knockout KatoIII cell lines was almost completely abolished ( Fig 3B ) . The genetically complemented strain ( P12ΔhopQ:hopQ[TEM-CagA] ) was restored in its ability for CagA translocation ( Fig 3A and 3B ) . In summary , these data support our results obtained with AGS cells . Furthermore they suggest that in AGS cells ( an ) other receptor ( s ) distinct from CEACAMs seem ( s ) to support the process of CagA translocation , as shown by infection assays with a P12ΔhopQ[TEM-CagA] strain ( Fig 3A ) . Such ( a ) receptor ( s ) is/are apparently absent in the KatoIII cell background , where CagA translocation seems to be mostly dependent on the HopQ-CEACAM interaction ( Fig 3B ) . We next performed an integrin profiling in each integrin depletion cell line to investigate whether the depletion of individual integrins can influence the expression levels of the remaining integrins . Especially increased expression levels of remaining integrins could be a reasonable explanation for the sustained CagA translocation efficiency in the different integrin deficient cell lines . To cover all kinds of integrin combinations , also aberrant expression of non-epithelial integrin heterodimers , we analyzed integrin knockout cell lines ( AGS and KatoIII background ) for expression of integrin αv , β1 , β2 , β3 , β4 , β5 , β6 , β7 and β8 by flow cytometry using specific antibodies ( S7 Fig ) . Among them , non-epithelial integrins β2 , β3 , β7 and β8 were not differently expressed by any of the mutant versus the wild type cell lines . Integrins β5 and β6 were absent only in αv knockout cells ( see black arrows ) , probably due to the loss of their exclusive alpha integrin binding partner ( S7 Fig and Fig 1A for scheme ) . In addition , the KatoIII β1 KO cell line showed a significant reduction in integrin αv and β5 surface localization as compared to wild type KatoIII cells ( S7B Fig , black arrows ) . The remaining integrins expressed on the surface of each integrin-deficient cell line exhibited similar expression levels as found on wild type AGS or KatoIII cells . This reduced CEACAM expression in KatoIII knockout cells could be responsible for a slightly but not significantly reduced CagA translocation capacity of the integrin double and triple knockout cells ( Fig 3B ) . Elevated expression levels of remaining integrins , or aberrant expression of non-epithelial integrin heterodimers could be a reasonable explanation for the sustained CagA translocation efficiency in the integrin or CEACAM deficient cell lines . However , we can safely exclude this possibility after intensive integrin profiling experiments . Most importantly , no unexpected additional integrin subunit ( s ) appeared on the cell surface , even in the triple integrin knockout KatoIII cells . This indicates that these cells do not bear any integrin on their surface to be exploited for CagA translocation by Hp . Thus , our data unequivocally demonstrate for the first time that the apparent complete absence of any integrin on the cell surface does not have a significant effect on the capacity of Hp to translocate CagA into these cells . So far our data clearly show that integrin receptors on the surface of AGS or KatoIII cells do not have a major impact for CagA translocation capacity of Hp , but the loss of the adhesin HopQ strongly reduced CagA translocation , especially in KatoIII cells . The question arose whether it is sufficient for CagA translocation to just mediate a physical tethering of the bacteria to the cell surface , or whether a special interplay between HopQ and CEACAMs , which may result in a very tight or close binding to the cell surface , is needed to facilitate CagA translocation ? To address these questions , we next generated in KatoIII cells a triple CEACAM knockout ( CEACAM1/5/6 knockout ) using the CRISPR/Cas system ( S8 Fig ) and verified the knockout status by flow cytometry , immunoblotting and sequencing ( Fig 4A and 4B and S9 Fig ) . Attempts to combine the integrin triple knockout and the CEACAM triple knockout mutations in KatoIII were not successful , since corresponding mutant cell lines did not survive . Notably , the triple integrin knockout KatoIII cells showed a 50% reduced expression of CEACAM5 as compared to the wild type cells ( Fig 4B and 4C ) , but a 4 . 5 fold increase in CEACAM1 expression , which might be explained by the fact that both surface receptors are usually found in the same lipid background and even might interact with each other [26] . This disturbance of CEACAM expression might have some influence on the slightly lower CagA translocation activity of Hp into the integrin double and triple knockout cells that we can always observe , although this difference is not statistically significant ( Fig 3B ) . For CEACAM1/5/6 KO KatoIII cells no significant changes in the intrinsic integrin αv or β expression pattern could be observed ( S10 Fig ) . We then infected the CEACAM triple knockout KatoIII cells with the P12[TEM-CagA] strain to quantify CagA translocation . As expected from the results with the P12ΔhopQ[TEM-CagA] strain , the CEACAM triple knockout showed a nearly complete loss of CagA translocation , comparable to and in support of the P12ΔhopQ[TEM-CagA] strain results ( Figs 3B and 5B ) . Similar results were obtained in the CagA tyrosine phosphorylation assay for strain P12 and other Hp lab strains ( Fig 5C and 5D ) . Thus , in KatoIII cells the HopQ-CEACAM interaction seems to be the major driver/mediator for CagA translocation . Next , an important question was how much of the total adhesion of the bacteria can be attributed to the HopQ-CEACAM interaction and is binding per se , independent of the type of host cell receptor , sufficient to allow CagA translocation . Interestingly , the binding capacity of a P12-GFP strain to KatoIII wild type versus the CEACAM triple knockout KatoIII cells was reduced to a level of about 75% ( Fig 5A ) , whereas the CagA translocation was nearly completely abolished under these circumstances ( Fig 5B ) . These data suggest that binding per se is not sufficient for Hp to induce CagA injection . It seems that the HopQ-CEACAM interaction mediates a ( n ) additional signal ( s ) to initiate CagA injection . To further study potential changes in the interaction of Hp with cells lacking all surface integrin receptors , or the relevant CEACAM receptors , we performed confocal microscopy studies using KatoIII wild type cells , KatoIIIΔαvβ1β4 and KatoIIIΔCEACAM1/5/6 cell lines infected with P12 wild type or P12ΔhopQ strains ( Fig 6A–6C ) . We typically find a reduced number of Hp binding to KatoIIIΔαvβ1β4 and KatoIIIΔCEACAM1/5/6 cell lines as compared to KatoIII wild type cells , and the binding pattern of Hp to integrin-deficient cells appears to be different . Interestingly , KatoIIIΔCEACAM1/5/6 cell lines produce large amounts of β1 integrin ( Fig 6C ) , and Hp is found closely attached to β1 integrin , although under these conditions very little CagA translocation was found ( Fig 5B ) . Thus , we see for each cell line an intimate interaction of the bacteria with the host cells , independent of the capacity for CagA translocation of the strain ( Fig 6A–6C , white arrowheads ) . Triple integrin knockout cells show a high number of adherent bacteria ( Fig 5A ) but a lower expression of CEACAM5 ( Fig 4B and 4C ) . This is also visible by a lack of CEACAM5 recruitment to the bacterial surface in the triple integrin knockout cells , which is in stark contrast to the CEACAM5 receptor recruitment seen in Hp-infected KatoIII wild type cells ( Fig 6A , versus B and C; yellow arrows ) . Notably , CagA translocation into these cells is not significantly reduced as compared to the KatoIII wild type cells ( Fig 3B and 3C ) .
It is well established that the cagPAI-encoded T4SS is a major Hp virulence determinant , the function of which has been implicated in severity of disease and increased risk of gastric cancer [27] . A major role of the cag-T4SS is the translocation of the CagA protein into various types of host cells , where CagA interferes in a phosphorylation-dependent and phosphorylation-independent manner with signaling events to manipulate fundamental processes in the gastric epithelium [28] . Major outcomes include the suppression of innate defense mechanisms [29] , changes in cell polarity and migration [30 , 31] , and putatively oncogenic events [32 , 33] . The involvement of a host cell integrin heterodimer ( α5β1 or any other β integrin heterodimer ) acting as receptor for the Hp T4SS , especially for the pilus-associated RGD containing CagL protein , was considered as a major requirement for CagA translocation [13 , 14] [34 , 35] . Several labs have provided data showing the interaction of integrin α5β1 or other αβ integrin heterodimers with different components of the cag-T4SS , especially CagL [13 , 34–41] , but also CagA [16] , CagI [42] and CagY [14 , 42] . Several previous studies suggested that integrins are required for CagA translocation . The major evidence for a functional role of β1 integrins as receptor for the cag-T4SS and translocation of CagA was coming from studies using β1 integrin-deficient murine fibroblast ( GD25 ) or epithelial ( GE11 ) cell lines , which did not support CagA translocation , whereas the corresponding β1 integrin complemented versions resulted in CagA translocation and phosphorylation [13 , 14] . We report now that integrin heterodimers are not required for Hp to translocate its CagA into gastric epithelial cell lines in vitro . We were interested in a better understanding of the role and the contribution of the recently identified CEACAM receptors versus the integrin receptors for CagA translocation . Therefore we generated a set of knockout cell lines and measured their ability for CagA translocation . The β-lactamase reporter system determines Hp CagA translocation into host cells in a very sensitive , reproducible and quantitative way [25] . Most important , it measures the translocation of CagA directly , rather than its tyrosine phosphorylation . The tyrosine phosphorylation depends on the activity of host cell kinases c-Src or c-Abl , which might be affected in their activity by manipulations of the cells , such as different growth conditions , buffer treatments or procedures like the gene knockout technology . Taking advantage of the CRISPR/Cas technology we started to knock out important integrins in an additive fashion to generate single ( β1 ) , double ( β1β4 , αvβ4 , β1αv , ) AGS and KatoIII cells and finally a triple integrin gene knockout cell ( αvβ1β4 ) in the KatoIII background . All CRISPR/Cas constructs and targeted cell lines for the generation of integrin-depletion AGS and KatoIII cell lines are summarized in S2 Table . AGS cells are adherent cells , which did not survive as triple knockout without any integrin on the surface . In AGS cells growing as attached cells on solid surfaces this phenomenon might be due to the induction of anoikis , a tissue architecture surveillance mechanism , which can be induced by the absence of integrin-ECM ligation to assure that dissociated and displaced cells are effectively eliminated , in order to prevent dysplastic growth [43 , 44] . KatoIII cells , which grow in a semi-adherent manner , were resistant to anoikis and allowed the generation of a triple integrin knockout . The triple integrin knockout KatoIII cell line , which is devoid of any integrin receptor on the surface , was still competent for CagA translocation . From these results we have to conclude that neither the direct interaction of components of the cag-T4SS with integrins , nor any integrin-mediated signaling event is necessary for CagA translocation . The data presented above seem to be in opposition to earlier publications [13 , 14] , which demonstrated that CagA translocation was possible in murine β1 integrin expressing GE11 or GD25 cells , but not in the corresponding integrin knockout cells . However , neither the murine GE11 and GD25 , nor the chinese hamster ovary ( CHO ) cell line , also used for such experiments , contain human CEACAMs for binding of the Hp adhesin HopQ . This might at least explain why in our earlier experiments only very low ( background ) levels of CagA translocation could be observed in the β1 integrin-complemented versions of these cells [14] . From our new perspective , the earlier integrin complementation data can be interpreted that the interaction of the cag-T4SS to integrins in cells without human CEACAM receptors has only a small supportive effect for CagA translocation . However , the integrin knockout data presented in this study clearly show that integrins are not necessary for CagA translocation in AGS or KatoIII gastric cell lines . Using hamster ( CHO ) cell lines devoid of human beta integrins , but genetically complemented with integrin genes encoding fully functional or partial integrins ( CHO K1 , CHOβ1TR ) we demonstrated that the extracellular part of β1 integrin was supportive for CagA translocation , but the cytoplasmic tail of β1 integrin was not necessary [14] . We also reported that neither the RGD motif in CagL for binding the β1 integrin heterodimer , nor the function of the integrin linked kinase ( ILK ) were essential for CagA translocation . From that we concluded that no integrin-mediated signaling is involved in this process [14] . Interestingly , also knockdown experiments of integrin α5β1 and ILK showed that both were dispensable for NF-κB activation during Hp infection , but the bacterial adhesin HopQ promoted canonical NF-κB activation in AGS and NCI-N87 cells [45] . Combined these data suggest that integrin-mediated signaling is neither needed for CagA translocation nor NFκB activation . Earlier data also showed that a recombinant protein of CagA covering the binding site of β1 integrin can interfere with CagA translocation and phosphorylation [16] . A similar effect on CagA phosphorylation was seen with the β1 integrin-specific antibody 9EG7 [14] . These data were interpreted as direct effects of β1 integrin on CagA translocation , suggesting that β1 integrin is essential . In the context of our results in this study we would explain these data as more indirect effects , e . g . by steric hindrance exerted by the binding of the recombinant protein , or the 9EG7 antibody , on the function of other receptors , such as CEACAMs , which usually reside in the same lipid domains as integrins [26] . With these novel results the question arises why components of the cag-T4SS bind specifically and in some cases with high affinity ( CagA , KD values in low nanomolar range ) to α5β1 integrin heterodimers [14 , 42] although this binding apparently has only a very minor , functional relevance for CagA translocation ? The binding of the cag-T4SS components to the extracellular domains of β1 integrin heterodimers may allow , by tethering of the T4SS to the host cell , a low level CagA translocation , but for a full CagA translocation , the HopQ adhesin–CEACAM binding is necessary . Furthermore , we cannot exclude that in an in vivo situation , when Hp interacts with primary gastric cells in tissue , integrin signalling via CagL might play a role . CagA translocation might happen independent from integrin interaction , as our in vitro data suggest , but the activation of Src kinase might be necessary in primary , untransformed cells , but dispensable in transformed cell lines , in which these kinases often are constitutively active . Our group as well as other labs have shown that certain CEACAMs act as receptors for Hp and support CagA translocation when they are reconstituted in a cell line deficient of CEACAM expression ( e . g . HEK293 , CHO ) [18–20] . The genetic complementation of CEACAM-negative cells , such as CHO or HEK293 cells , showed a drastic effect on CagA translocation [18–20] . However , in contrast to the integrin knockouts , a complete genetic knockout of CEACAM1/5/6 in an epithelial cell line ( KatoIII ) more or less completely abrogated CagA translocation . This clearly suggests that CEACAM receptors are essential for CagA translocation in certain cell types . Besides CEACAMs , other surface receptors seem to exist , which can support CagA translocation . Thus , in AGS cells we see a reduction in CagA translocation to approximately 50% when a HopQ-deficient versus a wild type Hp strain is used for infection ( Fig 3A ) [19] , which is in contrast to KatoIII cells suggesting that in AGS gastric epithelial cells an additional , so far unknown receptor might be expressed , which is probably targeted by another Hp adhesin to support CagA translocation . This receptor might be absent in other cell lines , such as KatoIII cells . Earlier work described a small BabA-Leb mediated but cag-T4SS-dependent effect on the production of proinflammatory cytokine mRNA expression ( IL-8 , CCL5 ) and a very minor effect on CagA translocation of Hp . This Leb dependent augmentation of cagPAI T4SS functions , which was seen in Leb-negative non-human and non-gastric CHO or Madin-Darby canine kidney cells ( MDCK ) transfected with several glycosyltransferase genes [46] , was independent of CEACAMs , since these cells do not produce CEACAMs recognized by HopQ . Besides the interaction of the bacteria with protein- or oligosaccharide cell surface receptors , the translocated effector protein CagA can bind phospholipids via a K-Xn-R-X-R motif , an amino acid sequence motif conserved among various pleckstrin homology ( PH ) domains directly involved in the interaction with acidic phospholipids , such as phosphatidylinositol ( PI ) and/or phosphatidylserine ( PS ) [47] . Murata-Kamiya and coworkers [48] reported that physical interaction of Hp CagA with host membrane PS , which is aberrantly externalized at the site of bacterial attachment by Hp , plays a key role in the delivery and intracellular localization of CagA . How the exploitation of CEACAM receptors by the adhesin HopQ , the functional buildup of the cag-T4SS secretion apparatus and the CagA binding to PS are coordinated and function to accomplish the internalization of CagA is still not well understood and the aim of intensive future research .
For TEM-1 reporter assays Hp wild type strain P12 [49] and defined P12 knockout mutants were used . To verify CagA translocation results into integrin-deficient AGS or integrin- or CEACAM-deficient KatoIII cells by tyrosine phosphorylation also other Hp strains were applied , such as 1-20A , TN2GF4 or G27 [50] . These strains harbor a functional cag pathogenicity island ( cagPAI ) in their genome , encoding the Cag T4SS [4] . Hp strains and mutants used in this study are listed in S3 Table . Hp strains were grown on GC agar plates ( Oxoid ) supplemented with vitamin mix ( 1% ) and horse serum ( 8% ) ( serum plates ) and cultured for 16–60 h in a microaerobic atmosphere ( 85% N2 , 10% CO2 , 5% O2 ) at 37°C . Escherichia coli strains Top10 and DH5alpha were grown on Luria–Bertani ( LB ) agar plates or in LB liquid medium [51] supplemented with antibiotics , as appropriate . Cell lines were cultivated under standard conditions [15] in 75 cm2 tissue culture flasks ( BD Falcon ) and subcultivated every 2–3 days in 6-well , 48- well ( tissue culture treated , Costar , Corning Inc . ) or 96-well microtiter plates ( black , transparent bottom , tissue culture treated , 4titude ) . Plasmids were introduced into Hp strains by transformation as described previously [52] . Hp transformants were selected on serum plates containing 6 mg l-1 chloramphenicol , 8 mg l-1 kanamycin , 10mg l-1 erythromycin or 250 mg l-1 streptomycin , as appropriate . AGS cells ( CRL-1739 ) and Kato III cells ( HTB-103 ) were obtained from ATCC . Cells were generally cultured in RPMI 1640 or DMEM supplemented with 10% ( vol/vol ) fetal calves serum ( FCS ) at 37°C and 5% CO2 . For passaging of the cells , the medium was removed and the cells were gently washed once with Dulbecco´s PBS ( DPBS , Life Technologies ) . To detach cells , 2 ml Trypsin-EDTA was added to a 75 cm2 flask for 3–5min incubation at 37°C . When detachment was observed under microscope , 8 ml of the pre-warmed RPMI1640 medium was added to neutralize the trypsin . After being gently pipetted up and down , cells were dissociated and were then reseeded into new flasks . Passages taken place in every 2–3 days with a split ratio of 1:5 or 1:8 . Cells were discarded when the passage number reached 80 . Generally , one day before transfection , 0 . 5 x 105 to 2 x 105 cells were plated in 500 μl of growth medium without antibiotics in a 24-well plate so that they would be 90–95% confluent at the time of transfection . For each transfection sample , 500 μg DNA and 2 μl lipofectamine 2000 was prepared according to the manufacturer´s instructions ( Lipofectamine 2000 , Invitrogen ) . After transfection the cells were incubated at 37°C in a CO2 incubator for 24–48 hours until they were ready to test for transgene expression . Cells were counted and added to a round-bottom 96-well plate with 2x105 cells per well . After centrifugation ( 300 g at 4°C ) primary antibodies were added to each well following the recommended concentration from the manufacturer . Dilutions of antibody , if necessary , were made in FACS buffer . Cells and antibodies were incubated at 4°C for 1 hour in the dark . Primary antibody stained cells were washed 3 times and resuspended in 200 μl to 1 ml of ice-cold FACS buffer . Subsequently , fluorochrome-labeled secondary antibodies were diluted in FACS buffer at the optimal concentration ( according to the manufacturer’s instructions ) and were added to each well , followed by 1 h incubation at 4°C in the dark and 3 times washing as described above . Cells were analyzed by flow cytometer right after washing or kept in the dark on ice until the scheduled time for analysis . The following antibodies were used: ITGA2 ( P1E6 , chemicon ) , ITGA3 ( P1B5 , chemicon ) , ITGAv ( P2W7 , Santa Cruz ) , ITGB1 ( AIIB2 , EMD Millipore ) , ITGB2 ( MEM-48 , antibodies-online GmbH ) ITGB3 ( VI-PL2 , antibodies-online GmbH ) , ITGB4 ( 439-9B , antibodies-online GmbH ) , ITGB5 ( AST-3T , antibodies-online GmbH ) , ITGB6 ( 437211 , antibodies-online GmbH ) , ITGB7 ( FIB504 , antibodies-online GmbH ) , ITGB8 ( 416922 , antibodies-online GmbH ) , CEACAM1 ( 8G5 , Genovac ) , ( CEACAM5 ( 26/3/13 , Genovac ) , CEACAM6 ( 9A6 , Genovac ) . This procedure is used for adherent cells , such as AGS cells . One day before infection , adherent cells were detached and 2 . 5 × 104 cells were seeded in each well in a 96-well plate with black wall and transparent bottom with low fluorescence background ( 4ortitude ) . The confluence of the cells was 80% to 90% on the day of infection . Before infection , Hp strains with fusion protein of beta-lactamase TEM-1 and CagA were collected as described before . Ideally , bacteria were resuspended and pre-incubated in sterile PBS containing 10% FCS at 37°C , 10% CO2 for 1 . 5 h . Subsequently , cells were infected by bacteria with an MOI of 60 for 2 . 5 h at 37°C , 5% CO2 as described above . Infections were stopped by placing the plates on ice and all the supernatants were removed . Prepared substrates mix is loaded immediately on the cell surface , followed by incubation at room temperature for 120 min in the dark . Plate reader filters are set to allow excitation of wavelength around 410nm , and detection of blue emission around 450nm and green emission around 520nm . Afterwards acquired data was normalized and analyzed following manufacturer’s instruction to obtain the blue to green fluorescence ratio . For suspension and semi-adherent cell lines , CagA translocation was detected by flow cytometry . The method of CagA translocation assay with flow-cytometry detection is very similar to the plate-reader detection except following procedures . Firstly , semi-adherent cells were detached after infection with room-temperature trypsin-EDTA before incubation with CCF4-AM fluorescence substrate mix . Secondly , incubation of cells with CCF4-AM mix were implemented at 27°C with constant shaking condition to allow even loading of cells with substrate and avoid cell sedimentation . Finally , cells were washed at least 2 times with PBS by centrifugation at 200–300 × g for 5 mins after incubation with CCF4-AM substrate . Cells were then analyzed by flow cytometry for Pacific Blue fluorescence and AmCyan green fluorescence . Rabbit polyclonal antisera AK268 and AK257 , directed against the CagA N-terminal and the CagA C-terminal region , respectively , have been described previously [5 , 53] . The mouse monoclonal antibody against TEM-1 β-lactamase was obtained from Abcam ( ab12251 ) . For immunoblotting the following antibodies were used: ITGAv , ( EPR16800 , abcam ) , ITGB1 ( LM534 , EMD Millipore ) , ITGB4 , ( 439-9B , abcam ) , CEACAM1/3/4/5/6 ( D14HD11 , Genovac ) . Standard infections of AGS and KatoIII cells with Hp strains and subsequent preparations for phosphotyrosine immunoblotting were performed as described previously [23] . Briefly , cells were plated one day before infection in 6-well plates . On the day of infection , cells were infected using an MOI of 60 for 2 . 5 h at 37°C and 5% CO2 . After washing with PBS , cells were collected by cell scrapers in the presence of 1 ml PBS* ( PBS containing 1 mM Na3V04 , 1 mM PMSF , 10 μg/ml leupeptin and 10 μg/ml pepstatin ) . Cells with adherent bacteria were collected by centrifugation and resuspended in SDS-PAGE sample solution . SDS PAGE and western blotting was performed as described previously [13] . For the development of immunoblots , polyvinylidene difluoride ( PVDF ) filters were blocked with 5% non-fat milk powder in TBS ( 50 mM Tris–HCl , pH 7 . 5 , 150 mM NaCl ) , 0 . 1% ( v/v ) Tween 20 ( TBS-T ) , and incubated with the respective antisera at a dilution of 1:1 . 000–1:15 . 000 in TBS-T with 1% non-fat milk powder . Horseradish peroxidase-conjugated anti-rabbit IgG antiserum was used to visualize bound antibody . Standard infections of AGS and KatoIII cells with Hp strains and subsequent preparations for phosphotyrosine immunoblotting were performed as described previously [5] . Tyrosine-phosphorylated proteins were analyzed by immunoblotting with the phosphotyrosine antibody PY99 ( Santa Cruz Biotechnologies ) . Single gel systems [54] were adapted for Stain-Free detection as described in protocol depository Protocols . io under dx . doi . org/10 . 17504/protocols . io . gipbudn . Western Blot data were quantified by densitometry using ImageJ . Band intensities of strain-free gel were normalized to the band intensity of KatoIII lane . CEACAM1 , 5 and 6 expression was measured as area percent of the respective lane and normalized to the CEACAM1 , 5 and 6 expression of KatoIII cells . Comparability between cell lines was achieved by standardizing each normalized CEACAM expression to the normalized loading controls . One day prior to experiments cells were seeded at 5 x 104 cells in a 24-well plate equipped with uncoated cover slides and grown overnight at 37°C and 5% CO2 . Cells were infected with Hp wild type or isogenic mutant strains with an MOI of 10 for 3h at 37°C and 5% CO2 . For immunostaining cells were fixed with 4% PFA for 10 min at room temperature . Cells were washed twice with Dulbecco´s PBS ( DPBS , Life Technologies ) and blocked overnight with 2% FCS in PBS at 4°C . Fixed cells were incubated with mouse anti-CEACAM5 ( 26/3/13 , Genovac , 1:300 ) , rabbit anti-Hp ( AK175 , 1:400 ) and rat anti-integrin beta1 ( AIIB2 , Millipore , 1:200 ) for 1h at room temperature . After washing secondary antibodies were applied ( goat anti-rat Alexa488 , goat anti-mouse Alexa555 and goat anti-rabbit Alexa647 all from Invitrogen , 1:1000 ) and incubated for 1h at room temperature in the dark . Cell nuclei were stained with DAPI ( 5μg/ml ) for 10 min . Samples were mounted on the cover slip with Fluorescent Mounting Medium ( DAKO ) . A cytospin3 ( Shandon ) was used to centrifuge suspension cells onto glass slides . Micrographs were taken with a confocal laser scanning microscope ( LSM880 , Zeiss ) with Airyscan Module using a 63x oil immersion objective . Statistical analysis was performed with GraphPad Prism 7 . 2 . Data were analyzed with One-way or Two-way analysis of variance ( ANOVA ) , as further specified in the legends of the corresponding figures . The significance level was set to 0 . 05 . If overall ANOVA tests were significant , a post hoc test ( Tukey’s HSD test or Bonferroni test ) was performed . Details for each experiment are described in the figure legends . | The Cag Type IV secretion system of Helicobacter pylori ( Hp ) interacts with host cell integrins and injects the bacterial oncoprotein CagA into host cells thereby contributing to inflammation and carcinogenesis during chronic infection . Binding of β1 integrin receptors by the CagA protein and the type IV secretion system is well described by many research groups , but its function for CagA translocation is not well understood . We report here that this interaction is not essential for the function of the secretion system and for CagA injection into the gastric epithelial cells lines AGS and KatoIII . However , the bacterial binding to a set of specific receptors called carcinoembryonic antigen-related cell adhesion molecules ( CEACAMs ) by the Hp outer membrane protein HopQ is a prerequisite for CagA translocation . Interestingly , other bacterial adhesins and the mediated binding events do not have a similar effect on CagA translocation , suggesting a specific feature associated with HopQ mediated binding . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"cell",
"binding",
"flow",
"cytometry",
"cell",
"physiology",
"bacteriology",
"phosphorylation",
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"viral",
"transmission",
"and",
"infection",
"molecular",
"probe",
"techniques",
"pathogens",
"microbiology",
"immunoblotting",
"epithelial",
"cells",
"integrins",
"secretion",
"systems",
"molecular",
"biology",
"techniques",
"cellular",
"structures",
"and",
"organelles",
"research",
"and",
"analysis",
"methods",
"microbial",
"physiology",
"animal",
"cells",
"cell",
"adhesion",
"proteins",
"extracellular",
"matrix",
"biological",
"tissue",
"molecular",
"biology",
"spectrophotometry",
"cytophotometry",
"bacterial",
"physiology",
"biochemistry",
"host",
"cells",
"cell",
"biology",
"anatomy",
"virulence",
"factors",
"post-translational",
"modification",
"virology",
"epithelium",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"spectrum",
"analysis",
"techniques"
] | 2018 | Integrin but not CEACAM receptors are dispensable for Helicobacter pylori CagA translocation |
L locus resistance ( R ) proteins are nucleotide binding ( NB-ARC ) leucine-rich repeat ( LRR ) proteins from flax ( Linum usitatissimum ) that provide race-specific resistance to the causal agent of flax rust disease , Melampsora lini . L5 and L6 are two alleles of the L locus that directly recognize variants of the fungal effector AvrL567 . In this study , we have investigated the molecular details of this recognition by site-directed mutagenesis of AvrL567 and construction of chimeric L proteins . Single , double and triple mutations of polymorphic residues in a variety of AvrL567 variants showed additive effects on recognition strength , suggesting that multiple contact points are involved in recognition . Domain-swap experiments between L5 and L6 show that specificity differences are determined by their corresponding LRR regions . Most positively selected amino acid sites occur in the N- and C-terminal LRR units , and polymorphisms in the first seven and last four LRR units contribute to recognition specificity of L5 and L6 respectively . This further confirms that multiple , additive contact points occur between AvrL567 variants and either L5 or L6 . However , we also observed that recognition of AvrL567 is affected by co-operative polymorphisms between both adjacent and distant domains of the R protein , including the TIR , ARC and LRR domains , implying that these residues are involved in intramolecular interactions to optimize detection of the pathogen and defense signal activation . We suggest a model where Avr ligand interaction directly competes with intramolecular interactions to cause activation of the R protein .
The plant immune system is based upon the ability to accurately perceive and appropriately respond to potential threats . In general , plants use membrane-spanning proteins with extracellular receptor domains to recognize common features of plant pathogens ( pathogen associated molecular patterns , PAMPs ) and intracellular receptors to detect pathogen effectors transferred into plant cells during infection [1] , [2] , [3] . Most intracellular immune receptors ( disease resistance proteins ) contain nucleotide-binding ( NB ) and leucine-rich repeat ( LRR ) domains; one subclass of these has a coiled-coil ( CC ) domain and the other possesses a TIR ( Toll , interleukin-1 receptor , resistance protein ) domain at the N-terminus [4] , [5] . Plant NB-LRR disease resistance proteins belong to the STAND ( signal transduction ATPases with numerous domains ) clade of AAA+ ( ATPase associated with diverse cellular activities ) proteins , and are similar to the Nod-like receptor ( NLR ) family of proteins that act as intracellular surveillance molecules in animal innate immunity [6] , [7] , [8] . The signature catalytic core of STAND proteins comprises an αβα NB domain , a four-helix ARC1 ( APAF-1 , R protein , CED-4 ) domain , and a winged helical ARC2 domain [9] , [10] , [11] . This domain is thought to function as a reversible molecular switch during signal transduction , with monomeric ADP-bound forms representing the off - or closed - state , and ATP-bound multimeric forms representing the on - or open – state [9] , [11] , [12] , [13] . Tight regulation of this switch is critical in plant NB-LRRs , because these proteins regulate an apoptotic process . The trigger for the conformational change to the open state is generated by signal perception , either directly when NB-LRRs bind effector proteins [14] , [15] , [16] , [17] , [18] , [19] , or indirectly when NB-LRRs detect the biochemical fingerprint of effector proteins as they attempt to carry out their virulence function [20] , [21] , [22] , [23] , [24] , [25] . This effector-mediated R protein activation is believed to ultimately lead to conformation changes that expose the N-terminal TIR or CC signalling domains , so they can interact with downstream signalling partner [12] , [26] . The C-terminal LRR domain of R proteins generally mediates signal perception [27] , [28] . This domain is composed of repeating LRR units that form stacking β-strands , resulting in a horseshoe-shape molecule with a continuous , parallel β-sheet on the inner concave surface [29] . Individual LRR units contain xxLxLxx motifs generating β-strand/β-turn structures in which the variable non-leucine residues form the concave , solvent-exposed surface of the horseshoe and are available for participation in protein-protein interactions [29] , [30] . This region of plant R proteins is often highly variable , as a result of diversifying selection , and a number of studies have demonstrated changes in specificity mediated by polymorphisms in the LRR domain [18] , [30] , [31] , [32] , [33] , [34] , [35] , [36] , [37] , [38] . TIR-NB-LRR resistance proteins in flax ( Linum usitatissimum ) confer resistance to the flax rust fungus Melampsora lini through recognition of effector proteins delivered into the host cell during infection [39] , [40] . For example , the L resistance locus consists of a single gene encoding 13 allelic protein variants ( L , L1 to L11 , and LH ) that recognise different matching avirulence proteins [32] . L alleles share greater than 90% amino acid sequence identity , with positively selected variation concentrated in the LRR domain . Domain-swap experiments between the L2 , L6 and L10 alleles showed that these recognition specificities are determined by the LRR domain [30] , [32] . Similarly , the L6 and L11 proteins differ by only 32 amino acids , all in the LRR domain , and a chimeric protein with 11 amino acid changes in the C-terminal region of the LRR displayed a novel specificity , with a reduced recognition spectrum [16] , [41] . The L5 , L6 and L7 proteins recognise allelic variants of the M . lini effector protein AvrL567 , a 127-amino acid secreted protein that is expressed in haustoria and translocated into host cells during infection [42] , [43] , [44] . Seven of the 12 variant forms of AvrL567 ( -A , -B , -D , -E , -F , -J , -L ) are avirulence alleles as they induce an L5 and/or L6 , and/or L7-dependent hypersensitive response ( HR ) in transient expression assays whereas the other 5 variants ( -C , -G , -H , -I , -K ) are virulence alleles as they do not induce an HR [16] . Yeast-two-hybrid ( Y2H ) assays demonstrated that AvrL567 and L5 , L6 , and L7 interact directly and that the specificity of this protein-protein recognition corresponds with that of the HR-inducing recognition in planta [16] . L6 and L7 are differentiated by just 11 polymorphisms found in the TIR domain and have identical AvrL567 recognition specificities , although L7 shows consistently weaker interaction in yeast , and a weaker HR in planta [26] , [30] . L5 and L6 are two of the most diverged L proteins , differing by 89 amino acid polymorphisms ( 61 in the LRR ) and four small indels , but nevertheless have overlapping recognition specificities . They are distinguished by L6 interacting with AvrL567-D , while L5 does not . Wang et al . [45] determined the structures of AvrL567-A and -D and identified four polymorphic surface-exposed amino acid residues that were important for their differential recognition . Here we have further investigated the role of these surface-exposed amino acids in recognition of AvrL567 . Single , double and triple mutations at these sites in a variety of AvrL567 variants showed additive effects on recognition strength , suggesting that multiple contact points are involved in the recognition event . We show by domain-swap experiments that the L5 and L6 specificities are determined by their corresponding LRR regions , with contributions made by seven and four N- and C-terminal LRR units , respectively , of a total of 26 , where most positively selected amino acid sites occur . This further confirms that multiple , additive contact points occur between AvrL567 variants and either L5 or L6 . However we also observed that recognition of AvrL567 is affected by co-operative polymorphisms between both adjacent and distant domains , including the TIR , ARC and LRR domains , implying that these residues are involved in intramolecular interactions to optimize detection of the pathogen and/or defense signal activation .
Sequence comparisons of the 12 AvrL567 variants suggested that polymorphisms at four positions ( 50 , 56 , 90 and 96 ) were associated with specificity differences [16] . Single amino acid substitutions at positions 50 ( T50I ) or 96 ( L96R ) were sufficient to restore recognition of AvrL567-D by L5 , while the I50T substitution almost completely blocked L5 and L6 recognition of AvrL567-A ( Table 1 ) [45] . To further evaluate the role of these residues in mediating interactions with L5 and L6 , we made reciprocal single , double and triple substitutions of these amino acids in a wider range of AvrL567 variants ( -A , -D , -E , -J and -C ) , which show varying recognition patterns ( Table 1 ) . Mutant AvrL567 proteins were assayed for recognition by L5 and L6 using both Y2H assays - to test for protein interaction - and by Agrobacterium-mediated transient expression in planta - to measure R gene-dependent cell death ( Figure 1 and Table 1 ) . With one exception ( see below ) , single amino acid changes at positions 56 , 90 and 96 in AvrL567-A did not alter recognition by L5 or L6 [45] . However , double and triple substitutions at these positions revealed that they all contribute additively to recognition . The K56D/S90I and K56D/R96L double mutants both substantially reduced recognition by L5 , while the K56D/S90I/R96L triple substitution blocked L5 recognition completely ( Figure 1 ) . Notably , none of these changes affected L6 recognition . Conversely , the single amino acid R96S substitution abolished L6 but not L5 recognition . This indicates that L5 and L6 recognise different molecular features of AvrL567 , but at similar positions . For AvrL567-D , Wang et al . [45] showed that either T50I or L96R substitutions were sufficient to allow interaction with L5; but we now found that the T50I/L96R double mutation shows an additive effect relative to the single mutants , which can be detected when the GAL4 AD and BD fusions are reversed ( Figure S1 ) . Further evidence for additive interactions comes from the context-dependent effects of several single substitutions . For instance , the presence of S or L at position 96 does not prevent L5 recognition of AvrL567-A or -J , but in AvrL567-D an R is required at this position to establish L5 recognition . Likewise , a S96 substitution destabilizes L6 recognition of both -A and -D , but is compatible with L6 recognition of -J ( Figure 1 and Table 1 ) . To complement these loss-of-function studies , we also tested the effect of reciprocal changes in the virulence allele , AvrL567-C , which is not recognised by L5 or L6 and found that multiple amino acid changes were required to restore full recognition ( Figure 1 , Table 1 ) . For instance , double substitutions at positions 50 and 56 or positions 50 and 96 were required to allow L5 recognition of this protein . L6 recognition could be restored weakly ( in yeast but not in planta ) by the single S96R substitution , but required the triple T50I/D56N/S96R substitution for full recognition . Interestingly , in the context of AvrL567-C , a K residue at position 56 was not compatible with L6 recognition , although it does not prevent recognition in the AvrL567-A or -D contexts ( Table 1 ) . The strong positive effect of isoleucine at position 50 was confirmed as the single T50I substitution in AvrL567-E restored its interaction with L5 and L6 . These data further support the additive roles of these amino acid positions in recognition . All AvrL567 mutant fusion proteins were stably expressed in yeast ( Figure S2 ) indicating differential recognition of mutants by L proteins was due to differences in their surface properties , resulting in physical changes in the interactions of these proteins . Data from Y2H and in planta HR analyses correlated well , apart from a few exceptions , which can all be explained by in planta HR induction being less sensitive than the Y2H interaction ( for instance , L6 interactions with AvrL567-A K56N or AvrL567-J S96L; Figure 1 ) . Overall , these data suggest that multiple contact points at disparate positions on the AvrL567 molecule are involved in interaction with the corresponding R proteins and make additive contributions to the strength of recognition . In addition , although L5 and L6 recognition of AvrL567 involves contacts with similar positions , they have different requirements for amino acid residue features at these positions . The cloned Avr genes and their derived mutants provide a sensitive set of test proteins to detect subtle changes of specificity of L5–L6 chimeric proteins described in the following sections . In order to correlate the recognition-determining residues in the AvrL567 proteins with variation in the L5 and L6 proteins , we conducted an analysis of positive selection on the coding sequences of all 12 cloned L genes . Previous analysis had found an excess of non-synonymous versus synonymous substitutions in the LRR domain [30] , and we used the program CODEML [46] to identify codons under positive selection . The M8 model allowing for positive selection provided a significantly better fit to the data than the M7 null hypothesis model , which excludes positive selection ( p<0 . 001; Table S1 ) and predicted 123 ( 9 . 5% ) codons as being under significant positive selection ( Figure 2 ) . This includes 86 of the 99 sites polymorphic between L5 and L6 ( Figure S3 ) . To examine the distribution of positively selected sites we considered the protein sequence in six regions: the TIR , NB , ARC1 , ARC2 and LRR domains and a short spacer region between ARC2 and LRR domains . In the LRR domain , 13 . 2% of codons ( 92 sites ) are under significant positive selection , compared to only 5 . 3% of codons ( 31 sites ) in the rest of the protein . These occurred mainly in the N-terminal and C-terminal portions of the LRR domain , with a lack of positively selected sites in the central portion of the LRR domain . This suggests that recognition specificity may be conferred mainly by interactions involving the two extremities of the LRR domain . The ARC1 domain and the spacer also showed elevated numbers of positively selected sites ( about 11% ) compared to the TIR , NB and ARC2 domains ( 2 to 5%; Figure 2 ) . The concentration of positively selected sites in the LRR domain of L proteins is consistent with the proposed role of this domain in recognition specificity . To test whether polymorphisms in the LRR domains of L5 and L6 are responsible for their different recognition specificity , we firstly generated chimeric proteins L6592L5 and L5592L6 , in which the complete LRR domains of L5 and L6 are exchanged at an engineered AvrII restriction site in codons 592–593 ( Figure S4 ) . The introduction of this site causes a W to R amino acid change at position 592 , but this substitution did not effect AvrL567-A or -D recognition by the modified L5 or L6 alleles ( Figure S5 ) and is also found in the functional L9 protein . The chimeric proteins were tested for recognition of AvrL567 variants and mutants by the Y2H assay ( Figure 3 ) . Both recombinant proteins were well-expressed in yeast ( Figure S2 ) , but L5592L6 was non-functional in that it did not interact with either AvrL567-A or -D ( Figure 5a ) . This may be related to the position of the exchange site within a seven-amino acid indel polymorphism ( Figure S3 ) . On the other hand , L6592L5 exhibited L5-like specificity , giving recognition of AvrL567-A but not -D , which indicated that this recognition pattern was determined by the LRR domain of L5 ( Figure 3b ) . However , when tested against the extended set of AvrL567 mutants , L6592L5 recognized only a subset of the wild-type L5 repertoire , and failed to interact with AvrL567-A K56D , K56D/S90I and K56D/R96L mutants , and with most AvrL567-C gain-of function mutants ( Figure 3b ) . We therefore extended the region swapped between the alleles further towards the N-terminus . The chimera L6493L5 , which contained the L5 LRR domain plus an additional three amino acid polymorphisms from the ARC2 domain and the entire wild-type spacer region ( Figure 3a , S3 ) , retained the full L5 recognition specificity across the pool of AvrL567 variants ( Figure 3b ) . The only exceptions were a slightly enhanced interaction with the AvrL567-A I50T and K56D/S90I mutants , which only weakly interacted with L5 , and a weak interaction with AvrL567-D L96S , which did not interact with L5 . Similarly , the chimera L5556L6 that included the L6 LRR domain and the spacer region had L6-like recognition specificity when tested against wild-type AvrL567 variants and mutants ( Figure 3c , d ) , although in some cases interactions were weaker than for L6 ( AvrL567-D , AvrL567-D N56K and L96R and the AvrL567-C mutants ) . In conclusion , the data are consistent with the L5–L6 specificity differences being contributed by polymorphisms in the LRR domain and spacer region , but with the strength of the R:Avr protein interactions modulated somewhat by interactions between these regions and the N-terminal TIR-NB-ARC region . The recognition repertoires conferred by the LRR domains of L5 and L6 can be qualitatively distinguished by their interactions with AvrL567-D ( interacts with L6 but not L5 ) and the AvrL567-A mutant R96S ( interacts with L5 but not L6 ) ( Figure 1 and 3 ) . To further understand the role of LRR domain polymorphisms in these differences , a series of L5–L6 chimeras with swaps within the LRR domain was generated . We designed swaps that would exchange groups of positively selected amino acids , as well as residues implicated in interaction in the docking-derived models of AvrL567 binding to a modelled L5 LRR structure presented by Wang et al . [45] ( Figure S3 ) . All the chimeric proteins were stably expressed in yeast ( Figure S2 ) and were evaluated for interactions with AvrL567-A , -D and the AvrL567-A-R96S mutant ( Figure 4 ) . These experiments allowed us to draw several inferences about residues controlling recognition specificity . Firstly , interaction with AvrL567-D , which discriminates the L6 specificity , requires L6 polymorphisms in the last four LRR units . Substitution of these four LRR units of L6 with the corresponding region of L5 ( L61193L5 ) , abolished the interaction with AvrL567-D , but not -A ( Figure 4a ) . This was also true for the same LRR exchange made in the context of the L5 TIR-NB-ARC ( L5556L61193L5; Figure 4b ) . Conversely , the reciprocal exchange in L5 ( L51193L6 ) allowed weak interaction with AvrL567-D ( Figure 4c ) , suggesting that the L6 polymorphisms in the last four LRR units are both necessary and sufficient to confer AvrL567-D interaction in these proteins . Interestingly , this interaction was stronger when the L6 TIR-NB-ARC region was also present ( L6592L51193L6; Figure 4d ) . Indeed , the L6 TIR-NB-ARC region enhanced the recognition of AvrL567-D for all the chimeras containing the L6 C-terminal LRRs ( compare Figures 4c and 4d ) . Secondly , the ability to interact with AvrL567-A-R96S , which discriminates the L5 specificity , requires L5-specific polymorphisms in the first seven LRR units . L6592L5 , containing the full-length L5 LRR domain , interacted with AvrL567-A-R96S ( Figure 3 ) , while proteins with chimeric L6-L5 LRR domains containing L6 N-terminal LRR polymorphisms ( L6793L5 , L6972L5 , L61125L5 and L61193L5 ) did not ( Figure 4a ) . Conversely , chimeras containing the reciprocal region from L5 swapped into the remainder of L6 ( L6592L5793L6 , L6592L5972L6 , L6592L51125L6 , and L6592L51193L6 ) acquired the capacity to interact with AvrL567-A-R96S ( Figure 4d ) . Thus , the L5 polymorphisms found between residues 594 and 793 in L5 are necessary and sufficient to determine the AvrL567-A-R96S interaction in these proteins . Intriguingly , two chimeric proteins ( L5793L6 and L5972L6 ) , which contain the critical 594-to-793 L5 residues along with the L5 TIR-NB-ARC domains did not interact with AvrL567-A-R96S ( Figure 4c ) . As above , this suggests that the presence of L6 TIR-NB-ARC region is required for strong Avr protein interactions in proteins containing the L6 C-terminal LRRs . Neither L6493L5972L6 nor L6493L51125L6 interacted with AvrL567-D ( Figure 4e ) , suggesting an additional positive contribution to the strength of the interaction between these LRR chimeras and AvrL567-D may be attributed to the presence of one or more of the three L6-specific ARC2 polymorphisms ( Figure S3 ) . We also observed that certain L6-L5 LRR chimeras lacked recognition function . For instance , swaps containing the N-terminal LRRs of L6 and the C-terminal LRRs of L5 gave rise to non-functional proteins when the junctions were made at positions 793 or 1125 , but not at 972 or 1193 ( Figure 4a and 4b ) . Similarly , proteins containing the N-terminal LRRs of L5 and the C-terminal LRRs of L6 gave rise to proteins with reduced functionality when the junctions were made at positions 793 , 972 or 1125 ( Figure 4c ) , although this could be overcome by the presence of polymorphisms found in the L6 TIR-NB-ARC region ( compare swaps in Figure 4c , d and e ) . These observations suggest a requirement for specific , co-operative combinations of polymorphisms within the LRR domain to allow interaction with the corresponding ligand , consistent with the interaction occurring across a large surface area . Because the strength of AvrL567 interaction of several chimeric proteins appeared to be influenced by whether the TIR-NB-ARC region is derived from L5 or L6 , we decided to examine the influence of polymorphisms in the N-terminal region on Avr protein interaction . A series of chimeras were generated in which various regions of the L6 TIR , NB , ARC1 and ARC2 domains were re-introduced into the L5556L6 protein , which exhibited L6-like specificity , but weaker AvrL567-D interaction , and tested for interaction with AvrL567-A and -D ( Figure 5a ) . All chimeric proteins were stably expressed in yeast ( Figure S2 ) . Introduction of increasing lengths of L6 sequence from the N-terminus ( L6185L5556L6 to L6431L5556L6 ) , including the TIR and NB regions , did not increase the interaction with AvrL567-D , but a further swap including the L6 ARC1 ( L6493L5556L6 ) restored strong interaction with AvrL567-D . This suggested that one or more of the six amino acid polymorphisms between 447 and 484 ( five in ARC1 and one in ARC2 ) were responsible for the reduced interaction . Consistent with this , inclusion of the ARC1 and ARC2 regions of L6 along with the L5 TIR-NB ( L5431L6 ) also restored recognition of AvrL567-D , again implicating this region in the reduced interaction . Chimeric proteins representing the inverse swaps and including the L5 LRR domain ( L6414L5 , L6431L5 , L6493L5 , L6592L5 , L5185L6592L5 and L5226L6592L5 ) did not interact with AvrL567-D but retained interaction with AvrL567-A ( Figure 5b ) , similar to L5 . This suggests that polymorphisms in the LRR domain of L6 are required to provide the specific recognition capacity to bind to AvrL567-D , while those in the ARC1/2 region may contribute to the strength of the interaction through intramolecular interactions . Interestingly , some other swaps in the TIR-NB region also led to reduced Avr protein interaction . Notably , while the L5185L6 chimera was functional , L5226L6 did not interact with either AvrL567-A or -D . Similarly , L6185L5556L6 failed to interact with the Avr proteins , while L6226L5556L6 did interact with AvrL567-A . However , both the reciprocal swaps ( L5185L6592L5 and L5226L6592L5 , Figure 5b ) interacted with AvrL567-A . This suggests that the two L5-derived amino acid polymorphisms in this region ( E216 and L218 ) interfere with recognition in the context of the L6 LRR domain . We therefore tested a series of constructs containing chimeric L5–L6 LRR domains in the context of the L5226L6 protein , to determine which part of the L6 LRR domain mediates this incompatibility ( Figure 6 ) . Introduction of the seven N-terminal LRR units from L5 was sufficient to restore AvrL567 interaction in this protein ( Figure 6a ) , while all chimeras containing this region from L6 failed to interact ( Figure 6b ) . A similar pattern was observed for chimeras containing a hybrid L6185L5 TIR-NB junction ( L6185L5556L6 , L6185L5556L6592L5793L6 and L6185L5556L6592L51193L6; Figure 5a and Figure 6c ) . This suggests that a negative interaction occurs between these TIR domain polymorphisms of L5 and polymorphic residues in the N-terminal region of the L6 LRR . Even though several of these constructs contained all of the polymorphisms from L6 that are required for strong recognition of AvrL567-D , the presence of the hybrid L5–L6 TIR-NB junction appears to have prevented this interaction , for example in L6185L5556L6592L5793L6 and L6185L5556L6592L51193L6 ( Figure 6c ) . As described above , chimeras L6493L51193L6 and L6592L51193L6 , which contain the N-terminal L5 specificity region and the C-terminal L6-specificity region , represent a novel and expanded recognition specificity , in that they recognise both AvrL567-A-R96S and AvrL567-D , which distinguish L5 and L6 ( Figure 4 ) . We therefore tested these further against the larger set of AvrL567 variants and mutants and compared this to the recognition repertoires of L5 , L6 , L6493L5 and L6592L5 ( Table 2 and Figure S6 ) . The L6592L51193L6 and L6493L51193L6 recognition specificities were largely L6592L5-like and L6493L5-like , respectively , however both were L6-like in regards to interactions with AvrL567-D and its derived mutants . Overall , L6493L51193L6 recognized more AvrL567 variants and mutants ( 23 ) than L6 ( 21 ) , L6493L5 ( 20 ) , L5 ( 19 ) , L6592L51193L6 ( 14 ) , or L6592L5 ( 13 ) ( Table 2 ) . We further tested the L6493L51193L6 chimera for its ability to trigger an AvrL567-dependent cell death in planta ( Figure 7 ) . Agrobacterium-mediated transient expression of the L5 , L6 and L6493L51193L6 cDNAs in transgenic tobacco also expressing AvrL567-A induced a strong HR-like cell death response ( Figure 7a ) , confirming that the L6493L51193L6 construct expresses a functional resistance protein . Co-expression of L6493L51193L6 with AvrL567-D or AvrL567-A-R96S also induced an HR , while L5 induced cell death only with AvrL567-A-R96 , and L6 only with AvrL567-D ( Figure 7b , c ) , thus recapitulating the recognition specificity observed in yeast . No HR was observed when L6493L51193L6 was expressed alone ( Figure 7d ) , indicating that this chimera is not autoactive .
Mutational analysis shows that multiple sites are involved in the interaction between AvrL567 variants and L5 and L6 , and that L5 and L6 have different specificity requirements at these positions ( Figure 1 ) . The additive nature of the interactions is shown by the effects of double and triple mutations in the AvrL567 proteins on recognition . For instance , while single mutations at positions 56 , 90 and 96 do not disrupt recognition by L5 , triple substitution abolishes L5 recognition . However , these changes do not disrupt recognition by L6 , highlighting the different sequence requirements of the two resistance proteins . Conversely , with the exception AvrL567-C-S96R ( which is weakly recognised by L6 ) , the virulence allele AvrL567-C required at least two to three mutations in combination to allow full recognition by L5 or L6 . Again , the requirements for the two resistance proteins were different . A double substitution at positions 50 and 56 was sufficient for L5 recognition , while full L6 recognition required a further substitution at position 96 , and showed a requirement for an asparagines residue at position 56 rather than lysine . Previously , Wang et al . [45] demonstrated that T50 in AvrL567 destabilizes interactions with L5 and L6 and our data further confirm that this residue is particularly important for recognition . For instance , the T50I substitution has a strong positive effect in AvrL567-E on stabilizing interactions with L5 and L6 ( Figure 1 ) . AvrL567-E and -J differ by only two polymorphisms ( H26D and T50I ) , but we previously found that the N-terminal region consisting of amino acids 26–37 of AvrL567-A could be deleted without affecting recognition [43] . Therefore , the T50I polymorphism is the critical residue that differentiates AvrL567-E and -J recognition specificities . The importance of position 50 can also be observed by the contribution a T50I substitution makes to allow L5 recognition of AvrL567-C when paired with D56N , D56K , or S96R substitutions , and to allow L6 to recognize AvrL567-C when associated with D56N and S96R in a triple substitution ( Figure 1 ) . The presence of a D residue at position 56 had a small negative effect on interactions with L5 but not L6 [45] , and we observed that this effect is much stronger when combined with either , or both , S90I and R96L in AvrL567-A . Similarly , neither S90I or R96L substitutions ( Table 1 ) , nor the double S90I/R96L substitution in AvrL567-A compromise recognition by L5; however , I90 and L96 both have stronger negative effects on interactions with L5 when combined with D56 , and completely disrupt L5 recognition when all three substitutions are present in AvrL567-A ( Figure 1 ) . Interestingly , in the context of AvrL567-C , K56 has a negative effect on recognition by L6 but not L5 , as the triple T50I/D56N/S96R substitution in AvrL567-C is recognized by both L5 and L6 in yeast , whereas the triple T50I/D56K/S96R substitution in AvrL567-C is only recognized by L5 . Position 96 in AvrL567 is important for interactions with both L5 and L6 , with R at this position favouring interactions with L5 , and an S disrupting interactions with L6 ( Figure 1 and [45] ) . For instance , the R96S substitution in AvrL567-A destabilizes interactions with L6 , whereas the reciprocal S96R substitution improves interaction of AvrL567-C with L6 both individually , and in combination with T50I or T50I/D56N substitutions , and with L5 when combined with the T50I , T50I/D56N or T50I/D56K substitutions . However , as with other polymorphisms in AvrL567 , the disruptive effect of S96 on L6 recognition is context-dependent , as both AvrL567-E and –J contain this polymorphism while maintaining interactions with L6 . Collectively , these data indicate that L5 and L6 interact with AvrL567 through multiple amino acid contact points , and support the hypothesis that recognition is mediated in an additive manner by the cumulative composition and context of their amino acid sequences . Chimeric L5–L6 proteins containing reciprocally swapped LRR domains showed AvrL567 interaction specificities consistent with the origin of the LRR domain ( Figure 3 ) , although some interactions were weaker than in the wild-type proteins . Positively selected amino acid sites in L alleles cluster at the N and C-terminal regions of the LRR ( Figure 2 ) , and docking analysis of AvrL567 to the modelled L5 LRR domain [45] suggested that the most likely binding site of AvrL567 is between the two ends of the LRR with most potential contact points in these N- and C-terminal regions . This hypothesis is supported by analysis of LRR chimeras ( Figure 4 ) , which showed that 13 polymorphisms in the last four LRRs of L6 are required for the recognition of AvrL567-D , while polymorphisms in the first seven LRRs of L5 are required for recognition of AvrL567-A-R96S . Some internal LRR fusions ( eg . L6793L5 and L61125L5; Figure 4a ) fail to interact with AvrL567 variants , or show a reduced interaction repertoire ( eg . L6493L5972L6 , L6493L51125L6; Figure 4e ) , suggesting that surfaces involved in specific interactions may have been disrupted by fusions at these junctions . Previously , Ellis et al . [41] showed that polymorphisms between L6 and L11 in the last three LRRs ( 24 , 25 and 26 ) are important for L6 recognition of AvrL567 . The L6L11RV chimera differs from L6 by 11 polymorphisms in these three LRR units , and recognizes only AvrL567-J , whereas L6L11B2 , with two additional polymorphisms in LRR 23 , does not interact with any AvrL567 variant ( Figure S7 ) . L5 is quite similar in sequence to L11 in this region , but a similar L6-L5 exchange in this region ( L61193L5 ) maintained interaction with many AvrL567 variants , including -A and -J ( Figure 4 ) . Comparison of the C-terminal sequences of these chimeras ( Figure S7 ) narrows the L11 polymorphisms responsible for these difference down to R1220 and K1222 in LRR 24 – contributing to the loss of Avr recognition ( other than -J ) in L6L11RV , and V1196 leading to the loss of -J recognition in L6L11B2 , because these are the only polymorphisms unique to L11 . It is important to note that these domain-swap experiments only examine the roles of polymorphic residues in determining recognition specificity and do not address the role of shared residues in AvrL567 interaction . Indeed , docking analysis identified 31 residues common to L5 and L6 that may be involved in protein contacts . Because chimeras with the N-terminal LRRs of L5 and C-terminal LRRs of L6 can interact with AvrL567-A , -A-R96S and -D , and reciprocal chimeras also retain interaction with at least AvrL567-A ( Figure 4 ) , we conclude that binding of AvrL567 to L5 and L6 occurs in the same basic orientation . The ability of the L5 N-terminal LRRs to allow binding to AvrL567-A-R96S may suggest that the AvrL567 surface region containing this residue makes contact with the N-terminal LRR region . Although in general domain-swaps involving the full LRR domain showed the expected specificity for the source of the LRR , some chimeras had weak or no interactions with AvrL567 variants ( Figure 3 ) . Because all of the polymorphisms found in L5 and L6 are compatible with AvrL567 interaction in their native context , this suggests that certain polymorphisms in L5 and L6 N-terminal regions occur in specific , co-operative combinations that are required for recognition function . We observed such co-adaptation between the spacer region and the LRR domain , between the ARC domains and the C-terminal region of the LRR domain , and between the TIR and LRR domains . While both L6493L5 and L6592L5 exhibited L5-like specificity ( Figure 3 ) , when tested on the wider set of AvrL567 mutants , L6592L5 recognised only a subset of the wild-type L5 repertoire . This suggests that some or all of the nine L5-specific amino acids in the ARC2 and spacer region ( Figure S3 ) are important for optimal recognition . Six of these residues represent a small insertional polymorphism in the spacer region , which may influence the relative positioning of the L5 LRR domain with respect to the rest of the protein . A wild-type spacer region is probably also required for L6 to adopt a functionally competent state , because an exchange within this indel in L5592L6 resulted in a non-functional protein , while L5556L6 ( in which the exchange occurs before this indel ) showed an L6-like recognition repertoire ( Figure 3 ) . Given its proximity to the LRR domain , it is possible that the spacer region also participates directly in AvrL567 binding . On the other hand , the observed co-adaptation between polymorphisms at the C-terminus of the TIR domain and the N-terminus of the LRR domain , and between the ARC and LRR domains , likely reflects an indirect effect on ligand affinity as a result of intramolecular interactions that obscure the ligand-binding site , rather than a direct effect on binding specificity . Such general effects on ligand accessibility would be expected to manifest themselves particularly in the case of those interactions that are close to the threshold of detection in Y2H assays . Sequence exchanges involving regions of the TIR domain suggested that two L5-derived amino acid polymorphisms in this region ( E216 and L218 ) interfere with recognition in the context of the L6 LRR . However , the TIR domain is not required for L6-AvrL567 interaction , and these two residues are exposed on the surface of the TIR domain structure in a region implicated in negative regulation of L6 through intramolecular interactions [26] . Indeed , TIR domain residues that are polymorphic between L6 and L7 also play a role in AvrL567 interaction and are responsible for the weak resistance phenotype of L7 [26] , [30] . Likewise , polymorphisms in the ARC1 and ARC2 domains of L6 strengthen AvrL567-D recognition , conferred by the polymorphisms found in the last four LRRs of L6 ( Figure 5 ) . Again , this appears to be a general ligand affinity effect , because swapping the L6 ARC domains into L5 does not generate recognition of AvrL567-D ( Figure 5b ) . Furthermore , the presence of the L6 TIR-NB-ARC region also strengthens interactions with AvrL567-A-R96S , mediated by polymorphisms found in the first seven LRRs of L5 ( Figure 4c and d ) . We previously showed that a P-loop mutation ( K271M ) in L6 , which would prevent nucleotide binding , also disrupted interaction with AvrL567 [16] , consistent with the idea that interactions between the NB-ARC and LRR are required to support ligand binding . Previous experiments in other systems have also demonstrated intramolecular interactions between R protein domains that are important for function . The CC , NB-ARC and LRR domains of potato Rx can interact and functionally complement each other when expressed as separate polypeptides [47] . Likewise , domain swaps have also implicated co-adaptation between domains of Rx , Mi-1 . 2 and I-2 in tomato , and Pm3 in wheat [34] , [48] , [49] , [50] . It has been suggested that ARC1 functions as a molecular scaffold , forming intramolecular interactions with the LRR domain , and that signal perception disrupts these interactions [34] . Subsequently , ARC2 may transduce this effect into defence protein activation [50] . However , these experiments do not distinguish between effects on ligand recognition , protein activation or downstream signalling . Our data on AvrL567 binding by L5/L6 recombinants indicate that these intramolecular interactions can have direct effects on ligand binding . This suggests a model of R protein activation in which ligand binding occurs in direct competition with intramolecular interactions , which presumably maintain the resting protein in an inactive signalling state . Rather than Avr binding directly destabilising intramolecular interactions , it is possible that R proteins exist in an equilibrium between active and inactive states , with the Avr protein preferentially binding to and stabilising the active state to induce signalling . This competition provides a mechanism for signalling activation , as well as for fine-tuning the triggering of the response . Weak R:Avr interactions require a very delicate trigger if they are to induce effective resistance , but this would come at a cost of increased autoactivity of the R protein . Conversely , stronger Avr interactions could compete with more stable inhibitory intramolecular interactions . To visualize structurally the L protein regions involved in AvrL567 recognition , we prepared a homology model of the L6 NB-ARC domain with the program Modeller [51] using the multiple sequence alignment of NB-ARC domains from different R proteins published in van Ooijen et al . [11] and the crystal structure of APAF-1 as a template [52] . In the model , four of the L6 polymorphisms involved in strengthening interactions with AvrL567-D ( A454 , E457 , E461 , and R465 ) map to a solvent exposed region in the last α-helix of the ARC1 domain ( Figure 8 ) . This α-helix ( H8 of HD1 ) , is part of the ARC1-ARC2 linker region , found in APAF-1 and CED-4 , that undergoes a drastic change during the switch from closed to open states [9] . Therefore in L6 , these residues are positioned appropriately to be involved in either , or both , inter- and intramolecular interactions , and may provide a putative link between effector perception and hypothetical conformational changes that lead to the open state . Likewise , Brunner et al . [48] found that polymorphic residues in the ARC2 domain that disrupt Pm3 function appear to be concentrated on one side of the ARC2 domain , and are largely solvent exposed , suggesting that they may be involved in intra- or intermolecular interactions . In the course of this chimeric protein analysis we generated some recombinant proteins that exhibited novel and expanded recognition specificities , through bringing together unique combinations of polymorphic sites . One of these ( L6493L51193L6 ) was shown to function to induce an HR in tobacco , recapitulating the expanded recognition observed in Y2H assays . This correspondence between Avr:R interaction in yeast and HR induction in planta , now observed across a large number of L-AvrL567 pairwise combinations ( Figure 1 and 7; [16] , [45] ) is consistent with the hypothesis of ligand interaction triggering signalling through competition with inhibitory intramolecular interactions . This suggests that recombination of existing polymorphisms through sequence exchanges is a powerful method for both generating changes in recognition specificity and fine-tuning the strength of defense response during evolution of resistance genes . This mechanism , along with induced mutation , may be adapted to engineering novel resistance genes that can be deployed in agriculture [53] . This process may require not only changes to LRR domain to generate new binding specificities , but also concomitant changes in N-terminal domains to optimise the defense signalling output .
Site-directed mutants of AvrL567 were constructed using the Gene-Tailor kit ( Stratagene ) according to the manufacturer's instructions . Chimeric L5–L6 proteins were constructed using native and introduced restriction sites , and/or by PCR-based fusion of overlapping sequences as described in Text S1 and Figure S5 . Chimeric proteins were either constructed directly in pGADT7 , or were sub-cloned in pBSK prior to construction in pGADT7 . All constructs were checked by restriction enzyme digests and DNA sequencing . GAL4-binding domain ( BD ) fusions , and transcriptional activation domain ( AD ) fusions to L5 , L6 , L5–L6 chimeric proteins and to AvrL567 mutants and variants , were prepared in the pGBT9 and PGADT7 vectors ( Clontech ) , as described [16] , [45] . Yeast transformation , lacZ and His growth assays were performed as described in the Yeast Protocols Handbook ( Clontech ) . Yeast proteins were extracted by the trichloro-acetic acid method , separated by SDS/PAGE , and transferred to nitrocellulose membranes ( Pall ) by electroblotting . Membranes were blocked with 5% skim milk and probed with anti-HA mouse monoclonal antibodies ( Roche ) , followed by goat anti-mouse antibodies conjugated with horseradish peroxidase ( Pierce ) . Labeling was detected with the SuperSignal West Pico chemiluminescence kit ( Pierce ) . DNA constructs encoding AvrL567 proteins lacking the signal peptide , or full-length L5 , L6 or L6493L51193L6 cDNAs , were cloned into the binary vector pTNotTReg between the cauliflower mosaic virus 35S promoter and ocs terminator sequences . Agrobacterium tumefaciens ( GV3101-pMP90 ) cells containing these constructs were grown for 36 h at 28°C in LB media supplemented with appropriate antibiotic selections . Cells were pelleted , resuspended in infiltration medium ( 10 mM MgCl2 , 200 µM acetosyringone ) , adjusted to OD600 nm = 1 and incubated for 2 h at room temperature . Resuspended cells were infiltrated with a 1-mL needleless syringe into the leaves of near isogenic lines of flax plants containing L5 , L6 ( cv . Bison ) or L6L11RV ( cv . Ward ) , or into leaves of 3-week-old tobacco plants ( W38 ) . Transgenic tobacco expressing AvrL567-A was described by Dodds et al . [42] . The non-synonymous/synonymous rate ratio parameter ω was estimated using the program CODEML [54] in phylogenetic analysis by ML v . 4 . 2 . Tests for positive selection were performed using the site class models that estimate ω for amino acid sites . Neutral sites have ω = 1 , those under purifying selection have ω<1 , and those under positive selection have ω>1 [55] . Likelihood ratio tests were performed using lnL values from the models M7 and M8 by comparing the test statistic 2ΔlnL = 2 ( lnLM7−lnLM8 ) , with the χ2 distribution ( d . f . = 2 ) . For model M8 , the empirical Bayes [55] procedure estimated the mean ω-value for each codon site , and the posterior probability that the site is under positive selection . | The biotrophic fungus Melampsora lini is the causal agent of flax rust disease . Flax produces immune-receptor proteins that recognize fungal effector proteins , and subsequently signal the activation of plant defense responses . Here we report the molecular details of interactions between L-locus immune-receptors and AvrL567-locus effectors , as well as the engineering of an enhanced flax immune-receptor . In order to investigate the role of AvrL567 amino acid residues hypothesized to mediate interactions with L-locus immune receptors , we generated a series of site-direct mutations in AvrL567 proteins . Conversely , to investigate the role of regions hypothesized to mediate interactions with AvrL567 effectors , we generated a series of chimeric L-locus immune-receptors that contain swaps between , and within protein domains . Interactions between modified immune-receptors and effector proteins were evaluated using the yeast-two-hybrid system and transient expression in planta . Our results revealed that interactions between L-locus immune receptors and AvrL567-locus effector proteins involve multiple surfaces , and that intramolecular interactions between , and within , domains of L-locus immune-receptors plays a crucial role in these interactions . Finally , the generation of an enhanced immune-receptor is an important proof-of-concept demonstrating the utility of protein engineering in generating novel disease resistance in agricultural crops . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"biotechnology",
"plant",
"biology",
"immunology",
"plant",
"science",
"crops",
"plant",
"pathology",
"genetically",
"modified",
"organisms",
"crop",
"diseases",
"biology",
"agriculture",
"immunity",
"innate",
"immunity",
"plant",
"pathogens",
"agricultural",
"biotechnology"
] | 2012 | Intramolecular Interaction Influences Binding of the Flax L5 and L6 Resistance Proteins to their AvrL567 Ligands |
New vector control paradigms expanding the use of spatial repellents are promising , but there are many gaps in our knowledge about how repellents work and how their long-term use might affect vector populations over time . Reported here are findings from a series of in vitro studies that investigated the plasticity and heritability of spatial repellent ( SR ) behaviors in Aedes aegypti exposed to airborne transfluthrin , including results that indicate a possible link between repellent insensitivity and insecticide resistance . A dual-choice chamber system was used to observe directional flight behaviors in Aedes aegypti mosquitoes exposed to passively emanating transfluthrin vapors ( 1 . 35 mg/m3 ) . Individual SR responder and SR non-responder mosquitoes were identified , collected and maintained separately according to their observed phenotype . Subsequent testing included re-evaluation of behavioral responses in some mosquito cohorts as well as testing the progeny of selectively bred responder and non-responder mosquito strains through nine generations . At baseline ( F0 generation ) , transfluthrin actively repelled mosquitoes in the assay system . F0 mosquitoes repelled upon initial exposure to transfluthrin vapors were no more likely to be repelled again by subsequent exposure 24h later , but repelled mosquitoes allowed to rest for 48h were subsequently repelled at a higher proportion than was observed at baseline . Selective breeding of SR responders for nine generations did not change the proportion of mosquitoes repelled in any generation . However , selective breeding of SR non-responders did produce , after four generations , a strain of mosquitoes that was insensitive to the SR activity of transfluthrin . Compared to the SR responder strain , the SR insensitive strain also demonstrated decreased susceptibility to transfluthrin toxicity in CDC bottle bioassays and a higher frequency of the V1016Ikdr mutation . SR responses to volatile transfluthrin are complex behaviors with multiple determinants in Ae . aegypti . Results indicate a role for neurotoxic irritation of mosquitoes by sub-lethal doses of airborne chemical as a mechanism by which transfluthrin can produce SR behaviors in mosquitoes . Accordingly , how prolonged exposure to sub-lethal doses of volatile pyrethroids might impact insecticide resistance in natural vector populations , and how already resistant populations might respond to a given repellent in the field , are important considerations that warrant further monitoring and study . Results also highlight the critical need to develop new repellent active ingredients with novel mechanisms of action .
New vector control tools and paradigms are desperately needed to complement existing approaches [1–3] , and there is growing evidence to support the expanded use of spatial repellents to help address this need [4–9] . The ultimate goal of public health interventions utilizing repellents is to exploit the behavior modifying effects of certain chemicals to prevent human-vector contact and , therefore , reduce disease transmission . Such approaches are among the most promising new strategies under investigation , with much progress already shown towards defining the parameters of spatial repellent-based interventions to control the global arbovirus vector Ae . aegypti [10–13] . However , there are gaps in our knowledge about how repellents work , including the exact molecular and physiological mechanisms by which various chemicals elicit SR behaviors in important vector species [5 , 14–17] and the hereditary basis by which SR behavioral traits are maintained in populations of disease vectors [18 , 19] . Spatial repellency ( SR ) is one of several behavior modifying effects of insecticides on mosquitoes that have been recognized for decades [6 , 8] and have been shown to contribute to disease reduction in many settings [5 , 20 , 13] . In outlining a new classification system to more accurately describe the actions of chemicals used for malaria vector control , Grieco et al . ( 2007 ) defined SR actions as those that stimulate “movement away from the chemical source without the mosquito making physical contact with the treated surface” [6] . An expanded concept of SR , which also includes chemical actions that interfere with host detection and/or otherwise disrupt the blood-feeding process , was established by WHO in 2013 to help determine guidelines for efficacy testing [9] . Taken together , it is clear that what is casually referred to as spatial repellency is really a set of complex and multifactorial behaviors which can be generally thought of as reactions to air-borne chemical stimuli that deter mosquitoes from entering a space to take a blood meal from an otherwise suitable host . Despite the complexities inherent in the modification of mosquito behavior , much evidence to date seems to indicate that olfactory mechanisms underlie many repellent behaviors [17 , 21 , 22] . For example , DEET , which is probably the most widely used and thoroughly studied mosquito repellent [23 , 24] , is thought to work either through direct olfactory stimulation [16 , 25] and/or through interference with normal host cue detection , essentially masking the presence of a potential blood meal [14 , 26] . Although DEET is typically found in products labeled for personal protection that are applied directly to the skin and is not , strictly speaking , a spatial repellent able to protect occupants of a defined area , knowledge of its mechanisms of action is likely to inform much of our view of how SR compounds function . Indeed , epidemiological and entomological evidence garnered from the use of indoor residual spraying with DDT for malaria control also supports a model whereby the SR action of the chemical results from a separate mechanism , likely olfaction , from that which produces neurotoxicity: SR activity is preserved in many locations where insecticide resistance is widely reported [27] . Similar observations have also been reported in pyrethroid tolerant mosquitoes that still demonstrate behavioral avoidance to sub-lethal doses of various pyrethroids [28 , 29 , 15] . Additionally , it has also long been observed that some proportion of mosquitoes continue to locate hosts and feed even in the presence of a repellent [30 , 31] , and in Ae . aegypti this DEET insensitivity has been shown to be a heritable trait with incomplete penetrance [19] associated with specific odorant receptor polymorphisms [32 , 26] . Less clear , however , is whether or not olfactory pathways are the only physiological drivers of SR behaviors in mosquitoes . For instance , Ogoma et al . ( 2014 ) have reported that airborne pyrethroids and DDT both elicit multiple behavioral effects on a given mosquito population at the same time , including deterrence ( the prevention of mosquito entrance into a structure ) , irritancy and excito-repellency ( eliciting the premature exit of mosquitoes from a structure via physical contact with an insecticide treated surface or with insecticide vapors , respectively ) , reduced blood feeding , increased 24h mortality and reduced fecundity [7] . Kawada et al . recently reported reduced pyrethroid ( permethrin and deltamethrin ) contact repellency in a strain of Anopheles gambiae s . s . with the L1014Skdr mutation , but not in strains of An . arabiensis or An . funestus s . s . with cytochrome P450 driven metabolic resistance traits , supporting a role for the non-lethal disruption of neuronal sodium ion channel function in eliciting the observed excito-repellency/irritancy behaviors [15] . While they did not evaluate SR behaviors specifically , these results are in line with previous knowledge that many pyrethroid compounds ( i . e . , permethrin , deltamethrin and alphacypermethrin ) can induce irritant and/or hyperactive responses in mosquitoes at sub-lethal concentrations [33 , 34] and this hyperactivity can promote the avoidance of insecticide treated nets [35] . It is clear that physical contact with surfaces treated with these pyrethroid insecticides can produce repellency behaviors through neurologically disruptive mechanisms . It is unknown , however , whether or not a highly active and more volatile pyrethroid insecticide like transfluthrin , which also has SR properties [36 , 37 , 12 , 7] , elicits the same physiological responses through airborne exposure . This question is especially important as residual pyrethroids are currently the most commonly used class of public health insecticide worldwide and there are growing concerns about the rapid expansion of pyrethroid resistance in key vector species [5 , 38 , 39] . Critically , it is unclear how the use of volatile compounds that could act through the same physiological pathways as the most commonly used residual insecticides might complicate the insecticide resistance landscape . Given the complex and multifactorial nature of SR behaviors in mosquitoes , the molecular and hereditary drivers of the behavior are likely to vary across different active ingredients and target organisms . Nonetheless , elucidating which mechanisms dominate in specific transmission settings is an important step to understanding how to best use spatial repellents in a public health context [40] and how their long-term use might impact vector populations over time [6 , 29] . Additionally , this data could be used to guide the rational design of new active ingredients that mitigate resistance driving mechanisms [5] . Here , we report on a series of in vitro experiments that first examined the plasticity and heritability of non-contact SR behaviors in Ae . aegypti that were exposed to airborne transfluthrin , and subsequently explored a link between SR insensitivity and reduced insecticide susceptibility in a selectively bred strain of this important arbovirus vector .
Aedes aegypti ( L . ) mosquitoes were colonized from wild-caught ( P1 ) larvae collected from discarded automobile tires near the Belize Vector and Ecology Center ( BVEC ) in Orange Walk Town , Belize ( 18°04 . 938’N , 88°33 . 390’W ) . The P1—F4 generations were reared and tested at the BVEC field laboratory at ambient light , temperature and humidity . Later generations ( F5—F10 ) and experimental crosses were reared and tested under climate controlled conditions ( 28°C , 60% RH , and 12L:12D light-dark schedule ) at the Uniformed Serviced University of the Health Sciences ( USUHS ) in Bethesda , MD . Larvae were fed Chiclid Gold fish pellets ( Kyorin Co . , LTD , Himeji , Japan ) and adults were provided 10% sucrose solution from soaked cotton ad libitum . Using CDC bottle bioassay methods , F0 adults exhibited greater than 90% susceptibility to transfluthrin , malathion and DDT at 60 minutes ( S1 Fig ) . SR Behavioral assays were performed using 5–12 day old mosquitoes , which were sorted into cohorts of 20 mosquitoes approximately 24h prior to testing . Female test mosquitoes were unmated , to allow for downstream selective breeding , and were sugar starved ( provided only water-soaked cotton ) for approximately 24h before testing , following standardized methods [41] . Because high mortality rates were observed in male mosquito populations , they were not sugar starved prior to testing . SR behavior was evaluated using a high throughput screening system ( HITSS-SRA configuration ) ( Fig 1 ) , previously described by Grieco et al . ( 2007 ) [41] and recently adopted by the WHO as a standard procedure for in vitro efficacy testing of spatial repellents [9] . The dual-choice chamber system , which allows the observation of directional mosquito movement in response to a single chemical stimulus outside the context of host cues , consists of a clear Plexiglas central unit connected at opposite ends to one treatment chamber housing repellent-treated netting and one control chamber housing a net treated with acetone only ( Fig 1 ) . Tests were conducted to evaluate Ae . aegypti SR responses to passively emanating transfluthrin ( 2 , 3 , 5 , 6-tetrafluorobenzyl ( 1R ) -trans-3- ( 2 , 2-dichlorovinyl ) -2 , 2-dimethyl cyclopropanecarboxylate ) ( S . C . Johnson and Son , Inc . , Racine WI ) , a volatile synthetic pyrethroid with widely demonstrated SR efficacy against mosquitoes [7 , 36 , 37 , 12] . Briefly , reagent grade ( unformulated ) transfluthrin was dissolved in 100% acetone ( Hofius Ltd . /Ace Hardware , Belize City and Fisher Scientific , Waltham MA ) . This solution was then applied evenly by micropipette across the surface of 11cm x 25cm pieces of nylon organdy netting ( No . I10N , G-Street Fabrics , Bethesda MD ) and allowed to air dry a minimum of 15 minutes before use . Industry guidelines ( M . C . Meier , personal communication , 16 August 2011 ) and concurrent experimental hut studies using transfluthrin in Belize [42] indicate a standard field application rate ( FAR ) of 1 . 35mg active ingredient per cubic meter of airspace to produce indoor SR activity against mosquitoes via passive emanation . Accordingly , HITSS treatment nets delivering 1x the FAR into the assay system were treated with 0 . 9mL of a 2 . 2x10-6 M ( 8 . 4x10-4 mg/mL ) solution . Concentrations tested ranged from 0 . 5xFAR to 1000xFAR . Control nets were treated with 100% acetone only . Cohorts of 20 mosquitoes were introduced into the central HITSS chamber and , after a 30 second acclimation period , butterfly valves situated at both ends of the central chamber were opened simultaneously to allow free movement of mosquitoes in either direction into either end chamber . After a ten minute exposure period , the butterfly valves were closed and the numbers of mosquitoes in each chamber were counted . Spatial repellency is measured by considering the number of mosquitoes that have moved into the untreated , control chamber ( away from the treated surface ) relative to the total number of mosquitoes that have moved in either direction using a weighted spatial activity index ( SAI ) , equal to [ ( Nc- Nt ) / ( Nc+ Nt ) ]x[ ( Nc+ Nt ) ]/N] where N is the total number of mosquitoes per replicate and Nc and Nt are the number of mosquitoes in the control and treatment chambers , respectively . Possible values for the weighted SAI range from 1 to -1 , with a value of 1 indicating the strongest SR response possible ( movement of all mosquitoes away from the chemical source ) , zero indicating no net response , and a value of -1 indicative of a strong attractive response ( movement of all mosquitoes towards the chemical source ) . To account for mosquito mortality , the total number of mosquitoes tested per each replicate was corrected using Abbott’s formula [43] . A SR dose-response curve was established using unselected ( control ) females by varying the dose of transfluthrin in the HITSS treatment chamber and measuring differences in corresponding SAI values and overall assay mortality ( S2 Fig ) . The dose corresponding to 1xFAR ( 1 . 35 mg/m3 ) produced the largest SAI value ( 0 . 10 , significantly greater than zero at P<0 . 02 ) and an overall non-contact mortality of only 2 . 8% and was selected for use in all subsequent HITSS SR replicates . Male and unmated nulliparous female mosquitoes were tested separately and , after each experimental replicate , were identified as either SRA responders ( SRA+ ) if they had escaped into the untreated control chamber or SRA non-responders ( SRA- ) if they either stayed in the central chamber or flew into the treatment chamber ( S3 Fig ) . Mosquitoes that were located in the treatment chamber at the end of a replicate ( i . e . had made physical contact with the transfluthrin treated netting ) were enumerated for statistical purposes but then discarded and not further processed or analyzed . Though both male and female mosquitoes were tested during these experiments , only female behavior was analyzed statistically and only female results are presented here . Typically , males were tested in fewer replicates only to provide sufficient numbers of each behavioral phenotype ( SRA+ responders and SRA- non-responders ) for selective mating purposes . To evaluate the plasticity of SR responses in unselected F0 females exposed to transfluthrin , test replicates were performed and mosquitoes were immediately collected and maintained separately based on their observed behavioral phenotype , i . e . SRA+ responders and SRA- non-responders . Mosquitoes were re-assayed on a subsequent day ( day 2 ) , after either a 24h or 48h resting period , and the weighted SAI for each phenotype cohort was compared to baseline ( day 1 ) results using Student’s t-test at 95% confidence . The heritability of SR behavioral responses was evaluated by performing test replicates and collecting mosquitoes based on their SR behavioral phenotype , as described above ( S3 Fig ) . SR responder females were then selectively mated with SR responder males to establish an SRA+ strain of Ae . aegypti , and non-responder females were mated with non-responder males to establish an SRA- strain . Changes in the SAI scores in test populations from each strain were followed for 9 generations and were compared using ANOVA with Dunnett’s test for multiple comparisons at 95% confidence . An additional control strain of Ae . aegypti originating from the same field collected P1 larvae but which was allowed to freely mate was also maintained and tested . In order to monitor relative changes in transfluthrin insecticide susceptibility over time and across different experimental populations , CDC bottle bioassay tests [43] were performed at various selection points , including the F0 , F5 and F8 generations and in progeny from an experimental cross between F9 SRA- females and newly colonized wild type F0 males . A discriminating dose of 94 ng transfluthrin ( 0 . 25 nm , approximately 0 . 125xFAR ) per bottle was established using F2 unselected control females ( S1 Table ) . Test replicates lasted one hour , with mosquito knockdown recorded every 15m and final mortality recorded at 24hr . Using the PCR genotyping approach developed by Linss et al . ( 2014 ) [44] , Ae . aegypti voltage gated sodium ion channel V1016I and F1534C kdr allele frequencies were estimated using cohorts of 30 mosquitoes each from the F9 Control , SRA+ and SRA- populations and the experimental cross progeny . Both target site mutations have been previously observed in Ae . aegypti populations from Latin America and the Caribbean and have been shown to contribute to pyrethroid resistance [45 , 46 , 44] . Unless otherwise noted , SAI scores were calculated for each test population at each time point using 180 total mosquitoes , consisting of 9 replicates of 20 mosquitoes each , following established procedures [9] . Herein , the term ‘test population’ is used to refer to a sample of mosquitoes from a unique generation ( e . g . F3 ) of a unique behavioral phenotype ‘strain’ ( e . g . SRA- , SRA+ or control ) . Raw data was organized and descriptive analyses were performed using Excel 2007 ( Microsoft Corp . , Albuquerque NM ) . A non-parametric signed rank test ( PROC UNIVARIATE ) in SAS v8 statistical software ( SAS Institute Inc . , Cary , NC ) was used to determine if mean SAI values were different from zero for each test population . SAI values were compared between populations via Student’s t-test and ANOVA with Dunnett’s test for multiple comparisons using SPSS Statistics 22 software ( IBM Corp . , Armonk NY ) . The kdr allele frequencies and herterozygosity were compared using Z-tests on the difference between sample proportions , and a chi-square test with one degree of freedom was used to evaluate deviations from Hardy-Weinberg equilibrium [47] . All analyses were performed at α = 0 . 05 .
Two variations of the behavioral plasticity experiment were performed using F0 mosquitoes , with differing results ( Table 1 and Fig 2 ) . During the first experiment , mosquito cohorts ( total n = 180 mosquitoes , average baseline SAI = 0 . 08 ±0 . 03 SEM ) were re-assayed after a 24 hour rest period and results indicated a large degree of plasticity in behavioral responses to the repellent: mosquitoes repelled on day one ( n = 29 ) were not more likely to be repelled again on day two ( SAI = 0 . 03 ± 0 . 02 ) ( Fig 2 ) . Mosquitoes not repelled on day one ( n = 129 ) were equally unlikely to be repelled on day two ( SAI = 0 . 03 ±0 . 04 ) ( Fig 2 ) . For the second experiment , mosquitoes ( total n = 280 , average baseline SAI = 0 . 05 ±0 . 04 ) were not re-assayed until the second day after the original test ( 48 hours post exposure ) . Unlike mosquitoes that were allowed to rest for 24hr , day one repellent responders from this cohort ( n = 60 ) were more likely to be repelled again on day two ( SAI = 0 . 30 ±0 . 08 , P<0 . 05 ) ( Fig 2 ) . As was observed in the first experiment , non-responding mosquitoes from this experiment ( n = 155 ) were also equally non-responsive on day two ( SAI = 0 . 06±0 . 04 ) ( Fig 2 ) . The baseline average SAI value for F0 female mosquitoes , which gave rise to all subsequent SRA+ and SRA- lineages , was 0 . 14 ±0 . 06 ( significantly greater than zero at P<0 . 02 ) , confirming that parental mosquitoes were actively repelled by volatile transfluthrin in the assay system . Selective breeding experiments were then carried out through the F9 generation ( Table 2 and Fig 3 ) . SAI results from the unselected control strain ( S4 Fig ) and the SRA+ strain ( Fig 3 ) did not indicate any changes in behavioral responses to volatile transfluthrin at any time point compared to baseline ( no significant differences at P = 0 . 05 ) . Results from the SRA- strain , on the other hand , showed a steady decrease in SAI scores , which reached statistical significance ( P<0 . 05 ) by the F4 generation ( SAI = -0 . 05 ±0 . 04 ) ( Fig 3 ) . This SR insensitive phenotype was confirmed in each subsequent SRA- generation , with the exception of the F7 cohort in which the reduced SAI value ( 0 . 02 ±0 . 03 ) was not significantly different from baseline at P = 0 . 05 ( Fig 3 ) . Baseline CDC bottle tests indicated greater than 95% susceptibility to transfluthrin toxicity ( 24hr mortality ) at the discriminating dose in the F0 parental mosquitoes that gave rise to all selectively bred strains ( Fig 4 ) . Insecticide susceptibility was then reevaluated in the F5 and F8 generations of colony and selectively bred mosquitoes ( Fig 4 ) . For the colony ( unselected control , S5 Fig ) and SRA+ ( responder , Fig 4 ) strains , no significant changes in insecticide susceptibility were noted by either time to knockdown or 24hr mortality . In the selectively bred SRA- repellent insensitive strain there was a moderate but significant ( P<0 . 05 ) 23% reduction in mortality observed in the F6 generation compared to the control strain ( 60% ±1% vs . 95% ±6% ) while the F8 SRA- test population was highly resistant with a mortality of just 14% ±11% , a significant ( P<0 . 01 ) 77% reduction in mortality compared to the unselected control ( Fig 4A ) . An additional round of selective breeding of F8 SRA- non-responders gave rise to F9 SRA- mosquitoes that continued to exhibit repellent insensitivity ( SAI = -0 . 04 ±0 . 05 ) ( Fig 3 ) as well as significantly decreased CDC bottle assay knockdown and 24h mortality ( 13% ±13% ) ( Fig 4 ) . Mating females from the F8 SRA- population with wild type F0 males newly colonized from the same location in Belize , however , restored both transfluthrin SR sensitivity ( SAI = 0 . 11 ±0 . 03 ) ( Table 2 and Fig 3 ) and insecticide susceptibility ( 24h mortality = 84% ±7% ) in the resulting progeny ( Fig 4 ) . Analysis of kdr allele frequencies was performed in the F9 control , F9 SRA+ , F9 SRA- , and experimental cross progeny cohorts . Results indicated that the V1016Ikdr allele was more frequent ( 50% ) in the SR insensitive , insecticide resistant SRA- population than in the susceptible SRA+ ( 16% , P<0 . 01 ) or the control ( 22% , P<0 . 02 ) cohorts ( Fig 5 ) . Overall V1016Ikdr allele frequency remained high in the experimental cross progeny in which SR sensitivity and insecticide susceptibility were both restored ( Fig 5 ) . However , there was a significant ( P<0 . 01 ) increase in the proportion of heterozygotes , from 27% in the SRA- population to 65% in the experimental cross offspring ( Fig 5 ) . The assumption of Hardy-Weinberg equilibrium was rejected in both of the SRA+ ( χ2 = 10 . 25 , P<0 . 01 ) and SRA- strains ( χ2 = 6 . 53 , P<0 . 02 ) , but not in either the control population or experimental cross progeny . There were no differences or changes in F1534Ckdr allele frequencies observed , with kdr prevalence over 90% for all cohorts tested .
Collectively , these results show that the in vitro SR responses observed here are complex behaviors with a mix of heritable and non-heritable determinants . Based on the link between the SR insensitive phenotype and decreased insecticide susceptibility , evidence also supports a model whereby sub-lethal doses of volatile transfluthrin can elicit SR responses in Ae . aegypti by inducing a hyperactive or agitated state via neurotoxic pathways , likely independent of olfactory stimulation or interruption . Care should be taken before extrapolating these results to other active ingredients or vector species . It should also be emphasized that these results do not indicate that transfluthrin elicits SR behaviors in Ae . aegypti exclusively by disrupting motor-neuron activity: olfactory and/or gustatory pathways may also play a role , whether via active detection and avoidance of odor cues or through the disruption of host detection and/or feeding , possibilities that should continue to be investigated using a variety of methods . Additionally , the appearance of decreased insecticide susceptibility and increased kdr allele frequency in the selectively bred offspring of mosquitoes exposed only to sub-lethal insecticide vapors raises some important questions about how the long-term use of repellents might impact vector populations over time . The answers to these questions will be dependent on several factors including which molecular mechanisms are driving specific repellent behaviors , the hereditary nature of repellent sensitivity and insensitivity , and other physiological effects of using sub-lethal concentrations of compounds that have insecticidal , as well as repellent , properties . Though the story is complex and further research is needed to better understand all of the physiological drivers of SR behaviors , evidence still supports the expanded use of spatial repellents in public health applications to control disease vectors , albeit with continued monitoring of potential changes in target vector repellent sensitivities and/or insecticide susceptibilities and a renewed emphasis on the need to develop new active ingredients with novel , non-toxic mechanisms of action . | There is growing evidence to support the expanded use of spatial repellents for vector control , but there are still many uncertainties about how repellents work and how their long term use may impact vector populations over time . Here , we conducted a series of in vitro experiments that investigated spatial repellent ( SR ) behaviors in Aedes aegypti mosquitoes exposed to airborne transfluthrin , a volatile pyrethroid commonly used in repellent products . We were able to show that repellent insensitivity is linked to reduced insecticide susceptibility and increased knock down resistance allele frequency , indicating that sub-lethal doses of airborne transfluthrin can elicit SR behaviors in mosquitoes by inducing an agitated state via neurotoxic pathways independent of olfactory stimulation . This raises questions about how the use of volatile pyrethroid repellents may impact insecticide resistance in target vectors over time , highlighting the need to further understand all of the physiological drivers of SR behaviors and emphasizing the requirement to develop new repellent active ingredients with novel , non-toxic mechanisms of action . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [] | 2015 | Insensitivity to the Spatial Repellent Action of Transfluthrin in Aedes aegypti: A Heritable Trait Associated with Decreased Insecticide Susceptibility |
To elucidate the history of living and extinct elephantids , we generated 39 , 763 bp of aligned nuclear DNA sequence across 375 loci for African savanna elephant , African forest elephant , Asian elephant , the extinct American mastodon , and the woolly mammoth . Our data establish that the Asian elephant is the closest living relative of the extinct mammoth in the nuclear genome , extending previous findings from mitochondrial DNA analyses . We also find that savanna and forest elephants , which some have argued are the same species , are as or more divergent in the nuclear genome as mammoths and Asian elephants , which are considered to be distinct genera , thus resolving a long-standing debate about the appropriate taxonomic classification of the African elephants . Finally , we document a much larger effective population size in forest elephants compared with the other elephantid taxa , likely reflecting species differences in ancient geographic structure and range and differences in life history traits such as variance in male reproductive success .
The technology for sequencing DNA from extinct species such as mastodons ( genus Mammut ) and mammoths ( genus Mammuthus ) provides a powerful tool for elucidating the phylogeny of the Elephantidae , a family that originated in the Miocene and that includes Asian elephants ( genus Elephas ) , African elephants ( genus Loxodonta ) , and extinct mammoths [1]–[8] . In the highest resolution study to date , complete mitochondrial DNA ( mtDNA ) genomes from three elephantid genera were compared to the mastodon outgroup . The mtDNA analysis suggested that mammoths and Asian elephants form a clade with an estimated genetic divergence time of 5 . 8–7 . 8 million years ago ( Mya ) , while African elephants diverged from an earlier common ancestor 6 . 6–8 . 8 Mya [8] . However , mtDNA represents just a single locus in the genome and need not represent the true species phylogeny since a single gene tree can differ from the consensus species tree of the taxa in question [9]–[11] . Generalizing about species relationships based on mtDNA alone is especially problematic for the Elephantidae because their core social groups ( “herds” ) are matrilocal , with females rarely , if ever , dispersing across groups [12] . This results in mtDNA genealogies in both African [13] , [14] and Asian elephants [15] that exhibit deeper divergence and/or different phylogeographic patterns than the nuclear genome . These observed discrepancies between the phylogeographic patterns of nuclear and mtDNA sequences have led to a debate about the appropriate taxonomic status of African elephants . Most researchers have argued , based on morphology and nuclear DNA markers , that forest ( Loxodonta cyclotis ) and savanna ( Loxodonta africana ) elephants should be considered separate species [13] , [16]–[19] . However , this notion has been contested [20] based on mtDNA patterns , which reveal some haplogroups with coalescent times of less than half a million years [21] that are shared across forest and savanna elephants , indicating relatively recent gene flow among the ancestors of these taxa . Taxonomies for African elephants based on mtDNA phylogeographic patterns have suggested anywhere from one to four species [20] , [22] , [23] , whereas analysis of morphology and nuclear data sets has suggested two species [13] , [16]–[19] . The study of large amounts of nuclear DNA sequences has the potential to resolve elephantid phylogeny , but due to technical challenges associated with obtaining homologous data sets from fossil DNA , no sufficiently large nuclear DNA data set has been published to date . Although a draft genome is available for woolly mammoth ( Mammuthus primigenius ) [5] and savanna elephant ( loxAfr; http://www . broadinstitute . org/ftp/pub/assemblies/mammals/elephant/ ) , comparative sequence data are lacking for Asian ( Elephas maximus ) and forest elephant , as well as for a suitable outgroup like the American mastodon ( Mammut americanum ) . Using a combination of next generation sequencing and targeted multiplex PCR , we obtained the first substantial nuclear data set for comparing these species .
We carried out shotgun sequencing of DNA from an American mastodon with a Roche 454 Genome Sequencer ( GS ) , using the same DNA extract from a 50 , 000–130 , 000-yr-old tooth that we previously used to generate a complete mtDNA genome sequence from the mastodon [8] . After comparing the 45 Mb of shotgun DNA data that we obtained to the Genbank database , and only retaining reads for which the best match was to sequences of the savanna elephant draft sequence ( loxAfr1 ) , we were left with 1 . 76 Mb of mastodon sequence ( Figure 1 and Figure S1 ) . To amplify the same set of loci across all species , we designed PCR primers flanking the regions of mastodon-elephant alignment , using the loxAfr1 savanna elephant sequence as a template ( Figure 1 ) ( a full list of the primers is presented in Dataset S1 ) . We used these primers in a multiplexed protocol [24] to amplify one or two Asian elephants , one African forest elephant , one woolly mammoth , and one African savanna elephant unrelated to the individual used for the reference sequence ( Figure 1 and Table S1 ) . We then sequenced the products on a Roche 454 GS to a median coverage of 41-fold and assembled a consensus sequence for each individual by restricting to nucleotides with at least 3-fold coverage . After four rounds of amplification and sequencing , we obtained 39 , 763 base pairs across 375 loci with data from all five taxa ( Text S1; Figure S2; Table S2 , Table S3 ) . We identified 1 , 797 nucleotides in this data set in which two different alleles were observed and used these sites for the majority of our analyses ( the genotypes are provided in Dataset S2 ) . A total of 549 of these biallelic sites were polymorphic among the elephantids , while the remaining sites were fixed differences compared to the mastodon sequence . To assess the utility of the data for molecular dating and inference about demographic history , we carried out a series of relative rate tests , searching for an excess of divergent sites in one taxon compared to another since their split , which could reflect sequencing errors or changes in the molecular clock [25] . None of the pairs of taxa showed a significant excess of divergent sites compared with any other ( Table 1 ) . When we compared the data within taxa , we found that the savanna reference genome loxAfr1 had a significantly higher number of lineage-specific substitutions than the savanna elephant we sequenced ( nominal P = 0 . 03 from a two-sided test without correcting for multiple hypothesis testing ) . This is consistent with our data being of higher quality than the loxAfr1 reference sequence , presumably due to our high read coverage . In contrast to our elephantid data , our mastodon data had a high error rate , as expected given that it was derived from shotgun sequencing data providing only 1-fold coverage at each position . To better understand the effect of errors in the mastodon sequence , we PCR-amplified a subset of loci in the mastodon , obtaining high-quality mastodon data at 1 , 726 bases ( Text S2 ) . Of the n = 23 sites overlapping these bases that we knew were polymorphic among the elephantids , the mastodon allele call always agreed between the PCR and shotgun data , indicating that our mastodon data are reliable for the purpose of determining an ancestral allele ( the main purpose for which we use the mastodon data ) . However , only 38% of mastodon-elephantid divergent sites validated , which we ascribe to mastodon-specific errors , since almost all the discrepancies were consistent with C/G-to-T/A misincorporations ( the most prominent error in ancient DNA ) [26]–[28] , or mismapping of some of the short mastodon reads ( 2 ) . Thus , our raw estimate of mastodon-elephantid divergence is too high , making it inappropriate to use mastodon for calibrating genetic divergences among the elephantids , as we previously did for mtDNA where we had high-quality mastodon data [8] . We estimated the relative genetic diversity across elephantids by counting the total number of heterozygous genotypes in each taxon , and normalizing by the total number of sites differing between ( S ) avanna and ( A ) sian elephants ( tSA ) . Within-species genetic diversity as a fraction of savanna-Asian divergence is estimated to be similar for savanna elephants ( 8±2% ) and mammoths ( 9±2% ) , higher for Asian elephants ( 15±3% ) , and much higher for forest elephants ( 30±4% ) ( standard errors from a Weighted Jackknife; Methods ) . This supports previous findings of a higher average time to the most recent common genetic ancestor in forest compared to savanna elephants ( Table 1 ) [13] , [17] . We caution that these diversity estimates are based on analyzing only a single individual from each taxon , which could produce a too-low estimate of diversity in the context of recent inbreeding . Encouragingly , however , in Asian elephants where two individuals were sequenced for some loci , genetic diversity estimates are consistent whether measured across ( 18±5% ) or within samples ( 15±3% ) . A further potential concern is “allele specific PCR” , whereby one allele is preferentially amplified causing truly heterozygous sites to go undetected [29] . However , we do not believe that this is a concern since we preformed an experiment in which we re-amplified about 5% of our loci using different primers and obtained identical genotypes at all sites where we had overlapping data ( Text S2 ) . We next inferred a nuclear phylogeny for the elephantids using the Neighbor Joining method ( Methods and Figure S3 ) . This analysis suggests that mammoths and Asian elephants are sister taxa , consistent with the mtDNA phylogeny [8] , and that forest and savanna elephants are also sister taxa . We estimate that forest-savanna genetic divergence normalized by savanna-Asian is tFS/tSA = 74±6% , while Asian-mammoth genetic divergence normalized by savanna-Asian tAM/tSA = 65±5% ( Table 1 ) . These numbers are all significantly lower than savanna-mammoth ( tSM/tSA = 92±5% ) , forest-Asian ( tFA/tSA = 103±5% ) , and forest-mammoth ( tFM/tSA = 96±7% ) normalized by savanna-Asian genetic divergence , which are all consistent with 100% as expected if they reflect the same comparison across sister groups ( Table 1 ) . An intriguing observation is that the ratio of forest-savanna elephant genetic divergence to Asian-mammoth divergence tFS/tAM is consistent with unity ( 90% credible interval 90%–138% ) , which is interesting given that forest and savanna elephants are sometimes classified as the same species , whereas Asian elephants and mammoth are classified as different genera [20] , [30] . To further explore this issue , we focused on regions of the genome where the genealogical tree is inconsistent with the species phylogeny , a phenomenon known as “incomplete lineage sorting” ( ILS ) [8] , [11] , [31] . Information about the rate of ILS can be gleaned from the rate at which alleles are observed that cluster taxa that are not most closely related according to the overall phylogeny . For example , in a four-taxon alignment of ( S ) avanna , ( F ) orest , ( E ) urasian , and mastodon , “SE” and “FE” alleles that cluster savanna-Eurasian or forest-Eurasian , to the exclusion of the other taxa , are likely to be at loci with ILS ( in what follows , we use the term “Eurasian elephants” to refer to woolly mammoths and Asian elephants , while recognizing that the range of the lineage ancestral to each species included Africa as well ) . Similarly , in a four-taxon alignment of ( A ) sian , ( M ) ammoth , ( L ) oxodonta ( forest plus savanna ) , and mastodon , “AL” or “ML” sites reveal probable ILS events . We find a higher rate of inferred ILS in forest and savanna elephants than in Asian elephants and mammoths: ( FE+SE ) / ( AL+ML ) = 3 . 1 ( P = 4×10−8 for exceeding unity; Table 2 ) , indicating that there are more lineages where savanna and forest elephants are unrelated back to the African-Eurasian speciation than is the case for Asian elephants and mammoths ( Table 2 ) . This could reflect a history in which the savanna-forest population divergence time TFS is older than the Asian-mammoth divergence time TAM , a larger population size ancestral to the African than to the Eurasian elephants , or a long period of gene flow between two incipient taxa . ( We use upper case “T” to indicate population divergence time and lower case “t” to indicate average genetic divergence time ( t≥T ) ) . To further understand the history of the elephantids , we fit a population genetic model to the data ( input file—Dataset S3 ) using the MCMCcoal ( Markov Chain Monte Carlo coalescent ) method of Yang and Rannala [32] . We fit a model in which the populations split instantaneously at times ΤFS ( forest-savanna ) , ΤAM ( Asian-mammoth ) , ΤLox-Eur ( African-Eurasian ) , and ΤElephantid-Mastodon , with constant population sizes ancestral to these speciation events of ΝFS , ΝAM , ΝLox-Eur , and ΝElephantid-Mastodon , and ( after the final divergences ) of ΝF , ΝS , ΝA , and ΝM ( Figure 2 ) . We recognize that elephantid population sizes likely varied within these time intervals , given recurrent glacial cycles [33] , changes in geographic ranges documented in the fossil record [15] , [30] , [34] , [35] , and mtDNA patterns suggesting ancient population substructure [13] , [15] . Nevertheless , the constant population size assumption is useful for inferring average diversity and obtaining an initial picture of elephantid history . MCMCcoal then makes the further simplifying assumptions that our short ( average 106 bp ) loci experienced no recombination and that they are unlinked ( the latter assumption is justified by the fact that when we mapped the loci to scaffolds from the loxAfr3 genome sequence , all but one pair were at least 100 kilobases apart; Text S3 ) . MCMCcoal then infers the joint distribution of the “T” and “N” parameters that is consistent with the data , as well as the associated credible intervals ( Table 3; Text S4 ) . The MCMCcoal analysis infers that the initial divergence of forest and savanna elephant ancestors occurred at least a couple of Mya . The first line of evidence for this is that forest-savanna elephant population divergence time is estimated to be comparable to that of Asian elephants and mammoths: ΤAM/ΤFS = 0 . 96 ( 0 . 69−1 . 36 ) ( Table 4 ) . Secondly , MCMCcoal infers that the ratio of forest-savanna to African-Eurasian elephant population divergence is at least 45%: ΤFS/ΤLox-Eur = 0 . 62 ( 0 . 45−0 . 79 ) ( Table 4 ) . Given that African-Eurasian genetic divergence ( TLox-Eur ) can be inferred from the fossil record to have occurred 4 . 2–9 . 0 Mya ( Text S5 ) , this allows us to conclude that forest-savanna divergence occurred at least 1 . 9 Mya ( 4 . 2 Mya × 0 . 45 ) . We caution that because MCMCcoal fits a model of instantaneous population divergence , our results do not rule out some forest-savanna gene flow having occurred more recently , as indeed must have occurred based on the mtDNA haplogroup that is shared among some forest and savanna elephants . However , such gene flow would mean that the initial population divergence must have been even older to explain the patterns we observe . We also used the MCMCcoal results to learn more about the timing of the divergences among the elephantids ( Figure 2 ) . To be conservative , we quote intervals that take into account the full range of uncertainty from both the fossil calibration of African-Eurasian population divergence ( TLox-Eur = 4 . 2–9 . 0 Mya; Text S5 ) , and the 90% credible intervals from MCMCcoal ( TFS/TLox-Eur = 45%–79% and TAM/TLox-Eur = 46%–74%; Table 4 ) . Thus , we conservatively estimate TFS = 1 . 9–7 . 1 Mya and TAM = 1 . 9–6 . 7 Mya . Our inference of TAM is somewhat less than the mtDNA estimate of genetic divergence of 5 . 8–7 . 8 Mya [8] . However , this is expected , since genetic divergence time is guaranteed to be at least as old as population divergence but may be much older , especially as deep-rooting mtDNA lineages are empirically observed to occur in matrilocal elephantid species .
Our study of the extant elephantids provides support for the proposed classification of the Elephantidae by Shoshani and Tassy , which divides them into the tribe Elephantini ( including Elephas—the Asian elephant and fossil relatives—and the extinct mammoths Mammuthus ) and the tribe Loxodontini ( consisting of Loxodonta: African forest and savanna elephants and extinct relatives ) [36] . This classification is at odds with previous suggestions that the extinct mammoths may have been more closely related to African than to Asian elephants [37] . Our study also infers a strikingly deep population divergence time between forest and savanna elephant , supporting morphological and genetic studies that have classified forest and savanna elephants as distinct species [13] , [16]– . The finding of deep nuclear divergence is important in light of findings from mtDNA , which indicate that the F-haplogroup is shared between some forest and savanna elephants , implying a common maternal ancestor within the last half million years [21] . The incongruent patterns between the nuclear genome and mtDNA ( “cytonuclear dissociation” ) have been hypothesized to be related to the matrilocal behavior of elephantids , whereby males disperse from core social groups ( “herds” ) but females do not [13] , [38] . If forest elephant female herds experienced repeated waves of migration from dominant savanna bulls , displacing more and more of the nuclear gene pool in each wave , this could explain why today there are some savanna herds that have mtDNA that is characteristic of forest elephants but little or no trace of forest DNA in the nuclear genome [13] , [14] , [39] , [40] . In the future , it may be possible to distinguish between models of a single ancient population split between forest and savanna elephants , or an even older split with longer drawn out gene flow , by applying methods like Isolation and Migration ( IM ) models to data sets including more individuals [41] . Our present data do not permit such analysis , however , as IM requires multiple samples from each taxon to have statistical power , and we only have 1–2 samples from each taxon . Our study also documents the highly variable population sizes across recent elephantid taxa and in particular indicates that the recent effective population size of forest elephants in the nuclear genome ( NF ) has been significantly larger than those of the other elephantids ( NS , NA , and NM ) ( Table 5 ) [13] , [17] , [19] . This is not likely due to the “out of Africa” migration of the ancestors of mammoths and Asian elephants as these events occurred several Mya [35] , and any loss of diversity due to founder effects would have been expected to be offset by subsequent accumulation of new mutations in the populations . The high effective population size in forest elephants could reflect a history of separation of populations into distinct isolated tropical forest refugia during glacial cycles [33] , which would have been a mechanism by which ancestral genetic diversity could have been preserved before the population subsequently remixed [1] , [2] , [23] . A Pleistocene isolation followed by remixing would also be consistent with the patterns observed in Asian elephants , which carry two deep mtDNA clades and where there is intermediate nuclear diversity . Intriguingly , our estimate of recent forest effective population size is on the same order as the ancestral population sizes ( NFS , NAM , and NLox-Eur ) ( Table 5 ) , providing some support for the hypothesis that forest elephant population parameters today may be typical of the ancestral populations ( a caveat , however , is that MCMCcoal may overestimate ancestral population sizes since unmodeled sources of variation across loci may inflate estimates of ancestral population size ) . An alternative hypothesis that seems plausible is that the large differences in intra-species genetic diversity across taxa could reflect differences in the variance of male reproductive success [42] ( more male competition in mammoth and savanna elephant than among forest elephants , with the Asian elephant being intermediate [43] ) . The results of this study are finally intriguing in light of fossil evidence that forest and savanna lineages of Loxodonta may have been geographically isolated until recently . The predominant elephant species in the fossil record of the African savannas for most of the Pliocene and Pleistocene belonged to the genus Elephas [30] , [34] , [35] . Some authors have suggested that the geographic range of Loxodonta in the African savannas may have been circumscribed by Elephas , until the latter disappeared from Africa towards the Late Pleistocene [30] , [34] , [35] . We hypothesize that the widespread distribution of Elephas in Africa may have created an isolation barrier that separated savanna and forest elephants , so that gene flow became common only much later , contributing to the patterns observed in mtDNA . Further insight into the dynamics of forest-savanna elephant interaction will be possible once more samples are analyzed from all the taxa , and high-quality whole genome sequences of forest and savanna elephants are available and can be compared with sequences of Asian elephants , mammoths , and mastodons .
For our sequencing of mastodon , we used the same DNA extract that was previously used to generate the complete mitochondrial genome of a mastodon [8] . We sequenced the extract on a Roche 454 GS , resulting in 45 Mb of sequences that we deposited in the NCBI short read archive ( accession: SRA010805 ) . By comparing these reads to the African savanna elephant genome ( loxAfr1 ) using MEGABLAST , we identified 1 . 76 Mb of mastodon sequences with a best hit to loxAfr1 that we then used in downstream analyses . To re-sequence a subset of these loci in the living elephants and the woolly mammoth , we used Primer3 to design primers surrounding the longest mastodon-African elephant alignments . A two-step multiplex PCR approach [24] was used to attempt to sequence 746 loci in 1 mammoth , 1 African savanna elephant , 1 African forest elephant , and 1–2 Asian elephants . After the simplex reactions for each sample , the PCR products were pooled in equimolar amounts for each sample and then sequenced on a Roche 454 GS , resulting in an average read coverage of 41× per nucleotide ( Text S1 ) . We carried out four rounds of PCR in an attempt to obtain data from as many loci as possible and to fill in data from loci that failed or gave too few sequences in previous rounds ( Text S1 ) . To analyze the data , we sorted the sequences from each sample according to the PCR primers ( 746 primer pairs in total ) and then aligned the reads to the reference genome ( loxAfr1 ) , disregarding sequences below 80% identity . Consensus sequences for each locus and each individual were called with the settings described by Stiller and colleagues [44] , with a minimum of three sequences required in order to call a nucleotide and a maximum of three polymorphic positions allowed per locus ( to filter out false-positive divergent sites due to paralogous sequences that occur in multiple loci in the genome ) . We finally generated multiple sequence alignments for each locus and called divergent sites when at least one allele per species was available . In the first experimental round we were not able to call consensus sequences for more than half of the loci , a problem that we found was correlated with primer pairs that had multiple BLAST matches to loxAfr1 , suggesting alignment to genomic repeats . Primer pairs for subsequent experimental rounds were excluded if in silico PCR ( http://genome . ucsc . edu/cgi-bin/hgPcr ) suggested that they could anneal at too many loci in the savanna elephant genome . Of the 1 , 797 biallelic divergent sites that were identified , we removed 22 to produce Tables 1 and 2 . The justification for removing these sites is that derived alleles were seen in both African and Eurasian elephants , which is unlikely to be observed in the absence of sequencing errors or recurrent mutation . For the MCMCcoal analysis we did not remove these divergent sites , since the method explicitly models recurrent mutation . To obtain standard errors , we omitted each of the 375 loci in turn and recomputed the statistic of interest . To compute a normally distributed standard error , we measured the variability of each statistic of interest over all 375 dropped loci , weighted by the number of divergent sites at the locus that had been dropped in order to take account of the variable amount of data across loci . This can be converted into a standard error using the theory of the Weighted Jackknife as described in [45] . For our relative rate tests , we compute the difference in the number of divergent sites between two taxa since they split , normalized by the total number of divergent sites . The number of standard errors ( computed from a Weighted Jackknife ) by which this differs from zero represents a z score that should be normally distributed under the null hypothesis and thus can be converted into a p value for consistency of the data with equal substitution rates on either lineage . To construct a Neighboring Joining tree relating the proboscideans in Figure S3 , we used MEGA4 [46] with default settings ( 10 , 000 bootstrap replicates ) . To prepare a data set for MCMCcoal , we used input files containing the alignments in PHYLIP format ( Dataset S3 ) [47] , restricting analysis to the loci for which we had diploid data from at least one individual from each of the elephantids we resequenced ( we did not use data from the loxAfr1 draft savanna genome , or from the second Asian elephant we sequenced at only a small fraction of loci ) . The diploid data for each taxon were used to create two sequences from each of the elephantids , allowing us to make inferences about effective population size in each taxon since its divergence from the others . We ran MCMCcoal with the phylogeny ( ( ( ( Forest1 , Forest2 ) , ( Savanna1 , Savanna2 ) ) , ( ( Asian1 , Asian2 ) , ( Mammoth1 , Mammoth2 ) ) ) Mastodon ) . Since MCMCcoal is a Bayesian method , it requires specifying a prior distribution for each parameter; that is , a hypothesis about the range of values that are consistent with previously reported information ( such as the fossil record ) . For the effective population sizes in each taxa ( NF , NS , NA , NM , NFS , NAM , NLox-Eur , and NElephantid-Mastodon ) we used prior distributions that had their 5th percentile point corresponding to the lowest diversity seen in present-day elephants ( savanna ) and their 95th percentile point corresponding to the highest diversity seen in elephantids ( forest ) . For the mastodon-elephantid population divergence time TElephantid-Mastodon we used 24–30 Mya [30] , [35] , [48]–[50] . For the African-Eurasian population divergence time ΤLox-Eur we used 4 . 2–9 Mya [30] , [35] , [51] . For the Asian-mammoth population divergence time ΤAM we used 3 . 0–8 . 5 Mya [30] , [35] , [52] . The taxonomic status of forest and savanna elephants is contentious . To allow us to test the hypotheses of both recent and ancient divergence while being minimally affected by the prior distribution , we use an uninformative prior distribution of TFS = 0 . 5–9 Mya . This prior distribution has substantial density at <1 million years , allowing us to test for recent divergence of forest and savanna elephants . A full justification for the prior distributions is given in Text S5 . MCMCcoal also requires an assumption about the mutation rate , which is poorly measured for the elephantids . We thus ran MCMCcoal under varying assumptions for the mutation rate , to ensure that our key results were stable in the face of uncertainty about this parameter . For each of the three mutation rates that we tested , MCMCcoal was run three times starting from different random number seeds with 4 , 000 burn-in and 100 , 000 follow-on iterations . Estimates of all parameters that were important to our inferences were consistent across runs suggesting stability of the inferences despite starting at different random number seeds ( we did observe instability for the parameters corresponding to mastodon-elephantid divergence , but this was expected because of the high rate of mastodon errors and is not a problem for our analysis as this divergence is not the focus of this study ) . We computed the autocorrelation of each sampled parameter over MCMC iterations to assess the stickiness of the MCMC . Parameters appear to be effectively uncorrelated after a lag of 200 iterations . Given that we ran each chain over 100 , 000 iterations , we expect to have at least 500 independent points from which to sample , which is sufficient to compute 90% credible intervals . The detailed parameter settings and results are presented in Text S4 . | The living elephants are the last survivors of a once highly successful mammalian order , the Proboscidea , which includes extinct species such as the iconic woolly mammoth ( Mammuthus primigenius ) and the American mastodon ( Mammut americanum ) . Despite numerous studies , the phylogenetic relationships of the modern elephants to the woolly mammoth , as well as the taxonomic status of the African elephants of the genus Loxodonta , remain controversial . This is in large part due to the fact that both the woolly mammoth and the American mastodon ( the closest outgroup to elephants and mammoths available for genetic studies ) are extinct , posing considerable technical hurdles for comparative genetic analysis . We have used a combination of modern DNA sequencing and targeted PCR amplification to obtain a large data set for comparing American mastodon , woolly mammoth , Asian elephant , African savanna elephant , and African forest elephant . We unequivocally establish that the Asian elephant is the sister species to the woolly mammoth . A surprising finding from our study is that the divergence of African savanna and forest elephants—which some have argued to be two populations of the same species—is about as ancient as the divergence of Asian elephants and mammoths . Given their ancient divergence , we conclude that African savanna and forest elephants should be classified as two distinct species . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"evolutionary",
"biology",
"evolutionary",
"biology/paleontology",
"evolutionary",
"biology/genomics"
] | 2010 | Genomic DNA Sequences from Mastodon and Woolly Mammoth Reveal Deep Speciation of Forest and Savanna Elephants |
The intrahepatic immune environment is normally biased towards tolerance . Nonetheless , effective antiviral immune responses can be induced against hepatotropic pathogens . To examine the immunological basis of this paradox we studied the ability of hepatocellularly expressed hepatitis B virus ( HBV ) to activate immunologically naïve HBV-specific CD8+ T cell receptor ( TCR ) transgenic T cells after adoptive transfer to HBV transgenic mice . Intrahepatic priming triggered vigorous in situ T cell proliferation but failed to induce interferon gamma production or cytolytic effector function . In contrast , the same T cells differentiated into cytolytic effector T cells in HBV transgenic mice if Programmed Death 1 ( PD-1 ) expression was genetically ablated , suggesting that intrahepatic antigen presentation per se triggers negative regulatory signals that prevent the functional differentiation of naïve CD8+ T cells . Surprisingly , coadministration of an agonistic anti-CD40 antibody ( αCD40 ) inhibited PD-1 induction and restored T cell effector function , thereby inhibiting viral gene expression and causing a necroinflammatory liver disease . Importantly , the depletion of myeloid dendritic cells ( mDCs ) strongly diminished the αCD40 mediated functional differentiation of HBV-specific CD8+ T cells , suggesting that activation of mDCs was responsible for the functional differentiation of HBV-specific CD8+ T cells in αCD40 treated animals . These results demonstrate that antigen-specific , PD-1-mediated CD8+ T cell exhaustion can be rescued by CD40-mediated mDC-activation .
Rapid clonal expansion of CD8+ T cells in response to antigenic challenge is a hallmark of adaptive immunity and a crucial element of host defense . Activation and differentiation of T cells are largely determined by their initial encounter with antigen-presenting cells ( APCs ) , and the resultant responses range from full activation and memory T cell differentiation to clonal exhaustion or deletion , depending on the nature and abundance of inductive signals that T cells decode from APCs during priming [1] , [2] . These events generally occur in secondary lymphoid organs because naïve T cells are usually not primed in nonlymphoid tissues [2] . The liver is , however , an exception to this rule , due to the unique architecture of the hepatic sinusoid which is characterized by a discontinuous endothelium , the absence of a basement membrane , and a very slow flow rate [3]–[5] , allowing circulating T cells to make prolonged direct contact with resident liver cells including hepatocytes [6] . Furthermore , the liver is replete with diverse and unique antigen presenting cell populations , including liver sinusoidal endothelial cells ( LSECs ) [7] , [8] , hepatic stellate cells ( HSCs ) [9] , Kupffer cells [10] , [11] , conventional and plasmacytoid dendritic cells [12]–[14] , all of which are capable of priming and/or tolerizing naïve T cells , at least in vitro . Thus , because of its unique immunological environment , antigens expressed and/or processed in the liver appear to be more accessible to T cells than those in other nonlymphoid organs [4] , [15] . The hepatitis B virus ( HBV ) is a noncytopathic , enveloped , double-stranded DNA virus that causes acute and chronic hepatitis and hepatocellular carcinoma [16] , [17] . Similar to other noncytopathic viruses , the clearance of HBV requires functional virus-specific CD8+ T cell responses [18] . Using the HBV transgenic mouse [19] as a model to study the impact of intrahepatic antigen recognition by HBV-specific CD8+ T cells , we have shown that adoptively transferred HBV-specific memory CD8+ T cells rapidly secrete IFNγ upon antigen recognition in the liver , thereby inhibiting HBV replication [20] . Subsequently , PD-1 is upregulated in the intrahepatic CD8+ T cells and they stop producing IFNγ , start expressing granzyme B ( GrB ) and undergo massive expansion [21] thereby mediating a necroinflammatory liver disease and terminating viral gene expression whereupon the intrahepatic CD8+ T cell population contracts , liver disease abates and IFNγ production returns [21] . While the foregoing studies illustrate the profound impact of intrahepatic antigen recognition on the distribution , expansion and effector functions of memory CD8+ T cells , they do not address the response of immunologically naïve CD8+ T cells to antigen recognition in the liver . Indeed , the literature reveals significant differences between naïve and memory CD8+ T cells in terms of the peptide:MHC complex concentration and costimulation required for activation and the development of their proliferative and cytokine secretion potentials , cytolytic activity and their migratory range [2] , [22] . While T cell priming to viruses that do not infect conventional pAPCs is believed to occur in lymphoid organs via cross-priming [1] , [2] , [23] , [24] , the consequences of naïve T cell priming by hepatocellularly expressed viral antigen are less well understood . In the current study , we used transgenic mice whose CD8+ T cells express T cell receptors ( TCRs ) specific for the HBV nucleocapsid ( COR ) and envelope ( ENV ) proteins to study the early intrahepatic immunological events that are likely to occur during HBV infection . By analyzing the response of naïve COR- and ENV-specific TCR transgenic CD8+ T cells to hepatocellularly presented HBV antigens in vivo after adoptive transfer into HBV transgenic mice whose hepatocytes produce all the HBV gene products and secrete infectious HBV virions [19] , and in vitro after cocultivation with primary HBV transgenic mouse hepatocytes , we show that HBV-specific naïve CD8+ T cells are primed in the liver by HBV+ hepatocytes and proliferate vigorously in situ , but do not differentiate into functional effector T cells unless PD-1 signaling is genetically ablated . Importantly , when the same T cells are transferred into HBV transgenic mice whose myeloid dendritic cells ( mDCs ) were simultaneously activated by agonistic antibodies against CD40 ( αCD40 ) , PD-1 induction is suppressed and the T cells differentiate normally , inhibit HBV antigen expression , and cause liver disease . Collectively , these results indicate that CD40-mediated activation of mDCs can rescue the effector functions of PD-1-inhibited naïve CD8+ T cells , apparently by suppressing the negative regulatory signals that are triggered by antigen recognition in the liver . These results imply that the balance achieved between these two opposing forces may regulate the pathogenesis and outcome of HBV and other hepatotropic virus infections .
A Kb-restricted CD8+ CTL clone ( BC10 ) that recognizes an epitope located between residues 93–100 in the HBV core protein ( MGLKFRQL ) ( COR93 ) was generated from a Balb/c ( H-2d ) by C57BL/6 ( H-2b ) F1 hybrid ( CB6F1 ) mouse that was immunized by standard DNA-prime/vaccinia boost immunization as previously described [21] , [25] . Importantly , when in vitro core peptide-activated BC10 T cells ( 1×107/mouse ) were adoptively transferred into HBV transgenic mice ( lineage 1 . 3 . 32 ) that express all of the HBV antigens and replicate HBV in the liver and kidney [19] , they inhibited HBV replication , and caused liver disease on day 1 after adoptive transfer ( data not shown ) as previously described after adoptive transfer of polyclonal COR93-specific effector memory CD8+ cells [21] . TCRα ( Vα13 . 1JαNEW06 ) and β ( Vβ8 . 1Jβ1 . 2 ) cDNA clones derived from BC10 were inserted into TCR expression cassettes [26] , and injected into fertilized CByB6F2 eggs to generate BC10 TCR transgenic mice . Two founders , BC10 . 1 and BC10 . 3 carrying both TCRα and β transgenes were derived , and lineage BC10 . 3 was chosen for further backcrossing based on its superior allelic exclusion rate ( data not shown ) . The BC10 . 3 TCR transgenic ( TCRtg ) mice were backcrossed more than 10 times onto C57BL/6 ( B6 ) background , and then mated once with CD45 . 1 mice ( H-2b ) so that the TCR transgenic T cells could be easily followed by anti-CD45 . 1 antibody staining . As shown in Figures 1A and 1B , >98% of the splenic CD8+ T cells ( 33 . 5% of total spleen cells ) in these mice were COR93-specific and CD45 . 1 positive as determined by staining with COR93-multimers and CD45 . 1 staining . As expected , they were phenotypically characterized as CD44− , CD62Lhigh , CD25− , CD69− , ( Figures 1C and 1D ) and fewer than 2% of them produced IFNγ or expressed Granzyme B ( GrB ) after 5 hours peptide stimulation in vitro ( Figure 1E ) , indicating that they were in fact naïve T cells . We also generated a lineage of transgenic mice whose CD8+ T cells express TCRs specific for the well-described Ld-restricted ENV28 epitope [27] , [28] . The TCRs of these mice consist of Vα4 . 1JαNEW and Vβ1 . 1Jβ2 . 5 chains cloned from CD8+ ENV28-specific CTL clone 6C2 , whose functional properties have been extensively characterized [20] , [27]–[29] . Lineage 6C2 . 36 was chosen for further characterization and backcrossed onto the Balb/c background for at least 6 generations and then mated once with CD45 . 1 mice ( H-2b ) . As shown in Figure 1F , approximately 83% of splenic CD8+ T cells ( 20% of total spleen cells ) in lineage 6C2 . 36 are ENV28-specific and all of them were CD45 . 1 positive ( Figure 1G ) . Again , virtually all the ENV28-specific CD8+ T cells were CD44− , CD62Lhigh , CD25− , CD69− , ( Figures 1H and 1I ) , and they did not express IFNγ or GrB after peptide stimulation ( Figure 1J ) , indicating that they are naïve T cells . To examine the response of HBV-specific naïve CD8+ T cells to hepatocellularly expressed HBV , we adoptively transferred 3–5×106 COR93-specific naïve CD8+ T cells from the spleen of BC10 . 3 TCR transgenic donor mice into HBV transgenic lineage 1 . 3 . 32 recipient mice [19] , [20] . Groups of 3–4 mice were sacrificed at various time points after adoptive transfer , and their intrahepatic , lymph nodal , and splenic lymphocytes were analyzed for the total number of COR93-specific CD8+ T cells and the extent to which they coexpress Granzyme B ( GrB ) and IFNγ either directly ex vivo or after in vitro stimulation by cognate COR93 peptide . To determine the functional capabilities of the adoptively transferred COR93-specific CD8+ TCR transgenic T cells during a systemic infection in vivo , we also studied their response to cognate HBcAg antigen produced by MHC-matched nontransgenic mice that had been infected 2 hours before adoptive transfer with 2×107 pfu of a recombinant vaccinia virus that expresses the HBV nucleocapsid protein ( cVac ) [30] . As shown in Figure 2 , COR93-specific CD8+ T cells were detectable in the liver of HBV transgenic mice as early as 1 hour after adoptive transfer , and they rapidly accumulated in the liver , constituting more than 17% of total intrahepatic lymphocytes on days 1 . 5 and 3 ( Figure 2A white bars ) and showing greater than a 20-fold increase in their absolute numbers between the 1 hour and 3 day time points ( Figure 2B white bars ) . After rapid expansion , the number of intrahepatic COR93-specific CD8+ T cells remained relatively stable up to day 10 , after which they decreased more than 10-fold by day 14 ( Figure 2B ) but still constituted a large fraction of total intrahepatic CD8+ T cells on day 28 ( Figure 2A ) . In contrast , the COR93-specific CD8+ T cells in the lymph nodes and spleen expanded much less vigorously between the 1 hour and 3 day time points than their counterparts in the liver , although the absolute number of COR93-specific CD8+ T cells was greater in the spleen than the liver at 1 hour and 4 hour time points ( Figure 2B ) . COR93 specific CD8+ T cells started disappearing from the lymph nodes and spleen on day 14 , and became almost undetectable on day 28 ( Figures 2A and 2B; gray and black bars ) . In contrast , in cVac infected nontransgenic C57BL/6 mice , the frequency of COR93-specific CD8+ T cells was similar in the liver , lymph nodes , and spleen ( Figure 2F ) , and fewer COR93-specific CD8+ T cells were detectable in the liver than the spleen ( Figure 2G ) at all time points tested . These results suggest that the COR93-specific CD8+ T cells were primed and accumulated preferentially in the antigen expressing liver of HBV transgenic mice rather than peripherally as in the cVac infected nontransgenic mice . Strikingly , despite vigorous expansion ( Figures 2A and 2B ) , the COR93-specific CD8+ T cells in the liver , lymph nodes and spleen of the HBV transgenic mice did not produce IFNγ either directly ex vivo ( data not shown ) or after 5 hours peptide stimulation ( Figure 2C ) at any time point examined , and their ability to express GrB was severely compromised as well ( Figure 2D ) . In contrast , the intrahepatic , lymph nodal and splenic COR93-specific CD8+ T cells in the cVac infected nontransgenic recipients were able to produce IFNγ in response to 5 hours COR93-peptide stimulation , and expressed GrB directly ex vivo ( Figures 2H and 2I ) . These data suggest that adoptively transferred HBV-specific naïve T cells preferentially expand in the HBV transgenic liver , but the expanding T cells are functionally impaired . Interestingly , virtually all the intrahepatic COR93-specific CD8+ T cells in HBV transgenic mice strongly expressed the co-inhibitory molecule PD-1 on day 1 . 5 ex vivo and remained so until day 28 ( Figure 2E ) , while PD-1 expression was virtually absent in their counterparts in cVac infected nontransgenic animals ( Figure 2J ) , suggesting that PD-1 signaling may have contributed to the functional impairment of intrahepatic COR93-specific CD8+ T cell responses in HBV transgenic mice . To determine if the dysfunctional T cell responses in the HBV transgenic liver reflect active suppression of functional differentiation by PD-1 signaling , the COR93-specific TCR transgene was crossed for two generations onto a MHC class I matched PD-1 deficient background ( kindly provided by Dr . Arlene Sharpe , Harvard Medical School ) [31] , yielding PD-1 deficient COR93-specific TCR transgenic animals . Equal numbers of PD-1 deficient and wild type COR93-specific naïve CD8+ T cells were adoptively transferred into HBV-transgenic mice , and analyzed for expansion , IFNγ producing ability and Granzyme B ( GrB ) expression on day 7 after adoptive transfer . The results were correlated with the degree of liver damage and HBV gene expression monitored by serum alanine aminotransferase ( ALT ) activity and HBV gene Northern Blot ( NB ) analysis , respectively . As shown in Figure 3A , PD-1 deficient COR93-specific CD8+ T cells expanded much more vigorously in the liver than wild type COR93-specific CD8+ T cells , and a larger fraction of PD-1 deficient T cells expressed IFNγ and Granzyme B ( Figure 3B and C ) and they induced a more severe liver disease , monitored as serum alanine aminotransferase ( ALT ) activity ( Figure 3D ) . Furthermore , HBV gene expression was significantly reduced in the recipients of PD-1 deficient COR93-specific CD8+ T cells but not in the wild type T cell recipients ( Figure 3E ) , reflecting the superior cytolytic and interferon gamma-producing activity of the PD-1 deficient cells . Collectively , these results indicate that PD-1 signaling suppresses the expansion and functional differentiation of HBV-specific CD8+ T cells after antigen recognition in the liver . Because the hepatocytes in HBV transgenic mice replicate HBV at high level and release viral particles and subviral antigens into the circulation [19] , HBV derived antigen could be presented to naïve T cells either by the hepatocytes themselves or by professional antigen presenting cells ( pAPCs ) that acquire virus particles and/or subviral antigens in the liver or in peripheral lymphoid organs . Therefore , the expansion of dysfunctional HBV-specific CD8+ T cells in the liver could reflect T cell priming and expansion in the liver , or the intrahepatic accumulation of T cells that were previously primed in the lymph nodes . To distinguish between these alternatives , we monitored the expression of activation markers ( CD69 and CD25 ) on HBV-specific CD8+ T cells in the liver , lymph nodes and spleen at very early time points ( 1 hour , 4 hours and day 1 ) after adoptive transfer into HBV transgenic mice , and the results were compared with the expression of these activation markers on the HBV-specific CD8+ T cells in the cVac infected nontransgenic animals . As shown in Figure 4A ( white bars ) , within 1 hour after adoptive transfer , approximately 85 . 0% of the intrahepatic COR93-specific CD8+ T cells in the HBV transgenic mice expressed the very early activation marker CD69 , suggesting that nearly all the COR93-specific T cells that entered the liver rapidly recognized antigen . By 4 hours , virtually all the intrahepatic COR93-specific CD8+ T cells in the transgenic mice were CD69 positive , and a large fraction of them also began to express CD25 ( Figure 4B , white bars ) , the IL-2α receptor that is required for high affinity binding of IL-2 [32] , suggesting that they were fully activated and prepared to proliferate . In contrast , CD69 expression by COR93-specific CD8+ T cells in the lymph nodes ( gray bar ) and spleen ( black bars ) occurred later ( Figure 4A ) than their intrahepatic counterparts ( Figure 4A ) , and fewer nodal and splenic COR93-specific CD8+ T cells expressed CD25 ( Figure 4B , gray and black bars ) , suggesting that naïve HBV-specific CD8+ T cell activation primarily occurred in the HBV-expressing liver and that these intrahepatically primed T cells subsequently trafficked to the lymph nodes and spleen . In contrast , COR93-specific CD8+ T cells in cVac infected nontransgenic recipients rapidly upregulated CD69 in the spleen and the liver as early as 1 hour after adoptive transfer ( Figure 4C ) . Interestingly , CD25 expression in cVac infected nontransgenic mice was mainly observed on the splenic COR93-specific CD8+ T cells ( Figure 4D ) , suggesting that the activation of COR93-specific CD8+ T cells during systemic vaccinia infection is largely splenic . None of these changes occurred in uninfected control nontransgenic recipients ( data not shown ) , indicating that they were antigen specific events . Collectively , these results suggest that hepatocellularly expressed HBV antigen primes naïve T cells in the liver . Next , groups of 4 HBV-transgenic mice received intraperitoneal injections of either saline or anti-CD62L antibodies ( αCD62L ) , that are known to block naïve T cell homing to the lymph nodes [33]–[35] , followed by naïve COR93-specific CD8+ T cells 16 hours later . The mice were sacrificed 1 hour after adoptive transfer and COR93-specific CD8+ T cells were isolated from the liver , lymph nodes , and spleen and analyzed for CD69 expression . As shown in Figure 5A , αCD62L administration completely abrogated the homing of COR93-specific CD8+ T cells to lymph nodes , but had no impact on the intrahepatic accumulation of the T cells . Despite the absence of T cell homing to the lymph nodes , COR93-specific CD8+ T cells in the αCD62L treated HBV-transgenic mice were fully activated in the liver , similar to those in saline treated recipients ( Figure 5B ) . These results confirm that intrahepatic T cell activation and expansion do not reflect redistribution of T cells that were activated in the lymph nodes . Intrahepatic priming of HBV-specific CD8+ T cells could reflect recognition of either endogenously synthesized hepatocellular antigen or of antigen that is released by the hepatocytes and internalized , processed and presented by liver sinusoidal endothelial cells ( LSEC ) , Kupffer cells , or dendritic cells that are capable of cross-presentation [8] , [10] , [12] , [36] . In an attempt to identify the antigen presenting cell population responsible for priming COR93-specific CD8+ T cells in the liver of HBV transgenic mice , we adoptively transferred COR93-specific naïve T cells into MHC-matched HBV transgenic mice lineages 1 . 3 . 32 and MUP-core 50 ( MC50 ) that produce a nonsecretable form of HBcAg , and compared T cell accumulation and activation 1 hour later . Lineage 1 . 3 . 32 replicates HBV and expresses HBcAg ( which is nonsecretable ) in their hepatocytes and it also secretes viral particles and HBeAg , a soluble viral protein that is highly cross-reactive with HBcAg [16] , [19] . In contrast , lineage MC50 express only HBcAg whose expression is restricted to hepatocytes [37] , reducing the likelihood of antigen presentation by professional antigen presenting cells that acquire secreted viral particles or subviral antigens . As shown in Figure 6 , COR93-specific CD8+ T cells accumulated similarly in liver , lymph nodes and spleen in both HBV-transgenic mouse lineages ( Figure 6A ) , and the fraction of CD69 positive COR93-specific CD8+ T cells in the liver and lymph nodes were comparable in these lineages ( Figure 6B ) . Since HBV core expression in MC50 transgenic mice is restricted to hepatocytes , these results suggest that naïve COR93-specific CD8+ T cells were primed by recognition of endogenously synthesized hepatocellular HBcAg . To test this notion , COR93-specific naïve T cells were co-cultured overnight with hepatocytes , LSECs , Kupffer cells , and dendritic cells that were isolated from the liver of HBV transgenic mice lineage 1 . 3 . 32 and nontransgenic controls , and then examined for CD69 expression . As shown in Figure 7 , approximately 25% of COR93-specific naïve T cells upregulated CD69 when they were cocultured with hepatocytes isolated from HBV transgenic mice ( Figure 7A and 7E ) , whereas fewer than 5% ( 2 . 5±1 . 8% ) did so when cocultured with transgenic LSEC , ( Figure 7B and 7E ) and virtually no T cells expressed CD69 when cocultured with transgenic DCs or KCs ( Figure 7C , 7D and 7E ) despite the ability of intrahepatic LSECs , DCs and KCs to activate the T cells if the APCs are pulsed with COR93-peptide prior to coculture ( Figure 7F–J ) . As expected , neither hepatocytes nor LSECs isolated from nontransgenic mice stimulated COR93-specific CD8+ T cells to express CD69 unless they were first pulsed with COR93-peptide ( Figures 7E and 7J , white bars ) indicating the antigen specificity of the T cell response to the HBV transgenic hepatocytes . Collectively , these data suggest that intrahepatic priming of COR93-specific CD8+ T cells was primarily mediated by endogenously synthesized antigen produced by the HBV-transgenic hepatocytes . To determine if intrahepatic priming of functionally defective CD8+ T cells is a general rule or restricted to COR93-specific TCR transgenic T cells , we adoptively transferred naive ENV28-specific T cells from CD45 . 1-6C2 . 36 TCRtg mice into MHC-matched HBV transgenic mice and nontransgenic littermates . Groups of 3 mice were sacrificed 4 hours , 3 days and 7 days after adoptive transfer to examine the ENV28-specific CD8+ T cell response in the liver ( Figure 8; white bars ) , lymph nodes ( Figure 8; gray bars ) and spleen ( Figure 8 , black bars ) . ENV28-specific naïve CD8+ T cells were rapidly activated in the liver of the HBV transgenic mouse recipients ( but not in nontransgenic recipients – not shown ) as early as 4 hours after adoptive transfer ( Figure 8A ) , suggesting that , like COR93-specific naïve CD8+ T cells , adoptively transferred ENV28-specific CD8+ naïve T cells are primed in the liver . The intrahepatically primed ENV28-specific CD8+ T cells expanded in the liver ( Figure 8B ) , but did not express IFNγ or GrB ( Figure 8C and 8D ) . These results recapitulate the immunological events observed after adoptive transfer of COR93-specific naïve T cells into HBV transgenic mice illustrated in Figures 2 and 4 , indicating that intrahepatic T cell priming and the expansion of functionally defective T cells occur irrespective of the antigen specificity or MHC restriction of the T cells . Thus , T cell hyporesponsiveness represents a general outcome induced by intrahepatic T cell priming to endogenously synthesized hepatocellular antigen . Ample evidence suggests that the induction of functional CD8+ T cell responses requires the activation of professional antigen presenting cells ( pAPCs ) , which in turn provide secondary signals to naïve T cells [1] , [2] , [38] . Since the results shown in Figure 7 indicate that HBV-specific naïve T cells were primed by hepatocytes that are not known to express co-stimulatory molecules [39] , it is possible that the dysfunctional HBV-specific T cell responses in the HBV transgenic liver reflected the absence of a second signal . To determine if the differentiation defect of intrahepatically primed HBV-specific CD8+ T cells can be rescued by products of the immune response to an exogenous pathogen , COR93-specific naïve T cells were adoptively transferred into HBV transgenic mice that were either treated with saline ( NaCl ) or infected with 2×107 of cVac 2 hours before transfer , and the results were compared with their differentiation after transfer into nontransgenic recipients that had been infected with 2×107 of cVac 2 hours before transfer . Three and seven days later , mice were sacrificed , and intrahepatic COR93-specific CD8+ T cells were analyzed for expansion , IFNγ producing ability and Granzyme B ( GrB ) expression . The results were correlated with the degree of liver damage and HBV gene expression monitored by serum alanine aminotransferase ( ALT ) activity and Northern Blot ( NB ) analysis , respectively . To monitor the impact of cVac infection per se on liver disease and HBV gene expression , HBV transgenic mice were infected with 2×107 of cVac without receiving COR-93-specific naïve CD8+ T cells , and they were sacrificed 3 and 7 days later . As expected , COR93-specific naïve T cells expanded vigorously in the HBV transgenic mouse liver but did not express IFNγ or Granzyme B ( Figures 9A–9C; white bars ) . In contrast , cVac infection of HBV transgenic mice triggered IFNγ ( Figure 9B; black bar ) and Granzyme B ( Figure 9C; black bar ) expression by a small but significant fraction of the transferred intrahepatic COR93-specific CD8+ T cells without significantly increasing their expansion in the liver ( Figure 9A; black bars ) . Note , however , that the frequency of IFNγ+ and Granzyme B+ CD8+ T cells was lower in cVac infected HBV transgenic mice ( Figures 9B and 9C , black bars ) than in cVac infected nontransgenic recipients ( Figures 9B and 9C , blue bars ) , suggesting that their effector functions were suppressed by continuous hepatocellular antigen recognition , similar to the response we have shown to occur when HBV-specific memory CD8+ T cells recognize antigen in the HBV transgenic mouse liver [21] . The COR93-specific CD8+ T cells induced only a modest elevation of serum ALT activity in saline injected HBV transgenic mice ( Figure 9D ) , and they had little or no effect on HBV gene expression ( Figure 9E ) in the liver . In contrast , the CD8+ T cell-mediated liver disease was more severe ( Figure 9D ) and intrahepatic HBV gene expression was strongly suppressed in cVac infected HBV transgenic mice compared to saline treated HBV transgenic controls ( Figure 9E ) . As expected , cVac infection per se did not induce liver disease ( Figure 9D , ; red bar ) , nor did it suppress HBV gene expression ( Figure 9E ) , suggesting that the induction of liver disease and the suppression of HBV gene expression in cVac infected HBV transgenic mice after COR93-specific CD8+ T cell adoptive transfer were mediated by the T cells . These results suggest that functional differentiation of HBV-specific CD8+ T cells in the HBV transgenic mouse liver is sufficiently restored in the context of a systemic virus infection to both cause hepatitis and inhibit viral gene expression . Activation of professional antigen presenting ( pAPC ) cells is believed to be essential for the induction of functional CD8+ T cell responses after virus infections , and several studies have demonstrated that ligation of CD40 induces pAPC activation , resulting in the induction of CD8+ T cell responses [40]–[42] . To examine whether CD40 activation could induce functional differentiation of HBV-specific CD8+ T cells in this model , we adoptively transferred naïve COR93-specfic CD8+ T cells into HBV transgenic mice and nontransgenic controls that had been injected intravenously either with 100 µg/mouse of an agonistic anti-CD40 antibody ( αCD40 ) [43] , [44] or with saline ( NaCl ) 16 hours before transfer . Seven days later , the mice were sacrificed , and intrahepatic COR93-specific CD8+ T cells were analyzed for expansion , IFNγ producing ability and Granzyme B ( GrB ) expression . The results were correlated with the degree of liver damage and HBV gene expression monitored by serum ALT activity and Northern Blot analysis , respectively . As shown in Figure 10A , by day 7 , αCD40 treatment increased the intrahepatic expansion of COR93-specific CD8+ T cells in HBV transgenic mice by 5 fold compared to the saline treated transgenic animals . Furthermore , by day 7 , approximately 40% of intrahepatic COR93-specific CD8+ T cells in the αCD40 treated animals produced IFNγ in response to 5 hours in vitro peptide stimulation ( Figure 10B ) , and almost all the COR93-specific CD8+ T cells expressed GrB directly ex vivo ( Figure 10C ) , contrasting strikingly to their counterparts in the saline treated animals . Interestingly the induction of T cell effector functions coincided with PD-1 downregulation in intrahepatic COR93-specific CD8+ T cells ( Figure 10D ) , suggesting that activation of CD40 signaling counteracted the PD-1 mediated negative signaling . In contrast to these observations , neither the expansion nor the functional differentiation of the COR93-specific CD8+ T cells were enhanced in αCD40 treated nontransgenic recipients ( not shown ) . Importantly , αCD40 treated HBV transgenic recipients displayed higher serum ALT activity ( Figure 10E ) and very strong suppression of intrahepatic HBV gene expression ( Figure 10F ) , after adoptive transfer of naïve COR93-specific CD8+ T cells compared to saline treated animals . The induction of severe liver disease and the suppression of HBV gene expression by αCD40 treatment reflect the vigorous expansion and functional differentiation of adoptively transferred COR93-specific CD8+ T cells , since these changes were not observed in αCD40 treated transgenic recipients that did not receive naïve COR93-specific CD8+ T cells ( Figure 10E; gray bars , and Figure 10F ) . These results suggest that CD40 activation during intrahepatic T cell priming converts T cell hyporesponsiveness into immunity . We then attempted to determine the role of professional antigen presenting cells ( pAPCs ) in αCD40 induced functional differentiation of HBV-specific CD8+ T cells . To do so , HBV-transgenic mice were crossed with CD11c . DOG mice that express the human diphtheria toxin ( DTX ) receptor on CD11c+ cells and thus allow depletion of dendritic cells after DTX administration with no signs of toxicity [45] . Groups of three CD11c . DOG-HBV transgenic mice were treated with DTX or saline ( NaCl ) every other day in combination with single administration of clodronate liposome ( CLL ) that is known to induce apoptosis of macrophages and DCs in vivo and in vitro [46] , [47] , or control liposomes ( NaCl-L ) , yielding 4 different groups of mice ( i . e . NaCl+NaCl-L , DTX+NaCl-L , CLL+NaCl , and DTX+CLL . ) On day 2 after CLL or NaCl-L treatment , we analyzed the numbers of myeloid dendritic cells ( mDCs; F480+CD11c+ ) , lymphoid dendritic cells ( lymDCs: F480−CD11c+ ) , Kupffer cells ( F480+CD11c− ) and B cells ( B220+ ) in the liver to determine the efficacy of pAPCs depletion . As shown in Figures 11A and 11B , DTX and CLL independently depleted mDCs in the liver and their effects were additive . ( Figure 11A ) , while intrahepatic lymDCs were depleted only by DTX treatment ( Figure 11B ) . Surprisingly , the number of Kupffer cells paradoxically increased when mice were treated with DTX or CLL alone and with both together ( Figure 11C ) . This might reflect that dendritic cell death stimulated proliferation and/or migration of Kupffer cells . None of these treatments significantly reduced the number of intrahepatic B cells ( Figure 11D ) . To examine the impact of pAPC-depletion on αCD40 induced functional differentiation of HBV-specific CD8+ T cells , CD11c . DOG-HBV transgenic mice that were pre-treated with DTX , CLL or DTX plus CLL , were injected with αCD40 , and 1 day later , adoptively transferred with COR93-specific naïve T cells . The mice were sacrificed on day 7 after adoptive transfer , and the intrahepatic COR93-specific CD8+ T cells were analyzed for expansion , IFNγ producing ability and Granzyme B ( GrB ) expression . The T cell responses were correlated with the degree of liver damage monitored by serum ALT activity . As shown in Figures 11E to 11G , expansion , IFNγ producing ability , and GrB expression of COR93-specific CD8+ T cells in αCD40 treated CD11c . DOG transgenic mice were directly correlated with the number of intrahepatic mDCs at the time of αCD40 administration , but not those of intrahepatic lymDCs , Kupffer cells , or B cells , suggesting that mDCs are required for αCD40 induced functional differentiation of HBV-specific CD8+ T cells . Taken together , these results suggest that activation of mDCs through the CD40 pathway can overcome PD-1-mediated suppression and induce functional CD8+ T cell responses in response to intrahepatically expressed HBV .
The current study examines the impact of hepatocellular antigen presentation on the expansion and functional differentiation of antigen-specific CD8+ T cells . Our results revealed that intrahepatic antigen presentation primes functionally defective antigen-specific CD8+ T cell responses , that the dysfunctional CD8+ T cell response reflects active suppression of expansion and functional differentiation by PD-1 signaling , and that such suppression can be overridden by activating myeloid dendritic cells ( mDCs ) through CD40 stimulation . COR93-specific naïve CD8+ T cells were rapidly activated in the liver after adoptive transfer into HBV transgenic mice as indicated by more rapid expression of activation markers CD69 and CD25 by intrahepatic COR93-specific CD8+ T cells than their lymph nodal and splenic counterparts ( Figures 4A and 4B ) . Hepatocellular activation of the COR93-specific CD8+ T cells was not due to hepatic migration of cells that had been activated in lymphoid organs , because the naïve T cells were equally activated when T cell homing to lymph nodes was prevented by anti-CD62L antibody ( αCD62L ) treatment ( Figure 5 ) and when HBcAg secretion was precluded ( Figure 6 ) . Rather , it reflected T cell priming in the liver by recognition of endogenously synthesized hepatocellular antigen ( Figure 7 ) . As a consequence of hepatocellular antigen presentation , the COR93-specific CD8+ T cells upregulated CD69 and CD25 expression ( Figure 4A and B ) and expanded vigorously ( Figures 2A and 2B ) but they were functionally impaired as they did not express IFNγ or Granzyme B ( GrB ) either directly ex vivo or after 5 hours in vitro peptide stimulation ( Figures 2C and 2D ) . Importantly , hepatocellular T cell priming and the expansion of functionally defective T cells also occurred when ENV28-specific naïve T cells were transferred into HBV-transgenic mice ( Figure 8 ) , indicating that hepatocellular T cell priming induces functionally defective T cells responses irrespective of antigen specificity and MHC restriction and illustrating the generality of these observations . The dysfunctional HBV-specific T cell responses likely reflected the suppression of functional differentiation by PD-1 signaling ( Figure 3 ) , but such suppression could be overcome by vaccinia virus infection ( Figure 9 ) or simultaneous activation of mDCs via the CD40 signaling pathway ( Figures 10 and 11 ) . While various hepatic cell populations have been shown to contribute to T cell priming in the liver [8] , [10] , [12] , [36] , [39] , our data suggest that HBV antigen-positive hepatocytes are responsible for priming of HBV-specific CD8+ T cells in our system . The fenestrated liver sinusoidal endothelium permits circulating T cells to make direct contact with underlying hepatocytes [6] . Indeed , when hepatitis B envelope ( ENV ) specific effector CD8+ T cells were adoptively transferred into hepatitis B virus ( HBV ) transgenic mice that express the ENV protein in their hepatocytes , renal tubular epithelium , and choroid plexus cells [48] , the ENV-specific effector CD8+ T cells were selectively sequestered and specifically activated in the liver where they caused a necroinflammatory disease [4] , [29] but not in the other tissues . Importantly , whereas they failed to recognize antigen in the kidney or the CNS when injected intravenously , the CTLs were highly cytopathic for ENV-positive renal tubules and choroid plexus epithelial cells when they were injected directly into those tissues [4] . Furthermore , a series of studies by Bertolino and colleagues suggest that hepatocytes can prime alloantigen specific naïve T cells [15] , [35] , [39] , suggesting that antigen expressed by hepatocytes is highly accessible to circulating native T cells . In those studies , however , the intrahepatic priming of alloantigen specific T cells by hepatocytes was shown to induce rapid T cell deletion [15] , [35] , [39] , contrasting strikingly to the vigorous expansion of HBV-specific CD8+ T cell responses described in this study . The basis for the difference is unclear , but it could reflect the different TCR affinity of transgenic T cells and the level of cognate antigen expression in the liver . Nonetheless , in Bertolino's hands and ours , intrahepatic antigen recognition fails to trigger CD8+ T cell functional differentiation . Strikingly , the intrahepatically activated HBV-specific CD8+ T cells in the HBV transgenic mice did not secrete IFNγ or display cytotoxic activity ( Figures 2C and 2D ) . Consequently , they did not inhibit HBV replication ( data not shown ) or gene expression ( Figure 3E ) or cause a necroinflammatory liver disease ( Figure 3D ) . The lack of effector functions was not due to intrinsic defects of HBV-specific TCR transgenic CD8+ T cells or the large number of adoptively transferred T cells , since the same transgenic CD8+ T cells differentiated into fully functional effector T cells in nontransgenic recipients that were infected with recombinant vaccinia viruses expressing the HBV core antigen ( Figures 2F–2I ) . Instead , our data suggest that the dysfunctional intrahepatic CD8+ T cell response reflects the impact of antigen-induced , PD-1-mediated negative signaling . The intrahepatic HBV-specific CD8+ T cells upregulated PD-1 ( Figure 2E ) in HBV transgenic mice but not in nontransgenic mice infected with cVac ( Figure 2J ) . Furthermore , PD-1 deficient COR93-specific naïve T cells expanded more vigorously than their PD-1 positive wild-type counterparts , and they differentiated into cytotoxic effector T cells in situ , caused severe liver damage , and inhibited HBV gene expression in the liver ( Figure 3 ) . These results suggest that HBV-specific CD8+ T cells can be primed in the liver by recognition of antigen expressed by hepatocytes but activation of their effector functions is suppressed by PD-1 signaling , consistent with the previous studies reported by us [21] , [49] and others [50] , [51] . Whether other negative signaling molecules such as CTLA-4 [52] , Tim3 [53]–[55] , 2B4 [56] , [57] , IL-10 [58] , [59] and TGFβ [60] , [61] that are known to suppress antiviral CD8+ T cell responses also contributed to the suppression of functional differentiation after intrahepatic priming remains to be determined . In chronic HBV patients , the inhibitory molecule 2B4 and Tim-3 are highly co-expressed with PD-1 on HBV-specific CD8+ T cells [62] , [63] . Similarly , CTLA-4 is highly expressed on HBV-specific CD8+ T cells that express high levels of pro-apoptotic molecule Bim [64] . Furthermore , in vitro blockade of CTLA-4 or Tim-3 signaling appears to restore effector functions of HBV-specific CD8+ T cells after in vitro peptide stimulation , and this effect was even enhanced when combined with PD-1 blockade , suggesting that HBV-specific CD8+ T cell responses in chronically infected patients are suppressed by several non-redundant mechanisms [62] , [64] . Importantly , our data suggest that such suppressive mechanism ( s ) can be overcome by myeloid dendritic cell ( mDC ) activation . HBV-specific CD8+ T cells differentiated into fully functional effector T cells in recipient HBV transgenic mice that were treated with agonistic anti-CD40 antibodies ( αCD40 ) , resulting in liver disease and the inhibition of HBV gene expression ( Figure 10 ) . Importantly , mDCs are required for αCD40 induced functional differentiation of HBV-specific CD8+ T cells ( Figure 11 ) . Collectively , these results suggest that activation of mDCs via CD40 signaling was essential to rescue HBV-specific CD8+ T cells from functional suppression through PD-1 and perhaps other regulatory molecules . Several studies with αCD40 established a model postulating that the αCD40 activates pAPCs that then provide secondary signals to naïve CD8+ T cells upon antigen presentation [38] , [40] , [42] . According to this model , HBV-specific CD8+ T cells were programmed to differentiate into functional effector T cells during cross-priming by αCD40-activated pAPCs that acquired circulating HBV particles , subviral antigens or HBV expressing hepatocytes or hepatocyte fragments . However , our preliminary data suggest that hepatic DCs isolated from αCD40 treated HBV-transgenic mice cannot stimulate COR93-specific native T cells to express CD69 ( data not shown ) , suggesting inefficient cross-priming by αCD40-activated pAPCs . Therefore , it is possible that αCD40-activated pAPCs released cytokines such as IL-12 and type I interferons that provide a third signal required for T cell functional differentiation [1] , [65] , [66] . In line with this notion , Maini and colleagues have recently showed that IL-12 potently augments the capacity of HBV-specific CD8+ T cells to produce IFNγ upon in vitro stimulation by cognate antigen in association with down-modulation of PD-1 [67] . Additional studies are required to test these hypotheses . It remains to be determined if similar events occur during natural HBV infection . Ample evidence suggests that HBV-specific CD8+ T cell responses are functionally impaired during chronic HBV infections [51] , [68] , [69] and the functional impairment is associated with PD-1 expression [68] , [70] by HBV-specific CD8+ T cells , similar to the transgenic T cells described in this study . Therefore , the expansion of functionally defective CD8+ T cells by hepatocellular priming may explain the weak CD8+ T cell responses observed during chronic HBV infections . In contrast , approximately 95% of adult onset acute HBV infection is characterized by a vigorous HBV-specific CD8+ T cell response . While our data indicate that activation of mDCs through the CD40 pathway can induce functional HBV-specific CD8+ T cell responses , it is currently unknown whether , and if so , how the mDCs are activated during natural HBV infection . While CD40 ligand ( CD40L ) is expressed on a variety of cells including platelets , mast cells , basophils , NK cells , and B cells , antigen-specific CD4+ T cells are primarily responsible for activating CD40 expressing cells , particularly professional antigen presenting cells ( pAPCs ) , and CD4 T cell mediated CD40 activation appears essential for cross-priming functional CD8+ T cell responses [40]–[42] . Indeed , early priming of HBV-specific CD4+ T cells before or during viral spread in HBV-infected chimpanzees appears to be necessary to initiate a functionally efficient CD8+ T cell response , and the depletion of CD4+ T cells before HBV infection precluded functional T cell priming and caused persistent infection in experimentally infected chimpanzees [70] . Experiments are currently underway to determine whether fully functional CD8+ T cell responses can be induced in this transgenic mouse model by providing HBV-specific T cell help . In summary , the data described herein demonstrate that endogenously synthesized hepatocellular antigen primes functionally defective HBV-specific CD8+ T cells via an instructional process involving PD-1 signaling that actively suppresses expansion and functional differentiation of hepatocellularly primed T cells . Importantly , such suppressive mechanisms can be overcome by activating mDCs through the CD40 pathway . Collectively , these results suggest that HBV specific CD8+ T cell responses are regulated by the balance between PD-1 mediated inhibitory signaling and stimulatory signals by activated DCs . More experiments are required to determine whether DC activation and/or PD-1 blockade may , individually or together , have therapeutic potential to terminate chronic viral infections of the liver and possibly other persistent viral infections as well .
All experiments involving mice were performed in the AAALAC accredited vivarium ( Vertebrate Animal Assurance No . A3194-01 ) at The Scripps Research Institute . All animal studies follow the guidelines in the NIH Guide for the Care and Use of Laboratory Animals and are approved by The Scripps Research Institute Animal Care and Use Committee ( Protocol # 08-0159 ) . HBV transgenic mouse lineage 1 . 3 . 32 ( inbred C57BL/6 , H-2b ) and lineage MC50 have been previously described [19] , [37] . Lineage 1 . 3 . 32 expresses all of the HBV antigens and replicates HBV in the liver and kidney at high levels without any evidence of cytopathology . Lineage MUP-core 50 ( MC50 ) ( inbred C57BL/6 , H-2b ) expresses the HBV core protein in hepatocyte under the transcriptional control of the mouse major urinary protein ( MUP ) promoter . In all experiments , the mice were matched for age ( 8 weeks ) , sex ( male ) , and ( for the 1 . 3 . 32 animals ) serum HBeAg levels in their serum before experimental manipulations . PD-1 deficient mice and CD11c . DOG mice ( both inbred C57BL/6 , H-2b ) , kindly provided by Drs . Arlene Sharpe and Günter Hämmerling , respectively , have been previously described [31] , [45] . All experiments were approved by The Scripps Research Institute Animal Care and Use Committee . Synthetic peptides corresponding to the previously described [27] , [28] , [71] HBV envelope ( ENV ) -specific CTL epitope , ENV28 ( Ld; IPQSLDSWWTSL ) and HBV nucleocapsid protein ( COR ) -specific CTL epitope , COR93 ( Kb; MGLKFRQL ) were purchased from Mimotope ( Victoria , Australia ) . Recombinant vaccinia viruses that express the nucleocapsid protein ( core ) ( subtype ayw ) of HBV ( designated cVac ) and the major envelope protein ( S ) ( subtype adw2 ) ( HBs4 ) were kindly provided by H . J . Schlicht [30] and B . Moss [72] , respectively . A CD8+ CTL clone termed BC10 , that is Kb-restricted and specific for an epitope located between residues 93–100 in the HBV core protein ( MGLKFRQL ) ( COR93 ) , was generated in Balb/c ( H-2d ) by C57BL/6 ( H-2b ) F1 hybrids ( CB6F1 ) that were immunized by standard DNA-prime/vaccinia boost immunization to induce an HBcAg specific CD8+ T cells response as previously described [21] , [25] . Fourteen days after the booster immunization , mice were sacrificed and spleen cells were collected . 4×106 spleen cells were cocultured with 1×105 of irradiated RBL5 cell transfectants that express the HBV core protein ( RBL5/c ) in complete RPMI 1640 medium ( GIBCO , Frederick , Md . ) containing streptomycin ( 100 µg/ml ) , penicillin ( 100 U/ml ) , 2-mercaptoethanol ( 5×10−5 M ) , 10% fetal calf serum , and 2 . 5% EL-4 supernatant in 24-well plates ( Costar , Cambridge , Mass . ) . The RBL5/c cell line was a gift from Dr . Jorg Reiman [71] . Seven days later , the spleen cells were restimulated with RBL5/c , and on day 14 , they were cloned in 96-well round bottom plates ( Costar ) at 1 cell/well . After 2 to 3 weeks of repetitive stimulation , wells containing growing cells were expanded and four H-2b restricted CTL clones were established . A CTL clone , termed BC10 , was chosen for further characterization based on its superior cytolytic activity against RBL5/c . The fine antigen specificity of BC10 was confirmed by intracellular IFNγ production in response to a dominant , Kb restricted COR93-CTL epitope [71] . TCR-α and -β cDNAs were synthesized from 20 ng of total messenger RNA extracted from COR93-specific CD8+ T cell clone BC10 and ENV28-specific CD8+ T cell clone 6C2 [27] , [29] , and amplified by PCR as described by Yoshida , et al [73] . The PCR products were cloned into the pGEM-T Easy Vector ( Promega . Madison , WI ) and then sequenced . The sequence analyses revealed that CTL clone BC10 expressed a TCR composed of Vα13 . 1JαNEW06 and Vβ8 . 1Jβ1 . 2 , while the CTL clone 6C2 expressed a TCR composed of Vα4 . 1JαNEW and Vβ1 . 1Jβ2 . 5 chains . Flanking primers were designed to amplify the rearranged Vα13 . 1JαNEW06 and Vβ8 . 1Jβ1 . 2 genomic DNA based on genomic sequence from these TCR loci . PCR products were sequenced again and then inserted into the TCR expression cassettes pTα and pTβ [26] , kindly provided by Dr . Diane Mathis . Prokaryotic DNA sequences were removed from both vectors and injected into fertilized CByB6F2 eggs as previously described [19] . Founders were screened by PCR and analyzed for the specific TCR expression on CD8+ T cells in the peripheral blood by FACS analysis . A founder , BC10 . 3 , expressed TCR specific for COR93 epitope and was bred against C57BL/6 mice ( H-2b ) for 6 generation and then against CD45 . 1 mice ( C57BL/6 background; H-2b ) for at least 6 more generations . A founder , 6C2 . 36 , expressed TCRs specific for ENV28 epitope and was bred against Balb/c mice ( H-2d ) for more than 6 generations before being mated with CD45 . 1 mice to produce H-2bxd F1 hybrids . Spleen cells were isolated from TCR transgenic mice BC10 . 3 ( CD45 . 1+;H-2b ) or 6C2 . 36xCD45 . 1 F1 hybrids ( CD45 . 1+; H-2bxd ) as previously described . Spleen cells from BC10 . 3 were transferred into either lineage 1 . 3 . 32 mice that were homozygous for HBV , or into MC50 mice heterozygous for the HBV core antigen , or into nontransgenic C57BL/6 mice ( H-2b ) . In selected experiments , the mice were either intravenously infected with 2×107 of recombinant vaccinia viruses expressing the HBV core antigen ( cVac ) or received saline as a control . Spleen cells from 6C2 . 36xCD45 . 1 F1 hybrids were transferred into HBV transgenic mice lineage 1 . 3 . 32× Balb/c F1 hybrids ( H-2bxd ) and syngeneic nontransgenic recipients . Groups of 3–4 mice were sacrificed at various time points after adoptive transfer and their livers , lymph nodes , and spleen were harvested for further analysis . The FGK45 hybridoma producing rat IgG2a mAb against mouse CD40 ( αCD40 ) was provided by Dr . A . Rolink ( Basel Institute for Immunology , Basel , Switzerland ) [43] . CD40 was purified from FGK45 culture supernatants as previously described [44] . Mice were intravenously injected with 100 µg of CD40 16 hours before adoptive transfer of HBV-specific naïve T cells . The mice were sacrificed at different time points after injection , and their livers were harvested for further analysis ( see below ) . A monoclonal anti-CD62L antibody ( clone Mel-14 ) was purchased from BD Bioscience , and mice were intraperitoneally administered with 100 µg of anti-CD62L mAb ( αCD62L; clone Mel14 ) in 200 µl of PBS at 16 , and 4 hours before adoptive transfer . If necessary , further doses of antibodies were administered on day 2 after adoptive transfer . Diphtheria Toxin ( DTX ) was purchased from Sigma-Aldrich , dissolved in PBS , and intraperitoneally administered ( 200 ng/mouse ) every other day . Clodronate Liposome and control Liposome were both purchased from Encapusula NanoSciences , and intravenously injected once ( 200 µl/mouse ) . Spleen cells , lymph node cells , and intrahepatic lymphocytes ( IHL ) were prepared as previously described [21] , [74] . Briefly , spleen cells and lymph node cells were isolated by pressing through a 70 µm cell strainer ( Becton Dickinson ) with the plunger of a 1-ml syringe and were washed three times with PBS and used for further analysis . For IHL isolation , livers were perfused with 10 ml of PBS via the portal vein to remove circulating lymphocytes and the liver cell suspension was pressed through a 70 µm cell strainer and digested with 10 ml of RPMI 1640 medium ( Life Technologies ) , containing 0 . 02% ( w/v ) collagenase IV ( Sigma ) and 0 . 002% ( w/v ) DNase I ( Sigma ) , for 40 minutes at 37°C . Cells were washed with RPMI 1640 and then overlaid on Percoll/Histopaque solution consisting of 12% Percoll ( Pharmacia ) and 88% Histopaque-1083 ( Sigma-Aldrich ) . After centrifugation for 20 min at 1500× g , the IHLs were isolated at the interface . The lymphmononuclear cells were washed twice with RPMI 1640 medium and used for further analysis The livers of HBV transgenic mice lineage 1 . 3 . 32 were perfused slowly via the inferior vena cava with 25 ml of warm Liver Perfusion Medium ( Gibco-Life Technologies ) at a rate of 5 ml/minute , and then digested with 50–75 ml of warm Liver Digest Medium ( Gibco-Life Technologies ) at a rate of 5 ml/min . Following complete digestion of the liver ( 10–15 min . ) , the gall bladder was removed and the liver carefully excised . Cells were collected from the liver by disrupting the liver capsule and swirling the tissue in a petri dish containing Liver Digest Medium . Liver nonparenchymal cells ( LNPCs ) containing LSECs were separated from hepatocytes by centrifuging the cell suspension at 100× g for 2 min at room temperature . For LSEC isolation , the supernatant containing LNPCs was washed twice with RPMI 1640 ( Cellgro ) , and the cell pellet was resupended with BD IMag Buffer at a concentration of 1×107/ml . The cell suspension was incubated with biotinylated antibody specific for lymphatic vessel endotherial hyaluronan receptor 1 ( LYVE-1 ) antibody ( eBioscience ) ( 20 µl for every 1×107 of LNPCs ) for 15 min on ice , washed twice , and resuspended in IMag buffer at a concentration of 2×107/ml . The cell suspension was then incubated with BD IMag Streptavidin Particles Plus-DM ( BD Bioscience ) ( 50 µl for every 1×107 of LNPCs ) for 30 min on ice , and washed twice , and resuspended in IMag buffer at a concentration of 2 to 8×107 . LYVE-1+ cells were then positively selected using BD IMagnet ( BD Bioscience ) following the manufactures instruction , and then stained with PE-conjugated CD147 , APC-conjugated CD31 , and Alexa 700 conjugated CD45 . The purity and viability of LSEC ( CD147+CD31+CD45− ) cells [75] , [76] was routinely greater than 80% and 90% , respectively . For hepatocyte isolation , the pellet from the initial centrifugation step were washed in DMEM containing 10% FCS and centrifuged at 100× g for 2 min at room temperature . This washing step was repeated until the supernatant was no longer cloudy . Hepatocyte viability was routinely higher than 80% . For intrahepatic Dendritic cell ( DC ) and Kupffer cell ( KC ) isolation , intrahepatic lymphocytes ( IHLs ) were isolated as described in the previous section , and DCs and Kupffer cells were positively selected using biotinylated CD11c and CD11b and streptavidin Magnetic Particles ( BD Biosciences ) , following the manufactures instruction . The purity of DCs and KCs were then analyzed by staining with PE-conjugated CD11b , APC-conjugated CD11c , PE-Cy7-conjugated F480 , and FITC-conjugated Ly6G . The purities of DCs ( CD11c+CD11b+Ly6G− ) and KCs ( CD11b+F480+CD11c−Ly6G− ) were routinely greater than 50% , and their viabilities higher than 90% . All antibodies were purchased from BD Bioscience and eBioscience . Lymphmononuclear cells isolated from the liver , spleen , peripheral blood and lymph nodes were incubated with a mixture containing the COR93-dimer or ENV28-dimer , APC- , or Pacific Blue-conjugated anti-mouse CD8+ , PE-Cy7 conjugated anti-mouse CD69 , APC-conjugated anti-mouse CD25 or CD62L , FITC-conjugated anti-mouse CD45 . 1 , and PE-conjugated PD-1 or CTLA-4 ( BD Bioscience ) for 1 hour on ice . After washing , the cells were incubated for 30 minutes with APC-conjugated anti-mouse IgG at 4°C to detect dimer positive cells . Dimer without peptide was used as a control . Intracellular cytokine staining ( ICS ) was performed using PE- or APC-conjugated anti-mouse IFNγ , APC-conjugated anti-mouse Granzyme B ( GrB ) ( Caltag ) after incubating for 5 hours at 37°C in the presence of brefeldin A ( BFA ) , as previously described [21] , [25] , [49] . All antibodies were purchased from BD Bioscience and eBioscience . Total liver DNA and RNA were analyzed for HBV replicative intermediates by Southern blot , and for HBV RNA by Northern blot , exactly as previously described [19] , [20] . The relative abundance of specific DNA and RNA molecules was determined by phosphor imaging analysis , using the Optiquant image analysis software ( Packard ) . The extent of hepatocellular injury was monitored by measuring sALT activity at multiple time points after treatment as previously described [20] . Student t test was performed using Microsoft Excel . Data are depicted as the mean ± SD , and P values<0 . 05 were considered significant: *P<0 . 05 , **P<0 . 01 , ***P<0 . 001 . | Hepatitis B virus ( HBV ) infection is responsible for more than 500 , 000 deaths annually as a result of the immune-mediated chronic liver damage it induces . The HBV specific CD8+ T cell response contributes to the pathogenesis of liver disease and viral clearance , and the failure to induce and/or sustain a vigorous CD8+ T cell response results in viral persistence and causes chronic necroinflammatory liver disease . To understand how the HBV-specific CD8+ T cell response is generated in response to intrahepatically expressed HBV , we generated T cell receptor transgenic mice whose CD8+ T cells are specific for HBV core or HBV envelope antigens . We find that these T cells are primed in the liver when they are adoptively transferred into HBV transgenic mouse recipients whose livers produce infectious virus particles , and that they proliferate vigorously in situ but do not differentiate into functional effector T cells after antigen recognition . Functional differentiation is suppressed by dominant negative regulatory signals , including PD-1 , unless they are suppressed by anti-CD40 activation of myeloid dendritic cells . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"medicine",
"infectious",
"diseases",
"hepatitis",
"hepatitis",
"b",
"immune",
"cells",
"gastroenterology",
"and",
"hepatology",
"virology",
"t",
"cells",
"immunology",
"biology",
"microbiology",
"viral",
"diseases",
"liver",
"diseases"
] | 2013 | CD40 Activation Rescues Antiviral CD8+ T Cells from PD-1-Mediated Exhaustion |
Usutu ( USUV ) and Zika ( ZIKV ) viruses are emerging arboviruses of significant medical and veterinary importance . These viruses have not been studied as well as other medically important arboviruses such as West Nile ( WNV ) , dengue ( DENV ) , or chikungunya ( CHIKV ) viruses . As such , information regarding the behavior of ZIKV and USUV viruses in the laboratory is dated . Usutu virus re-emerged in Austria in 2001 and has since spread throughout the European and Asian continents causing significant mortality among birds . Zika virus has recently appeared in the Western Hemisphere and has exhibited high rates of birth defects and sexual transmission . Information about the characteristics of USUV and ZIKV viruses are needed to better understand the transmission , dispersal , and adaptation of these viruses in new environments . Since their initial characterization in the middle of last century , technologies and reagents have been developed that could enhance our abilities to study these pathogens . Currently , standard laboratory methods for these viruses are limited to 2–3 cell lines and many assays take several days to generate meaningful data . The goal of this study was to characterize these viruses in cells from multiple diverse species . Cell lines from 17 species were permissive to both ZIKV and USUV . These viruses were able to replicate to significant titers in most of the cell lines tested . Moreover , cytopathic effects were observed in 8 of the cell lines tested . These data indicate that a variety of cell lines can be used to study ZIKV and USUV infection and may provide an updated foundation for the study of host-pathogen interactions , model development , and the development of therapeutics .
Usutu virus ( USUV ) , first identified in South Africa in 1959 , is a flavivirus belonging to the Japanese encephalitis complex [1 , 2] . In 2001 , USUV emerged in Austria and spread throughout the European and Asian continents [3–10] . Unlike USUV circulating in Africa , the new emergent strains caused significant mortality among European blackbirds , owls , and other wild and captive birds [3 , 11] . The life cycle of USUV is composed of transmission from primarily Culex mosquito vectors to avian reservoir hosts in a sylvatic transmission cycle [1] . Other than birds , evidence for USUV infection has been found in humans , horses , and bats [12–15] . Several human cases have been identified in Europe and Croatia [16–18] . Recently , USUV has been linked to neuroinvasive infections in 3 patents from Croatia [10] and has been detected in horses in Tunisia [14] . Zika virus ( ZIKV ) is an emerging , medically important arbovirus . There are two geographically distinct lineages of circulating ZIKV; African and Asian [19] . The Asian lineage has recently emerged in Micronesia where it was the cause of a large outbreak in 2007 [20] and currently in the Americas [21] . The natural hosts of ZIKV include humans , primates , and Aedes mosquitos [22–25] . Though no solid evidence exists of non-primate reservoirs of ZIKV [26] , antibodies to ZIKV have been detected in elephants , goats , lions , sheep , zebra , wildebeests , hippopotamuses , rodents , and other African ruminants [27 , 28] . Like many other tropical arboviruses , human infection with ZIKV typically presents as either asymptomatic or acute febrile illness with fever , rash , headache , and myalgia . The flavivirus , dengue virus ( DENV ) and the alphavirus , chikungunya virus ( CHIKV ) produce similar symptoms to ZIKV but are more commonly diagnosed . The high seroprevelance of ZIKV antibodies in human populations in Africa and Asia suggests the misdiagnosis of ZIKV for other arboviral illnesses is an ongoing problem [19] . There are several characteristics of ZIKV that distinguish it from other medically important arboviruses . In recent outbreaks , ZIKV has exhibited atypical symptoms including respiratory involvement and frequent conjunctivitis [20 , 29] . ZIKV also has the ability to spread from human to human through sexual and maternal-fetal transmission [30–32] . ZIKV has been linked to serious medical conditions such as microcephaly and other brain abnormalities in neonates and Guillain-Barré ( GB ) syndrome in adults [31–33] . While research in serology and genetic characterization are underway [19 , 20] , the recent changes in biology and distribution of these viruses warrant further investigation as many questions regarding the basic biology and ecology of ZIKV and USUV remain unanswered . To better understand the characteristics of USUV and ZIKV in vitro , we investigated the permissiveness of several cell lines .
Seventeen cell lines were obtained from the ATCC ( Manassas , VA ) and included Tb 1 Lu , DF-1 , Sf 1 Ep , EA . hy . 926 , CRFK , E . Derm , FoLu , Pl 1 Ut , OHH1 . K , OK , DN1 . Tr , PK ( 15 ) , LLC-MK2 , BT , MDCK , WCH-17 , and Mv 1 Lu ( Table 1 ) . These lines were selected to include domestic and peridomestic representatives of species found only in the Americas; specifically , North America . All cell lines were passaged 5 times after the initial expansion from the ATCC stock prior to experiments . All cells were cultured in Dulbecco's modified eagle medium ( DMEM ) supplemented with 10% ( v/v ) fetal bovine serum ( FBS ) , 4mM L-glutamine , 10 mM non-essential amino acids ( NEAA ) , 1 mM sodium pyruvate , 100 U/ml penicillin , 100 μg/ml streptomycin , and housed in a 37°C incubator with 5% CO2 . USUV ( SAAR-1776 ) , ZIKV ( MR766 –original , African ) , YFV ( 17D ) , Sindbis virus ( SINV EgAr 339 ) , CHIKV ( 181/25 ) , DENV-1 ( H87 ) , DENV-2 ( NGC ) , DENV-3 ( HI ) , and DENV-4 ( H241 ) were obtained from the World Reference Center for Emerging Viruses and Arboviruses ( Robert Tesh , UTMB , Galveston , TX ) . These viruses were of low passage stock that had been in storage since the mid-20th century . ZIKV ( PRVABC59 Puerto Rico 2015 , Asian ) was obtained from the American Tissue Type Collection and RNA was extracted directly from the sample upon arrival . WNV ( NY99 ) was obtained in 2001 from the National Veterinarian Services Laboratory and had undergone only 2 expansions prior to use . The titers of all viruses were determined via a plaque forming unit ( PFU ) assay in LLC-MK2 cells except DENV which , titer was determined with a focus-forming unit assay . All virus titrations were performed using 12-well standard cell culture plates seeded with cells to reach 100% confluency upon infection . Cells were inoculated with 10-fold serial dilutions of the recovered sample and were rocked at 37°C for one hour after which the inoculum was removed and replaced with an overlay of 1 ml of 1% methyl cellulose ( Sigma Catalog # M0512 ) mixed 1:1 with 2x MEM ( 20% FBS , 8 mM glutamine , 20 mM NEAA , 2% penicillin/streptomycin , 2 mM sodium pyruvate ) . Plates were placed in a 37°C incubator with 5% CO2 for 4 days . Cells were stained using 70% ethanol containing 1% wt/vol crystal violet . Plates were incubated for 15 minutes at 22°C after which the fixative was decanted . The plates were rinsed with cold water and dried overnight at room temperature . The titer of DENV was determined through a focus-forming unit assay . Briefly , cells were fixed and permealbilized using 1 ml of a 1:1 acetone/methanol solution with a 60 minute incubation at 4°C . Virus foci were detected using a specific mouse monoclonal antibody from hybridoma 2H2 ( Millipore catalog #MAB8705 ) , followed by a horseradish peroxidase-conjugated goat anti-mouse immunoglobulin ( Millipore catalog #AP124P ) , and developed using a 50mg tablet of 3 , 3’-Diaminobenzadine tetrahydrochloride ( Sigma catalog # D5905 ) dissolved in 20mL PBS with 8uL 30% hydrogen peroxide . All infections were performed using 24-well standard cell culture plates seeded with cells which , had reached a 90% confluence upon infection . Individual wells were inoculated with 1 , 000 PFU of virus ( ≈MOI 0 . 005 ) in 150ul of MEM and then rocked at 37°C for one hour after which the inoculum was removed , rinsed twice with sterile PBS , overlaid with 1 ml of DMEM ( 10% FBS , 4 mM glutamine , 10 mM NEAA , 100mg/ml penicillin/streptomycin , 1mM sodium pyruvate ) and incubated at 37°C incubator with 5% CO2 . Culture supernatants were collected at 1 and 72 hours post-inoculation ( PI ) . All infections were performed using 12-well standard cell culture plates seeded with cells which , had reached a 90% confluence upon infection . Individual wells were inoculated with 1 , 000 PFU of virus ( ≈ MOI 0 . 0025 ) in 200ul of MEM and then rocked at 37°C for one hour after which the inoculum was removed , rinsed twice with sterile PBS , overlaid with 1 ml of DMEM ( 10% FBS , 4 mM glutamine , 10 mM NEAA , 100mg/ml penicillin/streptomycin , 1mM sodium pyruvate ) and incubated at 37°C incubator with 5% CO2 . All cell lines were allowed to develop CPE for 7 days PI . Cells were stained using as described earlier . Images were obtained using Micron imaging software ( Westover Scientific ) and an inverted microscope at 40X magnification . Primers for USUV were designed against the USU181 sequence ( Genbank accession: JN257984 ) and amplify a 104 base pair fragment of the envelope protein gene starting at nucleotide position 1325 and ending at position 1428 ( Table 2 ) . Primers for ZIKV were designed against the MR766 strain ( Genbank accession: AY632535 ) and amplify a 128 base pair fragment of the envelope glycoprotein starting at nucleotide position 1398 and ending at position 1525 . Blasts for these primer sequences showed sequence homology to multiple strains of the respective virus but no homology to other viruses . This protocol did not detect RNA derived from ZIKV strain PRVABC59 , a Puerto Rican isolate from the 2015 outbreak . A standard curve for each virus was constructed in which , 10-fold serial dilutions of virus stock that had been titrated in LLC-MK2 cells via PFU assay , were compared to the cycle threshold ( Ct ) values from the real-time RT-PCR ( qRT-PCR ) . The USUV primer set could detect as few as 10 PFU per mL and the ZIKV primer set was able to detect as few as 100 PFU/mL . Both primer sets did not amplify other arboviruses tested including: WNV , SINV , YFV , DENV serotypes 1–4 , and CHIKV . Sequences for the primer sets are listed below: Of the cell lines tested , only LLC-MK2 cell lines consistently produced viral plaques . The FoLu cell line initially produced large round plaques at 3 days for both ZIKV and USUV but lost the ability to produce plaques after subsequent passaging of the cell line . In order to determine if , and how much , virus was being produced by these cells , qRT-PCR was employed . Viral RNA was extracted from cell culture supernatant using the Ambion MagMax-96 extraction kit ( Life Technologies: Grand Island , NY ) per manufacturer’s instructions . The qRT-PCR reactions were conducted using a BioRad Superscript One Step SYBR Green qRT-PCR kit ( Winooski , VT ) . The following cycling conditions were employed: reverse transcription at 50°C for 10 min , denaturation at 95°C for 5 min , followed by 40 cycles of denaturation and amplification at 95°C for 10 sec and 55°C for 30 sec . Cycle threshold values were used to estimate relative viral titers of infected cell lines according to a standard curve created using a serial dilution of known viral concentrations of virus that produced plaques in LLC-MK2 cells . Results are expressed as the average of 3 independent trials amplified in duplicate . A series of controls were included in each plate in order to identify true positives not related to background . A no-template control and a no-primer control were performed to verify that the reagents and equipment were working as expected . A positive virus control was used on each plate . A non-infected cell culture supernatant control was included to verify that there was no increase in non-specific binding from the PCR primers that could cause a higher background signal . Finally , the cell culture supernatants were collected 1 hour PI to ensure that qRT-PCR results , 72 hours PI , were not convoluted by input virus . RT-PCR data were analyzed using the ΔΔCt method . Replicates were pooled , averaged , and standard deviation was calculated . If a standard deviation was greater than 3 , any outliers were removed from the analysis . The LLC-MK2 cell line was used as the reference cell line . To determine if cell resistance to USUV or ZIKV was binding dependent , a virus: cell binding assay was performed as previously described by Thaisomboonsuk , et al [34] . Briefly , confluent LLC-MK2 cells in 6-well plates were rinsed 3 times with ice-cold PBS and then 3 ml of ice-cold binding medium ( DMEM containing 0 . 8% BSA and 25 mM HEPES , pH 6 . 0 ) was added to each well . Plates were incubated for 1 hour on ice . The medium was aspirated and 600ul of 10 , 000 PFU of virus ( ≈ MOI 0 . 01 ) in ice-cold binding medium was added to the cells and incubated on ice for 2 hours with rocking every 15 minutes . The inoculum was then removed and the cells rinsed 3 times with ice-cold PBS . An RNA extraction of the cell monolayers was immediately performed using the Qiagen RNeasy mini kit ( Valencia , CA ) per the manufacturer’s instructions and qRT-PCR was performed as described above . Results are expressed as the average of 3 independent trials amplified in duplicate .
Of the 17 cell lines tested , all showed quantifiable Ct values for both USUV and ZIKV based upon qRT-PCR data at 72 hours PI ( Fig 1 ) . The cell lines WHC-17 , E . Derm , BT , EA . hy . 926 , Tb 1 Lu , Sf 1 Ep , DNl . Tr , and Pl 1 Ut produced significantly less USUV than the LLC-MK2 cell line ( P = 1 . 27−07 , 1 . 71−03 , 1 . 4−08 , 3 . 97−07 , 2 . 48−04 , 1 . 84−06 , 1 . 7−08 , 7 . 64−08 ) ( Fig 1 ) . The cell lines OHH1 . K , MDCK , FoLu , Mv 1 Lu , and PK ( 15 ) were able to replicate USUV as well as the LLC-MK2 cell line ( P = 0 . 66 , 0 . 74 , 0 . 05 , 0 . 09 , 0 . 09 ) ( Fig 1 ) . The OK , CRFK , and DF-1 cell lines were able to produce significantly more USUV than LLC-MK2 cells ( P = 2 . 46−07 , 0 . 025 , and 0 . 021 ) ( Fig 1 ) . The cell lines OK , E . Derm , CRFK , OHH1 . K , FoLu , and PK ( 15 ) produced just as much ZIKV as the LLC-MK2 reference line ( P = 0 . 52 , 0 . 2 , 0 . 16 , 0 . 13 , 0 . 057 , and 0 . 56 ) ( Fig 1 ) . The BT , EA . hy . 926 , WCH-17 , DN1 . Tr , Tb 1 Lu , MDCK , Pl 1 Ut , DF-1 , Mv 1 Lu , and Sf 1 Ep cell lines produced significantly less ZIKV than the LLC-MK2 line ( P = 1 . 71−03 , 0 . 017 , 4 . 67−08 , 0 . 038 , 1 . 59−10 , 6 . 97−04 , 1 . 31−04 , 2 . 5−05 , 7 . 65−04 , 8 . 53−4 ) ( Fig 1 ) . USUV and ZIKV were detected in very low quantities ( ≈100pfu or less/ml ) in WCH-17 cells , over 12-fold less than LLC-MK2 cells ( P = 1 . 27−07 and 4 . 67−08 ) . Likewise , Tb 1 Lu cells produced more than 15-fold less ZIKV ( ≈100pfu or less/ml ) than LLC-MK2 cells ( P = 1 . 59-10 ) . CPE was not evident for either virus in either cell line . A virus: cell binding assay was performed in order to determine if cell receptors were present that would allow ZIKV or USUV to attach to the Tb 1 Lu or WCH-17 cell surface . The Ct values for all treatments express the amount of virus present in the sample . A Student’s T-test comparing the virus:cell binding of WHC-17 and Tb1 . Lu cells to LLC-MK2 cells indicated a lack of difference between the cell lines ( Table 3 ) . The statistical similarity of the data suggests that both ZIKV and USUV bind to WCH-17 cells and ZIKV binds to Tb 1 Lu cells as efficiently as they bind to the LLC-MK2 control cells ( Table 3 ) . Cytopathic effects were observed in CRFK , DN1 . Tr , E . Derm , EA . hy . 926 , FoLu , OHH1 . K , OK , PK ( 15 ) , Sf 1 Ep , and Mv 1 Lu cell lines from both ZIKV and USUV infection . Forms of CPE caused by USUV in CRFK cells included pyknosis ( Fig 2A ) , while ZIKV caused focal degeneration in addition to pyknosis . Dn1 . Tr cells exhibited pyknosis , koilocytes , enlargement , and rounding in response to ZIKV and USUV infection ( Fig 2B ) . E . Derm cells did not show consistent CPE when infected with USUV but did exhibit pyknosis in response to ZIKV ( Fig 2C ) . EA . hy . 926 cells produced pyknosis , koilocytes , rounding , and enlargement when infected with ZIKV or USUV however; ZIKV also produced focal degeneration ( Fig 2D ) . FoLu cells produced koilocytes and enlargement when infected with USUV and pyknosis when infected with ZIKV ( Fig 2E ) . USUV produced pyknosis and cellular enlargement in OHH1 . K cells and ZIKV produced koilocytes , enlargement and , infrequently , focal degeneration ( Fig 2F ) . OK cells produced pyknosis , koilocytes , enlargement , rounding , and focal degeneration when infected with USUV but only pyknosis and overgrowth when infected with ZIKV ( Fig 2G ) . Focal degeneration , pyknosis , and rounding were produced in PK ( 15 ) cells when infected with either USUV or ZIKV ( Fig 2H ) . Sf 1 Ep cells exhibited pyknosis , koilocytes , rounding , and enlargement when infected with either ZIKV or USUV with ZIKV also causing focal degeneration ( Fig 2I ) . Mv 1 Lu cells exhibited focal degeneration and koilocytes in response to USUV infection and cellular enlargement when infected with ZIKV ( Fig 2J ) . LLC-MK2 cells produced focal degeneration and cellular enlargement within 3 days PI and by 7 days PI , most cells were dead and detached ( Not pictured ) .
The data herein indicate that several cell lines can be used to culture and study USUV and ZIKV . The susceptibility for certain cell lines to USUV and ZIKV may provide a tool for characterizing these viruses and may provide an in vitro platform for the study of host-pathogen interactions , model development , and the development of therapeutics . Additional questions not addressed in this data included whether or not the broad host infectivity observed may be a function of the virus strains that were used for the experiments . These strains may not accurately reflect the characteristics of USUV or ZIKV currently circulating , or that of other laboratory-adapted strains . Finally , the behavior of USUV and ZIKV in the laboratory does not reflect the behavior of these viruses in their natural environment . | Usutu and Zika viruses are arboviruses of significant medical and veterinary outbreaks in recent years . Currently , standard laboratory methods for these viruses are limited to 2–3 cell lines . Here , our studies demonstrate that Zika and Usutu viruses are able to replicate in cells from a wide range of animal cell lines . The data will allow for further study of the potential for evolution of these viruses in other hosts . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"japanese",
"encephalitis",
"virus",
"dengue",
"virus",
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"togaviruses",
"pathogens",
"biological",
"cultures",
"microbiology",
"animals",
"alphaviruses",
"viruses",
"chikungunya",
"virus",
"rna",
"viruses",
"cell",
"cultures",
"insect",
"vectors",
"research",
"and",
"analysis",
"methods",
"medical",
"microbiology",
"epidemiology",
"microbial",
"pathogens",
"disease",
"vectors",
"insects",
"arthropoda",
"mosquitoes",
"west",
"nile",
"virus",
"arboviruses",
"flaviviruses",
"viral",
"pathogens",
"biology",
"and",
"life",
"sciences",
"organisms",
"zika",
"virus"
] | 2016 | Working with Zika and Usutu Viruses In Vitro |
Spike timing-dependent plasticity ( STDP ) has been shown to enable single neurons to detect repeatedly presented spatiotemporal spike patterns . This holds even when such patterns are embedded in equally dense random spiking activity , that is , in the absence of external reference times such as a stimulus onset . Here we demonstrate , both analytically and numerically , that STDP can also learn repeating rate-modulated patterns , which have received more experimental evidence , for example , through post-stimulus time histograms ( PSTHs ) . Each input spike train is generated from a rate function using a stochastic sampling mechanism , chosen to be an inhomogeneous Poisson process here . Learning is feasible provided significant covarying rate modulations occur within the typical timescale of STDP ( ∼10–20 ms ) for sufficiently many inputs ( ∼100 among 1000 in our simulations ) , a condition that is met by many experimental PSTHs . Repeated pattern presentations induce spike-time correlations that are captured by STDP . Despite imprecise input spike times and even variable spike counts , a single trained neuron robustly detects the pattern just a few milliseconds after its presentation . Therefore , temporal imprecision and Poisson-like firing variability are not an obstacle to fast temporal coding . STDP provides an appealing mechanism to learn such rate patterns , which , beyond sensory processing , may also be involved in many cognitive tasks .
STDP is now a well-established physiological mechanism of activity-driven synaptic regulation [1] , which can capture spiking information at a short timescale , down to milliseconds [2] , [3] . Although the relationship between the stimulating input structure and the resulting weight specialization has been investigated in a number of theoretical studies [4] , [5] , [6] , [7] , most of them have limited their scope to general and abstract input structures . A practical and fundamental question is to understand how , in natural or experimental situations , STDP can participate in the learning process . Importantly , although repeated stimulus presentations , or task trials , induce memorization ( e . g . [8] ) , the underlying neural mechanisms remain largely unknown . In this respect , a recent numerical study showed that a repeating arbitrary , but reliable , spatiotemporal spike pattern embedded in equally dense random activity can be learned and robustly detected by a single neuron equipped with STDP [9] . However , such reliable spike patterns have received scarce experimental evidence ( but see [10] , [11] , [12] , [13] , [14] , [15] , [16] ) and may constitute a very special case of activity . More generally , and across trials , spike trains often exhibit large variability , and can be described using an underlying probabilistic firing intensity , for example through inhomogeneous Poisson sampling . This hypothesis is tenable with most – if not all – experimental datasets , where the temporal spiking probability – or rate – is measured by a post stimulus time histogram ( PSTH ) [17] . PSTHs usually exhibit temporal peaks , whose spread width is of the order of ∼10–20 ms in many experimental findings [18] , [19] , [20] , [21] , [22] , [23] , [24] , [25] , [26] . Whether STDP is able to learn such rate patterns is currently unclear . Somewhat surprisingly , our study shows that such spread widths and Poisson-like firing variability are not an obstacle to STDP-based pattern learning , and fast and efficient detection afterwards . We consider a single postsynaptic neuron excited by presynaptic neurons ( Fig . 1Schema ) , of which an arbitrary and hidden number are involved in the repeating presentation of a given pattern , embedded in otherwise random spike trains . A main and novel contribution of this study concerns patterns generated using covariations of the input instantaneous rates , from which spikes are generated through an inhomogeneous Poisson process . To predict the evolution of the synaptic weights and the resulting neuronal selectivity , we analyze theoretically a dynamical system that describes the effect of STDP . We confirm these results using numerical simulations . We demonstrate that repeated presentations of such rate patterns induce spike-time correlations that are captured by STDP , even when rate peaks have a width up to 10–20 ms . In general , STDP favors synapses corresponding to early spikes in the pattern , resulting in fast response whenever the pattern is presented [9] . However , when the pattern exhibits sharp and/or large-amplitude peaks for several inputs , STDP tends to favor some of the corresponding synapses . Besides rate-modulated patterns , our theory also applies to spike patterns that we therefore include here for the sake of completeness .
Recent work has focused on generating spike trains with a given correlation structure , using a mixture of spike coordination and rate covariation [28] . Here these two mechanisms are used to generate each a class of patterns: spike patterns ( model S ) and rate-modulated patterns ( model R ) , the last mimicking PSTH-like probabilistic spiking activity . Throughout the present paper , we consider a single pattern that is presented to an unknown subset of among excitatory afferent ( or input ) plastic synapses that stimulate a single neuron ( Fig . 1Schema ) . The afferents involved and those not involved in the pattern will be denoted by pattern and non-pattern afferents , respectively . Pattern presentations occur randomly with frequency and duration , without overlapping . All pattern models rely on latencies for each pattern afferent and . Unless said otherwise ( cf . bimodal patterns below ) , the latencies are uniformly distributed in , i . e . , corresponding to a ( single ) realization of a homogeneous Poisson process with intensity rate for the duration for each input . Once determined these latencies ( possibly none ) for all pattern inputs , we generate the input spikes for each pattern presentation as follows: For each of the non-pattern inputs , as well as the pattern inputs outside pattern presentations , latencies similar to are generated using a homogeneous Poisson process with rate ; then the same spike generation applies according to each model . Note that the choice of Gaussian functions in model R is motivated by analytical tractability , but any peaked function could be used . We did not consider a probability of occurrence for the Gaussian peaks in model R ( similar to in model SD ) as variability was already present in the spike generation . In the remainder of the present paper , we sometimes refer to models S in general as spike patterns , which include model SJ , in contrast to model R . In the baseline simulations , and unless stated otherwise , we use , , Hz , Hz and ms . We use an abstract model of STDP where the weight change depends on the relative spike timing and the current value for the weight . In our model , each single spike and each pair of presynaptic and postsynaptic spikes contribute once to plasticity . A sole pair with respective times and induces a weight change determined by the following contributions ( 1 ) The rate-based contributions and account for the ( one-shot ) effect of each pre-and postsynaptic spikes , respectively [4] , [27] . They model homeostatic synaptic scaling mechanisms [29] ( see also Discussion ) and allow us to examine theoretically the weight specialization depending on individual spikes , while evacuating rate effects [27] . Those terms are not seen in typical STDP experiments , but they could be easily missed if its magnitude is lower than the STDP changes . Theoretically , can be chosen equal to zero provided depression dominates the STDP effects . Even though there exists a parameter range for which rate effects are small ( or even balance each other ) and STDP dominates the plasticity effects , we have used values for these rate-based coefficients that gave very robust results without any fine-tuning in our baseline numerical simulations . The STDP learning window function describes the effect of long-term potentiation ( LTP ) and long-term depression ( LTD ) on the weight as a function of the spike-time difference and weight . The learning rate determines the speed of learning . We consider a classic learning window function determined by a decaying exponential for each side ( see dashed line Fig . 2A ) : ( 2 ) where fits the mean effect observed in experimental data [30]; a complete list of parameters can be found in Text S1 ( Section S3 . 2 ) . The scaling functions determine the dependence of the change in the weight on its current value [31] . In the analysis and most numerical simulations , we use additive STDP , for which these functions are constant , namely , like in [9] , which leads to bimodal weight distributions [32] , and therefore strong resulting selectivity . However a slightly multiplicative STDP can be of interest , because it can ensure both competition and homeostasis [33] , without the need for the additional rate-based homeostatic terms ( ) . This will be used in the Results section ( ‘Influence of the STDP and neuronal parameters’ ) with the model proposed in [6] for which a parameter scales between additive ( ) and multiplicative ( ) STDP: ( 3 ) The constant is the upper “soft” bound , while the lower bound is set to zero . However , a too strong weight dependence weakens ( and eventually impairs ) the resulting specialization [6] , [33] . The Poisson neuron [4] is an abstract neuronal model where the spiking mechanism that generates the spike train is approximated by an inhomogeneous Poisson process . The latter is driven by a ( positive ) rate function that mimics the mean potential of a soma receiving pre-synaptic activity: ( 4 ) where is the positive part: . The time course of each EPSP following the -th spike arriving at synapse at time is described by the kernel function rescaled by the weight . The constant relates to other non-plastic input connections that are not considered in detail; they can be excitatory or inhibitory . Our analysis assumes for in order to preserve causality and to be normalized: . Note that the calculations are exact only when the soma potential is positive at all times . Numerical simulations use , where ms and ms are the decaying and rising time constants , respectively; denotes the Heaviside step function . To minimize false alarms in a detection scheme using a single neuron , we have used inhibition ( background activity ) , which leads to a subthreshold regime where the postsynaptic neuron has a low output firing rate . This complies with in vivo experiments where neurons receive excitatory and inhibitory inputs that almost balance each other [34] or even favor inhibition [35] . Using strong inhibition , a single Poisson neuron can be trained to be almost as reliable as a deterministic LIF neuron for pattern detection ( see Results section ) . Our choice for the models of additive STDP and Poisson neuron allows us to derive a dynamical system to analytically examine the weight dynamics . We draw on a previously developed framework [4] , [27] , where details can be found . Under the assumption of slow learning compared to other ( firing and synaptic ) mechanisms , an intermediate averaging period can be chosen between the two corresponding time scales . The expectation of the weight update corresponding to Equation ( 1 ) over the period can be evaluated using the firing rates and spike-time covariance of the input and output spike trains: ( 5 ) Here spikes are considered to be quasi-instantaneous events , so the spike trains and for of input and the neuron are modeled as a sum of delta functions ( Dirac combs ) . The angular brackets denote the ensemble average over the randomness of the spike trains , since we generate external inputs using stochastic processes . In Equation ( 5 ) , the terms and are associated with the mean ( time-averaged ) firing rates of the -th afferent and neuron , and , respectively , which are defined similar to: ( 6 ) The contributions of spike pairs that involve the STDP learning window function in Equation ( 1 ) is decomposed into two terms ( on the last line ) . The first one gives the product of the pre- and postsynaptic firing rates with , the integral value of . The second one gives the convolution of with the neuron-to-input ( time-averaged ) cross-covariance between the neuron and input : ( 7 ) The Poisson neuron described by Equation ( 4 ) leads to the following consistency equations for the neuronal output firing rate and neuron-to-input covariance , respectively: ( 8 ) where the input-to-input ( time-averaged ) cross-covariance is defined by ( 9 ) Equations ( 8 ) are exact only when the soma potential in Equation ( 4 ) is positive at all times; this is of importance when considering negative values for the constant . As will be justified in the following section , the input firing rates in Equation ( 6 ) and spike-time covariance in Equation ( 9 ) are independent of time , so we omit the latter variable thereafter ( except for the plastic weights when it is useful to precise ) . We combine Equations ( 5 ) and ( 8 ) to obtain Equation ( 29 ) in Results . Using matrix notation , it can be rewritten as a linear differential for the drift ( first stochastic moment ) of the synaptic weights in terms of the input properties: ( 10 ) The -column vector contains the input firing rates ; denotes the transpose of ; is the -row vector of the weights; we have also defined as the -column vector that has all its elements equal to one . The matrix absorbs the input correlations , neuronal and STDP parameters ( 11 ) where denotes the usual convolution of functions . The previous section showed how the weight evolution defined in Equation ( 1 ) is determined by the input firing rates in Equation ( 6 ) and spike-time covariances embodied in the coefficients of matrix in Equation ( 11 ) . In order to predict the weight evolution , we need to evaluate the respective variables and for input spike trains that convey pattern activity . The present study compares the two classes of patterns described at the beginning of Materials and Methods: model S with coordination of spike times; and model R with covariation of firing rates for inhomogeneous Poisson processes . We provide details of the calculations that lead to the following results in Text S1 ( Section 1 ) . In all the pattern models considered throughout this paper , input spike trains have a quasi-constant mean firing rate as defined in Equation ( 6 ) . This follows because we consider relatively short patterns ( ms ) compared to the averaging period and that the pattern spikes ( per input ) correspond to a firing rate comparable to the background rate . We focus on the “difficult” situation where the frequency of the pattern presentations is not too high , such that the condition is satisfied . In this case , the discrepancies between the numbers of pattern spikes for different inputs and different presentations do not affect the mean firing rates , which are almost identical for all inputs ( and roughly equal to the background rate ) : ( 12 ) For rate-modulated patterns of model R , Equation ( 12 ) requires that all rate peaks are normalized ( ) . Detailed calculations are provided in Text S1 ( Section 1 . 1 ) . Similar to the mean firing rates , the mean covariances are also practically independent of time . In spike patterns of model S ( with no jitter ) , pattern inputs repeatedly and consistently fire spikes with given latencies . Consequently , each pair of pattern input spike trains involves synchrony with time lags that are determined by the relative spike latencies . In other words , the corresponding spike-time correlogram defined in Equation ( 9 ) exhibits a peak for those preferred time lags , as detailed in Text S1 ( Section 1 . 2 ) . Namely , for two pattern inputs and with respective latencies and , we have ( 13 ) where is the frequency of the pattern presentation and is the probability for a spike at each latency to be fired during a pattern presentation . The approximation in Equation ( 13 ) neglects a term related to the “silence” for pattern inputs during the pattern presentation beside the spikes at latencies . The same term applies to all pairs among the pattern inputs and can thus be ignored when studying the emerging weight structure; this also partly explains discrepancies between theoretical predictions and numerical simulations ( see “Weight specialization by competition” ) . Jittering the spike times around the mean latencies amounts to replace the delta function in Equation ( 13 ) by the convolution of the jitter distributions . Consequently , Gaussian jitters with spread widths in model SJ lead to: ( 14 ) where is the normalized Gaussian function of width : ( 15 ) Details are provided in Text S1 ( Section 1 . 3 ) . Model SB is a particular case of model S , where inputs are partitioned into two groups . Each input in group has a single latency generated by a Gaussian distribution that is common to all inputs from the same group , namely with mean latency and variance . Recall that we constrain this special case of model S to and no jitter ( ) . After population average ( denoted by the overline ) , the mean cross-correlogram between input groups and in Equation ( 13 ) becomes: ( 16 ) Details are provided in later in Text S1 ( Section 1 . 5 ) . In a sense , the randomness over each population plays the same role as individual jitters in Equation ( 14 ) . For patterns of model R , the covariances are given in a similar manner by the convolution of the Gaussian kernels that determine the rate covariations for each spike train ( Fig . 1R1 ) : ( 17 ) where , and are the center , the width and the amplitude , respectively , of the corresponding rate peaks . Note that we take into account the whole Gaussian functions even outside the pattern of duration . See Text S1 ( Section 1 . 4 ) for details . The expressions in Equations ( 13 ) , ( 14 ) and ( 17 ) are actually particular cases of the same general formulation given in Equation ( 28 ) in Results . Once known the input firing rates and spike-time correlations for a given pattern , the weight dynamics can be analyzed using Equation ( 10 ) . We require STDP to produce a stable equilibrium for the mean input weight , which also stabilizes the neuronal output firing rate . This favors an effective weight specialization when maintaining the mean weight between the lower and upper bounds , which allows to potentiate ( select ) some weights while depressing ( discarding ) others . As detailed in Text S1 ( Section 2 . 1 ) , we average Equation ( 10 ) and ignore matrix to evaluate the dynamics of the mean weight . Note that this is equivalent to averaging over all inputs Equation ( 29 ) in Results and neglecting the correlation terms involving . The conditions ( 18 ) ensure stable fixed points both for at the equilibrium value ( 19 ) and output neuronal firing rate at ( 20 ) For the equilibrium to be realizable , the equilibrium value must be within the weight bounds ( e . g . , Fig . S1 in Text S1 ) , which implies in particular that ; a sufficiently large value for can ensure this condition to be satisfied [4] , [27] or a negative value can be used , as chosen here . Here the equilibrium is homeostatic [29] in the sense that the constraint in Equation ( 19 ) scales up the weights if their number decreases in order to maintain the level of the neuronal firing rate . This also guaranties that the neuron will not become silent when it is stimulated . For sufficiently strong input correlations ( i . e . , peaked correlograms in the right panels of Fig . 1 ) , the spike-based correlation terms become the leading order in Equation ( 10 ) , because the other terms roughly cancel each other so long as the homeostatic equilibrium remains satisfied . This causes competition between individual weights , but does not impair the stability of homeostasis . When the homeostatic equilibrium is realizable , the mean weight quickly converges close to the predicted equilibrium value and remains in the vicinity during the whole simulation time . Here the discrepancy of about 15–20% can be explained by the fact that the prediction does not involve the mean input spike-time correlation; cf . Section 2 . 1 in Text S1 for details . Meanwhile , a portion of individual synapses become potentiated , whereas most of the remainder become depressed almost to zero . The result is a bimodal weight distribution , where selected inputs have potentiated synapses , while the remainder hardly affects the output neuronal firing . A typical example can be found in Fig . S1 in Text S1 . Now we show how to predict such a competition between synaptic weights using Equation ( 10 ) . We adapt previous results [4] , [27] to the present context . The key to predict the weight evolution is thus the spectral analysis of matrix . More precisely , the weight specialization is determined by a divergent behavior of individual weights related to the eigenvalues of that have a positive real part . Meanwhile , the preservation of the homeostatic equilibrium is ensured by a largely negative ( real ) eigenvalue of that dominates the spectrum . In the simulations of section “General case of an arbitrary pattern” , this eigenvalue is roughly , about 100 times larger in magnitude than the other eigenvalues . Due to the fast time scale related to this large negative eigenvalue , we consider that homeostasis is attained well before specialization begins . To ensure a strong neuronal selectivity after learning , we have chosen the parameters such that the equilibrium mean input weight in Equation ( 19 ) is low compared to . Because this equilibrium value is low compared to the initial distribution ( uniform between zero and a maximal value , smaller than the upper bound ) , almost all weights are depressed at the beginning of the learning epoch . Consequently , we assume the weights to be roughly homogeneous before being specialized . This explains why the initial weight distribution has no impact in the present study . Note that homeostatic mechanisms ( in particular here ) save silent synapses from becoming static . The emerging weight structure is determined by the eigenvalues with largest positive real parts of matrix , which are actually closely related to those of in Equation ( 11 ) . Therefore , the weight specialization can be predicted with a satisfactory approximation by studying the spectra of alone . For this purpose , we use the expression in Equation ( 11 ) obtained when using the Poisson neuron . For both models S and R , the common expression for their input spike-time correlograms leads to similar dynamics; cf . Equations ( 28 ) and ( 30 ) in Results . Following Equation ( 11 ) , the elements of involve the convolution of with the Gaussian kernel function in Equation ( 15 ) that describes the temporal variability in our pattern models . This leads to the expression for in Equation ( 31 ) in Results . It follows that the temporal resolution of the pattern , measured by , affects the spectrum of , hence the weight dynamics . This effect is verified using numerical simulation in Results . The following section provides a simplified analysis for the weight dynamics for some particular distributions of latencies . However , in the general case , a more complete study of the spectrum of is necessary . Following this initial splitting , some weights start to win the competition to drive the neuronal output firing . More precisely , inputs with potentiated weights also have stronger cross-covariance with the output spike train in the “causal” range , namely for for . From Equation ( 8 ) , it is clear that the synaptic weights linearly scale the input cross-covariance structure in the expression of the neuron-to-input covariance structure . Because of the PSP response ( here embodied in the kernel function ) , peaks in become peaks in , but shifted toward more negative values of , i . e . , toward the causal side . Actually , is related to the driving of the neuronal output firing by input via peaks ( or positive values ) for . In other words , these correspond to spikes that perdict the neuronal output firing in the next instant [3] It follows that an STDP rule that induces potentiation for causal firing ( for ) results in more potentiation for the weights that are already strong ( and thus take a good part in driving the neuron ) . In the case of additive STDP , groups of weights diverge apart from each other until saturation at the upper bound or fading to zero . Then , the choice of the weight bounds affects the learned selectivity of the neuron . Since almost all weights asymptotically become either quiescent at zero or saturated at , the number of potentiated weights is roughly ( provided the homeostatic equilibrium is realizable ) . The portion of selected pattern inputs can thus be adjusted via the equilibrium value in Equation ( 19 ) . The previous study by Masquelier and colleagues used [9] , for which the equilibrium is not realizable . It follows that all input weights tend to be depressed so long as the neuron is not silent . Stronger input correlations ( e . g . , more frequent pattern presentations ) and tuning the STDP parameters are then necessary to obtain effective learning; further details are discussed in Text S1 ( Section S2 . 4 ) . For weight-dependent STDP , the above-mentioned trends are still qualitatively valid , even though the fixed point for the mean weight is also determined by the scaling functions in Equation ( 1 ) and individual weights saturate at stable intermediate values between the bounds . A too strong weight dependence of the learning window may prevent the splitting of the weight distribution [6] and thus compromise the resulting neuronal specialization . Until now , the analysis did not assume any specific distribution of latencies for the pattern models ( either S or R ) , except that the number of pattern spikes for each pattern input roughly corresponds the time-average input firing rate . Now we consider the default case of uniformly distributed latencies during the pattern presentation . We show that a general trend arise because of the temporal ( approximate ) antisymmetry of the STDP learning window considered here , namely inputs with early spikes are more likely to be selected compared to others . When has certain properties , the weight evolution can be described in a more intuitive way than studying its full spectrum . For this alternative analysis , we assume identical input firing rates and homogeneous initial weights . Following a previous study [4] , the relative evolution of two input weights and can be evaluated using the reformulation of the learning equation ( 29 ) in Results . Namely , the difference between their derivatives yields ( 21 ) If , for example , each matrix element of the column for input is stronger than the corresponding element for input , namely ( 22 ) then will increase . Because the homeostatic equilibrium remains satisfied during the weight specialization ( due to the large negative eigenvalue of that dominates the remainder of the spectrum ) , the sum of all input weights is constrained to . Combining these two trends , when inputs can be divided into two groups of inputs ( say , with respective indices and ) such that Equation ( 22 ) holds for all indices in each group , the weights are potentiated whereas are depressed . Now we consider patterns of either model S or R for which the latency for each input is chosen randomly with uniform density over the pattern duration . In other words , all latencies can be found across the population of inputs . In this case , the effect of STDP arises from the temporal antisymmetry of in Equation ( 31 ) and Fig . 2 , itself related to our choice for . For a relatively short pattern duration , each pattern input has only a few spikes or peaks with latencies . In the simple case of a single pattern spike per input , the contributions to the elements of are more likely to be positive for early latencies , since we have then for most : ( 23 ) As a general trend that extends the explanation above , when input has earlier spikes than input , Equation ( 23 ) is satisfied for most indices and input should win the competition over input . Further developments of this argument are presented in Text S1 ( Section S2 . 2 ) for the specific patterns of models SB and RB that are examined in the section “Influence of the spike distribution within the pattern” . A subsequent effect has been demonstrated in a previous numerical study [9]: when some weights saturate ( or are significantly large ) , inputs with spikes that come earlier tend to be potentiated . A Poisson neuron is more likely to fire a postsynaptic spike after each input spike cluster that corresponds to potentiated weights , which explains this reduction of the firing latency . For the deterministic LIF neuron , this effect is more pronounced . However , this prediction may not be valid for a general pattern , in particular with an inhomogeneous spike distribution ( see Sections ‘Influence of the spike distribution within the pattern’ and ‘General case of an arbitrary pattern’ in Results ) ; therefore , we did not develop the theory in that direction . A previous study used similar techniques to investigate the weight dynamics for a general class of time-varying inputs [36] . Apart from the specific input spike trains considered here , a crucial difference here lies in our use of the temporal ( approximate ) antisymmetry of function to extract the spiking information ( correlations ) from the pattern . In the present study , the weight dynamics is very robust , since then contains both positive and negative matrix elements and Equation ( 10 ) has a stronger drift . If function is rather symmetrical as considered by [36] , early spikes are not predicted to be potentiated , but patterns may still be learnable . For a given pattern input of model SD , we denote by the number of spikes involved in the pattern presentation . Each spike has probability to occur , leading to a number of spikes that follows a binomial distribution , with mean and variance . To compensate the missing spikes , a homogeneous Poisson spike train with rate is added to the aforementioned spike train . Because the pattern presentation has duration , the number of additional spikes has the same mean and variance . The random variables and being independent , the total spike count during pattern presentation gives the following Fano factor for the spike train of input : ( 24 ) Note that , for the mean value , the Fano factor simplifies to . When decreases from 1 to 0 , increases continuously from 0 to 1 , therefore interpolating from a reliable to a Poisson ( highly variable ) spike train . Jitters for Model SDJ do not affect the spike count and thus neither the Fano factor . For model R , the Fano factor for the spike count of any input is , since each spike train is generated using a single inhomogeneous Poisson process . Note that these Fano factors correspond to continuous time . The finite time step used in discrete-time simulations has an effect , which we minimize by taking sufficiently small time steps . In Fig . 3Conv , we use the convergence index: ( 25 ) where denotes the floor function and the absolute value . This index is positive and vanishes when all synapses are either maximally reinforced or completely depressed ( i . e . bimodal distribution ) . It was calculated every 50 s and a running average of the last four data points was used in order to remove noise in the plot . We use information theory to quantify how good the postsynaptic neuron is at detecting the beginning pattern after convergence ( section “General case of an arbitrary pattern” ) . Detection is considered to be successful when the neuron fires at least two times at the beginning of the stimulus presentation . Specifically , we discretize time into 25 ms bins . Each of those bins could either correspond to the first 25 ms of the pattern ( or stimulus ) , case referred to as , or not ( ) , and could contain at least 2 postsynaptic spikes ( ) or not ( ) . Note that for the LIF neuron the time window considered for counting spikes was actually shifted 10 ms backward since , because of , the potential rise actually starts before the theoretical start of the pattern presentation , which may shift backward the postsynaptic spikes signaling the pattern . The Poisson neuron does not require this because the rate increases mostly during the pattern presentation , and eventual preceding spikes have no influence . For both neurons the mutual information between the postsynaptic response and the presence of the stimulus is: ( 26 ) Note that in signal detection terms , the first term corresponds to “hits” , the second to “misses” , the third to “false alarms” , and the last one to “correct rejections” . A perfect detector would lead to , and . An upper bound on the mutual information is given by: ( 27 ) Details can be found in Equation ( S31 ) in Text S1 .
We focus on the situation where the pattern is difficult to learn and detect . Namely , relatively short duration and low presentation frequency imply that the mean firing rate ( averaged over hundreds of milliseconds ) is roughly the same across all afferents irrespective of the pattern presentations ( , which further requires for model R ) . For the example of model S , our choice of parameters implies that the discrepancies in the mean rate across inputs typically correspond to , synonymous with very low signal-to-noise ratio as far as learning is concerned . In this case , neuronal specialization cannot be achieved relying on a simple rate difference . The key to learn such patterns consists in extracting information contained at a fine timescale: the repetitive pattern presentations induce spike-time correlations between pattern inputs ( see Materials and Methods ) . Using our theoretical framework , we can derive a common equation for the cross-covariance between inputs and that is valid for all models S ( including SJ ) and R: ( 28 ) which is a sum of Gaussian kernel functions defined in Equation ( 15 ) centered around the latencies that determine the pattern . The ( time-averaged ) cross-correlogram in Fig . 1 ( right panels ) corresponding to contains information about the spike timing ( model S ) or spike distribution ( model R ) within the pattern , as illustrated by Fig . 1 . Both the relative latency positions and temporal resolution of the pattern determine peaks in the correlogram . For model SD where pattern spikes occur with a given probability at each presentation , it is clear from Equation ( 1 ) that a pattern with a frequency ( with ) and occurrence probability has the same correlogram as the same pattern with reliable spikes ( ) and frequency ( Fig . 1S2 and SD2 with ) . This means that non reliability of firing can be compensated by more frequent repetition , as far as spike-time correlations are concerned . For model SD , the Fano factor of the spike count scales between ( reliable ) for to ( Poisson ) for ( see Equation ( 24 ) in Materials and Methods ) and the influence of such a variability on learning will be examined . On the other hand , model R always gives as any Poisson process does . For both models SJ and R , the temporal precision can be varied through . In particular , with a same value for , they may only differ by the Fano factor . Fig . 1SJ2 and 1R2 show that Gaussian jitters for model S and Gaussian-peaked rate modulations for model R lead to similar correlograms when they have the same temporal accuracy . By comparing results for these models , we will assess the effect of the Fano factor on both learning and detection afterwards . Now we examine how this spiking information induced by pattern presentation can be captured by STDP . Our theoretical analysis is based on the additive STDP rule and Poisson neuron model , which allow the prediction of the weight specialization induced by STDP [4] , [27] . In our phenomenological model of synaptic plasticity , STDP is described by a temporal learning window ( Fig . 2A ) . The Poisson neuron linearly sums excitatory postsynaptic potentials ( EPSPs ) resulting from each incoming spike to determine the soma potential , which is used as an instantaneous rate function to generate an output spike train . The theoretical predictions will be verified numerically using both Poisson and leaky integrate-and-fire ( LIF ) neurons , and the effect of weight dependence in the STDP rule will be also examined . Details are provided in Materials and Methods . The first stochastic moment for the plastic weight follows a differential equation that is of the form: ( 29 ) cf . Materials and Methods before Equation ( 10 ) . Here is a function of the mean ( time-averaged ) firing rates and of the -th afferent and neuron , respectively; see Equation ( 6 ) for the detailed expression . The coefficients of matrix defined in Equation ( 11 ) involve the spike-time cross-correlations between input spike trains and . Note that for the pattern models S and R , the input firing rates and spike-time cross-correlograms are time invariant; the output neuronal firing rate only evolves as the consequence of the learning process . This dichotomy between mean firing rates on the one hand and spike-time correlation coefficients on the other hand highlights the separation of timescales in representing the information contained in spike trains [4] , [27] . In our framework , the firing rates and are low-pass filtered variables , but spiking information at a short timescale is still contained in the time-averaged spike-time correlations through the time variable ( see Equation ( 9 ) in Materials and Methods ) . Consequently , the weight evolution can be analyzed as a double dynamics that is a mixture of: A typical example is illustrated in Fig . S1 in Text S1 . Specialization leads to the potentiation of some weights at the expense of others , then the neuron behaves as a coincidence detector for the selected inputs . The weight selection can be predicted via the matrix , which is tractable when using the Poisson neuron model . Because of the similarity between the correlation structures that they induce , cf . Equation ( 28 ) , a common expression for the coefficients that appear in Equation ( 29 ) can be derived both for models S and R . It relies on the difference between the latencies and of inputs and , namely ( 30 ) Model S corresponds to ( in addition to for models S and SD , but not SJ ) and model R to . The kernel function is defined by ( 31 ) Note that , which is represented in Fig . 2A . As shown in Fig . 2B , the more spread the Gaussians are ( viz . larger value for ) , the smoother and smaller the function is as a function of the time difference . From the Equation ( 31 ) , it can be seen that the STDP effects are impaired if the STDP time constants are much smaller than those of the PSP response . This is not the case in the brain ( and in our simulations ) , where all these constants are in the 10–30 ms range . The spectral analysis mentioned above describes how an initially homogeneous weight distribution will begin to split as a result of STDP . The strength of cross-covariances between pattern inputs reflects their tendency to drive the firing of output spikes . When some weights become stronger compared to the remainder , this driving effect becomes stronger . It follows that the cross-covariances between the potentiated inputs and the postsynaptic neuron increase in a causal manner , in turn inducing further ( and stronger for additive STDP ) potentiation . This self-reinforcing mechanism ( described in more detail in Materials and Methods ) leads to a clear potentiation of some pattern inputs , until they either saturate at the upper bound ( additive STDP ) or reach an equilibrium value ( weight-dependent STDP ) . Competition between weights leads to the reinforcement of those with stronger correlation term , so the rule of thumb is that inputs with larger coefficients ( for all indices ) will be selected by STDP . Because is roughly antisymmetric and we do not consider slow synapses , in Equation ( 30 ) is such that negative arguments contribute positively to the sum in coefficients . It follows that pattern synapses with early spikes tend to be selected when the spike density is somehow constant for all pattern inputs ( see Materials and Methods ) . When some strong inputs start driving the neuronal firing ( as mentioned above ) , other pattern inputs corresponding to earlier spikes also tend to be reinforced for temporally Hebbian STDP , which further favors inputs with early spikes until the weight strengthening finally stabilizes . This is illustrated for models S and R by simulations using Poisson neurons in Fig . 3S1 , SD1 , SJ1 and R1 ( black clusters ) . Non-pattern synapses have also been completely depressed by STDP . While keeping the output neuronal firing rate low on average ( due to the homeostatic equilibrium ) , the potentiated synapses ( due to weight competition ) ensure a significant increase of the membrane potential ( solid curve ) at the beginning of each pattern presentation . This usually causes early postsynaptic spikes ( vertical dashed lines ) that can be used for fast pattern detection ( quantification with mutual information will be discussed later ) . The trained neuron is thus selective to the quasi-simultaneous arrival of the earliest pattern spikes , and can serve as “earliest predictor” of the subsequent spike events , at the risk of triggering a false alarm if these subsequent events do not occur , but with the benefit of being very reactive ( to learn the full pattern , several neurons in competition can be used [37] ) . Comparatively , the membrane potential before training ( dotted curve ) is similar during and outside pattern presentation . Model S with , Hz ( and ms ) , or with , Hz ( and still ms ) have the same correlation structure ( Fig . 1S2 and SD2 ) . Therefore , despite different Fano factors , they lead to the same final weights ( Fig . 3S1 and SD1 insets ) , at approximately the same speed ( Fig . 3Conv ) . However , after learning the increase of the summed EPSPs ( solid curve ) is stronger in the first case ( Fig . 3S2 ) than in the second ( Fig . 3SD2 ) because of the missing spikes . Likewise , model SJ with ms ( and , Hz ) and model R with ms ( and still Hz ) have similar correlation structures ( Fig . 1 SJ2 and R2 ) and thus lead to the same final weights ( Fig . 3SJ1 and R1 insets ) . Here too , they have approximately the same learning speed ( Fig . 3Conv ) despite the difference in their Fano factors . The use of ms induces a more spread rise of EPSPs , comparable in both cases ( Fig . 3SJ2 and R2 ) . Indeed , with model R the deviations from the mean spike counts at each pattern presentation tend to compensate across the n≈70 selected afferents ( with Poisson processes like here , the total spike count's coefficient of variation decreases in ) . Since the EPSP rise is more spread for model SJ and R than with model S , the detection performance is poorer , but it remains acceptable , even with a decision rule based only on the neuron's spiking output ( this will be discussed later ) . In our simulations spike count variability ( as measured by the Fano factor ) does not slow down the learning , whereas spiking temporal imprecision ( related to ) does . In terms of convergence speed , we have S∼SD>SJ∼R ( see Fig . 3Conv ) . A LIF neuron trained with model R ( Fig . 3RL ) also selects pattern synapses with early spikes , which results here in two postsynaptic spikes each time the pattern is presented . With our choice of parameters learning was found to be slower for the LIF neuron ( Fig . 3Conv ) . Note that , for model S with other parameters , selectivity can emerge in a few tens of pattern presentation [9] . The fact that the LIF neuron behaves similarly to the Poisson neuron justifies a posteriori the use of the latter for convenience in the theoretical analysis . In a regime where the LIF neuron is sensitive to volleys of almost coincident spikes , only the inputs with the very first spikes may remain potentiated at the end of the learning epoch , whereas the synapses corresponding to later spikes are depressed [9] . However , a thorough discussion of these effects is beyond the scope of the present paper . Here we focus on the Poisson neuron , for which new analytical results are presented below . Extensive numerical studies with the LIF neuron can be found in previous work [9] , [37] , [38] . The above-mentioned rule of thumb that inputs with early spikes are favored does not hold for all patterns . To illustrate how the spike distribution among pattern inputs affects the weight evolution , we use a specific configuration of models S and R , where pattern afferents are partitioned into two groups ( or populations ) with respective numbers of inputs and . More precisely , this bimodal distribution of input latencies is such that the afferents of group 1 tend to fire before those in group 2: The overline indicates group variables . In this way , we control the crucial parameters that determine the clustering of spikes within the early and late groups , namely their sizes and temporal resolutions , whose effect will be assessed against the difference between their latencies . In terms of population averages , both models have the same expression for the input spike-time correlations given in Equations ( 16 ) and ( 17 ) with respective mean latency and temporal spread , as illustrated in Fig . 4SB2 and 4RB2 . This illustrates another connection between these two pattern models despite their different spike generation mechanisms . Note that , for model R , the amplitudes and the number of inputs with clustered pattern spikes ( e . g . group size for model RB ) plays a similar role . We consider the situation where the inputs from the first group fire sufficiently early compared to the second group ( say ms ) . Our framework allows us to study the effect of the pattern parameters on the resulting competition between the two groups; detailed calculations are provided in Text S1 ( Section S2 . 2 ) . When both groups have similar size ( ) and spread width ( ) , STDP tends to select the first group in agreement to the previous section , see Fig . 4SB3 , 6 and 4RB3 , 6 . However , when the second group is more populated ( ) , STDP preferably selects the late group as shown in Fig . 4SB4 , 7 and 4RB4 , 7 . Now , when the second group has a narrower spread ( ) , while both groups have comparable size ( ) , STDP tends to select the second group as shown in Fig . 4SB5 , 8 and 4RB5 , 8 . This extends previous results on the effect of correlation spread for two input groups that have no correlation between them [39] . Simulations with models SB and RB exhibit similar trends , in agreement with the resemblance between their ( population averaged ) spike-time correlograms . In summary , potentiation of synapses corresponding to early pattern spikes competes with another trend that favors densely populated and narrow spike clusters , irrespective of the spike generation type within the pattern . Note that this does not affect the success of pattern detection , but only its timing . Now we go back to the case of a general pattern that has arbitrary latencies . We examine how the trends revealed by the analytical study of models SB and RB adapt here . A complete description of the weight evolution involves the spectrum of matrix , as Equation ( 28 ) can be rewritten as a linear differential matrix equation of the form ( see Equation ( 10 ) in Materials and Methods ) . The spectrum of this matrix ( circles ) is represented in Fig . 5ABC for a pattern of model R: most eigenvalues are close to those of ( pluses ) , which can thus be used to predict the weight evolution; the large negative eigenvalue of matrix roughly equal to and associated with the homeostatic equilibrium is not displayed there for clarity . Note that eigenvalues may be complex numbers . Likewise , the dominant left eigenvector ( s ) that determine the weight specialization are similar for both matrices ( Fig . 5DEF ) . As shown in Fig . 5GH , the weight evolution can be satisfactorily predicted using either the whole matrix ( label ‘whole A’ ) or its principal eigenvectors ( label ‘princ eig vect A’ ) both for models S and R . Note that our predictions slightly overestimate the number of potentiated weights here ( as in Fig . S1 , in Text S1 ) . Neglecting rate effects , more than 80% of the potentiated weight can be predicted ( label ‘A only’ ) , meaning that spike-time correlations dominate . In a similar manner to the simpler model RB when a pattern of model R has more spread rate peaks , the elements of go to zero ( Fig . 5ABC ) as the function becomes flatter when increases ( Fig . 2B ) . Consequently , the weight specialization weakens , which significantly decreases the quality of detection , as estimated by the mutual information defined in Equation ( 25 ) and plotted in Fig . 6 . With the same spread width , Gaussian jitters ( Fig . 6SJ ) and Gaussian rate peaks ( Fig . 6R ) give similar performance: because sufficiently many inputs are used spiking variability of model R ( within the pattern presentation ) hardly impairs the detection w . r . t . Model SJ with ( for ) . Together with results comparing models SB and RB in Fig . 4 , this supports our conclusion that the weight evolution induced by STDP is mainly determined by the input spike-time cross-correlograms , and that unreliability in the spike count ( as measured by the Fano Factor ) has little effect . Now , as an illustration of an arbitrary pattern that mimics real PSTHs , we consider a longer rate-modulated pattern ( ms ) . We assume that the inputs are modulated by some global strength ( e . g . image or movie contrast , sound volume ) , while individual rates encode local features . For this purpose , the rate functions are multiplied both for pattern and non-pattern inputs by a common envelope . The latter corresponds to superimposed 20-ms spread peaks distributed in time with a homogeneous Poisson process at 20 Hz , as illustrated in Fig . 7AB . For all inputs , the envelope is repeated identically during all pattern presentations . In our simulation , STDP selects a single cluster of rate peaks in the pattern ( e . g . third peak in Fig . 7CD ) and mainly reinforces inputs corresponding to that “population” peak . In agreement to the results for the simpler model RB , STDP favors clustered , high-amplitude and narrow modulations . Interestingly , even though the population firing rate in Fig . 7B may be much larger outside than during pattern presentations , the neuron fires almost only during the presentations ( Fig . 7D ) . To further evaluate the relative strengths of the mean input firing rates and spike-time correlations induced by the pattern , we also simulated a pattern with a lower firing rate than the noisy activity surrounding it: 15 Hz vs . 25 Hz , respectively . The output neuron could still be trained successfully , although not as well as for the control case of 20 Hz vs . 20 Hz ( not shown here ) . These results show that STDP can learn non-flat rate-modulated patterns relying on the same competition between afferent weights as described above for the more difficult case of quasi constant input firing rate . Realistic PSTHs can thus be learned provided sufficiently many inputs exhibit significant rate modulations on a “fast” time scale , namely ranging up to 10–20 ms with our parameters . In this case , correlation effects dominate the weight dynamics . Finally , we examine the influence of the STDP and neuronal parameters on both learning and detection afterwards for a typical pattern of model R ( similar results were obtained with model S ) . To estimate the quality of detection , we evaluate the neuronal response after training averaged over 10 pattern presentations ( in order to remove noise and display clear trends ) . We keep in mind that the goal is detection of each pattern presentation by the neuron , though . In our model , the mean output firing rate of the neuron is constrained by STDP close to an equilibrium value that depends on the rate-based contributions and , the ratio LTP/LTD ( related to ) , and the background excitation/inhibition ( cf . Equation ( 4 ) ) . For a good detection performance , in other words minimizing false alarms , in Equation ( 20 ) must be kept low , which can be achieved using a small value for , as illustrated in Fig . 8A . However , such a small value also decreases the magnitude of the response to the pattern . The use of and prevents the pitfalls where all synapses become depressed , leading to a quiescent neuron , or on the contrary maximally reinforced , leading to a very active non selective neuron . They also allow the use of asymmetric STDP where overall depression is stronger than the overall potentiation ( ) , relieving the requirement to tune depression only slightly higher than potentiation [9] , as illustrated in Fig . 8B where the ratio between depression and potentiation for STDP hardly affects the detection performance . We also looked at the effects of weight dependence using the model proposed in [6] , for which a parameter scales between additive ( ) and multiplicative ( ) STDP ( Fig . 8C ) . These simulations were done without the homeostatic terms and , aiming to show that learning is still possible without them . However , we had to fine tune some parameters: ( vs 0 . 1215 in the baseline simulation , cf . Text S1 Section S3 . 2 ) , ( vs 0 . 82 in the baseline ) . Convergence took about 4000 s ( vs 2000 s in the baseline ) . With these values learning fails with pure additive STDP ( ) , leads to good detection performance with slightly multiplicative STDP ( ) . This performance decreases with more multiplicative STDP ( ) , and collapses with , which leads to a unimodal final weight distribution . In comparison , additive STDP , also with , could hardly learn and detect patterns of type SJ with jitters larger than 2 ms [9] . Small values for in Fig . 8C ensure both homeostasis and strong competition for weight-dependent STDP [33] , which results in good detection for a jitter of 10 ms . Inhibitory background activity ( ) is used to obtain an equilibrium with both a low mean firing rate and a positive mean input weight ( see Equations ( 20 ) and ( 19 ) ) . As mentioned above , the first ensures that the trained Poisson neuron has a low number of false alarms . The second leads to successful detection with sufficiently many potentiated pattern inputs that imply a significant increase of the soma potential during each pattern presentation . In contrast , the case corresponds to a noisy neuron and the pattern response is then not so strong compared to , weakening detection as illustrated in Fig . 8D ( ) . In this case , inhibition helps to increase the sensitivity of the neuron to correlated inputs , both for learning and detection . This effect , together with that of , is predicted by the analytical evaluation of the mutual information in Text S1 ( Section S2 . 3 ) . We also verified the robustness of our scheme with respect to the number of pattern afferents . So far , we have used , which ensured that STDP could choose among many afferent candidates . Decreasing weakens the neuronal response , but down to as few as , the detection remains acceptable; only for the performance collapses ( Fig . 8E ) . Successful detection can be achieved provided there are sufficiently many pattern inputs such that the potentiation of “half” of them ( while depressing the others ) leads to a significant response of the soma potential . Last , Fig . 8F examines pattern of various durations ( between 3 and 100 ms ) . When , the pattern inputs behave as a narrowly correlated group which is selected by STDP [4] , but , not sufficiently many inputs have a rate peak in the pattern , so performance decreases . Increasing beyond the time scale of STDP does not significantly change detection , since only a portion of the pattern is learned ( e . g . the beginning in Fig . 8F ) . Patterns of duration ms fully use the temporal ( approximate ) antisymmetry of the STDP learning window . Taken together , these results show that the proposed STDP-based pattern learning/recognition mechanism works for a broad range of parameters . It also works similarly for Poisson and LIF neurons , which supports the theory that the proposed learning scheme relies on the STDP qualitative properties , namely the temporal antisymmetry of the learning window , rather than on a precise quantitative configuration or neuronal model .
A previous simulation study showed that a repeating arbitrary spatiotemporal spike pattern ( model S ) hidden in continuous equally dense distractor spike trains can be robustly detected and learned by a single neuron equipped with STDP [9] Here we have demonstrated that these results extend to the case of patterns of temporally modulated instantaneous firing rates ( model R ) , from which the spikes are generated at each presentation , e . g . , through an inhomogeneous Poisson process . To gain analytical insight , we developed a theoretical framework , which extends previous studies that showed how STDP favors correlated inputs [4] , [3] , [7] , [6] . We confirmed that Hebbian STDP ( cf . Fig . 2A ) tends to favor synapses corresponding to early spikes in the pattern , generalizing previous results on spike patterns [9] . An interesting corollary is that the neuronal response thus becomes faster presentation after presentation . However , this rule is not general when considering specific spike distributions within the pattern , both for spike and rate-modulated patterns . As summarized in Fig . 4 , inhomogeneities of the spike density also play an important role in the selection process , with a preference for densest and narrowest peaks , similar to spike “waves” for synchrony detection . In any case , what is important for successful detection is that some “nearly” coincident pattern inputs are strongly potentiated compared to the others . Whether the selected inputs correspond to an early part of the pattern or not changes only the response timing . These results are robust and hold for a broad range of STDP , neuronal and pattern parameters ( see Fig . 8 ) . The scheme proposed here draws on the approximate temporal antisymmetry of Hebbian STDP and its timescale . Changing the details of the STDP learning window ( see Fig . 2A ) does not compromise the detection , even though the learning dynamics is quantitatively affected . A larger time constant for STDP can be used to learn patterns with larger temporal spreads than those considered here . Even when input firing rates significantly vary over time ( cf . Fig . 7 ) or are inhomogeneous among inputs , spike effects tend to dominate rate effects in the learning dynamics to determine the neuronal specialization , provided sufficiently many inputs are involved . Similar results can be expected for general input spike trains , provided the firing rates and spike-time correlations are well defined ( cf . Equations ( 6 ) and ( 9 ) ) . In the brain , homeostatic synaptic scaling mechanisms ensure that firing rates remain in a suitable range [29] . As in previous studies [4] , [39] , [27] , we modeled those mechanisms with two additive terms occurring whenever an input spike is received ( ) or an output spike is emitted ( ) . However , even though they increase the learning robustness , simulations show that those terms are not required for its success ( Fig . 8C , [9] ) . Provided a single pattern keeps being presented , the emerged weight structure is preserved . However , changing the pattern presentation may cause the neuron to forget its specialization . For most weight-dependent models that produce a unimodal ( unspecialized ) weight distribution for uncorrelated inputs , the structure will be forgotten when the presentation frequency is too low [33] . Additive-like STDP rules can preserve a bimodal weight distribution over a long period even though inputs become uncorrelated [9] , although this behavior is somewhat parameter dependent [27] , [40] . When several patterns are presented , they compete to determine the emerging weight structure [37] and those with the strongest correlations dominate this competition . If one does not clearly dominate the others , the STDP-induced noise ( due to weights jumps ) may cause the neuron to switch its specialization between patterns of comparable correlation strength . This happens , for example , when the learning rate is sufficiently large [41] . One can distinguish two kinds of response variability , or lack thereof: reliability and precision [42] . When a neuron fires approximately the same number of spikes on each trial , it is said to be reliable , whereas when the spikes occur almost at the same time across trials it is said to be precise . Our study demonstrates that STDP-based pattern learning needs a precision of 10–20 ms , whereas it is almost insensitive to a lack of reliability ( see Section S1 . 6 and Fig . S2 and S3 in Text S1 for the case ) , provided the patterns involve at least ∼100 afferents , which is very probable as a typical cortical neuron has about 10 , 000 afferents . In particular , it can cope with the Poisson-like variability often observed experimentally [43] , [44] . Evidence for neural activity with the required precision abounds , at least in sensory systems exposed to dynamical stimuli [18] , [19] , [20] , [21] , [23] , [22] and , even more importantly , to naturalistic stimuli [24] , [25] , [26] . Moreover , for slowly varying or even static stimuli , learning with STDP may be still possible when neural activity oscillations are able to attribute to spikes stimulus-specific preferred phases [38] . Importantly , the mechanism we propose does not need an external time reference such as a stimulus onset . Patterns are learned and recognized is a clock-free system , as they are embedded in random but similar spiking activity . This is possible because STDP only relies on relative timing . A low absolute time precision with respect to the stimulus onset ( e . g . , estimated by a PSTH ) does not preclude far more precise relative latencies , in agreement with experimental observations [45] , [46] , [47] , [25] , [48] For example , this is the case with trial-dependent input fluctuations ( e . g . , correlated noise ) that affects spike times similarly . At a broader level , the rate-modulated patterns considered here can account for a large class of the naturalistic temporal stimuli [17] that a subject experiences in vision , audition , touch or multimodal integration , and we demonstrated that STDP genuinely enables learning and recognition for such patterns . This suggests an important role of this plasticity rule in the development and learning of receptive fields , in perceptual learning , which typically involves many repeated trials [49] and , beyond sensory processing , in other cognitive processes or tasks with fast time scales ( tens of ms ) . Importantly , the relatively low precision and high firing variability considered here correspond to a much weaker assumption than models of spike trains generally used with STDP , and the present results reconcile STDP with experimentally observed spike trains . A recent study has proposed a general scheme for the derivation of a linear-nonlinear Poisson ( LNP ) cascade model to reproduce the dynamics of a spiking neuron [38] In short , a stochastically firing LNP neuron model describes how the input current is filtered by the synapses and dendrites ( linear part ) , and then transformed to calculate the instantaneous rate ( nonlinear part ) . The Poisson neuron used here is a particular and simple case in the class of LNP models . In particular , we have not considered here with the Poisson neuron a linear filter ( described by the PSP kernel ) that diverges at the time origin , which is necessary to match the LIF dynamics . Further analysis using the LNP model could help to understand in more depth the general agreement and specific discrepancies between the LIF and the Poisson neurons in our scheme . For biological realism , a desirable extension of the present framework consists in the learning of a pattern ensemble with a recurrent neural network . This can be achieved with a simple connectivity scheme where afferent connections are trained similarly to the single neuron case: a differentiated specialization can be achieved using lateral inhibition , which prevents neurons from learning the same pattern features [37] In the case of a general recurrent network with plastic synapses , the learning and the neuronal dynamics are intricately coupled and new behavior will certainly appear [50] For a particular network configuration with plastic recurrent synapses , a numerical proof that STDP can successfully learn and retrieve spike patterns ( model S ) has been given [51] A layered network with both feed-forward and lateral STDP-plastic connections was shown to reliably transmit volleys of almost synchronous spikes by means of delay selection [52] with jitters up to 20 ms . Our results are consistent with that previous numerical study in the sense that STDP tends to select among all incoming connections those corresponding to denser clusters of spikes . This is indeed synonymous with a higher synchronization among the selected inputs , and because neurons within each layer have lateral connections , they tend to synchronize their output spike volley . A recurrent network where STDP operates can also generate spontaneous and repeating patterned activity [53] , [54] , [55] Our findings also suggest that similar results may be obtained with rate-modulated patterns ( model R ) . Beyond these particular results , the existence and the properties of a mapping between input patterns and output patterns of interconnected neurons whose connections are modified by STDP remains unknown . For example , the ongoing activity in a network as observed in the brain ( e . g . , due to background inputs ) can clearly affects the “signal-to-noise ratio” of pattern activity and should be taken into account . However , the present analysis is a necessary prerequisite to identify some important dynamical ingredients that may allow a network to learn , retrieve and generate a pattern ensemble . | In vivo neural responses to stimuli are known to have a lot of variability across trials . If the same number of spikes is emitted from trial to trial , the neuron is said to be reliable . If the timing of such spikes is roughly preserved across trials , the neuron is said to be precise . Here we demonstrate both analytically and numerically that the well-established Hebbian learning rule of spike-timing-dependent plasticity ( STDP ) can learn response patterns despite relatively low reliability ( Poisson-like variability ) and low temporal precision ( 10–20 ms ) . These features are in line with many experimental observations , in which a poststimulus time histogram ( PSTH ) is evaluated over multiple trials . In our model , however , information is extracted from the relative spike times between afferents without the need of an absolute reference time , such as a stimulus onset . Relevantly , recent experiments show that relative timing is often more informative than the absolute timing . Furthermore , the scope of application for our study is not restricted to sensory systems . Taken together , our results suggest a fine temporal resolution for the neural code , and that STDP is an appropriate candidate for encoding and decoding such activity . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"computer",
"science",
"mathematics",
"applied",
"mathematics",
"biology",
"computerized",
"simulations",
"neuroscience"
] | 2011 | STDP Allows Fast Rate-Modulated Coding with Poisson-Like Spike Trains |
Correlated inter-domain motions in proteins can mediate fundamental biochemical processes such as signal transduction and allostery . Here we characterize at structural level the inter-domain coupling in a multidomain enzyme , Adenylate Kinase ( AK ) , using computational methods that exploit the shape information encoded in residual dipolar couplings ( RDCs ) measured under steric alignment by nuclear magnetic resonance ( NMR ) . We find experimental evidence for a multi-state equilibrium distribution along the opening/closing pathway of Adenylate Kinase , previously proposed from computational work , in which inter-domain interactions disfavour states where only the AMP binding domain is closed . In summary , we provide a robust experimental technique for study of allosteric regulation in AK and other enzymes .
Conformational heterogeneity as a consequence of dynamics is an intrinsic feature of proteins linked to biological function . [1] An important aspect for our understanding of protein dynamics is a molecular characterization of the structural states that are populated and of how correlated conformational changes can mediate biological functions such as allostery and signal transduction . [2] , [3] The characterization of correlated conformational changes requires a description of how local structural heterogeneity translates into global conformational changes through collective motions . Recent developments in the analysis of various NMR parameters at atomic resolution , such as spin relaxation rates[4] and residual dipolar couplings ( RDCs ) , [5]–[7] have enabled the description of local structural heterogeneity . However , inferences of collective conformational changes from local correlated motions have relied on force fields or motional models . [7]–[9] RDCs are NMR parameters that report on both the local and global structural properties of weakly aligned macromolecules and , under the assumption that alignment does not alter the properties of the protein , can be used to study the amplitude of dynamics , especially when combined with simulations . [5] , [10] Alignment can be induced by steric and/or electrostatic interactions of the macromolecule with external media . Specifically , for a given conformation , RDCs depend on the geometrical properties of the environment of the nuclei in the molecular frame and on the direction and degree of alignment[11] , [12] ( Eq . 1 , Fig . 1 ) . ( 1 ) In Eq . 1 , for two nuclei P and Q , φiPQ is the angle between the inter-nuclear vector and axis i of the molecular frame , Sij is an element of the alignment tensor ( S ) , rPQ is the distance between P and Q , γX is the gyromagnetic ratio of nucleus X , µ0 is the magnetic susceptibility of vacuum and h is Planck's constant . The alignment tensor S is given by the alignment mechanism and , under steric alignment conditions , can be computed accurately and is closely related to the shape of the aligned macromolecule . [12]–[14] RDCs measured in steric alignment combine the local information contained in NMR parameters with the shape information contained in relaxation rates[15]–[17] and small angle X-ray scattering ( SAXS ) . [18] Since inter-domain motions alter the shape of proteins RDCs measured in steric alignment have , as we will show , the potential to act as reporters of inter-domain structural heterogeneity . RDCs can be used to study the structural heterogeneity of globular and disordered proteins by analysing the effect of ( sub-ms ) motional averaging on this NMR parameter . The effect of fast ( sub-ns ) local motions that do not directly alter the magnitude and main directions of alignment can be analysed in the molecular frame defined by the alignment tensor of the average structure . [19] , [20] The analysis of motions that change the shape of the protein , that are typically slower than the timescale of alignment ( 0 . 5 to 5 ns ) [21] needs on the contrary to explicitly consider that the various conformations in fast exchange that contribute to the measured RDC can have different alignment tensors . [20] , [22] Here we show that for a molecule undergoing conformational changes in two distal sites the RDCs depend on the degree of correlation of such changes . We exploit this to characterize the degree of correlation of the inter-domain conformational changes occurring in the substrate-free state of E . coli Adenylate Kinase ( AKe ) , an enzyme that undergoes conformational changes involving shape changes .
Our approach is based on the determination of ensembles that collectively agree with RDCs . The generation of the ensemble is divided in two steps ( see Methods and Supporting Information ) . First , an enumeration ( ca . 105 ) of inter-domain orientations is performed using unrestrained simulations . Secondly , the ensemble of minimal size that best fits the RDCs is identified by a genetic algorithm ( see Figs . S1 and S2 ) . [23] To maximize the coverage of conformational space in the first step we used PELE , an all atom Monte Carlo simulation algorithm with a move set designed to explore normal modes ( see Methods and Supporting Information ) . [24] The RDCs of each conformer are calculated via Eq . 1 using two independent methods to compute the S from knowledge of the structure . [12] , [13] In this case we used only the steric tensor because it is related to shape but it is principle possible to use also the electrostatic tensor , particularly when conformational changes modulate the charge distribution . AKe is an essential enzyme that catalyses the reversible conversion of ATP and AMP into two ADP molecules . The role of AKe is to maintain the energy balance in the cell , which is essential . AKe is a modular enzyme composed of three sub-domains: a CORE domain responsible for thermal stability[25] , [26] and two flexible substrate-binding domains referred to as LID and AMPbd ( Fig 2a ) . Crystallographic structures of AKe have been solved for the open ( inactive ) and closed ( active ) states . [27] Substrate-free AKe has been shown to sample a closed-like state using single-molecule Föster resonance energy transfer ( smFRET ) experiments . [28] , [29] We previously reported an analysis of AKe using RDCs in the substrate-free ( open ) and substrate-bound ( closed ) states where we fit the RDCs to the corresponding crystallographic structures . [30] We found that the fit of the closed state was better than that of the free state , which we interpreted as evidence for inter-domain dynamics in the substrate-free enzyme . To identify major conformational states sampled by substrate-free AKe we used backbone NH RDCs measured in steric alignment to select ensembles of conformations from a pool derived from simulations started from the open and closed X-ray structures[27] , [31] ( see Methods ) . The ensembles collectively agree with the RDCs ( Q≈0 . 26 ) even though the structures they contain do not ( Q>0 . 6 ) when considered individually . An ensemble size of 4 to 64 accounted for the data and equivalent distributions were observed regardless of ensemble size and method used to calculate the RDCs ( Figs S3 to S5 ) . It is worth noting that the LID domain of AKe is in equilibrium with a locally unfolded state with a population of ca . 5% at 37°C;[32] under the conditions used in this study ( 25°C ) this state has a low population ( <1% ) . Similarly , cracking of the AMPbd along the closing mechanism may play a role . [26] , [33] However , cracking occurs at the transition state , and therefore has a very low population . The states obtained for AKe are shown in Fig . 2b . An analysis of the results indicated that substrate free AKe samples three major states corresponding to i ) a closed-like state ( θLID ∼105° , θAMPbd ∼60° , population ∼0 . 5±0 . 1 ) , where the LID is closed and the AMPbd slightly opened ii ) the open state ( θLID ∼146° , θAMPbd ∼74° , population ∼0 . 25±0 . 1 ) , where both nucleotide binding domains are open , and iii ) conformations in which the degree of domain opening is intermediate between that of the fully closed and fully open . The estimate of the fraction closed AKe from our analysis is 0 . 5 , which is in agreement with numbers from single molecule FRET studies where fractions of 0 . 6[29] and 0 . 3[28] has been enumerated . We performed additional control simulations to assess the robustness of the distributions obtained . Scrambling the RDCs lead to a list of restraints that could not be fit to any distribution ( Fig . S6 ) , showing that the distribution obtained is encoded in the data and not biased by the population of the conformers in the pool . The distribution shown in Fig . 2a was found to be robust to various sources of error , including errors in the experimental RDCs ( Fig . S7 ) and in the prediction of the alignment tensor ( Fig . S8 ) . We also performed computational experiments to ensure that a single set of RDCs suffices to distinguish between ensembles with a distinct number of states and identify correlated structural changes ( see Figs . S9 to S13 ) . An analysis of the conformational ensemble allowed us to assess the presence and degree of inter-domain correlation , an important property of AKe . As shown in Fig . 2b , the ATP and AMP domain movements are correlated , disfavouring conformations where the LID is open and the AMPbd is closed . Our findings agree with free energy calculations[34] , molecular dynamics simulations[35] as well as with normal mode analysis . [33] , [36] Although it is impossible to derive the closure pathway from equilibrium data , the conformational states observed favour a step-wise mechanism in which LID closure takes place before AMPbd closure . Such a mechanism would be beneficial because the initial closure of the LID , together with the substantially higher affinity of this domain for its substrate ( ATP ) as compared to AMP ( 50 vs 1700 µM ) , [30] would reduce the probability of non-productive binding of AMP to the LID . [37] Single-molecule FRET[28] , [29] and NMR[28] have shown that the open and closed states of AKe are in equilibrium and that nucleotide binding shifts it towards the closed state . [30] Of particular interest are smFRET experiments , as these can resolve conformational states and provide a qualitative validation of the ensemble . Two different reaction coordinates , corresponding to distances between residues in the LID and the AMPbd ( Aquifex AK ) [28] and between residues in the LID and the CORE domain ( AKe ) , [29] have been studied . In the former the authors resolved two states in which open conformations were populated to ca . 70% , whereas in the latter the authors found that the open conformation was disfavoured , with a population of ca . 40% . These results can be rationalized based on the ensemble . Following the approach of Beckstein et al . , [38] where distances corresponding to the open and closed states are mapped in domain angle space , we estimated open populations along the LID-AMPbd and LID-CORE smFRET reaction coordinates >50% and ∼30% , respectively ( for LID-AMPbd large uncertainties are found due to difficulties in assigning open and closed states as quantified by smFRET ) . Overall , the ensemble is able to qualitatively reconcile two apparently contradictory smFRET studies , which also suggests that steric alignment does not significantly alter the structural heterogeneity of AKe . Here we have used the dependence of RDCs on global molecular shape via S to determine ensembles reporting on the amplitude and degree of correlation of inter-domain motions . Although RDCs are uniquely suited for this , this is a challenging endeavour due to their intrinsic degeneracies . [39] It is therefore relevant to discuss the factors that allowed the approach used in this work to alleviate them . The first factor is that our approach accounts for how shape changes alter the value of the RDC by computing S for each inter-domain orientation i . e . two inter-domain orientations that would have degenerate RDCs when the tensor is assumed to be constant can be differentiated if they have sufficiently large differences in shape ( Fig . S14 ) . [39] A second factor is the presence of structural constraints due to the covalent linkage , where steric clashes between domains restrict the available conformational space;[40] , [41] AKe is a well-structured proteins where this effect is particularly important but for multi-domain proteins with flexible linker sequences RDCs measured in multiple alignment media will be required . Alternatively , RDCs could be complemented with structural restraints derived from smFRET or SAXS data . Correlated motions in proteins are thought to mediate biochemical functionality . [42] Recently , we have identified weak long-range correlated motions in a surface patch of ubiquitin involved in molecular recognition . [7] Here , we move a step forward and find correlated domain movements in a representative multi-domain enzyme . The observed inter-domain correlation suggests functional roles in allowing ligand access , in adopting the inter-domain orientation necessary for catalysis as well as in binding allostery . Our results , therefore , reinforce that correlated inter-domain motions in proteins can mediate important biochemical processes .
NH RDCs used in this work for free AKe were measured in stretched polyacrylamide gels . [30] Two independent methods , PALES [12] and ALMOND[13] , [13]were used to calculate the alignment tensor using Eq . 1 . They gave equivalent distributions as shown in Figs . S3–S5 . In Figs . 2 we provide the distributions that lead the best agreement between calculated and experimental RDCs [43] ( Tab . S1 ) . For AKe it was obtained using ALMOND . The agreement between calculated and experimental ( or reference ) RDCs was assessed by the quality factor Q [43] defined in Eq . 2 . ( 2 ) The calculated RDCs were scaled , after averaging across the ensemble , to minimize Q to account for the difficulty of predicting the absolute concentration of alignment medium . We used RDCs corresponding to structured regions in the protein ( see Fig . S15 ) . For AKe , we found that RDCs in loop regions were more difficult to fit and therefore we decided to exclude them . However , we note that the ensemble derived including RDCs in the loop regions remains unaffected . We also evaluated the impact of the alignment media by monitoring changes in chemical shifts induced by the presence of the stretched polyacrylamide gel . As shown in the Supporting Figure S16 , no significant chemical shift perturbations were observed when apo AKe was immersed into the anisotropic phase . As a reference the chemical shift perturbations induced by binding of the inhibitor Ap5A to AKe are also displayed . A thorough analysis of the robustness of the procedure to various source of error is provided as Supporting Information . To select the optimum ensemble size we monitored the agreement with RDCs used to guide the genetic algorithm and to RDCs left out of this process or free RDCs . Ten sets of randomly removed RDCs ( 20% ) or free RDCs sets were used and 20 independent calculations were run for each set of randomly removed RDCs . The ensemble was qualitatively validated against two independent smFRET experiments ( see results and discussion ) . An extensive set of inter-domain conformations ( ca . 105 ) was generated for AKe . The X-ray structures 1ake [31] and 4ake [27] were used as seeds . Trajectories were generated where either the AMP or ATP binding domain open ( or closed ) . From each state along the trajectory a second trajectory was generated where the other domain was forced to open ( or to close ) . These simulations were performed with the molecular simulation package CHARMM c35 . [44] A time step of 1 . 0 fs was used . Simulations were performed at 300K . A shape-term ( biasing ) potential on the backbone atoms was used . The CHARMM commands "CONStraint HARmonic BEStfit COMParison" and the IC table tool ( CONStraint IC BOND ANGL IMPR DIHE ) were used to sample inter-domain orientations between open and closed conformations . The initial force was set to 0 . 1 and exponentially increased by a factor of 1 . 05 during 200 cycles of 100 steps each . The degree of opening ranged between ∼30–90° and ∼90–160° for the AMP- and ATP-binding domains ( see angle definitions for AKe ) . The angles sampled covered the range observed by known X-ray structures of AKe and related proteins . [38] An advanced sampling strategy based on the PELE [24] method was used ( see Protein Energy Landscape Exploration , see Supporting Information ) . This allowed identifying high-energy conformations which are unlikely to be of sufficiently low energy to be present in the crystalline state or sampled by conventional MD . An in-house genetic algorithm GA was developed to efficiently search within the pool of structures used to determine the experimental inter-domain orientation distributions of T4L and AKe . Initially 1000 ensembles were generated with structures randomly selected from a pool of ca . 105 conformations ( see reference pool of conformations above ) . Ensembles of size N = 1 , 2 , 4 , 8 , 16 , 32 and 64 were used . The 103 ensembles of size N were submitted to evolution through 1000 steps . At each step two new sets of 103 ensembles were generated by mutation and crossing operations , leaving 3×103 ensembles . At each step the best 103 ensembles , based on the value of Q ( see below ) , were retained . After 1000 iterations the best ensemble was saved and the complete procedure was repeated 200 times . Because too high mutation and recombination rates may lead to loss of good solutions and to premature convergence of the GA , these parameters evolved during the calculations . Initially , the mutation and crossing rates were set to 100% and 2% , respectively . At each step new mutation and crossing rates were determined by dividing/multiplying them by a factor of 1 . 001 , respectively . In Figs . S1 and S2 a scheme illustrating the GA and the convergence of the method are shown . The error in the population of the states observed were determined from the influence of RDC experimental uncertainty by performing 200 calculations with random Gaussian error added to each experimental RDC . We used a standard deviation of 1 . 0 Hz for AKe , three times the estimated experimental error . The definition used by Beckstein et al . [38] was adopted in this work . Briefly , the angle formed between the AMP-binding domain ( AMPbd ) and the core domain ( CORE ) was determined as the angle formed by two vectors: Vector 1 connects the centers of mass ( Cα atoms ) of CORE residues 90–100 with residues 115–125 ( “hinge region” between the CORE and LID domains ) . Vector 2 connects the centers of mass of CORE residues 90–100 and AMPbd residues 35–55 . The angle formed between the ATP-binding domain ( LID ) and the CORE was defined equivalently using the angle formed between two vectors: Vector 1 connects the centers of mass of residues 115–125 ( “hinge region” CORE–LID domains ) with CORE residues 179–185 . Vector 2 connects the centers of mass for residues 115–125 ( “hinge region” CORE–LID ) with LID residues 125–153 . | Most proteins contain several domains , and inter-domain motions play important roles in their biological functions . Describing the various inter-domain orientations that multi-domain proteins adopt at equilibrium is challenging , but key for understanding the relationship between protein structure and function . When more than two domains are present in a protein , correlated domain motions can be of fundamental importance for biological function . This type of behaviour is typical of molecular machines but is extremely challenging to characterize both from experimental and theoretical viewpoints . In this paper , we present a hybrid experimental/computational approach to address this problem by exploiting the information on molecular shape contained in nuclear magnetic resonance experiments to determine accurate conformation ensembles for the multi-domain enzyme adenylate kinase with help of advanced simulation methods . | [
"Abstract",
"Introduction",
"Results",
"and",
"Discussion",
"Methods"
] | [
"computational",
"chemistry",
"molecular",
"dynamics",
"biology",
"and",
"life",
"sciences",
"chemistry",
"physical",
"sciences",
"computational",
"biology",
"molecular",
"mechanics",
"biophysics",
"biophysical",
"simulations"
] | 2014 | Correlated Inter-Domain Motions in Adenylate Kinase |
All positive strand ( +RNA ) viruses of eukaryotes replicate their genomes in association with membranes . The mechanisms of membrane remodeling in infected cells represent attractive targets for designing future therapeutics , but our understanding of this process is very limited . Elements of autophagy and/or the secretory pathway were proposed to be hijacked for building of picornavirus replication organelles . However , even closely related viruses differ significantly in their requirements for components of these pathways . We demonstrate here that infection with diverse picornaviruses rapidly activates import of long chain fatty acids . While in non-infected cells the imported fatty acids are channeled to lipid droplets , in infected cells the synthesis of neutral lipids is shut down and the fatty acids are utilized in highly up-regulated phosphatidylcholine synthesis . Thus the replication organelles are likely built from de novo synthesized membrane material , rather than from the remodeled pre-existing membranes . We show that activation of fatty acid import is linked to the up-regulation of cellular long chain acyl-CoA synthetase activity and identify the long chain acyl-CoA syntheatse3 ( Acsl3 ) as a novel host factor required for polio replication . Poliovirus protein 2A is required to trigger the activation of import of fatty acids independent of its protease activity . Shift in fatty acid import preferences by infected cells results in synthesis of phosphatidylcholines different from those in uninfected cells , arguing that the viral replication organelles possess unique properties compared to the pre-existing membranes . Our data show how poliovirus can change the overall cellular membrane homeostasis by targeting one critical process . They explain earlier observations of increased phospholipid synthesis in infected cells and suggest a simple model of the structural development of the membranous scaffold of replication complexes of picorna-like viruses , that may be relevant for other ( + ) RNA viruses as well .
( + ) RNA viruses of eukaryotes are a very successful group of pathogens infecting organisms from unicellular algae to humans . In spite of adaptation to diverse hosts the basic processes of genome expression and replication are highly conserved among these viruses . One such feature shared among all ( + ) RNA viruses is the association of RNA replication machinery with cellular membranes . It has been proposed that assembly of replication complexes on membranes may facilitate infection in several ways: increase local concentration of viral proteins; provide structural scaffold for assembly of replication machinery; hide viral dsRNA replication intermediates from the cellular innate immunity mechanisms ( reviewed in [1] , [2] ) . Poliovirus ( PV ) is a prototype species of the Picornaviridae family . Its genome RNA of about 7500 nucleotides is directly translated into one polyprotein which is cleaved co- and post-translationally into a dozen of structural and replication proteins . Proteins encoded in the P2-P3 region of the viral genome as well as the intermediate products of the polyprotein processing are responsible for RNA replication . Other members of the Picornaviridae family share the same basic genome organization and expression strategy with minor modifications [3] . PV infection induces rapid development of new membranous agglomerates harboring viral replication complexes . The current models of the development of picornavirus replication structures suggest hijacking of either elements of the cellular secretory pathway or autophagy machinery [4] , [5] , [6] . However even closely related viruses vary greatly in their sensitivity to the inhibitors of the secretory pathway , and effects of manipulation of autophagy may vary even for the same virus [7] , [8] , [9] , suggesting that these cellular processes are not obligatory for the development of replication complexes . At the same time previously accumulated data show that diverse picornaviruses similarly induce strong stimulation of phospholipid biosynthesis , especially phosphatidylcholine ( PC ) , upon infection with [10] , [11] , [12] , [13] . PC constitutes about 50% of the total phospholipid content in eukaryotic membranes [14] . Phospholipids found in cellular membranes include fatty acids ( FAs ) with C16 and longer carbon atoms chains [15] . In mammalian cells fatty acid synthase can de novo synthetize palmitic acid ( C16:0 ) , which can subsequently be processed into other FA species [16] , [17] . However , most of the cells import the majority of long chain FAs from extracellular media . The mechanism of FA transport through plasma membrane is not yet completely understood , however it is believed that acyl-CoA synthetase activity plays a key role in this process . According to the current model of vectorial acylation , long chain FAs as hydrophobic molecules can freely diffuse through lipid bilayers , and inside the cell they are converted into hydrophilic acyl-CoAs that can no longer escape . Indeed most proteins that have been shown to facilitate FA uptake possess acyl-CoA synthetase activity and its inactivation prevented transport of FAs into cells [18] , [19] , [20] , [21] . Thus lipid biosynthesis is intrinsically dependent on acyl-CoA synthetases which activate FAs derived from either external or internal cellular sources . There are 26 different acyl-CoA synthetase genes in human genome [22] . Five members of the long chain acyl-CoA synthetase ( Acsl ) family; six proteins of the very long chain acyl-CoA synthetase family also known as fatty acid transport proteins ( Acsvl or FATP ) ; and two bubblegum acyl-CoA synthetases ( ACSBG ) can activate long chain fatty acids . Their differential tissue expression and sub-cellular localization , existence of multiple splice isoforms , and enzymatic preference towards certain classes of FA provide foundation for complex pattern of uptake and channeling of FA into different metabolic pathways [23] . In this study we show that PV infection results in fast up-regulation of long chain FA uptake due to activation of cellular long chain acyl-CoA synthetase activity , and we identify long chain acyl-CoA synthetase 3 ( Acsl3 ) as a novel host factor required for polio replication . We found that in mock-infected cells the newly-imported FAs are mostly channeled to lipid droplets , while in infected cells they are immediately utilized for highly up-regulated PC synthesis . The infected cells demonstrate preference for import of different FAs than mock-infected cells , resulting in significant changes in the spectrum of PC molecules . The enrichment of phosphatidylcholine species with short palmitoyl ( C16:0 ) moieties likely generates more fluid membranes with intrinsic capacity to assemble into convoluted tubular matrix of the membranous replication organelles . We find that stimulation of FA import requires PV protein 2A , but is independent of its protease activity , thus revealing a new important function this protein plays in alteration of the cell metabolism . The activation of FA import is observed upon infection of diverse picornaviruses in different cell types . Our work explains previous data on stimulation of membrane synthesis and morphology of replication structures in picornavirus-infected cells , and provides a new model of the development of the membranous scaffold of the replication organelles apparently shared by diverse picornaviruses .
The increase of phospholipid synthesis in PV-infected cells [10] , [12] , [13] should be sustained by sufficient supply of corresponding precursors including long chain FAs . To monitor FA import we pulse-labeled PV-infected HeLa cells with a fluorescent fatty acid Bodipy 500/510 C4 , 9 ( bodipy-FA ) which is believed to mimic FA with 18 carbon atoms backbone . This and similar molecules are extensively used in lipid metabolism research and it was previously shown to be rapidly utilized by cellular lipid synthesis machinery and incorporated into phospholipids , triglycerides and other natural lipids [24] , [25] , [26] , [27] , [28] . The cells were infected at a multiplicity of 50 PFU/cell to ensure simultaneous development of infection , and bodipy-FA was added for 30 min at 4 hours post infection ( h p . i . ) , in the middle of the infectious cycle . The infected cells showed strongly increased import of bodipy-FA ( Fig . 1A and B ) . In mock-infected cells the label was distributed into multiple round structures in the cytoplasm and was also found in intracellular ER-like staining ( Fig . 1C , mock ) . The round bright dots were identified as lipid droplets since they co-localized with a well-established lipid droplet marker ADRP [29] ( Fig . 1D ) . Note that some ADRP-positive structures did not accumulate bodipy-FA during 30 min labeling period ( Fig . 1D , arrow ) , consistently with the previous results that individual lipid droplets accumulate newly-synthesized lipids at different rates [30] . In infected cells , bright bodipy-FA fluorescence surrounded the nuclei and often occupied the total cytoplasmic area reflecting robust development of the poliovirus replication complexes ( note pycnotic nuclei in infected cells , characteristic of polio-induced cytopathic effect ) ( Fig . 1E ) . For the experiment shown on Fig . 1 the cells were incubated in serum-free media during the labeling period , so bodipy-FA was the only fatty acid available exogenously . We also monitored FA transport when cells were incubated in normal growth media supplemented with fetal bovine serum which provides ample supply of natural FA and other lipids . As expected , the level of fluorescent signal was lower in the presence of serum , due to competition with the fatty acids from serum , but the overall picture of strong stimulation of fatty acid import upon infection was the same ( not shown ) . Cells on Figure 1C and E are imaged directly after formaldehyde fixation without further detergent permeabilization which we found to deteriorate the fine structure of the distribution of the bodipy-FA label , especially in weakly labeled mock-infected cells . Staining of cells for a viral antigen 2B , a marker of membranous replication complexes , revealed extensive co-localization of bodipy-FA fluorescence with polio replication structures , especially in the cells where viral protein staining could still be visualized as discrete domains in the confocal plain ( Fig . 1F , arrowheads , also co-localization panel ) . With the further development of infection staining for both viral proteins and bodipy-FA tend to occupied all available perinuclear space reflecting massive development of membranous replication complexes . Staining for other membrane-targeted poliovirus replication proteins 2C and 3A revealed similar pattern of distribution of a viral antigen and bodipy-FA label ( not shown ) . Thus in PV-infected cells the import of FAs from media is strongly increased , their intracellular targeting is different from mock-infected cells , and they are used for building of viral membranous replication complexes . To investigate the metabolic targeting of the imported FAs we pulse-labeled cells with bodipy-FA for 30 min at 4 h p . i . , extracted the lipids and resolved them by thin layer chromatography ( TLC ) using solvent systems optimized for separation of either neutral or polar lipids . The chromatograms were first photographed in a fluorescence imager to reveal the newly synthesized lipids , and then developed with conventional stains to visualize the total lipid material . We did not recover noticeable amount of free bodipy-FA . Virtually all the fluorescence was found in newly synthesized complex lipids , thus validating the use of bodipy-FA in our system ( Figure S1 , compare bodipy-FA marker lane 8 on the fluorescent polar lipids panel with the fluorescent lipids extracted from the cells ( lanes 1–4 on the same panel ) ) . There were no dramatic differences due incubation of cells in the presence or absence of serum , but as expected the fluorescent signal recovered from the serum-free samples was higher , about 1 . 5× for mock-infected cells and ∼2× for virus-infected cells as quantitated from the fluorescence of the total lipid spots loaded at the start position before TLC resolution ( not shown ) . In the mock-infected cells incubated without serum significant amount of fluorescent label appeared in a spot on neutral lipids plate likely representing triglycerides with abnormal mobility due to the presence of bodipy-FA ( Fig . 2A , lane 4 ) , correlating with the microscopy observation of fluorescence accumulation in lipid droplets . Note that free bodipy-FA in the neutral lipid separation system moved much slower than the long chain free FA markers , while in the polar lipid separation system its mobility was close to C18 long chain free FAs ( Figure S1 , compare lanes 8 ( free bodipy-FA , white horizontal arrows ) and 5 ( stearic acid C18:0 ) , 6 ( palmitic acid C16:0 ) , 7 ( linoleic acid C18:2 ) on neutral and polar lipid plates ) . We cannot exclude that this spot represents some other type of neutral lipid , but in any case synthesis of this compound is active in mock-infected cells and shut down in infected cells . In the lipids isolated from infected cells almost no fluorescence was resolved in the non-polar system , ( Fig . 2A compare lanes 3 and 4 ) , indicating that synthesis of neutral lipids is shut down . In fact in the TLC system optimized for neutral lipids most of the labeled lipids isolated from infected cells remained as a bright spot at the loading position ( Fig . 2A , lanes 1 and 3 ) . In contrast , on the TLC plate resolved using the polar solvent system we observed a very strong signal for newly synthesized PC in infected cells ( Fig . 2B , compare lanes 1 , 3 and 2 , 4 ) . The staining of the TLC plates for total lipid content did not reveal significant differences between infected and mock-infected cells ( Fig . 2 ) . It should be noted that the samples were analyzed after 4 h p . i . , meaning that the period of up-regulated synthesis of new lipids was relatively short , and they apparently did not significantly change the overall lipid content of the cells . Thus , PV infection does not only increase the level of FA import but modifies their metabolic channeling by down-regulating synthesis of neutral lipids , and redirecting the newly imported FAs for the highly up-regulated production of PC . Import of FAs is inextricably connected to activity of acyl-CoA synthetases [23] . Transport of saturated or unsaturated long-chain fatty acids containing 18 or fewer carbons across biological membranes is rapid and not thought to be rate-limiting [31] , [32] . Thus the increased uptake of FFA probe suggests that long chain acyl-CoA synthetase activity must be up-regulated upon infection . To measure this activity we prepared lysates from HeLa cells infected at the multiplicity of 50 PFU/cell and incubated without serum for different times post-infection . The infected cells demonstrated elevated level of acyl-CoA synthetase activity as early as 2 h p . i which steadily increased at later times ( Fig . 3A ) . Cellular acyl-CoA synthetases have different preferences for the backbone length and degree of saturation of FA , although the substrate specificity is generally not very strict and one enzyme can activate multiple FA species [23] , [33] . To assess the substrate specificity of acyl-CoA synthetases activated upon infection we performed FA import competition assay by labeling the cells at 4 h p . i . with bodipy-FA in the presence of 125× molar excess of competitor FAs . If the competitor FA is a preferred substrate over the bodipy-FA probe , it should result in the corresponding reduction of fluorescence . The control samples showed that infected cells incorporated more than 250% of bodipy-FA relative to mock-infected cells , in agreement with microscopy and TLC data ( Fig . 3 B–E ) . In mock-infected cells incubated with serum no FA tested showed significant influence on the incorporation of bodipy-FA , likely because of the substantial amount of FAs already present in serum ( Fig . 3B , C ) . In mock-infected cells incubated without serum the strongest competition was shown by palmitic acid ( C16:0 ) ( ∼37% of mock control ) while myristic acid ( C14:0 ) demonstrated a weaker effect ( ∼22% of mock control ) ( Fig . 3D ) . The addition of unsaturated FAs to mock-infected cells incubated without serum actually significantly enhanced incorporation of bodipy-FA , up to more than 100% in case of linolenic ( C18:3 ) acid ( Fig . 3E ) , indicating that they were stimulating FA uptake by the cells starved without exogenous FAs for 4 hours . The virus-infected cells showed a completely different pattern . Unsaturated FAs: oleic ( C18:1 ) , linoleic ( C18:2 ) and linolenic ( C18:3 ) , mildly reduced bodipy-FA incorporation in the presence and in the absence of serum ( Fig . 3C , E ) . Palmitic acid ( C16:0 ) inhibited import of the fluorescent label in the presence of serum from ∼270% to ∼220% , and in the absence of serum from ∼250% to ∼190% ( Fig . 3A and C ) . The strongest competitor for the bodipy-FA incorporation in infected cells was myristic acid ( C14:0 ) inhibiting bodipy-FA incorporation in the presence of serum from ∼270% to ∼150% and from ∼250% to ∼130% in the absence of serum ( Fig . 3B , D ) . These data show that acyl-CoA synthetase activity in infected cells is strongly stimulated from the early time post infection and that its substrate preference is changed . The competition experiments suggest that the pool of acyl-CoAs available for new phospholipid synthesis should be different in infected and mock-infected cells . To investigate the changes in the spectrum of PC molecules , TLC-MALDI was used to couple the power of solvent resolution of phospholipids by TLC to the mass identification capacity of matrix assisted laser desorption-ionization time-of-flight mass spectrometry ( MALDI-TOF-MS ) ( Fig . 4A ) . PC was first identified by characteristic TLC migration , and reflectron positive mode MALDI-TOF-MS was used to scan the TLC lane . The mass to charge ratio ( m/z ) was used to secondarily identify the major PC molecules and acyl variants ( Fig . 4B ) . We observed a substantial drop in the diversity of the PC molecules containing long C18 chains , running in the high Rf chromatography zone , and correspondingly a rapid increase in the PCs with shorter acyl chains from the low Rf zone upon infection ( Fig . 4C ) . The analysis of the individual PC classes demonstrates a fast shift in the composition of PCs during infection . At 2 h p . i . there is already a significant increase in PCs with C18/C18 acyl chains , as well as C16/C18 ones , accompanied by a noticeable drop in the abundance of the C14/C16 and C16/C16 PCs , compared to mock-infected cells . This general trend continues later in infection with an especially strong increase in the C16/C18 PC species at 4 h p . i . ( Fig . 4D ) . Changes in lipid abundance at 6 h . p . i . do not follow the general trends observed at 2 and 4 h . p . i . likely due to the significant degree of cell lysis observed at this late stage of infection at high MOI . It should be noted that while the competition assay showed a strong preference for import of C14 myristic acid to infected cells , it only reflects the changes in the prevalent cellular acyl-CoA synthetase activity induced by polio infection , and cannot be directly interpreted as that myristic acid is the predominant imported FA in natural conditions . The actual composition of intracellular acyl-CoA pool will be shaped by the availability of the corresponding FA substrates . The resolution of TLC-MALDI is not sufficient to separate PC molecules with saturated and unsaturated FA chains with the same number of carbon atoms . Thus , PV infection does not only up-regulates the overall synthesis of PC but specifically changes the molecular composition of this structural phospholipid indicating that membranes of PV replication complexes are significantly different from the pre-existing cellular membranes . We investigated effect of different inhibitors of cellular metabolism and viral replication on the activation of FA import . The possibility that a PV protein ( s ) may have acyl-CoA synthetase activity is unlikely since all known acyl-CoA synthetases have signature of two conserved motifs [22] lacking in the PV polyprotein . PV replication proceeds in the cytoplasm and induces rapid shut-off of cellular mRNA translation , inhibition of nuclear transcription and disruption of nucleo-cytoplasmic barrier [3] . Indeed replication of poliovirus t . I Mahoney is not affected by actinomycin D ( AMD ) , an inhibitor of nuclear transcription [34] , [35] , our observations ( not shown ) . We assessed the effect of inhibition of cellular transcription on increase of fatty acids import upon polio infection . The cells were pre-incubated with AMD for 30 min before the infection , and the inhibitor was present in the media further on during the whole time of infection and bodipy-FA labeling . Consistent with the sufficiency of pre-existing cellular factors for poliovirus replication , we observed that actinomycin D had no effect on the activation of bodipy-FA import in infected cells ( Figure S2 , panels A and B ) . We also investigated if continuous synthesis of PV RNA and proteins are required to sustain the elevated FA uptake by infected cells . The infection was allowed to proceed normally for 3 . 5 h without the inhibitors , and then guanidine-HCl , a strong specific inhibitor of PV RNA replication [36] , or cycloheximide , a general inhibitor of translation , were added . After 30 min incubation with the inhibitors , the medium was replaced with the labeling media that contained bodipy-FA and the corresponding inhibitors , and the cells were incubated for another 30 min . Thus the labeling was performed when synthesis of the viral macromolecules was already inhibited for 30 min . The experiment shows that inhibition of polio RNA and protein synthesis did not prevent enhanced import of fatty acids ( Figure S2 , C , D and E ) . Since infection in the control sample effectively proceeded an hour more than in the inhibitor-treated samples ( 30 min pre-incubation +30 min labeling in the presence of inhibitors ) , control infected cells show higher bodipy-FA accumulation , consistently with the correlation between the amount of viral proteins and the level of FA import stimulation . Thus , increase of FA import in infected cells does not depend on new expression of cellular genes and likely relies on activation of pre-existing cellular factors by viral proteins . The human genome contains genes for 13 long and very long chain acyl-CoA synthetases that may facilitate FA uptake by the cells [22] . The data on expression profiles of these proteins as well as on their contribution to cellular metabolism are still very fragmentary and controversial [37] . First , we monitored by western blot several long-chain acyl-CoA synthetases for which the reliable antibodies were available . We observed specific proteolytic cleavage of FATP3 and to a lesser extent Acsl3 proteins in infected cells , suggesting that their activity is being actively regulated ( Figure 5A , arrows ) . Western blots of Acsl5 and FATP4 did not reveal obvious modifications of these enzymes in infected cells ( Figure 5A ) . To see if association of acyl-CoA synthetases with cellular components is changed upon infection we treated the cells with digitonin . The membrane-targeted viral proteins 2C and 2BC were virtually totally recovered from the permeabilized cells . At the same soluble proteins 3D and 3CD were mostly lost upon cell permeabilization confirming optimal permeabilization conditions ( Fig . 5A ) . Some amount of 3D and 3CD is expected to remain associated with the membrane-bound viral replication complexes . We observed significant loss of FATP3 protein from permeabilized cells at 4 and 6 h p . i . indicating that its association with cellular components is changing ( Figure 5A , arrowhead ) . FATP3 was previously shown to be an-ER-localized protein with its N-terminus inserted into the ER lumen [21] , and would be expected to remain in cells after digitonin treatment , as we see in the mock-infected cells . Thus its loss from the infected samples demonstrates that association of this protein with cellular structures is changing upon infection . Interestingly , FATP3 is one of the two long chain acyl-CoA synthetases undergoing proteolytic processing upon infection . To implement an unbiased approach to identify acyl-CoA synthetases that support replication of poliovirus we performed screen with siRNA pools targeting all 13 long chain acyl-CoA synthetases . Only siRNA against FATP5 showed significant toxicity in HeLa cells likely due to some non-specific effect ( Figure S3 ) since this protein is believed to be expressed only in liver [38] . Depletion of other acyl-CoA synthetases was well tolerated by the cells , the apparent slight toxic effect rather reflects somewhat slower growth of cells treated with certain siRNA pools ( Figure S3 ) . Our initial screen identified three siRNA pools that induced significant , ∼80% reduction of replication: anti-acyl-CoA synthetase Bubblegum 2 ( AcsBG2 ) , FATP3 and Acsl3 ( Figure S3 ) . Western blot analysis of the targeted proteins revealed that only effect of Acsl3 siRNA was specific . AcsBG2 was not expressed in our HeLa cells as expected , since this protein was previously shown to be specific for brain stem and testis [39] , and treatment of cells with either pooled or individual siRNAs against FATP3 did not result in significant reduction of the amount of the protein ( not shown ) . All siRNAs from the anti-Acsl3 pool resulted in reduction of the targeted protein and decreased replication of PV , siRNA #2 was the most potent . The specificity of the ACSL3 knock-down effect was confirmed by rescue of polio replication by expression of the ACSL3 with mutated siRNA #2 targeting sequence ( Figure S4 ) . The strongest reduction of polio replication with the least cellular toxicity was observed after treatment of cells with the all four anti-Acsl3 siRNAs pool ( Figure 5B ) . We also monitored PV infection in the cells expressing recombinant protein GFP-Acsl3-HA . Accumulation of viral proteins was significantly delayed in such cells , compared to cells expressing just EGFP , or transfected with an empty pUC plasmid ( Fig . 5C ) , suggesting that fusion protein GFP-Acsl3-HA works like a dominant-negative mutant in the context of polio infection . Note that transfection efficiency of HeLa cells is about 60–80% and protein accumulation is measured in the total cell population , therefore the actual reduction of polio replication in transfected cells only should be even stronger . Since knock-down of Acsl3 expression inhibits polio replication , it is impossible to directly examine the role of Acsl3 in activation of FA import upon infection . Thus we expressed poliovirus non-structural P2P3 polyprotein fragment ( Fig . 6A ) in cells treated with control or ACSL3-targeting siRNAs with the help of vaccinia virus expressing T7 RNA polymerase [40] . This system is independent of polio replication and is discussed in details in the section below . Expression of poliovirus proteins was induced by infection of cells with vaccinia-T7 virus ∼72 hours post siRNA transfection . At 4 hours post vaccinia infection bodipy-FA probe was added to the media for 30 min . The cells treated with control siRNA which were positive for a polio antigen showed strong activation of FA import ( Fig . 5D , arrows ) , while import of bodipy-FA in the cells with ACSL3 knock-down was significantly lower ( Fig . 5E , arrowheads ) . The statistical analysis confirmed that bodipy-FA fluorescence normalized to polio protein 2B signal strongly declined in ACSL3 knockdown cells ( Fig . 5D ) . Our data show specific limited proteolysis of Acsl3 and FATP3 in infected cells , accompanied by the loss of membrane association of FATP3 , which likely leads to modulation of activity of these enzymes , and demonstrate that functional Acsl3 is required for polio replication and is directly involved in import of FA upon expression of polio proteins . To identify a PV protein ( s ) responsible for activation of FA import we expressed fragments of the viral polyprotein with the help of the vaccinia virus expressing T7 RNA polymerase [40] . The cells are transfected with a plasmid coding for a viral protein under control of T7 RNA polymerase promoter , and the next day they are infected with a vaccinia virus expressing T7 RNA polymerase gene . Thus the gene of interest is only expressed when T7 RNA polymerase accumulates in vaccinia-infected cells . This system provides rapid expression of high amount of recombinant proteins independent of nuclear transcription and RNA processing machinery , thus allowing synthesis of poliovirus proteins uncoupled from replication of viral RNA , on the timescale similar to the normal polio infection . It was successfully used previously to assess effects of individual PV proteins on cellular membrane architecture [41] . The P1 region of poliovirus genome codes for the structural proteins which are dispensable for replication , so we focused on the non-structural proteins encoded in the P2P3 genomic region ( Figure 6A ) . The cells transfected with the plasmids coding for fragments of PV cDNA were infected with vaccinia virus , and bodipy-FA probe was added to the media for 30 min at 4 hours post vaccinia infection . All cells displayed significant vaccinia-induced CPE at that time ( Fig . 6 B and C ) . Three of the polio proteins: 2B , 2C , 3A have membrane localization domains , and they have long been implicated in membrane rearrangements in infected cells [5] , [6] , [42] . However individual expression of 2B , 2C , 2BC , 3A , as well as 3CD and 3D did not result in increased FA import ( not shown ) . Expression of the whole P2P3 polyprotein ( 2A-3D ) ( Fig . 6A ) induced strong increase in bodipy-FA import ( Fig . 6B 1 and 5 , arrows; and Fig . S5 ) . Expression of the 2B-3D or 2C-3D polyprotein fragments ( Fig . 6A ) never stimulated accumulation of bodipy-FA to the level comparable to the 2A-3D expressing cells ( Fig . 6B 2 , 6 and 3 , 7; and Figure S5 ) , showing that 2A is the protein responsible for triggering activation of FA import . Compared to the control cells infected with vaccinia-T7 virus after transfection with an empty vector , cells expressing 2B-3D fragment showed small , but reproducible increase in the baseline level of bodipy-FA accumulation ( Figure S5 ) . In the context of the P2P3 2A is expressed together with all the other non-structural poliovirus proteins and thus the activation of fatty acid import may depend on coordinated action of 2A and other viral factors . To investigate if expression of 2A alone can induce activation of FA import we generated a construct that expresses the 2A protein with an HA tag between amino-acids 144–145 since suitable anti-2A antibodies were not available . This position was previously identified to tolerate insertions in the context of the polio genome [43] . The full length polio RNA with 2A-HA had the same infectivity as the wt RNA , although it displayed somewhat smaller plaque phenotype ( not shown ) , showing that 2A-HA is fully functional in the viral life cycle . When we expressed the 2A-HA protein individually it did not induce activation of FA import on its own ( not shown ) . To investigate the requirement for 2A protease activity we engineered a point mutation in the 2A sequence substituting the catalytic amino acid C109 to A [44] . The lack of the protease activity of the 2A C109A mutant was confirmed by the absence of processing of eIF-4G , a well-established cellular target of 2A ( Fig . 6D ) . Expression of the P2P3 piece of the PV polyprotein with the inactive 2A induced activation of the FA import like the wt construct ( Fig . 6C ) , showing that complex role of 2A in modification of metabolism of infected cells is not restricted to the proteolitic processing of cellular proteins . Thus poliovirus protein 2A is necessary for activation of FA import , independent of its protease activity , but expression of 2A alone is not sufficient and requires contribution from other viral non-structural proteins from the P2P3 region . Viruses in the animal host encounter diverse cellular environments , even when their tropism is limited to a few specific tissues . At the same time the core essential processes of replication machinery are expected to operate similarly in every cell type permissive for viral infection . To see if the activation of long chain FA import is a universal attribute of picornavirus infection , we assessed FA import in different types of cells upon infection with different picornaviruses . PV replication induced strong activation of bodipy-FA import in Vero ( green African monkey kidney ) , 293HEK ( human embryonic kidney ) and SH-SY5Y ( human neuroblastoma ) cells similarly to what we observed previously in HeLa cells ( Fig . 7A–D ) . The Figure 7B shows that only Vero cells actively expressing polio proteins demonstrate high FA import phenotype . To see if different viruses induce activation of FA import we compared PV-infected HeLa cells with the cells infected with Coxsackie virus B3 ( CVB3 ) , another enterovirus related to polio; as well as with a significantly more distantly related encephalomyocarditis virus ( EMCV ) . These viruses efficiently replicate in HeLa cells with similar duration of their infection cycles ( not shown ) . The cells infected with all these viruses showed strong activation of the bodipy-FA import which was distributed into similar membranous structures ( Fig . 7 E–F ) . These data show that activation of long chain FA import is a universal mechanism of altering host cell membrane metabolism activated by diverse picornaviruses in different cell types .
The generation of the membranous replication platforms is an indispensable step in the life cycle of all ( + ) RNA viruses of eukaryotes . The double membrane compartments observed at certain conditions in PV-infected cells prompted a hypothesis of the autophagy contribution to their formation which is also supported by the processing of the LC3 protein , a hallmark of autophagosome development , in infected cells [5] , [45] . The available data also show that PV hijacks components of the cellular secretory pathway which carries coated vesicles between cellular organelles and plasma membrane . Rust et al . showed co-localization of PV protein 2B with the components of the COPII coat [6] . GBF1 , a guanidine-nucleotide exchange factor for small GTPase Arf1 that coordinates formation of COPI-coated vesicles was shown to be a critical host factor for PV and CVB3 replication [46] , [47] . Hsu et al . proposed that activation of Arf1 by GBF1 in infected cells results in recruitment of phosphoinositol-kinase 4 III β ( PI4KIIIβ ) instead of COPI coat . PI4KIIIβ generates PI4P lipid and diverts membranes from the secretory pathway to building viral membranous replication organelles [4] . However while the model of generation of viral replication membranes through subversion of the natural membrane remodeling machinery of the secretory cargo vesicles formation or autophagy seems logical and aesthetically appealing , it is difficult to reconcile with all the experimental data and it cannot explain membrane remodeling by even related viruses . Picornaviruses show drastically different sensitivity to inhibition of the secretory pathway . PV is very sensitive to BFA , an inhibitor of GBF1 [48] , [49] , while EMCV or FMDV are totally refractory to the drug , and other picornaviruses demonstrate intermediate sensitivity to the inhibitor [7] , [50] . It is likely that the components of the secretory pathway are necessary for the functionality of the poliovirus replication complexes , rather than for the development of the membranous scaffold of these structures . The characteristic membrane remodeling could be induced by expression of poliovirus proteins in the presence of BFA , showing that the functional GBF1-dependent pathways are not required to induce morphogenesis of the replication platforms [51] . Similarly picornavirus response to manipulation of autophagy varies greatly . Replication of human rhinovirus 2 was reported to be either dependent on induction of autophagy , or completely non-sensitive to manipulation of this pathway in different systems [8] , [9] . Moreover , electron tomography studies show that replication organelles of PV and CVB3 represent complex tubular structures , rather than clusters of vesicles expected to be generated by vesicle-forming machinery of the secretory pathway or autophagy [52] , [53] . The similar morphology of the replication structures of all picornaviruses strongly suggests that the mechanisms of their formation should be shared among different viruses [3] , [54] . An intriguing clue comes from the old observations on the phospholipid synthesis in infected cells . Both poliovirus and EMCV were shown to strongly stimulate phospholipid synthesis [13] , [55] , while the former is sensitive BFA and the latter is completely resistant to the inhibitor [7] , [48] , [49] . Our data presented here suggest that building of viral replication organelles to the large extent may rely on new membrane synthesis , unique to the infected cells , rather than on remodeling of pre-existing organelles through hijacking of membrane trafficking pathways . We demonstrate a rapid increase in long chain FA import into poliovirus-infected cells linked to activation of acyl-CoA synthetase activity . The overall cellular acyl-CoA synthetase activity was elevated as early as 2 hours post infection . Mock-infected cells largely incorporated FA in the lipid droplets , while in infected cells the imported FA was utilized mostly in highly activated PC synthesis , reflecting the rapid development of membranous replication platforms . The different substrate preference of acyl-CoA synthetase activity in infected cells translated into the overall significant perturbations in the composition of the PCs and increase of the diversity of PCs with shorter acyl chains . While we observed strong preference for import of C14 myristic acid in infected cells during the competition experiment , the cells preferentially accumulated C16/C18 PC species . This apparent controversy reflects the fact that during the competition assay the cells were incubated with bodipy-FA label and only one species of fatty acid was present in the media , while the cells assessed for changes in PC composition were incubated in standard media supplemented with serum . It is also possible that while myristic acid is being the preferred substrate for import into the infected cells , myristoyl-CoA may not be the preferred substrate for synthesis of PC . The actual composition of the pool of acyl-CoAs that eventually will be used for synthesis of new PC molecules is determined to a large extent by the availability of the corresponding FA substrates . Human serum for example may contain almost 30 times more palmitic ( C16:0 ) and stearic ( C18:0 ) acids than myristic acid [56] . The human genome contains 13 genes for acyl-CoA synthetases capable of activation of long and very long chain FAs ( C12–C26 ) [22] . They have different expression profiles and distinct , although often overlapping substrate specificity , contributing to the complex tissue-specific regulation of the FA metabolism [23] , [57] . siRNA knock-down and protein over-expression experiments show that polio replication requires functional Acsl3 . Interestingly , both siRNA knock-down of Acsl3 and over-expression of a fusion protein GFP-Acsl3-HA was detrimental for poliovirus replication . It is possible that GFP and/or HA tags of the fusion protein specifically interfered with the Acsl3 function required to support the viral infection . It was reported that addition of the HA tag could significantly change cellular localization of a lipid droplet protein [30] . On the other hand , it is possible that the inhibition of polio replication resulted from the excess of Acsl3 activity in the cells over-expressing GFP-Acsl3-HA construct . For example it was shown that both depletion and over-expression of a chaperon protein DNAJC14 reduces replication of flaviviruses [58] , suggesting that at least some cellular factors can support viral replication only at a narrow range of concentrations . We cannot exclude that other acyl-CoA synthetases are supporting viral replication . In this study we followed only the results of the siRNA screen that showed the most significant reduction of polio replication . It is possible that during the siRNA treatment targeting one acyl-CoA synthetase the cells would compensate this loss by increasing synthesis of other related proteins . Our siRNA data underscore that validation of the knock-down results on protein level is very important and that without it the popular high-throughput siRNA-based screens for host factors important for viral replication should be interpreted with caution . The available data on expression , cellular localization and activity of long chain acyl-CoA synthetases is still fragmentary and often frustratingly contradictory . Acsl3 was reported to contribute to FA uptake by mammalian cells [20] , [29] . In our experiments Acsl3 appeared to be directly involved in the activation of import of FA upon expression of polio proteins . Interestingly , rat Acsl3 preferentially activates short saturated FAs lauric ( C12:0 ) and myristic ( C14:0 ) in a biochemical assay [59] consistent with the strong effect of myristic acid in our competition experiments . Import of FAs into infected cells and polio replication were relatively insensitive to triacsin C ( not shown ) , an inhibitor of rat acyl-CoA synthetases 1 , 3 , and 4 [60] , [61] . However sensitivity of human and rat proteins to triacsin C may not be the same , moreover , as was reported for Acsl5 , conflicting results could be observed in different assays for the same enzyme [62] , [63] . It is also possible that sensitivity to the inhibitor may change upon interaction of Acsl3 with viral factors . We also observed proteolytic cleavage of Acsl3 and FATP3 and loss of association of FATP3 with cellular structures upon infection , suggesting that modulation of acyl-CoA synthetase activity in infected cells is very specific . Poliovirus proteases 2A and 3C recognize YG or FG , and GQ bonds respectively , but the actual utilization of the cleavage sites depends on protein conformation and surrounding sequences [3] . Potential cleavage sites for the viral proteases that can generate the observed fragments are present in the Acsl3 and FATP3 sequences , but whether 2A or 3C actually directly cleave Acsl3 and/or FATP3 or if the proteolysis is performed by cellular proteases requires further investigation . The size of the small proteolytic products ( ∼30 KDa for FATP3 and ∼37 KDa for Acsl3 ) suggest that the cleavage site is located between the two conserved motifs characteristic of acyl-CoA synthetases [64] , therefore the cleavage likely inactivates the enzymes . It should be noted that accumulation of the small cleavage products of FATP3 and Acsl3 was not accompanied by a significant decrease of the full length proteins , suggesting that proteolysis affects limited fractions of these proteins , possibly only in specific cellular locations which may help to redirect fatty acids from triglyceride synthesis pathway to production of PC . Since poliovirus infection induces rapid shut-off of cellular cap-dependent mRNA translation and nuclear transcription [3] , the most plausible mechanism of stimulation of FA import in infected cells is activation of pre-existing cellular acyl-CoA synthetases rather than the up-regulation of de novo expression of the enzymes . The lack of increase of the total amount of at least some acyl-CoA synthetases in infected cells , as well as resistance of stimulation of FA import to actinomycin D strongly support the idea of activation of pre-existing cellular factors upon infection . Very little is known about the rapid posttranslational regulation of long chain acyl-CoA synthetases . It was shown that insulin may modulate acyl-CoA synthetase activity within rat adipocytes on the timescale of minutes , but the mechanism of this regulation is unknown [65] . The fast activation of acyl-CoA synthetase activity in PV-infected cells required the viral protein 2A . 2A is responsible for the first cleavage of the polyprotein releasing the precursor of capsid proteins [3] . It also cleaves cellular proteins rendering cell environment favorable for viral replication . 2A-mediated cleavage of eIF4G results in inhibition of translation of cellular mRNAs and reorganization of translation apparatus to support IRES-driven translation of the PV RNA [66] . 2A proteases of PV and rhinoviruses can directly cleave nucleoporines leading to rapid disintegration of nucleo-cytoplasmic barrier in infected cells [67] , [68] . The requirement of 2A for activating FA import independently of its protease activity represents a novel function of this protein in virus-cell interaction . Interestingly , previous studies showed that 2A has some role in polio replication unrelated to its protease function , but the nature of this contribution remained unclear [69] . Our data explain previous observations on expression of the PV proteins with known membrane-targeted sequences such as 2C and 2BC that failed to activate synthesis of new lipids in spite of producing complex membrane rearrangements [41] , and shows that expression of the individual membrane-targeted proteins does not fully recapitulate complex modulation of membrane metabolism in infected cells . Previous data show that replication of diverse ( + ) RNA viruses is intrinsically connected to the metabolism of long chain FAs . Replication of some picornaviruses as well as other ( + ) RNA viruses was shown to be sensitive to the inhibitors of the cellular FA synthase [70] , [71] , [72] , [73] . Replication of brome mosaic virus in a yeast model depends on the activity of Delta-9 fatty acid desaturase [74] , [75] and acyl-CoA binding protein Acb1p [76] , consistent with the requirement for specific long chain FA for viral replication . Changes in lipid composition of the replication membranes compared to the membranes in non-infected cells was reported for diverse ( + ) RNA viruses [74] , [77] , supporting to the idea that viral replication complexes represent products of mostly de novo synthesis of new membranes with unique characteristics . Our data provide a foundation for a simple model that may explain the structural development of the membranous replication organelles shared by at least picorna-like viruses ( Fig . 8 ) . Certainly many aspects of this model are hypothetical at this point and require further investigations to elucidate the mechanistic details . The elevated acyl-CoA synthetase activity in infected cells inevitably increases import of FAs from the extracellular media but also would activate FAs released from intracellular sources , thus allowing utilization of resources in different cell types or growth conditions . The resulting excess of long chain acyl-CoAs would stimulate further steps in phospholipid synthesis and result in continuous extrusion of new membranes . Moreover the preference of acyl-CoA synthetase activity in infected cells for shorter FAs would result in generating membranes with higher fluidity with the intrinsic propensity to assemble into tight tubular structures ( myelin figures ) [78] surprisingly similar to the picornavirus replication membranes [52] , [53] . These new membranes would need to be decorated with the necessary viral and cellular factors to make them capable of supporting viral replication , but the generation of the structural scaffold seems to be a unique process activated in infected cells , independent of the elements of the secretory pathway or autophagy . The distinct properties of the FA metabolism in infected cells and the widespread reliance of diverse ( + ) RNA viruses on their activation represent an attractive target for development of future broad spectrum antiviral therapeutics .
HeLa and 293HEK cells were maintained in DMEM medium supplemented with 10% heat-inactivated fetal bovine serum ( FBS ) , Vero cells were grown in Eagle's MEM medium with heat-inactivated 10% FBS . Poliovirus type 1 Mahoney , Coxsackie B3 virus and encephalomyocarditis virus were propagated on HeLa cells and their titer was determined in a standard plaque assay . For experimental infections cell monolayer was washed once with serum-free DMEM , the virus was added according to the desired multiplicity of infection for 30 min at room temperature in serum-free DMEM buffered with 50 mM Hepes pH 7 . 4 , after that the cells were incubated in standard or serum free media in standard growth conditions according to the experimental design . Plasmids pTM-2A-3D , pTM-2B-3D , pTM2C-3D coding for the corresponding fragments of poliovirus cDNA under transcriptional control of T7 promotor and translational control of EMCV IRES were a gift from Dr . Natalya Teterina ( NIH ) . pXpA-RenR plasmid coding for polio replicon with Renilla luciferase gene substituting capsid region of polio genome was described elsewhere [79] Plasmid pTM-2Amut-3D containing mutation C109A in the 2A sequence was produced by point mutagenesis of the 2A sequence in pXpA-P2P3 [80] and recloning the PstI-SpeI fragment into pTM-2A-3D . For plasmid pXpA-SH-PV2A-HA coding for the full length polio cDNA containing 2A with HA tag the SpeI-SnaBI fragment containing 2A-HA was generated by two sequential overlapping PCRs and recloned into pXpA-SH [80] . Plasmid pTM-2A-HA for vaccinia-based expression of 2A protein with HA tag PmlI-SalI fragment containing part of EMCV IRES and the whole coding sequence for 2A-HA was synthesized by Life technologies and was cloned into pTM-1 vector ( gift from Dr . Natalya Teterina , NIH ) . All polio constructs were verified by sequencing . Plasmids pGFP-ACSL3-HA and pcDNA3-ACSL3-HA were generously provided by Dr . Joachim Füllekrug , University of Heidelberg , Germany . Plasmid pCI-ACSL3 was made by cloning the PCR fragment coding for the wt ACSL3 sequence obtained from pcDNA3-ACSL3-HA into pCI expression vector ( Promega ) . pCI-ACSL3r was produced by mutating the anti-ACSL3 siRNA #2 targeting sequence ( bold ) GGAAATGGATTACAGATGTTG into GGGAACGGCCTTCAAATGCTG coding for the same amino acids GNGLQML . Plasmid HsCD00078974 coding for human ADRP protein ( GenBank: BC005127 . 1 ) was purchased from DNASU plasmid collection ( Arizona State University ) . Plasmid pmCherry-ADRP was constructed by cloning the PCR fragment coding for human ADRP into pmCherry-C1 vector ( Clontech ) . All constructs were verified by sequencing . Purified recombinant vaccinia virus expressing T7 RNA polymerase ( VT7-3 [40] ) was a gift from Dr . Natalya Teterina , NIH . The day before experiment HeLa cells were transfected with pTM- or pXpA-based plasmids coding for fragments of polio cDNA under transcriptional control of T7 RNA polymerase promoter and translational control of EMCV or polio IRES respectively , with Mirus 2020 DNA transfection reagent according to manufacturer's protocol . The next day cells grown on glass cover-slips were infected with vaccinia VT7-3 virus at a multiplicity of 10 PFU/cell in serum-free DMEM for 1 hour at 37 C and then incubated for 4 more hours in standard growth media . After that the media was changed to pre-warmed serum-free media supplemented with 5 µM of fluorescent fatty acid analog Bodipy 500/510 C4–C9 and the cells were incubated for 30 more minutes , then washed once with PBS and fixed with 4% formaldehyde in PBS and processed for microscopy analysis . Anti-Acsl3 mouse polyclonal antibodies were from Abnova; anti-Acsl5 mouse monoclonal antibodies were from Abcam; rabbit polyclonal anti-FATP3 , anti-FATP4 , anti-Acs bubblegum 2 were described in [81] , [82] , [83] . Mouse monoclonal anti-polio 2B and 3A antibodies were a gift from Prof . K . Bienz , University of Basel , Switzerland . Rabbit polyclonal anti-polio 3D antibodies were made by Chemicon using recombinant polio 3D protein as immunogen . Secondary anti-mouse and anti-rabbit highly cross-adsorbed goat antibodies conjugates with Alexa 350 , Alexa 498 and Alexa 595 were from Molecular Probes . Fluorescent fatty acid Bodipy 500/510 C4–C9 ( bodipy-FA ) was from Molecular probes . Unlabeled long chain fatty acids , Coenzyme A ( CoA ) , α-cyclodextrin , ATP and neutral and phospholipid standards were from Sigma-Aldrich . Phosphatidylinositol 4 phosphate from porcine brain was from Avanti Polar Lipids . Fatty acid stock solutions were prepared in DMSO . Neutral and phospholipids were dissolved in chloroform . siGenome siRNA pools and corresponding individual siRNA oligos targeting human long chain acyl-CoA synthetases were from Dharmacon . HeLa cells were plated at 10000 cells/well in a 96 well plate and transfected with siRNA with Dharmafect 1 transfection reagent ( Dharmacon ) according to manufacturer's recommendations . After 72 hours of incubation with siRNA the cells were used for polio replicon replication assay . Toxicity of siRNA treatment was assessed with Cell-Titer Glo luminescent assay ( Promega ) . Polio replicon assay was performed as described in [51] with minor modifications . Briefly , HeLa cells grown on 96-well plates were transfected with polio replicon RNA with mRNA trans-It kit ( Mirus ) and incubated in TECAN Infinite M1000 plate reader at 37 C for 16 hours in standard growth media supplemented with 30 µM live cell Renilla substrate EnduRen ( Promega ) . Measurements of Renilla luciferase activity were taken every hour post transfection , and the signal for each sample is averaged from at least 8 wells of the 96 well plate . For immunofluorescent microscopy cells were fixed with 4% formaldehyde in PBS for 20 min . The cells labeled with Bodipy 500/510 C4–C9 were incubated with primary and secondary antibodies in 0 . 02% saponin in PBS with 5% FBS for 1 hour and were washed 3 times with PBS after each incubation . Secondary antibodies conjugated to Alexa 350 were used for detection of viral antigens in Bodipy 500/510 C4–C9-labeled cells since we found that Bodipy 500/510 emits strong fluorescence in both green and red spectra after detergent treatment required for immunodetection of proteins , apparently due to the red shift in fluorescent spectrum of Bodipy when the fluorescent groups come in close contacts ( Molecular Probes manual ) . Without detergent permeabilization red fluorescence of Bodipy-labeled lipids was negligible , thus allowing use of pmCherry-fused ADRP protein for detection of lipid droplets . Images were taken with Zeiss Axiovert 200M fluorescent microscope equipped with Axiocam Mrm monochrome digital camera; confocal images were obtained with a Zeiss ApoTome AxioImager M2 microscope . Colors were artificially assigned to black and white microscope images using Adobe Photoshop software according to the imaging channel . All software processing was applied universally to the whole image and images from different samples were taken and processed using identical conditions . Fluorescence intensity in individual cells was evaluated with ImageJ software ( NIH ) . Co-localization pattern was visualized with ImageJ software ( NIH ) co-localization module by Pierre Bourdoncle , Institut Jacques Monod , Service Imagerie , Paris . HeLa cells grown on 12-well plate were infected with poliovirus and incubated for 4 hours in standard growth conditions . Permeabilization was performed at room temperature . The cells were washed once with KHM buffer ( 110 mM K-acetate , 2 mM MgCl2 , 20 mM HEPES-KOH , pH 7 . 4 ) and incubated for 5 min in 50 µg/ml fresh digitonin solution in KHM ( KHM buffer without digitonin for control cells ) . After that the cells were washed twice with KHM and lysed with mild lysis buffer ( 0 . 1M Tris-HCl pH 7 . 8; 0 . 5% Triton-×100 ) supplemented with protease inhibitors cocktail ( Sigma-Aldrich ) . The lysate cleared by low-speed centrifugation was used for western blot analysis . Multiple western blots were performed after stripping the membrane with Re-Blot Plus solutions ( Chemicon ) according to manufacturer's recommendations . Western blots were developed with ECL prime or ECL select chemiluminescence kits ( GE Healthcare ) . For fatty acid import assay infected or mock-infected cells were incubated in standard or serum-free media depending on experiment design . After indicated time post infection the media was replaced for new pre-warmed media with 0 . 4 µM Bodipy 500/510 C4–C9 and after 30 min incubation the cells were fixed with 4%formaldehyde in PBS for 20 min , washed with PBS and used for microscope imaging and/or fluorescence reading on TECAN Infinite M1000 plate reader . Cells labeled with bodipy-FA as described above were detached from the plate with versen solution , fixed for 1 h with 1% paraformaldehyde in PBS and analyzed with FACSAriaII cell sorter ( BD ) . The total of 10000 events were used for fluorescence analysis with FloJo software after gating in FCS and SSC mode for single cell population . HeLa cells were grown on a 96 well tissue culture plate with transparent bottom overnight and the next day infected with poliovirus at 50 PFU/cell . After infection the cells were incubated either in standard or serum-free media as indicated . At 4 h p . i . the media was replaced with the same type of pre-warmed media supplemented with 0 . 4 µM of bodipy-FA label , 50 µM of the competitor fatty acid and 1 µg/ml of cell-permeable DNA stain Hoechst 33342 . After 30 min incubation the cells were washed with PBS and fixed with 4% formaldehyde in PBS and the Hoechst and bodipy-FA fluorescence were read in TECAN Infinite M1000 plate reader at excitation/emission 340/455 and 490/520 respectively . The bodipy-FA signal was normalized to Hoechst to account for variability in cell density and the data for each fatty acid are averaged form 12 wells . The data are expressed as percentage of the signal from control mock-infected cells incubated without any competitor fatty acid . Statistical analysis was performed with GraphPad PRIZM software . HeLa cells grown on T75 flasks were harvested , washed 3 times with cold PBS and re-suspended in STE buffer ( 8 . 5% sucrose , 10 mM Tris-HCl pH 8 . 0 . 5 mM EDTA ) supplemented with protease inhibitors cocktail ( Sigma-Aldrich ) . Cells were lysed by twice freeze-thawing and the protein concentration of the lysates was determined by Bradford method . The assay mix containing 500 nM Bodipy 500/510 C4–C9 substrate solubilized with α-cyclodextrin ( 10 mg/ml in 10 mMTris-HCI pH 8 . 0 ) , 40 mM Tris-HCI pH 7 . 5 , 10 mM ATP , 10 mM MgCl2 , 0 . 2 mM CoA , 0 . 2 mM dithiothreitol was assembled on ice , and the reaction was started by addition of an aliquot of cell suspension containing 60 µg of total protein . Duplicate reactions were incubated at 37 C for 20 min and terminated by the addition of Dole's solution ( isopropanol∶heptane∶2N H2SO4 40∶10∶1 ) . Newly synthesized fluorescent acyl-CoA was recovered in aqueous phase after 4 extractions with heptane and the fluorescence was measured by TECAN Infinite M1000 plate reader . Statistical analysis was performed with GraphPad PRIZM software . Lipid extraction was performed according to Folch method [84] . Cells grown on T75 flasks were harvested , resuspended in STE buffer ( 8 . 5% sucrose , 10 mM Tris-HCl pH 8 . 0 . 5 mM EDTA ) and the protein concentration of cell suspension was determined by Bradford method . An aliquot of cell suspension containing 1500 µg protein was adjusted to 250 µl by STE in a glass tube and 3 . 75 ml of chloroform∶methanol ( 2∶1 ) with 5 mM HCl mix was added . The tube was vortexed for 30″ , then 0 . 75 ml H2O was added , and the tube was vortexed again for 30″ and centrifuged at 1500 rpm in a tabletop centrifuge for 5 min . The top aqueous phase was discarded and lower organic phase was extracted 3 more times with Folch theoretical upper phase ( chloroform∶methanol∶H2O 3∶48∶47 ) . After extractions lipid-containing organic phase was dried under a stream of nitrogen gas . Lipids were re-suspended in 50 µl of chloroform before loading on TLC plates . Thin layer chromatography glass silica plates ( Analtech ) were prewashed with chloroform∶methanol ( 1∶1 ) and air-dried . Polar lipids were separated in chloroform∶ethanol∶water∶triethylamine ( 30∶35∶7∶35 ) and neutral lipids were separated in hexane∶ether∶acetic acid ( 80∶20∶1 ) . Phospholipids were stained with Phospray ( Sigma Aldrich ) and neutral lipids were detected with bromothymol blue spray ( Sigma Aldrich ) . Total lipid extracts of infected HeLa cultures were reconstituted in equal volumes 2∶1 chloroform∶methanol ( v∶v ) . 20% of the reconstituted lipids were spotted onto aluminum-backed silica gel 60 F254 TLC plates ( Merck , Darmstadt , Germany ) . TLC plates were developed in an equilibrated chamber of 65∶25∶4 chloroform∶methanol∶ammonium hydroxide ( v∶v∶v ) , then removed and dried under a gentle nitrogen stream . Plates were prepared for MALDI by spray-coating with at least 5 mL of a 20 mg/mL solution of Norharmane ( 9H-Pyrido[3 , 4-b]indole ) in 2∶1 chloroform∶methanol ( v∶v ) . Plates were dried under a gentle nitrogen stream and mounted to a TLC-MALDI adapter . TLC-MALDI software within Compass 1 . 3 was used to define lane length , width , raster distance , and total shots ( 300 per position , 900 shots summed from 3 x-step , width , 1 y-step per millimeter , height ) . Plates were read in reflectron-positive mode with a mass window of 660–1060 m/z . Peak intensity values were used to summate PC intensity . These intensities were transformed to ratios of PC subset/PC total . To accommodate for lane-to-lane variation these ratios were normalized to mock infected ratios , represented as a percentage change . To evaluate changes in the diversity of PC species during infection the total PC spot was deconstructed by Rf value and split into 2 groups , low Rf PC and high Rf PC . For each group the number of unique m/z values detected were tallied within an experimental group . Unique low or high values are represented as a percentage over total PC diversity . These unique m/z values likely represent acyl chain unsaturation patterns , whose abundance , detection threshold , and exact identity cannot be adequately confirmed with this technique . All reagents were obtained through Sigma-Aldrich ( St . Louis , MO ) unless otherwise noted . Commercial phosphatidylcholine ( PC ) standards were obtained from Avanti Polar Lipids , Inc . ( Alabaster , Alabama ) containing different acyl chain configurations . These PC standards were used to define TLC migration in the respective solvent system and used as mass standards for MALDI identification , 2 carbon unit fatty acid changes are easily identified first by m/z 28 unit changes and secondarily by subtle changes in migration . Autoflex Speed MALDI-TOF/TOF-MS , ImagePrep , and MS-related software were sourced from Bruker Daltonics ( Billerica , MA ) . | Eukaryotic cells feature astonishing complexity of regulatory networks , yet control over this fine-tuned machinery is easily overrun by viruses with expression of just a handful of proteins . One of the striking examples of such hostile take-over is the rewiring of normal cellular membrane metabolism by ( + ) RNA viruses towards development of new membranous organelles harboring viral replication machinery . ( + ) RNA viruses of eukaryotes infect organisms from unicellular algae to humans . Many of them induce diseases resulting in significant economic losses , public health burden , human suffering and sometimes fatal consequences . We show how picornaviruses reorganize cellular lipid metabolism by targeting long chain acyl-CoA synthetase activity . This induces increased import of fatty acids in infected cells and up-regulation of phospholipid synthesis , resulting in formation of replication organelles different from the pre-existing cellular membranes . This mechanism is utilized by diverse viruses and may represent an attractive target for anti-viral interventions . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"molecular",
"cell",
"biology",
"virology",
"membranes",
"and",
"sorting",
"viral",
"replication",
"complex",
"biology",
"microbiology",
"viral",
"replication"
] | 2013 | Increased Long Chain acyl-Coa Synthetase Activity and Fatty Acid Import Is Linked to Membrane Synthesis for Development of Picornavirus Replication Organelles |
Neglected Tropical Diseases ( NTDs ) not only cause health and life expectancy loss , but can also lead to economic consequences including reduced ability to work . This article describes a systematic literature review of the effect on the economic productivity of individuals affected by one of the five worldwide most prevalent NTDs: lymphatic filariasis , onchocerciasis , schistosomiasis , soil-transmitted helminths ( ascariasis , trichuriasis , and hookworm infection ) and trachoma . These diseases are eligible to preventive chemotherapy ( PCT ) . Eleven bibliographic databases were searched using different names of all NTDs and various keywords relating to productivity . Additional references were identified through reference lists from relevant papers . Of the 5316 unique publications found in the database searches , thirteen papers were identified for lymphatic filariasis , ten for onchocerciasis , eleven for schistosomiasis , six for soil-transmitted helminths and three for trachoma . Besides the scarcity in publications reporting the degree of productivity loss , this review revealed large variation in the estimated productivity loss related to these NTDs . It is clear that productivity is affected by NTDs , although the actual impact depends on the type and severity of the NTD as well as on the context where the disease occurs . The largest impact on productivity loss of individuals affected by one of these diseases seems to be due to blindness from onchocerciasis and severe schistosomiasis manifestations; productivity loss due to trachoma-related blindness has never been studied directly . However , productivity loss at an individual level might differ from productivity loss at a population level because of differences in the prevalence of NTDs . Variation in estimated productivity loss between and within diseases is caused by differences in research methods and setting . Publications should provide enough information to enable readers to assess the quality and relevance of the study for their purposes .
Most of the people affected by Neglected Tropical Diseases ( NTDs ) are impoverished and marginalized populations , with low visibility and little political voice . They are not considered a priority market for pharmaceutical manufacturers or a health risk for the wealthier parts of the world . [1–3] Nevertheless , NTDs have an important impact on child development , school attendance , learning , nutritional status , pregnancy outcomes , and worker productivity , especially in poor rural settings , where physical labor is the major subsistence mode . As any other disease , they can lead to productivity loss in many ways , including reduced productivity at work ( presenteeism ) , absence from work ( absenteeism ) or even job loss , depending on the type , severity and duration of the disease . [2–12] Many publications in the literature describe the epidemiological and physical aspects of NTDs . In contrast , the impact of NTDs on paid and unpaid work and the productivity of individual men and women has been less frequently studied . Most of the data about the economic burden of NTDs come from small studies in restricted geographical areas . [13] The costs of treatment , mainly long-term ones , can inflict further economic difficulties in populations already struggling to live with less than US$ 1 a day . Besides the obvious advantages of decreasing the healthcare costs due to lack of care or delayed care , investments in health improvement would also help to increase economic growth of the affected regions since healthier populations are more economically productive . [14–16] As part of the movement to increase the attention given to NTDs , a coalition of many stakeholders gathered in January 2012 to discuss the importance of reaching the 2020 WHO goals for this group of diseases . As a result , the London Declaration was signed by many partners , committed to eradicate Guinea worm disease , eliminate three NTDs ( lymphatic filariasis , leprosy , African sleeping sickness ( human African trypanosomiasis ) and blinding trachoma ) and control the others ( schistosomiasis , soil-transmitted helminths , Chagas disease , visceral leishmaniasis and river blindness ( onchocerciasis ) ) . [17 , 18] A better understanding of the effect that NTD have on people’s economic livelihood would be an additional argument in favor of controlling or eliminating them . With this in mind , we performed a systematic literature review to identify and examine publications describing the impact of the London Declaration NTDs . Here we present the results for the five most prevalent ones , which are the ones eligible for preventive chemotherapy ( PCT diseases ) : lymphatic filariasis , onchocerciasis , schistosomiasis , soil-transmitted helminths ( ascariasis , trichuriasis , and hookworm infection ) and trachoma on productivity loss in adults . [7 , 19 , 20]
We performed a comprehensive search of the literature relating to the economic impact of all of the NTDs included in the London Declaration . Databases searched included Embase , Medline ( OvidSp ) , Web of Science , Scopus , CINAHL , PubMed publisher , Cochrane , Popline , Lilacs , Scielo and Google Scholar . The search terms aimed at identifying articles about direct costs of treatment ( such as consultation fees , medication , transport , food , assistance , accommodation ) , as well as indirect labor costs arising from decreased working hours and reduced economic activity attributable to morbidity . The search strategy included the names of the ten London Declaration NTDs ( since many articles mention more than one ) and words such as: ‘economic' , 'financ' , 'cost' , 'productivity' , 'absenteeism' , 'employment' , and 'cost' . A detailed list of the keywords used for each database is found in Supporting Information ( S1 File ) . The search only considered title and abstracts , did not use any time restriction , and was restricted to the English language . The main database search was conducted in November 2013 . There is no review protocol registered . This search included not only productivity loss , but also direct costs for all 10 London Declaration NTDs for a larger project . The results found in this article are limited to the results of the literature search regarding productivity loss from PCT NTDs . The databases were merged according to the order shown in Table 1 . Duplicates were removed automatically using Endnote and the remaining articles were then compared manually using author , year , title , journal , volume and pages to identify any additional duplicates . [21] After duplicates were excluded , we selected the articles that were related to each particular disease and screened the abstract and title of all papers to identify the ones that might provide information on productivity or indirect costs . The full-text versions of all remaining articles were then examined . Articles that did not contain any information on productivity , or only qualitative information on productivity loss ( without any quantitative measures ) were excluded , as well as articles that investigated productivity loss in children . Since the number of relevant publications was expected to be small , no restrictions were made regarding populations ( participants ) , interventions , comparisons , outcomes , study design , or length of follow-up . Articles that could not be retrieved through their respective journals , contacting libraries , or after contacting the authors were classified as ‘not available’ and excluded from the selection . Any additional relevant articles identified when reading the full-text articles or checking their reference lists ( i . e . , the ‘snowball’ search strategy ) were screened using abstract and title and then examined in more detail if they were considered potentially relevant . In addition to searches using databases relating to the ‘white’ literature , we also searched the grey literature by screening websites of relevant organizations ( i . e . World Health Organization , the Centre for Neglected Tropical Diseases , the Carter Center ) ( see S2 File ) . The list of selected articles for each disease was sent to disease experts identified in the literature and from institutions researching/combating NTDs , to check if the selection was comprehensive . Data were extracted from selected articles independently , using a standardized Excel sheet , for the variables: author , year , study design , population , sample size , follow-up period , country , region , disease sequela , definition of productivity loss and results . Disease sequelae are disease manifestations , which for this review were defined by the Global Burden of Disease 2010 study ( see S1 Table ) . [22] No summary measure was chosen beforehand . Instead , the results were presented separately per disease and study and described as they were reported in the articles; results were not statistically combined . If the productivity loss was not already described in percentages of annual productivity in the articles , we calculated it whenever the unit of measurement made it possible , for the sake of comparability between studies and diseases . A working year was assumed to consist of 300 working days . [23] Since the outcome of interest was productivity loss , various study designs were expected . The studies were therefore critically appraised regarding general criteria of selection , performance , attrition , detection , and reporting biases , as specified in the Cochrane Handbook for Systematic Reviews of Interventions . [24 , 25] Therefore , each article was given a rating regarding the risk of bias ( possible options: low , high or unclear ) for each criterion as well as a summary rating . [24 , 25] We added an extra criterion about the degree of relevance that the study outcomes defined as productivity loss had in terms of quantifying productivity loss in adults due to an NTD . This ‘relevance’ criterion was also rated as low or high . This review was conducted according to the PRISMA checklist for systematic reviews .
Table 1 provides an overview of the databases searched and the number of articles identified through each of them . In total , 11 , 449 articles regarding all 10 NTDs were identified using the database searches . Of these , 5 , 316 articles remained after duplicates were removed . There was no duplication across the various NTDs . Sixty percent of the selected articles had a high overall risk of bias ( 26 articles of 42 ) , mostly due to detection bias ( 24 of 42 articles ) , selection bias ( 21 articles of 42 ) , and attrition bias ( 10 of 42 articles ) . Twenty-two articles were rated as relevant , and of these studies , two-thirds ( 14/21 ) had a high overall risk of bias , 2 had a low overall risk of bias and 6 had an unclear overall risk . Only 6 articles had a low overall risk of bias , of which only 2 were relevant , and 9 had an unclear summary rating , of which 6 were relevant ( as described before ) . No particular trend was observed , regarding over- or underestimation of results due to bias . For the complete risk of bias assessment table , please refer to S3 Table .
Various studies have examined productivity loss in patient populations having one of the five most prevalent NTDs . While is clear that these diseases reduce productivity , the actual impact depends on the type , severity and duration of the NTD as well as on the setting . Variation in estimated productivity loss between and within diseases is caused by differences in the different definition of productivity loss , research methods and setting . It is therefore important to examine the literature carefully to understand what was actually observed in order to draw conclusions about the generalizability of the studies . Since productivity loss is an important aspect of the burden of diseases , further research on better estimates of the magnitude of the productivity loss caused by NTDs would enable a more complete picture of their economic burden to individuals , countries , and globally , adding an additional persuasive argument in favor of their control . This review already contributes to a better perception of the magnitude of the effect of an NTD on people’s working and economic situation , and can already offer additional arguments in favor of controlling and eliminating them . However , there is still much room for further research in this field to improve the understanding on NTDs’ effects on individuals’ productivity loss . | Neglected Tropical Diseases ( NTDs ) not only have impact on health and life expectancy of mostly disadvantaged populations , but can also lead to economic consequences , including reduced ability to work . Investments in health improvement of the populations affected by NTDs would also help to increase economic growth of the affected regions , since healthier populations are more economically productive . We performed a systematic literature review to better understand how much NTDs affect people’s economic welfare . Here we present the results for the NTDs that are controlled with preventive chemotherapy ( PCT ) : lymphatic filariasis , onchocerciasis , schistosomiasis , soil-transmitted helminths ( ascariasis , trichuriasis , and hookworm infection ) and trachoma . Our findings show that PCT NTDs clearly affect productivity , although the actual impact depends on the type and severity of the NTD as well as on the context where the disease occurs . Variation in estimated productivity loss is also caused by differences in research methods . Publications should provide enough information to enable readers to assess the quality and relevance of the study for their purposes . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"and",
"health",
"sciences",
"tropical",
"diseases",
"database",
"searching",
"parasitic",
"diseases",
"bacterial",
"diseases",
"filariasis",
"eye",
"diseases",
"neglected",
"tropical",
"diseases",
"onchocerciasis",
"lymphatic",
"filariasis",
"blindness",
"research",
"and",
"analysis",
"methods",
"infectious",
"diseases",
"helminth",
"infections",
"schistosomiasis",
"visual",
"impairments",
"database",
"and",
"informatics",
"methods",
"ophthalmology",
"trachoma",
"soil-transmitted",
"helminthiases"
] | 2016 | Productivity Loss Related to Neglected Tropical Diseases Eligible for Preventive Chemotherapy: A Systematic Literature Review |
Homologous recombination ( HR ) mediates one of the major mechanisms of trypanosome antigenic variation by placing a different variant surface glycoprotein ( VSG ) gene under the control of the active expression site ( ES ) . It is believed that the majority of VSG switching events occur by duplicative gene conversion , but only a few DNA repair genes that are central to HR have been assigned a role in this process . Gene conversion events that are associated with crossover are rarely seen in VSG switching , similar to mitotic HR . In other organisms , TOPO3α ( Top3 in yeasts ) , a type IA topoisomerase , is part of a complex that is involved in the suppression of crossovers . We therefore asked whether a related mechanism might suppress VSG recombination . Using a set of reliable recombination and switching assays that could score individual switching mechanisms , we discovered that TOPO3α function is conserved in Trypanosoma brucei and that TOPO3α plays a critical role in antigenic switching . Switching frequency increased 10–40-fold in the absence of TOPO3α and this hyper-switching phenotype required RAD51 . Moreover , the preference of 70-bp repeats for VSG recombination was mitigated , while homology regions elsewhere in ES were highly favored , in the absence of TOPO3α . Our data suggest that TOPO3α may remove undesirable recombination intermediates constantly arising between active and silent ESs , thereby balancing ES integrity against VSG recombination .
Trypanosoma brucei proliferates in the bloodstream of its mammalian host and periodically escapes the antibody-mediated immune response . A single species of variant surface glycoprotein ( VSG ) is expressed at a given time , from among >1 , 000 VSG genes and pseudogenes [1] , [2] , and ∼10 million VSG molecules homogenously coat the surface of a parasite . Switching the expressed VSG causes antigenic variation ( reviewed in [3]–[5] ) . VSG genes are found in 15 expression sites ( ESs ) — polycistronic transcription units that are transcribed by RNA Polymerase I [3] , [6]–[8] — of the Lister 427 strain [9] . These VSGs are located 40–60 kb downstream of their ES promoters and are flanked by 70-bp and telomere repeat sequences . Several expression-site-associated genes ( ESAGs ) with mostly unknown functions , and ESAG and VSG pseudogenes , are located between the promoter and the 70-bp repeat region . Only one ES is transcriptionally active at any time and the rest are silent . Many VSGs are found upstream of telomere repeats in minichromosomes but most are thought to reside in ‘telomere-distal’ arrays . Minichromosomal and telomere-distal VSGs lack promoters , but small numbers of 70-bp repeats are present upstream of these VSGs . By analyzing switched variants , two major pathways of antigenic switching have been identified in T . brucei: in situ ES transcription switching and recombination-mediated switching [4] , [5] , [10] . In situ switching occurs by silencing the active ES and activating a silent ES , without DNA rearrangement [11] , [12] . Recombination-mediated switching occurs mainly by gene conversion ( GC ) and can involve just the VSG or larger regions of the ES . VSG GC can occur by recombination between the active VSG and a silent ES-associated VSG , a minichromosomal VSG , or a telomere-distal VSG [13]–[18] . Gene conversion between larger regions can result in the duplication of an entire ES , including its VSG [12] . Crossover switches , where two VSGs are exchanged , have also been observed infrequently [19]–[22] . Deficiency of RAD51 or RAD51-3 ( RAD51-related gene ) , or BRCA2 , a mediator for RAD51 filament formation , decreased switching frequency in T . brucei [23]–[25] . Mre11 is essential for DNA damage response , as a sensor of double strand breaks ( DSBs ) that can be repaired by homologous recombination ( HR ) or non-homologous end joining ( NHEJ ) [26]–[28] . As in yeast and mammals , T . brucei mre11 null mutants exhibited growth defects , hypersensitivity to a DNA damaging agent , and gross chromosomal rearrangements ( GCR ) , but no detectable decrease in VSG switching [29] , [30] , indicating that , although antigenic variation shares core features with classic HR , specific roles for recombination factors in antigenic variation remain to be determined . Mitotic crossover can be detrimental , leading to unequal exchanges . Sgs1 , a RecQ family helicase in yeast , is one of the major factors that control spontaneous crossovers [31] . Sgs1 forms a complex with Top3 ( type IA topoisomerase ) and Rmi1 ( RecQ-mediated genome instability ) , and plays major roles in the suppression of genome instability by influencing mitotic and meiotic recombination , replication fork stability , and telomere maintenance [32]–[38] . At least one mechanism of crossover-suppression appears to involve ‘dissolution’ of double Holliday Junction ( dHJ ) intermediates . Sgs1-Top3-Rmi1 , also known as the RTR ( RecQ-Top3-Rmi1 ) complex , is well conserved in humans as the BLM ( Bloom mutated ) -TOPO3α-BLAP75/18 ( Bloom associated protein 75kDa/18kDa , or RMI1/2 ) . Mutations in any member of the RTR complex increase recombination frequency and crossover [31] , [32] , [39]–[43] . Defects in the BLM pathway are associated with elevated levels of sister chromatid exchanges ( SCEs ) , chromosomal breaks and translocations [40] , [41] , [44]–[46] . Crossover has rarely been observed in VSG switching . Suppression of crossover is intriguing because , in principle , the outcome of duplicative VSG conversion holds no apparent advantage over crossover events , as re-expressing a VSG , either exchanged or duplicated , will be lethal in vivo . Given the similarities between HR and VSG switching , we hypothesized that certain yeast hyper-recombination mutants could be hyper-switchers in trypanosomes . Using new recombination and VSG switching assays , we took advantage of a potential member of T . brucei Sgs1 pathway , TbTOPO3α ( Tb11 . 01 . 1280 ) , to get better insights on how trypanosomes employ recombination factors to control antigenic variation .
Type IA topoisomerases cleave DNA by covalent attachment of one of the DNA strands through a 5′phosphodiester bond to a tyrosine residue in their catalytic domains [47] . In many organisms , type IA topoisomerases function in cooperation with helicases , as a combination of Top3-Sgs1 in yeasts and TOPO3α-BLM in humans . T . brucei expresses a 102 . 5-kDa TOPO3α protein with 918 amino acids . Figure 1 shows an alignment of TbTOPO3α with human TOPO3α and S . cerevisiae and S . pombe Top3 . The primary sequences are well aligned at the N-terminal catalytic domain including the active site tyrosine . Both E . coli Top1 and human TOPO3α contain Zn-binding motif ( s ) in their C-terminal regions . E . coli Top3 and two yeast Top3 lack a Zn-binding domain ( reviewed in [47] ) . TbTOPO3α seems to have a Zn-binding motif in the C-terminus ( four cysteine residues written in red ) , although this region does not align well with human TOPO3α . The sequences of TOPO3α are very well conserved in T . brucei , T . cruzi and Leishmania major ( Supporting Figure S1 ) . T . brucei also has a type IA TOPO3β ( http://www . genedb . org/genedb/tryp ) , but its function has not been studied . To explore the role of TOPO3α , we sequentially deleted both alleles . We used deletion-cassettes containing hygromycin ( HYG ) or puromycin ( PUR ) resistance genes fused to Herpes simplex virus thymidine kinase ( HSVTK or TK ) and flanked by loxP sites , allowing the markers to be removed by transient expression of Cre-recombinase and reused [48] . Deletion of both alleles was confirmed by PCR analyses ( Supporting Figure S2 ) . Loss of Top3 causes a severe growth defect in budding yeast and is lethal in fission yeast [43] , [49] . The absence of TOPO3α or TOPO3β results in embryonic lethality or shortened life span in mice [50] , [51] . In contrast , TOPO3α null mutants exhibited only a minor growth defect in T . brucei ( Figure 2A ) . Yeast Top3 is important for the maintenance of genome integrity . top3 mutants are sensitive to DNA-damaging agents and show defects in the activation of the cell-cycle checkpoint kinase Rad53 ( CHK2 in mammals ) , in response to genotoxic stresses [52]–[54] . We therefore asked whether T . brucei TOPO3α is required for the DNA damage response , by assessing sensitivity to the DSB-inducing agent phleomycin or the replication inhibitor hydroxyurea ( HU ) . Cells were treated with phleomycin for 24 hours and single cells were distributed in 96-well plates . The color of the medium turns from red to yellow when the culture becomes saturated . Yellow wells were counted after 7–8 days and the percent viability was calculated by normalizing to the untreated samples . In the null mutants , viability was reduced by 3-fold at 0 . 3 µg/ml and 10-fold at 0 . 6 µg/ml phleomycin ( Figure 2B ) . Viability of the HU-treated null mutants was reduced by 3-fold in 0 . 04 mM HU ( Figure 2C ) . topo3α−/+ was comparable to the wild type in both experiments . We conclude that TOPO3α is required for the response to DNA damage and replication block , similar to the roles of yeast Top3 . top3 was isolated as a hyper-recombination ( ‘hyper-rec’ ) mutant in a genetic screen designed to identify mutations that increase recombination frequency at SUP4-o locus in budding yeast [43] . We therefore hypothesized that Tbtopo3α could be a ‘hyper-rec’ mutant and this phenotype could be reflected in the frequency of recombination-mediated antigenic switching . To test whether TOPO3α deficiency increases recombination frequency , we established a new recombination assay . Thus far , transfection-based recombination assays have been predominantly used , in which trypanosomes are transfected with linear DNA containing a selection marker flanked by targeting sequences , and the recombination frequency is calculated from the number of drug-resistant clones that arise . Although this method can give reliable measurements , it requires a high rate of recombination at the target site and is subject to variations in transfection efficiency . To allow a more convenient , natural and reliable measure of recombination efficiency , we established an assay ( Figure 3A ) in which HYG-TK can replace one allele of what we will call TbURA3 ( the bifunctional orotidine-5-phosphate decarboxylase/orotate phosphoribosyltransferase Tb927 . 5 . 3810 ) on chromosome V . The frequency of loss of either the HYG-TK or TbURA3 allele represents the rate of gene conversion at this locus . The frequency of HYG-TK loss can be measured with gancyclovir ( GCV ) , a nucleoside analog , as only the cells that had lost the TK gene can grow in the presence of GCV . The loss of TbURA3 can be measured with 5-FOA ( 5-fluoroorotic acid ) , as only the ura3− cells can grow in the presence of 5-FOA . To remove the HYG-TK and PUR-TK markers that were used for the deletion of TOPO3α , Cre-recombinase was transiently transfected into the topo3α−/− cells and GCVR HYGS PURS clones were selected ( Supporting Figure S2 ) . One allele of TbURA3 was then replaced with HYG-TK and the targeting was confirmed by PCR . Gene-conversion frequencies were determined by counting total GCRR and FOAR cells , in three wild-type and five topo3α−/− independent HYG-TK clones . As shown in Figure 3B , Tbtopo3α gave indeed a hyper-recombination phenotype . Total gene-conversion frequency was increased 6-fold in topo3α−/− ( 5 . 12±0 . 15×10−5 ) compared to the wild type ( 0 . 87±0 . 70×10−5 ) . To investigate the roles for TOPO3α in VSG switching , we generated a VSG switching reporter strain in which we could easily measure switching frequency and score different switching mechanisms . As illustrated in Figure 4 , the parental strain expresses VSG 427-2 ( 221 ) in ES1 , which was doubly marked with a blasticidin-resistance gene ( BSD ) downstream of the promoter and PUR-TK at the 3′ end of the 70-bp repeat region , without disrupting the co-transposed region ( CTR ) , disruption of which has been shown to induce rapid VSG switching [55] . The 5′ boundaries for recombination-mediated VSG switching have been mapped at regions upstream of CTRs that are located between the 70-bp repeats and the VSG . Therefore , the PUR-TK gene will either be lost or repressed in switched cells . This will allow switchers , but not the parental cells , to grow in the presence of GCV . Doubly marked wild-type and topo3α−/− cells were maintained in media containing blasticidin and puromycin , to exclude switchers from the starting population . The cells were allowed to switch in the absence of drugs for 3–4 days . Un-switched VSG 427-2-expressing cells were depleted by magnetic-activated cell sorting ( MACS ) [56] . The column flow-through , highly enriched with switchers , was serially diluted in medium containing 4 µg/ml GCV and distributed into 96-well plates . Switching frequency was determined as the ratio of GCV-resistant cells to the total number of cells prepared for the MACS column experiments . We analyzed three independent wild-type cultures and four topo3α−/− cultures . As shown in Figure 5A , TOPO3α deficiency caused a 10–40-fold increase in switching frequency ( 26±16×10−5 ) compared to wild type ( 1 . 01±0 . 45×10−5 ) . This is the only known example of an increase in VSG switching frequency when a repair factor is deleted . To confirm that the column-mediated depletion of VSG 427-2-expressing cells was not biasing our results , other batches of cells were directly diluted in GCV-containing media and distributed into 96-well plates . Switching frequency was 10–30-fold increased in the absence of TOPO3α ( data not shown ) . We have determined the switching frequency in a strain without the TK marker but with a PUR marker inserted downstream of VSG 427-2 and obtained similar frequencies , ∼1×10−5 , in wild type . In two different but closely related cell lines , with the same genotype except that one line has PUR-TK inserted at the 70-bp repeat and the other just PUR , again similar switching frequencies , ∼1×10−5 , were observed [56] ( personal communication with Nina Papavasiliou ) . Reintroduction of wild-type TOPO3α complemented the hyper-switching phenotype of topo3α−/− ( −/−/+ in Figure 5B ) , confirming that this phenotype is associated with the TOPO3α deficiency . The results were obtained from three complemented clones ( −/−/+ ) and two cultures each of wild type and topo3α mutant . RAD51-dependent recombination intermediates accumulate in top3 mutants and the removal of persistent intermediates requires the cleavage activity of Top3 [57] , [58] . We examined whether the hyper-switching phenotype of topo3α−/− is dependent on RAD51 . Both RAD51 alleles were sequentially deleted in the wild-type and topo3α−/− strains . We analyzed four independent cultures of rad51−/− and two of topo3α−/− rad51−/− . RAD51 deletion reduced the switching frequency of the wild type by 2-fold and abolished the hyper-switching phenotype of topo3α−/− ( Figure 5C ) . Collectively , we concluded that TOPO3α functions as an important regulatory factor for recombination-mediated VSG switching and that , in the absence of TOPO3α , recombinogenic structures may accumulate between the active ES and VSG donors , and could then be resolved to give rise to switched variants . In other organisms , Top3 defects are associated with elevated crossover as well as hyper-recombination [32]–[34] , [45] , [46] . To learn how individual switchers had undergone antigenic variation , we analyzed total 296 cloned switchers . The rationales for the double marking of parental cells are as follows ( Figure 4 ) . First , switchers can be effectively counter-selected using GCV ( Figure 4 and 5 ) . Second , transcription is initiated at silent ESs but elongation is prematurely terminated [59]: genes that are located closer to silent ES promoters are not completely silenced . Therefore , in-situ switchers can be distinguished from recombination-mediated switchers using different concentrations of blasticidin . Based on our titration for blasticidin concentration , in-situ switchers can grow in 5µg/ml blasticidin but not in 100 µg/ml , while ES gene conversion ( ES GC ) switchers cannot grow in either concentration . VSG gene conversion ( VSG GC ) and VSG-exchange ( crossover ) switchers will be resistant to 100 µg/ml blasticidin , and these alternatives can be distinguished by the absence or presence of VSG 427-2 , respectively , which can be analyzed by PCR . The strategies to score individual switching mechanisms are summarized in Table 1 and examples of PCR analyses are shown in Figure 6A ( right ) . We analyzed cloned switchers isolated from six independent cultures and were able to discriminate among the alternative switching mechanisms . The results are summarized in Table 2 and Figure 6A . Switchers from cultures 1 and 2 were isolated by the column method and switchers from culture 3 by directly plating in GCV . Switching occurred largely by gene conversion ( Figure 6A ) . In both wild type and topo3α mutants , 64∼77% of switching exploited VSG GC . Crossovers were rare in wild type ( ∼3% ) but , on average , 20% of switchers exchanged their VSGs in topo3α−/− . These data suggest that , in the absence of TOPO3α , recombination intermediates may be accumulated and these could be repaired mostly by duplicative VSG GC and crossover . In a previous study designed to examine in-situ switching , using a cell line with TK marker inserted next to the active ES promoter , frequent loss of entire active ES was observed . This could be caused by duplicative transposition of a silent ES ( ES GC ) or by deletion of the active ES coupled with transcriptional activation of a silent ES [60] . In our experiments , ES GC and ES loss cannot be distinguished , as switchers that lost both BSD and VSG 427-2 genes could arise either by duplicative transposition of a silent ES or by ES breakage coupled with an ES transcriptional switch . The ‘ES GC or ES loss’ events were rather frequently detected in wild-type cells ( average ∼30% ) , while they were either not detected ( culture 1 and 2 ) or detected at a low frequency ( 4 our of 51 cloned switchers in culture 3 ) in the absence of TOPO3α ( Figure 6A and Table 2 ) . Interestingly , RAD51 deletion significantly decreased ‘ES GC or ES loss’ frequency ( unpublished data ) , indicating that ‘ES GC or ES loss’ events are mainly under the control of RAD51-dependent recombination . We noticed that some switched variants had growth disadvantages . Depending on how long it took to saturate the medium , wild-type switchers were categorized as ‘fast’ , ‘medium’ or ‘slow’ . ‘ES GC or ES loss’ switchers were prevalent in clones that grew up more slowly ( data not shown ) . The functions of ESAGs are mostly unknown , but expressing different ESAGs might be advantageous when entering different hosts [61] . The slower-growth phenotype of some of these switchers may reflect impaired function of one or more ESAGs in the bovine serum-containing culture medium , which appears to favor stable transcription of the VSG 427-2-containing ES1 . In-situ switchers were rare in our assay . This phenotype is different from previous reports [24] , [62] , for reasons we do not understand . In our hands , in-situ switchers generally grew slower than VSG-GC switchers , so VSG-GC switchers would quickly take over the switched population if it was initially mixed , although this is unlikely because our switching population was initiated at 500–1000 cells/ml , while it was at 5 , 000–10 , 000 cell/ml in previous assays . Before this seeding , cells were grown in the presence of drugs that prevented switching . The 70-bp repeat unit has been proposed to be a recombination hot spot , possibly as a potential target for a site-specific endonuclease playing a similar role to that of the HO-endonuclease in yeast . Such an endonuclease has not been identified in trypanosomes . The 70-bp repeats could serve as switching hot-spots because of their structural features [63] , rather than require cleavage by a specific endonuclease . Early experiments suggested that the overall VSG switching-frequency was not reduced in the absence of 70-bp repeats or by inversion of the repeats although , when present in the correct orientation , the repeats were used more than 10% of the time [62] . More recently , however , it has been shown that the 70-bp repeats of the actively transcribed ES are prone to break , which could induce recombination-mediated switching , and that the switching frequency was greatly increased when breaks were experimentally induced at the 70-bp repeats , but not when induced elsewhere in the ES or in the absence of 70-bp repeats [56] . We mapped the region where the recombination occurred ( or resolved ) in the VSG-GC switchers from wild type and topo3α mutants , to learn whether the 70-bp repeat unit is the hot spot of duplicative VSG GC and whether TOPO3α can redirect this preference . ESAG1 genes are located immediately upstream of the 70-bp repeats , and their sequence polymorphisms allowed us to design ES1-specific-ESAG1 oligonucleotides for PCR analysis . PCR results from several VSG-GC switchers were shown in Figure 6A ( right ) . The presence of ES1-specific ESAG1 in VSG-GC switchers indicates that gene conversion occurred at 70-bp repeat regions , and its absence indicates that recombination occurred upstream of ESAG1 ( Figure 6B ) . Crossover and ‘ES GC or ES loss’ switchers were used to verify that the PCR primer set was amplifying only the ES1-specific ESAG1 gene . The ES1-specific ESAG1 was lost in all ‘ES GC or ES loss’ switchers but was detected in all crossover switchers examined , as expected . The ES1-specific ESAG1 gene was amplified in ∼63% of VSG-GC switchers in wild-type cells but ∼81% of VSG-GC switchers lost the ES1-specific ESAG1 gene in topo3α−/− , indicating that , in the absence of TOPO3α , the active ES recombined mostly with silent ESs upstream of ESAG1 , rather than within the 70-bp repeats , but not with minichromosomal or telomere-distal VSGs . We concluded that the 70-bp repeat region is an important but not an essential element for recombination-mediated switching . Gene conversion upstream of 70-bp repeats , at ESAG2 , has also been reported [64] . The primary function of TOPO3α may be to prevent accumulation of recombination intermediates constantly arising between the active and silent ESs , to maintain the integrity of ESs . Recombination by a one-strand invasion event could replace VSGs by break-induced replication ( BIR ) [56] . Alternatively , a second strand invasion at homologous sequences within or downstream of the VSG could generate VSG-GC switchers . Duplication of a telomere-distal VSG into an active ES is a relatively rare event , at least in the modest extent to which switching events have been characterized experimentally , but it appears to serve as an important switching mechanism in later stage of infection and as a mechanism to further expand the expressed VSG repertoire [22] , [65] , [66] . The few telomere-distal VSG arrays so far characterized contain only short stretches of 70-bp repeats but lack telomeric repeats . To determine how VSG GC occurred , we analyzed the sequences downstream of the 3′ homology region of VSG 427-2 by PCR in all VSG-GC switchers ( Supporting Figure S3 ) . If the second strand invaded at this 3′ homology region , downstream sequences should be unchanged . We found , however , that the ES1-specific downstream sequences were lost in all the VSG-GC switchers obtained from wild-type and topo3α cells , indicating that VSG-GC switchers were most likely repaired by BIR , consistent with a recent report [56] , and that internal-VSG duplication is extremely rare . PCR results from a selection of VSG-GC switchers were shown in Figure S3 . To confirm the duplicative translocation of newly expressed VSGs to the VSG 427-2 ES and to examine whether minichromosomal VSGs contribute to antigenic switching , 32 VSG-GC switchers from wild-type cells were further analyzed . Minichromosomes terminate with telomeres , VSGs and 70-bp repeats . Gene conversion with minichromosomal VSGs occurs frequently [56] , but only when recombination is initiated at the 70-bp repeats . Therefore , we cloned and sequenced newly activated VSGs from VSG-GC switchers that utilized 70-bp repeats . From 32 switchers that had undergone at least one type of switching , VSG GC at the 70-bp repeats , we obtained eight different newly activated VSGs ( Supporting Figure S5 , left ) . It is possible that we have underestimated the number of independent switching events as these switchers may have used different sequences within or near the 70-bp repeats , which should be counted as independent . Some switchers might have arisen earlier than others , for examples VSG 427-32 , as these were presented more often than others . Among these eight newly expressed VSGs , four were novel VSGs , 427-32 , 33 , 34 and 35 , full or partial sequences of which can be found in the following website ( http://tryps . rockefeller . edu ) . Switchers expressing VSGs 427-3 , 11 , 32 , 33 and 35 were examined by rotating agarose gel electrophoresis ( RAGE ) and Southern blot [56] . As shown in Supporting Figure S5 ( right panel ) , VSG 427-2 was lost in all the switchers and all newly expressed VSGs were duplicated and translocated to the 427-2 ES , except for 427-33 , an intermediate chromosomal ( IC ) VSG . The original copy of 427-33 may be lost after recombination . VSGs 427-32 and 35 came from megabase chromosomes ( MBC ) . We have not isolated any minichromosomal VSGs in these switchers , indicating that recombination between ES-associated VSGs was the major source for VSG switching .
As illustrated in Figure 7 , ES structures seem to play a particular role in VSG switching . ES-associated VSG genes are located between the 70-bp and telomeric repeats . ESAGs and some pseudogenes are present upstream of the 70-bp repeats in all ESs , sometimes duplicated and sometimes missing [3] , [9] . Strong sequence homologies are present throughout the ESs , with the exception of most of the VSG coding sequence and the immediately upstream ‘co-transposed region’ ( CTR ) . VSG sequences are highly dissimilar except for ∼200-bp encoding the C-terminus and within the 3′ UTR [67] . The reason why every VSG cassette contains a unique CTR is unknown . The purpose of CTR could be to insulate the individuality of VSG cassettes , so that the VSG sequences can evolve separately from other regions in ESs , which maintain their sequences to serve for VSG recombination . When HR occurs , the CTR could block branch migration of HJ or dHJ downstream of the 70-bp repeats . What roles does TOPO3α play in this scheme ? Our study shows that TOPO3α deficiency increases VSG switching , especially VSG GC and crossover , and that the hyper-switching phenotype requires RAD51 . The accumulation of toxic recombination intermediates accounts for the slow growth phenotype of yeast top3 mutants , which is suppressed by mutations in SGS1 or in the RAD51-pathway [43] , [68] , [69] . Recombination intermediates accumulate in cells over-expressing dominant-negative Top3-Y356F in response to methylmethane sulfonate in a RAD51-dependent manner [58] . The function of TOPO3α is not restricted to the 70-bp repeats in antigenic switching , as its absence appears to cause promiscuous recombination throughout the ESs . We therefore propose that TOPO3α removes recombinogenic structures constantly arising between ESs so as to maintain the albeit limited individuality of different ESs . In the absence of TOPO3α , recombination intermediates would accumulate during VSG switching and unresolved intermediates would have to be repaired either by GC associated with crossover or by placing a new duplicated VSG into the active ES by BIR ( Figure 7 ) . Suppression of crossover in recombination-mediated VSG switching is an interesting result , considering that there are probably more than 200 potential VSG donors: ∼20 ESs with extensive sequence homology and ∼200 minichromosomal VSGs . Antigenic variation probably requires balancing preservation and variation of VSG information , but we cannot explain how suppression of crossover would be important for maintaining this balance . However , we think that by favoring duplicative GC over crossover , rather than crossover over GC , trypanosomes could slowly accumulate VSG diversity without abrupt loss of their functionalities , because duplicative GC requires VSG DNA synthesis , during which point mutations could be incorporated into newly synthesized VSGs , but VSG crossover does not require VSG DNA synthesis . TOPO3α deficiency increased VSG GC far more than GC at the URA3 locus ( Figures 3 and 6 ) . GC at these two loci is probably mediated by different pathways . Recombination at URA3 locus would prefer flanking homologies , rather than BIR . In contrast , BIR would present a better option for VSG GC , as only one end homology appears to be involved ( supporting Figure S3 ) [56] . It is possible that a second invasion could occur within the telomere repeats , but this is impossible to determine . The higher VSG GC rate could also be because the active ES is less stable than URA3 locus . Alternatively , TOPO3α may specifically suppress BIR-mediated VSG switching . The role of TOPO3α in BIR has not been extensively characterized elsewhere . Our results show a novel function of TOPO3α in VSG switching , which could be an excellent system to study BIR . DNA recombination involves many factors , of which only a few have been studied in the context of antigenic variation: RAD51 , RAD51-related genes , BRCA2 , KU70/80 , MRE11 , and MSH2/MLH1 [23]–[25] , [29] , [30] , [70] , [71] . Among these , only the deletion of RAD51 , RAD51-3 , and BRCA2 decreased VSG switching , in wild-type cells that already had a very low switching rate . Our findings on TOPO3α in VSG switching suggest potential roles for numerous DSB-HR response factors in antigenic variation . Two RecQ family helicases are annotated in the T . brucei gene database ( http://www . genedb . org/genedb/tryp ) . Rmi1 is required to load Top3 onto the substrates and stimulate its activity through the physical interaction [72] . We have identified a TbRMI1 homologue . All the phenotypes that we have examined in Tbrmi1 mutants were identical to those in topo3α mutants ( unpublished data ) . Therefore , we believe that RecQ , TOPO3α and RMI1 are likely to function as a complex in antigenic variation in T . brucei . Synthetic-lethality screens with sgs1 in budding yeast identified three pathways working in parallel with Sgs1 [73]; Mus81-Mms4 , Slx1-Slx4 , and Slx5-Slx8 . Synthetic lethality of sgs1 mus81 or sgs1 mms4 requires HR factors [74] . Mus81-Mms4 is a structure-specific endonuclease that cleaves 3′ flap , replication fork , or HJ substrates [74]–[76] . Resolvase , an endonuclease , symmetrically cleaves HJs and the products can be resolved with crossover or non-crossover . Human and yeast resolvases have recently been characterized [77] . MUS81 appears to be present in T . brucei but a resolvase remains to be identified . Although we do not yet have functional data for these proteins , we propose , based on the studies from other organisms , that the regulation of antigenic variation is similar to that of mitotic HR . When present , TOPO3α could dissolve dHJs to prevent the ES instability , consequently generating non-crossover recombinants ( no switching ) . In the absence of TOPO3α , resolvase ( Figure 7a , grey box ) or MUS81 may cleave the accumulated recombination intermediates arising between the ESs and generate crossover switchers . Alternatively , stalled replication forks can be cleaved by MUS81 and the broken leading strand can invade a silent ES to generate VSG-GC switchers ( Figure 7b , grey box ) . Although VSG switching has similarities with mitotic HR , it appears that specific elements are present for its regulation . A hyper-recombination phenotype does not always correlate with hyper-switching phenotype . The mismatch repair ( MMR ) pathway can abort recombination during strand exchange between non-identical substrates and mmr mutants can increase recombination frequency ( reviewed in [78] ) . Consistent with their roles in repair and recombination , Tbmsh2 or Tbmlh1 mutants increased recombination frequency but did not change switching frequency [71] . Recombination is closely linked with DNA replication and checkpoint pathways as well [32] , [57] , [58] , [79] . Therefore , we believe that roles for DNA replication , checkpoint , and recombination factors and their interactions need to be determined to fully understand the mechanisms of antigenic variation . Measuring VSG switching has , until now , been time-consuming and not very reproducible . Our new switching assay circumvents previous technical difficulties and can effectively assign specific roles to individual proteins . It has recently been shown that a DSB introduced at the active 70-bp repeats by the I-SceI endonuclease causes a 250-fold increase in VSG switching and that the DSBs were repaired by BIR [56] . However , it is unknown whether the VSG switching is activated by targeted DSBs or by random chromosomal breaks , or whether recombinogenic ssDNA is a primary cause for the initiation of VSG switching . HR can be instigated by many different sources; random breaks , endonuclease cleavage at specific target sites , replication fork instability , unusual secondary DNA structure , or transcription . The Mre11 complex , which consists of Mre11 , Rad50 , and Xrs2 ( NBS1 in mammals ) , plays a central role in the DSB-HR response [26]–[28] . MRE11 deficiency , however , did not change the VSG switching frequency [29] , [30] , promoting the idea that ssDNA regions may generate recombinogenic structures for the initiation of switching . Uncoupling of leading and lagging strand DNA synthesis caused by DNA lesions can destabilize a replication fork , leaving ssDNA gaps behind the fork , which could be processed into recombinogenic structures . If an ssDNA gap is a major trigger for recombination-mediated switching , switching frequency should increase in cells suffering from replication challenge . To address this issue , we treated cells with aphidicolin , an inhibitor of lagging strand DNA synthesis , and HU , and measured the switching frequency in parallel ( Supporting Figure S4 ) . Cells were treated with the drugs at a sub-lethal dose to exclude a possibility of chromosome break-induced switching . No significant correlation was observed between these treatments and switching frequency . Therefore , an ssDNA gap may not be a major initiating factor for VSG switching . Rather , random breaks might be responsible for switching induction , consistent to a previous study [56] . However , it is still difficult to rule out the possibility that an ssDNA gap triggers switching , as ssDNA gaps might not be extensive enough to create recombinogenic structures at the low doses of aphidicolin or HU . The best way to test this hypothesis would be to use conditional mutants associated with replication defects . Unfortunately , we do not yet have such genetic tools , as nuclear DNA replication has not been studied in T . brucei . A high transcription level can stimulate recombination , a mechanism known as transcription-associated recombination ( TAR ) ( reviewed in [80] ) . Transcription has been shown to promote recombination in T . brucei [81] , [82] . Interestingly , it was shown in budding yeast that transcription- and DSB-induced recombination events were similar , indicating that transcription affects only the initiation of recombination , not the mechanism of recombination [83] . ssDNA regions exposed in the active ES during transcription could be readily accessible by recombination factors . Alternatively , transcription-replication collision causes replication fork stalling , which could also induce switching . Studies of mammalian cells have shown that TAR is dependent on replication [84] , and that transcription increases recombination frequency when a replication fork converges with transcription [85] . The active ES is more fragile than silent ESs [56] . The high level of transcription may explain why the active ES breaks more frequently , and this may induce VSG switching . The 70-bp repeat has been proposed to be a potential endonuclease target site to induce switching , but such an enzyme has not been found . Instability of the 70-bp repeat [63] may play a role in the initiation of switching and could lead to template switching . However , according to our results and previous studies [62] , [64] , switching is not completely dependent on the 70-bp repeats . With the available data , it would be reasonable to conclude that random breaks may occur throughout the active ES but more frequently at 70-bp repeats , and these could initiate various switching events . Gene conversion is used by several other pathogens , including Borrelia hermsii and Anaplasma marginale , as an evasion mechanism [10] , [86] . Our study suggests that exploring how trypanosomes manipulate the HR machinery to gain advantage against their host's immunity , while successfully preserving their genomes , may reveal weaknesses that can be exploited to control infectivity and virulence .
Trypanosoma brucei bloodstream forms ( strain Lister 427 antigenic type MITat1 . 2 clone 221a ( VSG 427-2 ) ) were cultured in HMI-9 at 37°C . The cell lines constructed for this study are listed in Supporting Table S1 , and they are of ‘single marker’ ( SM ) background that expresses T7 RNA polymerase and Tet repressor ( TETR ) [87] . Stable clones were obtained and maintained in HMI-9 media containing necessary antibiotics at the following concentrations , unless otherwise stated: 2 . 5µg/ml , G418 ( Sigma ) ; 5µg/ml , blasticidin ( Invivogen ) ; 5µg/ml , hygromycin ( Sigma ) ; 0 . 1µg/ml , puromycin ( Sigma ) ; 1µg/ml , phleomycin ( Invivogen ) . Plasmids used for this study are listed in Supporting Table S2 . TOPO3α genes were sequentially deleted using deletion-cassettes containing either puromycin or hygromycin-resistance gene fused with HSVTK , Herpes simplex virus thymidine kinase ( TK ) , PUR-TK and HYG-TK . These fusion genes are flanked by loxP sites so that the markers can be removed by transient expression of Cre recombinase ( pLew100-Cre ) . The entire open reading frame ( ORF ) of TOPO3α was deleted by transfecting ‘single marker’ ( SM ) cells with a deletion-cassette that was amplified with primer 35 and 36 using pHJ18 ( PUR-TK ) as a template . Primer 35 and 36 contains 70 nt homologies to the target sites . This topo3α ‘single knock-out’ cells ( sKO , HSTB-97 ) were used to PCR amplify a cassette containing a marker ( PUR-TK ) along with 453 nt upstream and 402 nt downstream sequences of TOPO3α gene . The PCR fragment was inserted into pGEM-easy-T vector by TA cloning to create pHJ63 . pHJ64 was constructed by replacing a PUR-TK marker with a HYG-TK from pHJ17 . topo3α ‘double knock-out’ ( dKO ) was generated by transfecting NotI-digested pHJ64 into topo3α sKO , HSTB-97 . Deletion of both TOPO3α alleles was confirmed by PCR analyses . To remove the selection markers , topo3α dKO cells were transfected with pLew100-Cre to transiently express Cre-recombinase , and the cells that lost both HYG-TK and PUR-TK were selected in 50µg/ml ganciclovir ( GCV ) . Loss of markers was confirmed by resistance to puromycin and hygromycin , and by PCR analysis . The sequences of primers used here are available upon request . pLHTL-pyrFE [48]-linearized by PvuII digestion was transfected into wild-type ( HSTB-188 ) and topo3α−/− ( HSTB-328 and HSTB-330 ) cells , to replace one allele of TbURA3 with HYG-TK . The integration was confirmed by PCR analysis with primers 48 and 49 . Three or five independent HYGR clones from wild-type or topo3−/− cells were analyzed . Cells were grown in the absence of hygromycin for 2 days to allow recombination to occur . Approximately 500 , 000 cells were diluted in HMI-9 media containing 30 µg/ml GCV or 6 µg/ml FOA , and distributed into 96-well plates . Yellow wells ( phenol red indicating acidification due to growth ) containing GCVR or FOAR cells were counted after 7–8 days of incubation and the GC frequency was determined . The sequences of primers used for genotyping are available upon request . To create a doubly-marked switching reporter strain ( Figure 4 ) , pHJ23 was linearized by KpnI-NotI digestion and integrated downstream of the ES1 promoter , to confer resistance to blasticidin . These cells were then marked with PUR-TK at the 3′ end of 70-bp repeats by transfecting a PCR-amplified PUR-TK cassette . Ten µg/ml of puromycin , 100 times higher than normal usage , was added to select clones targeted specifically at the active ES . When determining switching frequency , the parental cells were maintained in the presence of blasticidin and puromycin to exclude switchers from the starting population . Cells were then allowed to switch in the absence of selection for 3–4 days . Switchers were enriched using a MACS [56] . Flow-through enriched with switchers was collected and serially diluted in media containing 4 µg/ml GCV , and distributed into 96-well plates . The switching frequency was determined by counting GCVR clones . Alternatively , switching frequency was determined without the column enrichment step . Cells were diluted in GCV-containing media and directly distributed into 96-well plates . Non-switchers that carry spontaneous mutation ( s ) in TK gene but not in PUR were ruled out by examining puromycin resistance . Non-switchers that carry mutations in PUR and TK were ruled out by western blot analysis using antibodies against VSG 427-2 . To determine switching mechanisms , cloned switchers were analyzed for blasticidin sensitivity at 5 µg/ml and 100 µg/ml concentrations . Genomic DNA was prepared from 296 switchers and PCR-analyses were performed at four regions: BSD , VSG 427-2 , ESAG1 , and VSG 427-2 downstream . The primer set designed for BSD-PCR can also amplify TETR ( Tet Repressor ) gene , which was used as a control for PCR analyses . The sequences of primers used here are available upon request . Wild type ( SM ) , topo3α−/+ ( HSTB-97 ) , and topo3α−/− ( HSTB-226 and HSTB-227 ) cells were incubated with indicated concentration of phleomycin for 24 hours . The same number of cells was distributed into 96-well plates . All the plating was duplicated . The wells that contain viable cells were counted after 7–8 days of incubation at 37°C and the viability was calculated by normalizing to untreated samples . Sensitivity to HU and aphidicolin was determined similarly . Cells were incubated with HU or aphidicolin for 2 or 3 days . The viability was calculated by normalizing to untreated samples . Database ID numbers ( http://www . genedb . org and http://tritrypdb . org ) for TOPO3α discussed in this paper are Tb11 . 01 . 1280 , LmjF36 . 3200 and Tc00 . 1047053511589 . 120 . What we refer to as TbURA3 is the bifunctional orotidine-5-phosphate decarboxylase/orotate phosphoribosyltransferase Tb927 . 5 . 3810 . | Trypanosoma brucei , the causative agent of African sleeping sickness , escapes the host immune response through a mechanism known as the antigenic variation . Each individual trypanosome expresses a single species of surface antigenic protein at any time yet possesses an infinite potential to express different surface antigens by transcriptional and recombinatorial switching . Periodic switching to a different antigen allows parasites to escape the antibody-mediated host immune response and causes chronic infection , eventually overwhelming the host's immune system and leading to death . DNA recombination factors are critical for the protection of chromosome integrity . One of the major antigen-switching mechanisms exploits particular recombination pathways to achieve its purpose . We have used a new switching assay to study a regulator of recombination and to demonstrate that antigenic variation is a complex mechanism balancing chromosome integrity and antigen diversity by suppressing and promoting particular recombination events . Recombination is used in evasion or virulence mechanisms by several pathogens . Exploring how Trypanosoma brucei manipulates the recombination machinery to gain advantage against their host will help us understand pathogenesis in various organisms and may reveal weaknesses that can be exploited to control infectivity and virulence . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"genetics",
"and",
"genomics/genetics",
"of",
"disease",
"molecular",
"biology/recombination",
"molecular",
"biology/dna",
"repair",
"infectious",
"diseases/neglected",
"tropical",
"diseases"
] | 2010 | TOPO3α Influences Antigenic Variation by Monitoring Expression-Site-Associated VSG Switching in Trypanosoma brucei |
Leishmaniasis is an intracellular parasitic infection transmitted to humans via the sandfly . Approximately 350 million people are at risk of contracting the disease and an estimated 1 . 6 million new cases occur annually . Of the two main forms , visceral and cutaneous , the visceral form is fatal in 85–90% of untreated cases . This literature review aims to identify and evaluate the current evidence base for the use of various preventative methods against human leishmaniasis . A literature search was performed of the relevant database repositories for primary research conforming to a priori inclusion and exclusion criteria . A total of 84 controlled studies investigating 12 outcome measures were identified , implementing four broad categories of preventative interventions: animal reservoir control , vector population control , human reservoir control and a category for multiple concurrently implemented interventions . The primary studies investigated a heterogeneous mix of outcome measures using a range of different methods . This review highlights an absence of research measuring human-specific outcomes ( 35% of the total ) across all intervention categories . The apparent inability of study findings to be generalizable across different geographic locations , points towards gaps in knowledge regarding the biology of transmission of Leishmania in different settings . More research is needed which investigates human infection as the primary outcome measure as opposed to intermediate surrogate markers , with a focus on developing a human vaccine .
Leishmania is of the protozoan genus trypanosomatida . The parasite resides intracellularly causing the disease leishmaniasis , and is transmitted between hosts by the bite of the female sandfly ( genus Phlebotomus in the Old World localities of Europe , Africa and Asia , and Lutzomyia in the New World – Americas and Oceania ) . The primary hosts are vertebrates and commonly infected animals include humans , domestic dogs and cats , opossums , the crab-eating fox and the common black rat [1] . The sandfly comprises five genera and over 700 species . Approximately 30 species are thought to be implicated in transmission of Leishmania parasites [2] . Sandflies are found around human settlements and breed in organic matter such as leaf litter , manure and in rodent burrows [3] , and are approximately a third of the size of mosquitos , measuring between 2–3 mm in length [4] . They are often categorised by virtue of where they bite ( being classified as either endophagic ( biting indoors ) or exophagic ( biting outdoors ) ) , as well as where they rest ( either endophilic ( resting indoors ) or exophilic ( resting outdoors ) ) [5] . In humans , the clinical forms of the leishmaniases are broadly categorised into visceral leishmaniasis ( VL ) and cutaneous leishmaniasis ( CL ) ( cutaneous forms presenting in a spectrum ranging from cutaneous , mucosal and diffuse cutaneous ) [6] . Leishmania occurs in five continents and is endemic in 98 countries [7] . The World Health Organisation ( WHO ) estimate that 350 million people are at risk of contracting leishmaniasis [6] . Approximately 58 , 000 cases of visceral leishmaniasis and 220 , 000 cutaneous cases are officially reported each year . However it is thought that only around two thirds of countries actually report incidence data , with the sparsest data from Africa [7] . Based on assessments of under-reporting , 0 . 2–0 . 4 million new cases of VL and 0 . 7–1 . 2 million new cases of CL are estimated to occur every year [7] . There are several reasons for under-reporting of official Leishmania cases [2]; After malaria and African trypanosomiasis ( sleeping sickness ) , the leishmaniases are the third most important vector-borne disease and are ranked ninth in terms of global burden of disease of all infectious and parasitic diseases [8]; accounting for more than 57 , 000 deaths per year and an estimated 2–2 . 4 million Disability Adjusted Life Years ( DALYs ) lost [8] , [9] . When discussing the eco-epidemiology of leishmaniasis , there are considered to be four main forms of disease which are named with respect to the mode of transmission; Zoonotic Visceral Leishmaniasis ( ZVL ) , Anthroponotic Visceral Leishmaniasis ( AVL ) , Zoonotic Cutaneous Leishmaniasis ( ZCL ) and Anthroponotic Cutaneous Leishmaniasis ( ACL ) . In anthroponotic forms , which are mostly located in Old World foci , humans are considered to be the main reservoir for infection whereas in zoonotic forms ( mainly in New World areas such as Brazil ) , animals are thought to be the major source of parasites [10] . Following the 2007 World Health Assembly , where a resolution by member states to improve research on prevention , control and management of leishmaniasis was approved , the WHO convened the Expert Committee on Leishmaniasis in March 2010 . The resulting technical report “Control of the Leishmaniases” [6] was the first updated report in more than 20 years . It details recommendations for diagnosis , treatment and prevention . Although this report makes a start on describing the problem , as well as bringing together and directing global efforts to reduce the burden of the Leishmaniases , it highlights the fact that the disease is very complicated . There are various different diagnostic modalities , treatments and preventative methods , being more or less useful depending on the Leishmania species , the vector characteristics , the host immunity levels , the main reservoir , as well as the socio-economic and political makeup of the locality . The fact that the true burden of disease is not accurately known ( due in part , to reliance on passive case detection , as well as an array of issues with different diagnostic modalities ) , further complicates any efforts to deploy efficient methods of prevention and disease management and to secure funding for research . Human VL , is mainly caused by two species of Leishmania parasites , each having a characteristic regional distribution , as described by Gill & Beeching [11] VL may also be caused by L . tropica in the Old World and L . amazonesis in the New World , and is fatal in 85–90% of untreated cases and up to 50% of treated cases [11] . Approximately 90% of CL occurs in Afghanistan , Pakistan , Syria , Saudi Arabia , Algeria , Islamic Republic of Iran , Brazil , and Peru [12] . CL occurs in a spectrum of clinical presentations ranging from ulceration of the skin only ( cutaneous ) , to various degrees of mucosal involvement ( diffuse cutaneous and mucocutaneous ) . Distinct sub-genii of Leishmania are thought to cause different cutaneous clinical presentations , usually divided into Old and New World regions . However Leishmania species which are implicated in VL can cause cutaneous disease ( and vice versa , especially when the individual has co-infections ) [11] . Mortality associated with CL is not significant; however the morbidity , in the form of disfigurement , with subsequent social stigmatisation which arises from cutaneous lesions and the resulting scars is very important . In endemic areas many people have the belief that CL can be transferred through physical contact [12] resulting in restriction of social participation . Arguably , equally as important in terms of burden of disease as the health and economic effects , are the detrimental impacts on quality of life and mental health resulting from social stigma [13] . Clinical diagnosis of VL is often confused with other diseases such as malaria , schistosomiasis , African trypanosomiasis , miliary tuberculosis and malnutrition , and for CL , with tropical ulcers , leprosy and skin cancer [14] . Importantly , infection does not always result in clinical presentation of symptoms . The ratio of asymptomatic infections to clinical infections is thought to vary between 1∶2 . 6 to 50∶1 [15] . This presents a major problem for organisations relying on passive case detection when determining the true burden of disease and the size of the reservoir for future infections in areas of anthroponotic transmission . It is also interesting for future disease control developments to understand the genetically determined immunological factors that regulate clinical manifestation in humans [15] . The gold standard for confirmation of Leishmania infection is visualisation of parasites by microscopy; in a tissue smear such as a splenic aspirate , bone marrow or liver biopsy for VL [14] , and scrapings or fluid from cutaneous sores in the case of CL [3] . Potential complications with obtaining tissue smears and biopsies , as well as the need for specialist medical staff and equipment mean that less invasive but equally as sensitive and specific diagnostic tools are needed for diagnosis of VL . Polymerase Chain Reaction ( PCR ) detection of parasite DNA in blood or organs is highly sensitive and specific but the high cost and need for specialised equipment and staff limit its use to hospitals and research centres [15] . Serological diagnostic methods are increasingly being used for diagnosis of VL but are not suitable for diagnosis of CL . One major problem with these methods is that serum antibodies remain in the body after successful treatment for several years , and therefore relapses cannot be detected using the same method [15] . Another problem is that a proportion of the population of an endemic area will test positive for serum antibodies even though they have no history of clinical leishmaniasis due to asymptomatic infection [16] . This makes it difficult to differentiate asymptomatic infected individuals from successfully treated , and from apparently cured individuals who may relapse in the future . There are a range of serologic assays including ELISA , IFAT , DAT , rK39 . More recently new techniques including loop-mediated isothermal amplification ( LAMP ) , Nucleic Acid Sequence-Based Assay ( NASBA ) and Latex Agglutination Test ( KAtex ) have been developed but as yet have not been used in the trials reviewed here . Skin hypersensitivity tests such as the Montnegro or Leishmanin skin test ( MST/LST ) are used to detect cell-mediated immunity using intradermal injection of Leishmania antigen . A negative response is normally seen during active infection with VL , with a switch to a positive skin test after cure . A positive result is also seen after asymptomatic infection [6] . A lack of sensitivity ( 14% ) has been seen in LST for diagnosis of VL in India , limiting its utility [17] . The number of diagnostic tests available , as well as the variation in antibodies and antigens used in tests , coupled with the appearance of counterfeit immunochromatographic tests on the Indian subcontinent [6] , shows the need for not only more reliable field-appropriate diagnostics , but standardisation and regulation of those already in use . In areas of anthroponotic leishmaniasis , effective treatment of VL will help to decrease the human reservoir , however in areas of zoonotic transmission , treatment of humans will not solve the problem of potential human reinfection from an infected animal reservoir . With the lack of effective drugs , prohibitive treatment costs and the possibility of relapse and resistance , clearly there is a need for a more effective way to stop the cycle of infection in both zoonotic and anthroponotic transmission . Potential areas of intervention include targeting the vector or animal reservoir population , or attempting to either prevent humans being bitten , being infected , or from developing clinical symptoms . Figure 1 illustrates the potential areas of intervention . This literature review aims to provide a comprehensive account of the methods used to prevent human infection with Leishmania by performing a systematic search of all interventions aimed at reducing human disease incidence .
Medline , EMBASE , CENTRAL , Web of Science , LILACS and WHOLIS were searched using terms relating to the keywords “leishmania” , “leishmaniasis” , “kala azar” ( see Supplemental Materials Text S1 for full search terms for each database ) . Hand-searching of references of relevant studies and review articles was also performed and relevant articles retrieved . Numbers of studies retrieved , included and excluded were documented and recorded using a flow chart of stages of inclusion following the recommendations of the Preferred Reporting Items for Systematic Reviews and Meta-Analyses ( PRISMA ) group [18] ( Figure 2 ) . After all titles and abstracts resulting from the electronic and hand searching were assessed against the predetermined inclusion and exclusion criteria by use of a standardised study eligibility form , the full article was retrieved . The full paper was assessed against the inclusion/exclusion criteria . Any paper excluded at this stage was documented with a reason for the exclusion . In order to extract relevant data as systematically as possible , an electronic data extraction template form was used . Due to the anticipated heterogeneity of the study types and outcomes investigated , the template section headings were designed to be broad in order to capture as much information as possible . Based on scoping searches , it was not anticipated that meta-analysis would be possible due to the array of interventions and outcome measurements . Studies were analysed by narrative synthesis , making use of subgroup divisions based on which species ( human , animal reservoir or vector ) was the focus of the intervention . Within these groups there were further subdivisions based on the outcome measurements investigated . The primary measurement of interest was human infection rates . In many cases however , intermediate surrogate measurements were investigated which have differing applicability to potential changes in human incidence/prevalence of disease . Studies were grouped and analysed based on these outcome measures .
This section includes animal elimination , canine vaccines , use of insecticides on dogs; including insecticide-impregnated dog collars , spot-on insecticides ( drops applied to skin underneath hair on neck of dog ) , as well as whole-body insecticide use . A total of 34 studies were retrieved , of which; Only four of these studies investigated human-specific outcomes as a result of interventions directed at animal reservoir control . Table 1 provides a summary of human outcomes for animal reservoir control studies . One rodent culling intervention in Iran measured human CL by parasitological confirmation of skin lesion smear [28] . One insecticide-impregnated dog collar study investigated child VL seroconversion using DAT in Iran [29] . One dog culling intervention in Brazil studied human VL seropositivity using ELISA [30] , and another dog culling intervention identified paediatric VL cases by passive case detection of clinical disease using health records in Brazil [31] . Three of the four studies reported statistically significant reductions in human Leishmania infection between the intervention and control areas . Two of these were animal elimination programmes , and one was an insecticide-impregnated dog collar study . The only study to measure human infection using the gold standard of parasite visualisation was that of Ershadi et al . [28] who used poisoned grain to eliminate the rodent population around one intervention village and compared human infection to one control village where rodents were not eliminated . This study suffers from lack of generalizability not only to areas of anthroponotic transmission but also to areas of zoonotic transmission where transmission is believed to be driven by dogs . As with all the animal culling studies reviewed here , as well as all but one insecticide study , the report makes no reference to randomisation of areas . Ershadi et al . do however state that pre-intervention rates of infection were similar between the two areas investigated , but no data to support this assertion were presented . Pre-intervention numbers of active rodent burrows vary between intervention and control areas ( between two and 18 times more burrows were reported in the control area than in the intervention areas ) . Dietze et al . [30] reported that dog elimination in two valleys in Brazil did not result in a significant difference in human seropositivity as measured by ELISA when compared to one control valley . Although the major mode of transmission in Brazil is thought to be zoonotic ( with dogs as the major reservoir ) , the authors hypothesised a greater than previously thought role for humans as the significant reservoir for VL , adding to the uncertainty in the literature that dogs are the most important driver of transmission in areas historically considered zoonotic . The group used active case detection by serology census of humans , using ELISA at six and 12 months following culling of all seropositive dogs in two intervention areas and following no intervention in one control area . The authors did not specify if the eliminated dogs were domestic or feral and no reference to randomisation of intervention and control areas is made . Absence of data on initial numbers of dogs in each valley and how many were eliminated makes it difficult to know if the areas are comparable . If only domestic animals were studied , this would leave a large section of the feral dog population unaccounted for and if the feral population was included , this would make follow-up even more difficult . Ashford et al . [31] studied paediatric VL cases using passive case detection of clinical disease using health records following dog elimination . The authors used the measurement of cases per 1000 inhabitants , however the actual numbers of inhabitants ( and therefore the actual numbers of cases ) in each of the two neighbourhoods studied are not stated . In the four years of follow up , the authors reported nine cases per 1000 in the intervention area compared to 35 per 1000 in the control area . As with all studies using passive case detection , potential bias is introduced if one area is systematically better or worse at detecting and reporting cases . This may be linked to other factors such as educational level and socio-economic status of the population studied as well as quality and accessibility of healthcare . There is also a possibility , as with any intervention study , that increased activity focussing on preventative methods may influence the population to take other precautions against acquisition of the disease . None of the studies included in this section reported these kinds of population data . Interestingly , the decrease in incidence of dog infection , as measured by seroconversion , did not differ significantly between the two groups thus adding to the lack of clarity regarding the relationship between canine and human Leishmania infection . Like Dietze et al . [30] , Ashford et al . [31] do not specify if the elimination programme included domestic and/or feral dog populations . One of the major problems faced by groups carrying out any kind of animal elimination programme is the lack of control over the numbers of animals actually present in an area . In the case of the rodent elimination study , Ershadi et al . [28] used poisoned grain around rodent burrows . Active burrows were counted and treated every six months for three years if the number of re-opened burrows was 30% or more of initial numbers . Only once in three years did the researchers not need to re-bait burrows with poison meaning that burrows were re-opened rapidly and that the population of rodents may have attained original numbers quite quickly post-intervention . The same is seen of the dog elimination studies by Ashford et al . [31] and Diezte et al . [30] where follow-up of dogs was compromised by virtue of the dog population being both poorly understood through lack of regular censuses , and being dynamic with variable births/deaths/inward and outward movement . It is not clear from either dog culling study whether the elimination included all dogs or just domestic animals . The only other study in this section which investigated human infection was an insecticide-impregnated dog collar study by Gavgani et al [29] which used DAT to detect seroconversion in children . The group used a matched-cluster randomised trial of 18 villages paired on pre-intervention child VL prevalence . Collars were only fitted to domestic dogs , however the authors note that due to elimination of stray dogs being a generic disease control practise in Iran , the feral dog population is very small and was not considered an issue . Although the intervention is associated with a statistically significant decrease in child VL prevalence during the one year follow up , the actual numbers of seroconversions are low – 17 in the nine intervention villages , and 26 in the nine control villages , and because of this the authors recommend caution be used when interpreting the results of the study . The lack of studies measuring human infection following intervention resulted in only four out of 34 studies being included in this part of the discussion . Three of those reported a positive effect of the intervention they studied and one reported no statistically significant difference in the intervention and control group . Because of the small number of studies and the potential for bias , limited conclusions can be drawn as to the efficacy of interventions aimed at reducing animal reservoir infection with Leishmania on reducing disease burden in humans . In order to address this gap in knowledge , more and larger studies investigating human infection are needed . Ideally these would be cluster randomised in order to attempt to account for known and unknown confounders . Although the fundamental aim would be to reduce the burden of human disease , a secondary outcome would be to clarify whether the canine population is actually driving Leishmania transmission in areas historically described as zoonotic , or whether the human reservoir is more important . Without this evidence , it is not possible to determine if there is any use in allocating resources towards controlling animal reservoir populations . The search for vector population control interventions identified studies relating to indoor and area-wide insecticide spraying , and a range of different interventions broadly termed here ‘Environmental Management’ . A total of 22 studies were retrieved , of which; Of the 22 studies included in this section , only four investigated human-specific outcome measures following interventions directed at controlling the vector population . All of these involved insecticide spraying of houses and other buildings . Table 2 provides a summary of human outcomes for vector population control studies . Two studies in Brazil measured seroconversion by ELISA; one measured VL in children under 12 years of age [32] whilst the other included all age groups [33] . One study in the Peruvian Amazon measured human CL by clinical signs [34] , and one measured human CL by self-reporting in Afghanistan [35] . Two of the four studies reported statistically significant reductions in human Leishmania infection between the intervention and control areas , and two did not . As a general criticism of studies in this category , only four measured human outcomes , and none studied human infection by parasite visualisation . Only three studies used a cluster randomised study design , with the majority not randomising study areas at all . Both studies using the most robust diagnostic method ( serology ) reported no significant difference between the control and intervention groups . Souza et al . [32] studied VL in children under 12 years in three control and three intervention areas subject to intradomiciliary pyrethroid spraying every six months . The control group is described by the authors as ‘normal activity’ - however this is not defined , nor is it specified if the study area used any preventative interventions as part of their ‘normal’ activity . The researchers needed parental consent to enrol in the study; it is not reported if there was any systematic difference between families who allowed their children to be enrolled , compared to those who did not , or whether there were differences in enrolment between intervention and control areas . Children were tested serologically for leishmaniasis using ELISA every 12 months for 2 years . Two different pyrethroid insecticides are used but there is no mention of which were used when and where . No reference is made to randomization of areas although the authors do state that background infection rates were approximately equal in all three areas . Nery Costa et al . [33] used random allocation of 34×200 m2 areas into four interventions ( see also multiple interventions section ) in order to study VL in the population living in each area . The intervention area used an unspecified insecticide applied to the inside walls of houses and animal pens whereas the control area used an unspecified insecticide applied to the inside walls of houses only . Background VL prevalence was 42% in the intervention areas and 31% in the control area . It is not clear whether this initial difference was accounted for in the final analysis . Of the 213 seronegative individuals included in the study ( authors do not specify how many individuals were in each group ) , 120 ( 56% ) were followed up during the 6 month study . No indication of drop-out rates by intervention is given and so potential bias could be introduced if the majority of drop-outs occurred in one particular group . Davies et al . [34] used semi-randomisation ( some were randomised , some were matched based on pre-intervention measures ) of houses in 3 villages allocated to either intervention ( sprayed ) or control ( unsprayed ) arms . Spraying did not occur at the same time for all areas ( one village was first sprayed one year later than the other two ) . Houses allocated to the intervention group were sprayed at six-monthly intervals on four occasions for two villages and only three occasions for the third village . No explanation for differences in randomisation or spraying regimens was given . Cases were detected by active case finding with clinical diagnosis of active lesions . The group reported a significant difference in CL cases over two years between the control houses and the intervention areas ( 24 versus nine respectively ) ; the differences were more pronounced when cases detected within three months and then six months of the start of the study were excluded . The authors' rationale for removing cases diagnosed within three and six months is that cases within the first six months may reflect infection events which occurred prior to the commencement of the intervention programme . No evidence of existence of a period of time between infection occurring and a positive serology result ( pre-patent period ) is referenced by the authors . Such post hoc analytical decisions may introduce bias . There is evidence in the literature of varied infection rates over time in the same area [36] and epidemics occurring in certain years [37] . Therefore removing data without evidence of a significant pre-patent period of infection , as well as starting the interventions at different times for data that would later be pooled , may not have been appropriate . One general criticism relating to many of the studies included in this section is that reports exist of interventions ( especially residual insecticide spraying ) having the effect of directing sandflies into non-sprayed areas [38] . This could be a particular issue when the unit of study is a room or a house , or where areas allocated to intervention and control are directly adjacent . Sandflies are thought to be able to travel anywhere up to 960 m in 36 hours [39] . Therefore , adjacent areas may be more at risk of increased sandfly abundance , and therefore risk of infection , if an intervention has a repellent , but not harmful , effect on sandfly activity . Finally , Reyburn et al . [35] conducted a large ( 3666 participants ) cluster randomised trial of management of CL in Afghanistan , with four arms ( see also multiple interventions section ) . The sample size was chosen based on power calculations ( which none of the other studies reported ) and follow-up house visits were conducted at eight , 10 and 15 months post-intervention . The units studied were blocks of 10 houses . Like Nery Costa et al . 17 , Reyburn et al . , also included a safeguard against interventions overlapping into adjacent blocks by including ‘reservoir’ ( untreated and not included in the study ) blocks . All self-reported lesions were visually inspected before being reported as a case . The authors do state that the intention to confirm diagnosis by smear was not possible due to deterioration of the security situation in Kabul . All survey workers were blinded to the intervention , and a census was conducted prior to the programme to ensure areas were matched for pre-intervention prevalence . Intervention houses were sprayed only once during the 15 month study period and control and reservoir houses were also offered insecticide spraying of the living areas of their houses although the authors do not report how many houses accepted the intervention . This fact calls into question the validity of the control areas to which the intervention is being compared , however the authors report that since the concentration of the insecticide applied to control houses was so low , it likely did not have an effect on results . The group added that if sandflies had been diverted from intervention to control houses then the protective effect of spraying could have been over-estimated . A loss to follow-up of approximately 45% was reported due to the security situation , however possible bias introduced by this was not considered . Although Davies et al . , [34] do not specify that case detection was self-reported , it is difficult with CL diagnosis that relies on clinical signs , to be certain that all cases have been reported . Reyburn et al . , [35] rely on household questionnaire and follow up of clinical signs . Ultimately if someone wanted to hide the fact that they had a skin lesion and the lesion was not on an exposed part of the body , the information could be concealed from researchers , even when active case detection was employed . None of the papers included in this section , or indeed in this review , address the fact that CL diagnosis , especially because of the social stigma lesions can bring to the sufferer [13] , could be particularly susceptible to bias in case finding , in comparison to population-wide serological diagnostics . In summary , only four studies measure human outcome following insecticide spraying of houses and buildings . Two of these use robust diagnostic methodologies and show no difference between intervention and control , and two use less robust methods of diagnosis and show a statistically significant effect . None of the studies measuring environmental management study human-specific outcomes . Based on this evidence , there is not enough information to determine if insecticide spraying is advantageous in reducing the burden of disease in humans . In terms of generalizability , indoor insecticide spraying programmes are only useful where the sandfly is likely to come into contact with the walls that are actually sprayed , which necessitates endophagic or endophilic ( either biting or resting indoors ) sandfly species . It is not clear from any of the papers reviewed here that the authors took this into consideration . Further research would be needed in order to characterise the feeding , resting and breeding habits of different sandfly species in each endemic area , prior to implementation of a preventative intervention . In addition , more studies are needed which measure human infection after vector control interventions . Finally , there is evidence in the literature that sandfly resistance to insecticides is emerging , especially to DDT [4] which is still used in large-scale spraying programmes in India [40] . There is also evidence that spraying campaigns are more effective directly before transmission seasons which vary depending on location [40] . Care needs to be taken to ensure that spraying programmes are thoroughly investigated with respect to transmission seasons and sandfly activity in order to have the highest likelihood of being effective . Insecticide spraying is still used during leishmaniasis epidemics [37] and misuse may result in wide-spread resistance which would render them useless when they are needed most . This section includes studies describing personal protection measures for humans such as treated and untreated bed nets , barrier nets , insecticide-impregnated fabrics such as curtains , clothing and bed sheets , and use of soap containing insecticide . It also includes studies on human vaccines . Table 3 provides a summary of human outcomes for human reservoir control studies . Based on the search criteria , a total of 34 studies were retrieved and included in this section . Five studies which investigated insecticide-impregnated bed nets or other fabrics have multiple arms in order to examine multiple interventions , often including non-impregnated nets and fabrics as a control group . Because of this , there are more interventions than studies . One study [41] investigated the concomitant use of insecticide-impregnated bed nets and curtains and so will be included in the multiple interventions section . Due to the heterogeneity of study designs and outcome measures , a full and formal quality assessment of studies was not carried out . Although the variable methodological quality seen in the studies reviewed here is addressed in this discussion section , this was not carried out systematically . It may be advantageous for future work to systematically rate methodological quality and take this into consideration when reporting results . The authors are aware of one potentially relevant paper published in a Chinese language journal but were unable to obtain the reference . At the time of writing , published protocols highlight the intention to investigate aspects of prevention of Leishmania infection in humans [59]–[61] . To date however , there are no published reviews which address preventative methods in their entirety . This review provides a comprehensive overview of all interventions against human leishmaniasis , highlighting fundamental gaps in knowledge , and suggesting directions for future research . Four broad categories of preventative interventions were identified in this review , investigating a heterogeneous mix of outcome measures and using a variety of different methods . This review emphasizes the absence of high quality evidence demonstrating the impact of interventions on the prevalence or incidence of human Leishmania infection assessed using reliable diagnostic modalities . Research identified within this review includes intervention strategies ranging from protection of humans against infection , to interventions aimed one stage upstream of human infection ( targeting the sandfly vector ) , and even further , to interventions targeting animal reservoir species . Conflicting data on the impact of dogs in transmission of leishmaniasis to humans , along with lack of generalizability of interventions directed at vector control , point towards gaps in fundamental knowledge of the biology of transmission . Despite this weak evidence base , many countries continue to invest heavily in preventative methods focussed on control of leishmaniasis in dogs . Based on our current lack of understanding of the transmission of Leishmania , it seems salient to focus scant resources on prevention of human infection , as opposed to interventions which attempt to address upstream risk . The absence of a promising prophylactic human vaccine candidate , along with the scarcity of human vaccine studies , indicates that an effective ( and cost-effective ) human vaccine is unlikely to be forthcoming in the near future . Nonetheless , with no reliable intervention having been identified and the current treatment options for leishmaniasis being expensive , with serious side effects and emerging resistance , there is clearly a need for a more integrated focus within the international community to direct resources towards development of a human vaccine . | Leishmaniasis is a vector-borne parasite infection , transmitted to humans by sandflies . It is estimated to cause 1 . 6 million new cases of disease annually . Of the two main forms , so-called “visceral” and “cutaneous” , the visceral form is fatal in 85–90% of untreated cases . This literature review provides a comprehensive summary of all the available evidence relating to the impact of interventions against infection on the burden of leishmaniasis in people and highlights the absence of high quality evidence demonstrating an effect . Four broad categories of preventative interventions are identified , investigating a range of strategies , from protection of humans against infection , to interventions aimed one stage upstream of human infection ( targeting the sandfly vector ) , and even further , to interventions targeting animal reservoir species . Based on the current lack of understanding of the dynamics of transmission of Leishmania , we conclude that scant resources might be best directed toward prevention of human infections , with a focus on development of a human vaccine . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"and",
"Discussion"
] | [
"public",
"health",
"medicine",
"preventive",
"medicine",
"infectious",
"diseases",
"global",
"health",
"leishmaniasis",
"parasitic",
"diseases"
] | 2013 | A Review of Preventative Methods against Human Leishmaniasis Infection |
BRCA1 promotes DNA repair through interactions with multiple proteins , including CtIP and FANCJ ( also known as BRIP1/BACH1 ) . While CtIP facilitates DNA end resection when de-acetylated , the function of FANCJ in repair processing is less well defined . Here , we report that FANCJ is also acetylated . Preventing FANCJ acetylation at lysine 1249 does not interfere with the ability of cells to survive DNA interstrand crosslinks ( ICLs ) . However , resistance is achieved with reduced reliance on recombination . Mechanistically , FANCJ acetylation facilitates DNA end processing required for repair and checkpoint signaling . This conclusion was based on the finding that FANCJ and its acetylation were required for robust RPA foci formation , RPA phosphorylation , and Rad51 foci formation in response to camptothecin ( CPT ) . Furthermore , both preventing and mimicking FANCJ acetylation at lysine 1249 disrupts FANCJ function in checkpoint maintenance . Thus , we propose that the dynamic regulation of FANCJ acetylation is critical for robust DNA damage response , recombination-based processing , and ultimately checkpoint maintenance .
The hereditary breast cancer associated gene product , BRCA1 is an essential tumor suppressor . To promote genomic stability , BRCA1 interacts with multiple protein partners . In particular , through its C-terminal BRCT repeats , BRCA1 directly interacts with Abraxas , CtIP and FANCJ ( also known as BRIP1 or BACH1 ( BRCA1-associated C-terminal helicase 1 ) ) . These BRCT-interacting proteins contribute to the function of BRCA1 in the DNA damage response ( DDR ) . Abraxas serves to localize BRCA1 to sites of DNA damage and CtIP promotes the initiation of DNA end resection , which is critical for HR [1]–[3] . FANCJ also participates in localizing BRCA1 to sites of DNA damage , in DNA repair , and in checkpoint signaling; however , its distinct function is less clear . Elucidating how FANCJ functions in the DDR is important , as mutations in the FANCJ gene are associated with hereditary breast cancer as well as with the rare cancer prone syndrome Fanconi anemia ( FA ) within the FANCJ patient complementation group ( FA-J ) [4] . As a DEAH-family helicase , it is expected that FANCJ metabolizes DNA substrates to facilitate DNA repair . Consistent with this idea , recombinant-FANCJ is a 5′-3′ helicase and translocase that can unwind D-loops and displace RAD51 [5] . In cells , FANCJ also localizes to sites of DNA damage . Furthermore , when FANCJ is absent , catalytically inactive , or lacks BRCA1 binding , cells display defects in double strand break repair ( DSBR ) and HR [6]–[9] . Recently , FANCJ was identified as a factor essential for maintaining the DNA damage induced checkpoint in response to ionizing radiation [10] . Despite these findings , FANCJ-deficient cells are only mildly sensitive to agents that induce DSBs [11] . To explain these findings , it has been proposed that FANCJ functions in DSBR , but has a more significant role in processing replication forks stalled at lesions , such as DNA interstrand crosslinks ( ICLs ) . In support of this idea , FANCJ-null cells , similar to other FA patient cells , are extremely sensitive to agents that induce ICLs , such as cisplatin , melphalan , or mitomycin C ( MMC ) [7] , [12] , [13] . This sensitivity is reversed by complementation of FA-J cells with wild-type FANCJ ( FANCJWT ) , but not with catalytically inactive FANCJ mutants [6] , [8] , [14] . Interestingly , the mechanism by which FANCJ mediates ICL processing is regulated by BRCA1 binding . HR is favored when BRCA1 binds FANCJ . When BRCA1 binding is prevented , lesion bypass is favored by a mechanism requiring the translesion synthesis polymerase polη [9] . Thus , complementation of FA-J cells with a BRCA1-interaction defective mutant FANCJS990A reverses ICL sensitivity but does not fully restore FANCJ function . Here , we present evidence that FANCJ contributes to lesion processing by promoting a robust DDR . Essential for this function is FANCJ acetylation on a specific lysine residue . As such , preventing FANCJ acetylation suppresses DNA end resection that normally serves to engage recombination-based processing . Thus , both BRCT-interacting proteins , CtIP and FANCJ undergo DNA damage induced changes in acetylation that further regulates their function in the DDR to promote genomic stability .
As observed for CtIP , FANCJ binds directly to the BRCT domains of BRCA1 [6] , [9] , [15] . Given that CtIP function is inactivated by acetylation [16] , we addressed whether FANCJ was similarly modified . For this analysis , myc-tagged FANCJ was co-transfected with various Flag- or HA-tagged histone acetyltransferases . In an immunoblot probed with a pan-acetyl lysine antibody , we found that the precipitated FANCJ was acetylated only when CBP was over-expressed ( Figure 1A ) . Moreover , FANCJ acetylation was induced by CBP in a dose dependent manner ( Figure 1B ) . FANCJ acetylation was preserved most effectively by the inclusion of two types of deacetylase inhibitors , trichostatin-A ( TSA ) and nicotinamide ( NAM ) ( Figure 1C ) . Thus , we considered that more than one class of histone deacetylase ( HDAC ) could deacetylate FANCJ . TSA inactivates class 1 and class II HDACs , whereas NAM inactivates the nicotinamide adenoine dinucleotide ( NAD+ ) -dependent sirtuin ( class III ) family of HDACs ( including SIRT1 to SIRT7 ) [17] . FANCJ acetylation was reduced more when either Flag-tagged-HDAC3 or SIRT1 were overexpressed in 293T cells than observed upon overexpression of HDAC1 , HDAC2 , or SIRT6 ( Figure 1D ) . Titration of the SIRT1 expression vector transfected into 293T cells revealed that 0 . 01 µg of the SIRT1 construct matched the expression level of 4 µg of the HDAC3 construct . At this similar level of expression , HDAC3 more efficiently deactylated FANCJ than did SIRT1 ( Figure 1E ) . Together , these data implicate that FANCJ can be acetylated by CBP and deacetyated by HDAC3 as well a SIRT1 when over-expressed . To identify the FANCJ acetylation site ( s ) , myc-tagged C-terminal FANCJ truncation mutants were co-transfected with CBP into 293T cells . By Immunoblot analysis using the pan-acetyl antibody , we found that acetylation of FANCJ required amino acids 1239 to 1249 ( Figure 2A , 2C ) . Consistent with this region being modified , a C-terminal domain of FANCJ similar to a C-terminal p53 control was acetylated in vitro by a HAT-domain protein ( Figure 2B , 2C ) . To determine , which of three lysine ( K ) residues in this C-terminal region were required for acetylation , we generated three independent FANCJ mutant constructs that converted lysines 1240 , 1242 , or 1249 to arginine ( R ) . Further transfection experiments revealed that the K1249 was the dominant site for FANCJ acetylation , a lysine that is not conserved in chicken or C . elegans FANCJ species ( Figure 2D , 2E ) . Next , we sought to provide more conclusive evidence that CBP-induced acetylation on FANCJ was at the K1249 site . We purified FANCJ from 293T cells transfected with a C-terminal myc-tagged FANCJWT or the FANCJK1249R mutant species by immunoprecipitation using a myc antibody . Isolated proteins were then digested with trypsin and subjected to tandem mass spectrometry analysis ( LC-MS/MS ) . FANCJ-derived peptides covering the entire sequence were analyzed , and acetylation sites were identified using MASCOT search algorithm . Most of the acetylated lysine residues were detected in overlapping peptides derived from at least two independent protein preparations . In the FANCJWT , one of these sites was K1249 ( Figure 2F ) . Interestingly , even though by antibody detection , the FANCJK1249R mutant scores unmodified as in Figure 2D; FANCJWT and FANCJK1249R mutant had three additional acetylation marks detected by mass spectrometry ( Figure S1 ) . Furthermore , the K1249R mutant had five additional acetylated lysines not found in wild-type FANCJ , suggesting that these sites are not available when K1249 is acetylated ( Figure S1 ) . Thus , immunoblot and mass spectrometry analysis confirm that the very last amino acid of FANCJ , lysine 1249 is acetylated . Given that DNA damage reduces CtIP acetylation [16] , we addressed whether DNA damage could alter FANCJ acetylation . Endogenous FANCJ acetylation was enhanced in MCF7 cells treated with zeocin , camptothecin ( CPT ) , or hydroxyurea ( HU ) as compared to ultraviolet radiation ( UV ) , MMC , or methyl methanesulfonate ( MMS ) at the dose and time-post treatment analyzed ( Figure 3A ) . Notably , zeocin had a more robust induction of FANCJ acetylation despite the dose of zeocin , CPT , or UV having similar affect on cell survival ( Figure S2; data not shown ) . As found previously , DNA damage did not measurably alter FANCJ co-precipitation with BRCA1 with the exception of UV damage , which could reflect the UV-induced BRCA1 degradation [11] , [18] ( Figure 3A ) . DNA damage also induced FANCJ acetylation in HeLa cells , in response to not only CPT , but also MMC ( Figure 3A ) . In response to DNA damage , we also noted that FANCJ protein levels were sometimes enhanced ( Figure 3A ) . To clarify whether acetylation or our ability to detect acetylation was induced by DNA damage , we sought to induce DNA damage in cells in which our ability to detect FANCJ acetylation was not limiting . Indeed , the amount of acetylation on similar levels of exogenous FANCJWT achieved with low dose CBP expression was considerably enhanced following treatment with zeocin or CPT ( Figure 3B , 3C ) . Interactions with BRCA1 and MLH1 were not required for the CBP-induced acetylation of FANCJ , because BRCA1- and MLH1-interaction-defective mutants , FANCJS990A and FANCJK141/142A were readily modified ( data not shown ) . In contrast , following treatment with CPT , acetylation was not detected on the FANCJK1249R mutant ( Figure 3C ) , indicating that DNA damage-induced FANCJ acetylation requires the C-terminal K1249 residue . It remains to be determined , however if FANCJ acetylation is induced by a distinct type of DNA damage . The enhanced FANCJ acetylation following DNA damage led us to hypothesize that this modification facilitated FANCJ function in DNA repair . To address this possibility , we made use of this lysine to arginine FANCJK1249R mutant that prevents acetylation and also generated a lysine to glutamine FANCJK1249Q mutant to structurally mimic acetylation . Consistent with these mutants being functional , the purified recombinant proteins displayed similar catalytic activities as FANCJWT ( Figure S3 ) . In addition , they were expressed at similar levels as FANCJWT in FANCJ-null FA-J cells ( Figure 4A ) . Similar to FANCJWT , FANCJK1249R and FANCJK1249Q precipitated with known FANCJ interacting partners , BRCA1 and MLH1 [6] , [8] ( Figure 4B ) . In addition , the mutants co-localized with BRCA1 in response to DNA damage and the FA-J cells expressing FANCJWT or mutants had similar asynchronous cell cycle profiles ( Figure 4C , 4D ) . The acetylation mutants also restored MMC resistance and the ability of FA-J cells to exit from an abnormal G2/M accumulation , albeit in a manner slightly more robust than FANCJWT ( Figure 4E , 4F ) . Together , these findings suggested that the mutants were enzyme active and functional in vivo; however the mechanism by which the FANCJ mutants restore ICL resistance could be distinct from FANCJWT . Previously , complementation of FA-J cells with a BRCA1-binding defective mutant , FANCJS990A gave the semblance of FANCJWT function . In particular , MMC resistance was restored [8] . However , in contrast to FANCJWT , FANCJS990A provides resistance to MMC by a mechanism dependent on the DNA damage tolerance pathway . Within this tolerance pathway , translesion synthesis polymerases can bypass DNA lesions such as unhooked ICLs and intra-strand crosslinks generated by UV , but not DSBs generated by zeocin . Evidence that FANCJS990A skewed lesion processing towards DNA damage tolerance was based on several findings . First , the sensitivity to MMC in these cells was restored upon depletion of the essential tolerance factor , Rad18 or the translesion polymerase polη , but not upon depletion of the HR protein , Rad54 . Second , in comparison to FANCJWT , cells expressing FANCJS990A were hyper-resistant to UV , a phenotype that was reversed upon polη-depletion . Third , in comparison to FANCJWT , FANCJS990A-expressing cells were sensitive to zeocin , indicating reduced DSBR [9] . Thus , we sought to determine whether similar to the BRCA1-binding mutant , the acetylation mutants also functioned differently from FANCJWT . To test this idea , the FA-J cell lines were left untreated or treated with increasing doses of MMC , zeocin , or UV . In comparison to the other FA-J cell lines , the FA-J cell line expressing the acetylation mutant FANCJK1249R was hyper-resistant to UV , but unable to restore normal levels of zeocin resistance . In contrast , the FA-J cell line expressing the acetylation mimic FANCJK1249Q displayed greater resistance to zeocin ( Figure 5A; Figure S2 ) . Thus , in response to UV and zeocin , cells expressing the acetylation mutants are distinct from each other as well as from cells expressing FANCJWT . To further validate these results , we targeted recombination or DNA damage tolerance pathways by using siRNA reagents to Rad54 or polη . Significantly , depletion of Rad54 suppressed the zeocin resistance of the FA-J cell line expressing FANCJK1249Q ( Figure 5A , 5C ) . Likewise , depletion of polη suppressed the UV hyper-resistance of the FA-J cell line expressing FANCJK1249R ( Figure 5A , 5C ) . Furthermore , depletion of polη , but not Rad54 reversed the MMC resistance of the FA-J cell line expressing FANCJK1249R ( Figure 5B ) . In contrast , depletion of Rad54 , but not polη reduced the MMC resistance of the FA-J cell line expressing FANCJK1249Q ( Figure 5B ) . Together , these results indicate that the acetylation of FANCJ at lysine 1249 contributes to the mechanism of lesion processing; preventing acetylation favors DNA damage tolerance and constitutive acetylation favors recombination . How could FANCJ acetylation affect lesion processing ? Because both CtIP and FANCJ are acetylated and directly bind to the BRCA1-BRCT domain , we speculated that FANCJ might similarly have a role in DNA end resection . In particular , the affect of CtIP acetylation on DNA end resection was analyzed in response to CPT [16] . We found RPA foci formation at 1 h post-CPT was more robust ( 64% and 65% ) in the FANCJWT and FANCJK1249Q FA-J cell lines as compared to vector and FANCJK1249R FA-J cell lines that had 47% and 29% , respectively ( Figure 6A ) . Thus , as measured by RPA foci formation , FANCJWT and the acetylation mimic FANCJK1249Q were more active in DNA end resection . RPA loading onto ssDNA also leads to its subsequent phosphorylation on Ser4 and Ser8 [3] . We found that the FA-J cell lines had a similar phosphorylation of Chk2 and γ-H2AX following exposure to two different dose of CPT , indicating that FANCJ or its ability to be acetylated is not required for DSB formation in response to CPT ( Figure 6B ) . Likewise , at 1 h post-CPT treatment , Chk1 phosphorylation was detected ( Figure 6B ) . In contrast , RPA phosphorylation was most robust in the CPT-treated FANCJWT and FANCJK1249Q FA-J cell lines ( Figure 6B ) . In support of these findings , reduced RPA phosphorylation was also detected in CPT-treated FANCJ-deficient U2OS cells generated by siRNA reagents ( Figure S4 ) . Furthermore , at 4–24 h post CPT treatment , we noted diminished RPA phosphorylation in FANCJK1249R as compared with FANCJWT and FANCJK1249Q FA-J cell lines ( Figure 6C ) . At this time , RPA phosphorylation in the FANCJK1249R FA-J cells was also reduced compared to vector FA-J cells that had gained considerable RPA phosphorylation as compared to 1 h post-CPT ( Figure 6B , 6C ) . In the response to zeocin , which induces DSBs independent of replication , RPA phosphorylation was similar in FA-J cell lines with or without FANCJWT ( Figure S5 ) . Together , these results suggest a role for FANCJ and its acetylation in DNA end resection at stalled replication forks as induced by CPT . To address whether the contribution of FANCJ acetylation to DNA end resection was sufficient to enhance HR , we next analyzed Rad51 foci formation . In response to CPT , we found that Rad51 foci were the most robust in FA-J cells complemented with FANCJWT or the FANCJK1249Q mutant . Instead , Rad51 foci in the FA-J cells with vector or FANCJK1249R were more anemic ( Figure 7A ) . Furthermore , a greater number of FANCJK1249Q expressing FA-J cells were positive for Rad51 foci as quantitated between 2–16 h after CPT treatment ( Figure 7A ) . In contrast , the γ-H2AX foci did not have a significant difference between the FA-J cells lines . Thus , a greater proportion of γ-H2AX co-staining Rad51 foci were detected in FANCJK1249Q or FANCJWT , as compared to vector or FANCJK1249R expressing FA-J cells ( Figure 7A merge ) . Together , these findings demonstrate that in response to CPT , FANCJ and its acetylation at 1249 promote DNA end processing events that enhance RPA phosphorylation , and both RPA and Rad51 focal accumulation . Given these findings and the recent identification that FANCJ promotes checkpoint maintenance [10] , we considered that FANCJ acetylation could be essential for maintaining the checkpoint . Defects in checkpoint maintenance were evaluated by determining if CPT treated FA-J cells traversed prematurely to mitosis . FA-J cells lacking FANCJWT entered mitosis by 24 h post-CPT as indicated by a positive histone H3 phosphorylation ( Figure 7B ) . These results are consistent with FANCJ acetylation supporting checkpoint maintenance . However , FA-J cells expressing FANCJK1249R or FANCJK1249Q also failed to maintain the checkpoint , showing H3 phosphorylation by 24 h ( Figure 7B ) . Substantiating this finding , at time points greater than 4 h post-CPT treatment , both mutants had reduced Chk1 phosphorylation as compared to FA-J cells expressing FANCJWT ( Figure 7B ) . Collectively , these findings suggest that FANCJ acetylation enhances the initial DDR to facilitate recombination-based repair and limit translesion synthesis . Checkpoint maintenance however , requires FANCJ and its dynamic regulation by acetylation ( Figure 7C ) .
Here we identify acetylation as a DNA damage-dependent regulator of the BRCA-FA protein , FANCJ . We show that acetylation at lysine 1249 is a critical regulator of FANCJ function during cellular DNA repair . We analyzed the expression of two FANCJ mutants that mimicked either the constitutive deacetylated FANCJK1249R or acetylated FANCJK1249Q protein isoforms . While the mutants functioned similar to FANCJWT in several assays and restored MMC resistance and exit from an abnormal G2/M checkpoint response to FA-J cells , the mutants were distinct from FANCJWT with respect to lesion processing . Notably , FA-J cells expressing the acetylation mutants differentially relied on repair and tolerance factors for resistance to DNA damaging agents . Our findings further demonstrate that FANCJ has the ability to potentiate HR and DNA damage induced acetylation is important for this function . Another BRCA1-BRCT interacting protein , CtIP is acetylated and functions in DNA end resection . Thus , we considered that recombination-based lesion processing by the FANCJ acetylation mimic , FANCJK1249Q resulted from a function for FANCJ acetylation in DNA end resection . To test this idea , the FA-J cells were treated with CPT , which generates breaks in S-phase . Indeed , FA-J cells expressing the acetylation mutants were distinct in the initial response to CPT . Specifically , FA-J cells expressing the FANCJK1249Q , but not FANCJK1249R , promoted DNA end resection post-CPT exposure as measured by presentation of RPA foci or its phosphorylation at serine residues 4 and 8 . Furthermore , FA-J cells expressing the FANCJK1249Q , as compared to FANCJK1249R had 2 . 5-fold more cells with CPT-induced Rad51 foci . This more robust DDR could reflect a role for FANCJ acetylation in loading RPA , as shown for FANCJ in response to HU [19] . Our data do not implicate a global role for FANCJ in DNA end resection given that FANCJ-deficiency did not affect the amount of RPA phosphorylation following zeocin , an agent that induces DSBs independent of replication . However , FANCJ is acetylated when cells are exposed to CPT or zeocin . Thus , DSB-induced FANCJ acetylation that is not associated with stalled or broken replication forks may contribute to some other aspect of the DDR , such as checkpoint maintenance . In fact , we find that FANCJ as well as its acetylation are essential for checkpoint maintenance . Specifically , in the absence of FANCJ or its DNA damage induced acetylation , Chk1 phosphorylation was induced , but not maintained and correspondingly cells underwent a more rapid transit into mitosis post-CPT . Interestingly , we found that similar to FA-J cells expressing the acetylation mutant FANCJK1249R , FA-J cells expressing the acetylation mimic FANCJK1249Q failed to maintain the checkpoint despite an initial DDR to CPT . Thus , some other aspect of checkpoint signaling is perturbed in FA-J cells that express the acetylation mimic . Perhaps this mutant fails to mediate a protein interaction or act upon a DNA substrate important for checkpoint maintenance . Instead FANCJ acetylation could serve as a switch , in which acetylation and de-acetylation is essential to maintain the checkpoint ( Figure 7C ) . Consistently , a role for FANCJ in checkpoint maintenance was reported in a recent study [10] . It follows that defects in initiating the DDR , engaging HR , and maintaining the checkpoint impact cellular DNA damage resistance . Reduced DNA repair and/or checkpoint maintenance defects could explain why FA-J cells expressing the acetylation mutant FANCJK1249R were sensitive to zeocin . Defects in repair and in maintaining the checkpoint may not increase cellular sensitivity if backup lesion processing mechanisms serve to process or bypass the lesion . Compensatory pathways could explain the lack of CPT-sensitivity in the FA-J cells with or without acetylation mutants ( Figure S2 ) . In support of this idea , our data reveal that FA-J cells expressing the acetylation mutant were resistant to DNA damage by relying on tolerance factors . As such , depletion of polη in FA-J cells expressing the non-acetylatable FANCJK1249R mutant reversed the UV and MMC resistance . Instead , FA-J cells expressing the acetylation mimic FANCJK1249Q maintained zeocin and MMC resistance in a Rad54-dependent manner . These findings suggest that the toxicity to ICLs lesions as found in cells deficient for FANCJ is avoided because FANCJ enzyme active acetylation mutants facilitate recombination in S phase or translesion synthesis bypass of unhooked ICL lesions perhaps in mitosis . In the absence of a maintained checkpoint , however recombination similar to translesion synthesis bypass is likely to be error-prone . Previously , we found that BRCA1 binding to FANCJ altered FANCJ function in HR and translesion synthesis pathways . Indeed , we find that similar to FA-J cells expressing the acetylation FANCJK1249R mutant , FA-J cells expressing the BRCA-interaction defective mutant , FANCJS990A were hyper-resistant to UV induced damage , sensitive to zeocin induced damage , and relied on polη for MMC resistance [7] . Data also indicate that similar to FANCJK1249R , the FANCJS990A mutant fails to maintain the checkpoint . In response to melphalan treatment , FA-J cells expressing the FANCJS990A mutant , as compared to FANCJWT , underwent a reduced and more rapid G2/M checkpoint exit [7] . These similar outcomes do not reflect common defects in BRCA1 binding or acetylation . Indeed , the FANCJS990A mutant was acetylated upon co-transfection of CBP to levels similar to those observed for FANCJWT ( data not shown ) . Moreover , co-precipitation experiments demonstrated that the FANCJK1249R mutant bound BRCA1 as well as FANCJWT . Thus , BRCA1 binding and acetylation of FANCJ may be distinct events . Nevertheless , defects in BRCA1 binding at serine 990 or acetylation at lysine 1249 could have similar outcomes for FANCJ function because both mutants fail to maintain a robust checkpoint and Rad51-based repair is reduced [9] . A stalled replication fork with exposed single stranded and double stranded regions could provide an ideal DNA substrate for FANCJ . Indeed , FANCJ requires several nucleotides for binding and metabolizing DNA [20] . FANCJ function in replication fork processing could also be similar to other 5′-3′ DNA helicase/translocases such as Ecoli RecD and yeast Rad3 . Rad3 facilitates exonucleolytic degradation of DNA ends , which restricts recombination between short homologous sequences [21] . Interestingly , RecD regulates resection and recombination by changes in helicase speed , which can also facilitate a polymerase swap , in which bypass polymerases diminish fork break down [22] . Conceivably , enhanced FANCJ enzyme activity or altered substrate preference due to acetylation could generate more single-stranded DNA to elicit checkpoint responses such as RPA loading as proposed [19] . Alternatively , checkpoint maintenance could require reduced FANCJ enzyme activity so that FANCJ does not displace proteins from lesions , such as RAD51 or interacting partners BRCA1 , RPA and BLM helicase [6] , [23]–[25] . In this context , it is worth noting that changes in motor speed have been associated with FANCJ clinical mutants . The breast cancer associated mutant , M299I is enzyme activating and both unwinds and translocates DNA more efficiently than FANCJWT , whereas the P47A mutant is enzyme inactivating [26] , [27] . Whether changes in FANCJ function derive from acetylation and/or partners that bind via this modification remains to be determined . Furthermore , based on our current data , it is unclear if distinct DNA lesions selectively induce FANCJ acetylation . In summary , our findings indicate that FANCJ has the ability to potentiate HR through dual roles in DNA end processing and checkpoint maintenance . These two functions require FANCJ lysine 1249 , a site not conserved in FANCJ orthologues such as chicken FANCJ and C . elegans Dog-1 . Interestingly , unlike in human cells , FANCJ does not function in HR in chicken and C-elegans systems [28] , [29] . It is not surprising that regulators of FANCJ acetylation state , HDACIII , SIRT1 , and CBP have roles in DNA repair and genomic stability [30]–[32] . It remains to be determined , however , whether associated repair defects are related to failure to regulate FANCJ acetylation . Complicating this analysis , HDACIII , SIRT1 , and CBP have many other histone and non-histone protein substrates that also have role in DNA repair and genomic stability . For example , SIRT1 deacetylation plays an important role in regulating the function of DNA double strand break repair proteins , such as Ku70 [33] , WRN [34] , and NBS1 [35] . Moreover , p300/CBP functions to regulate the activities of multiple proteins at the replication fork including PCNA [36] . CBP also regulates the activity of other helicases , including WRN [37] . Whether HDAC or HAT associated defects derive from a failure to regulate FANCJ acetylation will be an important question for future studies .
MCF7 , HeLa , and 293T cells were grown in DMEM supplemented with 10% fetal bovine serum and penicillin/streptomycin ( 100 U/mL each ) . FA-J ( EUFA30-F ) cells were cultured with 15% fetal bovine serum and penicillin/streptomycin ( 100 U/mL each ) . FA-J cells were infected with the POZ retroviral vector [38] containing no insert , WT , K1249R , or K1249Q FANCJ inserts . Stable FA-J POZ cell lines were selected as before [8] . Cells were harvested , lysed , and processed for Western blot analysis as described previously using an NETN lysis buffer ( 20 mM Tris , 150 mM NaCl , 1 mM EDTA , and 0 . 5% NP-40 ) containing 10 mM NaF and 1 mM NaVO3 [7] . For acetylation detection , unless otherwise noted cells were lysed with 150 mM NETN buffer supplemented with 10 µM TSA and 5 mM nicotinamide . For γ-H2AX detection , cell pellets were collected and dissolved and boiled in 2× lysis buffer ( 50 mM Tris pH 6 . 8 , 2% SDS , 1% B-ME ) . Antibodies used for immunoprecipitation ( IP ) and Western blot assays include FANCJ polyclonal Abs E67 [26] , β-Actin ( Sigma ) , pRPA S3/4 ( Bethyl ) , RPA ( Bethyl ) , pChk1 S317 ( Bethyl ) , Chk1 ( Bethyl ) , pChk2 ( Cell signaling ) , Chk2 ( Cell Signaling ) , γ-H2AX S139 ( Millipore ) , H2AX ( Bethyl ) , Flag ( Sigma ) , HA ( 12C4 ) , pan-acetylated lysine ( Cell signaling ) , MLH1 ( BD Bioscience ) , BRCA1 monoclonal ( ms110 ) , pH 3 ( Millipore ) , H3 ( Abcam ) , polη ( Abcam ) , Rad54 ( Abcam ) , Rad51 ( Abcam ) , and Myc monoclonal ( 9E10 ) . FA-J stable cell lines were either mock treated or treated with 0 . 25 µg/ml of melphalan ( Sigma ) and incubated for various times . Cells were fixed with 90% methanol in PBS overnight and then incubated 10 min with PBS containing 30 µg/ml DNase-free RNase A and 50 µg/ml propidium iodide . 1×104 cells were analyzed using a FACs Calibur instrument ( Becton-Dickinson , San Jose , CA ) . Aggregates were gated out and the percentage of cells in G2/M was calculated using Flow Jo software . The pCDNA3-myc . his vector ( Invitrogen ) was digested by Not1/Apa1 and different FANCJ fragments generated by PCR and digested by Not1/Apa1 were inserted . Primers are available upon request . Reverse primers used for K1249R-pCDNA3 and K1249RQ-pCDNA3 are 5′TTTTGGGCCCCCTAAAACCAGGAAACATGCC3′ and 5′TTTTGGGCCCCTGAAAACCAGGAAACATGCC3′ , respectively . The K1249R and K1249Q pOZ vectors were generated with the QuickChange Site-Directed Mutagenesis Kit ( Stratagene , La Jolla , CA ) by using the FANCJ- pCDNA3-myc . his or FANCJ-pOZ as a template and the following primers: ( K1249R-pOZ-Forward ) 5′GGCATGTTTCCTGGTTTTAGGGCGGCCGCTGGAGGAGCA3′ and ( K1249R-pOZ-Reverse ) 5′GTCTCCTCCAGCGGCCGCCCTAAAACCAGGAAACATGCC3′; ( K1239Q-pOZ-Forward ) 5′GGCATGTTTCCTGGTTTTCAGGCGGCCGCTGGAGGAGAC3′ and ( K1239Q-pOZ-Reverse ) 5′GTCTCCTCCAGCGGCCGCCTGAAAACCAGGAAACATGCC3′; Recombinant FANCJ protein production was made in insect cells using the PVL13 . 2 vector as before [26] . Full-length WT FANCJ was used as a template to generate the acetylation mutants using the following primers: ( K1249R-PVL132 Forward ) 5′GGCATGTTTCCTGGTTTTAGGGACTACAAGGAGACG3′ and ( K1249-PV132 Reverse ) 5′CGTCGTCCTTGTAGTCCCTAAAACCAGGAAACATGCC3′ . ( K1249Q-PVL132 Forward ) 5′ GGCATGTTTCCTGGTTTTCAGGACTACAAGGACGACG3′ and ( K1249Q-PVL132 Reverse ) 5′ CGTCGTCCTTGTAGTCCTGAAAACCAGGAAACATGCC3′ . The pGEX-5X vector ( GE Healthcare Life Sciences ) was digested by Sal1/Not1 and the FANCJ C-terminal fragment was generated by PCR and digested by Sal1/Not1 and inserted . Primers are available upon request . All DNA constructs were confirmed by DNA sequencing . Stable FA-J cell lines were untransfected or transfected with siRNA previously described against Luc , Rad54 , or polη [9] . Cells were seeded onto 6 well plates and incubated overnight . Seeded cells were either untreated or treated with increasing dose of MMC ( 1 h , serum free ) , UV , CPT , ( 1 h , serum free ) , or zeocin ( 1 h , serum free ) . To assay for percent survival , cells were counted 5–8 days post infection and percent survival was calculated as before [9] . Helicase assay reaction mixtures ( 20 µl ) contained 40 mM Tris-HCl ( pH 7 . 4 ) , 25 mM KCl , 5 mM MgCl2 , 2 mM dithiothreitol , 2% glycerol , 100 ng of bovine serum albumin/µl , 2 mM ATP , 10 fmol of 19-bp duplex DNA substrate ( 0 . 5 nM ) , and the concentrations of FANCJ ( acetylated or non acetylated ) indicated in the figures . Helicase reactions were initiated by the addition of FANCJ , and the reaction mixtures were incubated at 30°C for 15 min unless otherwise indicated . Reactions were quenched with the addition of 20 µl of 2× Stop buffer ( 17 . 5 mm EDTA , 0 . 3% SDS , 12 . 5% glycerol , 0 . 02% bromophenol blue , 0 . 02% xylene cyanol ) . For standard duplex DNA substrates , a 10-fold excess of unlabeled oligonucleotide with the same sequence as the labeled strand was included in the quench to prevent reannealing . Reaction products were resolved on nondenaturing 12% ( 19∶1 acrylamide-bisacrylamide ) polyacrylamide gels , and quantitated as described previously [27] . Stable FA-J cell lines were seeded onto 6 well plates and incubated overnight . Cells were either untreated or treated with 1 mM HU ( 24 h ) or 0 . 25 µM CPT ( 1 h ) . Cells were fixed with 3% paraformaldehyde/2% sucrose for 10 min at RT , and permeabilized with 0 . 5% Triton X-100 in 20 mM HEPES for 5 min on ice . Incubation with antibodies and washes were described previously [6] . For Rad51 staining , cells were fixed with 3% paraformaldehyde/2% sucrose for 10 min at RT , permeabilized with ice-cold methanol for 30 min , and blocked with 4% BSA for 1 h . Staining was as described previously [6] . The acetyltransferase assays were performed in 30 µl of reaction , which includes reaction buffer ( 50 mM HEPES ( ph 8 . 0 ) , 10% glycerol , 1 mM DTT , 1 mM PMSF , 10 mM Na-butyrate ) , 1 µL [3H]-acetyl-CoA , 1 µl recombinant HAT domain of p300 ( gift of Dr . Luo ) , and recombination FANCJ-CT or p53-CT [39] . Reaction were carried out at 30°C for 1 h and separated by SDS-PAGE , analyzed by autoradiography . Concentrations of recombinant proteins were determined by comassie staining from Invitrogen . Gel bands containing FANCJ1 were de-stained twice with 25 mM ammonium bicarbonate in 50% acetonitrile for 30 min in 37°C , reduced with 7 . 6 mg/ml dithiothreitol at 60°C for 10 min , and alkylated with 18 . 6 mg/mL iodoacetamide at room temperature for 1 hour . The bands were then washed twice with 25 mM ammonium bicarbonate in 50% acetonitrile for 15 min at 37°C prior to shrinking with 50 µL acetonitrile for 10 min at room temperature . 100 ng trypsin ( Promega ) was added to each sample and 25 mM ammonium bicarbonate was added until the gels were fully swollen ( ∼10–50 µL ) and the digestion proceeded overnight at 30°C . Following digestion , peptide extracts were transferred into new tubes and the gels were further extracted with 50 µL of 50% acetonitrile containing 5% formic acid ( v/v ) and following 15 min were added to the initial extracts . The latter process was repeated for a total of three extractions . Extracts were then dried on a SpeedVac and reconstituted in 20 µL of 0 . 1% formic acid for LC-MS/MS analysis . Tryptic peptides ( 2 µL ) were directly loaded at 4 µL/min for 7 min onto a custom-made trap column ( 100 µm I . D . fused silica with Kasil frit ) containing 2 cm of 200 Å , 5 µm Magic C18AQ particles ( Michrom Bioresources ) . Peptides were then eluted using a custom-made analytical column ( 75 µm I . D . fused silica ) with gravity-pulled tip and packed with 25 cm 100 Å , 5 µm Magic C18AQ particles ( Michrom ) . Peptides were eluted with a linear gradient from 100% solvent A ( 0 . 1% formic acid∶acetonitrile ( 95∶05 ) ) to 35% solvent B ( acetonitrile containing 0 . 1% formic acid ) in 35 min at 300 nL/min using a Proxeon Easy nanoLC system directly coupled to a LTQ Orbitrap Velos mass spectrometer ( Thermo Scientific ) [40] . Data were acquired using a data-dependent acquisition routine of acquiring one mass spectrum from m/z 350–2000 in the Orbitrap ( resolution 60 , 000 ) followed by tandem mass spectrometry scans in the LTQ linear ion trap of the 10 most abundant precursor ions found in the mass spectrum . Charge state rejection of singly-charged ions and dynamic exclusion was utilized to minimize data redundancy and maximize peptide identification [41] . The raw data files were processed and searched against the human index of the SwissProt database ( version 09/21/11 ) containing both the mutant and wild-type forms of FANCJ1 with Mascot ( version 2 . 3 . 02; Matrix Science ) using parent mass tolerances of 15 ppm and fragment mass tolerances of 0 . 5 Da . Full tryptic specificity with 2 missed cleavages was used and variable modifications of acetylation ( protein N-term and lysine ) , pyro-glutamination ( N-term glutamine ) , and oxidation ( methionine ) , and fixed modification of carbamidomethylation ( cysteine ) were considered . Mascot search results were also loaded into Scaffold ( Version 3 . 3 . 1; Proteome Software ) for comparative analyses using spectral counting of tandem mass spectra and full annotation of the data [42] . | The BRCA1–Fanconi anemia ( FA ) pathway is required for both tumor suppression and cell survival , particularly following treatment with DNA damaging agents that induce DNA interstrand crosslinks ( ICLs ) . ICL processing by the BRCA–FA pathway includes promotion of homologous recombination ( HR ) and DNA damage tolerance through translesion synthesis . However , little is known about how the BRCA–FA pathway or these ICL processing mechanisms are regulated . Here , we identify acetylation as a DNA damage–dependent regulator of the BRCA–FA protein , FANCJ . FANCJ acetylation at lysine 1249 is enhanced by expression of the histone acetyltransferase CBP and reduced by expression of histone deacetylases HDAC3 or SIRT1 . Furthermore , acetylation on endogenous FANCJ is induced upon treatment of cells with agents that generate DNA lesions . Consistent with this post-translation event regulating FANCJ function during cellular DNA repair , preventing FANCJ acetylation skews ICL processing . Cells have reduced reliance on HR factor Rad54 and greater reliance on translesion synthesis polymerase polη . Our data indicate that FANCJ acetylation contributes to DNA end processing that is required for HR . Furthermore , resection-dependent checkpoint maintenance relies on the dynamic regulation of FANCJ acetylation . The implication of these findings is that FANCJ acetylation contributes to DNA repair choice within the BRCA–FA pathway . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"biology"
] | 2012 | FANCJ/BACH1 Acetylation at Lysine 1249 Regulates the DNA Damage Response |
Nuclear domain 10 ( ND10 ) components are restriction factors that inhibit herpesviral replication . Effector proteins of different herpesviruses can antagonize this restriction by a variety of strategies , including degradation or relocalization of ND10 proteins . We investigated the interplay of Kaposi's Sarcoma-Associated Herpesvirus ( KSHV ) infection and cellular defense by nuclear domain 10 ( ND10 ) components . Knock-down experiments in primary human cells show that KSHV-infection is restricted by the ND10 components PML and Sp100 , but not by ATRX . After KSHV infection , ATRX is efficiently depleted and Daxx is dispersed from ND10 , indicating that these two ND10 components can be antagonized by KSHV . We then identified the ORF75 tegument protein of KSHV as the viral factor that induces the disappearance of ATRX and relocalization of Daxx . ORF75 belongs to a viral protein family ( viral FGARATs ) that has homologous proteins in all gamma-herpesviruses . Isolated expression of ORF75 in primary cells induces a relocalization of PML and dispersal of Sp100 , indicating that this viral effector protein is able to influence multiple ND10 components . Moreover , by constructing a KSHV mutant harboring a stop codon at the beginning of ORF75 , we could demonstrate that ORF75 is absolutely essential for viral replication and the initiation of viral immediate-early gene expression . Using recombinant viruses either carrying Flag- or YFP-tagged variants of ORF75 , we could further corroborate the role of ORF75 in the antagonization of ND10-mediated intrinsic immunity , and show that it is independent of the PML antagonist vIRF3 . Members of the viral FGARAT family target different ND10 components , suggesting that the ND10 targets of viral FGARAT proteins have diversified during evolution . We assume that overcoming ND10 intrinsic defense constitutes a critical event in the replication of all herpesviruses; on the other hand , restriction of herpesviral replication by ND10 components may also promote latency as the default outcome of infection .
Human Herpesvirus 8 ( HHV8 ) , or Kaposi's Sarcoma-Associated Herpesvirus ( KSHV ) , belongs to the subfamily Gammaherpesvirinae and is grouped together with the closely related prototypic Saimiriine herpesvirus 2 , or Herpesvirus saimiri ( HVS ) , into the genus Rhadinovirus . KSHV was first discovered in 1994 from patients with Kaposi's Sarcoma ( KS ) lesions [1] . KSHV is the etiologic agent of three human diseases; a ) KS , a skin tumor of endothelial origin , and two B cell malignancies , b ) Primary Effusion Lymphoma and c ) Multicentric Castleman's Disease . KS exists in several forms , classical KS , which is endemic in the Mediterranean and Middle East [2] , African endemic KS , iatrogenic KS , which is associated with renal transplantation and AIDS-associated KS , which is among the leading causes of death in AIDS patients [3] . The promyelocytic leukemia protein ( PML ) is the main component of a subnuclear structure called nuclear domain 10 ( ND10 ) . ND10 components like PML , Sp100 , Daxx and ATRX were identified as cellular restriction factors that are able to inhibit the replication of several DNA viruses , including herpesviruses . The restriction is assumed to be mediated by the recruitment of cellular proteins like heterochromatic protein 1 ( HP1 ) , DNA methyltransferases or histone deacetylases that induce a repressive chromatin status on the herpesviral genome [4] . This antiviral function of PML , however , is counteracted by most human herpesviruses through viral effector proteins using a variety of strategies , including degradation or relocalization of PML . So far , herpesviruses were mainly found to target the major ND10 component PML , as exemplified by the herpes simplex virus regulatory protein ICP0 , which has been shown to induce the proteasomal degradation of PML either via acting as a SUMO-targeted ubiquitin ligase or via SUMO-independent ubiquitination [5]–[7] . Similarly , the IE1 protein of human cytomegalovirus targets PML via inducing a de-SUMOylation of this major ND10 component [8] , [9] . Since SUMOylation of PML is essential for the integrity of ND10 , this leads to a dispersal of the subnuclear structure . PML expression is induced by the interferon pathway [10] , [11] and the protein is detected in high amounts in KSHV infected PEL cells [12] . KSHV vIRF3 ( LANA2 , K10 . 5 ) has been shown to colocalize with PML and its overexpression is associated with a redistribution of PML [13] . Another KSHV gene , ORF K8 encodes an early lytic protein , which shares homology to the EBV BZLF1 gene . K8 is a SUMO interaction motif ( SIM ) dependent , SUMO-2/3-specific SUMO E3 ligase [14] that colocalizes with PML [15] . K8 recruits p53 and probably sequesters it to PML bodies but has no function in dispersing PML bodies [12] . We recently demonstrated that infection of fibroblasts with HVS results in restriction of immediate-early gene expression , which can be alleviated by siRNA mediated knock down of PML [16]; PML is thus also a restriction factor for HVS infection . In contrast to all other Herpesviruses , HVS exclusively degrades the cellular ND10 component Sp100 whereas other factors like PML or Daxx remain intact . The ORF3 tegument protein of HVS was identified as being responsible for this effect . ORF3 induced the proteasomal degradation of Sp100 , and a mutant HVS lacking the orf3-gene was no longer able to mediate Sp100 degradation [16] . The ORF3 of HVS shares homology with the cellular formylglycineamid-ribotideamidotransferase enzyme ( FGARAT , EC 6 . 3 . 5 . 3 ) . All gammaherpesviruses encode one to three FGARAT-homologous proteins in their genome . Interestingly , recent studies showed that ORF3-homologous proteins from murid gammaherpesvirus 68 ( MHV-68 ) and from Epstein-Barr virus ( EBV ) modulate the ND10 proteins PML and ATRX , respectively , suggesting that the ND10 targets of the viral FGARAT protein family have diversified during evolution [17] , [18] . The KSHV orf75 gene comprises one distinct member of the vFGARAT-family . The putative tegument protein ORF75 has been shown to be involved in NF-kB coactivation with KSHV K13/vFLIP [19] . This study addresses the role of ND10 components in the restriction of lytic replication of KSHV infection . We describe the essential protein ORF75 as a new viral effector , adding a new piece to the intricate puzzle of herpesvirus coevolution with cellular intrinsic immunity .
We recently demonstrated that PML acts as a restriction factor for infection with the gammaherpesvirus HVS [16] . To investigate whether ND10 components exert a similar effect on KSHV infection , we analyzed KSHV infection in primary human foreskin fibroblast ( HFF ) cells carrying an siRNA mediated knock-down of ND10 components . Using recombinant rKSHV . 219 [20] we detected significantly more GFP-positive cells in the absence of PML ( siPML ) or Sp100 ( siSp100 ) compared to control cells ( siC ) ( Figure 1A and 1B ) . Moreover , we detected a moderate increase in GFP expression from the viral genome after knock-down of Daxx ( siDaxx ) compared to control cells . However , there was no enhancement of KSHV infection as indicated by GFP expression after depletion of ATRX ( siATRX ) ( Figure 1A and 1B ) . KSHV-restriction was in a similar range as the restriction of HVS , suggesting that analogous mechanisms are active on these evolutionary closely related viruses ( Figure 1C ) . Accordingly , ND10 components may compromise KSHV infection in primary human cells . We next wanted to elucidate the effect of KSHV de novo infection on major ND10 components . Therefore , epithelial iSLK cells , which promote lytic replication of KSHV [21] , [22] , were infected with rKSHV . 219 and subjected to immunofluorescence analysis . In KSHV infected GFP positive iSLK cells , the dot like staining of ATRX characteristic for ND10 association was lost ( Figure 2A ) . In addition , we observed a loss of dotted Daxx staining pattern in infected cells ( Figure 2B ) . Using Western Blot experiments from iSLK cells infected with high multiplicity of infection ( MOI ) , we could demonstrate that KSHV infection leads to a reduction of ATRX protein levels . However , this coincided only with a minimal decrease of Daxx levels in Western Blot experiments ( Figure 2C ) . Both Western blot and immunofluorescence analysis revealed no effect of KSHV infection on the ND10 components PML and Sp100 in KSHV-infected cells , which were identified by GFP expression . PML and Sp100 were still detectable in numerous foci being characteristic for their assembly in ND10 structures ( Figure 2 ) . Thus , KSHV infection leads to the disappearance of ATRX and to the dispersion of Daxx from ND10; total Daxx protein in immunoblots is unchanged , however . In order to further corroborate that the observed effects are due to an incoming tegument protein , cells were infected in the presence of cycloheximide to inhibit de novo protein synthesis , or infected with UV-inactivated virus ( Figure 3 ) . ATRX also disappeared under both conditions , underlining that de novo protein synthesis is not necessary and the effect is caused by a viral tegument protein . In an additional set of experiments , treatment with the proteasome inhibitor MG-132 did also not reverse the effect of KSHV infection on ATRX , while the treatment stabilized the phosphorylated form of Cyclin D1 ( CCND1-pT286 ) as expected ( Figure 3A , lower panel ) . Next we wanted to know what happens after lytic induction of KSHV . Therefore we used epithelial iSLK . 219 cells [21] , [22] , a cell line that harbors multiple copies of KSHV strain rKSHV . 219 per cell . The KSHV strain rKSHV . 219 expresses GFP under control of a constitutive EF1a promoter situated in the BACmid backbone , and red fluorescent protein ( RFP ) under control of the PAN promoter , which is switched on by the KSHV initiator of lytic viral replication RTA encoded by ORF50 . Thus , infection per se results in GFP expression , and induction of early lytic gene expression can be easily monitored by RFP expression . Similar to our observation after de novo infection , KSHV lytic induction leads to a decrease of ATRX in cells expressing RFP ( Figure 4 , panel i to p ) ; it is also able to overcome an apparent initial increase in ATRX expression after sodium-butyrate treatment of cells , which can be seen in cells with GFP-expression alone . Interestingly , we also observed the disappearance of Sp100 from ND10 in lytically infected cells Fig . 4 , panels f to k ) . This indicates that a lytic gene product of KSHV may antagonize Sp100 . We had recently found that the lytic orf3 gene-product of the related primate rhadinovirus HVS induces the proteasomal degradation of Sp100 [16] . Thus , we speculated that the homologous ORF75 protein of KSHV may be involved in the depletion and relocalization of ND10 components . To investigate this , primary human fibroblasts were transfected with expression constructs for ORF75 , and the integrity of ND10 components was visualized by multicolor immunofluorescence . The ORF75 protein shows a colocalization with the main ND10 component PML ( Figure 5A and B ) . Moreover , compared to untransfected cells , ORF75 induced the relocalization of PML . However , we never observed a complete loss of spotted PML staining but a decrease in the number of ND10 dots per nucleus , and the remaining dots showed an increase in size ( Figure 5 ) . In addition , ORF75 induced the removal of ATRX , which is in accordance with ATRX disappearing after de novo KSHV-infection ( Figure 5B , panels f to k and Figure 4 ) . Furthermore , the spotted ND10-associated appearance of Daxx was lost in cells expressing ORF75 compared to untransfected cells ( Figure 5A panels f , g ) . Quantification of ND10-dots per nucleus revealed a significant reduction in the number of PML- , Sp100- , Daxx- and ATRX-dots per cell ( Figure 6A ) . It is worth to note , that the variation in dot number per nucleus in staining of different ND10 components is most likely related to the signal/noise ratio of used antibodies . The loss of ATRX and Daxx dots per nucleus seemed to be dependent on ORF75 levels; cells expressing high amounts of ORF75 showing both nuclear and cytoplasmic staining of ORF75 had a complete loss off ATRX and Daxx dots , whereas cells with ORF75 staining restricted to nuclei showed only reduced numbers of dots ( Figure 5 ) . Along this line , by transfection of myc-tagged ORF75 into 293T cells , we could demonstrate that ORF75 leads to a dose dependent loss of ATRX expression as detected by Western Blotting ( Figure 6B ) . In contrast to ATRX , Daxx expression levels remained unaffected ( Figures 6B+6C ) . Moreover , we could demonstrate that transfection of increasing amounts of ORF75-myc had no influence on Sp100 protein levels , indicating that ORF75 does not mediate a degradation of Sp100 ( Figure 6C ) . In cells expressing high amounts of ORF75 , we detected a relocalization and frequently also a complete loss of spotted Sp100 staining; PML was relocalized to more intense spots but not dispersed . This suggests that ORF75 is involved in the dispersion of both Sp100 and Daxx during lytic replication but not in Sp100 degradation ( Figure 5A , panel e–h; 5B , panels b–d; g–i , Figure 6 ) . The depletion of ATRX in de novo infected human fibroblasts , which are not fully permissive for KSHV , hinted at the involvement of a virion protein . We then wanted to address the question whether ORF75 is truly a tegument protein and therefore part of the virion , as it was assumed in previous studies [23] , [24] . To this end , recombinant viruses were generated for further analysis of the role of KSHV ORF75 ( Figure 7A ) . The orf75 gene is located close to the right end of the L-DNA . We generated recombinant KSHV mutants by using recombination-mediated , site-directed mutagenesis of cloned KSHV BACmid Bac16 [25] . Mutant viruses were verified using XhoI restriction enzyme digestion , PCR , PFGE , sequencing of recombination junctions , and Illumina sequencing of the Bac16-75-Stop-revertant and -orf75-Flag ( data not shown ) . Pulsed field gel electrophoresis shows expected restriction patterns of wild type and deletion mutant BACs , including the full maintenance of the terminal repeat sequences ( data not shown ) . All of these experiments proved that the recombinant genomes contained the correct deletion , with no other detectable rearrangements , and were therefore used for further analysis . iSLK cells were transfected with KSHV Bac16 or Bac16-orf75-Flag DNA , selected with hygromycin B , and lytic viral replication induced with doxycyclin and sodium-butyrate . Three days post induction , cells were harvested and Flag-tagged ORF75 was stained by immune electron microscopy ( IEM ) with a Flag-tag specific antibody and a secondary antibody coupled to 12 nm gold-particles . Electron microscopy imaging revealed a tegument specific gold staining of virions in iSLK-Bac16-Flag cells , which was absent in iSLK-Bac16 wt cells , indicating that the ORF75 protein is truly part of the viral particle ( Figure 7B ) . We could not detect virus particle production by IEM in iSLK-Bac16-orf75-Stop cells after lytic viral induction despite multiple attempts ( Figure 7B ) . Recombinant viruses were reconstituted by transfection of BAC DNA into iSLK cells followed by selection with hygromycin B . Cells were treated with sodium-butyrate and doxycyclin to induce lytic virus production . We were able to reconstitute recombinant viruses carrying Flag-tagged or YFP-tagged versions of the ORF75 protein , however , despite multiple attempts we failed to reconstitute virus carrying a STOP codon at the beginning of the orf75 gene . Nine individual iSLK-Bac16-orf75-Stop clones were tested for the production of infectious virus by transfer of supernatant from induced iSLK-Bac16-orf75-Stop clones onto 293T cells; GFP-expression resulting from KSHV infection was measured by flow cytometry two days post infection . While the control showed infection rates above 50% , we could not detect infection of 293T cells using supernatant of any of the nine orf75-Stop clones ( Figure 8A ) . Furthermore , KSHV ORF75 appears to be essential for lytic viral induction , since we could not detect lytic viral gene expression in the absence of ORF75 ( Figure 8B ) . iSLK cells harboring different recombinant Bac16 clones were induced with doxycyclin and sodium-butyrate for 2 days and lytic gene expression determined by Western Blotting . Intriguingly , we could not detect expression of KSHV genes K8 , K8 . 1 , and v-IRF3 in iSLK-Bac16-orf75-Stop cells . In contrast , a Bac16-orf75-revertant , constructed by repairing the engineered stop-codon of the orf75-Stop mutant , showed lytic gene expression comparable to that of the wildtype , indicating that the defects observed in the orf75-Stop mutant are in fact due to a lack of orf75 expression . Further support for the essential role of ORF75 comes from knockdown experiments of ORF75 using siRNA and rescue of infectious virus from the iSLK-Bac16-orf75-Stop cells by lentiviral transduction with tet-on ORF75 ( Figures S3 and S4 ) . In addition , all cell lines showed regular expression levels of latent genes Lana/orf73 and GFP expressed from the viral genome which increased after lytic induction in all cell lines except iSLK-Bac16-orf75-Stop cells . Finally , when tested for viral DNA replication in iSLK cells by quantitative real time PCR , we could detect only a minimal increase in viral copy numbers per cell in iSLK-Bac16-orf75-Stop cells after lytic viral induction , whereas the controls showed a normal increase in copy numbers corresponding to lytic replication ( Figure 8C ) . This might indicate that , in addition to being a tegument protein important for infectious virion production , ORF75 could have an additional role in generating a permissive environment for viral lytic DNA replication . Accordingly , orf75-mRNA can be detected as early as RTA-mRNA after infection of SLK cells ( Figure 8D ) . In addition , ORF75 protein is present in infected cells during the entire viral replication cycle , starting from 6 h post infection up to 72 h post infection ( Figure 8E ) . In order to further clarify the role of KSHV-ORF75 during lytic replication , we used the recombinant viruses expressing ORF75 either with C-terminal Flag- or YFP-tag . After lytic induction of iSLK cells harboring the respective BACs , ORF75-YFP as well as ORF75-Flag colocalize with PML ( Figure 9A and 9B ) . Moreover , only cells undergoing lytic replication and subsequently showing ORF75 expression exhibited a loss of both ATRX and Sp100 expression , revealing the role of ORF75 in ND10 antagonization . KSHV vIRF3/LANA2 overexpression has previously been associated with redistribution of PML , and a moderate decrease in the number of ND10 dots per cell [13] . Therefore , to separate the effects of ORF75 and vIRF3 on ND10 composition , we generated a recombinant virus expressing ORF75-YFP with additional deletion of the vIRF3 gene . After lytic induction of iSLK-Bac16-Δvirf3-orf75-YFP cells , ORF75-YFP colocalizes with PML ( Figure 10 ) . In addition , this experiment recapitulated the loss of both ATRX and Sp100 expression ( Figure 10 ) , and clearly shows that vIRF3 is not required for the ORF75-mediated alterations of ND10 composition . We demonstrated that KSHV ORF75 mediates the depletion of ATRX and is essential for lytic viral replication . To this end , we addressed the question whether a knock-down of ATRX can re-establish lytic viral replication in the absence of ORF75 . iSLK-Bac16-orf75-Stop cells were transduced with a lentiviral shRNA vector targeting ATRX ( shATRX ) and a respective non-silencing control vector ( shC ) . The ATRX-specific shRNA mediates an almost complete knock-down of ATRX protein in iSLK cells ( Figure 11A ) . After lytic induction of knock-down cells with doxycyclin and sodium-butyrate for three days and transfer of supernatant onto freshly seeded 293T cells , we could not detect infection of cells with supernatant of iSLK-Bac16-ORF75-Stop cells treated with shC ( Figure 11B ) . However , supernatants from iSLK-Bac16-orf75-Stop cells treated with shRNA specific for ATRX contained infectious virus capable of infecting 293T cells , indicating that a viral ORF75 deletion mutant can be rescued by knock down of ATRX ( Figure 11B ) . ATRX knockdown also leads to enhanced lytic viral gene expression . In iSLK-Bac16-orf75-Stop cells , shRNA mediated knockdown of ATRX is able to rescue expression of the lytic gene K8 . 1 , emphasizing the role of ORF75 mediated ATRX counteraction for viral replication ( Fig . 11C+D ) .
The counteraction of ND10-instituted intrinsic immunity as a critical step during the replication cycle of DNA viruses and particularly of herpesviruses has become more and more evident over the last decade . Several ND10 components , like PML , Sp100 , Daxx and ATRX have been shown to be part of a cellular defense mechanism against viral infection and therefore became targets of viral effector proteins . Recently , members of the viral formylglycineamid-ribotideamidotransferase ( vFGARAT ) family have been added to the list of herpesviral ND10-antagonists . We demonstrated that one of the two vFGARATs of Herpesvirus saimiri - the ORF3 protein - leads to a selective proteasomal degradation of the ND10 component Sp100 [16] . Moreover , another vFGARAT , the BNRF1 protein of Epstein-Barr virus , has recently been shown to interact with the ND10 component Daxx , resulting in a release of the chromatin remodeling factor ATRX from ND10 , and thereby facilitating viral replication [18] . One out of three viral FGARAT proteins of the murid gammaherpesvirus 68 ( MHV-68 ) , ORF75c , has been shown to fully disperse cellular ND10 complexes by inducing the proteasomal degradation of PML; it further reduced SP100 isoforms and SUMOylation , and reduced Daxx levels [17] , [26] . Thus , although vFGARAT homologs BNRF1 , ORF75c and ORF3 all target ND10 components , a comparative analysis of vFGARAT proteins revealed that the ORF3 FGARAT homologs , the rhadinoviral ORF75 homologs , and the lymphocryptoviral BNRF1 homologs cluster within different branches of a phylogenetic tree [16] . Apparently , viral FGARAT homologs have evolved in different ways , probably reflecting different viral survival strategies . Whereas MHV-68 is able to antagonize PML , which drastically enhances viral lytic replication , KSHV and HVS are not able to counteract PML-mediated restriction per se ( Figure 1C ) . It is notable in this context that the outcome of KSHV infection in vitro is by default latency , whereas the situation in vivo in humans is very difficult to assess . Therefore , it is highly remarkable that there is an increase in viral gene expression after depletion of both PML and Sp100 , as shown by de novo GFP expression ( Figure 1 ) . According to the current model that ND10 components act as restriction factors by creating a repressive environment for viral gene-expression , this increase in GFP expression in PML- and Sp100-knock-down fibroblasts is most likely due to increased viral lytic gene expression in the absence of ND10 factors . KSHV infection leads to an initial selective depletion of ATRX and relocalization of Daxx , whereas PML and Sp100 remain unaffected ( Figures 1D and 2A and B ) . This is in coincidence with our data of KSHV infection of ATRX knock-down cells: since ATRX is efficiently antagonized by the virus , there is no further enhancement of KSHV infection after ATRX depletion ( Figure 1 ) . We could further observe a modest increase in GFP-expression after Daxx depletion , indicating that Daxx can only be partially antagonized by KSHV . Alpha- and beta-herpesviruses , like herpes simplex virus and human cytomegalovirus , have evolved very efficient counter-mechanisms that dissolve ND10 , while primate gammaherpesviruses are lacking the full ability to counteract a central part of the ND10 intrinsic immunity . Therefore , there is good evidence to support the hypothesis that ND10 components like PML may contribute to the efficient establishment of latent infection in target cells , leading to the accumulation of secondary mutations over time and ultimately resulting in oncoprotein mediated cell transformation , as it is the case for most gamma-herpesviruses . On the other side , lytic disease and lytic reactivation is the observed outcome for most alpha- and beta-herpesviruses that can overcome ND10 restriction . KSHV encodes a single vFGARAT homolog , the ORF75 protein . ORF75 has been described as a tegument protein , due to its homology to other vFGARATs and its presence in preparations of purified virions [23] , [24] . We could confirm Flag-tagged ORF75 within KSHV virions by immune electron microscopy ( Figure 7B ) . In addition , in a genome-wide screen for NF-kB activation by KSHV proteins , ORF75 has been shown to activate the NF-kB signaling pathway [19] . The role of ORF75 during lytic viral replication had not been assessed so far . Here we could demonstrate that the ORF75 protein of KSHV is involved in the antagonization of ND10-mediated immunity . Isolated expression of ORF75 in primary human cells led to alterations in ND10 composition reminiscent of the situation during lytic viral replication ( Figures 4 , 5 and 6 ) : i ) relocalization of ND10 components including PML; ii ) depletion of ATRX; iii ) relocalization and dispersion of Sp100; iv ) dispersion of Daxx . Using quantification of immunofluorescent images and immunoblotting we could show that ORF75 acts in a dose dependent manner on ND10 components Daxx and ATRX ( Figure 6 ) . In addition we could also observe a strong colocalization of ORF75 and PML . These findings could also be confirmed during lytic viral replication , using either Flag-tagged or YFP-tagged ORF75 protein expressed by recombinant viruses ( Figures 9 and 10 ) . Cells undergoing lytic replication showed a relocalization of PML , a colocalization of ORF75-YFP and PML , exclusion of Sp100 and Daxx from ND10 , and loss of ATRX protein expression; this further corroborates the role of ORF75 in ND10 antagonization . For the generation of infectious viruses from recombinant KSHV genomes , we used iSLK cells transfected with BAC DNA . Although it has recently been demonstrated that iSLK cells are not of endothelial origin as it has been assumed previously [22] , this cell line can be used for efficient virus production . According to our experience , the internal repeat and multiple TR sequences present in KSHV mandate extra caution . In contrast to other BAC-cloned herpesvirus genomes , which seem to be rather stable during homologous recombination in E . coli , we frequently observed shortening of terminal repeat sequences from Bac16 recombinants . Since this cannot be detected by conventional gel electrophoresis and DNA sequencing , it is notable that we routinely verify KSHV BAC16 recombinants by pulsed field gel electrophoresis ( PFGE ) as well as Illumina deep sequencing of final constructs . It would have been of great interest to see whether the ablation of ORF75 expression affects the antagonization of ND10 components during infection and lytic replication . Intriguingly , we were unable to reconstitute infectious virus and could detect neither lytic gene expression nor viral DNA replication in the absence of ORF75 . The expression of latent genes Lana/orf73 and GFP was not affected in unstimulated iSLK cells . Thus , to our own surprise , ORF75 seems to be essential for lytic replication at multiple levels . We could show for HVS that individual deletion of vFGARATs ORF3 and ORF75 , respectively , leads to an impaired viral replication , but did not entirely abrogate viral replication [16] . However , for HVS , we were so far not able to reconstitute a virus with deletion of both ORF3 and ORF75 , hinting at the important role of at least one vFGARAT in the genome of a rhadinovirus ( Full and Ensser , unpublished data ) . In this context , it would also be of great interest to elucidate the role of ORF75 of rhesus rhadinovirus ( RRV ) , the closest known relative of KSHV , during lytic replication . The loss of virus production in the absence of ORF75 hints at multiple functions of the protein within the viral life cycle , as it has also been shown for ORF75c of MHV68 [26] . Since ORF75 is a tegument protein ( Figure 7B ) , it may well be that it is an integral part of the virion and necessary for assembly of viral particles . However , the loss of an important structural component of the virion would not explain the complete absence of lytic gene expression and viral replication . In ORF75c-deleted MHV68 viral capsids are still made and accumulate , while no particles are apparent in orf75-stop KSHV ( Figure 7B ) . It is thus remarkable that we cannot detect lytic gene expression despite overexpression of RTA in induced iSLK cells ( Figure 8B ) . It is also unlikely that the loss of K13/vFLIP-mediated NF-kB ( co ) activation by ORF75 leads to this effect , as the KSHV genome comprises multiple potent activators of NF-kB signaling [19] . Our data hint at a role of ORF75 in the initiation of lytic gene expression . As a prerequisite , ORF75 protein must be expressed during latency or at least early during initiation of the lytic cycle . This is indeed the case ( Figure 8D+E ) : after infection , ORF75 is released from the viral tegument into the cell . As soon as immediate early gene expression starts , de novo ORF75 protein is made in small amounts and at later timepoints in increasing amounts until new viral particles are released form the cell . Thus ORF75 protein is present for the entire viral replication cycle . The orf75 gene is situated within a region at the right end of the KSHV genome that is transcribed during latency , and genome-wide analysis of epigenetic modifications of latent KSHV genomes revealed histone modification patterns in the putative orf75 promoter region which reflect actively transcribed chromatin [27] , [28] . Therefore , it would be very interesting , but beyond the scope of this article , to test for potential epigenetic mechanisms underlying the loss of viral DNA replication in the absence of ORF75 , and the role of ORF75 during latency of KSHV . Concerning the role of ORF75 during lytic replication , we could clearly show that ORF75 is essential for viral replication . Moreover the dispersion of ATRX by ORF75 seems to be crucial for lytic replication of KSHV; a KSHV-orf75 deletion mutant that is not capable of lytic viral replication can at least be partially rescued by knock-down of ATRX ( Figure 11 ) , clearly emphasizing the role of ORF75 mediated ATRX relocation . The fate of ATRX after ORF75 mediated removal from ND10 remains elusive . Proteasomal degradation of ATRX can be excluded ( Figure 3A+B ) . Although difficult to prove , we hypothesize that ATRX might be translocated to a highly insoluble nuclear fraction , and therefore can no longer exert its antiviral function . KSHV infection leads to a dispersion of the ND10 component ATRX ( Figures 1D and 2 ) . We could not detect a degradation of Daxx , although KSHV infection also led to a loss of spotted Daxx localization ( Figures 1D , 2 , 3 ) . Apparently , Daxx is also dispersed from ND10 . Whether the Daxx dispersion is a direct or indirect effect of KSHV-induced loss of ATRX could not be demonstrated . Using co-immunprecipitation experiments , we could not detect a direct interaction of ORF75 with Daxx as it has been shown for BNRF1 of EBV [18] . This is also in accordance with findings by Tsai et al [18] , indicating that EBV- and KSHV-FGARATs target the same proteins but have evolved different mechanisms in the antagonization of ND10 components . Isolated expression of ORF75 after transfection led to alterations in ND10 composition similar to observations after lytic replication . In order to separate the functions of vIRF3 and ORF75 , we constructed a recombinant virus harboring a deletion of the vIRF3 gene in combination with either YFP-tagged ORF75 ( Figure 10 ) or Flag-tagged ORF75 protein ( data not shown ) . It has been shown that vIRF3 is SUMOylated , and that overexpression of vIRF3 leads to a moderate decrease in ND10 foci per cell [13] , [29] . However , the phenotype of vIRF3-deletion mutants compared to wildtype virus with respect to ND10 was indistinguishable . Since it is known , that ND10 components can be induced by IFN , the decrease in ND10 dots per cell may also be a result of IFN antagonization by vIRF3 [30]–[32] . In summary , our results clearly revealed that ORF75 is essential for viral replication . Thus , ORF75 could be a new and specific target for antiviral therapy , since inhibition of ORF75 function should abrogate viral lytic replication . This is of particular interest also for therapy of KSHV tumors , since it is known that lytic replication of KSHV is of particular importance for KS tumorigenesis in vivo . KS tumors regularly show lytic viral replication at low levels [33] , [34] , and a clinical trial demonstrated that the development of new KS lesions in patients can be blocked by administration of ganciclovir , an inhibitor of the viral DNA polymerase which is required for lytic viral replication [35] , [36] .
Primary human foreskin fibroblasts ( HFF ) , were prepared from human foreskin-tissue as described previously [37] and HFF with a small interfering RNA-mediated knockdown of PML ( siPML cells ) , Daxx ( siDaxx cells ) , Sp100 ( siSp100 cells ) , ATRX ( siATRX cells ) and the respective control cells ( siC cells ) were cultured in Dulbecco's minimal essential medium ( DMEM , LifeTechnologies , Germany ) supplemented with 10% fetal calf serum and 5 µg/ml puromycin [38]–[40] . HeLa cells and HEK 293T cells were maintained in Eagle's minimal essential medium ( LifeTechnologies ) supplemented with 10% fetal calf serum . iSLK cells were maintained in DMEM supplemented with 10% fetal calf serum , 2 . 5 µg/ml puromycin and 250 µg/ml G418 . KSHV from BAC-DNA was obtained by reconstitution of infectious viruses after transfection of BAC16 DNA into iSLK cells . iSLK cells were transfected in 6-wells with 5 µg BACmid DNA with 9 µl ExtremeGene ( Roche ) according to manufacturer's instructions , and selected with 200 µg/ml hygromycin B until all cells were GFP positive . Lytic replication of KSHV was induced in iSLK harboring KSHV Bac16-DNA cells by plating cells in the absence of antibiotics and adding 900 µM sodium-butyrate and 1 µg/ml doxycyclin . UV inactivation was done by application of 120 . 000 microjoules/cm2 using the Stratalinker UV Crosslinker 1800 ( Stratagene , Amsterdam , the Netherlands ) . Supernatants containing infectious virus were harvested 3 d post induction and stocks of wildtype and recombinant viruses were stored at −80°C in aliquots . Viral replication in induced iSLK cells was monitored by quantitative PCR of the KSHV orf26 locus and compared to cellular ccr5 DNA . The KSHV Bac16 clone was used for recombination-mediated genetic engineering of recombinants by a two-step , markerless λ-red-mediated recombination strategy using the kanamycin gene as a first selection marker [41] . Linear fragments for homologous recombination were generated by PCR using Phusion High Fidelity DNA polymerase ( Finnzymes ) ; DpnI was added to digest the methylated plasmid template , and the amplification product was purified from an agarose gel with the Nucleobond Gel Extraction Kit ( Macherey&Nagel , Düren , Germany ) . Primers that were used to generate linear fragments for the manipulations were purchased as Ultramers™ from Integrated DNA Technologies . For homologous recombination , the PCR fragment was then transformed into Escherichia coli strain GS1783 ( gift of G . Smith , NW University , USA ) [42] harboring BAC16 , and Red recombination was performed as described earlier [41] , [43] . Cells were then plated on agar plates containing 15 µg/ml kanamycin ( first recombination ) or 15 µg/ml chloramphenicol and 1% arabinose ( second recombination ) and incubated at 32°C for 1–2 days . Bacterial colonies growing on these plates were further analyzed . Reconstitution of recombinant KSHV using purified BAC DNA was performed in iSLK cells as described above . BAC DNA was isolated by standard alkaline lysis from 5 ml liquid cultures . Subsequently , the integrity of BACmid DNA was analyzed by digestions with restriction enzyme XhoI and separation in 0 . 8% PFGE agarose ( Bio-Rad ) gels and 0 . 5×TBE buffer by pulse field gel electrophoresis at 6 V/cm , 120 degree field angle , switch time linearly ramped from 1 s to 5 s over 16 h ( CHEF DR III , Bio-Rad ) . For characterization of insertion or deletion of the aphaI selection marker , recombinant BACmids were analyzed by PCR , and the recombined regions within the BAC DNA were controlled by sequence analysis ( ABI 3130XL genetic analyzer , Weiterstadt , Germany ) in order to confirm the correct deletion sequences and to exclude accidental mutations . Furthermore , the integrity of the complete viral genome of representative Bac16 recombinants was confirmed by next-generation sequencing using Nextera DNA Sample Preparation system and the MiSeq Reagent Kit , 300 Cycles on the Illumina MiSeq system . NGS data were analyzed by Genome Workbench 5 ( CLCbio , Aarhus , DK ) . Endogenous PML was detected with mouse monoclonal antibody PG-M3 ( IgG1 , Santa Cruz Biotechnology ( SCBT ) or 5E10 ( gift of R . van Driel , University of Amsterdam , The Netherlands ) . Human Sp100 was detected with a rabbit polyclonal antiserum ( ProteinTech Group , USA ) or a mouse polyclonal antibody ( MaxPAP , Abnova , USA ) . A rabbit polyclonal antibody from Sigma-Aldrich ( Germany ) or the mouse monoclonal antibody MCA2143 ( IgG1 , Serotec , Germany ) were used for detection of hDaxx . ATRX protein expression was detected with mouse monoclonal antibody clone D-5 ( IgG2a , SCBT ) . Rabbit polyclonal antibody for detection of beta-actin , was purchased from Sigma-Aldrich , GFP-specific rabbit polyclonal antibody was purchased from Genescript ( Piscataway , NJ ) . Epitope-tagged proteins were either detected using the anti-FLAG tag antibody M2 ( Sigma-Aldrich , Germany ) or the anti-myc tag antibodies clone 9E10 ( IgG1 , Sigma-Aldrich ) or clone 9B11 ( IgG2a , Cell Signaling Technologies , Germany ) . Horseradish-peroxidase-conjugated anti-mouse and anti-rabbit secondary antibodies for Western blot analysis were obtained from Dianova ( Hamburg , Germany ) , while Alexa Fluor 488- , Alexa Fluor 555- , and Alexa Fluor 647-conjugated secondary antibodies for indirect immunofluorescence experiments were purchased from LifeTechnologies ( Germany ) . For detection of KSHV proteins , KSHV-antibodies to K8 ( clone 8C12G10G1 , SCBT , USA ) , K8 . 1 ( clone 4A4 , SCBT , USA ) , vIRF3 ( rat monoclonal , kindly provided by Frank Neipel ) and ORF75 KSHV ( polyclonal rabbit serum , Genescript ) were used . For immunofluorescence analysis , iSLK cells or HFF cells were plated onto coverslips . Lytic replication of KSHV in iSLK cells was induced as described above . Cycloheximide was used at 10 µg/ml , MG132 at 10 µM . At indicated timepoints post induction , infection or transfection , cells were washed three times with phosphate-buffered saline ( PBS ) , followed by fixation with 4%paraformaldehyde for 10 min at room temperature . Then , cells were permeabilized with PBS-0 . 2% Triton X-100 , 5% FCS for 1 hour , followed by incubation with the respective primary or secondary antibodies for 30 min at 37°C . Finally , the cells were mounted by using Mowiol mounting medium ( Fluka , Germany ) plus 4′ , 6-diamidino-2-phenylindole ( DAPI; Vector Laboratories , USA ) or Hoechst ( Sigma-Aldrich , Germany ) . The samples were examined by using a Leica TCS SP5 confocal microscope , with 405 nm , 488 nm , 543 nm or 633 nm laser lines , scanning each channel separately under image capture conditions that eliminated channel overlap . The Fiji distribution of ImageJ was used to count nuclear dots in ORF75 expressing human fibroblasts [44] . Briefly , nuclei were identified by watershedding and the analyze particles function in the DAPI image , then defined as ROI; after setting an appropriate noise tolerance , single point maxima in the channel corresponding to ND10 proteins were then counted within each nuclear ROI and results copied to MS Excel . Statistical analyses were conducted using the unpaired t-test for SP100 and PML , and one way ANOVA test for multiple comparisons for Daxx and ATRX , respectively ( GraphPad Prism 6 for Windows , GraphPad Software , USA ) . Extracts from infected cells were prepared either directly in a sodium dodecyl sulfate-polyacrylamide gel electrophoresis ( SDS-PAGE ) loading buffer or in a RIPA-lysis buffer as described previously , separated either on sodium dodecyl sulfate-containing 8% , 10% 12% polyacrylamide gels or 4–12% bis-tris gradient gels ( Life Technologies , Germany ) , and transferred to polyvinylidene fluoride membranes ( Millipore ) . Enhanced Chemi-luminescence was detected according to the manufacturer's protocol ( GE Healthcare ) . HFF cells or 293T cells were detached using PBS supplemented with 0 . 1% EDTA , pH 7 . 4 , spun down ( 300×g , 5 min ) , supernatant removed and then fixed in PBS supplemented with 2% paraformaldehyde for 30 min . GFP expression of cells was analyzed on a LSRII flow cytometer ( BD Biosciences ) and data were analyzed with FCS Express 3 software ( De Novo Software , Canada ) . For cell cycle analysis , cells were harvested and fixed in 80% ethanol at least over night . Staining after RNase treatment ( 50 µg/ml ) was done with propidium iodide ( 20 µg/ml ) and analyzed on a BD LSR2 flow cytometer . Cell cycle data were modeled with ModFitLT 3 . 3 for Windows ( Verity Software House , Topsham , ME ) . Induced iSLK cells were harvested in Dulbeco's modified Eagle's medium ( DMEM ) 48 h post induction . Cells were fixed with DMEM containing 4% paraformaldehyde , dehydrated at −20°C and embedded in LR-White at −20°C . Polymerization was induced by heat for 2 days at 60°C prior to sectioning using an ultramicrotome ( Leica , Wetzlar , Germany ) . The sections were transferred to nickel grids and immunostaining was performed with anti-FLAG antibody ( clone M2 , Sigma-Aldrich ) and secondary anti-mouse antibody conjugated with 12 nm gold ( Dianova ) . The grids were stained for 10 min with 1% ( w/v ) uranyl acetate in 40% Ethanol followed by lead citrate staining for additional 10 min prior to analysis by electron microscopy . Specimens were observed using a TecnaiTM G2 Scanning transmission electron microscope ( FEI Company , Eindhoven , The Netherlands ) operated at 120 kV . | Kaposi's Sarcoma-Associated Herpesvirus ( KSHV ) establishes a lifelong persistent infection in humans and is associated with tumors and lymphoproliferative disease , particularly upon immunosuppression . The virus has to overcome cellular intrinsic immunity in order to initiate viral protein expression and genome replication in primary infection . We demonstrated that KSHV is restricted by a cellular intrinsic immunity complex called nuclear domain 10 ( ND10 ) and identified a critical role of the KSHV ORF75 protein , which is part of the viral particle , in this process . We found that ORF75 is essential for viral replication and that ORF75 leads to disappearance of the ND10 protein ATRX . Furthermore , it induces the relocalization of several other ND10 components . Noteworthy , all herpesviruses studied so far have evolved mechanisms for ND10 counteraction , indicating the importance of this step for herpesviral replication . The individual mechanisms , however , including the extent of ND10-antagonization , are of considerable variation between different herpesviruses . We speculate that , in contrast to efficient lytic replication of alphaherpesviruses , less effective ND10 counteraction may represent a doorway for gammaherpesviruses to latent infection . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"medicine",
"viral",
"transmission",
"and",
"infection",
"immunology",
"microbiology",
"cancers",
"and",
"neoplasms",
"epstein-barr",
"virus",
"infectious",
"mononucleosis",
"oncology",
"hiv",
"opportunistic",
"infections",
"infectious",
"diseases",
"bone",
"and",
"soft",
"tissue",
"sarcomas",
"hiv",
"biology",
"kaposi",
"sarcoma",
"host",
"cells",
"clinical",
"immunology",
"immunity",
"virology",
"innate",
"immunity",
"hiv",
"clinical",
"manifestations",
"viral",
"diseases"
] | 2014 | Kaposi's Sarcoma Associated Herpesvirus Tegument Protein ORF75 Is Essential for Viral Lytic Replication and Plays a Critical Role in the Antagonization of ND10-Instituted Intrinsic Immunity |
Mammary gland development commences during embryogenesis with the establishment of a species typical number of mammary primordia on each flank of the embryo . It is thought that mammary cell fate can only be induced along the mammary line , a narrow region of the ventro-lateral skin running from the axilla to the groin . Ectodysplasin ( Eda ) is a tumor necrosis factor family ligand that regulates morphogenesis of several ectodermal appendages . We have previously shown that transgenic overexpression of Eda ( K14-Eda mice ) induces formation of supernumerary mammary placodes along the mammary line . Here , we investigate in more detail the role of Eda and its downstream mediator transcription factor NF-κB in mammary cell fate specification . We report that K14-Eda mice harbor accessory mammary glands also in the neck region indicating wider epidermal cell plasticity that previously appreciated . We show that even though NF-κB is not required for formation of endogenous mammary placodes , it is indispensable for the ability of Eda to induce supernumerary placodes . A genome-wide profiling of Eda-induced genes in mammary buds identified several Wnt pathway components as potential transcriptional targets of Eda . Using an ex vivo culture system , we show that suppression of canonical Wnt signalling leads to a dose-dependent inhibition of supernumerary placodes in K14-Eda tissue explants .
The murine mammary gland development initiates at around embryonic day 10 . 5 ( E10 . 5 ) with the establishment of bilateral milk or mammary lines [1] . Between E11-E12 , five pairs of mammary placodes , local thickenings of the epithelium , emerge at conserved positions . By E13 . 5 , the placodes have transformed via hillock stage to buds that have submerged downward and are surrounded by several layers of a specialized dermis , the primary mammary mesenchyme [1] . As the tip of the primordium begins to elongate , at E15 . 5 , it forms a primary sprout that invaginates into the more distal secondary mammary mesenchyme . Branching morphogenesis begins a day later , and by birth a small ductal tree with several branches has formed . The murine mammary line is not externally visible but only detectable from histological sections or molecularly identifiable by expression of Wnt pathway genes such as Wnt10b or TOP-gal , a transgenic reporter of the canonical Wnt pathway [2 , 3] . Initially , the milk line is not a continuous structure but instead three independent Wnt10b-positive stripes arise: in axillary and inguinal regions , and the third one in the flank between the fore and hind limb buds . The axillary milk line gives rise to placode 1 , inguinal to placode 5 , and placodes 2 , 3 and 4 form from the milk line of the flank [2] . Establishment of placodes is asynchronous and expression analysis of the Wnt pathway mediator Lef1 revealed a designated order: 3 , 4 , 1/5 and 2 [4] . As the placodes form , low level of Wnt10b expression transiently combines all three milk lines but by E12 . 5 Wnt10b expression becomes confined to mammary buds [2 , 3] . Placode morphogenesis is thought to rely mainly on migration of the progenitor cells along and from the immediate vicinity of the milk line and not on proliferation [5 , 6] . Similar to other ectodermal appendages such as hair follicles and teeth , reciprocal interactions within and between the epithelium and the underlying mesenchyme are a necessity for proper development and pattering of mammary glands [1 , 7 , 8] . These interactions are mediated by conserved signaling pathways , of which at least the fibroblast growth factor ( Fgf ) , Wnt/β-catenin , and Neuregulin ( Nrg ) /ErbB pathways regulate mammary placode formation . Mammary gland initiation relies on a complex interplay between these pathways and transcription factors Gli3 and Tbx3 and their absence disrupts formation of one or more placode pairs ( reviewed in [9] ) . Fgf10 , emanating from the tip of the thoracic somites and the limb buds , has been proposed to function as one of the earliest signals for milk line specification . In the absence of Fgf10 or its receptor FgfR2b , only placode pair four develops [2 , 4] . Wnt/β-catenin signaling is required for all mammary placodes to form . Ectopic ectodermal expression of the secreted Wnt inhibitor Dkk1 abolishes all signs of mammary placodes [3] . Disruption of the Hedgehog pathway mediator Gli3 leads to loss of placodes 3 and 5 [10 , 11] . In Tbx3 null embryos , all mammary placodes are absent with the exception of occasional presence of placode 2 [12] . Tbx3 has been proposed to act both up- and downstream of Fgf and Wnt pathways but the details of these interactions are not well understood [12–14] . Finally , hypomorphic Nrg3 mutant mice display frequently missing or hypoplastic placode 3 but also supernumerary placodes [15] , whereas ectodermal overexpression of Nrg3 induces multiple supernumerary mammary glands along and adjacent to the milk line [16] . Another important player in embryonic mammary gland development is the tumor necrosis factor ( Tnf ) superfamily ligand Ectodysplasin-A1 ( hereafter Eda ) and its receptor Edar . The Eda pathway has a well characterized role in the development of diverse set of ectodermal organs [17 , 18] . The ectodermal appendage phenotype of Eda null mice ( Tabby mice ) and mice with compromised activation of transcription factor NF-κB is highly similar [19] , and biochemical and genetic studies have confirmed the importance of NF-κB downstream of Eda [18 , 20] . In humans , mutations in the genes encoding EDA , EDAR , or the cytosolic signal mediator EDARADD cause a condition known as hypohidrotic ectodermal dysplasia ( HED ) . In addition to tooth , hair , and sweat and salivary gland defects , breast anomalies such as hypoplastic/absent/supernumerary nipples and even absence of breast tissue have been reported in HED patients [21–23] . Studies using Eda loss- and gain-of-function mouse models have shown that Eda regulates embryonic and prepubertal mammary gland branching morphogenesis via NF-κB [24] . However , all five mammary glands form in Eda null mice suggesting that Eda is dispensable for mammary placode formation [24 , 25] . Strikingly , ectodermal overexpression of Eda ( K14-Eda mice ) leads to formation of supernumerary mammary placodes along the milk line , in particular in the region between mammary buds 3 and 4 , and give rise to supernumerary mammary glands in the adult [26 , 27] . Beyond this , little is known about the importance of Eda in the initial stages of mammary gland development . We report here that NF-κB is dispensable for mammary placode induction , yet it is necessary for the ability of Eda to induce supernumerary mammary primordia . Using an unbiased genome-wide approach , we identify several transcriptional targets of Eda . We provide evidence indicating that Eda promotes mammary cell fate by enhancing canonical Wnt signaling activity . Furthermore , we find that Eda induces supernumerary mammary glands not only between the endogenous mammary glands , but also in the neck region . Based on analysis of wild-type and K14-Eda embryos we propose that the murine mammary line extends more anteriorly than previously recognized .
Embryonic mammary primordia exhibit high Eda-dependent NF-κB activity from E12 onwards [24 , 25] . To gain further insights on the role of the Eda/NF-κB pathway in early mammogenesis , we assessed NF-κB signaling activity with reporter mice expressing β-galactosidase under an NF-κB–responsive element in control and K14-Eda embryos . We detected NF-κB activity in the mammary placode forming region from E11 onwards ( Figs 1 and S1 ) . At E11 . 0 , a low level reporter expression was detected in the region of future mammary placode 3 and the interface of the forelimb bud and the thorax where placode 1 will later appear ( Figs 1A and S1 ) . At E11 . 25 faint expression was detected also at the border of the hind limb bud and ventrum ( prospective placode 5 ) , as well as at the site of future primordium 4 ( Figs 1B and S1B ) . By E11 . 5 reporter expression had intensified in placode 3 and become more condensed at placodes 1 , 4 , and 5 ( Fig 1C ) . Dispersed X-gal-positive cells were detected at the location of prospective placode 2 ( Figs 1C and S1D ) . In addition , modest amount of reporter positive cells were observed along the entire milk line , from placode 1 to 5 . At E12 . 0 high localized reporter expression was confined to the mammary buds and low level NF-κB activity was found throughout the dorsal side of the embryo whereas the ventrum appeared devoid of reporter expression ( Fig 1D ) . At these early stages , the reporter expression was constantly stronger in K14-Eda background ( Fig 1A–1D ) . Similar to previous reports on expression of the Wnt pathway genes and TOP-gal reporter , NF-κB reporter positive cells disappeared from the milk line between E12 . 5 and E13 . 5 in control embryos ( Fig 1E and 1F ) . In K14-Eda embryos , elevated NF-κB signaling was observed along the milk line , as well as in the dorsum , yet they exhibited no obvious focal clustering of X-gal-positive cells between buds 3 and 4 , i . e . at the site of prospective supernumerary primordia , until at ~E12 . 5 ( Fig 1C–1E ) . These foci were more pronounced at E13 . 5 , although reporter expression was markedly less intense than in endogenous buds ( Fig 1F ) . The milk line , the area possessing mammary inductive capacity , is considered to extend from the axilla to the groin [1] . To our surprise , we observed faint NF-κB reporter activity from E11 . 0 onwards also in the neck area , anterior to mammary bud 1 , which was substantially more pronounced in K14-Eda embryos ( arrowheads in Fig 1 ) . In K14-Eda embryos , reporter expression was confined to one or up to four small foci suggesting that supernumerary placodes were induced in the neck region . In situ hybridization analysis revealed high focal expression of Wnt10b , as well as Dkk4 , another placode marker [28] in endogenous placodes of E11 . 25 wild type and K14-Eda embryos , as well as in the neck region ( Fig 1G and 1H ) . The latter coincided with the site of ectopic placodes marked by NF-κB reporter expression in K14-Eda embryos ( compare Fig 1G and 1H to Fig 1A–1C ) . To analyze more in detail NF-κB activity , we sectioned whole mount stained reporter embryos ( Fig 2 ) . NF-κB activity was present throughout the developing mammary epithelium in control and K14-Eda embryos at E12 . 5 , similar to expression of Edar ( Fig 2A and 2B ) . At E13 . 5 , NF-κB reporter activity was mainly confined to the basal cells in control embryos ( Fig 2B ) , but remained high throughout the bud in K14-Eda embryos ( Fig 2C ) . Further , sectioning confirmed that supernumerary neck placodes were truly thickened at E12 . 5 ( Fig 2D , left column ) and showed that supernumerary mammary buds , in particular those between buds 3 and 4 , consisted of both reporter positive and negative cells ( Fig 2D , right column ) . Supernumerary mammary placodes forming between gland 3 and 4 give rise to nipples with an associated ductal system in K14-Eda adults , and are responsive to pregnancy hormones [26] . As suggested by embryonic analyses ( Fig 1 ) , a nipple was observed also in the neck region and was often accompanied by accessory , smaller nipple-like structures ( Fig 3A ) . However , the nipple-like structures in the neck region did not express keratin 2e , a specific marker of nipple epithelium [29] indicating defective differentiation of the nipple epithelium ( Fig 3B ) . Surprisingly , the neck region was also capable of supporting ductal morphogenesis ( Fig 3C ) . Similar to the supernumerary glands located between glands 3 and 4 [26] , the ductal trees in the neck were considerable smaller than those of the endogenous glands and displayed typical pregnancy-associated morphological changes ( Fig 3D and 3E ) . As discussed above , engagement of Edar leads to activation of NF-κB . In unstimulated cells , inhibitory IκB proteins , most commonly IκBα , retain NF-κB is in the cytosol [30] . Ligand binding leads to phosphorylation and degradation of IκBα thereby releasing NF-κB . To elucidate the importance of NF-κB signaling in the embryonic mammary placode development , we utilized the IκBαΔN mouse strain which displays suppressed NF-κB activity as a result of ubiquitous expression of a non-degradable IκBα [19] . Analysis of NF-κB reporter expression in IκBαΔN embryos at E11 . 25 revealed absence of reporter expression ( Fig 4A ) . This indicates that NF-κB signaling is fully suppressed in this mouse model at the time of mammary placode induction , similar to later developmental stages ( E12-E16 ) [24] . In situ hybridization analysis of Lef1 , which is expressed both in the mammary epithelium and the mesenchyme at E12 . 5 [31 , 32] , confirmed the presence of normal number of mammary primordia in IκBαΔN embryos ( Fig 4B ) , yet Wnt10b expression suggested that mammary buds may be somewhat smaller ( Fig 4C ) . Taken together , these data show that NF-κB activity is dispensable for mammary placode induction . To address the necessity of NF-κB for the ability of Eda to induce supernumerary placodes , we crossed K14-Eda strain with the IκBαΔN mice . Mammary placode markers Tbx3 , Wnt10b , and PTHrP [2 , 24 , 33] were expressed in mammary buds of wild type , IκBαΔN , K14-Eda and compound K14-Eda;IκBαΔN embryos at E13 . 5 ( Fig 5A–5C ) . Expectedly expression of PTHrP and Wnt10b appeared slightly downregulated in IκBαΔN background as they have been identified to be transcriptional targets of Eda/NF-κB [24 , 34] . All three were also detectable in the supernumerary primordia of K14-Eda embryos , Tbx3 showing a circular expression pattern around the placodes though . Expression of all marker genes was completely abolished in the supernumerary placode forming region in the compound mutants ( Fig 5A–5C ) . Analysis with scanning electron microscope ( SEM ) showed no morphological signs of supernumerary placodes in K14-Eda;IκBαΔN mutants ( Fig 5D ) . Further , supernumerary nipples or ductal trees were never observed in the adult compound mutants . Our findings show that even though NF-κB is not needed for the formation of endogenous mammary placodes , it is indispensable for formation of Eda-induced supernumerary mammary placodes . In order to identify the immediate downstream targets of Eda/NF-κB , we performed microarray profiling of genes expressed in Eda-/- E13 . 5 mammary buds exposed to control medium or to recombinant Fc-Eda protein . Using the same setup , but quantitative real-time reverse-transcriptase–PCR ( qRT-PCR ) and candidate gene approach , we have previously shown that Eda upregulates expression of Wnt10a , Wnt10b , Dkk4 , and PTHrP in mammary buds [24] . Altogether 245 probes were upregulated ( including Wnt10a , Wnt10b , Dkk4 , and PTHrP ) and 78 probes downregulated by Eda treatment ( Tables 1 and S1 ) . Genes in several different signaling pathways including Wnt , Fgf , Tnf , Tgfβ , chemokine , and hedgehog pathways were differently expressed . In addition , adhesion molecules Madcam1 and Icam1 , extracellular matrix degrading metalloproteinases Adamts15 and Mmp9 , chloride channel proteins clca1 and clca2 ( recently reannotated as a1 and a2 variants of clca3 , respectively ) , and transcription factor Foxi3 were among the upregulated genes ( Table 1 ) . To validate the microarray results , we performed qRT–PCR analysis and in situ hybridization ( ISH ) or immunostaining of selected genes , both strongly and modestly induced ones ( Table 1 , Figs 6A , 6B and S2 ) . Of the 7 genes tested all showed the same tendency as in the microarray , the difference between control and Eda-treated specimen being statistically significant for 5 genes . Both Madcam1 and Icam1 are known to be expressed in hair placodes and their transcripts are upregulated by Eda in E14 back skin [35] . We did not detect Madcam1protein or Icam1 , Adamts15 , or Mmp9 transcripts in developing mammary primordia of control embryos by whole mount analysis , yet Madcam1 and Mmp9 ( but not Icam1 or Adamts15 ) were readily observed in K14-Eda embryos ( S2A and S2B Fig ) suggesting that they lie downstream of Eda in mammary buds . Mmp9-deficient mice have no overt mammary gland phenotype [36] possibly owing to redundancy with other Mmps . There are no reports on the function of the other genes in mammary gland development . Transcription factor Foxi3 was one of the most highly induced genes by Eda . Foxi3 is mutated in several dog breeds , a condition described as canine ectodermal dysplasia [37] . We have previously identified Foxi3 as an Eda-induced gene in developing hair follicles and teeth and shown augmented expression in K14-Eda mammary buds in vivo [38] . The finding prompted us to analyze whether Foxi3 could play a role in mammary gland induction . However , the mammary glands of Foxi3 null embryos were indistinguishable from control littermates and formation of Eda-induced supernumerary mammary primordia was unaffected by loss of Foxi3 ( S3A–S3D Fig ) . Our microarray and previous qRT-PCR analyses revealed that several Wnt pathway genes are induced by Eda ( Fig 5 , [24] ) . Further , we have earlier reported that Lef1 is expressed very early on in the emerging supernumerary placodes of K14-Eda embryos [27] . Given the importance of the Wnt pathway in mammary placode formation , we wanted to study more closely whether expression of the Wnt pathway genes is altered in response to diverse levels of Eda by comparing Eda-/- , wild type , and K14-Eda embryos at E12 . 5 , when ectopic placodes are becoming apparent between buds 3 and 4 . Wnt10b , one of the earliest markers of the milk line , becomes gradually restricted to the placodes as they emerge [2 , 3] . Kremen2 ( Krm2 ) is a transmembrane protein that inhibits Wnt signaling in the presence of Dkk proteins [39] whereas Lgr4 is a receptor for R-spondins , which are potent Wnt pathway stimulators [40 , 41] . Both Krm2 and Lgr4 have been localized to E12 . 5 mammary buds [3 , 42] . Wnt10a , Wnt10b , Krm2 and Lgr4 were all present in the endogenous mammary buds of all three genotypes ( Fig 7A–7D ) . Expression of all four genes revealed a correlation with Eda levels: reduction in Eda-/- and up-regulation in K14-Eda mammary buds . Notably , Wnt10b and occasionally Lgr4 and Kremen2 were clearly upregulated as a continuous streak in K14-Eda embryos at the site where supernumerary placodes form . Further , we analyzed expression of β-catenin , which also exhibited a streak-like expression pattern between buds 3 and 4 in K14-Eda embryos ( Fig 7E ) . Next , we studied the expression of other genes critical for mammary placode formation . Tbx3 and Nrg3 are first detected in the mesenchyme but at the onset of placode formation they become upregulated ( Tbx3 ) or completely restricted ( Nrg3 ) to the mammary epithelium [12 , 15 , 33] . Expression of both genes was expectedly found in the endogenous mammary buds in all three genotypes ( Fig 7F and 7G ) . However , neither of them could be detected in the ectopic mammary forming region in K14-Eda embryos at E12 . 5 ( Fig 7F and 7G ) , yet Tbx3 was observed in the ectopic primordia at E13 . 5 ( [24]; Fig 5 ) . Our results show that all Wnt pathway genes studied exhibited early upregulation in the region where ectopic mammary placodes arise , whereas expression of other genes implicated in mammary placode formation ( Tbx3 , Nrg3 ) was detectable in this region only at a later developmental stage . Although different probes cannot be directly compared with each other , these data might suggest that especially Wnt pathway activation is critical for induction of ectopic placodes downstream of Eda . Further , our microarray associated Eda with several Wnt pathway genes , but revealed no link between Eda and Tbx3 or Nrg3 . Two Fgf ligands ( Fgf17 and Fgf20 ) were upregulated by Eda , but these Fgfs are thought to signal mainly via the mesenchymally expressed c isoforms of Fgfrs , not Fgfr2b [43] , and are thus unlikely to function in a manner similar to Fgf10 . In order to be able to manipulate and follow mammary placode formation more precisely , we developed an ex vivo tissue culture setup . In brief , ventrolateral skin explants containing the milk line region were dissected from E12 . 5 embryos and grown in a Trowell-type culture system as described previously [44] . Explants isolated from K14-Eda and control littermate embryos were cultured for a period of two days . After one day ( E12 . 5+1d ) , wild type and K14-Eda samples appeared almost identical ( Fig 8A and 8B ) . By E12 . 5+2d , K14-Eda explants were clearly distinguishable from controls due to the presence of supernumerary bud-like structures that had formed between buds 3 and 4 , and occasionally also between buds 2 and 3 ( Fig 8A and 8B ) . Typically , 2–3 supernumerary primordia formed between buds 3 and 4 . The explants thus recapitulated the in vivo phenotype very closely [27] . Next , we tested whether recombinant Fc-Eda protein had the capacity to induce formation of ectopic buds ex vivo . After one day ( E12 . 5+1d ) , control and Eda-treated samples appeared fairly similar ( Fig 8C and 8D ) although incipient supernumerary placodes were observed in Eda-treated specimen . A day later , similar to K14-Eda explants , several ectopic bud-like structures had developed within the milk line in Eda-treated specimen , whereas the controls showed no morphological changes in this region ( Fig 8C and 8D ) . Increased NF-κB reporter activity was evident in response to Eda treatment at the sites of presumptive supernumerary placodes ( Fig 8E and 8F ) . The endogenous mammary buds of both control and Eda-treated explants expressed Wnt10b , Krm2 and PTHrP . As in vivo , expression of these genes was observed in Eda treated samples between buds 3 and 4 ( S4A–S4C Fig ) . Sonic hedgehog ( Shh ) is a hair lacode-specific marker whose expression is barely detectably in mammary buds [45] . No Shh expression was observed in endogenous buds or in the region where supernumerary mammary primordia formed in control or Eda-treated samples ( S4D Fig ) . Upregulation of several Wnt pathway genes by Eda suggests involvement of canonical Wnt signaling in the induction of supernumerary placodes . However , the effects of Wnt pathway would be difficult to assess genetically due to several putative target genes of Eda that could act redundantly . Instead , we cultured E12 . 5 wild type and K14-Eda explants in the presence of XAV939 , an inhibitor of the canonical Wnt-pathway [46] . Application of XAV939 on tissues of TOP-gal Wnt reporter embryos confirmed significant downregulation of Wnt signaling in all treated explants ( 17/17 explants ) ( Fig 9A ) . Supernumerary placodes were always observed in non-treated K14-Eda samples at E12 . 5 + 2d whereas their formation was greatly reduced by low ( 10 μM ) and almost completely inhibited by high ( 40 μM ) concentration of XAV939 , respectively ( Fig 9B–9F ) . At these concentrations , XAV939 had no apparent effect on endogenous buds in wild-type or K14-Eda explants . We also tested the effect of XAV939 on endogenous placodes at the time when they emerge ( E11 . 0 ) and visualized forming mammary primordia with the aid of K17-GFP transgene [47] . 40 μM of XAV939 did not prevent formation of endogenous placodes , although placode size was clearly reduced ( S5 Fig ) indicating that supernumerary placodes are more sensitive to Wnt inhibition than endogenous ones . In conclusion , these data suggest that Eda signaling upregulates Wnt activity within the milk line which leads to formation of ectopic mammary placodes .
We report here that mice overexpressing the Tnf-like ligand Eda develop supernumerary mammary glands not only along the milk line [26 , 27] , but additionally in the neck region , anterior to mammary gland 1 . Further , based on pregnancy-associated morphological changes , these glands are functional . Traditionally the region possessing mammary potential has been thought to be limited to the area between the axilla and the genital tubercle [1] . The murine mammary line has been identified by a streak of Wnt10b expressing cells [2] . Initially three separate streaks form: a central streak between the limbs appears first followed by independent stripes at the ventral border of each of the limbs where placodes 1 and 5 form [2] . We also observed similar stripes of NF-κB reporter expressing cells in the axilla and groin , and as reported for Wnt10b [2] , they only later became connected with the central streak of the milk line . In addition to the mammary line , a separate streak of Wnt10b-positive cells , named the dorsal line , has been noted but the importance of these cells has remained elusive [2] . This streak is located dorsally to the milk line , encircles the fore limb bud from the dorsal side and ends at the anterior edge of the fore limb bud . Intriguingly , this is exactly where ectopic placodes form in K14-Eda embryos . We observed a discernible cluster of Wnt10b-positive and Dkk4-positive cells , and a less pronounced aggregate of NF-κB reporter expressing cells , in this location also in control embryos . Inspection of published pictures reveals that this domain is also positive for several other placode markers including Wnt6 , Tbx3 , and s-SHIP-GFP [2 , 33 , 48] . Further , it is characterized by high TOP-Gal activity [3 , 49] . Collectively , these data and our new findings indicate that the murine milk line extends past the axillary area . Our data suggest that Eda induced mammary cell fate supporting signals result in the maintenance of a normally transient group of mammary cells at the far end of the dorsal line . Similarly , Eda has been proposed to sustain a transient signaling center in the dental lamina resulting in the formation of an ectopic tooth in K14-Eda mice [27] . On the other hand , in wild-type embryos the area between buds 3 and 4 , another region where supernumerary mammary glands develop in K14-Eda mice , is characterized by scattered rather than clustered Wnt10b-positive cells . This may explain why Eda can readily overcome the developmental threshold for placode induction in the neck region leading to the appearance of supernumerary placodes in this position substantially earlier than elsewhere in the milk line . At the time when supernumerary placodes arise between buds 3 and 4 , the streak of Wnt10b-positive cells in no longer detectable in control embryos , but it is maintained/reappears in K14-Eda embryos . However , similar to the endogenous buds , Wnt10b expression becomes later confined to the newly formed buds . This suggests that similar mechanisms account for the formation of both endogenous and supernumerary mammary primordia . Our analysis on Eda-/- embryos showed that lack of Eda does not interfere with the patterning of endogenous mammary placodes . NF-κB is thought to be activated by all Tnf receptors , but also JNK and p38 pathways can be employed by many Tnfrs [50] . JNK pathway has been suggested to mediate Edar signaling , at least in some cultured cell lines [51] . Though NF-κB can be activated by multiple stimuli we found no evidence that other NF-κB activating cues besides Eda operate during mammary placode formation . As NF-κB was shown to be dispensable for mammary gland induction , it was surprising that formation of Eda-induced supernumerary placodes was NF-κB -dependent . We find it plausible that Eda/NF-κB has a role in the formation of endogenous mammary placodes as well , but loss of function may be compensated for by other pathways , in particular those that enhance Wnt signaling activity ( see also below ) . Nrg3/ErbB4 is one potential pathway that could exert this function as Nrg3-coated beads induce Lef1 expression concomitant with ectopic placodes-like structures in vitro [15] . Our microarray profiling of genes differentially expressed in mammary buds upon short exposure to recombinant Eda protein revealed members of several signaling pathways . This implies that Eda may act by tinkering the activity of multiple mammary-associated pathways during morphogenesis . We found significant changes in expression of Wnt , Fgf , Tnf , and chemokine pathway genes , similar to our previous findings in Eda-regulated genes in hair placodes [35 , 52] . Although these two studies cannot be directly compared due to different microarray platforms used , it seems that the gene regulatory network governed by Eda is largely shared between hair follicles and mammary glands . We found changes in several Wnt pathway genes upon Eda treatment; both agonists ( Wnt10a , Wnt10b , Lef1 , and Lgr4 ) and antagonists ( Lrp4 , Kremen2 and Dkk4 ) were upregulated , as also observed in hair placodes [24 , 34 , 52] . Proper spacing of many ectodermal appendages is believed to be achieved by combinatory regulation of positive and negative signals [53 , 54] . The reaction-diffusion model suggests that soluble factors that either promote or inhibit placode fate are co-expressed in placodes . However , unequal diffusion/stability of these substances may result in higher activator activity in the placodes whereas in the surrounding tissue , the opposite , higher inhibitor to activator ratio , prevents acquisition of placode fate . Further , it seems plausible that these cues are fine-tuned by several different pathways . The ability of Eda to modulate the expression of both placode activators and inhibitors in combination with the input from other signaling pathways may also explain the puzzling finding that HED patients may have both missing and supernumerary nipples [21 , 23] . We propose that maintenance and/or enhancement of Wnt pathway activity is the critical molecular mechanism whereby Eda induces the formation of mammary placodes . Our conclusion is based on the following findings: 1 ) Canonical Wnt signaling is absolutely necessary for mammary placode induction and genetic deletion of Wnt pathway antagonists ( Lrp4 , Sostdc1 ) causes more epidermal cells to adopt mammary cell fate along the mammary line [3 , 31 , 49]; 2 ) Several Wnt pathway genes show an early upregulation at the prospective site of ectopic mammary placodes in K14-Eda embryos; and 3 ) Pharmacological inhibition of Wnt signaling suppresses formation of supernumerary placodes in K14-Eda mammary explants in a dose-dependent manner at doses that however , do not yet prevent formation of endogenous placodes . Eda and Wnt signaling pathways are intertwined during development of several ectodermal organs [24 , 34 , 52 , 55 , 56] . In primary hair placodes , Wnt/β-cat signaling enhances Edar expression which in turn is required for upregulation of Wnt10a/b to levels high enough for placode morphogenesis to proceed . In the absence of Eda , placode formation is halted at a rudimentary ‘pre-placode’ stage characterized by severely reduced levels of Wnt activation [19 , 34 , 52 , 57] . However , mammary placodes are largely insensitive to loss of Eda . Interestingly , Lgr4 deficient embryos display a similar defect in primary hair placodes as Eda null embryos [58] but all mammary glands form in Lgr4 deficient mice [59] . Collectively , these data indicate that in the absence of Eda , cues other than Eda/NF-κB are responsible for maintenance of Wnt10a/b/Lgr4 expression and thereby sufficient Wnt signaling activity to support early mammary morphogenesis . Alternatively , other Wnt ligands/Lgrs that are insensitive to Eda levels may have a more critical role in mammary primordia than in hair placodes . The number of mammary glands is usually considered to be a species-typical invariant trait [60–62] indicating that the balance between mammary fate promoting signals must be tightly balanced with inhibitory cues to ensure the development of the correct number of mammary glands . Yet , in some species such as the pig , dairy cattle , and multimammate mice ( a . k . a . African soft-furred rat ) , Mastomys natalensis and its closely related species , a notable intraspecific variability has been observed [60 , 63 , 64] . Even in humans , accessory nipples and breast tissue are found at relatively high prevalence ( estimates ranging from 0 . 2% to 5 . 6% ) [65 , 66] . These findings show that the mammary line has the capacity to produce more than the species-typical number of organs . Misregulation of the Eda/Wnt pathway could offer an explanation for some sporadic cases of polythelia or absence of breast . The number and location of mammary glands vary widely between mammals [60 , 62 , 67] . In humans and other primates mammary glands are located at the thoracic region , in most ungulates at the inguinal region , in mice , cats and dogs at both regions , whereas pigs have them along the entire length of the milk line . Usually the number of pairs corresponds to the average number of offspring born at a time [61 , 67] . Highest mammary gland numbers are seen in some marsupials and domesticated pigs , whereas mice and rats have maximally 6 pairs [60–62] . Multimammate mice are a striking exception with 8 to 12 pairs , or even more , scattered throughout the mammary line [64 , 68] thereby greatly resembling K14-Eda mice . However , it is not known whether the milk line is expanded anteriorly as in K14-Eda mice . Changes in the Eda pathway activity have been linked to intraspecies evolutionary adaptations in the numbers of skin appendages in two species: the amount of armor plates in marine vs . freshwater threespine sticklebacks and the sweat gland density in modern human populations [69 , 70] . It is tempting to speculate that differential expression levels of the Eda pathway components account for some of the interspecific differences observed in the number and position of mammary glands .
The generation and genotyping of the following mouse strains have been described: K14-Eda [26] , IκBαΔN [19] , Eda null ( Tabby ) ( Jackson Laboratories; stock no . 000314 ) , TOP-gal ( Jackson laboratories; stock no . 004623 ) , K17-GFP , Foxi3-deficient , and NF-κB reporter mice [24 , 47 , 71 , 72] . K14-Eda , Foxi3-deficient , K17-GFP , and NF-κB rep mice were maintained on the C57Bl/6 background . IκBαΔN mice were bred in the C57BL/6 or a mixed C57BL/6 and FVB background . Eda null and TOP-Gal mice were on B6CBA and NMRI backgrounds , respectively . The appearance of a vaginal plug was considered the embryonic day ( E ) 0 . 5 . The age of the embryos were further staged according to the limb morphogenesis [73] and other external criteria . All mouse experiments were approved by the local ethics committee and National Animal Experiment Board of Finland under licenses KEK13-020 and ESAVI/2984-04 . 10 . 07–2014 . The mice were sacrificed with CO2 followed by cervical dislocation . Embryos or dissected tissues were fixed overnight in 4% PFA at 4°C , processed through rising ethanol series and xylene into paraffin and sectioned at 5 μm . Whole mount X-gal staining was done according to a published protocol [74] . The samples were postfixed with 4% PFA . When sectioned , the counterstain was performed with Nuclear fast red . Processing of the mammary glands and the Carmine alum staining was performed as previously described [24] . Whole embryos and tissues were photographed using the Olympus SZX9 stereomicroscope and slides with the Zeiss Imager . M2 . For the keratin2e immunostaining , sections were deparaffinised and citrate-treated in 6mM sodium-citrate buffer ( pH 6 ) . The blocking was done with 5% goat serum in 3% BSA in PBS . The samples were incubated overnight with a primary mouse antibody against keratin2e ( 10R-C166a , 1:200 , Fitzgerald ) followed by a goat anti-mouse-HRP secondary antibody ( 1:500; Jackson Immuno Research ) . Detection was done with the Vectastain Elite ABC Kit ( Vector Laboratories ) and counterstain with haematoxylin . The whole mount immunostaining for Madcam1 was performed with a primary rat antibody against Madcam1 ( 550556 , 1:25; BD Pharmingen ) and a secondary anti-rat-HRP antibody ( 1:200 , Santa Cruz Biotechnology ) . The DAB substrate kit for peroxidase ( Vector Laboratories ) was used for detection . Unspecific staining was blocked with 1% dry milk in 1xPBS/0 . 1% Tween-20 . The ventrolateral skins that contained the mammary forming region and at least the endogenous buds 2 , 3 and 4 were dissected from E12 . 5 embryos and half embryo explants were prepared from E11 . 0 embryos as indicated in the text . The explants were cultured for 1 to 2 days in a Trowell-type culture setting [24 , 52] . The medium consisted of 1:1 mixture of DMEM and F12 ( Ham’s Nutrient Mix: Life Technologies ) and was supplemented with 10% ( vol/vol ) FCS ( PAA Laboratories ) , 2 mM l-glutamine , penicillin-streptomycin and ascorbic acid ( 75 mg/L ) . When indicated , recombinant Eda protein ( Fc-Eda-A1 ) [75] was added to the growth medium to achieve a final concentration of 250ng/mL . Wnt inhibitor XAV939 in DMSO ( Stemgent ) was used as 10 μM or 40μM concentrations . Two separate stock solutions were generated for the inhibitor in order to avoid DMSO concentrations higher than 0 . 25% in the culture medium . Each time , one side of the embryo was used as the control and the other was treated with XAV939 . The embryos or tissue culture samples were fixed overnight in 4% PFA at 4°C and processed for whole mount in situ hybridization or for paraffin-embedding . The whole mount in situ hybridization was performed with inSituPro robot ( Intavis AG ) . The following digoxigenin-labelled RNA probes were used: PTHrP [76] , Wnt10b [77] , Wnt10a [78] , Lef1 , β-catenin , Shh [79] , Tbx3 , Nrg3 [15] , Kremen2 ( nucleotides 1306–1703 of NM_028416 . 2 ) , Mmp9 ( nucleotides 527–1131 of NM_013599 . 3 ) and Lgr4 ( nucleotides 3408–3823 of NM_172671 . 2 ) . The detection was achieved by using BM Purple AP substrate Precipitating Solution ( Boehringer Mannheim ) . Radioactive in situ hybridization on paraffin sections was carried out according to previously described protocols using 35S-UTP labelled ( Amersham ) probe specific to Edar [80] . The hanging drop culture has been described in detail elsewhere [24 , 52] . In short , two pools of 15–20 E13 . 5 Eda-/- mammary buds from 4 or 5 embryos were collected for each sample pair: one pool was treated with 250 ng/mL of Fc-Eda for 4h , whereas the other one was maintained in a control medium for 4h . RNA extraction and cDNA synthesis was performed as described previously [24 , 52] . qRT-PCR was done in a LightCycler 480 ( Roche , Indianapolis , IA ) and the following analysis was done with software provided by the manufacturer . The expression data were normalized against Ranbp1 gene . For primer sequences see ( S2 Table ) . Unpaired Student’s t- test was used for statistical analysis of all data . P-values of ≤0 . 05 were considered to be statistically significant . E13 . 5 mammary buds were dissected from Eda-/- embryos and used either as a control or exposed to 250 ng/mL of Fc-Eda as described above . 15–20 mammary buds were pooled in one sample , and three biological replicates were collected . RNA was extracted as previously described [24 , 28] and RNA quality was monitored using a 2100 Bioanalyzer ( Agilent Technologies ) . RNAs were processed and hybridized on Affymetrix Mouse Exon 1 . 0 ST arrays ( Santa Clara , CA ) in the Biomedicum Functional Genomics unit ( University of Helsinki , Finland ) . Significance analysis between treated and control samples was done using three statistical tests . In each test , a paired t-test ( pairing was done over treatment-control pairs ) was applied to the data . Differentially expressed genes were detected using Limma , IBMT ( intensity based moderated t-test ) , and Cyber-T . All methods were applied with default parameters . Obtained p-values were adjusted for multiple testing using Storey’s q-value method . The data discussed in this publication have been deposited in NCBI's Gene Expression Omnibus [81] and are accessible through GEO Series accession number GSE69781 ( http://www . ncbi . nlm . nih . gov/geo/query/acc . cgi ? acc=GSE69781 ) | Mammary glands are the most characteristic feature of all mammals . The successful growth and function of the mammary glands is vital for the survival of offspring since the secreted milk is the main nutritional source of a new-born . Ectodysplasin ( Eda ) is a signaling molecule that regulates the formation of skin appendages such as hair , teeth , feathers , scales , and several glands in all vertebrates studied so far . In humans , mutations in the EDA gene cause a congenital disorder characterized by sparse hair , missing teeth , and defects in exocrine glands including the breast . We have previously shown that excess Eda induces formation of supernumerary mammary glands in mice . Here , we show that Eda leads to extra mammary gland formation also in the neck , a region previously not thought to harbor capacity to support mammary development . Using Eda loss- and gain-of-function mouse models and transcriptional profiling we identify the downstream mediators of Eda . The presence of extra nipples is a fairly common developmental abnormality in humans . We suggest that misregulation of Eda or its effectors might account for some of these malformations . Further , the number and location of the mammary glands vary widely between different species . Tinkering with the Eda pathway activity could provide an evolutionary means to modulate the number of mammary glands . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2015 | Ectodysplasin/NF-κB Promotes Mammary Cell Fate via Wnt/β-catenin Pathway |
Genome-wide association studies ( GWAS ) have identified 14 tagging single nucleotide polymorphisms ( tagSNPs ) that are associated with the risk of colorectal cancer ( CRC ) , and several of these tagSNPs are near bone morphogenetic protein ( BMP ) pathway loci . The penalty of multiple testing implicit in GWAS increases the attraction of complementary approaches for disease gene discovery , including candidate gene- or pathway-based analyses . The strongest candidate loci for additional predisposition SNPs are arguably those already known both to have functional relevance and to be involved in disease risk . To investigate this proposition , we searched for novel CRC susceptibility variants close to the BMP pathway genes GREM1 ( 15q13 . 3 ) , BMP4 ( 14q22 . 2 ) , and BMP2 ( 20p12 . 3 ) using sample sets totalling 24 , 910 CRC cases and 26 , 275 controls . We identified new , independent CRC predisposition SNPs close to BMP4 ( rs1957636 , P = 3 . 93×10−10 ) and BMP2 ( rs4813802 , P = 4 . 65×10−11 ) . Near GREM1 , we found using fine-mapping that the previously-identified association between tagSNP rs4779584 and CRC actually resulted from two independent signals represented by rs16969681 ( P = 5 . 33×10−8 ) and rs11632715 ( P = 2 . 30×10−10 ) . As low-penetrance predisposition variants become harder to identify—owing to small effect sizes and/or low risk allele frequencies—approaches based on informed candidate gene selection may become increasingly attractive . Our data emphasise that genetic fine-mapping studies can deconvolute associations that have arisen owing to independent correlation of a tagSNP with more than one functional SNP , thus explaining some of the apparently missing heritability of common diseases .
Genome-wide association ( GWA ) studies of colorectal cancer ( CRC ) have so far identified 14 common , low-risk susceptibility variants [1] . Of these 14 variants , 3 are close to loci that are secreted members of the bone morphogenetic protein ( BMP ) signalling pathway: GREM1 ( rs4779584 ) ; BMP4 ( rs4444235 ) ; and BMP2 ( rs961253 ) . In the colon , GREM1 is one of several BMP antagonists produced by sub-epithelial myofibroblasts ( ISEMFs ) . GREM1 binds to and inactivates the ligands BMP2 and BMP4 that are primarily produced by inter-cryptal stromal cells . Our GWA studies have utilised a primary phase of genome-wide typing of tagging single nucleotide polymorphisms ( tagSNPs ) , followed by larger validation phases of those SNPs with the strongest signals of association . We have previously used relatively stringent statistical thresholds to take SNPs forward into the final validation phases [1] . Whilst such a design has been cost-effective , the use of a lower threshold may have led to the discovery of more CRC SNPs , albeit at the cost of a relatively high type I error rate . One means of reducing false positives might be to select SNPs using a less stringent threshold where there is a priori evidence for candidacy . We reasoned that the best candidate loci were those already identified as harbouring CRC risk alleles . Of those 14 loci , we prioritised GREM1 , BMP2 and BMP4 for further analysis owing to their strongly-related functions . The GWA studies had identified a single tagSNP associated with CRC risk close to each of GREM1 , BMP2 and BMP4 [1] . Examination of the regions around these genes in public databases such as HapMap ( http://www . hapmap . org/ ) showed in all cases that the coding sequence and predicted surrounding regulatory regions were present within more than one linkage disequilibrium ( LD ) block . For each of the 3 genes , therefore , it was possible that there were additional genetic determinants of CRC risk , independent of the already-identified SNPs . We proceeded to test this hypothesis in large sets of CRC cases and controls of European origin .
In order to refine the location of CRC-associated functional variation close to the GREM1 , BMP4 and BMP2 loci , we genotyped 442 SNPs close to rs4779584 , rs4444235 and rs961253 in 4 , 878 CRC cases and 4 , 914 controls from the UK2 and Scotland2 sample sets and imputed other SNPs within these regions . No significant localisation of a functional variant was achieved for rs4444235 or rs961253 ( Figure S1 ) , but at GREM1 , rs16969681 ( chr15:30 , 780 , 403 bases ) had a notably stronger signal of association than rs4779584 ( pairwise LD: r2 = 0 . 18 , D′ = 0 . 70 ) ( Figure 1 and Figure S2 ) . We genotyped rs16969681 in additional independent CRC case-control series ( UK1 , UK4 , VQ58 , Helsinki , Cambridge and EPICOLON; see Methods ) . After combined analysis , a significant association between the minor allele at rs16969681 and CRC risk was seen ( P = 5 . 33×10−8; Table 1 ) . Unconditional logistic regression analysis , incorporating sample series as a co-variate , showed that rs16969681 was more strongly associated with CRC than rs4779584 , but that the signals were non-independent ( for rs16969681 , OR = 1 . 16 , 95% CI 1 . 07–1 . 25 , P = 1 . 91×10−4; for rs4779584 , OR = 1 . 08 , 95% CI 1 . 02–1 . 14 , P = 5 . 27×10−3 ) . Akaike information criteria metrics for rs16969681 and rs4779584 respectively were 25608 and 25922 , consistent with a superior fit of the risk model incorporating the former SNP . Intriguingly , we found that rs16969681 maps to a site of open chromatin in GREM1-expressing CRC cell lines , raising the possibility that it may be directly functional ( Figure S3 ) . Haplotype risk analysis ( Table S2 ) provided evidence that rs16969681 alleles do not capture all the CRC risk associated with rs4779584 . In brief , data from UK2 and Scotland2 showed that the risk alleles at rs16969681 and rs4779584 were defined by a TGGTC haplotype at rs16969681-rs16969862-rs12594722-rs4779584-rs9888701 . The TT rs16969681-rs4779584 haplotype was at a frequency of 0 . 063 in cases and 0 . 052 in controls ( P = 6 . 29×10−5 ) . However , there appeared to be a residual effect of the T allele at rs4779584 , since there was also an elevated risk associated with the CT rs16969681-rs4779584 haplotype ( P = 0 . 026 ) . We therefore tested the hypothesis that rs4779584 tags two independent risk SNPs at GREM1 . We used reverse stepwise logistic regression to search the set of GREM1 SNPs genotyped in the UK2 and Scotland2 samples ( Table S1 ) for associations that were independent of rs16969681 genotype and that captured the residual rs4779584 signal . This analysis led to elimination of rs4779584 from the regression model and identification of a model in which only rs16969681 ( P = 1 . 04×10−4 ) and another SNP , rs11632715 ( P = 1 . 00×10−3 ) , produced independent signals . rs11632715 ( chr15:30 , 791 , 539 ) is in low LD with rs16969681 ( r2 = 0 . 009 , D′ = 0 . 31 ) and modest LD with rs4779584 ( r2 = 0 . 18 , D′ = 0 . 90; Figure S2 ) . Through genotyping of additional case-control series , we showed that rs11632715 was significantly associated with CRC risk ( P = 2 . 30×10−10; Table 1 ) . Unconditional logistic regression in the 21 , 139 samples typed for both rs11632715 and rs16969681 provided confirmatory evidence of the independence of the signals ( for rs16969681 , P = 1 . 84×10−6 and for rs11632715 , P = 6 . 36×10−7 ) ; these associations were of very similar magnitude to those obtained when each SNP was analysed individually in those sample sets ( Figure S2 ) . Incorporation of rs4779584 into the logistic regression model showed that this SNP had a weaker effect than that of either rs16969681 or rs11632715 and did not significantly improve the model fit ( Table S3 ) . Inspection of the region containing rs4779584 , rs16969681 and rs11632715 ( Figure S2 ) showed that rs4779584 lay within a recombination hotspot . This finding was consistent with our discovery that rs4779584 tags two independent functional variants that are , in turn , tagged by rs16969681 and rs11632715 . The regions analysed for fine mapping encompassed only a minority of the transcribed and flanking regions of GREM1 , BMP4 and BMP2 . We therefore tested for further independent CRC-associated SNPs around these loci ( Table S4 ) by undertaking a pooled analysis of data from 5 CRC GWA studies ( UK1 , Scotland1 , VQ58 , CCFR , Australia ) and from the UK2 and Scotland2 samples that had been genotyped at 55 , 000 SNPs with the strongest evidence of association from meta-analysis of UK1 and Scotland1 ( Figure S4 ) [1] . Since each of the 7 sample sets had been genotyped using different , but overlapping , SNP panels , we performed the combined analysis irrespective of the number of studies in which any SNP had been typed . Figure 2 shows the resulting signals of association from single SNP analysis in this discovery phase . We prioritised SNPs for further assessment in the replication data sets if they passed two thresholds . First , we required SNPs to show association with CRC at P<1×10−4 under the allelic or Cochran-Armitage tests; this was a less stringent threshold than that used in our previously-reported hypothesis-free GWA studies [1] , [2] , [3] , reflecting the fact that GREM1 , BMP4 and BMP2 were strong candidate susceptibility genes . Four SNPs at BMP4 , 3 at BMP2 and 9 at GREM1 fulfilled this criterion ( Figure 2 ) . Second , since our aim was to test for novel , independent disease variants rather than to refine existing signals of association , we required that SNP genotypes were not correlated with each other or with previously identified risk SNPs ( r2<0 . 05 , D′<0 . 10 ) . After applying these criteria , one SNP at BMP4 ( rs1957636 ) and one at BMP2 ( rs4813802 ) were retained for subsequent analyses . rs1957636 and rs4813802 were then genotyped in the validation sample sets ( Figure S4 ) , comprising 15 , 075 CRC cases and 13 , 296 controls from six independent European case-control series ( COIN/NBS , UK3 , UK4 , Scotland3 , Cambridge , Helsinki ) . After combined analysis , significant associations ( Table 2 ) were shown for both rs1957636 , P = 1 . 36×10−9 ( OR = 1 . 08 , 95% CI: 1 . 06–1 . 011 , Phet = 0 . 009 , I2 = 54% ) and rs4813802 , P = 7 . 52×10−11 ( OR = 1 . 09 , 95% CI: 1 . 06–1 . 012 , Phet = 0 . 42 , I2 = 3% ) . In case-only analysis , neither SNP showed any evidence of association with age or sex ( P>0 . 05 , details not shown ) . We used unconditional logistic regression , adjusting for sample series , to test the independence of the two pairs of SNPs at BMP4 and at BMP2 . In both cases , each signal remained independent , reflecting the existence of recombination hotspots between the pairs of SNPs at each locus ( Figure S5 and Figure S6 ) . For rs4444235 and rs1957636 , association P-values were respectively 2 . 09×10−8 ( I2 = 47 . 7% ) and 3 . 93×10−10 ( I2 = 0% ) ) . For rs961253 and rs4813802 , P-values were 1 . 89×10−15 ( I2 = 5% ) and 4 . 65×10−11 ( I2 = 5% ) ) . Thus , all 4 SNPs represented independent signals of association with CRC . Further imputation around BMP4 and BMP2 provided no evidence for the alternative possibility that a single variant was tagged by the two SNPs in each region ( details not shown ) . rs1957636 ( chr14: 53 , 629 , 768 ) is 136 kb upstream of the transcriptional start site of BMP4 , 150 kb telomeric to the previously-identified CRC susceptibility SNP , rs4444235 ( chr14:53 , 480 , 669 ) , which is downstream of BMP4 . There is a recombination hotspot at chr14:53 , 510 , 000 ( Figure S5 ) and LD between rs1957636 and rs4444235 is very weak ( r2 = 0 . 004 , D′ = 0 . 073 from UK1 ) . rs1957636 is within a region of LD flanked by SNPs rs431669 ( chr14:53 , 512 , 418 ) and rs10150369 ( chr14:53 , 873 , 515 ) . This region contains no known transcripts , and the nearest gene apart from BMP4 is CDKN3 ( transcriptional start site , chr14:53 , 933 , 476 ) . Using SNAP ( http://www . broadinstitute . org/mpg/snap/ ) to search HapMap3 release 2 and 1000 Genomes Pilot 1 , we identified 265 SNPs were in moderate or greater LD ( r2>0 . 20 ) with rs1957636 in Europeans . Of those SNPs , several mapped to sites of potential functional importance in BMP4 transcription ( H3K4Me1 , H3K4Me3 , DNAseI hypersensitivity , transcription factor ChIP-Seq ) , as evidenced by the ENCODE regulation tracks ( http://genome . ucsc . edu/cgi-bin/hgTrackUi ? hgsid=171775907&c=chr14&g=wgEncodeReg ) of the UCSC Genome Browser . For example , rs12432287 ( r2 = 0 . 60 , D′ = 1 . 00 with rs1957636 ) and rs728425 ( r2 = 0 . 69 , D′ = 1 . 00 ) lie within a region of apparently high transcriptional regulatory activity at chr14:53 , 642 , 340–53 , 652 , 937 . Another SNP , rs8011813 ( r2 = 0 . 822 , D′ = 0 . 811 ) , maps within a similar region at chr14:53 , 728 , 957–53 , 731 , 647 . Although none of the SNPs in the region around rs1957636 is the location of a reported eQTL ( http://eqtl . uchicago . edu/cgi-bin/gbrowse/eqtl/ ) , no studies relating transcription to SNP genotype have yet been undertaken in the colorectum . rs4813802 maps to chr20:6 , 647 , 595 , about 49 kb upstream of BMP2 and 295 kb telomeric of the previously-identified BMP2 CRC susceptibility SNP , rs961253 ( chr20:6 , 352 , 281 ) . There is very little LD between these two SNPs ( r2 = 0 . 000 , D′ = 0 . 017 from UK1 ) owing to a recombination hotspot at chr20:6 , 587 , 000 ( Figure S5 ) . rs4813802 lies within a region of LD flanked by rs727689 ( chr20:6 , 636 , 405 ) and rs6117401 ( chr20:6 , 664 , 097 ) . This region contains 3 spliced ESTs , BX107852 , BG822004 and DB094697; none of these has any known functional role or homology to other human or non-human transcripts or genes . The nearest gene to rs4813802 apart from BMP4 is FERMT1 ( transcriptional start site , chr20:6 , 052 , 191 ) . From HapMap3 release 2 and 1000 Genomes Pilot 1 , 29 SNPs were found to be in moderate or greater LD ( r2>0 . 20 ) with rs4813802 in Europeans . Of those SNPs , several in the region chr20:6 , 636 , 405–6 , 647 , 595 mapped to sites of potential functional importance in BMP2 transcription . None of the SNPs in the area around rs4813802 is the location of a reported eQTL . Using a case-control logistic regression design , we searched for pairwise gene-gene interactions between 5 SNPs associated with CRC risk ( rs4444235 , rs1957636 , rs961253 , rs4813802 and rs4779584 ) . Risks were additive and no evidence of epistasis was detected ( P>0 . 2 for all SNP pairs ) . We also searched for evidence of CRC susceptibility alleles at tagSNPs close to other BMP pathway genes . Using the transcribed regions of flanking genes as boundaries , we identified 4 , 361 tagSNPs mapping to 37 BMP agonist , antagonist and receptor loci ( Table S5 ) . However , we found no statistically significant evidence of associations with disease ( P>10−3 in all cases ) .
We have identified two new CRC predisposition tagSNPs close to BMP4 ( rs1957636 ) and BMP2 ( rs4813802 ) . To date , few other loci have been shown at stringent levels of significance to harbour more than one , independent cancer susceptibility variant . One notable exception is the locus proximal to MYC on chromosome 8q24 . 21 that contains multiple regions independently associated with the risk of prostate and other cancers [4] . Low-penetrance cancer predisposition loci are becoming increasingly hard to identify , owing to small effect sizes and/or low risk allele frequencies – and a return to candidate gene-based approaches may become increasingly attractive . It is true that in the past , candidate gene approaches have generally been unsuccessful at identifying cancer risk loci , but it is now possible to make use of information , such as expression quantitative trait locus identification , that increasingly permits a more considered approach . We have also found good evidence that the original CRC-associated SNP near GREM1 , rs4779584 [5] , tags two independent functional SNPs , represented by association signals at rs16969681 and rs11632715 . This finding emphasises that genetic fine-mapping studies are valuable not only for detecting stronger association signals , but also for deconvoluting tagSNP associations that have arisen owing to independent correlation of the tagSNP with more than one functional SNP . The original rs4779584 tagSNP signal could be described as an example of “synthetic association” , a term that has been used to describe a situation in which multiple , sometimes rare , variants underlie a tagSNP signal [6] , [7] . Synthetic association can explain some of the apparently missing heritability of complex diseases . Here , we estimate that the 6 SNPs close to the 3 BMP pathway genes contribute approximately 2% of the heritability of CRC , about double that estimated before this study . Finally , our data provide evidence that GREM1 , BMP4 and BMP2 are the targets of the functional variation in each region . Multiple , independently-acting variants close to these loci contribute to CRC risk . Perhaps unexpectedly , there are no detectable genetic interactions among these variants . If the downstream SMAD effectors that function within both the BMP and TGF-beta pathways are included , the components of BMP signalling involved in CRC risk might comprise up to 3 high-penetrance predisposition genes ( SMAD4 , BMPR1A , GREM1 ) and 8 low-penetrance variants at GREM1 , BMP4 , BMP2 , SMAD7 and LAMA5 ( tagged respectively by rs16969681 and rs11632715 , rs4444235 and rs1957636 , rs961253 and rs4813802 , rs4939827 , and rs4925386 ) [1] , [2] , [3] , [5] , [8] , [9] , [10] , [11] . Collectively these data emphasise the potential importance of genetic variants in the BMP pathway for CRC predisposition .
Collection of blood samples and clinico-pathological information from patients and controls was undertaken with informed consent and ethical review board approval in accordance with the tenets of the Declaration of Helsinki . The study had two main components: ( i ) refinement of existing GWAS signals at the GREM1 , BMP4 and BMP2 loci using a dense genotyping and imputation approach in several thousand cases and controls previously used for GWAS validation; and ( ii ) a search for new , independent CRC tagSNPs at the same three loci using a less stringent threshold for validation than used previously , combined with multiple validation sample sets . UK1 ( CORGI ) [1] comprised 922 cases with colorectal neoplasia ( 47% male ) ascertained through the Colorectal Tumour Gene Identification ( CoRGI ) consortium . All had at least one first-degree relative affected by CRC and one or more of the following phenotypes: CRC at age 75 or less; any colorectal adenoma ( CRAd ) at age 45 or less; ≥3 colorectal adenomas at age 75 or less; or a large ( >1 cm diameter ) or aggressive ( villous and/or severely dysplastic ) adenoma at age 75 or less . The 929 controls ( 45% males ) were spouses or partners unaffected by cancer and without a personal family history ( to 2nd degree relative level ) of colorectal neoplasia . Known dominant polyposis syndromes , HNPCC/Lynch syndrome or bi-allelic MYH mutation carriers were excluded . All cases and controls were of white UK ethnic origin . Scotland1 ( COGS ) [1] included 980 CRC cases ( 51% male; mean age at diagnosis 49 . 6 years , SD±6 . 1 ) and 1 , 002 cancer-free population controls ( 51% male; mean age 51 . 0 years; SD±5 . 9 ) . Cases were for early age at onset ( age ≤55 years ) . Known dominant polyposis syndromes , HNPCC/Lynch syndrome or bi-allelic MYH mutation carriers were excluded . Control subjects were sampled from the Scottish population NHS registers , matched by age ( ±5 years ) , gender and area of residence within Scotland . VQ58 comprised 1 , 832 CRC cases ( 1 , 099 males , mean age of diagnosis 62 . 5 years; SD±10 . 9 ) from the VICTOR [12] and QUASAR2 ( www . octo-oxford . org . uk/alltrials/trials/q2 . html ) trials . There were 2 , 720 population control genotypes ( 1 , 391 males , ) from the Wellcome Trust Case-Control Consortium 2 ( WTCCC2 ) 1958 birth cohort ( also known as the National Child Development Study ) , which included all births in England , Wales and Scotland during a single week in 1958 [13] . The Colon Cancer Family Registry ( CCFR ) data set [14] comprised 1 , 332 familial CRC cases and 1 , 084 controls Colon Cancer Family Registry ( Colon-CFR ) ( http://epi . grants . cancer . gov/CFR/about_colon . html ) . The cases were recently diagnosed CRC cases reported to population complete cancer registries in the USA ( Puget Sound , Washington State ) who were recruited by the Seattle Familial Colorectal Cancer Registry; in Canada ( Ontario ) who were recruited by the Ontario Familial Cancer Registry; and in Australia ( Melbourne , Victoria ) who were recruited by the Australasian Colorectal Cancer Family Study . Controls were population-based and for this analysis were restricted to those without a family history of colorectal cancer . The Australian study [15] comprised 591 patients treated for CRC at the Royal Melbourne , Western and St Francis Xavier Cabrini Hospitals in Melbourne from 1999 to 2009 . The 2 , 353 controls were derived from Queensland or Melbourne: for the former , the controls came from the Brisbane Twin Nevus Study [16]; for the latter , individuals were participants in the Genes in Myopia study [17] . There was no overlap between the CFR and Australian data sets . Owing to potential residual ethnic heterogeneity within the Melbourne population , for the Australian cohort only we performed an additional screen to minimise heterogeneity after performing principal components analysis ( PCA ) to remove individuals who clustered with non-CEU individuals ( see below ) . We achieved this by performing PCA on the Australian cases and controls without reference samples of known ancestry . We then paired each case with a control in a 1∶1 ratio based on a maximum separation of 0 . 050 using the first and second eigenvectors . All unpaired samples were excluded , leaving 441 cases and 441 controls in the study . The genomic inflation factor , λGC , was 1 . 02 after this filtering . UK2 ( NSCCG ) [1] consisted of 2 , 854 CRC cases ( 58% male , mean age at diagnosis 59 . 3 years; SD±8 . 7 ) ascertained through two ongoing initiatives at the Institute of Cancer Research/Royal Marsden Hospital NHS Trust ( RMHNHST ) from 1999 onwards - The National Study of Colorectal Cancer Genetics ( NSCCG ) and the Royal Marsden Hospital Trust/Institute of Cancer Research Family History and DNA Registry . The 2 , 822 controls ( 41% males; mean age 59 . 8 years; SD±10 . 8 ) were the spouses or unrelated friends of patients with malignancies . None had a personal history of malignancy at time of ascertainment . All cases and controls had self-reported European ancestry , and there were no obvious differences in the demography of cases and controls in terms of place of residence within the UK . Scotland2 ( SOCCS ) [1] comprised 2 , 024 CRC cases ( 61% male; mean age at diagnosis 65 . 8 years , SD±8 . 4 ) and 2 , 092 population controls ( 60% males; mean age 67 . 9 years , SD±9 . 0 ) ascertained in Scotland . Cases were taken from an independent , prospective , incident CRC case series and aged <80 years at diagnosis . Control subjects were population controls matched by age ( ±5 years ) , gender and area of residence within Scotland . UK3 ( NSCCG ) [1] comprised 7 , 912 CRC cases ( 65% male; mean age at diagnosis 59 years , SD±8 . 2 ) and 4 , 398 controls ( 40% male; mean age 62 years , SD±11 . 5 ) ascertained through NSCCG post-2005 . Scotland3 ( SOCCS ) [1] comprised 1 , 145 CRC cases ( 50% male; mean age at diagnosis 53 . 2 years , SD±15 . 4 ) and 2 , 203 cancer-free population controls ( 47% male; mean age 51 . 8 years , SD±11 . 5 ) . Controls were recruited as part of the Generation Scotland study . UK4 ( CORGI2BCD ) [1] consisted of 621 CRC cases ( 46% male; mean age at diagnosis 58 . 3 years; SD±14 . 1 ) and 1 , 121 cancer-free population or spouse controls ( 45% male; mean age 45 . 1 years , SD±15 . 9 ) . Cambridge/SEARCH consisted of 2 , 248 CRC cases ( 56% male; mean age at diagnosis 59 . 2 years , SD±8 . 1 ) and 2 , 209 controls ( 42% males; mean age 57 . 6 years; SD±15 . 1 . Samples were ascertained through the SEARCH ( Studies of Epidemiology and Risk Factors in Cancer Heredity , http://www . cancerhelp . org . uk/trials/a-study-looking-at-genetic-causes-of-cancer ) study based in Cambridge , UK . Recruitment started in 2000; initial patient contact was though the general practitioner . Control samples were collected post-2003 . Eligible individuals were sex- and frequency-matched in five-year age bands to cases . The COIN samples [18] were 2 , 151 cases derived from the COIN and COIN-B clinical trials of metastatic CRC . Median age was 63 years . COIN cases were compared against genotypes from 2 , 501 population controls ( 1 , 237 males , ) from the WTCCC2 National Blood Service ( NBS ) cohort ( 50% male; mean age at diagnosis 53 . 2 years , SD±15 . 4 ) . The Helsinki ( FCCPS ) study ( http://research . med . helsinki . fi/gsb/aaltonen/ ) comprised 988 cases from a population-based collection centred on south-eastern Finland and 864 population controls from the same collection . EPICOLON [19] included 1 , 410 cases matched with the same number of controls collected in a prospective fashion from centres in Spain . Exclusion criteria were Mendelian CRC syndromes and a personal history of inflammatory bowel disease . In all cases CRC was defined according to the ninth revision of the International Classification of Diseases ( ICD ) by codes 153–154 and all cases had pathologically proven adenocarcinomas . DNA was extracted from samples using conventional methods and quantified using PicoGreen ( Invitrogen ) . The VQ , UK1 , Scotland1 and Australia GWA cohorts were genotyped using Illumina Hap300 , Hap370 , or Hap550 arrays . 1958BC and NBS genotyping was performed as part of the WTCCC2 study on Hap1M arrays . The CCFR samples were genotyped using Illumina Hap1M or Hap1M-Duo arrays . In UK2 and Scotland2 , genotyping was conducted using custom Illumina Infinium arrays according to the manufacturer's protocols . Some COIN SNPs were typed on custom Illumina Goldengate arrays . To ensure quality of genotyping , a series of duplicate samples was genotyped , resulting in 99 . 9% concordant calls in all cases . Other genotyping was conducted using competitive allele-specific PCR KASPar chemistry ( KBiosciences Ltd , Hertfordshire , UK ) , Taqman ( Life Sciences , Carlsbad , California ) or MassARRAY ( Sequenom Inc . , San Diego , USA ) . All primers , probes and conditions used are available on request . Genotyping quality control was tested using duplicate DNA samples within studies and SNP assays , together with direct sequencing of subsets of samples to confirm genotyping accuracy . For all SNPs , >99% concordant results were obtained . We excluded SNPs from analysis if they failed one or more of the following thresholds: GenCall scores <0 . 25; overall call rates <95%; MAF<0 . 01; departure from Hardy-Weinberg equilibrium ( HWE ) in controls at P<10−4 or in cases at P<10−6; outlying in terms of signal intensity or X∶Y ratio; discordance between duplicate samples; and , for SNPs with evidence of association , poor clustering on inspection of X∶Y plots . We excluded individuals from analysis if they failed one or more of the following thresholds: duplication or cryptic relatedness to estimated identity by descent ( IBD ) >6 . 25%; overall successfully genotyped SNPs <95%; mismatch between predicted and reported gender; outliers in a plot of heterozygosity versus missingness; and evidence of non-white European ancestry by PCA-based analysis in comparison with HapMap samples ( http://hapmap . ncbi . nlm . nih . gov/ ) . We excluded 6 duplicate samples using PCA ( see below ) within the UK samples that had undergone analysis of over 200 SNPs ( UK1 , Scotland1 , UK2 , Scotland2 , VQ , 1958BC , NBS , COIN ) . We excluded duplicates from other UK cohorts on the basis of names ( or initials where release of names was not possible ) and dates of birth . No duplicates were found from the CCFR or Australian sample sets . To identify individuals who might have non-northern European ancestry , we merged our case and control data from all sample sets with the 60 European ( CEU ) , 60 Nigerian ( YRI ) , and 90 Japanese ( JPT ) and 90 Han Chinese ( CHB ) individuals from the International HapMap Project . For each pair of individuals , we calculated genome-wide identity-by-state distances based on markers shared between HapMap2 and our SNP panel , and used these as dissimilarity measures upon which to perform principal components analysis . Principal components analysis was performed using Eigenstrat/SmartPCA using CEU , YRI and HCB HapMap samples as reference . The first two principal components for each individual were plotted and any individual not present in the main CEU cluster ( that is , >5% of the PC distance from HapMap CEU cluster centroid ) was excluded from subsequent analyses . We had previously shown the adequacy of the case-control matching and possibility of differential genotyping of cases and controls using Q-Q plots of test statistics in STATA . The inflation factor λGC was calculated by dividing the mean of the lower 90% of the test statistics by the mean of the lower 90% of the expected values from a χ2 distribution with 1 d . f . Deviation of the genotype frequencies in the controls from those expected under HWE was assessed by χ2 test ( 1 d . f . ) , or Fisher's exact test where an expected cell count was <5 . Regions selected for fine mapping were: chr15:30 , 733 , 560–30 , 802 , 752; chr14:53 , 430 , 973n 53 , 530 , 761; and chr20:6 , 292 , 730–6 , 402 , 661 . These corresponded to the haplotype blocks and immediately flanking regions harbouring rs4779584 , rs4444235 , and rs961253 . To define these haplotype blocks and the recombination hotspots harbouring these CRC-associated SNPs , we used Haploview and SequenceLDHot . From dbSNP ( build 128 ) , we selected all SNPs between the recombination hotspots flanking the haplotype block . All these SNPs were submitted to Illumina for assay design and those with a design score>0 . 3 were genotyped on custom arrays in the UK2 and Scotland2 case-control series . In total , we genotyped 81 , 42 and 60 SNPs in the 15q13 . 3 , 14q22 . 2 and 20p12 . 3 regions respectively . A list of these SNPs is shown in Table S1 . Association statistics , using an additive model , were obtained with SNPTEST v2 ( www . stats . ox . ac . uk/~marchini/software/gwas/snptest . html ) . We used genotype data from the 1000 Genomes CEPH ( http://www . 1000genomes . org/ ) and HapMap3 CEPH and TSI samples ( www . hapmap . org/ ) and the IMPUTE v2 software ( https://mathgen . stats . ox . ac . uk/impute/impute_v2 . html ) to generate in silico genotypes at additional SNPs in all three regions . This imputation resulted in the addition of 74 , 113 and 255 markers in the chromosome 15q13 . 3 , 14q22 . 2 and 20p12 . 3 regions respectively ( for details on imputed and genotyped markers see Table S1 ) . Association meta-analyses only included markers with proper_info scores >0 . 5 , imputed call rates per SNP >0 . 9 and minor allele frequencies ( MAFs ) >0 . 01 . Meta-analyses of the two sample sets were carried out with Meta ( http://www . stats . ox . ac . uk/~jsliu/meta . html ) using the genotype probabilities from IMPUTE v2 , where a SNP was not directly typed . To test for the presence of additional independent risk alleles in each region , we carried out logistic regression analysis within each region , both pairwise with the original tagSNP and then in a backwards analysis that included all SNPs with evidence of association in the meta-analysis at P<5×10−4 . Association between SNP genotype and disease status was primarily assessed in STATA v10 ( http://www . stata . com/ ) and PLINK v1 . 07 ( http://pngu . mgh . harvard . edu/~purcell/plink/ ) using allelic and Cochran-Armitage tests ( both with 1df ) respectively , or by Fisher's exact test where an expected cell count was <5 . Genotypic ( 2df ) , dominant ( 1df ) and recessive ( 1df ) tests were also performed . The risks associated with each SNP were estimated by allelic , heterozygous and homozygous odds ratios ( ORs ) using unconditional logistic regression , and associated 95% confidence intervals ( CIs ) were calculated . Joint analysis of data generated from multiple phases was conducted using standard methods for combining raw data based on the Mantel-Haenszel method in STATA and PLINK . The reported meta-analysis statistics were derived from analysis of allele frequencies , and joint ORs and 95% CIs were calculated assuming fixed- and random-effects models . Tests of the significance of the pooled effect sizes were calculated using a standard normal distribution . Cochran's Q statistic to test for heterogeneity [20] and the I2 statistic [21] to quantify the proportion of the total variation due to heterogeneity were calculated . Large heterogeneity is typically defined as I2≥75% . Where significant heterogeneity was identified , results from the random effects model were reported . Alongside , we also performed meta-analysis based on allele dosage ( 0 , 1 , 2 ) and incorporated age and sex as co-variates . Although age and sex are associated with colorectal cancer risk , they were not associated with SNP genotype and did not materially affect the significance of any of the 6 reported associations ( details not shown ) . We used Haploview software v4 . 2 ( http://www . broadinstitute . org/haploview ) to infer the LD structure of the genome in the regions around GREM1 , BMP2 and BMP4 . The combined effects of pairs of loci identified as associated with CRC risk were investigated by multiple logistic regression analysis in PLINK to test for independent effects of each SNP and stratifying by sample series . Evidence for interactive effects between SNPs ( epistasis ) was assessed by likelihood ratio test assuming an allelic model in PLINK . The sibling relative risk attributable to a given SNP was calculated using the formulawhere p is the population frequency of the minor allele , q = 1−p , and r1 and r2 are the relative risks ( estimated as OR ) for heterozygotes and rare homozygotes , relative to common homozygotes [22] . Assuming a multiplicative interaction , the proportion of the familial risk attributable to a SNP was calculated as log ( λ* ) /log ( λ0 ) , where λ0 is the overall familial relative risk estimated from epidemiological studies of CRC , assumed to be 2 . 2 [23] . UK2/NSCCG2 samples were used for this estimation . The Akaike information criterion was calculated using the swaic command in STATA . Genome co-ordinates were taken from the NCBI build 36/hg18 ( dbSNP b126 ) . | Genome-wide association studies ( GWAS ) have identified several colorectal cancer ( CRC ) susceptibility polymorphisms near genes that encode proteins in the bone morphogenetic protein ( BMP ) pathway . However , most of the inherited susceptibility to CRC remains unexplained . We investigated three of the best candidate BMP genes ( GREM1 , BMP4 , and BMP2 ) for additional polymorphisms associated with CRC . By extensive validation of polymorphisms with only modest evidence of association in the initial phases of the GWAS , we identified new , independent CRC predisposition polymorphisms close to BMP4 ( rs1957636 ) and BMP2 ( rs4813802 ) . Near GREM1 , we used additional genotyping around the GWAS-identified polymorphism rs4779584 to demonstrate two independent signals represented by rs16969681 and rs11632715 . Common genes with modest effects on disease risk are becoming harder to identify , and approaches based on informed candidate gene selection may become increasingly attractive . In addition , genetic fine mapping around polymorphisms identified in GWAS can deconvolute associations which have arisen owing to two independent functional variants . These types of study can identify some of the apparently missing heritability of common disease . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"oncology",
"genome-wide",
"association",
"studies",
"medicine",
"genetic",
"causes",
"of",
"cancer",
"cancer",
"genetics",
"gastroenterology",
"and",
"hepatology",
"genetics",
"cancer",
"risk",
"factors",
"biology",
"gastrointestinal",
"cancers",
"genetics",
"of",
"disease",
"genetics",
"and",
"genomics"
] | 2011 | Multiple Common Susceptibility Variants near BMP Pathway Loci GREM1, BMP4, and BMP2 Explain Part of the Missing Heritability of Colorectal Cancer |
Escherichia coli ST131 is a globally disseminated , multidrug resistant clone responsible for a high proportion of urinary tract and bloodstream infections . The rapid emergence and successful spread of E . coli ST131 is strongly associated with antibiotic resistance; however , this phenotype alone is unlikely to explain its dominance amongst multidrug resistant uropathogens circulating worldwide in hospitals and the community . Thus , a greater understanding of the molecular mechanisms that underpin the fitness of E . coli ST131 is required . In this study , we employed hyper-saturated transposon mutagenesis in combination with multiplexed transposon directed insertion-site sequencing to define the essential genes required for in vitro growth and the serum resistome ( i . e . genes required for resistance to human serum ) of E . coli EC958 , a representative of the predominant E . coli ST131 clonal lineage . We identified 315 essential genes in E . coli EC958 , 231 ( 73% ) of which were also essential in E . coli K-12 . The serum resistome comprised 56 genes , the majority of which encode membrane proteins or factors involved in lipopolysaccharide ( LPS ) biosynthesis . Targeted mutagenesis confirmed a role in serum resistance for 46 ( 82% ) of these genes . The murein lipoprotein Lpp , along with two lipid A-core biosynthesis enzymes WaaP and WaaG , were most strongly associated with serum resistance . While LPS was the main resistance mechanism defined for E . coli EC958 in serum , the enterobacterial common antigen and colanic acid also impacted on this phenotype . Our analysis also identified a novel function for two genes , hyxA and hyxR , as minor regulators of O-antigen chain length . This study offers novel insight into the genetic make-up of E . coli ST131 , and provides a framework for future research on E . coli and other Gram-negative pathogens to define their essential gene repertoire and to dissect the molecular mechanisms that enable them to survive in the bloodstream and cause disease .
Escherichia coli O25b:H4-ST131 ( E . coli ST131 ) is a recently emerged , globally disseminated clone that is often multidrug resistant and is responsible for a high proportion of community- and nosocomially-acquired urinary tract and bloodstream infections [1]–[6] . E . coli ST131 strains are also capable of causing complicated infections including acute pyelonephritis , osteomyelitis , septic arthritis and septic shock [7] , [8] . E . coli ST131 are commonly associated with production of the CTX-M-15 enzyme , currently the most widespread extended spectrum β-lactamase ( ESBL ) of its type in the world [1] , [9] . In addition to resistance against oxyimino-cephalosporins ( i . e . cefotaxime , ceftazidime ) , and monobactams , E . coli ST131 strains are often co-resistant to fluoroquinolones [3] , [10] . Indeed , most fluoroquinolone-resistant E . coli strains belong to a recently emerged and dominant subgroup of ST131 strains [11] . Some E . coli ST131 strains have also been reported to produce carbapenemases [12]–[14] , thus severely limiting treatment options that are currently available against this clinically predominant clone [15] . E . coli ST131 strains , like many other uropathogenic E . coli ( UPEC ) strains , are derived from phylogenetic group B2 [3] . Typically , UPEC strains possess a large and diverse range of virulence factors that contribute to their ability to cause urinary tract and bloodstream infections , including adhesins , toxins , siderophores and protectins [15] , [16] . Several studies have demonstrated that E . coli ST131 strains possess a similar suite of virulence factors and cause invasive disease , leading to the hypothesis that the widespread pathogenic success of E . coli ST131 strains may be in part due to enhanced virulence [7] , [8] , [17] . However , it has become clear from recent studies that E . coli ST131 strains do not possess a heightened virulence potential compared to other UPEC or B2 E . coli strains in causing invasive infections [18] or infections in nematodes and zebrafish embryos [19] . Thus , other factors such as enhanced metabolic capacity have been proposed to contribute to the fitness and pathogenic success of this dominant clone [20] , [21] . The genome sequence of one of the best-characterized E . coli ST131 strains , EC958 , was recently described [22] . E . coli EC958 is a member of the pulsed-field gel electrophoresis ( PFGE ) defined UK epidemic strain A , which represents one of the major pathogenic lineages ( PFGE strains A–E ) of ESBL producing E . coli causing urinary tract infections ( UTI ) across the UK [23] . E . coli EC958 is resistant to eight antibiotic classes , including oxyimino-cephalosporins , fluoroquinolones and sulphonamides . E . coli EC958 colonizes the bladder of mice in a type 1 fimbriae-dependent manner [22] , can invade into bladder epithelial cells and form intracellular bacterial communities , can establish both acute and chronic UTI [24] and can inhibit the contraction of ureters , in vitro [25] . The ability to resist the bactericidal activity of serum , and thus survive in the bloodstream , represents an essential virulence trait for UPEC and other extra-intestinal E . coli ( ExPEC ) strains , including E . coli ST131 [26]–[28] . In E . coli , several mechanisms have been shown to contribute to serum resistance . The importance of O-antigens and K capsules in resistance to serum has been recognized since the 1960s and 1980s , respectively [29]–[31]; and their multiple types , combinations and length contribute differently to serum resistance [32]–[35] . The major outer membrane protein OmpA [36] , plasmid-encoded proteins TraT [37] , [38] and Iss [39] , and the phage membrane protein Bor [40] have also been reported to contribute to serum resistance in E . coli . Notably , each of these resistance mechanisms has been studied in isolation and in different strain backgrounds . Thus , while serum resistance is clearly a complex phenotype determined by multiple elements , little is known about the combination of factors that contribute to resistance in a single strain . High-throughput transposon mutagenesis combined with genome-wide targeted sequencing was used recently to study the essential genes in Salmonella enterica serovar Typhi and Caulobacter crescentus [41] , [42] . Langridge et al . also used their transposon directed insertion-site sequencing ( TraDIS ) method to assay every gene for its role in the survival of S . Typhi in the presence of bile salts [42] . Similar approaches ( INSeq , HITS , Tn-seq ) have also been applied to a range of organisms to study gene requirements for survival in particular niches [43]–[46] . Here , we adapted TraDIS and designed a multiplexing method to define the essential genes required for in vitro growth ( i . e . Luria-Bertani agar media supplemented with 30 µg/ml Cm at 37°C ) and the serum resistome ( i . e . genes required for resistance to human serum , in E . coli EC958 ) . We show that the essential gene list of E . coli EC958 comprises 315 genes , 231 of which are shared with E . coli K-12 . We also define for the first time a comprehensive inventory of genes required for resistance to human serum . Our study provides a molecular blueprint for understanding the mechanisms employed by E . coli ST131 to survive , grow in the bloodstream and cause disease .
Approximately 1 million mutants were generated in the E . coli ST131 strain EC958 [22] using an in-house miniTn5 transposon carrying a chloramphenicol ( Cm ) resistance gene derived from the pKD3 plasmid [47] . A primer comprising four functional regions was designed to facilitate specific sequencing of the transposon insertion sites on the Illumina HiSeq 2000 platform while allowing for intra-lane multiple sample indexing ( Figure S1 ) . This primer contained ( 5′-3′ ) : ( i ) the P5 sequence to bind to TruSeq flowcells , ( ii ) the Illumina read 1 sequencing primer binding site , ( iii ) a 6-bp index sequence for multiple sample barcoding and ( iv ) a 25-bp transposon specific sequence designed to amplify the last 12 bp of the transposon and its adjacent genomic sequence . Using this custom primer , we successfully sequenced the transposon insertion sites for 6 samples on both TruSeq version 2 and version 3 flowcells ( Figure 1A ) . Each sample yielded from 6 . 8 million to 15 million reads that were tagged with transposon specific sequence , 71% of which were reliably mapped to EC958 draft chromosome ( excluding unscaffolded contigs and plasmids ) ( Table 1 ) . All experiments were performed in duplicate , with the correlation coefficient for the number of insertions per gene for each pair of samples close to 1 ( R2>0 . 99 ) and thus demonstrating a high level of reproducibility for each experiment ( Figure 1B ) . We initially used our saturated random insertion mutant library to determine the ‘essential genes of EC958’ , defined as the set of genes required for growth on LB agar supplemented with Cm 30 µg/ml . We extracted genomic DNA ( in duplicates: input A and B ) directly from the library pool and sequenced using our multiplexed TraDIS protocol . We combined the reads from input A and B to maximize the coverage resulting in 16 million transposon-tagged reads , of which 11 million uniquely mapped to the EC958 chromosome , resulting in 502 , 068 unique insertion sites . This equates to an average of one insertion site every 9 . 92 bp , with a very low probability of having 100 consecutive bp without interruption by chance ( P = 4 . 2×10−5 ) . The essential gene list was identified using a statistical analysis similar to that described by Langridge et al . , which recognized two distinct distributions of insertion indexes ( number of insertions divided by gene length ) for non-essential genes ( gamma ) and essential genes ( exponential ) and called those with insertion indexes less than or equal to the intercept of the two distributions as essential [42] . In our data , an insertion index cut-off of 0 . 0158 , resulted in the identification of 315 genes as essential ( Table S1 ) . This cut-off is equivalent to a log2-likelihood-ratio ( LLR ) of −3 . 6 , which means that our essential genes are at least 12 times more likely to belong to the exponential distribution ( essential ) than the gamma ( non-essential ) distribution . The functional category of each gene was identified based on the COG ( Clusters of Orthologous Groups ) numbers from the EC958 annotation ( accession number PRJEA61443 ) . Figure 2 shows an overview of essential functions in EC958 compared with the total number of genes in each functional category . Genes involved in translation , ribosomal structure and biogenesis account for 25% of the total number of essential genes in EC958 , which is 42% of the total number of genes in this category . The second most abundant category in the essential gene list comprised genes involved in cell wall/membrane/envelope biogenesis ( 12% ) , followed by genes involved in coenzyme transport and metabolism . There were 23 essential genes with functions not identified in the COG database . To investigate the conservation of EC958 essential proteins among different E . coli pathotypes , we performed tfastx alignment ( FASTA v36 ) between the essential protein sequences and translated DNA from 50 E . coli complete genomes ( Table S2 ) . There were 270 ( 86% ) proteins conserved across all genomes investigated . An additional 17 proteins were also present in more than 90% of the genomes . Only 6 proteins were specific for EC958 ( not found in 50 genomes ) ( Table S3 ) . Saturated transposon mutagenesis in combination with next-generation sequencing is a powerful tool for whole genome , high-throughput identification of all candidate genes involved in a particular phenotype . Here , we used our transposon mutant library in combination with TraDIS to identify genes from EC958 involved in resistance to human serum , thus enabling us to define the serum resistome of EC958 . We designed a mutant selection procedure in which 1 million mutants were exposed to pooled fresh human serum for 90 minutes and then allowed to grow in LB broth for 4 hours before genomic DNA extraction . This procedure permitted the growth of serum resistant mutants while eliminating or inhibiting mutants that were sensitive to serum ( Figure 3A ) . The procedure was performed in parallel with control samples where fresh serum was replaced by inactivated serum that lacked bactericidal activity ( data not shown ) . The genomic DNA from test and control samples were sequenced using our modified Illumina multiplexed TraDIS procedure ( Figure 3B ) to generate multiple datasets ( Figure 1A ) that were analysed by the Bioconductor package edgeR after filtering out genes identified as essential [48] . The serum resistance genes were identified as genes that have significant reduction in read counts in the test samples compared to the control samples ( i . e . less mutants survived after serum treatment ) ( Table S4 ) . A stringent threshold of log2 fold change of read counts ( logFC ) less than −1 and an adjusted p-value less than 0 . 001 was used to identify significant genes that are involved in serum resistance ( Figure S2 ) . Figure 4 shows the names and genomic locations of the 56 genes that satisfied these stringent criteria . Twenty-two ( 39 . 3% ) of the genes belong to three operons responsible for LPS biosynthesis ( including both O-antigen biosynthesis and lipid A-core biosynthesis ) and enterobacterial common antigen ( ECA ) biosynthesis . This result represents the first layer of evidence demonstrating the importance of the O25 antigen as well as ECA in an E . coli ST131 background for protection from the bactericidal activity of human serum . Detailed characterisation of the O25 antigen gene cluster is discussed in subsequent sections . ECA is common to all Enterobacteriaceae and is expressed in both serum resistant ( smooth ) and sensitive ( rough ) strains , except for rough strains that are defective in the shared biosynthesis pathway affecting both O-antigen and ECA [49] , [50] . Seven out of 12 genes in the ECA operon were required for serum resistance as determined by the TraDIS technique . The remaining 60 . 7% of genes in the serum resistome identified by TraDIS included genes encoding lipoprotein , membrane proteins , regulators and hypothetical proteins ( Table 2 ) . Some of these also affect LPS such as rfaH ( EC958_4322 ) , encoding a known regulator required for LPS biosynthesis [51] , [52] and virulence of pathogenic E . coli strains [53] , [54] , whilst others represent genes that have not previously been shown to be associated with serum resistance . The murein lipoprotein gene lpp ( EC958_1897 ) showed the greatest difference between the test and control samples ( logFC of −10 ) , followed by pgm ( EC958_0806 ) , encoding phosphoglucomutase . Four hypothetical proteins were also identified , two of which ( EC958_0460 and EC958_0461 ) were further characterized in this study ( see below ) . As mentioned above , we employed a stringent threshold combining fold change and statistical significance to define the set of 56 genes in the EC958 serum resistome . In order to validate these findings , we attempted to test all 56 genes independently for their role in serum resistance . Using a modified lambda red mediated-homologous recombination approach [22] , [47] we successfully generated defined knock-out mutants for 54 genes ( 96 . 4% ) in EC958 . We were unable to obtain mutants for the remaining 2 genes ( acrA and EC958_2373 ) despite multiple attempts ( Table 2 ) . The 54 defined mutants were subjected to serum susceptibility testing , whereby the number of surviving colonies after a 90-minute exposure to fresh pooled human serum was compared to the number of colonies prior to treatment ( Table 2 ) . A mutant was defined as susceptible to serum when its log difference was at least 1 ( i . e . 10 fold reduction after exposure to serum ) . Forty-one genes ( 75 . 9% ) contributed to serum resistance in EC958 using this assay . In the case of the remaining 13 mutants , it is possible that the lack of susceptibility to human serum was a reflection of the assay , suggesting that survival in serum within a mixed population of one million mutants may be very different from survival of a pure population carrying the same defective mutation . Therefore , to better mimic the condition of TraDIS library serum selection , a competitive assay was devised where EC958 wild-type was mixed equally with a mutant before exposure to serum and the competitive index of the mutant was measured . Using this competitive assay , five of the twelve mutants were significantly attenuated compared to the wild-type EC958 strain ( Table 2 ) . Thus , the overall number of validated susceptible mutants was 46 out of 54 tested ( 85 . 2% ) . We hypothesized that one mechanism associated with enhanced sensitivity to human serum could be due to decreased membrane integrity caused by destabilization of the outer leaflet of the outer membrane . In order to test this , we examined the survival of the 54 mutants in response to outer membrane stresses ( i . e . SDS ) and osmotic potential ( i . e . NaCl ) . In total , 50 . 0% ( 27/54 ) of the mutants displayed enhanced sensitivity to SDS and 11 . 1% ( 6/54 ) of the mutants displayed enhanced sensitivity to NaCl ( Table 2 ) . A comparative analysis of these phenotypes in the context of serum sensitivity is presented below . It is well established that O-antigen represents the main determinant for serum resistance in E . coli . However , there are more than 180 different O-antigens that have been defined in E . coli and these may contribute differently to serum resistance in individual bacterial strains [32] , [33] . Furthermore , direct genetic evidence linking O-antigen biosynthesis to serum resistance is only available for a small number of specific O-antigen types . With this in mind , a detailed characterization of the O25b biosynthesis genes was performed using sequence comparison for function prediction in combination with analysis of LPS composition to deduce the role of each gene in resistance to serum . Similar to most E . coli strains , the O-antigen biosynthesis operon is located between the galF and gnd genes in EC958 . Figure 5A shows a comparison of the EC958 O-antigen operon with the equivalent operon from the K-12 strain MG1655 and the O25 serotype E . coli strain E47a [32] , [55] . LPS gel analysis was performed on all 54 defined mutants to identify genes that contribute to serum resistance by affecting LPS ( Figure S3 ) . The normal LPS pattern of EC958 consists of 12 bands including a thick bottom band representing the lipid A-core and an 11-band laddering pattern of lipid A-core bound O-antigen polymers , followed by approximately 6 thick bands of very long O-antigen chain length ( Figure 5C ) . In addition to the 6 genes involved in O-antigen biosynthesis mentioned above , the LPS patterns of 20 mutants were altered in comparison to wild-type EC958; 6 of these mutants ( waaLKYJ , wecA and rfaH ) only produced a lipid A-core ( Table 2 and Figure S3 ) . Mutation of the hyxA and hyxR genes in EC958 resulted in the modulation of O-antigen chain length ( Figure 6B ) . The EC958 hyxA mutant exhibited an increased proportion of O-antigen chain of 2 to 6 units with maximum number at 3–5 units and reduction in very high chain length polymer . The EC958 hyxR mutant had an increased proportion of O-antigen polymer of 2–4 units . The hyxA and hyxR genes are located in a pathogenicity island ( PAI-X ) consisting of 4 genes ( fimX and hyxRAB ) as previously described in the UPEC strain UTI89 [70] . The hyxB gene ( EC958_0459 ) has also been named upaB due to its function as an autotransporter [71] , and we prefer to maintain this nomenclature . This island is present ( in several variations ) in 24 out of 50 E . coli genomes across all pathotypes ( Figure 6A ) and current sequence data suggest that it is exclusive to E . coli . No functional prediction was found for hyxA , and thus this is the first report of hyxA involvement in serum resistance by regulating O-antigen chain length . The hyxR gene encodes a LuxR-like response regulator that suppresses the nitrosative stress response and contributes to intracellular survival in macrophages by regulating hmpA , which encodes a nitric oxide-detoxifying flavohaemoglobin [70] . The expression of the hyxR gene is regulated through bidirectional phase inversion of its promoter region by the upstream gene fimX , which encodes a tyrosine-like recombinase [70] . It is also worth noting that the contribution of hyxA to serum resistance was greater than hyxR , as demonstrated by the 3-log reduction in viability by the hyxA mutant compared to the hyxR mutant . A serum sensitive phenotype for the hyxR mutant was only observed in mixed competition assays , and both mutants did not exhibit altered sensitivity to SDS or NaCl ( Table 2 ) . A major advantage of whole genome approaches such as TraDIS lies in their power of discovery . Out of 46 genes that define the serum resistome of EC958 , 21 ( 46% ) genes were confirmed to be required for serum resistance independent of altered LPS patterns ( Table 2; Figure S3 ) . The function of these genes ranged across 10 COG functional categories and included ‘Carbohydrate transport and metabolism’ ( 4 genes ) , ‘Cell wall/membrane/envelope biogenesis’ ( 3 ) , ‘Posttranslational modification , protein turnover , chaperones’ ( 3 ) and many others ( Table 2 ) . Twelve of these genes were also associated with enhanced sensitivity to SDS and one with enhanced NaCl sensitivity ( Table 2 ) Of the non-LPS genes required for serum resistance , the most notable was lpp ( logFC −10 , log difference = 6 ) ( Table 2 ) . The lpp gene encodes one of the most abundant proteins in E . coli and is responsible for the stabilisation and integrity of the bacterial cell envelope [72] . Mutation of the lpp gene results in the formation of outer membrane blebs , leakage of periplasmic enzyme ribonuclease , decreased growth rate in media of low ionic strength or low osmolarity and hypersensitive to toxic compounds [73] , [74] . Indeed , the EC958 lpp mutant was more sensitive to SDS , suggesting decreased membrane integrity ( Table 2 ) . To the best of our knowledge , this study is the first to show a direct link between Lpp and serum resistance . Another set of genes notable in our TraDIS analysis include tolQAB , which encode three of the six proteins ( YbgC-YbgF-TolQ-R-A-B-Pal ) that make up the Tol-Pal system of E . coli cell envelope . The Tol-Pal system is responsible for maintaining the integrity of the outer membrane . TolQRA form an inner membrane complex in which TolQR is necessary for its stability [75] . TolB , a periplasmic protein , connects the inner membrane complex with the peptidoglycan-associated lipoprotein , Pal , which is anchored to the outer membrane [76] . In our study , mutation of the tolA and tolQ genes caused sensitivity to human serum and increased sensitivity to SDS , while the tolB mutant did not ( Table 2 ) . Our results demonstrate that the Tol-Pal system is important for resistance to human serum , and thus describe a novel function for this important cell wall complex . BamB is a lipoprotein that is part of the BamABCD complex . Mutation in bamB results in increased outer membrane permeability , thus enhancing sensitivity to rifampin and dramatically reducing growth on SDS and novobiocin [77] . Our data showed that mutation of the bamB gene in EC958 resulted in increased sensitivity to both human serum and SDS ( Table 2 ) . Our TraDIS experiment also indicated that the modification of lipid A with L-Ara4N was important for serum resistance . Three ( arnDEF ) of the seven genes involved in the biosynthesis and attachment of L-Ara4N to lipid A-core were identified as part of the serum resistome of EC958 and their role was confirmed by mutagenesis ( Table 2 ) . This mechanism is known to confer resistance to polymixin B by preventing its binding to lipid A [Reviewed in 78] , [79] , [80] . ArnD catalyzes a deformylation step to generate UDP-L-Ara4N before it is transported across the inner membrane by ArnEF [79] , [81] . The requirement of ArnDEF for serum resistance indicates that EC958 requires L-Ara4N modification to evade the antimicrobial activity of cationic peptides present in human serum . Interestingly , only the arnD mutation conferred sensitivity to SDS ( Table 2 ) , which might suggest a role of UDP-L-Ara4N in maintaining membrane integrity . Further investigation is needed to understand why ArnT , the final enzyme required for transferring the L-Ara4N residue to the 4′-phosphate group of lipid A-core , was not identified in our TraDIS-defined serum resistome . We also identified three genes encoding catabolic enzymes that contributed to the serum resistance phenotype of EC958 ( gmm , pgi and fbp ) and confirmed their role by mutagenesis ( Table 2 ) . Of these genes , only the pgi mutant displayed enhanced sensitivity to SDS ( Table 2 ) . Gmm is a GDP-mannose mannosyl hydrolase capable of hydrolyzing both GDP-mannose and GDP-glucose [82] . This enzyme contributes to the biosynthesis of GDP-fucose , a component of colanic acid , possibly by influencing the concentration of GDP-mannose or GDP-glucose in the cell and thus regulating cell wall biosynthesis [82] . Both Pgi and Fbp catalyze the production of D-fructose-6-phosphate from β-D-glucose-6-phosphate and fructose-1 , 6-bisphosphate , respectively [83] , [84] . D-fructose-6-phosphate is a precursor for the biosynthesis of UDP-GlcNAc , which in turn is required for peptidoglycan , lipid A and ECA biosynthesis . Thus , these three enzymes may catalyse key reactions that , if disrupted , could adversely affect the downstream biosynthesis of cell surface components including colanic acid , peptidoglycan , lipid A and ECA . Of all the chromosomal genes previously attributed to serum resistance , the only gene that was not identified in our TraDIS screen was ompA . To examine this further we constructed an EC958 ompA mutant and indeed observed it was sensitive to killing by human serum ( Figure 7 ) . One way to explain this discrepancy is that the phenotype of an ompA mutant could be complemented in trans by other ompA-intact bacteria in a mixed population such as the mutant library . In fact , OmpA inhibits serum-mediated killing by binding to C4b-binding protein ( C4BP ) to prevent the activation of C3b via the classical complement pathway [85] , and OmpA is known to be released from E . coli cells when treated with serum [86] . In our mutant library , ompA mutants only accounted for approximately 0 . 02% of the total bacterial cells , and thus we hypothesized that the release of OmpA from 99 . 98% of the cells , when treated with serum , provided OmpA in trans to complement ompA mutants . We tested this hypothesis by mixing the ompA mutant with wild-type EC958 at various ratios and indeed showed that ompA mutants were protected from serum killing if the proportion of ompA mutants was less than 15% ( Figure 7 ) . This result strongly suggests that in trans complementation of OmpA prevents the identification of ompA as a serum resistance gene in our assay . To further demonstrate the function of non-LPS genes in serum resistance , we selected three orphan genes ( acnB , greA and fbp ) that do not belong to an operon to perform genetic complementation . For these experiments , the selected gene was amplified by PCR , cloned into the low copy number plasmid pSU2718G , and transformed into the respective mutant strain for complementation . In each case , the phenotype of the complemented strain exactly matched that of wild-type EC958 ( Table 4 ) . Taken together , these results confirm the role of acnB , greA and fbp in serum resistance and provide a further layer of evidence to support the use of techniques such as TraDIS in functional gene discovery .
The rapid advancement of new sequencing technologies has created novel opportunities to interrogate biological systems that were not previously possible . TraDIS was first described as a method that combined high-density mutagenesis with Illumina next generation sequencing technology to study the essential genes of S . Typhi and the conditional essential genes required for survival in bile [42] . The increased data output afforded by next generation DNA sequencing is particularly useful and cost-effective for small bacterial genomes , however it presents technical and bioinformatical challenges for applications such as TraDIS that utilize low complexity DNA libraries . Here we present the application of a modified version of TraDIS that is amenable to multiplexing using the Illumina HiSeq 2000 platform , and we demonstrate its effectiveness by using it to define the essential gene repertoire and the serum resistome of a multidrug resistant strain from the globally disseminated E . coli ST131 lineage . The multiplexed TraDIS protocol utilizes a newly designed custom oligonucleotide in the library enrichment step of the Illumina library preparation protocol ( Figure S1 ) . This oligonucleotide incorporates the Illumina sequencing primer-binding site into transposon specific DNA fragments , enabling the use of the standard Illumina sequencing primer and eliminating the need to design and optimize another sequencing primer for each new transposon sequence . The 6-bp barcode immediately after the sequencing primer-binding site allows 12-sample multiplexing within one lane . The use of 12 barcodes at the first 6 nucleotides of read 1 increased the complexity of the library compared with the original TraDIS protocol , thus reducing data loss due to mis-identification of clusters [87] , [88] . However , the number of useable reads from our sequencing runs was still low ( 15–20% of total reads ) . We believe non-specific amplification in the enrichment step was the main cause , and further optimization of the enrichment PCR conditions is required . Similar approaches combining transposon mutagenesis with high-throughput sequencing ( Tn-seq [45] , INSeq [43] , HITS [44] ) have also been used to address different scientific questions , including the identification of essential genes and genes associated with enhanced fitness in specific growth conditions [41] , [42] , [45] , determination of niche-specific essential genes [43] , [44] , [89] , identification of genes associated with tolerance to various agents/conditions [42] , [90] and many other applications as reviewed elsewhere [91] , [92] . In terms of insertion density , we achieved 502 , 068 independent insertion sites with a density of 1/10 bp ( i . e . one insertion every 10 bp ) , which is comparable with the work by Christen et al . ( 1/8 bp in C . crescentus ) [41] , Langridge et al . ( 1/13 bp in Salmonella Typhi ) [42] and Barquist et al . ( 1/9 bp in Salmonella Typhimurium ) [46] . The identification of the essential gene set for a single organism is challenging due to several factors , including the presence of transposon insertion cold spots ( i . e . regions of low transposon insertion frequency ) , the difficulty in distinguishing mutations that prevent growth from those that severely reduce growth rate , pre-existing gene duplications and the specific growth conditions used in the experiment [93]–[95] . In this study , we define essential genes as those genes that , when mutated by transposon insertion , either prevent or severely attenuate growth on LB agar media supplemented with 30 µg/ml Cm at 37°C . The cut-off value to determine whether a gene is essential was defined as the intercept of two distributions of the insertion index of each gene: the exponential distribution representing essential genes and the gamma distribution representing non-essential genes [42] . This means that our essential genes also include those genes that can tolerate insertions but were severely attenuated in the input pool . Out of 315 essential genes , 64 genes had no transposon insertions , 178 genes had 1 to 5 transposon insertions and 73 genes had more than 5 transposon insertions ( Table S1 ) . The high density of insertion sites achieved in our study provided reliable data for the identification of essential genes within the EC958 genomes with a minimal probability of false positive calls due to transposon insertion cold spots . The identification of essential genes has previously been performed using several approaches in E . coli K-12 ( strains MG1655 and W3110 ) [96] , [97] . Baba et al . generated null mutations by lambda-red recombination in 3985 E . coli W3110 genes ( the Keio library ) , but were unable to mutate 303 candidate essential genes [97] . This set of essential genes was further consolidated by manual literature review on the EcoGene website ( www . ecogene . org ) , which reduced the set to 289 genes . Of the 315 essential genes identified for EC958 in this study , 231 genes ( 73% ) matched those previously described in the EcoGene list ( Table S1 ) . There were 84 essential genes specific for EC958 , twenty-four of which do not have homologs in the MG1655 genome . In contrast , 58 genes previously identified as essential for E . coli K-12 were either not present in EC958 or not identified in our analysis . The majority of essential genes in EC958 are conserved with 91% of the genes present in more than 90% of E . coli complete genomes available . In this study , we provided two layers of evidence for the role of each serum resistance gene: by simultaneously assaying a large mutant library and by generation of defined mutants for independent phenotypic testing . Indeed , using defined mutagenesis we were able to confirm a role for 46 of the 56 genes identified by TraDIS in serum resistance . To the best of our knowledge , this represents the first large scale follow-up of TraDIS data in this manner and highlights the effectiveness of the technique in large-scale functional genomics . Our study also revealed that trans complementation of specific mutants can occur in a large mutant library population , as demonstrated by our findings with an ompA mutant . We also demonstrated complete complementation of mutants containing deletions in the acnB , greA and fbp genes , corroborating their novel role in serum resistance independent of LPS alterations . Finally , we provided further insight into the mechanistic action of the serum resistance genes identified in EC958 by examining the survival of the respective mutants to outer membrane stresses that affect antimicrobial access and osmotic potential . The search for genetic determinants of serum resistance in bacteria has been ongoing since the 1970s [98] . Our current understanding of the mechanisms that promote bacterial resistance to human serum include a role for surface structures such as O antigens , K antigens , outer membrane proteins ( OmpA ) , and plasmid-encoded proteins ( TraT , Iss ) [37]–[39]; notably , however , not all of these mechanisms are required for resistance in a single bacterial strain [32] , [33] . Our study represents the first report to simultaneously investigate the entire serum resistome of one strain . Our results demonstrated that both the lipid A-core and O25 antigen are crucial for serum resistance in EC958 , while K antigen does not contribute to serum resistance . Out of 54 defined mutants investigated , half had changes in their LPS gel patterns , all of which resulted in serum sensitivity . In contrast , none of the K capsular biosynthesis genes were identified in our TraDIS screen . This result is similar to that reported for the O75:K5 UPEC strain GR-12 , where alterations in O75 LPS affected serum resistance more than a K5 null mutation [34] . It is likely , however , that there are strain-specific differences for the role of O antigen and K capsule in serum resistance , and that this reflects differences in the make-up of these structures . For example , previous analysis of an E . coli O4:K54:H5 blood isolate revealed that the K54 antigen contributes more to serum resistance than the O4 antigen [99] . Other K antigens such as the K1 and K2 capsules have also been shown to play an important role in serum resistance [100] , [101] . The K antigen expressed by EC958 has not been typed but genomic analysis shows that EC958 has a group 2 capsular gene cluster that conforms to the conserved structure of this group [Reviewed in 102] . However , region 2 , which encodes glycosyltransferases specific for each K type , shares such low similarity with available sequences in the GenBank database that deducing its K type in silico was not possible . The O25 antigen gene cluster was further characterized using sequence analysis , targeted mutation and LPS profiling . All of the dTDP-α-L-rhamnose biosynthesis genes ( rmlCADB ) were required for serum resistance . However , EC958 lacks the biosynthesis genes for UDP-FucNAc , a component of O25 antigen unit . If , based on the cross reaction of antiserum against the O25 antigen with O25b expressing cells , we assume that EC958 has the same O-antigen repeat unit as the O25 determined from previous studies [57] , [58] , then EC958 must possess a novel mechanism for the synthesis or uptake of UDP-FucNAc . Two additional surface antigens that contribute to serum resistance in EC958 are the enterobacterial common antigen and colanic acid ( M antigen ) . Mutations in five ECA biosynthesis genes rendered EC958 susceptible to serum . While mutation of three of these genes ( wecA , wzzE and wecD ) affected LPS and sensitivity to SDS , mutation of wecE and wecF did not change LPS ( although a wecF mutant was more sensitive to SDS ) , suggesting that the ECA may be involved in serum resistance , perhaps indirectly via its role in membrane integrity . Our data also suggest the involvement of colanic acid in serum resistance . Three genes encoding for colanic acid biosynthesis ( wcaI , gmm and wcaF ) were identified by TraDIS . Gmm is most likely to be involved in the biosynthesis of colanic acid [82] , while mutation of the wcaI gene did not confer serum resistance . The product of wcaF was predicted to be an acetyltransferase [103] required for colanic acid production [69] . The EC958 wcaF mutant possessed an altered LPS pattern with a reduced amount of O-antigen ( especially very long chain length O-antigen ) and was sensitive to both SDS and high osmolarity . The enhanced sensitivity of the EC958 wcaF mutant could therefore be explained by a number of factors , including altered LPS , altered colonic acid and reduced overall membrane integrity . A number of other genes were identified that contributed to serum resistance in an LPS-dependent manner . The gene nagA encodes N-acetylglucosamine-6-phosphate deacetylase , an enzyme important for the metabolism of N-acetyl-D-glucosamine [104] . It catalyzes the first step in producing UDP-GlcNAc , a nucleotide sugar required for ECA , lipid A and peptidoglycan biosynthesis , by deacetylating N-acetylglucosamine-6-phosphate to glucosamine-6-phosphate [Reviewed in 105] . However , NagA is not solely responsible for the production of UDP-GlcNAc because glucosamine-6-phosphate can also be obtained via GlmS from fructose-6-phosphate or taken up from the environment by ManXYZ [105] . Indeed , the LPS banding pattern of the nagA mutant was different to that of the parent strain ( i . e . it possessed thicker second and third bands from the bottom of the gel; Figure S3 ) , suggesting its enhanced sensitivity phenotype may be associated with a predominantly shorter O antigen . Pgm is a phosphoglucomutase that catalyses the reversible conversion of glucose-1-phosphate to glucose-6-phosphate , an important step in galactose and maltose catabolism [106] . A pgm mutant has several phenotypes; it is defective in swimming and swarming mobility [107] , it possesses an aberrant ( shorter and wider ) cell morphology , is sensitive to detergents [108] and it stains blue with iodine when grown in the presence of galactose [106] . An EC958 pgm mutant produced little full length O-antigen; the majority of its LPS condensed into a thick band of incomplete lipid A-core and a thin clear band of lipid A-core plus one unit of O-antigen . This feature is consistent with the high serum and SDS sensitivity phenotypes observed for this mutant . GalE is a well-studied enzyme that catalyzes the interconversion of UDP-galactose and UDP-glucose [Reviewed in 109] . Both nucleotide sugars are required for colanic acid biosynthesis . Furthermore , UDP-glucose is used in three steps to synthesize the LPS outer core ( catalyzes by WaaG , WaaI and WaaJ ) . LPS patterns of the galE mutant exhibited a very thick band of lipid A-core , suggesting that the lipid A-core in this strain has multiple sizes . This may be explained by the limiting effect of UDP-glucose in the three steps involved in its incorporation into the outer core . UDP-glucose can also be synthesized by GalU from glucose-1-phosphate [110] , which may explain why an EC958 galE mutant could still make LPS ( Figure S3 ) . Despite being able to make LPS , however , the galE mutant was sensitive to human serum . Whether this sensitivity can be attributed to the effect a galE mutation has on LPS or colanic acid remains to be determined . We have demonstrated the successful application of multiplexed TraDIS for a functional genomics study targeted at E . coli EC958 , a prototype strain from the globally disseminated and multidrug resistant E . coli ST131 lineage . This approach enabled the first description of an essential gene set from an ExPEC strain . Our work has also defined the serum resistome in E . coli EC958 . This comprehensive inventory of E . coli EC958 genes that contribute to this phenotype provides a framework for the future characterization of virulence genes in ExPEC as well as other Gram-negative pathogens that cause systemic infection .
Approval for the collection of human blood was obtained from the University of Queensland Medical Research Ethics Committee ( 2008001123 ) . All subjects provided written informed consent . E . coli EC958 was isolated from the urine of a patient presenting with community UTI in the Northwest region of England and is a representative member of the UK epidemic strain A ( PFGE type ) , one of the major pathogenic lineages causing UTI across the United Kingdom [23] . EC958Δlac , which contained a mutation in the lac operon , was used in competitive assays . This strain had an identical growth rate to wild-type EC958 . Strains were routinely cultured at 37°C on solid or in liquid Luria Broth ( LB ) medium supplemented with the appropriate antibiotics ( Cm 30 µg/ml or gentamicin 20 µg/ml ) unless indicated otherwise . A miniTn5-Cm transposon containing a Cm cassette flanked by Tn5 mosaic ends ( sequence from Epicenter ) was PCR amplified from pKD3 plasmid DNA ( NotI digested ) using primers 2279 5′- CTGTCTCTTATACACATCTcacgtcttgagcgattgtgtagg-3′ and 2280 5′- CTGTCTCTTATACACATCTgacatgggaattagccatggtcc-3′ . The PCR reactions were performed using Phusion High-Fidelity DNA polymerase ( New England BioLabs ) . The amplicon was purified using the QIAGEN MinElute PCR purification kit before being phosphorylated using T4 polynucleotide kinase ( New England BioLabs ) and subjected to the final purification step . A total of at least 800 ng of this miniTn5-Cm transposon DNA was incubated in an 8 µl reaction containing 4 µl of EZ-Tn5 transposase ( Epicenter Biotechnologies ) at 37°C for 1 h then stored at −20°C . Bacterial cells were prepared for electroporation as previously described [42] . Briefly , cells were grown in 2×TY broth to an OD600 of 0 . 3–0 . 5 , then harvested and washed three times in 0 . 5× volume of 10% cold glycerol before being resuspended in a 1/1000× volume of 10% cold glycerol and kept on ice . A volume of 60 µl cells was mixed with 0 . 2 µl of transposomes and electroporated in a 2 mm cuvette using a BioRad GenePulser set to 2 . 5 kV , 25 µF and 200Ω . Cells were resuspended in 1 mL SOC medium and incubated at 37°C for 2 hours , then spread on LB agar plates supplemented with Cm 30 µg/mL . After incubation overnight at 37°C , the total number of colonies was estimated by counting a proportion from multiple plates . Chloramphenicol resistant colonies were resuspended in sterilised LB broth using a bacteriological spreader before adding sterile glycerol to 15% total volume and stored in −80°C . Each batch of mutants contained an estimated 32 , 000 to 180 , 000 mutants . The final library of 1 million mutants was created by pooling 11 mutant batches , resulting in a cell suspension of 2×1011 CFU/ml . Freshly pooled human serum was collected from at least two healthy individuals on the day of the experiment . Ten milliliters of blood was collected from each person and centrifuged at 4000 rpm for 10 minutes to collect the serum . Approximately 2×108 viable mutants were incubated in 1 ml of 50% freshly pooled human serum in LB broth at 37°C for 90 minutes . The control samples were prepared the same way but were incubated with inactivated serum ( Millipore ) instead of fresh serum . Both control and test experiments were performed in duplicate . The cells were then washed twice with sterile 1×PBS to remove serum , transferred to 100 ml LB broth and allowed to grow at 37°C with 250 rpm shaking for 4 hours . The genomic DNA was then extracted from 5 ml of each culture using Qiagen 100-G genomic tips . Genomic DNA was standardized to 3 . 6 µg in a volume of 120 µl before being sheared by Covaris S2 according to the Illumina TruSeq Enrichment gel-free method ( TruSeq DNA sample preparation v2 guide ) . The subsequent steps of DNA end repair , DNA end adenylation and adapter ligation were also done following the Illumina TruSeq v2 instructions . The adapter-ligated fragments containing transposon insertion sites were enriched using a transposon-specific indexing forward primer and the Illumina reverse primer Index 1 ( Table S5 ) at 500 nM each per reaction . This 89 bp forward primer binds specifically to miniTn5-Cm transposon ( 25 bp ) and carries 64 bp overhang which includes 6 bp index sequence , 33 bp binding site for Illumina read 1 sequencing primer and 23 bp P5 sequence for binding to the flowcell . The enrichment step was done using the KAPA Library Amplification kit ( KAPA Biosystems ) at an annealing temperature of 60°C for 22 cycles . The KAPA Library Quantification kit was used to measure the concentration of DNA fragments in the enriched library . Twelve libraries from 12 samples were pooled to equimolar concentrations to achieve a cluster density of 850 K/mm2 when 10 nM of library pool was loaded onto the flowcell . The 12-plex pool was loaded on 3 lanes of TruSeq v2 and 3 lanes of TruSeq v3 flowcells for sequencing using a 100 cycles , paired-end protocol to access the reproducibility and read quantity among lanes and flowcells . The data from six samples ( Figure 1A ) were presented in this study . The TraDIS sequence data from this study was deposited on the Sequence Read Archive ( SRA ) under the BioProject number PRJNA189704 . Sequence reads from the FASTQ files were split according to twelve 6 bp index sequences combined with the 37 bp transposon-specific sequence using fastx_barcode_splitter . pl ( total length of 43 bp as barcodes , allowing for 2 mismatches ) ( FASTX-Toolkit version 0 . 0 . 13 , http://hannonlab . cshl . edu/fastx_toolkit/index . html ) . The barcode matching reads were trimmed off the 43 bp barcode at the 5′end and 25 bp of potential low quality at the 3′ end , resulting in high quality sequence reads of 31 bp in length that were used to map to the EC958 chromosome ( PRJEA61443 ) by Maq version 0 . 7 . 1 [111] . Subsequent analysis steps were carried out as previously described [42] to calculate the number of sequence reads ( raw read counts ) and the number of different insertion sites for every gene , which were then used to estimate the threshold to identify essential genes . The read counts and insertion sites were visualized using Artemis version 13 . 0 [112] . The circular genome diagram was generated by CGView [113] and linear genetic comparison was illustrated using Easyfig version 2 . 1 [114] . We identified genes required for survival in human serum by comparing the differences in read abundance of each gene between the inactivated serum control and active serum test samples using the Bioconductor package edgeR ( version 2 . 6 . 10 ) [48] . The raw read counts from two biological replicates of each treatment were loaded into the edgeR package ( version 2 . 6 . 12 ) using the R environment ( version 2 . 15 . 1 ) . Genes that have very low read counts in all the samples ( essential genes ) were removed from further analysis . The composition bias in each sequence library was normalized using the trimmed mean of M value ( TMM ) method [115] . We then used the quantile-adjusted conditional maximum likelihood ( qCML ) for negative binomial models to estimate the dispersions ( biological variation between replicates ) and to carry out the exact tests for determining genes with significantly lower read counts in the test samples compared to the control samples [116] , [117] . Stringent criteria of log fold-change ( logFC ) ≤−1 and false discovery rate ≤0 . 001 were chosen to define a list of the most significant genes for further investigation by phenotypic assays . Chromosomal DNA purification , PCR and DNA sequencing of PCR products was performed as previously described [118] . Defined mutations were made using the λ-Red recombinase method with some modifications [22] , [47] . In brief , the final PCR products were fused and amplified from three fragments containing two 500-bp homologous regions flanking the gene of interest and a Cm cassette from pKD3 plasmid ( see Table S5 for list of primers ) . The fused PCR products were then electroporated into EC958 harbouring a gentamicin resistant plasmid carrying the λ-Red recombinase gene . Mutants were then selected and confirmed by sequencing . Complementation was done by cloning the gene of interest into a gentamicin resistant derivative of pSU2718 [119] at BamHI-XbaI cut sites ( primers listed in Table S5 ) . The construct was then transformed into the respective mutant and induced using 1 mM IPTG before and during phenotypic assays . Overnight bacterial cultures were washed in phosphate buffered saline ( PBS ) and then standardized to an OD600 of 0 . 8 . Equal volumes ( 50 µL ) of standardized cultures and pooled human sera were mixed and incubated for 90 min at 37°C ( in triplicates ) . Viable counts were performed to estimate the number of bacterial cells prior to serum treatment ( t = 0 min ) and post serum treatment ( t = 90 min ) . E . coli MG1655 was used as a control as it is completely killed by serum . Serum and PBS only samples served as sterility controls . Competitive serum resistance assays were performed in the same manner , except that a 50∶50 mixture of wild-type ( EC958Δlac ) and mutant strains were used . Viable counts were performed on MacConkey agar , which allowed the differentiation of EC958Δlac ( non-lactose fermenter ) and the mutant strains . The MICs of SDS and NaCl were determined by broth microdilution method as previously described [120] . We used five concentrations for SDS including 0 . 125% , 0 . 0625% , 0 . 031% , 0 . 016% and 0 . 008% in LB . For NaCl , the range of concentration was 0 . 8 M , 0 . 6 M , 0 . 5 M , 0 . 4 M and 0 . 3 M . LPS was extracted from bacterial strains and LPS patterns were determined by Tricine-SDS Polyacrylamide gel electrophoresis ( TSDS-PAGE ) and visualized by silver staining as previously described [121] , [122] . | The emergence and rapid dissemination of new bacterial pathogens presents multiple challenges to healthcare systems , including the need for rapid detection , precise diagnostics , effective transmission control and effective treatment . E . coli ST131 is an example of a recently emerged multidrug resistant pathogen that is capable of causing urinary tract and bloodstream infections with limited available treatment options . In order to increase our molecular understanding of E . coli ST131 , we developed a high-throughput transposon mutagenesis system in combination with next generation sequencing to test every gene for its essential role in growth and for its contribution to serum resistance . We identified 315 essential genes , 270 of which were conserved among all currently available complete E . coli genomes . Fifty-six genes that define the serum resistome of E . coli ST131 were identified , including genes encoding membrane proteins , proteins involved in LPS biosynthesis , regulators and several novel proteins with previously unknown function . This study therefore provides an inventory of essential and serum resistance genes that could form a framework for the future development of targeted therapeutics to prevent disease caused by multidrug-resistant E . coli ST131 strains . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2013 | The Serum Resistome of a Globally Disseminated Multidrug Resistant Uropathogenic Escherichia coli Clone |
The Unfolded Protein Response ( UPR ) maintains homeostasis in the endoplasmic reticulum ( ER ) and defends against ER stress , an underlying factor in various human diseases . During the UPR , numerous genes are activated that sustain and protect the ER . These responses are known to involve the canonical UPR transcription factors XBP1 , ATF4 , and ATF6 . Here , we show in C . elegans that the conserved stress defense factor SKN-1/Nrf plays a central and essential role in the transcriptional UPR . While SKN-1/Nrf has a well-established function in protection against oxidative and xenobiotic stress , we find that it also mobilizes an overlapping but distinct response to ER stress . SKN-1/Nrf is regulated by the UPR , directly controls UPR signaling and transcription factor genes , binds to common downstream targets with XBP-1 and ATF-6 , and is present at the ER . SKN-1/Nrf is also essential for resistance to ER stress , including reductive stress . Remarkably , SKN-1/Nrf-mediated responses to oxidative stress depend upon signaling from the ER . We conclude that SKN-1/Nrf plays a critical role in the UPR , but orchestrates a distinct oxidative stress response that is licensed by ER signaling . Regulatory integration through SKN-1/Nrf may coordinate ER and cytoplasmic homeostasis .
The endoplasmic reticulum ( ER ) is responsible for multiple functions in protein synthesis and processing , lipid metabolism , xeno/endobiotic detoxification , and Ca2+ storage ( reviewed in [1] , [2] ) . The ER forms a continuous structure with the nuclear envelope and maintains extensive contact with mitochondria [3] , [4] . Consequently , the ER is well positioned to sense and respond to changes in the cellular environment . All secretory and membrane-bound proteins are synthesized in the rough ER , a process that is highly regulated so that only properly folded and modified proteins are released to the Golgi [1] , [2] , [5] , [6] . Maturation and folding of these proteins involves glycosylation and formation of appropriate Cys-Cys crosslinks . When its protein folding capacity is exceeded ( ER stress ) , the ER protects itself through the Unfolded Protein Response ( UPR ) ( Figure S1A ) [2] , [5] , [6] . This signaling and transcription program decreases protein translation , expands ER size and folding capacity , and directs misfolded proteins to be degraded in the cytosol . The UPR functions continuously to maintain ER homeostasis , but is amplified and diversified under ER stress conditions [5] , [7]–[10] . In response to severe ER stress , the UPR promotes ER absorption through autophagy and ultimately may induce cell death . ER stress and the UPR have been implicated in many human diseases , including diabetes , inflammatory disease , neurodegenerative disease , secretory cell malignancies , and other cancers [6] , [11] , [12] . The canonical metazoan UPR is orchestrated by three major ER transmembrane signaling proteins ( IRE1 , PERK , and ATF6 ) , and three bZIP-family transcription factors ( XBP1 , ATF4 , and cleaved ATF6 ) ( Figure S1A ) [2] , [5] , [6] . The most ancient of these transmembrane proteins , IRE1 , is a cytoplasmic endoribonuclease and kinase that senses unfolded proteins in the ER . In response to ER stress , the IRE1 RNAse initiates cytoplasmic splicing of the mRNA encoding XBP1 , the transcription factor that is most central to the UPR . The IRE1 kinase contributes to ER homeostasis by regulating the IRE-1 endonuclease activity , and transmits signals through JNK , p38 , and other pathways . The kinase PERK phosphorylates the translation initiation factor eIF2α , thereby globally decreasing translation . This reduces the ER protein-folding load , but also favors translation of mRNAs that encode protective proteins , including ATF4 . ATF6 resides in the ER membrane but is transported to the Golgi and cleaved in response to ER stress . The activation status of these transmembrane proteins is influenced by their interactions with the ER chaperone BiP ( HSP-3/-4 in C . elegans ) . The ER lumen maintains an oxidative environment , in contrast to the cytoplasm , because the ER enzyme systems that form disulfide bonds generate reactive oxygen species ( ROS ) [1] , [13] , [14] . Accordingly , ER stress may eventually lead to cellular oxidative stress and activation of oxidative stress defense genes [15] . Metazoan oxidative and xenobiotic stress responses are orchestrated mainly by the Nrf bZIP-family transcription factors ( Nrf1 , 2 , 3 in mammals ) . Nrf-family proteins regulate genes involved in various small molecule detoxification processes , including glutathione biosynthesis and conjugation , and have been implicated in longevity assurance in invertebrates and mammals [16]–[21] . These transcription factors have recently been shown to function in proteasome regulation , stem cell maintenance , and metabolism , suggesting that they may control a wider range of processes than previously realized [22]–[26] . It has been reported that mammalian Nrf1 and Nrf3 associate with the ER membrane and nuclear envelope [27]–[30] , and that Nrf2 is phosphorylated by PERK [31] , [32] . While these last observations are intriguing , it is unknown whether Nrf-family proteins might actually be involved in ER stress defenses , either through mobilizing an oxidative stress response or participating in the UPR itself . The nematode C . elegans has been a valuable system for investigating how Nrf proteins function and are regulated in vivo , because of its advantages for employing genetics to elucidate regulatory networks , and performing whole-organism analyses of stress resistance and survival . The C . elegans Nrf ortholog SKN-1 plays a critical role in resistance to oxidative and xenobiotic stress , and in various pathways that extend lifespan [16] , [17] , [19] , [23] , [33] . Here we describe a comprehensive analysis of whether SKN-1 might be involved in the UPR . We found that under ER stress conditions SKN-1 directly activates many genes involved in ER function , including canonical ER signaling and transcription factors that in turn induce skn-1 transcription . Importantly , this response is distinct from that which SKN-1 mobilizes under oxidative stress conditions . SKN-1 is required for resistance to ER stress , including reductive stress , a surprising finding given the importance of SKN-1 for oxidative stress defense . Unexpectedly , UPR signaling is needed for SKN-1 to mobilize an oxidative stress response , suggesting that the ER has a licensing and possibly sensing role during oxidative and xenobiotic stress responses .
Several observations led us to investigate whether SKN-1/Nrf might be involved in ER stress defenses . Expression profiling that we performed in C . elegans under normal and oxidative stress conditions suggested that SKN-1 regulates a number of genes that are involved in UPR or ER functions [21] . These included atf-5 ( UPR transcription factor ATF4 ) , ckb-4 ( choline kinase ) , pcp-2 ( prolyl carboxypeptidase ) , and many genes encoding xenobiotic metabolism enzymes that localize to the smooth ER ( Table S1 ) . Moreover , a genome-wide Chromatin Immunoprecipitation ( ChIP ) analysis of C . elegans L1 stage larvae ( MOD-ENCODE ) [34] detected binding of transgenically expressed SKN-1 at the predicted regulatory regions of numerous genes involved in UPR- or ER processes , including UPR signaling and transcription ( ire-1 , xbp-1 , pek-1 , and atf-6 ) , Ca++ signaling , and protein folding and degradation ( Table S1 ) . To investigate whether SKN-1 might be involved in the UPR , we first used quantitative ( q ) RT-PCR to investigate whether it is needed for expression of representative ER stress-induced or ER maintenance genes , many of which are predicted to be SKN-1 targets ( Table S1 ) . In these initial gene expression studies we induced ER stress by treating C . elegans with the N-linked glycosylation inhibitor tunicamycin ( TM ) , at a concentration that readily induces the UPR but does not cause detectable toxicity ( 5 µg/ml , Figure S1B ) [15] . TM treatment resulted in skn-1-dependent upregulation of numerous canonical or predicted UPR- or ER-related genes ( Figures 1A and 1B , Table S1 ) . skn-1 was also required for the basal expression of psd-1 , R05G6 . 7 , and cnb-1 , even though these genes were not activated by TM ( Figures 1A and 1B ) . TM-induced ER stress also upregulated two direct SKN-1 targets that are involved in glutathione metabolism ( gcs-1 and gst-4 ) [19] in a skn-1–dependent manner , and transgenic reporter analysis detected gcs-1 activation in the intestine , the C . elegans counterpart to the gut , liver , and adipose tissue ( Figures 1C and 1D ) . Importantly , however , ER stress did not activate various other genes that are typically induced by SKN-1 under oxidative stress conditions ( Figure S1C ) . Taken together , the data indicate that SKN-1 mediates a response to ER stress , but also that this response does not correspond simply to its oxidative stress defense function . To investigate whether SKN-1 activates genes directly during ER stress , we used ChIP to detect endogenous SKN-1 and markers of transcription activity at pcp-2 , atf-5 , and gst-4 , each of which is flanked by SKN-1 binding sites and upregulated by oxidative and ER stress in a skn-1-dependent manner [21] ( Figures 1B and 1C ) . SKN-1 was readily recruited to these genes in response to either TM-induced ER stress or Arsenite ( AS ) -induced oxidative stress ( Figures 2A , 2E , 2I , and S2A-S2C ) . During transcription , RNA Polymerase II ( Pol II ) is phosphorylated on Ser 2 of its C-terminal domain ( CTD ) repeat ( P-Ser2 ) [35] . At each gene we examined , ER stress increased Ser 2 phosphorylation levels ( Figures 2B , 2F , and 2J ) . Also consistent with transcriptional activation , at these loci ER stress increased acetylation of Histone H3 , another marker of transcription activity [36] , but reduced overall Histone H3 occupancy ( Figures 2C , 2D , 2G , 2H , 2K , and 2L ) . Taken together , our findings suggest that SKN-1 directly activates a major transcriptional response to ER stress . We next investigated whether SKN-1 might regulate expression of core UPR signaling and transcription factors , as predicted by the MOD-ENCODE data [34] . XBP-1 is central to the UPR , and in mammals it controls transcription of other core UPR genes ( atf4/atf-5 , and BiP/hsp-4 ) along with many downstream genes [6] , [37] . During the UPR , xbp-1 expression is regulated at the level of transcription , as well as through cytoplasmic splicing of its mRNA by the IRE-1 endoribonuclease ( Figure S1A ) [5] , [6] . The spliced form of the xbp-1 mRNA ( xbp-1s ) encodes the transcriptionally active form of XBP-1 ( XBP-1s ) . When SKN-1 was lacking , ER stress failed to induce accumulation of each xbp-1 mRNA form and , remarkably , decreased the ratio of xbp-1s to the unspliced xbp-1 form ( xbp-1u ) ( Figures 3A , 3B , and S3A ) . The xbp-1 locus includes a predicted SKN-1 binding site ( not shown ) , and ChIP results indicated that endogenous SKN-1 accumulates at the xbp-1 site of transcription in response to ER stress ( Figure 3C ) . This evidence that SKN-1 directly regulates xbp-1 could account for the reduction in total xbp-1 mRNA , but not the apparent effect of SKN-1 on xbp-1 splicing . A plausible explanation is that lack of SKN-1 also reduced basal and ER stress-induced expression of ire-1 ( Figures 3D and 3E ) . Moreover , we observed that SKN-1 is recruited to the ire-1 locus in response to ER stress ( Figure 3F ) , consistent with MOD-ENCODE evidence that ire-1 may be a SKN-1 target [34] . SKN-1 was also required for expression of other core UPR genes . Mutation or RNAi knockdown of skn-1 prevented ER stress-induced expression of the unfolded protein chaperone and sensor HSP-4 ( BiP ) ( Figure S1A ) ( Figures 3G , S3B , and S3C ) . Binding of SKN-1 at hsp-4 was not detected in the MOD-ENCODE study of L1 larvae [34] , but our ChIP evidence indicated that both SKN-1 and XBP-1 bind directly to the hsp-4 locus ( Figures S3D and S3E ) , which includes predicted SKN-1 binding sites ( not shown ) . SKN-1 similarly contributed to expression of the core UPR factors pek-1 and atf-6 ( Figures 3D and 3E ) . Our evidence that SKN-1 is important for transcriptional induction of core UPR signaling and regulatory factors predicts that it should be important for C . elegans survival under ER stress conditions . Treatment with TM at a 7-fold higher concentration ( 35 µg/ml ) than is sufficient to induce the UPR impaired the survival of skn-1 mutants but not wild type animals ( Figure 3H and Table S2 ) . We conclude that SKN-1 plays a critical role in the UPR through its direct transcriptional regulation of core UPR factors , along with many downstream genes . We next examined whether expression of skn-1 itself is increased when the ER becomes stressed , and whether various conditions that cause ER stress affect SKN-1 activity . Treatment with TM increased the levels of multiple mRNA species that encode SKN-1 isoforms ( Figure 4A and S4A ) . In addition , non-lethal treatment with either the Ca++ pump inhibitor thapsigargin ( Thap ) or the proteasome inhibitor Bortezomib upregulated transcription of skn-1 , and various SKN-1-regulated genes ( Figures 1 , 4A , and S4B–S4C ) . Finally , knockdown of either the ER chaperone hsp-4 or the UPR transcription factor atf-6 resulted in transcriptional upregulation of skn-1 and many of its ER stress targets in the absence of drug treatment , presumably because of an elevated level of ER stress ( Figures 4A , S4D and S4E ) . We conclude that skn-1 transcription and activity are increased in response to a variety of conditions that are associated with ER stress . An important hallmark of the UPR is a decrease in the overall levels of translation [5] , [6] . This relieves stress on the ER , and allows translation of atf4 and other protective genes to be maintained or even increased . We investigated whether SKN-1 translation is similarly “spared” under ER stress conditions . Supporting this idea , TM treatment increased SKN-1 protein levels , a trend that was observed in Western and IP-Western analyses of whole animals with two specific SKN-1 antibodies ( Figures 4B and S4F–S4I ) . Based upon its size , this approximately 85 kD SKN-1 species is likely to represent SKN-1a , the largest SKN-1 isoform . While this size is larger than the expected SKN-1a MW of 70 kD , SKN-1 is phosphorylated and predicted to be glycosylated , as is characteristic of Nrf1 and Nrf3 ( not shown ) [17] , [28] , [38]–[40] . Our finding that SKN-1 protein levels are increased by ER stress is consistent with earlier evidence that SKN-1 translation seemed to be preserved when translation initiation was inhibited [41] . Prolonged ER stress leads to accumulation of reactive oxygen species ( ROS ) and induction of an oxidative stress response [15] , [42] , making it important to determine whether ER stress treatments might activate SKN-1 simply through a secondary response to oxidative stress . Arguing against this interpretation , even though SKN-1 is well known to defend against oxidative stress , we found that reductive ER stress also induced a SKN-1-dependent response . The reducing agent dithiothreitol ( DTT ) initiates the UPR through reduction of Cys-Cys bonds in the ER [43] . DTT treatment resulted in transcriptional induction of skn-1 and many of its target genes , and increased SKN-1 protein levels ( Figures 4C and S4J ) . SKN-1 appeared to be required for its downstream targets to be activated by DTT-induced reductive stress ( Fig . S4K ) , and knockdown of either skn-1 or hsp-4 rendered C . elegans comparably sensitive to reductive stress from DTT ( Figure S4L and Table S3 ) . Another way to reduce oxidation in the ER is through inhibiting expression of the oxidase ERO-1 , which promotes Cys-Cys crosslinking [43] . ero-1 RNAi decreases ROS levels , initiates the UPR , and extends lifespan [15] . As observed with DTT , ero-1 RNAi transcriptionally activated skn-1 and several of its downstream targets ( Figure 4D ) . Additional lines of evidence support the idea that SKN-1 acts in the UPR independently of its role in oxidative stress defense . Many genes that are activated by SKN-1 under oxidative stress conditions were not upregulated by ER stress ( Figures S1C and S4M ) . Oxidative stress from AS treatment induced the SKN-1::GFP ( green fluorescent protein ) fusion to accumulate to high levels in intestinal nuclei , as previously described ( Inoue , et al . , 2005 ) , but this did not occur in response to ER stress ( Figure S4N ) . Finally , we did not observe increased levels of oxidized proteins under conditions of TM-induced ER stress ( Figure S4O ) . Taken together , the data show that ER stress directs SKN-1 to activate a specific set of its target genes independently of any secondary oxidative stress response . If ER signaling pathways regulate SKN-1 , then key UPR signaling and transcription factors should be required for ER stress to activate SKN-1 and its target genes . Accordingly , RNAi or mutation of ire-1 , atf-5 , pek-1 , or hsp-4 essentially prevented ER stress from inducing transcription of skn-1 and several of its target genes ( Figure 5A ) . Knockdown of xbp-1 under control conditions increased background expression of some SKN-1 isoforms and target genes ( skn-1b , pcp-2 , gst-4 , hsp-4 ) , possibly because ER stress was increased , but also interfered with ER stress-induced activation of several of these genes ( skn-1a , pcp-2 , gcs-1 , hsp-4 ) ( Figure S5A ) . RNAi against ire-1 , which is essential for XBP-1s expression [5] , [6] , also blocked TM-induced accumulation of SKN-1 , Pol II , or P-Ser2 at the gst-4 , pcp-2 , and atf-5 loci ( Figures 5B–5E , S5B and S5C ) . Knockdown of hsp-4 or pek-1 had a similar effect ( Figure S5D–S5G ) . The evidence indicates that , in general , core UPR factors are required for ER stress to upregulate expression of SKN-1 and its target genes . The most straightforward mechanism through which ER stress could increase skn-1 transcription is through the direct regulation of skn-1 by one or more of the canonical UPR transcription factors . During the UPR , downstream gene transcription is controlled largely by XBP1 and ATF4 , which may regulate each other directly , with ATF-6 playing a more specialized role [8] , [15] , [37] . The skn-1 locus contains possible XBP-1 and ATF-6/XBP-1 binding elements ( not shown ) , and genome-wide ChIP studies suggest that mammalian Nrf3 may be a direct XBP1 target [37] . We determined that XBP-1 binds within the skn-1 locus in response to ER stress , suggesting direct regulation ( Figure 5F ) , a remarkable parallel to the direct regulation of xbp-1 by SKN-1 ( Figure 3C ) . Moreover , ATF-6 was also recruited to the skn-1 locus in response to ER stress ( Figure 5G ) . In mammals , XBP-1 may regulate its own expression [37] . Our ChIP analysis indicated that SKN-1 also binds to its own locus with ER stress ( Figure 5H ) , suggesting that SKN-1 , XBP-1 , and ATF-6 together regulate skn-1 transcription . ER stress also resulted in XBP-1 and ATF-6 recruitment to the direct SKN-1 targets pcp-2 and gst-4 ( Figures S5H–S5K ) . Together , the evidence suggests that SKN-1 , XBP-1 , and ATF-6 may function together to regulate several downstream genes . We conclude that SKN-1 is transcriptionally integrated into the UPR , in which it functions upstream , downstream , and in parallel to the known core UPR transcription factors . The mammalian SKN-1 orthologs Nrf1 and Nrf3 have been detected in association with the ER ( see Introduction ) , raising the question of whether this might also be true for a proportion of SKN-1 . Consistent with this idea , Nrf1 and the SKN-1a isoform each contain a predicted transmembrane domain [27] ( Figure S6A ) . To investigate whether SKN-1 might be present at the ER , we asked whether it might be detected in association with the ER-resident chaperone BiP ( HSP-3/-4 ) ( Figure S1A ) . We performed co-immunoprecipitation ( IP ) analyses of intact worms that had been crosslinked with formaldehyde as in our ChIP experiments . These conditions capture direct and indirect in vivo interactions that occur within approximately 2 Å , and allow for high-stringency detergent and salt-based washings that minimize non-specific binding [44] , [45] . Under both normal and ER stress conditions , association between HSP-4 and SKN-1 was readily detected by high-stringency IP performed in either direction ( Figure 6A and 6B ) . As in Figure 4B , the size of this SKN-1 species suggested that it may correspond to SKN-1a . The data suggest that some SKN-1 may be produced at the ER and might remain associated with this organelle . Given that BiP has been found in other cellular locations besides the ER [46] , we also investigated whether SKN-1 is present in a cellular fraction that is enriched for the ER ( Figure S6B ) . SKN-1 was readily detectable in an ER fraction that included HSP-4 , but not the cytoplasmic protein GAPDH ( Figures 6C and 6D ) . The interaction between endogenous SKN-1 and HSP-4 was confirmed within this ER fraction by a co-IP that was performed without crosslinking ( Figure 6E ) . Together , our findings suggest that the association of SKN-1/Nrf proteins with the ER is evolutionarily conserved . Our finding that UPR factors are required for SKN-1 activity to be increased under ER stress conditions raised a related question: might UPR-related mechanisms also be involved in SKN-1 responses to oxidative stress ? Surprisingly , we found that RNAi or mutation of core UPR signaling and transcription factors ( atf-5 , pek-1 , ire-1 , hsp-4 and xbp-1 ) impaired oxidative stress ( AS ) -induced activation of several SKN-1 target genes , including skn-1 itself ( Figures 7A , 7C , and S7A ) . Similarly , ire-1 RNAi attenuated activation of the gcs-1::GFP reporter in the intestine ( Figure S7B ) . This impairment of the oxidative stress response is particularly striking because ire-1 RNAi actually increased oxidized protein levels , in contrast to the mild AS treatment conditions used for gene expression analyses ( Figure S4O ) . Importantly , oxidative stress from AS did not simply activate the canonical UPR . Many SKN-1-regulated genes that were induced by oxidative stress were not upregulated by ER stress , and vice-versa ( Figures S1C , S4M , and S7C ) . This shows that SKN-1 mobilizes distinct transcriptional responses to oxidative and ER stress , even if these responses overlap to an extent . Moreover , AS primarily increased accumulation of the unspliced xbp-1 mRNA form ( xbp-1u ) , in striking contrast to the increase in xbp-1s levels that is characteristic of ER stress ( Figures 3A and 7C ) . Treatment with the oxidative stressor tert-butyl hydrogen peroxide ( tBOOH ) induces a SKN-1-dependent response that overlaps with the AS response , but includes SKN-1-independent activation of many genes that are otherwise SKN-1-dependent [21] . Knockdown of ire-1 or hsp-4 inhibited tBOOH from upregulating skn-1 and some SKN-1 targets ( Figure 7B ) , but did not eliminate activation of other genes ( gcs-1 , sdz-8 , and gst-10; not shown ) . The data suggest that core UPR factors are needed for SKN-1 to function properly under oxidative stress conditions , in addition to the setting of ER stress . The extensive regulatory integration that exists among UPR transcription factors , as described by others and in this study ( Figures 7A , 7B , and S7A ) [8] , [15] , [37] , could explain why multiple UPR-associated signaling and transcription factors are needed for skn-1 expression to be increased in response to oxidative stress . However , we considered that the UPR might also influence SKN-1 regulation at a post-translational level . In the C . elegans intestine SKN-1 is predominantly cytoplasmic under normal conditions , but accumulates in nuclei in response to oxidative stress from AS treatment [38] . This nuclear accumulation was dramatically reduced in animals that had been exposed to ire-1 RNAi ( Figure S7D ) . The presence of SKN-1 in intestinal nuclei is dependent upon its phosphorylation by the p38 kinase , which is activated by oxidative stress [23] , [38] , [47] . The IRE-1 kinase activity transmits signals through the JNK and p38 MAPK pathways [6] , [48]–[50] , and we determined that ire-1 knockdown largely prevented the increase in p38 signaling that occurs in response to oxidative stress ( Figures 7D and S7D ) . Taken together , these data suggest that IRE-1 is required for oxidative stress to activate SKN-1 post-translationally . If UPR signaling and transcription factors are required for SKN-1 to mobilize appropriate oxidative stress responses , then oxidative stress sensitivity should be increased when these canonical UPR factors are lacking . Accordingly , RNAi or mutation of these genes significantly increased sensitivity to oxidative stress from exposure to AS , paraquat , or t-BOOH ( Figures 7E , S7E , and S7F; Table S4 ) . We conclude that signaling from the ER is required for SKN-1 to respond to oxidative stress , and therefore that UPR-mediated regulation of SKN-1 plays a central role in the homeostatic integration of ER and oxidative stress responses .
It is well-established that the canonical UPR transcription factors XBP1 , ATF4 , and ATF6 control overlapping sets of downstream genes and processes [5] , [6] , but much less is known about how their responses to ER stress might be integrated with other mechanisms that maintain cellular stress defense and homeostasis . We have determined that the oxidative/xenobiotic stress response regulator SKN-1/Nrf functions as a fourth major UPR transcription factor in C . elegans . Without SKN-1 , ER stress failed to increase the expression of core UPR signaling and transcription factors , many of which are regulated directly by SKN-1 ( ire-1 , xbp-1 , atf-5 , and hsp-4; Figures 1 , 2 , 3 and S3 ) . It was particularly striking that SKN-1 was disproportionally required for production of spliced xbp-1 mRNA ( xbp-1s ) , presumably because of its importance for IRE-1 expression ( Figures 3D–F ) . SKN-1 was also needed for ER stress to upregulate numerous genes that are known or predicted to be involved in various ER- or UPR-related processes , including ER homeostasis ( ero-1 , pdi-2 ) , chaperone-mediated protein folding ( hsp-3 , hsp-4 , dnj-28 , T05E11 . 3 ( HSP-90/GRP94 ) ) , autophagy ( lgg-1 , lgg-3 ) , calcium homeostasis ( sca-1 , crt-1 ) , ER membrane integrity ( ckb-4 ) , and a pathway that defends against ER stress when the canonical UPR is blocked ( abu-8 , abu-11 [51] ) ( Figure 1 , 3G and Table S1 ) . Together , our data indicate that SKN-1 regulates transcription of essentially the entire core UPR apparatus and many downstream ER stress defense genes in vivo . We were surprised to find that SKN-1 was so broadly important for UPR transcription events . A trivial explanation for our findings would be that skn-1 mutants did not need to induce the UPR robustly because they were resistant to ER stress . This explanation was ruled out , however , by our finding that skn-1 mutants are actually sensitized to ER stress from diverse sources ( Figures 3H and S4L ) . Importantly , our ChIP studies and MOD-ENCODE data [34] indicate that SKN-1 controls many core and downstream UPR genes directly by binding to their promoters ( Figures 2 , 3 , and S3E , Table S1 ) . We also found that ER stress induces SKN-1 , XBP-1 , and ATF-6 to bind promoters directly to regulate many of the same genes , including skn-1 itself ( Figures 5 , S3 , and S5 ) . In addition , under ER stress conditions , UPR signaling increased levels of skn-1 mRNA and protein ( Figures 4 and S4 ) , indicating that SKN-1 is controlled by the UPR and is an active participant in this response . Together , our data reveal that a remarkable degree of regulatory and functional integration exists between SKN-1 and the three canonical UPR transcription factors ( Figures 7F and S1A ) . Although ER stress increases skn-1-dependent transcription and SKN-1 occupancy at several downstream gene promoters , it did not detectably alter the overall levels of SKN-1 in intestinal nuclei , at least as indicated by levels of a transgenic GFP fusion protein ( Figure S4N ) . While this might seem paradoxical , we observed a similar situation with reduced TORC1 signaling [19] . Under conditions of low TORC1 activity SKN-1 target genes were activated in a skn-1-dependent manner , and this was accompanied by increased SKN-1 binding to their promoters , but not by an obvious increase in the bulk levels of SKN-1 in nuclei . Our finding that SKN-1 binds to downstream UPR genes together with other UPR transcription factors suggests a paradigm that could explain this phenomenon . If SKN-1 binds cooperatively with UPR factors or other co-regulators to some of its targets , this could shift the binding equilibrium to allow those targets to be activated by SKN-1 that is already present in the nucleus , without it being necessary to “flood” the nucleus with higher levels of SKN-1 . This scheme might be important for fine-tuning of SKN-1 downstream functions , and for allowing SKN-1 to activate different targets in different situations , as we have observed in this study . In performing these analyses , we were mindful of the concern that the involvement of SKN-1 in the UPR might derive from its possible role in a secondary oxidative stress response . Several lines of evidence argued against this interpretation . For example , the direct involvement of SKN-1 in regulating multiple core UPR signaling and transcription factors during the UPR ( Figures 3 and S3 ) is not consistent with its UPR functions deriving simply from a secondary oxidative stress response . Moreover , under our ER stress conditions SKN-1 was required for accumulation of the spliced form of the xbp-1 mRNA , whereas oxidative stress increased levels of the unspliced xbp-1 message ( Figures 3A , 3B , and 7C ) . It was particularly striking that SKN-1 defended against reductive ER stresses ( Figures 4C , 4D , S4J , S4K , and S4L ) , given the extensively described role of SKN-1/Nrf proteins in oxidative stress responses . These last observations indicated that SKN-1 defends against ER stress per se , and not only against oxidative conditions . Importantly , ER stress and the UPR directed SKN-1 to activate some of its target genes that are induced by oxidative stress , but not others ( Figure S1C and S4M ) . On the other hand , many genes that SKN-1 activated under ER stress conditions were not induced by oxidative stress ( Figure S7C ) . Taken together , the data show that SKN-1 does not simply activate oxidative stress defenses in the context of ER stress , but orchestrates a specific transcriptional ER stress response that is integrated into the broader UPR . Our finding that SKN-1 mobilizes overlapping but distinct responses to ER and oxidative stress defines a new function for this surprisingly versatile transcription factor . It also supports our model that SKN-1/Nrf proteins do not control the same genes under all circumstances , but instead induce protective responses that are customized to the challenge at hand [19] , [26] . The idea that SKN-1 works together with canonical UPR transcription factors at downstream genes may provide a model for understanding how particular SKN-1 functions can be mobilized under different conditions , if these proteins and other SKN-1 “partners” guide its activities . Consistent with reports that Nrf1 and Nrf3 are present at the ER [27]–[30] , we found that some SKN-1 also localizes to the ER . We detected association between SKN-1 and the ER chaperone HSP-3/4 ( BiP ) in crosslinking analyses of intact animals , the presence of SKN-1 within an ER fraction , and association between SKN-1 and HSP-3/4 within that fraction ( Figure 6 and S6 ) . Each of these experiments involved analysis of endogenous proteins . These strategies would have detected either direct or indirect interactions , so they do not demonstrate that SKN-1 binds directly to HSP-3/4 ( BiP ) , but they do show that these proteins reside very close to each other at the ER . Apparently , association between SKN-1/Nrf proteins and the ER is evolutionarily conserved . The example of ATF-6 , which is activated through cleavage in the Golgi ( Figure S1A ) , predicts that ER-associated SKN-1 might have a signaling function in which it is cleaved in response to ER stress . However , the relative instability of SKN-1 and the presence of smaller isoforms have so far confounded the resolution of this question ( not shown ) . We recently determined that some SKN-1 also localizes to mitochondria and that SKN-1 can promote a starvation-like state when overexpressed , a function that also appears to be conserved in Nrf proteins [26] . Given the extensive communication between the ER and mitochondria [4] , [52] , our results suggest that SKN-1/Nrf might respond directly to the status of each of these organelles . Consistent with this notion , SKN-1 is required for expression of the C . elegans ortholog of mitofusin ( fzo-1 ) ( Figure 1A ) , which mediates mitochondrial fusion and mitochondria-ER interactions [4] . Taken together , our findings show that processes controlled by SKN-1/Nrf proteins are critical for ER stress defense and homeostasis , and that SKN-1 is extensively intertwined with the UPR in vivo . While differences could exist between C . elegans and mammals with respect to regulatory networks , the extent of the functional interactions we have observed predicts that mammalian Nrf proteins are likely to play an important role in the UPR that is distinct from their familiar function in oxidative stress responses . Perhaps our most surprising finding was that core UPR signaling and transcription factors were required for SKN-1 to mount a transcriptional response to oxidative stress ( Figures 7 and S7 ) . Cooperative interactions between SKN-1 and UPR transcription factors could account for some of these findings , through their effects on SKN-1 expression , but it was striking that ire-1 was needed for AS to induce SKN-1 nuclear accumulation , a phenomenon that does not occur under ER stress conditions ( Figures S4N and S7D ) . Moreover , ire-1 was required for the AS-induced p38 signal that is needed for SKN-1 to be present in nuclei ( Figure 7D ) . These last findings indicate that IRE-1 affects the oxidative stress response at a step upstream of SKN-1 . One speculative possibility for further investigation is that the IRE-1 kinase activity might be needed to initiate the oxidative stress-induced p38 signal . Together , our data show that signaling from the ER is required to “license” the oxidative/xenobiotic stress response , and suggest that the ER might function in effect as a stress sensor . This importance of the UPR for SKN-1 activity may have implications for our understanding of aging and longevity assurance . SKN-1/Nrf not only defends against resistance to various stresses , but is also important in pathways that affect longevity , including insulin-like , TORC1 , and TORC2 signaling , and dietary restriction [16] , [17] , [19] , [20] . IRE-1 and XBP-1 have each been implicated in longevity [53] , [54] , making it important to determine the extent to which these UPR-based mechanisms might influence aging through regulation of SKN-1/Nrf and its functions . Why would such extensive integration have arisen , in which SKN-1/Nrf is essential for the UPR , and signaling from the ER is needed for SKN-1/Nrf activities that are distinct from the UPR ( Figure 7F ) ? SKN-1/Nrf controls cellular processes that profoundly influence the ER . Its target genes drive synthesis of glutathione , the major redox buffer within the ER , and encode many endobiotic and xenobiotic metabolism enzymes that reside on or within the smooth ER ( Table S1 ) [20] , [21] , [55] . Under some circumstances SKN-1/Nrf also regulates proteasome expression and activity , and numerous chaperone genes [20] , [21] , [23]–[25] . One possibility is that the influence of SKN-1 could attune the UPR to events taking place in the cytoplasm . It might be advantageous to mount a robust transcriptional UPR if the cytoplasm is under duress , for example , and to moderate the UPR when cytoplasmic stress is low . Under these conditions , SKN-1 activity would be relatively high and low , respectively . SKN-1 activity is also comparatively low when translation rates are high [19] , [23] . If the ER becomes stressed under growth conditions it might be useful to limit the transcriptional UPR initially , because a reduction in translation rates might largely suffice to restore homeostasis . Again , under these conditions low SKN-1 activity could act as a brake on the transcriptional UPR . With respect to the oxidative/xenobiotic stress response , it could be important for the ER to have a “vote” on its intensity , given the profound influence of SKN-1/Nrf on cellular redox status and resources devoted to the ER . It seems likely , therefore , that the ER not only manages its own homeostasis , but through SKN-1/Nrf has a broader impact on cellular stress defense networks that is likely to be critical in their normal and pathological functions .
For each condition studied , RNA was extracted from approximately 100 µl of packed mixed-stage worms that were collected in M9 at the indicated time point . To induce UPR-associated gene expression , at day three of adulthood worms were treated with 5 µg/ml TM ( Sigma ) for 16 hours [15] , or at day four with 5 mM DTT ( Sigma ) [54] for two hours , 5 µM thapsigargin ( Enzo ) [56] for two hours , or 5 µM Bortezomib ( proteasome inhibitor , LC Labs ) for six hours ( similar to published C . elegans MG132 proteasome inhibitor treatment [57] ) . In each case , these treatments were non-lethal . For arsenite ( AS ) and tBOOH exposure , up to 100 µl of packed worms were collected and nutated in 5 mM AS or 12 mM tBOOH for 1 hour ( a non-lethal duration ) . Each of these treatments was performed in a volume of 1 ml , and was followed by pelleting . RNA was analyzed by qRT-PCR as described , with values normalized to an internal standard curve for each amplicon [19] , [44] . The same treatment conditions were used for ChIP experiments . Expression or nuclear accumulation of transgenic GFP proteins was scored as “low , ” “medium , ” or “high” essentially as published [19] , or were quantified using ImageJ 1 . 45S . ChIP was performed essentially as described [19] , [44] . 2 ml of packed mixed-stage worms were crosslinked with formaldehyde at room temperature for 20 minutes . After quenching , lysis , and determination of protein concentration , 1 mg/ml samples were frozen as aliquots at −80°C . The resolution of the assay was approximately 250–500 bp [44] . The monoclonal antibody FC4 [58] was used for SKN-1 ChIP experiments , as in previous ChIP analyses [19] . Other antibodies are described in the Supplemental Experimental Procedures . Analyses of intergenic regions and control genes ( not shown ) indicated that average signals of 14% , 11% , 26% , 4% , 11% , 7% , and 8% represent thresholds for specific presence of SKN-1 , Pol II , PSer2 , and H3-AcK56 , XBP-1 , ATF-6 , and Histone H3 respectively . Worms from five confluent 20 cm2 plates were collected in M9 with or without TM treatment ( 5 µg/ml ) for 16 hours , in order to generate 2× 1 ml of packed mixed-stage animals . Worms were sonicated 3× for 20 seconds in homogenization buffer ( supplied by IMGENEX kit , supplemented with HDAC inhibitors , protease inhibitors , phosphatase inhibitors , and MG132 ) with the Branson midiprobe 4900 Sonifer before fractionation with the IMGENEX Endoplasmic Reticulum Enrichment Kit ( Cat No . 10088K ) [59] . Mitochondrial and ER fractions were washed 3× with 1 ml PBS and resuspended in 400 µl PBS ( supplemented with HDAC , protease , and phosphatase inhibitors and MG132 ) . Up to 100 µl of the ER or cytoplasmic fractions were used for each IP . Controls for a polyclonal rabbit antiserum raised against SKN-1c ( JDC7 , referred to as pSKN-1 ) are shown in Figures S4F–S4J . HSP-3/4/BiP was detected with either C-terminal Drosophila Hsc3 [60] ( Figures 6A and 6B ) or N-terminal human BiP antibody ( Sigma et21 ) [61] , [62] ( Figures 6C and 6E ) . Note that both BiP antibodies recognized the same 75 kD band . ATF-6 ( Abcam ab11909 ) , Tubulin ( Sigma #9026 ) , and GAPDH ( Santa Cruz sc25778 ) antibodies were also used . Phosphorylated p38 was detected using an antibody from Cell Signaling T180/Y182 as described previously [23] . For Western blotting , antibodies were used at the following dilution: 1∶200 FC4 monoclonal αSKN-1 , 1∶200 polyclonal αSKN-1 , 1∶1000 αPol II , and 1∶1000 for αHsc3 . All other antibodies were used at manufacturer's recommended concentrations . For IPs , the indicated antibodies ( 50 µl FC4 monoclonal αSKN-1 or polyclonal αSKN-1 , 10 µl Hsc3 ( BiP ) or 20 µl BiP ( Sigma ) ) and pre-blocked Salmon Sperm DNA/Protein A beads ( Zymed ) were added to lysates or samples from the fractionation described above . The final volume was brought to 500 µl in 1× PIC , 1× PMSF , and 1∶1000 MG132 diluted in 1× PBS . Samples were nutated overnight at 4°C and washed three times for 5 minutes at 4°C the next day with NP-40 wash buffer . Beads were spun down at 3000 rpm and resuspended in 4× SDS Laemmli Buffer . Samples were boiled for 15 minutes with 20 µl β-mercaptoethanol and 50 µl 4× SDS Laemmli . Samples were loaded ( 50 µl each ) onto NuPAGE Novex Bis-Tris 10% Gels . Pierce ECL or Femto Western Blotting Substrate was used for detection . Other methods are available in Text S1 ( Supplementary Materials and Methods ) . | Proteins that are placed in membranes or secreted are produced in a cellular structure called the endoplasmic reticulum ( ER ) . An accumulation of misfolded proteins in the ER contributes to many disease states , including diabetes and neurodegeneration . The ER protects against a toxic buildup of misfolded proteins by activating the unfolded protein response ( UPR ) , which maintains ER homeostasis by slowing protein synthesis and enhancing ER functions such as protein folding and degradation . Many of these processes are controlled by three canonical ER/UPR gene regulatory factors . Here we identify the gene regulator SKN-1/Nrf as also playing a critical role in the UPR . SKN-1/Nrf is well known for its functions in oxidative stress defense and longevity . We now report that SKN-1/Nrf mobilizes an ER stress gene network that is distinct from its oxidative stress response , and includes regulation of other central UPR factors . Surprisingly , we also find that ER- and UPR-associated mechanisms are needed to “license” SKN-1/Nrf to defend against oxidative stresses . Our findings show that UPR and oxidative stress defense mechanisms are integrated through SKN-1/Nrf , and suggest that this integration may help maintain a healthy balance between ER and cytoplasmic functions , and stress defenses . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2013 | Integration of the Unfolded Protein and Oxidative Stress Responses through SKN-1/Nrf |
Cutaneous leishmaniasis due to L . braziliensis ( CL ) is characterized by a positive delayed type hypersensitivity test ( DTH ) leishmania skin test ( LST ) and high IFN-γ production to soluble leishmania antigen ( SLA ) . The LST is used for diagnosis of CL and for identification of individuals exposed to leishmania infection but without disease . The main aim of the present study was to identify markers of exposure to L . braziliensis infection . This cohort study enrolled 308 household contacts ( HC ) of 76 CL index cases . HC had no active or past history of leishmaniasis . For the present cross-sectional study cytokines and chemokines were determined in supernatants of whole blood culture stimulated with SLA . Of the 308 HC , 36 ( 11 . 7% ) had a positive LST but in these IFN-γ was only detected in 22 ( 61 . 1% ) . Moreover of the 40 HC with evidence of IFN-γ production only 22 ( 55% ) had a positive LST . A total of 54 ( 17 . 5% ) of 308 HC had specific immune response to SLA . Only a moderate agreement ( Kappa = 0 . 52; 95% CI: 0 . 36–0 . 66 ) was found between LST and IFN-γ production . Moreover while enhancement of CXCL10 in cultures stimulated with SLA was observed in HC with DTH+ and IFN-γ+ and in patients with IFN-γ+ and DTH− , no enhancement of this chemokine was observed in supernatants of cells of HC with DTH+ and IFN-γ− . This study shows that in addition of LST , the evaluation of antigen specific IFN-γ production should be performed to determine evidence of exposure to leishmania infection . Moreover it suggests that in some HC production of IFN-γ and CXCL10 are performed by cells not involved with DTH reaction .
American tegumentary leishmaniasis ( ATL ) is caused predominantly by Leishmania braziliensis , L . guaynensis , L . mexicana and L . amazonensis [1] , [2] . It is endemic in South and Central America and cutaneous leishmaniasis ( CL ) , characterized by well delimited ulcers with raised borders , is the most common clinical picture of ATL . The main characteristics of the immunological response in CL are a strong Th1 type immune response to soluble leishmania antigen ( SLA ) , demonstrated by positive delayed type hypersensitivity ( DTH ) reaction to the leishmania skin test ( LST ) , as well as lymphocyte proliferation and production of high levels of IFN-γ and TNF-α [3] , [4] . Only a few parasites are found in the lesions due to L . braziliensis and because of this the leishmania skin test ( LST ) is widely used for diagnosis of ATL . A positive test in a patient with a typical cutaneous lesion has a high predictive value [5] , [6] . The LST has been also used to measure exposure to leishmania infection , and a positive LST in the absence of clinical manifestations of ATL has been documented in up to 17% of healthy individuals living in endemic areas of L . braziliensis [6] . Individuals with a positive LST who do not develop leishmaniasis are considered as having a subclinical L . braziliensis infection [7] . Although a concordance between DTH and in vitro tests of cell mediated immune response is expected , discordant results between IFN-γ production and tuberculin skin test ( TST ) have been shown in individuals with latent tuberculosis [8]–[10] . Therefore it is important to determine whether this discordance also occurs in subclinical L . braziliensis infection as well as to evaluate if other tests may be indicative of exposure to leishmania infection . As the ratio from infection to disease based on LST is 3 . 7 to 1 , about 25% of individuals who are exposed to L . braziliensis will develop cutaneous leishmaniasis [7] . It is known that early events after penetration of leishmania in the skin are important to determine the outcome of leishmaniasis . Therefore characterization of immune response early after the infection is highly relevant . Moreover , an early detection of individuals exposed to L . braziliensis will allow a comparative analysis between individuals who will develop or not develop disease . The aim of this study was to establish and follow a prospective cohort of household contacts of CL patients , to evaluate initially markers of exposure to leishmania infection and ultimately to identify markers that are associated with resistance or susceptibility to develop disease . Our initial evaluation of this cohort indicates based in a cross-sectional study that more than one test is needed to determine exposure to L . braziliensis . In addition to LST , IFN-γ production in SLA stimulated cultures should be determined . Moreover , the production of CXCL10 , a chemokine associated with recruitment and activation of T cells , gives support that in some cases CXCL10 and IFN-γ are produced by cells not involved with DTH .
This study was approved by the Ethical Committee of the Federal University of Bahia . Written informed consent was obtained from all enrolled subjects . This study was conducted in Corte de Pedra , a rural region in Northeastern Brazil endemic for ATL , where we have performed clinical and immunological studies for over 25 years [3] , [11] , [12] . The area was previously dominated by Atlantic rainforest and is now a mostly deforested agricultural community . Lutzomyia whitmani and Lu . intermedia sandflies that transmit L . braziliensis are endemic in the local fauna [2] , [13] . The health post of Corte de Pedra was created in 1986 and is a reference center for diagnosis and treatment of CL and is staffed by medical personnel from the Federal University of Bahia . This is a cohort study enrolling household contacts ( HC ) of patients with history of CL . Index cases ( IC , N = 76 ) were recruited at the Corte de Pedra health post . An index case was defined as a patient with confirmed CL diagnosed at the Corte de Pedra health post within two years prior to enrollment in the study , living within a 10 km radius of the health post . Patients with evidence of mucosal or disseminated leishmaniasis were not considered for enrollment as index cases . Researchers visited index cases in their homes to recruit HC from January to April 2010 . Household contacts ( N = 533 ) were defined as individuals without history of any type of leishmania infection who were living in the same home as the index case at the time of enrollment in the study and at the time of diagnosis of CL by the index case . Cutaneous leishmaniasis ( CL ) patients were diagnosed based on a typical clinical leishmaniasis lesion , associated with a positive leishmania skin test ( LST ) and documentation of parasites in culture or by histopathology . After obtaining informed consent , a negative history of CL in HC was established by a medical interview , assessing for symptoms consistent with previous CL or ML infection , and negative physical exam looking for scars consistent with past CL or ML on the skin , nose and soft palate . Exclusion criteria for HC were age less than 2 or more than 65 years , or frequent stays outside of the endemic area . After the initial immunologic studies a cross-sectional study was performed first comparing epidemiological and chemokine data among HC with evidence or without evidence of immune response to soluble leishmania antigen ( SLA ) . Second , comparing cytokine data among groups according evidence of IFN-γ production and response to LST . SLA for the skin test was prepared with an isolate of L . braziliensis as previously described [14] . Briefly , promastigotes of L . braziliensis were grown in Schneider's medium supplemented with 10% fetal bovine serum and 2% human urine . The promastigotes were washed in sterile phosphate buffered saline ( PBS ) , resuspended in lysis solution ( Tris HCl , EDTA and leupeptin ) , immersed in liquid nitrogen , and thawed at 37°C . After freeze-thaw procedure , they were sonicated . The disrupted parasites were centrifuged at 14 , 000G and assayed for total protein ( BCA Protein Assay Kit , Thermo Scientific ) . For in vitro testing the filtrate was adjusted to a concentration of 500 µg/mL with sterile PBS . For the LST , the filtrate was adjusted to a concentration of 250 µg/mL with sterile PBS containing Tween 80 and phenol at final concentrations of 0 . 0005% ( w/v ) and 0 . 28% ( w/v ) respectively . Once a negative history of CL was established , heparinized peripheral blood ( 10 mL ) was collected immediately before LST was performed . About 6 h after collection , 1 mL of whole blood aliquots was dispensed into a 24-well tissue plate . SLA at 20 µg/mL , 50 µl of phytohemaglutinin ( GIBCO , Grand Island- NY ) as positive control or medium ( negative control ) were added to each well and incubated at 37°C 5% CO2 for 72 hours . Plasma supernatants ( 300–400 uL ) were collected , and samples were stored at −20°C . The levels of IFN-γ released were quantified by ELISA , using commercially available reagent ( BD OpTEIA , San Diego-CA ) . A standard curve was used to express the results in pg/ml . A positive test was defined as any detectable IFN-γ level after subtracting the SLA stimulated IFN-γ levels by unstimulated IFN-γ levels . The levels of CXCL9 , CXCL10 , and CCL2 were measured by ELISA using commercially available reagents ( BD OpTEIA , San Diego-CA ) . A standard curve was used to express the results in pg/ml . Chemokine data are summarized separately as spontaneous production ( medium ) and SLA-induced production ( Ag ) . The LST was performed after collection of blood for the in vitro test to avoid the possible influence of the skin test reaction on the immune response determined in vitro . SLA for the skin test was prepared with an isolate of L . braziliensis as previously described [14] . For LST , 0 . 1 mL ( 25 µg/mL ) of the SLA was injected intracutaneously on the volar surface of the forearm , and the greater diameter of induration was measured 48–72 h later . Induration of ≥5 mm was defined as a positive reaction . HC were considered to have been exposed to leishmania if they demonstrated a positive immune response to L . braziliensis antigen in either the in vitro and in vivo test . The analysis of the concordance between LST and IFN-γ production was performed by calculating a κ statistic for agreement with 95% confidence interval . Demographic characteristics were compared across subject groups as follows: For continuous variables , one-way ANOVA or Kruskal-Wallis tests were used and if the overall P-value was <0 . 05 , pair-wise Bonferonni or Dunn's post-hoc tests were performed . For categorical variables , chi-square or Fisher's exact test were performed . Student T test was used to compare the means of variables following normal distribution . For analysis of IFN-γ and chemokines production in unstimulated and stimulated culture , the Wilcoxon rank-sum test was used . Correlations were performed by the method of Spearman . STATA version 11 ( College Station , TX ) and GraphPad InStat3 ( La Jolla , CA ) statistical software were used for all the analyses .
Figure 1 depicts the design of the cohort study and distribution of study subjects . A positive LST was observed in 36 ( 11 . 7% ) of the 308 HC tested and IFN-γ production was detected in 40 HC ( 12 . 9% ) . However in the 36 individuals with positive LST , production of IFN-γ was only observed in 22 ( 61 . 1% ) , and in the 40 HC with evidence of production of IFN-γ , only 22 ( 55% ) had a positive LST . Considering evidence of immunological response to leishmania antigen as a positive LST and/or IFN-γ production in cultures stimulated with SLA , 54 ( 17 . 5% ) of 308 HC had specific immunological responses to SLA . IFN-γ production was documented in supernatants from cells of all HC stimulated with phitohemaglutinin . The demographic and epidemiological aspects of index cases and HC with and without evidence of immune response to SLA are shown in Table 1 . There was no difference among the index cases and HC with evidence of immune response in all variables analyzed ( age , gender , occupation , years living in the area as well as in the same house ) . There was also no difference among the three groups regarding the time of arriving at home . However , differences were found between HC with and without evidence of exposure to leishmania infection regarding age , occupation , time living in the endemic area and time living in the same house . Household contacts without evidence of immune response were younger and the majority was students; consequently , they had less time living in the same house and in the same endemic area of index cases than HC with evidence of immune response . In order to determine the best test to detect exposure to leishmania infection , we compared the ability of the LST and IFN-γ production to identify exposure to L . braziliensis . When assessing the concordance between LST and IFN-γ production , among those who had at least one test positive , 61 . 1% of the HC were positive for both tests and there was moderate agreement beyond that expected by chance ( Kappa = 0 . 49; 95% CI: 0 . 34–0 . 64 ) . When the two tests were analyzed in the whole population a moderate concordance between IFN-γ and LST ( Kappa = 0 . 52; 95% CI: 0 . 36–0 . 66 ) was confirmed . As chemokines are produced early in leishmania infection and they participate in the immune response by activation and recruitment of T cells , expression of chemokines related to lymphocyte and monocyte recruitment were determined in all individuals with evidence of immune response ( N = 54 ) and in a sub-group ( N = 51 ) of HC without evidence of immune response . These 51 HC without evidence of immune response were selected among individuals living in the same house of HC with evidence of immune response . The IFN- γ , CXCL9 , CXCL10 and CCL2 levels in IC , HC with evidence of immune response and HC without evidence of immune response are shown in Table 2 . As expected IFN-γ levels were higher in IC than in the other groups . High levels of CXCL9 and CCL2 were found in unstimulated cultures of both IC and HC with and without evidence of immune response . While there was no difference regarding the production of CCL2 in unstimulated culture , the levels of this chemokine were higher in HC with evidence of immune response in cultures stimulated with SLA when compared with the control group ( p<0 . 001 ) . With regard to CXCL9 and CXCL10 , levels of SLA induced chemokines were higher among the IC and HC with evidence of immune response than the levels observed in the control group ( P<0 . 05 ) . To evaluate if the chemokine production was associated with a positive LST or ability to produce IFN-γ in vitro upon SLA stimulation , the 105 HC who had chemokines determined were divided into 4 sub-groups: 1 ) LST positive and evidence of IFN-γ production ( LST+ IFN-γ+ ) ; 2 ) LST+ and absence of IFN-γ production ( LST+ IFN-γ− ) ; 3 ) LST negative and IFN-γ positive ( LST− IFN-γ+ ) ; 4 ) Both LST and IFN-γ negative ( DTH− LST− ) . The production of CXCL9 and CXCL10 in both spontaneous and in SLA stimulated cultures is shown in Figure 2 . The levels of CCL2 were increased in cultures stimulated with SLA in all individuals who had a positive LST or IFN-γ production as well as in HC without evidence of immune response ( data not shown ) . However , the high levels of CCL2 in cultures of individuals without evidence of immune response prevent the use of CCL2 as an indicator of evidence of immune response to SLA . An increase in the production of CXCL9 in cultures stimulated with SLA was documented in IC and HC with evidence of IFN-γ production and with positive LST ( Figure 2A ) . In the majority of HC , positivity in both tests ( LST and production of IFN-γ ) was required for a high production of CXCL9 in SLA stimulated cultures . Production of CXCL10 in unstimulated and SLA stimulated cultures is shown on Figure 2B . IC and subjects with positive LST and IFN-γ production , as well as those who were LST negative but produced IFN-γ , had increased CXCL10 in SLA stimulated cultures . However , one striking observation was that while CXCL10 production was detected in SLA stimulated cultures of HC with a negative LST but producers of IFN-γ , absence or very low production of CXCL10 was observed in SLA stimulated cultures of individuals who , despite having a positive DTH , had an absence of IFN-γ production to SLA . Moreover , there was a positive correlation between production of IFN-γ and CXCL10 in SLA stimulated culture ( Figure 3 ) .
Epidemiologic and clinical studies of ATL have focused on determining the influence of host , parasite and environmental factors on the development of the different clinical forms of leishmaniasis [15] , [7] , [11] , [16] . However , there is a lack of studies in individuals who have evidence of exposure to leishmania but may or may not develop disease . The documentation of a positive LST in the absence of current or past history of CL is the only test that has been used to identify individuals who have a subclinical form of L . braziliensis infection [6] , [7] . Herein we showed that the use of in vitro immunologic tests such as production of IFN-γ not only increases the number of individuals with evidence of exposure to leishmania infection , but we also demonstrate discordant results between LST and in vitro IFN-γ production by cells of HC . Moreover , the increase in CXCL10 production in SLA stimulated cultures was predominantly associated with IFN-γ production rather with a positive LST . The primary objective of this study was to identify individuals recently exposed to leishmania infection . A limitation of this type of study is that we cannot be sure when exposures to leishmania have occurred . However , the comparative analysis of demographic characteristics of the index cases and of the HC with and without evidence of immune response suggested that HC with evidence of immune responses more closely resembled index cases , supporting the contention that exposure to leishmania infection in this population likely occurred close to the period that the CL index cases acquired the parasite . The DTH and in vitro immunologic response determined by lymphocyte proliferation or cytokine production to recall antigens have been widely used to determine evidence of cell mediated immune response [17]–[20] , [9] . In patients with CL as well as in patients with mucosal leishmaniasis due to L . braziliensis infection there is a strong association between the positivity of LST and production of IFN-γ as well as evidence of lymphocyte proliferative response to SLA [3] , [17] , [15] . Because of the high predictive value of LST in the diagnosis of CL , this test has also been used to identify exposure to leishmania infection among healthy individuals living in areas of Leishmania sp transmission [21] , [6] , [7] . Individuals with a positive LST and absence of current or past history of leishmaniasis in areas of L . braziliensis transmission are considered as having subclinical L . braziliensis infection . We have previously shown that these individuals with a positive LST have a lower production of IFN-γ in SLA stimulated cultures than patients with CL , and in some of them even no detectable IFN-γ levels were found in supernatants of lymphocytes stimulated with SLA [7] , [16] . In the present study in addition to showing that HC that have a positive LST may not produce IFN-γ upon SLA stimulation , there were also HC who produced IFN-γ but had a negative LST . Discordance between DTH and IFN-γ production has been observed in subjects with latent tuberculosis [22] , [8] , [9] , [10] , and several factors may explain the discordance between LST and in vitro IFN-γ production: 1 ) Presence of suppressor factors in vivo that prevent the documentation of DTH; 2 ) Production of IFN-γ by non T cells; 3 ) Lack of effector or effector memory T cells but presence of memory T cells . Discordance between DTH and in vitro tests have been documented in patients with active tuberculosis as well as in individuals with subclinical L . chagasi infection [9] , [23] . In such cases malnutrition as well as the presence of soluble suppressor factors may explain the absence of response in vivo but the occurrence of response in vitro . Usually in this case restoration of the DTH test occurs after specific therapy as well as with improvement in nutritional status [24] , [25] . Regarding memory , studies in an experimental model of leishmaniasis have shown that after control of leishmania infection memory effector cells may not be found but animals remain with central memory T cells [26] . IFN-γ in patient with CL is predominantly secreted by effector T cells or memory effector T cells [27] . However , the disappearance of these cells from the peripheral blood may make the in vitro test become negative . Since in DTH tests antigens are inoculated intradermally and immune response is evaluated 48 to 72 hours after the test , there is time for memory effector T cells that remain in lymph nodes or in other tissues to migrate to the site of the injection of SLA and react in vivo to leishmania antigen . As the participants of this study were healthy , well-nourished , and likely recently exposed to L . braziliensis based on the epidemiologic data , the more likely explanation for the discordance between the LST and IFN-γ production in the LST negative IFN-γ positive subjects is the production of IFN-γ by cells not involved with DTH . Giving support to this hypothesis is our data that CXCL10 was produced in HC with evidence of IFN-γ production but not in those with only a positive LST . Different cell types such as neutrophils , NK cells and NKT cells may produce IFN-γ [28]–[31] and future studies will address this subject . This study shows that in L . braziliensis infection in addition to the LST , the documentation of exposure to leishmania antigen should be also evaluated by antigen specific IFN-γ production as it increases the evidence of exposure to leishmania infection from the 11 . 7% ( as documented by LST ) to 17% , when LST or IFN-γ were positive . The discordance between IFN-γ and LST was highlighted by the observation that production of CXCL10 was associated with IFN-γ production but not with a positive LST . | Both control of L . braziliensis infection and development of cutaneous leishmaniasis ( CL ) are dependent on the host immunological response . Due to the difficulty of finding parasites in leishmanial lesions , a delayed type hypersensitivity reaction - leishmania skin test ( LST ) , is widely used to diagnose CL . In areas of L . braziliensis transmission a positive LST is also documented in up to 18% of individuals without disease , who are considered to be putatively resistant to leishmania infection . However the mechanisms involved in the control of parasite grow is not known . The aim of this study is to identify tests that could determine in house contact of CL ( HC ) without past or current evidence of leishmaniasis exposure to leishmania infection . We found that of the 308 HC , 36 ( 11 . 7% ) had a positive LST but in these IFN-γ was only detected in 22 ( 61 . 1% ) . Moreover of the 40 HC with evidence of IFN-γ production only 22 ( 55% ) had a positive LST . Therefore at least the two tests , the LST and IFN-γ production , should be used to determine exposure to L . braziliensis . Identification of subjects exposed to leishmania infection that may or may not develop CL is highly relevant to understand pathogenesis of L . braziliensis infection . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"infectious",
"diseases",
"public",
"health",
"and",
"epidemiology",
"epidemiology",
"immunology",
"biology"
] | 2012 | IFN-γ Production to Leishmania Antigen Supplements the Leishmania Skin Test in Identifying Exposure to L. braziliensis Infection |
Kindler Syndrome ( KS ) , characterized by transient skin blistering followed by abnormal pigmentation , skin atrophy , and skin cancer , is caused by mutations in the FERMT1 gene . Although a few KS patients have been reported to also develop ulcerative colitis ( UC ) , a causal link to the FERMT1 gene mutation is unknown . The FERMT1 gene product belongs to a family of focal adhesion proteins ( Kindlin-1 , -2 , -3 ) that bind several β integrin cytoplasmic domains . Here , we show that deleting Kindlin-1 in mice gives rise to skin atrophy and an intestinal epithelial dysfunction with similarities to human UC . This intestinal dysfunction results in perinatal lethality and is triggered by defective intestinal epithelial cell integrin activation , leading to detachment of this barrier followed by a destructive inflammatory response .
Kindler Syndrome ( KS; OMIM:173650 ) is a rare , recessive genodermatosis caused by mutations in the FERMT1 gene ( C20ORF42/KIND1 ) [1] , [2] . KS patients suffer from varying skin abnormalities that occur at distinct phases of their life [3] . Skin blisters develop and disappear after birth , followed by skin atrophy , pigmentation defects and finally skin cancer . The severity of the individual symptoms varies extensively among individual patients . FERMT1 mutations are distributed along the entire gene and can give rise to different truncated Kindlin-1 proteins [4] . Interestingly , the different courses of the KS cannot be linked to mutations within specific regions of the FERMT1 gene [3] suggesting that additional environmental and/or genetic factors contribute to the disease course . Kindlin-1 belongs to a novel family of cytoplasmic adaptor proteins consisting of three members ( Kindlin-1-3 ) [5] . Kindlins are composed of a central FERM ( band 4 . 1 , ezrin , radixin , moesin ) domain , which is disrupted by a pleckstrin homology ( PH ) domain . They localize to cell-matrix adhesion sites ( also called focal adhesions , FAs ) where they regulate integrin function . In line with the role of Kindlins in integrin function , keratinocytes from KS patients and Kindlin-1-depleted keratinocytes display impaired cell adhesion and delayed cell spreading [6] , [7] . The mechanism how Kindlin-1 regulates integrin function is not understood and controversial . Kindlin-2 ( Fermt2 ) and Kindlin-3 ( Fermt3 ) were shown to bind to the membrane distal NxxY motif of β1 ( Itgb1 ) and β3 ( Itgb3 ) integrin cytoplasmic domains . This binding , in concert with Talin ( Tln1 ) recruitment to the membrane proximal NPxY motif , leads to the activation ( inside-out signaling ) of β1 and β3 class integrins enabling them to bind to their ligands . Following ligand binding , Kindlin-2 and Kindlin-3 stay in matrix adhesion sites where they link the ECM to the actin cytoskeleton by recruiting ILK and Migfilin ( Fblim1 ) to FAs ( outside-in signaling ) . Consistent with this adaptor function of Kindlins , keratinocytes from KS patients and keratinocytes depleted of Kindlin-1 display impaired cell adhesion and delayed cell spreading [6] , [7] . Importantly , however , Kindlin-1 was reported to have different properties than Kindlin-2 and -3 , since it was shown to bind like Talin to the proximal NPxY motif of β1 integrin tails [6] . The Kindlins have a specific expression pattern . Kindlin-1 is expressed in epithelial cells , while Kindlin-2 is expressed almost ubiquitously . They are both found at integrin adhesion sites and/or cadherin-based cell-cell junctions . Kindlin-3 is exclusively expressed in hematopoietic cells , where it controls a variety of functions ranging from integrin signaling in platelets [8] to stabilizing the membrane cytoskeleton in mature erythrocytes [9] . Although the FERMT1 gene is expressed in epithelial cells of almost all tissues and organs [5] , only abnormalities of the skin and the oral mucosa are associated with KS . Recently it has been reported , however , that some KS patients also develop ulcerative colitis ( UC ) [4] , [10] , [11] , which together with Crohn's disease belongs to idiopathic inflammatory bowel disease . UC usually occurs in the second or third decade of life , although the incidence in pediatric patients is steadily rising [12] , [13] . UC is restricted to the colon and is characterized by superficial ulcerations of the mucosa . It is currently believed that the mucosal ulcerations are triggered by the release of a complex mixture of inflammatory mediators leading to severe inflammation and subsequent epithelial cell destruction [12] . In line with this paradigm a large number of murine colitis models occur when the innate or adaptive immune response is altered [12] . Genetic linkage analysis in man led to the identification of several susceptibility loci [14] , [15] . In line with the UC disease course most of the KS patients develop their first UC symptoms in adulthood . Interestingly , however , one of them suffered from a severe neonatal form of UC and was diagnosed with a null mutation in the FERMT1 gene after developing trauma-induced skin blistering [10] . Since only a few KS patients were reported to develop intestinal symptoms , it is currently debated whether UC development in these patients is directly linked to FERMT1 gene mutations or secondarily caused by a microbial infection or an abnormal inflammatory response . In this study we directly investigated the role of Kindlin-1 in vivo by generating mice carrying a constitutive null mutation in the Kindlin-1 gene . We demonstrate that Kindlin-1 deficient mice develop skin atrophy and a lethal intestinal epithelial dysfunction , resembling the reported UC in KS patients . The intestinal epithelial dysfunction is caused by defective intestinal epithelial integrin activation leading to extensive epithelial detachment followed by a severe inflammatory reaction .
To unravel the consequences of loss of Kindlin-1 in vivo , we established a mouse strain with a disrupted Fermt1 gene , leading to a complete loss of Kindlin-1 mRNA and protein ( Figure 1A–C ) . Heterozygous Kindlin-1 mice ( Kindlin-1+/− ) had no phenotype and were fertile . Kindlin-1-deficient mice ( Kindlin-1−/− ) were born with a normal Mendelian ratio ( 29 . 6% +/+; 44 . 1% +/−; 26 . 3% −/−; n = 203 at P0 ) and appeared normal at birth . Two days postnatal ( P2 ) , all Kindlin-1−/− mice analyzed so far were dehydrated ( Figure 1D ) , failed to thrive ( Figure 1E ) and died between P3-P5 ( Figure 1E ) . Blood glucose and triglyceride levels of Kindlin-1−/− mice were normal suggesting a normal absorption of nutrients in the small intestine ( Figure S1 ) . Their urine showed an increased osmolarity and protein content further pointing to severe dehydration ( Figure 1F , Figure S2A and B ) . Histology of Kindlin-1−/− kidneys at P3 revealed normal morphology of glomeruli and tubular systems ( Figure S2C ) . Thus , these findings suggest that the peri-natal lethality is not caused by a renal dysfunction . Next we analyzed whether skin abnormalities caused the perinatal lethality . Although Kindlin-1−/− mice showed features of KS like skin atrophy and reduced keratinocyte proliferation ( Figure 2A and B ) , adhesion of basal keratinocytes to the basement membrane ( BM ) was unaltered ( Figure 2A and Figure S3 ) . Histology of backskin sections from different developmental stages revealed normal keratinocyte differentiation ( Figure S4 ) , normal development of the epidermal barrier ( Figure S5A and Figure 2C and D ) , and comparable epidermal thickness at E18 . 5 and P0 ( Figure S5B ) . In line with the progressing proliferation defect quantified by the number of keratinocytes positive for the cell cycle marker Ki67 ( Mki67 ) , a reduction of the epidermal thickness was first observed at P1 ( Figure S5B ) . Interestingly despite the mild in vivo phenotype , Kindlin-1−/− keratinocytes displayed severe adhesion and spreading defects in culture ( Figure S6A and B ) further indicating that Kindlin-1−/− keratinocytes from mouse and man display similar defects [7] . These results indicated that another defect is responsible for their perinatal lethality . When the stomach and the intestine of Kindlin-1−/− mice were examined , they were swollen and filled with milk and gas ( Figure 3A ) suggesting severe intestinal dysfunction might be the cause of death . The presence of milk in the stomach together with the normal histology of the oral and esophageal mucosa suggested that the lethality is not caused by impaired feeding ( Figure S7 ) . At P2 , the terminal ileum and colon were shortened and swollen and strictures were evident in the distal colon , which are signs of acute inflammation ( Figure 3B ) . By P3 , when the majority of mutant mice were dying , more than 80% of the colonic epithelium was detached ( Figure 3C and Figure S8A–C ) , became apoptotic ( Figure S9 ) and infiltrated by macrophages , granulocytes ( Mac-1; Itgam staining ) and T cells ( Thy staining ) ( Figure 3D ) . The shortened colon was neither a consequence of increased apoptosis , which was only seen in detached epithelium , nor a result of reduced intestinal epithelial cell ( IEC ) proliferation ( Figure S9 ) . The epithelial detachment and severe inflammation extended into the ileum ( Figure 3C ) . In contrast , the proximal small intestine ( duodenum and jejunum ) had no evidence of IEC detachment or inflammation ( Figure 4 ) . The phenotype of Kindlin-1−/− mice for the most part phenocopied the intestinal disease observed in the patient with a complete loss of Kindlin-1[10] . To define the cell type affected by loss of Kindlin-1 we localized Kindlin-1 in the normal intestine by immunostaining . Similar to the situation in man [4] , Kindlin-1 is present throughout the cytoplasm of IECs of the colon and at the basolateral sites of both IECs of the colon ( Figure 5A ) and the small intestine ( Figure 5B ) . The anti-Kindlin-1 polyclonal antibody produced some weak unspecific background signals in the intestinal mesenchyme of wild type and Kindlin-1−/− mice ( Figure 5B ) . Kindlin-2 was exclusively found in cell-cell contacts and did not change its distribution in the absence of Kindlin-1 ( Figure 5C and D ) . Focal adhesion ( FA ) components such as Talin and Migfilin as well as filamentous actin ( F-actin ) were expressed normally in Kindlin-1−/− colonic epithelium that was still adhering to the BM ( Figure 5C and data not shown ) . Next we determined the time point when mutant mice began developing intestinal abnormalities . At E18 . 5 the ileum and colon of Kindlin-1−/− mice were histologically normal and electron microscopy revealed an intact epithelium and basement membrane ( BM ) ( Figure 6A ) . Shortly after birth ( P0 ) , wild-type and Kindlin-1−/− mice began to suckle and accumulated milk in their stomachs . Within the first hours after birth nursed Kindlin-1−/− mice contained colostrum in the intestinal lumen and displayed extensive epithelial cell detachment ( Figure 6B; see Colon P0 ) without infiltrating immune cells ( Figure 6B and C ) in the distal colon . No epithelial detachment occurred when Kindlin-1−/− mice were delivered by Caesarean section and incubated in a heated and humidified chamber for up to 7 hours ( Figure 6B Colon CS ) indicating that mechanical stress applied by stool caused IEC detachment . However , inflammatory infiltrates were clearly visible between the detached epithelial cells and the underlying mesenchyme at around 12 hours after birth in fed mice ( Figure 6B; see Colon P0 . 5 ) and further increased during the following day ( Figure 6D ) . The steady immune cell infiltrate was accompanied with increased expression of the proinflammatory cytokines TNF-α ( Tnf ) and IL-6 ( Il6 ) and a reduction in goblet cell mucins ( Figure 6D and E ) . The wide range of TNF-α and IL-6 expression levels in Kindlin-1−/− mice likely reflects the different severities of inflammation at the time tissues were prepared . Although inflammation extended into the ileum at P2 and P3 ( Figure 3B and C ) , the epithelial cells of the ileum were never detached suggesting that Kindlin-1−/− mice develop a so-called backwash ileitis caused by stool “washed back” from the colon into the ileum [13] . These analyses revealed that the epithelial detachment begins at P0 in the distal parts of the colon and subsequently expands proximally . An important question is how Kindlin-1 deficiency leads to detachment of intestinal epithelial cells . One potential explanation could be loss of support by a disrupted BM as reported for skin of KS patients [1] , [2] , [3] . Moreover , it is known that BM digestion and epithelial detachment in inflammatory bowel disease ( IBD ) can be triggered via the secretion of MMPs by epithelial and/or infiltrating immune cells [17] . This possibility could be excluded , since Kindlin-1−/− mice at P1 showed a continuous BM with all major components present , both in areas of the colon where IECs were still adherent as well as in areas where IECs were detached ( Figure 7A ) . Interestingly , also the skin of Kindlin-1−/− mice showed a normal BM distribution by immunostaining ( Figure S3 ) . An alternative explanation for the IEC detachment could be a reduction of integrin levels , or a dysfunction of integrins , similar to that reported for Kindlin-3-deficient platelets [8] and Kindlin-2-deficient primitive endoderm [18] . The normal distribution of β1 integrin ( Figure 7B ) and the comparable levels of β1 and αv ( Itgav ) integrins ( Figure 7C and D and data not shown ) excluded a defect in expression and/or translocation of integrins to the plasma membrane . Flow cytometry of primary IECs with the monoclonal antibody 9EG7 , which recognizes an activation-associated epitope on the β1 integrin subunit , showed significantly reduced binding ( Figure 8A ) suggesting that loss of Kindlin-1 decreases activation ( inside-out signaling ) of β1 integrins . Primary keratinocytes from Kindlin-1−/− mice also showed normal localization ( Figure S10A ) and surface expression of β1 integrins ( Figure S10B ) . Interestingly , 9EG7 staining revealed reduced , although not statistically significant , activation of β1 integrins in these cells ( Figure S10C ) . Since it is difficult to culture and maintain primary murine IECs , we depleted Kindlin-1 in a human colon carcinoma cell line ( HT-29 ) using RNAi ( HT-29siKind1; Figure 8B ) to show that integrin-mediated cell adhesion and shear stress induced detachment were also perturbed in a colon cell line . HT-29siKind1 cells were unable to adhere to Fibronectin ( FN; Fn1 ) and showed strongly reduced adhesion to Laminin-332 and Collagen IV ( Figure 8C ) and easily detached from FN upon exposure to low ( 0 , 5dyn/cm2 ) as well as high shear stress ( up to 4dyn/cm2; Figure S11 ) . The remaining adhesion to Laminin-332 and Collagen IV is likely mediated by other Laminin- and Collagen-binding receptors on colonic epithelial cells such as α6β4 integrins ( Itga6 , Itgb4 ) and discoidin domain receptors [19] , [20] , which are both known to function independent of Laminin and Collagen binding β1 integrins [21] . These findings indicate that ( i ) loss of Kindlin-1 impairs integrin activation , which compromises adhesion of colonic epithelial cells , that ( ii ) Kindlin-2 cannot rescue Kindlin-1 loss in colonic epithelial cells , and that ( iii ) the residual IEC adhesion to Laminin and Collagen is suspended by shear forces exerted for example , by the feces . It has been reported that Kindlin-1 associates with the membrane proximal NPxY motif of β1 and β3 integrins [6] . This observation , however , is in contrast with observations made with Kindlin-2 and -3 , both of which bind the membrane distal NxxY motifs of β1 and β3 integrins to trigger their activation [8] , [18] , [22] . To explore the mechanism whereby Kindlin-1 induces integrin activation , we performed pull down experiments with recombinant GST-tagged cytoplasmic β integrin tails in IEC and keratinocyte lysates . The results confirmed that Kindlin-1 associated with the cytoplasmic domains of β1 and β3 ( Figure 8D ) . Since substitutions of the tyrosine residues in the proximal NPxY motifs with alanine ( β1Y788A; β3Y747A ) allowed Kindlin-1 binding , while tyrosine to alanine substitutions in the distal NxxY motif of β1 and β3 integrin tails ( β1Y800A; β3Y759A ) abolished Kindlin-1 binding , we conclude that the binding and functional properties are conserved among all Kindlins . This was further confirmed with direct binding assays , which showed that the recombinant His-tagged C-terminal FERM domain of Kindlin-1 ( aa 471–677 ) containing the phosphotyrosine binding ( PTB ) motif bound GST-tagged β1 but not the Y800A mutated β1 integrin cytoplasmic tail ( Figure 8E ) . It is well established that Talin can induce activation of integrins , and for a long time it was believed that it is sufficient for the execution of this task . This important function of Talin was discovered by overexpressing Talin or its FERM domain in CHO cells stably expressing the human platelet integrin αIIbβ3 ( Itga2b , Itgb3 ) [23] , which shifted the inactive conformation of αIIbβ3 integrin to a high affinity state , as demonstrated by increased binding of the PAC1 antibody recognizing activation associated epitopes on αIIbβ3 integrin ( Figure 8F ) . In contrast to Talin , overexpression of Kindlin-1 failed to trigger activation of αIIbβ3 integrin in these cells ( Figure 8F ) . Interestingly , as described for Kindlin-2 [18] , [22] , overexpression of both the Talin FERM domain and Kindlin-1 doubled PAC1 binding when compared with cells expressing only the Talin FERM domain ( Figure 8F ) . The synergism between Talin and Kindlin-1 depends on a Kindlin-1 and β integrin tail interaction , as a PTB mutant of Kindlin-1 ( QW611/612AA ) failed to bind β integrin tails ( Figure 8G ) and the synergistic effect with Talin was lost ( Figure 8F ) . These findings suggest that Kindlin-1 is not sufficient for integrin activation but is required for inducing Talin-mediated integrin activation . This notion was confirmed with CHO cells , in which endogenous Kindlin-2 levels were depleted by RNAi ( Figure 8H ) . Furthermore , overexpressing Talin failed to induce integrin activation in these cells , but expression of Kindlin-1 restored this function ( Figure 8I ) . These findings show ( i ) that Kindlin-1 and -2 require Talin for integrin activation , ( ii ) that Talin requires Kindlins for integrin activation , and ( iii ) that Kindlin-1 and Kindlin-2 have redundant functions in vitro as both Kindlin-1 and -2 are recruited to FAs where they exert similar functions on integrin cytoplasmic tails . However , in vivo this is not the case as Kindlin-2 is recruited to cell-cell contacts in IECs and apparently does not compensate Kindlin-1 loss .
In the present study we show that a null mutation in the Fermt1 gene gives rise to skin atrophy and an acute and fulminant , neonatal intestinal epithelial dysfunction . We demonstrate that the primary defect is a loss of the intestinal epithelial barrier that secondarily leads to inflammatory cell infiltrates and the development of a severe colitis . Furthermore , we show that loss of the intestinal epithelial barrier is caused by a severe adhesion defect of intestinal epithelial cells to the underlying BM , which in turn is caused by the inability of integrins to become activated and to bind BM components . It is possible that in addition to defective integrin activation and epithelial detachment , Kindlin-1 exerts other yet unidentified functions that could contribute to the phenotype in Kindlin-1−/− mice . Kindler syndrome ( KS ) is thought to be primarily a skin disease with a disease course that is characterized by epidermal atrophy and followed by epidermal blistering , pigmentation defects and skin cancer . The complex disease syndrome is difficult to diagnose at the disease onset due to similarities with other forms of skin blistering diseases ( also called epidermolysis bullosa; EB ) that are caused by mutations in keratin and BM genes [24] . Recent case reports showed that KS may involve more organs than only the skin , as several KS patients also suffer from intestinal symptoms . One patient with a severe form of KS developed a severe postnatal UC . Interestingly , this patient was diagnosed with a null mutation in the FERMT1 gene after developing trauma-induced skin blistering [10] . In line with this severe UC case of KS , we found that the null mutation of the Fermt1 gene in mice also leads to a dramatic and lethal intestinal epithelial dysfunction very shortly after birth . Lethality is usually not seen in KS patients , which is most likely due to the immaturity of the murine intestine at birth , making it more vulnerable to injury [25] . The intestinal epithelial dysfunction of Kindlin-1-deficient mice is characterized by flat and superficial ulcerations in the colon , as the epithelium detaches from an intact BM . The defects begin in the rectum and extend along the entire colon finally leading to a severe pancolitis . The ulcerations and epithelial cell detachments are restricted to the colon , although the ileum shows signs of a secondary inflammation at later stages of the disease . In vitro studies with primary IECs from Kindlin-1−/− mice and Kindlin-1-depleted HT-29 cells showed that the cell detachment is caused by impaired activation of integrins leading to weak adhesion of IECs to the underlying BM . Mechanistically Kindlin-1 requires direct binding to the β1 and β3 integrin cytoplasmic domains to promote the activation of the two integrin subfamilies . These biochemical and functional properties are conserved among the three members of the Kindlin family . Kindlin-2-mediated binding and activation of β1 and β3 integrins critically support the attachment of endoderm and epiblast cells to the underlying BM in peri-implantation embryos [18] , while Kindlin-3 plays a central role for the activation of platelet integrins [8] . The findings of this report also demonstrate that Talin function crucially depends on the activity of Kindlin-1 . Depletion of Kindlin-2 ( the only Kindlin expressed in CHO cells ) completely prevents overexpressed Talin from activating integrins . Re-expression of either Kindlin in Kindlin-2-depleted CHO cells , however , recovers the ability of Talin to trigger the activation of integrins . It will be important to next investigate how Kindlins become activated and why Kindlin-2 is unable to take over the function of Kindlin-1 in Kindler Syndrome and Kindlin-1-deficient mice . The conclusion that the observed phenotype is triggered by IEC detachment rather than by a primary inflammatory defect in Kindlin-1 deficient mice is based on the observation that epithelial cell detachment always occurred prior to immune cell infiltration . We would therefore , argue that the detachment of IECs resembles an intestinal wound , which secondarily triggers a strong wound healing response leading to immune cell infiltrates and release of a cytokine storm . In line with this hypothesis , epithelial cell detachment and induction of inflammatory reactions can be completely prevented when Kindlin-1 pups are delivered by Caesarian section and subsequently incubated in a humidified and temperature controlled chamber . Mechanical stress applied by the colostrum is likely inducing the detachment of the weakly adhering epithelial cells in the colon . The vast majority of mouse models reported to develop colitis so far have an abnormal immune system [12] , [26] . This fact as well as the identification of several susceptibility loci in human patients [14] , [15] led to the conclusion that defects in the immune system are of central importance for UC development . Severe adhesion defects of IECs leading to a massive wound response may represent an alternative etiology for UC development . Although adhesion is severely compromised in the colon of Kindlin-1-deficient mice , they are born without skin blisters . This is in line with KS patients , who are also born without skin blisters even when they are delivered by the vaginal route but develop blisters postnatally at trauma prone sites . Interestingly , Kindlin-1 deficient mice did not show defective adhesion of basal keratinocytes to the BM even after application of mechanical stress . The different severity of the adhesion defect in skin and colon could be reflected by the functional properties of the distinct set of integrins expressed in the two organs and the absence of classical hemidesmosomes in intestinal epithelial cells [27] . Another pronounced skin defect in Kindlin-1-deficient mice as well as KS patients is skin atrophy , which seems to be due to reduced proliferation of interfollicular keratinocytes . This finding raises several questions; first , regarding the mechanism underlying the molecular control of cell proliferation by Kindlin-1 . The mechanism is unknown and could result from a diminished cross talk between integrin and growth factor signaling . Second , it is also unclear how a molecular player that supports proliferation is giving rise to cancer at a later stage . It is possible that the localization of Kindlins in different cellular compartments , i . e . cell-matrix adhesion sites , cell-cell adhesion sites and in certain instances in the nucleus , equips them with different functions that become evident at different time points in life . Kindlin-1 and -2 are co-expressed in epidermal cells as well as epithelial cells of the colon . Interestingly , we found that Kindlin-2 cannot compensate Kindlin-1 function in vivo , neither in the colon nor in skin . Since Kindlin-2 normally localizes to cell-cell adhesions in both cell types and does not translocate to integrin adhesion sites in mutant intestinal and epidermal epithelial cells , it is unable to compensate for the loss of Kindlin-1 , although both Kindlins are capable of performing the same tasks at the integrin tails ex vivo and in vitro [27] . Hence , a therapeutic strategy to reroute some of the Kindlin-2 from cell-cell to the integrin adhesion sites may represent a promising approach to prevent ulceration in KS patients with severe UC .
The Kindlin-1−/− mice were obtained by replacing the ATG-containing exon 2 with a neomycin resistance cassette ( detailed information on the cloning of the targeting construct can be obtained from Faessler@biochem . mpg . de ) . The construct was electroporated into R1 embryonic stem ( ES ) cells ( passage 15 ) and homologous recombination was verified with southern blots . Genomic DNA was digested with EcoRV , blotted and then hybridized with a 5′ probe or digested with BglII , blotted and hybridized with a 3′ probe ( Figure 1A ) . Targeted ES cells were injected into blastocysts and transferred into foster mice . Mice were housed in a special pathogen free mouse facility . All animal experiments have been approved by the local authorities . For H&E stainings intestinal segments were either PFA fixed and embedded in paraffin , or frozen on dry-ice in cryo-matrix ( Thermo ) . Immunhistochemistry of paraffin embedded sections was carried out as previously described [5] . Sections of 8 µm thickness were prepared and stained following routine protocols . Cryo sections were fixed in 4% PFA/PBS except for the Kindlin-1 staining where sections were fixed with 1∶1 methanol/acetone at −20°C . Subsequently tissue sections were blocked with 3% BSA/PBS , incubated with primary antibodies in a humidity chamber over night at 4°C , with fluorescently labeled secondary antibodies for 1 h at RT and finally mounted in Elvanol . Pictures were taken with a Leica DMIRE2 confocal microscope with a 100× or 63× NA 1 . 4 oil objective . Recombinant GST-β1 , β1Y788A , β1Y800A , β3 , β3Y747A , β3Y759A cytoplasmic tails were expressed and purified from E . coli under non denaturing conditions . 5 µg of recombinant tails were incubated with 500 µg IEC lysate ( in 50 mM Tris pH 7 . 4 , 150 mM NaCl , 1 mM EDTA , 1% Triton-X-100 ) overnight . GST-constructs were precipitated with glutathione beads ( Novagen ) . Subsequent western blots were probed for Kindlin-1 and GST . A polyclonal peptide antibody against Kindlin-1 was raised against the peptide YFKNKELEQGEPIEK as previously described [5] . The following antibodies were used at the given concentration indicated for western blot ( W ) , immunoprecipitation ( IP ) , immunfluorescence ( IF ) , immunhistochemistry ( IHC ) : Kindlin-1 ( W: 1∶5000 , IF cells: 1∶200 , IF tissue: 1∶1000 ) , Kindlin-2 ( W: 1∶1000 , IF cells 1∶200 , IF tissue: 1∶200 ) , E-cadherin ( Cdh1; Zymed , W: 1∶5000 ) , Migfilin ( W: 1∶5000 , IF cells 1∶100 , IF tissue: 1∶100 ) , GAPDH ( Chemicon; W: 1∶10000 ) , phalloidin Tritc ( Sigma; IF cells: 1∶800 , IF tissue: 1∶800 ) , Mac-1 ( EuroBioscience; IF tissue: 1∶100 ) , GR-1 ( Ly6g; eBioscience; IF tissue: 1∶100 ) , Thy1 . 2 ( PharMingen; IF tissue: 1∶100 ) , GST ( Novagen; W: 1∶10000 ) , His ( CellSignaling; W: 1∶1000 ) , PAC1 ( BD; FACS: 1∶100 ) , α6 integrin ( Itga6; PharMingen; IF tissue: 1∶100 ) , CollagenIV ( a gift from Dr . Rupert Timpl; IF tissue: 1∶100 ) , Laminin-332 ( a gift from Dr . Monique Aumeilley; IF tissue: 1∶200 ) , Perlecan ( Hspq2; a gift from Dr . Rupert Timpl; IF tissue: 1∶100 ) , β1 integrin ( Chemicon; WB: 1∶3000 , IF tissue: 1∶600 ) , 9EG7 ( PharMingen; FACS: 1∶100 ) , EGFP ( Abcam; WB: 1∶10000 ) , β1 integrin ( PharMingen; FACS: 1∶200 ) , αv integrin ( PharMingen; FACS: 1∶200 ) . Keratin10 ( Krt10; Covance; IHC: 1∶600 ) , Keratin14 ( Krt14; Covance; IHC: 1∶600 ) , Loricrin ( Lor; Covance; IHC: 1∶500 ) Ki67 ( Dianova; IHC: 1∶50 ) , cCaspase3 ( CellSignaling , IHC: 1∶100 ) . Pregnant mice were sacrificed by cervical dislocation when embryos were at E18 . 5–E19 of gestation . The uterus was removed and cut open . Embryos were taken out and the umbilical cord was cut . Mice were subsequently dried and kept in an incubator at 37°C and high humidity . Total RNA from whole colons was extracted with a PureLink Micro to Midi RNA extraction kit ( Invitrogen ) following the manufacturers instructions . cDNA was prepared using the iScript cDNA Synthesis Kit ( Biorad ) . Real Time PCR using a Sybr Green ready mix ( Biorad ) was performed in an iCycler ( Biorad ) . Each sample was measured in triplicates and values were normalized to GAPDH . Following primers were used; TNFα fwd: AAAATTCGAGTGACAAGCCTGTAGC , TNFα rev: GTGGGTGAGGAGCACGTAG . IL-6 fwd: CTATACCACTTCACAAGTCGGAGG IL-6 rev: TGCACAACTCTTTTCTCATTTCC . RT-PCR for Kindlin-1 and GAPDH was performed as previously described [5] . Neonatal mouse intestine was removed and flushed with 1 ml PBS . The intestine was longitudinally cut open , rinsed with PBS and incubated for 40 min . in IEC isolation buffer ( 130 mM NaCl , 10 mM EDTA , 10 mM Hepes pH 7 . 4 , 10% FCS and 1 mM DTT ) at 37°C on a rotor . The epithelium was shaken off and pelleted by centrifugation at 2000rpm for 5 min . For WB analysis cells were washed once with PBS and subsequently lysed . For flow cytometry cells were washed once with PBS and trypsinized with 2× trypsin ( GIBCO ) for 10 min . at 37°C . Trypsin was inactivated with DMEM containing 10%FCS . A single cell suspension was prepared by passing cells through a cell strainer ( BD ) . Primary keratinocytes were isolated from P3 mice as described previously [28] . Cells were cultured on a mixture of ColI ( Cohesion ) and 10 µg/ml FN ( Invitrogen ) coated plastic dishes in keratinocyte growth medium containing 8% chelated FCS ( Invitrogen ) and 45 µM Ca2+ . IEC's and keratinocytes were stained with 9EG7 antibody in Tris buffered saline [29] . For the PAC1 binding assay CHO cells were stained for 40 min . at RT as described previously [23] . Cells were gated for viability by excluding propidium iodide-positive cells . CHO cells transfected with EGFP-tagged constructs were additionally gated for highly EGFP-positive cells . Measurements were performed with a FACS Calibur ( BD ) and data evaluation was done with FlowJo software . The EGFP-Kindlin-1 expression plasmid was previously described [5] . The cDNA encoding His-Kindin-1 C-terminus ( aa 471–677 ) was amplified by PCR and cloned into the pQE-70 vector ( Qiagen ) . The Kindlin-1 PTB mutation QW611/612AA was introduced with a site directed mutagenesis kit ( Stratagene ) following the manufacturers recommendations . All EGFP constructs were cloned into the pEGFP-C1 vector ( Clontech ) and subsequently sequenced . EGFP-Talin head was previously described [8] . CHO and HT-29 cells were maintained in DMEM containing penicillin/streptomycin , non-essential amino-acids and 10% or 20% FCS , respectively ( GIBCO ) . Cells were transfected with 2 µg of each DNA in six well plates using Lipofectamine 2000 following the manufacturers' instructions ( Invitrogen ) . To deplete Kindlin-1 constitutively from HT-29 cells , an shDNA corresponding to the cDNA sequence GTAAGTCCTGGTTTATACA of hKindlin-1 and a control cDNA with the sequence AGCAGTGCATGTATGCTTC were cloned into the pSuperRetro vector ( OligoEngine ) . Viral particles were prepared as described previously [30] . HT-29 cells were infected and subsequently selected with 2mg/l puromycin . Transient knockdown of Kindlin-2 from CHO cells was achieved by transfection of the siRNAs; Kind2_1: GCCUCAAGCUCUUCUUGAUdTdT and Kind2_2: CUCUGGACGGGAUAAGGAUdTdT , and a control siRNA ( purchased from Sigma ) using Lipofectamine 2000 ( Invitrogen ) , following the manufacturers instructions . Cells were harvested and assayed 24 hours after transfection . The adhesion assays were performed as previously described [29] , using 40000 cells per well in serum free DMEM ( HT-29 ) or MEM ( primary keratinocytes ) . Osmolarity was measured from 50 µl urine using an Osmomat 030 from Gonotec . Embryonic skin barrier formation was determined as previously described [31] . Slides with a 1 µm diameter ( ibidi BioDiagnostics ) were coated overnight with 5 µg/ml FN and then blocked with 1% BSA . 100 . 000 cells were seeded onto the slides and incubated for 2 . 5 hours in a cell culture incubator . Cells were subsequently exposed to increasing amounts of shear force in two minute intervals ( as indicated in the Figure S11 ) and pictures were taken every second . CHO cells were transiently transfected with the indicated EGFP constructs . Approximately 1mg of protein lysate was immunoprecipitated using the μMACS Epitope Tag Protein Isolation Kit for EGFP tags ( Miltenyi Biotec ) following the manufacturers instructions . Electron microscopy was performed as previously described [29] . Analyses were performed with GraphPad Prism . If not mentioned otherwise in the figure legends , Gaussian distribution of datasets was determined by a D'Agostino & Pearson omnibus normality test . If samples were not Gaussian distributed a Mann-Whitney test was performed . Gaussian distributed samples were either compared with a one way ANOVA and a Tukey's multiple comparison post test or an unpaired two-tailed t-test . | Mutations in FERMT1 , coding for the Kindlin-1 protein , cause Kindler Syndrome in humans , characterized by skin blistering , atrophy , and cancer . Recent reports showed that some Kindler Syndrome patients additionally suffer from ulcerative colitis . However , it is unknown whether this is caused by loss of Kindlin-1 or by unrelated abnormalities such as infections or additional mutations . We ablated the Fermt1 gene in mice to directly analyze the pathological consequences and the molecular mode of action of Kindlin-1 . Kindlin-1–deficient mice develop a severe epidermal atrophy , but lack blisters . All mutant mice die shortly after birth from a dramatic , shear force-induced detachment of intestinal epithelial cells followed by a profound inflammation and organ destruction . The intestinal phenotype is very similar to , although more severe than , the one observed in Kindler Syndrome patients . In vitro studies revealed that impaired integrin activation , and thus impaired adhesion , to the extracellular matrix of the intestinal wall causes intestinal epithelial cell detachment . Therefore , we demonstrate that intestinal epithelial cells require adhesive function of integrins to resist the shear force applied by the stool . Furthermore , we provide evidence that the colitis associated with Kindler Syndrome is caused by a dysfunction of Kindlin-1 rather than by a Kindlin-1–independent event . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"biochemistry",
"dermatology/pediatric",
"skin",
"diseases,",
"including",
"genetic",
"diseases",
"dermatology/photodermatology",
"and",
"skin",
"aging",
"immunology/immune",
"response",
"gastroenterology",
"and",
"hepatology/inflammatory",
"bowel",
"disease",
"genetics",
"and",
"genomics/disease",
"models",
"genetics",
"and",
"genomics/genetics",
"of",
"disease",
"cell",
"biology/cell",
"adhesion"
] | 2008 | Loss of Kindlin-1 Causes Skin Atrophy and Lethal Neonatal Intestinal Epithelial Dysfunction |
The epidemic spread of infectious diseases is ubiquitous and often has a considerable impact on public health and economic wealth . The large variability in the spatio-temporal patterns of epidemics prohibits simple interventions and requires a detailed analysis of each epidemic with respect to its infectious agent and the corresponding routes of transmission . To facilitate this analysis , we introduce a mathematical framework which links epidemic patterns to the topology and dynamics of the underlying transmission network . The evolution , both in disease prevalence and transmission network topology , is derived from a closed set of partial differential equations for infections without allowing for recovery . The predictions are in excellent agreement with complementarily conducted agent-based simulations . The capacity of this new method is demonstrated in several case studies on HIV epidemics in synthetic populations: it allows us to monitor the evolution of contact behavior among healthy and infected individuals and the contributions of different disease stages to the spreading of the epidemic . This gives both direction to and a test bed for targeted intervention strategies for epidemic control . In conclusion , this mathematical framework provides a capable toolbox for the analysis of epidemics from first principles . This allows for fast , in silico modeling - and manipulation - of epidemics and is especially powerful if complemented with adequate empirical data for parameterization .
Despite huge efforts to improve public health , the spread of infectious diseases is still ubiquitous at the beginning of the 21st century , and there is considerable variability in epidemic patterns between locations . Although the recent influenza pandemic has been a global challenge , there have nonetheless been differences in its timing in the northern and southern hemisphere due to seasonal effects [1] , [2] . Another prominent example for epidemic variability is the prevalence of sexually transmitted diseases ( STDs ) , specifically HIV infections . Although HIV is endemic in many populations at low levels or restricted to high-risk groups , it has become highly endemic in other parts of the world [3] , [4] . As a consequence , the spread of infectious diseases cannot be understood globally but understood only as the result of several local factors , such as climate and hygiene conditions , population density and structure , and cultural habits and mobility . Epidemic models aim to capture the mechanisms that link these factors to the emergent epidemics and to promote an understanding of the underlying dynamic processes as a prerequisite for intervention strategies [5] , [6] . A useful abstraction in this context is to regard individuals that may be infected as nodes of a network in which the links are the potentially infectious contacts among individuals ( or nodes in network notation ) . A major remaining challenge in modern epidemiology is to link the variability of transmission networks to the corresponding emergent epidemics . Models that are flexible and can be adapted to specific epidemic situations best meet these challenges . Because we focus on the interplay between transmission network topology and epidemics , we will restrict ourselves to diseases caused by agents that lead to either immunity or death in their host , i . e . , in which infection can occur only once . These epidemics can be described by Susceptible-Infected-Recovered or SIR models [5] , [6] . We refer to the mathematically closely related case , where infection eventually leads to the death of the host , as a SID model ( Susceptible-Infected-Death ) . The original , or classical , SIR model [7] assumes a mass-action type dynamic and as a consequence describes epidemics in homogeneous , well-mixed populations . Because this is generally not a good approximation of real world situations , current epidemic models strive for an integrated approach that considers both information about the course of disease ( i . e . , susceptible , infected and recovered stages ) and the relevant transmission network [8] , [9] . The models vary in their assumptions , attention to detail , computational costs , and as a consequence , their fields of application . Compartmental SIR models consider different contact patterns in sub-populations and link them via a contact matrix [10] , providing a coarse-grained , but often adequate , representation . Network-based SIR models consider the distribution in each individual's number of infectious contacts in the transmission network ( i . e . each node's degree in network notation ) [11]–[13] . These models allow for the study of transmission networks with strong heterogeneity in the number of contacts among individuals , which in some cases also means that they consider correlations in the way contacts are made [14] , [15] , or clustering [16]–[18] . Although these approaches focus on static networks , a recent approach considers networks with arbitrary degree distributions and transient contacts and allows for the derivation of the temporal evolution in the number of susceptible and infected nodes from a closed set of equations [19]–[21] . Finally , pair models are a very general approach [22] for studying SIR epidemics on heterogeneous networks . They provide a large amount of flexibility in considering the way contacts are made ( correlations or clustering ) and maintained [23]–[25] , but , as a trade-off , they quickly become very computationally demanding [26] . An assumption often implicitly made in epidemic models is that the epidemic sweeps through the population at much shorter time scales than the time scale of background demographic processes , i . e . , natural birth and death processes are neglected . This is a good approximation in cases such as the yearly influenza epidemics , but it is hardly adequate for HIV epidemics , which span decades . To compensate for this limitation , we integrated demographic background processes into recent network epidemic models [19] , [20] . With HIV in mind as a case study , we focus on disease epidemics that lead to death after infection of susceptible individuals , possibly after undergoing several stages of the disease . In addition to earlier work , our approach also allows for an in depth study of the interplay between epidemic spreading and the structure and dynamics of the underlying transmission network . Strictly speaking , all approaches discussed only predict the mean behavior of epidemics within the limit of an infinite host population . However , our comparisons with finite size , agent-based simulations show that this is a good and computationally efficient approximation already for moderate population sizes .
Assuming knowledge about infectious contacts , hosts ( or nodes ) can be assembled into subgroups according to their number of infectious contacts ( or degree in network notation ) . Following the ansatz of [19] , [20] , evolution equations for each of these subgroups are then derived which allows us to model arbitrary heterogeneity in the number of infectious contacts between hosts . Susceptible hosts get infected at a rate that is proportional to their number of contacts , , the transmission rate per contact , , as well as the probability that a contact made by a susceptible individual links to an infected individual , . The number of hosts with a given number of contacts changes either by birth and death or by birth and death of the hosts' contacts , where death may occur both due to disease ( at a rate ) or due to other causes ( at a rate ) . Individuals entering the population at a rate establish contacts with probability ( corresponding to the probability generating function ) . Table 1 summarizes the parameters and notation of the model which allows us to compile the equations for the numbers of susceptible and infected individuals with contacts and as: Note that it is implicitly assumed that new individuals enter the population at a rate have contacts with probability by which they link randomly to those individuals already present . All presented data are based on this dynamic for the establishment of new nodes' contacts , as this can be considered as the most basic dynamic model . However , the approach is not restricted to this case and can naturally be extended to other dynamics , for example preferential attachment of links with respect to the target nodes' degree ( See Text S1 ) . It is further assumed that each individual in the population of size can give birth to a susceptible individual at the same rate which may be conditioned on the individuals' health status in future models . Individuals dying from natural causes at a rate are assumed to have the same average number of contacts as found in the whole population without preferences for susceptible or infected individuals . In terms of the total number of susceptible and infected individuals the equations read ( 1 ) ( 2 ) To close this set of equations and to take into account local clustering of infectious cases , additional equations have to be derived for and , the probabilities that a contact made by a susceptible or infected individual points to an infected individual . We apply the techniques developed in [19] ( See Text S1 for detailed calculations and discussion of the underlying approximations ) and conclude ( 3 ) ( 4 ) The occurrence of the probability generating functions ( PGFs ) in equations ( 3 ) and ( 4 ) which represent the degree distributions of susceptible and infected individuals makes it clear that the way links are maintained between susceptible and infected nodes depends on the contact behavior within these subgroups , as well as the changes therein in response to the epidemic . In other words , the set of equations can only be closed if the time evolution of the PGFs and is considered . With , this results in ( 5 ) ( 6 ) Given that equations ( 3–4 ) show and its derivatives only for might suggest that the set of partial differential equations can be reduced to a set of ordinary differential equations including equations for the time evolution of the moments expressed by and . In fact , this leads to a hierarchy of equations for the time evolution of all higher order moments of . With the equations given in Table 1 we are able to investigate the interplay between epidemic processes and transmission network topology . In particular , it is possible to follow the degree distribution within the subgroups of susceptible and infected individuals in terms of the probability generating functions and . However , transmission networks do not only change their structure as a result of birth and death processes but also because of changes in contact partners . Analogous to [20] , we consider swapping of contact partners at a rate which affects the quantities and via additional terms and , respectively . Agent-based simulations were performed using NetLogo V4 . 0 . 4 [27] , in which some code fragments from the model “Virus on Network” included in the software's model library were used [28] . Poisson networks were generated by assigning links between randomly chosen nodes; other random networks were generated based on their degree sequence [13] , [29] ( See Text S1 for details ) . The numerical solution to the partial differential equations was obtained using Mathematica V6 . 0 . 2 [30] with the function NDSolve and the numerical method of lines [31] .
The SID model allows for the study and analysis of epidemics using transmission networks with a broad range of topological features . In particular , any distribution of the number of infectious contacts per host can be implemented within the framework by providing the corresponding probability generating function as an input parameter . Fig . 1 compares the predictions derived from the set of partial differential equations ( 1–6 ) with the observations from agent based simulations for several exemplary topologies ( See the section on Methods and Text S1 for details ) . Although equations ( 1–6 ) hold for the mean behavior in the limit of infinite network size , they are already a good approximation for the agent-based simulations with moderate population sizes and in particular reproduce well-characterized behaviors . Epidemics in heterogeneous networks spread faster but are more restricted ( Fig . 1 , column 1 vs . 2 ) [32] . Contrarily , transient contacts lead to a larger epidemic size when contacts are maintained sufficiently long to ensure transmission ( Fig . 1 , column 1 vs . 4 ) . The re-growth in the average number of contacts per person ( node ) in the scenario with birth and death processes ( Fig . 1 , column 3 ) is reminiscent of re-emergent , or persistent , infections that can be observed in this context . Finally , the method can be naturally extended to epidemics of diseases with several infected stages before death ( See Text S1 ) . To model HIV epidemics , we have to extend the SID model to take into account the heterogeneous infectious profile of an HIV infection . The course of the disease is characterized by a short , but highly infectious , period of primary infection . This is followed by a prolonged period of latent infection with a much lower infectiousness ( sometimes referred to as an asymptomatic or chronic phase ) [33] before the onset of AIDS as the final stage of the disease . Our focus will be on the primary infection , , which ceases at a rate of 4 . 1 per year and is associated with a transmission rate of 2 . 76 per year as opposed to the latent infection , , with a progression rate of 0 . 12 per year and a transmission rate of 0 . 1 per year . We neglect the role of the final stage of disease in transmission with the assumption that health conditions prevent a further transmission of HIV . The evolution equations for , and are determined analogously to the SID case to describe the numbers of individuals with contacts as being susceptible , primarily infected and latently infected . In addition , equations describing the contacts among individuals of different epidemic groups are derived ( i . e . , , , , and ) . Finally , the set of equations is closed by a derivation of the probability generating functions , and which describe the contact patterns and their temporal changes within each group ( See Text S1 for details ) . The resulting equations allow for a much faster ( and flexible ) assessment of epidemic scenarios than agent-based simulations . Again , the structure of the equations emphasizes the mutual influence between the epidemic process and the underlying transmission network . This is exemplarily illustrated in Fig . 2 , which shows the spreading of HIV in a “synthetic” population with a scale free distribution in the number of potentially infectious contacts . Agent-based simulations and the numerical solution of the set of partial differential equations agree well and show how the epidemic saturates within a few decades with a few percent of latently infected individuals ( as opposed to a few per mill of primarily infected individuals ) . During the epidemic expansion phase , the average number of contacts among infected individuals grows sharply while the average number of contacts among susceptible individuals decreases slightly . Individuals with more potentially infectious contacts are at a higher risk of infection and accumulate among primarily infected individuals . Their average number of contacts decreases during the following latent stage due to the increased mortality in their infected neighbors . This hierarchy in the average number of contacts from primarily infected through latently infected to susceptible individuals can still be observed during the saturation phase of the epidemic . The temporal evolution in the network topology can be studied in more detail from the probability generating functions and the degree distributions ( ) describing the contact behavior within the epidemic subgroups in Fig . 2 ( panels C and D ) . Although the contact behavior of individuals newly entering the population is not time dependent and corresponds to the original distribution found before the onset of the epidemic ( ) , contact patterns change specifically in the epidemic subgroups . One can observe that the fraction of single individuals grows among susceptible individuals as it does among latently infected individuals in the saturation phase of the disease due to the increased mortality rate among their infected contacts . This loss in contacts is not observed in the short period of primary infection , which can be seen in the maintenance of the high average number of contacts in this subgroup ( originally acquired due to the hazard of infection growing with the number of contacts ) . These features have been observed in actual HIV epidemics , for example in the Eighties' San Francisco MSM cohort [34] . Note that a persistent epidemic cannot be generally expected if new individuals enter the population with the current contact behavior instead of the initial contact behavior , i . e . with instead of . Because high-risk individuals have a higher risk of death due to the epidemic , this will successively lead to an introduction of individuals with lower risk behavior ( lower average number of contacts ) until eventually the network becomes sub-critical and the epidemic ceases [35] . The analysis has already taken into account that the course of an HIV infection is intimately related to the dynamics of its epidemic spreading . While the brief period of primary infection is associated with a largely increased infectiousness , a much lower infectiousness is observed during the prolonged period of latent infection ( also referred to as asymptomatic or chronic infection ) , which grows again in the late stages of disease [33] , [36]–[38] . Therefore , it cannot easily be determined , which phase of the disease results in most new infections , making this a topic of ongoing debate [39] , [40] . Various studies on the contribution of the initial and latent stage to HIV incidence in different populations [41] , [42] reveal that the contribution of either stage of the disease to HIV incidence is very context dependent . There is , however , agreement that primary infection becomes a more important epidemic driver when risk behavior increases ( number of casual/concurrent contacts or partners ) , whereas the incidence obtained from latent infections becomes more important during the saturation phase of an epidemic ( as opposed to its expansion phase ) , which is supported by modeling approaches [43] , [44] . This question can certainly not be conclusively answered by our study due to a lack of adequate empirical data , but we can provide a tool that helps researchers to better understand under which circumstances infections dominantly originate from either primarily or latently infected individuals . Therefore , we do not aim to parameterize the model in the most realistic way but rather to investigate two possible scenarios that contain some general features found in epidemic networks of sexually transmitted HIV . There is generally a large heterogeneity in the numbers of sex partners [45]–[47] , the frequency of partner change and the level of concurrency in partnerships [48] . Acknowledging that not only the number of infectious contacts but also their timing [49]–[51] is important for epidemic spreading , we investigate two scenarios that we depict as having weak and strong concurrency with parameters given in Fig . 3 . Both scenarios assume an average number of 10 lifetime partners during a lifetime of 50 years ( p . a . , neglecting delays due to childhood/adolescence before sexual debut ) . However , in the scenario of weak concurrency , a lower average number of concurrent partners , , is exchanged for a higher partner change rate , , in comparison to the scenario of strong concurrency . The numerical solution of the set of PDEs shown in Fig . 3 makes it immediately apparent that not only the number of partners but also the level of their concurrency have a profound impact on both the time scale and width of an epidemic's expansion . A much more severe epidemic is seen in the case of strong concurrency , which agrees with the earlier findings discussed above . This effect is still present if constant transmission rates are assumed throughout the course of disease ( See Fig . 2 in Text S1 ) . A common feature of both epidemics is the hierarchy in the average number of contacts for primarily , latently infected and susceptible individuals; this gives direction for targeted intervention strategies . To understand which stage of disease drives the epidemics , it is of particular interest to study the relative risk of infections derived from primary versus latent infections ( 7 ) which is shown in the bottom right panel of Fig . 3 . While infections from the primary stage of disease dominate in the expansion phase of the epidemic in a scenario with strong concurrency in contacts , infections from the latent stage of disease dominate as soon as the epidemic matures . In the case of weak concurrency ( being a closer approximation to serial monogamy ) infections from the latent stage of disease dominate throughout the whole epidemic . This case study confirms that the question which stage of the disease drives HIV epidemics cannot be answered without in depth knowledge of the topology and dynamics of the underlying transmission network , as well as knowledge about the saturation stage of the epidemic .
A highly flexible mathematical framework has been introduced that allows for the investigation of the interplay between the topology and dynamics of transmission networks and emergent epidemics within a closed set of equations . The approach is focused on pathogens that lead to death of their hosts after some time of infection ( potentially in several stages ) . HIV epidemics have been considered as an area of application in which the newly developed method helps to understand the complex interdependencies between the HIV epidemic profile , its transmission network , and the epidemic process . Several scenarios in synthetic populations have been investigated , showing that the transmission network is not a static support that shapes the epidemic but , on the contrary , is itself shaped by the epidemic . This becomes clear in the changing contact behavior of infected and susceptible individuals quantified either by degree distributions or probability generating functions . The mathematical framework further provides a capable tool to address the question of whether the group of primarily or of latently infected individuals is the main driver of HIV epidemics . The case studies emphasize that the answer depends both on the maturation stage of the epidemics and the structure of the relevant transmission network . These findings are relevant for the implementation of targeted intervention strategies ( e . g . , promotion of behavioral changes or vaccination programs if available ) , and are particularly relevant to the ongoing debate on public health policies [41] , [42] . The capacity of the modeling approach has been illustrated by example applications . However , its real strength is the development of a framework that allows for a quantitative and systematic assessment of the interdependencies and feedback mechanisms between transmission network dynamics and the spread of an epidemic . In particular , the detailed tracing of contact behavior in all epidemic groups , which may also undergo a flexible demographic process , goes beyond earlier approaches . The method can naturally be extended to other settings with a more complex infectious profile or to epidemics with a classical SIR dynamic . This makes it useful for a very broad spectrum of epidemic scenarios , which may include improved modeling of SIR-like infections such as measles , rubella , pertussis or influenza . The broad applicability of the approach makes it worthwhile to consider further improvements to stretch its limits towards an increasingly realistic description of the epidemics we face day-to-day . The finite size agent-based simulations shown in Fig . 1 and Fig . 2 for validation already indicate that the current approach is designed for the limit of infinite population sizes . Although the mean behavior is well represented already for moderate population sizes the approach does not account for obvious fluctuations . Recent research has shown that stochastic fluctuations may have a strong influence on real world epidemic phenomena such as re-emergent epidemics [52] , [53] which in combination with other recently developed techniques makes this an exciting direction of future research [54] . It is further assumed that contacts are made randomly [13] without taking into account any preferences or correlations influencing their establishment . Social networks usually show clustered communities [55] and some degree of assortativity , i . e . , individuals tend to mix with their likes [10] , [14] , [56] . This often results in the generation of core groups that sustain and drive an epidemic . Ongoing research efforts [17] , [18] , [25] , [57] , [58] in this field give directions for an extended model , ideally in combination with a more realistic description of transient contacts . Moreover , demographic change with random assignment of new contacts results in increasingly homogeneous contact behavior of older individuals in the network ( See Fig . 3–5 in Text S1 ) . A straight-forward extension of the approach would be to study other modes of contact establishment , such as preferential attachment with respect to degree . This will allow for a study of more complex topological evolution and its consequences . Finally , it should be considered that the transmission network is not only shaped by the epidemic process but also by active behavioral changes , such as social distancing or vaccination [59]–[64] . In conclusion , we have presented a new mathematical framework that allows researchers to closely monitor both the epidemic process and its transmission network for general SIR-like infections in an computationally efficient manner . The current method allows for great flexibility accounting for variability in transmission network topology and dynamics , as well as pathogen specific features . Nonetheless , it will be important to assess the method's limitations in the field after parameterization with appropriate empirical data . An exciting challenge for future research is to further expand its limits . | The way potentially infectious contacts are made strongly influences how fast and how widely epidemics spread in their host population . Therefore , it is important to assess changes in contact behavior throughout an epidemic; these may occur due to external factors , such as demographic change , or as a side effect of the epidemic itself , leading to an accumulation of individuals with risky behavior in the infected population . We have developed a mathematical framework that allows for the study of the mutual interdependencies between epidemic spread and changes in contact behavior . The method is used to study HIV epidemics in model populations . We address the question of whether HIV is primarily spread by highly contagious initially infected hosts or by less contagious latently infected hosts who will , however , encounter more situations for potential transmission . The answer to this question strongly depends on the concurrency of contacts and the maturation stage of the epidemic . Initially infected hosts are major epidemic spreaders in populations with strongly concurrent contacts and during epidemic expansion whereas otherwise latently infected hosts play a more important role . The availability of better data for parameterization will make this approach relevant for public health considerations . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"physics/interdisciplinary",
"physics",
"evolutionary",
"biology/sexual",
"behavior",
"mathematics",
"biophysics/theory",
"and",
"simulation",
"computational",
"biology/evolutionary",
"modeling",
"infectious",
"diseases/hiv",
"infection",
"and",
"aids",
"public",
"health",
"and",
"epidemiology/epidemiology",
"public",
"health",
"and",
"epidemiology/infectious",
"diseases",
"evolutionary",
"biology/bioinformatics",
"computational",
"biology/systems",
"biology",
"infectious",
"diseases/epidemiology",
"and",
"control",
"of",
"infectious",
"diseases"
] | 2010 | Untangling the Interplay between Epidemic Spread and Transmission Network Dynamics |
The interdependence of selective cues during development of regulatory T cells ( Treg cells ) in the thymus and their suppressive function remains incompletely understood . Here , we analyzed this interdependence by taking advantage of highly dynamic changes in expression of microRNA 181 family members miR-181a-1 and miR-181b-1 ( miR-181a/b-1 ) during late T-cell development with very high levels of expression during thymocyte selection , followed by massive down-regulation in the periphery . Loss of miR-181a/b-1 resulted in inefficient de novo generation of Treg cells in the thymus but simultaneously permitted homeostatic expansion in the periphery in the absence of competition . Modulation of T-cell receptor ( TCR ) signal strength in vivo indicated that miR-181a/b-1 controlled Treg-cell formation via establishing adequate signaling thresholds . Unexpectedly , miR-181a/b-1–deficient Treg cells displayed elevated suppressive capacity in vivo , in line with elevated levels of cytotoxic T-lymphocyte–associated 4 ( CTLA-4 ) protein , but not mRNA , in thymic and peripheral Treg cells . Therefore , we propose that intrathymic miR-181a/b-1 controls development of Treg cells and imposes a developmental legacy on their peripheral function .
Regulatory T cells ( Treg cells ) expressing the lineage-defining transcription factor forkhead box protein P3 ( Foxp3 ) form an integral part of the adaptive immune system and function to prevent unwanted immune responses [1 , 2] . Treg cells are generated during T-cell development in the thymus ( thymic [t]Treg cells ) as well as via peripheral induction of naive T cells ( induced [i]Treg cells ) . Development of tTreg cells depends on strong T-cell receptor ( TCR ) signals [3] . Accordingly , tTreg cells are generated when a TCR of a developing T cell recognizes a self-antigen with high affinity , as has been demonstrated in mouse models transgenic for both a TCR and its cognate antigen [4 , 5] and by analysis of polyclonal superantigen-reactive T cells [6 , 7] . tTreg cells can develop through two distinct precursor ( prec ) stages . Some Treg-cell precursors are found within a cluster of differentiation ( CD ) 4 single-positive ( SP ) Foxp3− , glucocorticoid-induced tumor necrosis factor receptor-related protein high ( GITRhi ) , CD25+ population [8] . These cells are the first precursors generated in double transgenic TCR/cognate-antigen mouse models [9 , 10] . More recently , an additional CD4SP Foxp3+CD25− Treg-cell precursor has been described [11] . These cells are phenotypically less mature than tTreg cells , are generated with similar kinetics as tTreg cells upon induction of T-cell development in vivo , and efficiently become tTreg cells in vitro and in vivo [10 , 11] . Generation of both precursors is dependent on strong TCR signals , although on average , Foxp3+CD25– Treg-cell precursors have received somewhat weaker signals than their Foxp3−GITRhiCD25+ counterparts [10] . Further differentiation into mature Foxp3+CD25+ tTreg cells is then dependent on γc cytokines [8 , 10–12] . The level of TCR signal strength required for tTreg cell generation in comparison to TCR signals resulting in clonal deletion have not been fully established . Data from a TCR signaling reporter as well as repertoire studies suggest that signal strength required for tTreg-cell development overlaps with that inducing clonal deletion in other autoreactive thymocytes [3 , 13–15] . However , reduction of major histocompatibility complex ( MHC ) ligand levels on medullary thymic epithelial cells rescued autoreactive T cells from clonal deletion but resulted in a concomitant increase in Treg-cell development , suggesting that at least some tTreg cells are generated through weaker TCR signals than those inducing clonal deletion [16] . Treg cells suppress T-cell immune responses using multiple molecular mechanisms , including consumption of interleukin-2 ( IL-2 ) and production of suppressive cytokines , as well as expression of the coinhibitory receptor cytotoxic T-lymphocyte–associated protein 4 ( CTLA-4 ) [17 , 18] . Mice specifically lacking the inhibitory receptor CTLA-4 in Treg cells succumb to fatal autoimmune disease , indicating that CTLA-4 plays a major role in suppressive function [19] . It has been proposed that CTLA-4 on Treg cell surfaces acts through capture of the costimulatory ligands CD80 and CD86 on antigen-presenting cells , thereby curtailing full activation of conventional T cells [20] . MicroRNAs ( miRNAs ) play a critical role in immune homeostasis and tolerance . Global loss of miRNAs results in defective development of tTreg cells [21] . However , no individual miRNA has been demonstrated to control intrathymic generation of Treg cells . miRNA miR-181a is the most prominently expressed miRNA in double-positive ( DP ) thymocytes , and it has been shown in vitro that miR-181a serves as a rheostat for TCR signals in T cells and thymocytes through targeting a combination of tyrosine and dual-specificity phosphatases , including protein tyrosine phosphatase , nonreceptor type ( Ptpn ) 11 , Ptpn22 , and dual specificity phosphatase 6 ( Dusp6 ) [22–24] . Deletion of miR-181a/b-1 in mice resulted in an almost complete failure in development of invariant natural killer T ( iNKT ) cells and Mucosal-Associated Invariant T ( MAIT ) cells [25–27] due to a defect in thymic agonist selection [26 , 28] . Furthermore , loss of miR-181a/b-1 caused altered selection of conventional T cells in a TCR transgenic model with a shift towards positive selection [29] . However , counterintuitively , miR-181a/b-1−/− mice display increased resistance to experimental autoimmunity , which has not been fully explained [29] . Here , we tested the hypothesis that miR-181a/b-1 controlled intrathymic development of Treg cells . De novo production of miR-181a/b-1−/− tTreg cells was impaired because of altered sensitivity to TCR signals during selection . Generation of Treg cells in the absence of miR-181a/b-1 resulted in elevated expression of CTLA-4 , which penetrated into the periphery despite the fact that peripheral WT Treg cells express very low amounts of miR-181a . As a consequence , miR-181a/b-1−/− Treg cells had an increased suppressive capacity .
First , we tested the hypothesis that miR-181a/b-1 might play a role during Treg-cell development . Using a recombination activating gene 1–green fluorescent protein ( Rag1GFP ) knock-in allele to discriminate between nascent and mature thymus-resident or recirculating Treg cells , we found that frequencies and absolute numbers of de novo generated Rag1GFP-positive Treg cells were reduced by 2- to 3-fold in miR-181a/b-1−/− mice when compared to control ( ctrl ) , indicating that expression of miR-181a/b-1 is required for normal Treg-cell development in the thymus ( Fig 1A ) . Competitive bone marrow ( BM ) chimeras with 1:1 mixtures of donor cells from wild-type ( WT ) and miR-181a/b-1−/− mice revealed a disadvantage in Treg-cell generation , but not more immature double-negative and DP as well as CD4SP populations , from thymocytes of miR-181a/b-1−/− origin , indicating that miR-181a/b-1 controls Treg-cell formation in a cell-intrinsic manner ( Fig 1B and S1A Fig ) . In order to test how miR-181a/b-1 influenced developmental progression towards Foxp3+CD25+ Treg cells , we analyzed inducible Rag1 ( InduRag1 ) miR-181a/b-1−/− mice , in which a wave of T-cell development can be induced by transient initiation of Rag1 gene expression through a tamoxifen-inducible Cre recombinase ( Cre ) [30] . At day 7 after Rag1 induction , only a few CD4SP thymocytes were generated , precluding robust analysis of Treg cell development ( S1B and S1C Fig ) . However , we noted reduced frequencies of postselection DP thymocytes as well as postselection CD4SP thymocytes at this time point in miR-181a/b-1−/− mice when compared to ctrls ( S1D Fig ) . These data support the notion that selection processes are altered in the absence of miR-181a/b-1 . At day 14 and day 28 , frequencies of Treg cells within CD4SP thymocytes were lower in miR-181a/b-1−/− mice when compared to ctrls ( Fig 1C ) . Consistent with findings at steady state , these data indicate that Treg-cell formation is less efficient in the absence of miR-181a/b-1 . Next , we analyzed miR-181a/b-1–dependent formation of Foxp3−CD25+ ( prec 1b ) or Foxp3+CD25− ( prec 1a ) Treg-cell precursors in the InduRag1 model . At day 14 after induction , both intermediates were present at somewhat similar frequencies in miR-181a/b-1−/− mice when compared to ctrl , whereas at day 28 , we observed reduced frequencies of Foxp3+CD25− precursors and an accumulation of Foxp3−CD25+ precursors , suggesting that precursor generation is not restricted by a developmental block ( Fig 1C ) . Taken together , these data indicate that in the absence of miR-181a/b-1 , Treg cells are formed with slower kinetics rather than being subject to a defined developmental block . Treg-cell development in the thymus follows a somewhat different course in the absence of a full CD4SP compartment [10] . To account for such differences , we complemented the analysis of InduRag1 mice by taking advantage of the Rag1GFP knock-in allele described above to temporally separate Treg-cell development at steady state . Green fluorescent protein ( GFP+ ) cells were arbitrarily gated into 5 populations based on different GFP levels , with loss of GFP expression indicating increasing amounts of time since cessation of Rag gene expression , which occurs in DP thymocytes . Frequencies of precursor 1b were elevated in GFPhi cells from miR-181a/b-1−/− mice when compared to ctrls . In contrast , generation of precursor 1a as well as Treg cells was delayed in miR-181a/b-1−/− mice when compared to ctrls ( Fig 1D ) , which is consistent with data obtained in the InduRag1 model . We conclude that loss of miR-181a/b-1 results in an overall delay of Treg-cell formation , which is likely to be initiated prior to the emergence of defined Treg-cell precursors and cannot be compensated for by increased frequencies of Foxp3–CD25+ precursors . In order to test whether TCR signal strength differed in thymocytes from miR-181a/b-1−/− mice , we assessed expression of Nuclear hormone receptor NUR/77 ( Nur77 ) as a surrogate marker . Of note , miR-181a/b-1−/− DP cells expressed lower levels of Nur77 prior to stimulation when compared to ctrls ( Fig 2A ) . Furthermore , ex vivo stimulation of miR-181a/b-1−/− DP cells failed to induce WT levels of Nur77 , together suggesting that miR-181a/b-1−/− DP cells received weaker TCR signals and failed to respond to TCR triggering with the same sensitivity as their WT counterparts ( Fig 2A ) . In order to test whether TCR signaling was impaired prior to Treg-cell generation , we assessed surface expression of CD5 , which correlates with TCR signal strength , on thymocyte subsets [31] . At steady state , total DP thymocytes , the majority of which have not undergone selection , displayed similar levels of surface CD5 in the presence and absence of miR-181a/b-1 ( Fig 2B ) . However , CD4SP thymocytes from miR-181a/b-1−/− mice displayed lower surface levels when compared to ctrls . tTreg cells from either genotype expressed similar levels of Nur77 transcripts ( Fig 2C ) , together indicating that miR-181-a/b-1 limits TCR signal strength prior to the emergence of Treg cells . Next , we took advantage of a system mimicking increased TCR signal strength during Treg-cell development via inducible nuclear translocation of the Nur77 family member Nr4a2 [32] . BM chimeric mice were generated to carry miR-181a/b-1–deficient or miR-181a/b-1–sufficient ovalbumin-specific MHC class II-restricted alpha beta ( OT-II ) TCR transgenic cells expressing inducible Nr4a2 . Transduction with ctrl virus resulted in generation of low frequencies of Treg cells from miR-181a/b-1–sufficient mice , and even fewer Treg cells emerged from miR-181a/b-1−/− donor BM cells ( Fig 2D ) . Upon activation of Nr4a2 , frequencies of Treg cells generated from miR-181a+/− donors were slightly , albeit not significantly , reduced , suggesting that activation of Nr4a2 promotes a shift towards clonal deletion . In contrast , in the absence of miR-181a/b-1 , Treg-cell development was rescued upon activation of Nr4a2 , supporting the hypothesis that limited TCR signal strength accounts for defective Treg-cell differentiation in miR-181a/b-1−/− mice . To corroborate these data , we analyzed chimeric mice generated using OT-II TCR transgenic miR-181a/b-1–sufficient ( OT-II-ctrl ) or deficient ( OT-II-knockout [KO] ) donor cells transferred into RIPmOVA recipients . RIPmOVA mice express the cognate antigen for the OT-II TCR in the thymus , resulting in clonal deletion of OT-II TCR transgenic cells as well as generation of low numbers of Treg cells . OT-II-KO>RIPmOVA chimeras showed lower levels of clonal deletion and generated considerably lower numbers of Treg cells when compared to OT-II-ctrl>RIPmOVA chimeras ( Fig 2E ) . OT-II-ctrl>WT as well as OT-II-KO>WT chimeras failed to generate sizeable numbers of OT-II Treg cells and showed no signs of clonal deletion of OT-II thymocytes . We conclude that impaired generation of Treg cells in miR-181a/b-1–deficient mice is due to restricted TCR signal strength during thymic selection . To address potential consequences of impaired tTreg-cell development in the absence of miR-181a/b-1 , we next determined frequencies of Treg cells in the periphery . We did not observe any differences in frequencies and absolute numbers of peripheral Treg cells in spleens from miR-181a/b-1−/− mice compared to ctrls ( Fig 3A ) . Consistently , frequencies of recirculating or thymus-resident ( Rag1GFP-negative ) Treg cells in the thymus were largely unaffected in the absence of miR-181a/b-1 , but we observed reduced frequencies of Rag1GFP-positive recent thymic emigrants ( RTEs ) among peripheral Treg cells ( S2A and S2B Fig ) . Equilibration of Treg-cell numbers in the periphery can occur through homeostatic expansion of tTreg cells or preferential peripheral induction from naive T cells . Spleens of miR-181a/b-1–sufficient and deficient mice contained comparable frequencies of RTEs ( TCRβ+Rag1-GFP+ ) cells , which are enriched in peripheral Treg cell precursors ( S2B Fig ) [33] . Furthermore , CD4+CD25− RTEs from miR-181a/b-1−/− mice did not produce more iTreg cells upon transfer into lymphopenic interleukin-7 receptor α gene ( Il7r ) −/− hosts when compared to ctrls , suggesting that Treg-cell induction is not the primary mechanism to equilibrate peripheral Treg-cell numbers in miR-181a/b-1−/− mice ( S2C Fig ) . Chimeric mice generated with 1:1 mixtures of miR-181a/b-1−/− and WT BM showed that miR-181a/b-1−/− Treg cells had a competitive disadvantage in the periphery when WT Treg cells were present , indicating that niche availability permits homeostatic expansion of tTreg cells in miR-181a/b-1−/− mice ( Fig 3B ) . This conclusion was supported by Helios staining and TCR repertoire analysis . Elevated Helios expression has been associated with Treg-cell activation and proliferation [34] . Comparison of tTreg cells from miR-181a/b-1–sufficient and deficient mice showed low and similar expression of Helios between genotypes ( Fig 3C ) . Staining in peripheral lymphoid organs ( spleen , subcutaneous lymph node ( scLN ) , and mesenteric lymph node [mLN] ) revealed elevated numbers of Helios+ Treg cells in miR-181a/b-1−/− mice , indicating that in these mice , more Treg cells are in an activated/proliferative state ( Fig 3C ) . We predicted that limited de novo generation and increased peripheral expansion resulted in reduced TCR diversity in peripheral Treg cells in the absence of miR-181a/b-1 . Comparison of numbers of unique CDR3 sequences as well as calculation of effective number of species as a measure for repertoire diversity showed that TCR diversity in peripheral Treg cells from miR-181a/b-1−/− mice was significantly reduced ( Fig 3D and S3A Fig ) . In contrast , in the thymus , Treg cells from miR-181a/b-1−/− mice displayed comparable TCR diversity as their ctrl counterparts ( Fig 3E and S3A Fig ) . Thus , these data indicate that , as a consequence of less efficient generation , fewer clones egress from the thymus to be available for peripheral expansion . Next , we assessed whether impaired generation in the thymus altered the phenotype of Treg cells in the absence of miR-181a/b-1 . Expression of miR-181a progressively decreases after thymocytes exit the DP stage , with CD4+SP thymocytes and thymic Treg cells expressing 2-fold and 15-fold lower levels , respectively ( Fig 4A ) . In the periphery , expression of miR-181a was further reduced to 30-fold and 75-fold lower levels for CD4+ conventional T cells ( Tconv ) and Treg cells , respectively , when compared to DP thymocytes . Peripheral miR-181a/b-1−/− Treg cells showed a virtually identical global gene expression profile to their ctrl counterparts . Notably , this was also the case for key Treg-cell signature genes ( Fig 4B ) . Comparison of transcriptomes of thymic miR-181a/b-1−/− Treg cells with their WT counterparts also revealed no significant differences both globally as well as with regard to Treg-cell signature genes ( Fig 4C ) . Given that miRNAs may also act on a post-transcriptional level , we next assessed expression levels of Treg-cell signature receptors and transcription factors . Peripheral and tTreg cells from miR-181a/b-1−/− mice showed similar expression levels of most surface receptors and transcription factors analyzed as WT ctrls ( S4A , S4B and S4C Fig ) . Notably , we detected strongly increased levels of total CTLA-4 protein in both peripheral and tTreg cells from miR-181a/b-1−/− mice when compared to ctrls ( Fig 4D ) . Although the coding sequence of Ctla4 mRNA contains putative miR-181 binding sites , direct modulation of CTLA-4 expression by miR-181 could not be observed in luciferase assays ( S5A and S5B Fig ) . We conclude from these data that thymic generation in the absence of miR-181a/b-1 results in post-transcriptionally controlled up-regulation of CTLA-4 protein in Treg cells while other Treg-cell signature genes remain unaffected . Loss of miR-181a expression in WT peripheral Treg cells suggests that elevated expression of CTLA-4 in these cells is imprinted during development . In order to understand the underlying mechanisms of how elevated levels of CTLA-4 protein are maintained in peripheral miR-181a/b-1−/− Treg cells , we analyzed its intracellular distribution using confocal microscopy . We confirmed elevated expression of CTLA-4 protein in the absence of miR-181a/b-1 ( Fig 5A ) . Next , we determined intracellular localization of CTLA-4 by costaining for Ras-related in brain protein 11 ( Rab11 , marking recycling endosomes ) , lysosome-associated membrane protein 2 ( LAMP2 , late endosomes ) , early endosome antigen 1 ( EEA1 , early endosomes ) , and cis-Golgi matrix protein 130 ( GM130 , Golgi ) . We detected no differences in the extent of colocalization of CTLA-4 with early endosomes and the Golgi apparatus in Treg cells from miR-181a/b-1−/− mice compared to ctrls ( Fig 5B ) . However , we noted reduced colocalization of CTLA-4 with recycling endosomes but a marked increase in colocalization with late endosomes in miR-181a/b-1−/− Treg cells when compared to ctrls ( Fig 5C ) . In order to assess whether aberrant localization of CTLA-4 in miR-181a/b-1−/− Treg cells affected protein degradation , we stimulated Treg cells in the absence or presence of the translation inhibitor cycloheximide . Inhibition of translation for 2 h reduced CTLA-4 protein to similar levels in miR-181a/b-1−/− Treg cells and ctrls ( Fig 5D ) . Given the higher protein levels when translation is active , these data imply that degradation rates of CTLA-4 are higher in the absence of miR-181a/b-1 , which is consistent with its preferential localization in late endosomes rather than recycling endosomes . Furthermore , these data predict that if protein degradation is intact , elevated levels of CTLA-4 protein arise as a result of increased rates of protein translation . To test this prediction , we assessed accumulation of CTLA-4 protein in Treg cells in the presence of bafilomycin , an inhibitor of lysosomal protein degradation , in vitro . Over the course of 3 hours , miR-181a/b-1−/− Treg cells accumulated significantly more CTLA-4 protein when compared to their WT counterparts ( Fig 5E ) . Together , these data indicate that elevated levels of CTLA-4 in peripheral Treg cells in the absence of miR-181a/b-1 are due to increased rates of translation . Increased protein translation in the absence of altered mRNA levels may be induced by loss of miRNAs other than miR-181a/b-1 . To test this possibility , we performed small RNA sequencing ( small RNAseq ) of miR-181a/b-1–sufficient and deficient peripheral Treg cells . Consistent with the overall small changes in the transcriptome , we identified 4 miRNAs ( miR-15b , miR-150 , miR-342 , and lethal ( let ) -7g ) that were moderately down-regulated in Treg cells from miR-181a/b-1−/− mice ( S5C and S5D Fig ) . However , in silico analysis of Ctla4 mRNA using Targetscan7 and RNA22 provided no evidence for the existence of either canonical or noncanonical binding sites for any of these miRNAs , suggesting that elevated protein levels of CTLA-4 are not caused by reduced miRNA expression . Next , we assessed whether peripheral expansion contributed to sustained expression of CTLA-4 in peripheral Treg cells . To this end , we generated mixed BM chimeras and analyzed CTLA-4 levels on miR-181a/b-1−/− and WT Treg cells isolated from the same mice . CTLA-4 levels were consistently higher in miR-181a/b-1–deficient Treg cells despite competition by their WT counterparts , indicating that CTLA-4 levels are regulated cell-intrinsically and do not depend on peripheral expansion ( Fig 5F ) . Finally , we tested whether alterations in tonic TCR signaling might result in elevated expression of CTLA-4 . Freshly isolated peripheral Treg cells from miR-181a/b-1−/− and WT mice expressed comparable levels of Nr4a1 mRNA , suggesting that tonic signaling through the TCR is similar ( Fig 5G ) . Taken together , these findings further support the hypothesis that miR-181a/b-1–dependent ctrl of CTLA-4 expression is elicited in the thymus and subsequently sustained in the periphery . In order to test the functional consequences of elevated levels of CTLA-4 in miR-181a/b-1−/− Treg cells , we assessed their suppressive capacity . First , loss of miR-181a/b-1 resulted in reduced levels of tumor necrosis factor ( TNF ) α , IL-4 , and IL-2 by conventional CD4+ T cells ( Fig 6A ) . In contrast , expression of IL-17 and IL-10 remained unaffected . Whereas levels of TNFα were also reduced in miR-181a/b-1−/− Treg cells , we did not observe additional significant miR-181a/b-1–dependent alterations in cytokine production by Treg cells or CD8+ T cells ( Fig 6A and S6A Fig ) . Such alterations in cytokine profiles might be due to cell-intrinsic effects or due to modulation of Treg-cell suppressive capacity . Therefore , we next assessed in vivo suppressive capacity of Treg cells by transfer of congenically marked 1:1 mixtures of miR-181a/b-1–sufficient or miR-181a/b-1–deficient Treg cells ( CD45 . 2 ) and conventional T cells ( CD45 . 1 ) into Rag1−/− recipients . Suppression of lymphopenia-driven expansion of Tconv cells is dependent on CTLA-4 [35] . At 14 days after transfer , homeostatic expansion of Tconv cells was assessed . We observed lower frequencies of Tconv cells ( CD45 . 1 ) in spleens of mice co-transferred with miR-181a/b-1−/− Treg cells when compared to those co-transferred with miR-181-a/b-1–sufficient Treg cells ( Fig 6B ) . This reduction in frequency was reflected by lower absolute numbers of Tconv cells recovered in the presence of miR-181a/b-1−/− Treg cells , whereas absolute numbers of Treg cells recovered were similar in both conditions ( Fig 6C ) . Together , these data indicate that in the absence of miR-181a/b-1 , Treg cells have a stronger capacity to suppress lymphopenia-driven expansion of Tconv cells in vivo . Of note , we did not observe significant alterations in suppressive capacity of miR-181a/b-1−/− Treg cells in vitro ( S6B Fig ) . Taken together , our data indicate that miR-181a/b-1 controls intrathymic Treg cell development in a TCR-dependent manner . Impaired Treg-cell development in the absence of miR-181a/b-1 is associated with post-transcriptional up-regulation of CTLA-4 , which penetrates into the periphery and results in increased suppressive capacity . Low levels of miR-181a in peripheral WT Treg cells suggest that the effects of loss of miR-181a/b-1 are imprinted during intrathymic development .
Here , we demonstrated that intrathymic generation of Treg cells depends on miR-181a/b-1 via establishing signaling thresholds to adequately respond to strong TCR signals . In the absence of miR-181a/b-1 , de novo generation of Treg cells was attenuated and resulted in Treg cells expressing elevated levels of CTLA-4 . Homeostatic expansion resulted in a completely filled peripheral Treg-cell compartment while CTLA-4 levels remained elevated via a post-transcriptional mechanism , resulting in Treg cells with increased suppressive capacity . Treg cells develop from CD4SP thymocytes through two possible intermediates , Foxp3−CD25+ and Foxp3+CD25− [8 , 11] . It has been suggested that generation of these precursors occurs through a TCR-dependent step , whereas further maturation into mature Foxp3+CD25+ Treg cells is dependent on the cytokines IL-2 and IL-15 [8 , 10 , 11] . Analysis of InduRag1 mice as well as a Rag1GFP virtual timer indicated that miR-181a/b-1 predominantly affects formation of Foxp3+CD25− precursors , whereas Foxp3−CD25+ are more frequent in miR-181a/b-1–deficient mice . Nevertheless , these precursors cannot compensate for the partial loss of Foxp3+CD25− precursors , suggesting that the major route of Treg-cell development is through the latter . Indeed , it has been shown that in WT mice , only approximately 20% of CD4+Foxp3−CD25+ cells ultimately give rise to Treg cells [8 , 10] . Treg-cell development via Foxp3−CD25+ intermediates predominantly occurs in double transgenic mouse lines expressing a transgenic TCR plus its cognate antigen [9] . Furthermore , these cells express higher levels of a Nur77GFP reporter than either Foxp3+CD25− precursors or mature Treg cells [10] . Thus , it has been suggested that Foxp3−CD25+ intermediates arise at the extreme end of the TCR affinity spectrum and might increase in frequency by an influx of cells otherwise targeted for clonal deletion . Accordingly , reduction in MHCII levels on thymic epithelial cells in a monoclonal system diverted thymocytes from clonal deletion into the Treg cell lineage [16] . TCR repertoire analyses and autoreactivity suggest that TCR signal strength required for tTreg-cell generation overlaps both with that of positively selected thymocytes as well as that of cells normally undergoing clonal deletion [3 , 13 , 14] . Rescue experiments performed in this study agree with both non-mutually exclusive models . Fewer donor-derived miR-181a/b-1−/− OT-II Treg cells developed in RIPmOVA antigen transgenic mice compared to WT OT-II Treg cells . Concomitantly , clonal deletion of miR-181a/b-1−/− OT-II cells was also impaired in RIPmOVA mice , suggesting that in this particular paired TCR–antigen model , TCR signal strength is reduced through loss of miR-181a/b-1 to limit both Treg-cell formation and clonal deletion . Conversely , induced expression of Nr4a2 promoted Treg-cell production in the absence of miR-181a/b-1 but resulted in somewhat limited production of Treg cells in the presence of miR-181a/b-1 , suggesting that in the latter case , clonal deletion might be favored over Treg-cell development . Intrathymic development of Treg cells depends on CD28-mediated costimulatory signals [36] . Thus , it might be possible that elevated expression of CTLA-4 by Treg cells in the thymus contributes to impaired development . Costimulation via CD28 is required for efficient generation of Foxp3−CD25+ Treg-cell precursors but less so during later Treg-cell development , suggesting that CD28 signaling protects thymocytes from clonal deletion [36–39] . Loss of CD28 signaling does not result in export of autoreactive cells into the periphery , indicating that it does not simply act as an amplifier of TCR signal strength [39] . Consistently , loss of CD28-mediated costimulation and loss of miR-181a/b-1 generate phenotypically distinct developmental defects , also supporting the notion that elevated levels of CTLA-4 in miR-181a/b-1−/− Treg cells are a consequence rather than cause of inefficient generation of Treg cells in these mice . Consequences of altered TCR signal strength in the thymus have been previously analyzed in mice carrying hypomorphic mutations of key signal mediators , such as zeta-chain–associated kinase of 70 kD ( Zap-70 ) , or reduced numbers of immunoreceptor tyrosine-based activation motifs ( ITAMs ) within CD3ζ molecules [40–42] . Collectively , these studies showed that alterations in TCR signal affected positive and negative selection as well as Treg-cell formation , albeit in a manner that is not easily predictable . Thus , these data indicate that the quantitative relationship between proximal TCR signaling and effcient thymic selection needs to be tightly balanced . Furthermore , mutations characterized in these studies equally affect T-cell activation and tonic signaling in the periphery , precluding analysis of developmental consequences of altered TCR signaling exclusively occurring in the thymus . In contrast , expression levels of miR-181a/b-1 in peripheral Treg cells are very low and should therefore allow WT-like levels of tonic TCR signaling . Notably , peripheral TCR signaling controls Treg-cell homeostasis and helps to maintain functional Treg cells [43 , 44] . For instance , in the absence of peripheral TCR expression , levels of CTLA-4 protein are reduced , and suppressive capacity is compromised [44] . Our data suggest that alterations in thymic selection caused by the absence of miR-181a/b-1 have long-term impact and are translated to increased suppressive activity of peripheral Treg cells . We therefore propose that the developmental legacy of TCR signal strength during agonist selection determines Treg-cell function in the periphery . Thus , altered TCR thresholds during selection might affect a Treg cell’s responsiveness to tonic signaling . Similar observations have previously been reported for both CD4 and CD8 Tconv cells [45 , 46] . The avidity of positively selecting self-peptides and thus strength of the TCR signal during selection determines the outcome of a T-cell immune response even of T cells recognizing the same foreign antigen with an identical avidity [45] . In contrast to Treg cells , differential reactivity to self-peptide by CD8 Tconv cells was accompanied by clear differences in gene expression profiles [46] . Although in Tconv cells , the capacity for tonic signaling in the periphery contributes to distinct responsiveness to pathogens , thymically predetermined levels of the feedback regulator of TCR signaling CD5 are likely to help control tonic signals [45] . Thus , the quality of protective T-cell responses as well as Treg-cell mediated suppression appear to be preset during thymic selection . How TCR signal strength during thymic agonist selection confers long-term changes in CTLA-4 protein expression remains unclear . Our study supports a model in which expression of CTLA-4 is cell-intrinsically sustained in the periphery through a post-transcriptional mechanism controlling its translation rate . Translational control can be exerted at multiple different levels , including changes in mRNA composition through alternative splicing and miRNAs as well as RNA binding proteins . Given the lack of umambiguous evidence for one of these mechanisms being predominant , our data suggest that multiple factors may act in concert to control CTLA-4 protein . Tight control of CTLA-4 expression is likely to be paramount for tuning suppressive capacity of Treg cells . Our study establishes miR-181a/b-1 as a central regulator of agonist-selected αβT cells . Earlier studies showed that miR-181a/b-1 is critical for the development of innate-like T cells expressing semi-invariant TCRs , such as iNKT cells and MAIT cells ( but not their γδTCR-expressing counterparts ) [25 , 26 , 47] . Here , we demonstrated that the role of miR-181a/b-1 can be extended to highly diverse polyclonal T-cell populations . This finding was not anticipated because it might be expected that a shift in integrated signal strength might be compensated for by a shift in the repertoire . Such compensation might partly explain why the effect of miR-181a/b-1 deletion on Treg cells is somewhat milder when compared to other agonist selected T cells . Finally , based on the dramatic down-regulation of miR-181a after the DP stage , our study implies that the lineage fate decision to become a Treg cell manifests itself early during selection .
All experiments were performed in accordance with German law on care and use of laboratory animals or with the institutional and ethical guidelines of the University of Edinburgh and have been approved by the Niedersächsisches Landesamt für Verbraucherschutz und Lebensmittelsicherheit ( LAVES ) ( 33 . 12-42502-04-08-1480; 11/0533; 12/0869; 13/1224; 15/1846 ) , the Regierungspräsidium Darmstadt ( V54–19c20/15–FU/1119; FU/1155; FU/1159; FU/1178 ) or under a project license granted by the UK Home Office ( PPL60_4510 procedure 19b 5 ) , respectively . Animals were killed by CO2 inhalation , in some cases followed by cervical dislocation . MiR-181a/b-1−/− and miR-181a/b-1+/− mice ( B6 . Mirc14tm1 . 1Ankr ) were generated as described in [26] and bred at the Hannover Medical School and Goethe University , Frankfurt , Germany . OT-II mice ( B6 . Cg-Tg ( TcraTcrb ) 425Cbn/J ) were purchased from The Jackson Laboratory , crossed with miR-181a/b-1−/− and miR-181a/b-1+/− mice , and bred at the Hannover Medical School . RIPmOVA mice ( C57BL/6-Tg ( Ins2-TFRC/OVA ) 296Wehi/WehiJ ) were purchased from The Jackson Laboratory and crossed with B6 CD45 . 1 . F1 mice ( RIPmOVA CD45 . 1/2 ) were bred at the Hannover Medical School . C57BL/6J mice ( CD45 . 2 ) , B6 . SJL-PtprcaPepcb/BoyJ mice ( termed “B6 CD45 . 1” throughout this manuscript ) , and IL-7Rα–deficient mice ( B6 . 129S7-Il7rtm1Imx/J ) were purchased from Charles River or The Jackson Laboratory or bred at the animal facility of Hannover Medical School . ( C57BL/6J × B6 CD45 . 1 ) F1 mice ( CD45 . 1/CD45 . 2 heterozygous ) and IL-7Rα−/− CD45 . 1 mice were bred at the animal facility of Hannover Medical School . InduRag1fl/fl × Rosa26-CreERT2 mice ( termed InduRag1 here ) were obtained by crossing InduRag1fl/fl mice with Rosa26-CreERT2 mice ( kindly provided by Prof . Anton Berns , The Netherlands Cancer Institute , Amsterdam , The Netherlands ) and bred at the Helmholtz Centre for Infection Research [30] . These mice were crossed to miR-181a/b-1−/− ( InduRag1 × miR-181a/b-1−/− ) and miR-181a/b-1+/− ( InduRag1 × miR-181a/b-1+/− ) mice bred and maintained at the Helmholtz Centre for Infection Research in Braunschweig . Foxp3hCD2 × Rag1GFP mice were obtained by crossing Foxp3hCD2 mice [48] with Rag1GFP mice [49] ( kindly provided by Nobuo Sakaguchi , Kumamoto , Japan ) . These mice were crossed to miR-181a/b-1−/− and miR-181a/b-1+/− mice and bred at the Helmholtz Centre for Infection Research in Braunschweig . Rag1−/− mice and B6 CD45 . 1 mice used in the in vivo suppression assay were bred and maintained at the University of Edinburgh . All mice were used between 7–12 weeks of age and were maintained under specific-pathogen–free conditions . Treg-cell development was induced in 6- to 8-week–old InduRag1 × miR-181a/b-1−/− and InduRag1 × miR-181a/b-1+/− mice by oral administration of tamoxifen ( Ratiopharm , Ulm , Germany ) in corn oil ( Sigma Aldrich , St . Louis , MO , USA ) , 0 . 6 mg/400 μl/mouse every 4 days . To ensure proper tamoxifen solubility , in the first step , it was dissolved in 100% ethanol prewarmed to 55 °C and later in prewarmed corn oil . Before each oral administration , the mixture was always prewarmed to 37 °C . B6 CD45 . 1+ recipient mice were lethally irradiated ( 9 Gy ) . Donor BM from both WT ( CD45 . 1/2+ ) and miR-181a/b-1+/− or miR-181a/b-1−/− ( CD45 . 2+ ) mice mixed at a 1:1 ratio was injected into the lateral tail vein within 24 h postirradiation ( a total of 4 × 106 cells/mouse ) . Mice were provided with antibiotic-containing water and were housed in sterile microisolator cages . Analysis of BM chimeras was performed 8–12 weeks after transplantation . BM cells were isolated from the tibia and femur of age-matched ( 7–10 weeks ) OT-II × miR-181a−/− and OT-II × miR-181a+/− mice . First , lineage-negative ( lin− ) cells were enriched by incubation with the cocktail of purified rat-anti-mouse antibodies ( CD19 [1D3] , CD11b [M1/70] , Gr-1 [RB6-8C5] , NK1 . 1 [PK136] , Ter-119 , CD4 [RM4 . 4] , and CD8α [53–6 . 7]; all from eBioscience , Thermo Fisher Scientific , Waltham , MA , USA ) followed by incubation with sheep-anti-rat dynabeads ( Dynal; Invitrogen , Carlsbad , CA , USA ) and magnetic separation . Cells were stained with a phycoerythrin ( PE ) -Cy7–conjugated cocktail of lin antibodies ( CD19 [6D5] , CD11b [M1/70] , Gr-1 [RB6-8C5] , NK1 . 1 [PK136] , Ter-119 , CD4 [GK1 . 5] , and CD8α [53–6 . 7]; all from eBioscience ) and Sca-1 ( E13–161 . 7 ) PacificBlue , c-kit ( CD117 ) allophycocyanin ( APC ) , and sorted as lin−Sca-1+c-kit+ to 98% purity . Cells were cultured overnight at a density of 5 × 104 cells/100 μl ( on 96-well plate ) with mouse IL-6 ( 20 ng/ml ) , mouse IL-7 ( 25 ng/ml ) , mouse Flt-3L ( 25 ng/ml ) , and mouse SCF ( 50 ng/ml; all obtained from PeproTech , Rocky Hill , NJ , USA ) , in DMEM containing 10% FBS . Next day , cells were infected by centrifugation at 700 g for 45 min at 32 °C on retronectin-coated plates ( 50 μg/ml , 4 °C overnight ) loaded ( 3× 100 μl/well , 2 , 000 g , 20 min at 32 °C ) with Nr4a2-ΔLBD-ERT2–expressing retrovirus or ctrl retrovirus ( both carrying Thy1 . 1 tag detectable by flow ctometry ) in the presence of polybrene ( 8 μg/ml ) and the abovementioned cytokines . Fresh medium was supplemented after 24 h . After 48 h , cells were collected , and transduction efficacy ( frequency of Thy1 . 1+ cells ) was determined . Routinely , it was approximately 60% , and thus cells were later sorted to 100% of Thy1 . 1+ . B6 CD45 . 2+Thy1 . 2+ recipient mice were lethally irradiated ( 9 Gy ) . Sorted cells were intravenously injected ( 1 × 105/recipient ) . Mice were provided with antibiotic-containing water and were housed in sterile microisolator cages . Six weeks after injection , mice were orally administered 0 . 6 mg/400 μl of tamoxifen ( Ratiopharm ) dissolved in corn oil ( Sigma Aldrich ) per recipient mouse for 5 consecutive days . Tamoxifen preparation was performed as for induction of Treg-cell development in InduRag1 mice . Twelve hours after the final administration , mice were analyzed . Thymocytes isolated from miR-181a+/− and miR-181a−/− mice were left untreated or were stimulated for 3 h at 37 °C in the presence of plate-bound αCD3 ( 17A2 , 2 . 5 μg/ml ) and soluble αCD28 ( 37 . 51 , 2 . 5 μg/ml ) . Next , cells were stained for surface CD4 and CD8α and intracellular Nur77 with intracellular staining buffer set ( eBioscience ) . BM cells were isolated from the tibia and femur of age-matched ( 7–10 weeks ) OT-II × miR-181a/b-1−/− and OT-II × miR-181a/b-1+/− mice . Red blood cells were lysed , and cells were counted and injected ( 5 × 106/recipient ) into the lateral tail vein of lethally ( 2 × 4 . 5 Gy ) irradiated RIPmOVA ( CD45 . 1/2+ ) recipient mice . Mice were provided with antibiotic-containing water and were housed in sterile microisolator cages . Analysis of BM chimeras was performed 8–12 weeks after transplantation . Monoclonal antibodies specific for CD4 ( RM4 . 4 , GK1 . 5 ) , CD8α ( 53–6 . 7 ) , CD25 ( PC61 . 5 , eBio7D4 for sort ) , CD24 ( M1/69 ) , CD27 ( LG . 3A10 ) , CD28 ( 37 . 51 ) , CD103 ( M290 ) , CD122 ( TM-β1 ) , CD127 ( A7R34 ) , Gr-1 ( RB6-8C5 ) , erythroid cell marker ( Ter-119 ) , CD19 ( 1D3 , 6D5 ) , CD11b ( M1/70 ) , CD45 . 1 ( A20 ) , CD45 . 2 ( 104 ) , CD117 ( c-kit ) ( 2B8 , ACK2 ) , Sca-1 ( E13–161 . 7 ) , NK1 . 1 ( PK136 ) , CD11c ( N418 ) , TCRβ ( H57-597 ) , Foxp3 ( MF23 , FJK-16s ) , human CD2 ( RPA-2 . 10 ) , Qa2 ( 1-1-2 ) , Nur77 ( 12 . 14 ) , Vβ5 . 1/5 . 2 TCR ( MR9-4 ) , Vα2 TCR ( B20 . 1 ) , CTLA-4 ( UC10-4B9 , UC10-4F10-11 ) , Helios ( 22F6 ) , TGF-βR ( RI/ALK-5 ) , GITR ( DTA-1 ) , PLZF ( Mags . 21F7 ) , Gata3 ( TWAJ ) , Egr2 ( erongr2 ) , Irf4 ( 3E4 ) , c-Rel ( 1RELAH5 ) , and KLRG1 ( 2F1 ) were used as biotin , PacificBlue , fluorescein isothiocyanate ( FITC ) , Alexa Fluor 488 , PE , peridinin chlorophyll protein-Cy5 . 5 ( PerCP-Cy5 . 5 ) , PE-Cy7 , APC , APC-Cy7 , APC-eFluor780 , Brilliant Violet 421 , eFluor450 , and Alexa Fluor 647 conjugates . Antibodies were purified from hybridoma supernatants using standard procedures or were purchased from eBioscience , BD Biosciences ( San Jose , CA , USA ) , BioLegend ( San Diego , CA , USA ) , or R&D Systems ( Minneapolis , MN , USA ) . For intracellular stainings , an intracellular staining buffer set and a Foxp3/transcription factor staining buffer set ( both eBioscience ) were used according to the manufacturer’s protocol . RNA flow cytometry was performed using the PrimeFlow system ( Thermo Fisher Scientific ) . Cells were stained with Nr4a1-AF647 . Samples were acquired on LSRII ( BD Biosciences ) cytometers and sorted on a FACS Aria II ( BD Biosciences ) . Data were analyzed with FlowJo software , v . 9 . 4 . 9 ( Tree Star , Ashland , OR , USA ) . For analysis , dead cells and debris were excluded by gating of forward and side scatter . Sorted cells were of 98% or higher purity , as determined by reanalysis . RTEs depleted of Treg cells were sorted as Rag1GFP+CD4+CD8α−CD25− from the spleens of Foxp3hCD2 × Rag1GFP × miR-181a/b-1−/− and Foxp3hCD2 × Rag1GFP × miR-181a/b-1+/− mice . 5 × 105 cells were injected into the lateral tail vein of lymphopenic IL-7Rα−/− ( CD45 . 1 ) recipients . Spleens , peripheral lymph nodes ( pLNs: inguinal , brachial , axiliary , and cervical ) , and mLN were analyzed for the induction of Foxp3 within donor population ( CD45 . 2 ) 28 days later . Splenic antigen-presenting cells ( CD45 . 1+ ) were purified using sheep-anti-rat magnetic beads and depletion of CD19 , CD3 , NK1 . 1 , and Gr-1–positive cells as described above . Cells were plated on a U-bottom–shaped 96-well plate coated with αCD3 antibody ( 17A2 , 10 μg/ml ) at 4 °C , overnight , at the density 1 × 104 cells/100 μl/well . Next , they were loaded with OVA323–339 peptide ( 2 μg/ml; AnaSpec , Fremont , CA , USA ) for 45 min at 37 °C and washed 3 times to remove unbound peptide . Naïve , CD4+ OT-II cells ( CD45 . 1/2+ ) were purified using a CD4+T cell negative depletion kit ( Invitrogen ) . Cells were further labeled with 1 μM CFSE ( Molecular Probes , Eugene , OR , USA ) for 10 min at 37 °C and washed twice . They were added to peptide-loaded antigen-presenting cells at a density of 1 × 105 cells/well ( antigen-presenting cell/OT-II ratio 1:10 ) . Treg cells ( CD45 . 2+ ) were sorted as CD4+CD25+ cells from spleens of miR-181a/b-1+/− and miR-181a/b-1−/− mice ( purity always >98% ) , and their graded concentrations were added to antigen-presenting cell/OT-II cocultures in the presence of murine IL-2 ( 100 U/ml , PeproTech ) . Assay was analyzed after 48–60 h by flow cytometry . To analyze the function of Treg cells in vivo , sorted populations of naïve CD45 . 1+ ( 1 × 105 to 4 × 105 cells ) together with CD45 . 2+ miR-181a/b-1+/− or miR-181a/b-1−/− Treg cells were injected intravenously into Rag1−/− recipients and analyzed after 14 days . Total RNA was isolated using the RNeasy Mini Kit ( Qiagen , Hilden , Germany ) from 1 × 105 sorted thymic CD3+CD4+CD25+ miR-181a/b-1+/− or miR-181a/b-1−/− Treg cells and 4 × 105 sorted splenic CD3+CD4+CD25+ miR-181a/b-1+/− or miR-181a/b-1−/− Treg cells . Purity was above 98% as determined by reanalysis . Four independent sorts were performed ( pooled 4–5 mice/genotype ) , and so were 2 independent sequencing experiments . cDNA templates were synthesized using SuperScript II reverse transcriptase ( Invitrogen ) according to the manufacturer’s recommendation . To generate template libraries of rearranged TCR CDR3 regions from Treg-cell cDNA for the Genome Sequencer FLX system ( 454 sequencing; Roche , Basel , Switzerland ) , we used primers spanning the variable region between constant Cα and V elements of the Vα8 family ( comprising TRAV12-1*01 , TRAV12-1*03 , TRAV12-1*04 , TRAV12-1*05 , TRAV12D-2*01 , TRAV12D-2*02 , TRAV12D-2*03 , TRAV12D-2*04 , TRAV12D-2*05 , TRAV12D-3*01 , TRAV12D-3*02 , and TRAV12D-3*03 ) [50] . Forward and reverse primers contained at their 5′ ends the universal adapter sequences and a multiplex identifier ( MID ) , respectively . Amplicons were purified by agarose gel electrophoresis and QIAquick Gel Extraction Kit ( Qiagen ) and quantified by Quant-iT dsDNA HS Assay Kit ( Invitrogen ) . Single PCR amplicon molecules were immobilized onto DNA capture beads within an oil–water emulsion to enable clonal amplification in a second PCR process with universal primers . The emulsion was then disrupted , and isolated beads were loaded onto PicoTiterPlates . Sequencing reactions were performed by ultradeep 454 pyrosequencing on the Genome Sequencer FLX system ( Roche ) . Productive rearrangements and CDR3α regions were defined by comparing nucleotide sequences to the reference sequences from IMGT , the international ImMunoGeneTics information system ( http://www . imgt . org ) [51] . Rearrangements were analyzed and CDR3α regions were defined using IMGT/HighV-QUEST [52] . CD4+ T cells were purified from spleens and pLN of miR-181a/b-1+/− and miR-181a/b-1−/− mice using CD4+ T-cell negative isolation kit ( Dynal , Invitrogen ) . Next , cells were plated on glass coverslips ( Assistent , 0 . 13–0 . 16 mm , Thermo Fisher Scientific ) coated with poly-L-lysine ( Sigma Aldrich ) or Histogrip ( Invitrogen ) for 2 h at 37 °C . In some experiments , glass coverslips were additionally coated with αCD3 antibody ( 10 μg/ml , 17A2 ) and αCD28 antibody ( 10 μg/ml , 37 . 51 ) . After 2 h , cells were fixed in 3% ( v/v ) electron-microscopy–grade PFA ( Electron Microscopy Sciences ) . CTLA-4 , Foxp3 , LAMP2 , EEA1 , GM130 , and Rab11 were stained using indirect immunofluorescence . The following primary antibodies were used: purified α-mouse CTLA-4 ( UC10-4F10-11 , BD Biosciences ) labeled with DyLight650 antibody labeling kit ( Pierce , Thermo Fisher Scientific ) according to the manufacturer’s protocol , AlexaFluor488-labeled rat-α-mouse Foxp3 ( MF23 , BD Biosciences ) , purified rat-α-mouse LAMP2 ( Hybridoma Bank ) , α-mouse EEA-1 ( 14/EEA1 , BD Biosciences ) , α-mouse GM130 ( 35/GM130 , BD Biosciences ) , and α-mouse Rab11 ( D4F5 , Cell Signaling , Danvers , MA ) . The following secondary antibodies were used , all conjugated to AlexaFluor488 ( all Molecular Probes ) : mouse-α-rat , rat-α-mouse , and goat-α-rabbit . Nuclear staining was performed using DAPI ( Molecular Probes ) . Coverslips were mounted on glass slides using aqueous mounting medium ( DakoCytomation , Glostrup , Denmark ) . Samples were analyzed by confocal fluorescent microscopy using a Leica SP5 inverted microscope ( Leica Microsystems , Wetzlar , Germany ) . During imaging , a single focal plane was monitored in x-y-z scanning mode using 63×/1 . 4–0 . 6 NA oil HCX PL APO lambda blue DIC oil objective , UV laser ( 405 nm ) , argon laser ( 488 nm ) , diode-pumped solid-state ( DPSS ) laser ( 561 nm ) , and helium–neon ( HeNe ) laser ( 633 nm ) at a scanner frequency of 400 Hz , line averaging 4 . In order to avoid fluorescence overlap , sequential scans were performed . Images were analyzed using LAS AF Lite ( Leica Microsystems ) and ImageJ software . Quantification of fluorescence intensity and image analysis was performed using ImageJ software . CD4+CD25+ Treg cells were enriched from spleens and LNs of miR-181a/b-1+/− or miR-181a/b-1−/− mice using a MACS isolation kit ( Miltenyi Biotec , Bergisch Gladbach , Germany ) . Cells were incubated in the absence or presence of 50 nM bafilomycin ( InvivoGen , San Diego , CA , USA ) and collected after 30 , 60 , 120 , and 180 min . Samples were stained with αCD4 , αCD25 , and αhCD2 ( Foxp3 ) antibodies and intracellularly for CTLA-4 . For intracellular stainings , the Foxp3/transcription factor staining buffer set ( eBioscience ) was used according to the manufacturer’s protocol . Samples were acquired on LSRII ( BD Biosciences ) . Data were analyzed with FlowJo software , v . 9 . 4 . 9 ( Tree Star ) . For analysis , dead cells and debris were excluded via staining with Zombie Aqua reagent ( BioLegend ) prior to fixation and permabilization of cells . RNA isolation , cDNA preparation , and DNA microarray analysis of gene expression were performed at the Microarray Genechip Facility of the University of Tübingen ( MFT Services ) . Fluorescent images of hybridized microarrays ( MOE-430 version 2 . 0; Affymetrix , Santa Clara , CA , USA ) were obtained using an Affymetrix Genechip Scanner . Microarray data were analyzed using BioConductor Suite 2 . 1 software . All samples were repeated two times with individually sorted cells and averaged . Treg cells ( 1 × 105 ) sorted from 3 pooled WT and miR-181a/b-1−/− thymi were stored in RNAprotect Cell Reagent ( Qiagen ) . Small RNAseq was performed by Admera Health ( South Plainfield , NJ , USA ) using the SMARTer smRNA-Seq Kit ( Takara , Kusatsu , Japan ) . Adapters were trimmed with Flexbar 3 . 4 , and rRNA was removed using Bowtie 2 . The remaining reads were aligned using STAR aligner and counted using HTSeq . Differential expression analysis was performed in R using the DESeq2 package . Three biological replicates per genotype were analyzed . Part of the CTLA-4 coding sequence ( CTLA-4WT , 294 bp ) and a version devoid of the putative miR-181a binding site ( CTLA-4del , 271 bp ) were synthesized by GeneArt ( Regensburg , Germany ) and cloned into PsiCheck2 . 0 Vector ( Promega , Madison , WI , USA ) . 3T3 cells overexpressing murine miR-181a or ctrl cells were established by retroviral transduction and sorted to 100% purity ( GFP reporter ) . Viral particles were produced in HEK293T cells by co-transfection with pCLEco ( coexpressing MLV gag , pol , and env ) and the plasmids MDH1-PGK-GFP_2 . 0 ( Addgene plasmid #11375 ) or MDH1-miR-181a-1-PGK-GFP ( Addgene plasmid #11376 ) , which were gifts from Chang-Zheng Chen [53] . 3T3 cells were cultured in complete DMEM ( Life Technologies , Carlsbad , CA , USA ) ( 10% FCS , 100 U PenStrep , 1 mM Na-pyruvate , 25 μg/mL Geneticin [Life Technologies] ) . 250 , 000 cells were electroporated with 0 . 1 μg of PsiCheck2 . 0 ( 250 V , 950 μF , Biorad Gene Pulser II ) and cultured for 24 h on 6-well plates . Dual-Luciferase Reporter ( DLR ) assays ( Promega ) were conducted according to the manufacturer’s instructions . Luciferase activity was measured using Lumat LB 9507 machine ( Berthold ) . RNA was prepared using the miRNeasy Kit according to the manufacturer’s instructions ( Qiagen ) . RT reaction was performed using TaqMan MicroRNA Reverse Transcription Kit ( Applied Biosystems , Thermo Fisher Scientific , Waltham , MA , USA ) and miRNA-specific primers according to the manufacturer’s protocol . Quantitative RT-PCR analysis of miRNA expression was carried out using the following Taqman probes: hsa-miR-181a , TM: 000480; mmu-miR-15b , TM: 000390; mmu-miR-150-5p , TM: 000473; mmu-miR-342 , TM: 002260; mmu-let-7g-5p , TM: 002282 ( Applied Biosystems ) . Fold differences were calculated using the ΔCt method normalized to snoRNA412 as housekeeping miRNA gene ( Applied Biosystems , TM: 001243 ) . All analyses were performed using GraphPad Prism software . Data are represented as mean ± SD . Statistical analyses of significance were performed using unpaired or paired Student’s t test , multiple t test , or two-way ANOVA with Bonferroni post t test or Sidak’s multiple comparison test . | T cells are pivotal in orchestrating an adaptive immune response . They are produced in the thymus and undergo selection processes resulting in elimination of nonfunctional and self-reactive cells in order to prevent autoimmune disease . One type of T cells , called regulatory T cells ( Treg cells ) , is generated either in the thymus or in the periphery through a process termed agonist selection . Peripheral Treg cells are crucial for the suppression of unwanted immune responses . However , too much suppressive function of Treg cells can also impair immune responses directed against tumors . Here , we have analyzed the mechanisms that regulate this process and show that a microRNA , miR-181a/b-1 , controls de novo generation of Treg cells by modulating agonist selection . We show that thymic and peripheral Treg cells deficient in miR-181a/b-1 contain higher levels of the key effector molecule CTLA-4 , and as a consequence , such Treg cells display increased suppressive capacity . We observe that the expression of miR-181a/b-1 in peripheral Treg cells is much lower when compared to thymocytes; thus , our study implies that peripheral Treg-cell effector function is imprinted during intrathymic differentiation . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"blood",
"cells",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"forkhead",
"box",
"natural",
"antisense",
"transcripts",
"gene",
"regulation",
"immunology",
"micrornas",
"stem",
"cells",
"tcr",
"signaling",
"cascade",
"immune",
"system",
"proteins",
"white",
"blood",
"cells",
"animal",
"cells",
"proteins",
"gene",
"expression",
"t",
"cells",
"thymus",
"hematopoietic",
"progenitor",
"cells",
"immune",
"system",
"biochemistry",
"signal",
"transduction",
"t",
"cell",
"receptors",
"thymocytes",
"rna",
"cell",
"biology",
"nucleic",
"acids",
"genetics",
"protein",
"domains",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"regulatory",
"t",
"cells",
"immune",
"receptors",
"non-coding",
"rna",
"cell",
"signaling",
"signaling",
"cascades"
] | 2019 | miR-181a/b-1 controls thymic selection of Treg cells and tunes their suppressive capacity |
Routine entomological monitoring data are used to quantify the abundance of Ae . aegypti . The public health utility of these indicators is based on the assumption that greater mosquito abundance increases the risk of human DENV transmission , and therefore reducing exposure to the vector decreases incidence of infection . Entomological survey data from two longitudinal cohort studies in Iquitos , Peru , linked with 8 , 153 paired serological samples taken approximately six months apart were analyzed . Indicators of Ae . aegypti density were calculated from cross-sectional and longitudinal entomological data collected over a 12-month period for larval , pupal and adult Ae . aegypti . Log binomial models were used to estimate risk ratios ( RR ) to measure the association between Ae . aegypti abundance and the six-month risk of DENV seroconversion . RRs estimated using cross-sectional entomological data were compared to RRs estimated using longitudinal data . Higher cross-sectional Ae . aegypti densities were not associated with an increased risk of DENV seroconversion . Use of longitudinal entomological data resulted in RRs ranging from 1 . 01 ( 95% CI: 1 . 01 , 1 . 02 ) to 1 . 30 ( 95% CI: 1 . 17 , 1 . 46 ) for adult stage density estimates and RRs ranging from 1 . 21 ( 95% CI: 1 . 07 , 1 . 37 ) to 1 . 75 ( 95% CI: 1 . 23 , 2 . 5 ) for categorical immature indices . Ae . aegypti densities calculated from longitudinal entomological data were associated with DENV seroconversion , whereas those measured cross-sectionally were not . Ae . aegypti indicators calculated from cross-sectional surveillance , as is common practice , have limited public health utility in detecting areas or populations at high risk of DENV infection .
Dengue virus ( DENV ) , which is transmitted by the bite of female Aedes aegypti mosquitoes , causes more human morbidity and mortality than any other arthropod-borne virus [1] . Since the 1950s , dengue has spread via the globalization of trade and travel , rapid urbanization and expansion of vector habitats [2] . The four serotypes ( DENV1 , DENV2 , DENV3 and DENV4 ) occur throughout the tropics and infect approximately 390 million persons per year [1] . Until effective DENV vaccines become broadly commercially available , vector control will remain the primary prevention strategy in most dengue endemic settings [3] and even as vaccines become accessible vector control will be needed to supplement vaccine efforts [4] , as well as control of other arboviruses also vectored by Ae . aegypti . The World Health Organization recommends monitoring vector abundance for the targeting and evaluation of vector control interventions [5] . Ae . aegypti monitoring was first employed in yellow fever control programs in the first half of the 20th century [6 , 7] . Since then , over two dozen indicators have been proposed to quantify abundance of Ae . aegypti . Entomological monitoring data are typically collected from households sampled from neighborhoods or blocks on a routine or ad hoc basis [8] . The frequency of entomological data collection also varies by setting , and WHO guidelines recommend implementation occur at a frequency from “weeks to months” [5] . As such , entomological monitoring surveys impose cross-sectional measurement of the highly dynamic Ae . aegypti population . Monitoring indicators vary by mosquito life stage ( adults , larvae and/or pupae ) , availability of larval development sites ( infestation indices ) , and process of collection ( fixed trap or human-based surveys such as adult aspirator collections , household inspection for larvae ) [9] . The public health utility of these indicators is based on the assumption that greater mosquito abundance increases the risk of DENV transmission , and therefore reducing exposure to the vector decreases incidence of infection . Further , by identifying “hot spots” of Ae . aegypti infestation , targeted vector control would be an efficient use of limited intervention resources [10] . To date , studies have not shown a consistent association between various indices and infection or disease outcomes [7] . This may be due to several limitations inherent to the large-scale measurement of Ae . aegypti densities . First , there is no established threshold of Ae . aegypti density associated with an increased risk of human DENV infection [11] . Second , entomological survey techniques may not capture the fine spatial and temporal variability in an urban setting due to the constraints dictated by household-based monitoring , and the fact that indices are calculated from cross-sectional prevalence measures , not derived from continuous monitoring . Third , while adequate sampling of immature and adult populations requires consideration of vector dynamics [12] and spatial relationships [13] , the data do not capture the daily productivity of individual larval development sites or the activity of individual mosquitoes over their lifespan [8 , 14] . Finally , previous attempts to quantify the association between vector abundance and dengue outcomes may also have been biased due to measurement error caused by operational constraints and collection procedures [9] , and methodological issues , such as restricting the analysis outcomes to infected people who sought treatment or small sample size [7] . Ae . aegypti densities may also fail to describe risk of DENV infection due to the complexity of transmission . The probability of transmission is dependent on human movement to introduce DENV into mosquito populations and the presence of susceptible individuals that mosquitoes infect to perpetuate new rounds of transmission [15] . Because Ae . aegypti are daytime-biting mosquitoes that are highly adapted to the human urban environment [16] , their frequent biting contact with susceptible human hosts is mediated by social and economic [17] factors that govern human movement through times and spaces where they encounter mosquitoes [18] . While high concentrations of Ae . aegypti within or around a household present an opportunity for clustered DENV transmission , it ignores transmission occurring in other places [19 , 20] . To help predict risk and direct public health interventions , there is substantial interest in an improved understanding of the utility of Ae . aegypti monitoring measures in terms of an association with DENV infection , according to mosquito life stage and spatial scale of measurement . We aimed to systematically examine measures of entomological risk collected through routine household surveillance with human DENV infection using longitudinal entomological and human serology data to test associations between Ae . aegypti indices and the 6-month risk of DENV seroconversion .
Written informed consent ( and assent for children 8–17 years of age ) was obtained for all individuals providing serological data . Written informed consent was provided by parents or guardians for children under 18 years of age . Written consent ( 1999–2003 ) or oral consent ( 2008–2010 ) was obtained from an adult head of household for entomological surveys as approved by the institutional review boards . Oral consent for entomological surveys was documented upon obtaining access to the household and heads of households were provided information sheets describing the data collection procedures . Data collection procedures were approved by the University of California , Davis ( Protocols 2002–10788 and 2007–15244 ) , Instituto Nacional de Salud , and Naval Medical Research Center Institutional Review Boards ( Protocols NMRCD . 2001 . 0008 and NMRCD2007 . 0007 ) . This ancillary analysis was approved by the Institutional Review Board at the University of North Carolina at Chapel Hill ( Study # 14–3151 ) . The analytical cohort was constructed using entomological and serological data collected between 1999–2003 and 2008–2010 from two longitudinal cohort studies implemented in Iquitos , Peru . Iquitos , the largest city in the Peruvian Amazon , has a population of approximately 350 , 000 [21] . DENV1 is presumed to have been introduced in 1990–1991 [22] , followed by DENV2 in 1995 [23] , DENV3 in 2001 [24] , and DENV4 in 2008 [25] . Seasonal epidemic levels of DENV transmission occurred throughout this period [21 , 25] . From 1999–2003 , study activities were implemented in four city districts: Maynas , Punchana , Belen and San Juan . During the period 2008–2010 , data were collected from two neighborhoods: Maynas and Tupac Amaru ( located within the Maynas and Punchana districts ) . Procedures for entomological data collection were previously described [13 , 21] . Briefly , once households were enrolled , two-person study teams collected entomological data following a circuit to survey neighboring households on the same and/or neighboring block ( with block defined as a group of households that shared a common perimeter defined by city streets ) within an approximately two-week period . The entire study area required approximately four months to complete data collection , upon which entomological surveys resumed following the same schedule . Adult Ae . aegypti were collected using CDC backpack aspirators ( 1999–2009 ) [26] or Prokopack aspirators ( 2009–2010 ) [27] in both the exterior and interior of the participating household by passing the vacuum tube over common Ae . aegypti resting sites , outside walls , vegetation , and the entrance of potential larval habitats . Pupae and larvae were collected via enumeration of all wet containers or other larval development sites that contained water upon inspection . During surveys , all observed pupae and a sample of larvae were collected in small plastic Whirlpack bags; larval density was estimated as one of four levels ( 0 , 1–10 , 11–100 , >100 ) . All adult , larval and pupal samples were transported to and examined at the study laboratory , counted and identified to species and sex . Pupal data were recorded as observed counts . The total number of adult male and female Ae . aegypti mosquitoes collected in the interior and exterior of the dwelling were recorded . Household demographic data were collected for variables including enumeration of household residents by age and sex , household water source , sanitation facility , presence of electricity , type of building material , roof structure , and any reported use of insecticide or larvacide . The indicators were classified by scale ( household or block ) and life stage ( adult , pupal and/or larval ) . Household-level indicators were calculated using the observed survey data . To construct block-level indicators , all household survey data were first aggregated by block using a unique block identification number and circuit schedule . Indicators were then calculated using the aggregated block-level Ae . aegypti data . Block-level measures were then linked back to individual households by matching on block identifier and date of collection . The household-level indicators and their definitions are summarized in Table 1 and block-level indicators are summarized in Table 2 . Since the distribution of Ae . aegypti counts across all life stages is narrow in most settings , we dichotomized the continuous indicators to determine if categorical characterization of mosquito abundance would reflect a better fit to the data . To test categorical ( dichotomous ) versions of continuous indicators , a preliminary analysis was conducted to identify cut-off values by estimating the sensitivity and specificity of the mosquito density in terms of DENV infection at different levels ( data not presented ) . There is no consensus in the literature as to what categorical values of mosquito density measures correlate with DENV infection , therefore we used the following systematic approach to select categories and then test for an association . This approach was used to allow the distribution of the continuous indicator value to inform categorization without data mining for an association . To choose a categorical cutpoint , the sensitivity of the mosquito density indicator to identify a DENV seroconversion was calculated for increments of five ( e . g . , a Breteau Index of 0 , 5 , 10 , etc . ) , with the exception of the Potential Container Index , which was estimated for increments of two . The cutpoint was chosen as >0 if the sensitivity was less than 50% at that value . If cutpoints greater than zero had a sensitivity >50% , then the cutpoint ( not zero ) with the highest sensitivity was chosen for evaluation . Once a categorical variable was defined , it was then tested for an association with risk of DENV seroconversion . Categorical classification of continuous indicators tested is listed in Tables 1 and 2 . Data on eggs or exact larval counts were not collected in the parent study; therefore , indices relying on this information could not be tested . Cross-sectional entomological indicators were calculated using vector data from a single entomological survey observation . Longitudinal household-level indicators were calculated as an average of entomological data observed within the 12 months preceding the start of the seroconversion interval ( up to three survey visits collected approximately every four months ) . If a paired sample interval began before any entomological data collection , the cross-sectional measure of mosquito density was used . For block measures , indicators were calculated by averaging block-level densities calculated from surveys conducted within 12 months from the start of the seroconversion interval . In the parent study , members of households selected for entomological monitoring were asked to provide blood samples every six months [21 , 33] . Samples were collected at the participant’s home , stored in ice and transported to the study laboratory within four hours of collection . Sera were tested at two ( 1999–2003 ) and four ( 2008–2010 ) serum dilutions plaque reduction neutralization test ( PRNT ) [34] at the United States Naval Medical Research Unit No . 6 laboratory in Lima , Peru . To identify seroconversion to DENV , a serum sample was considered positive for DENV if a dilution neutralized 70% of the test virus ( PRNT70 ) [21 , 33] . The primary outcome of interest in this analysis was seroconversion to any circulating DENV serotypes as determined by PRNT70 . The longitudinal serological samples used in this analysis were previously reviewed to determine seroconversion [33] . In brief , to minimize misclassification of serological data , the full serological profile of subjects was reviewed as follows: if the increase in titer that reduced DENV plaques between a negative sample and a subsequent sample was at least 20% and all subsequent samples were positive , the subject was determined to have seroconverted . However , if subsequent PRNT results were not consistent with respect to seroconversion ( e . g . , negative-positive-negative ) , the subject was classified as not having seroconverted . For this study , serological results for all paired samples were classified as a binary outcome ( any seroconversion versus no seroconversion ) . Fig 1 illustrates the construction of the analysis cohort . Serological data were reviewed to identify paired sample observations taken approximately six months apart that could be linked to household entomological data . To account for operational constraints around serology collection , the at-risk interval was defined as 140 to 220 days . Each paired sample interval for which a subject was susceptible to any of the circulating DENV serotypes ( DENV1 and DENV2: all study years; DENV3: 2001–2010; DENV4 , 2008–2010 ) was included in the risk set . For household-level indicators , entomological data were matched by the date nearest the end of ( but within ) each paired serological sample interval . For block-level indicators , datasets were constructed by restricting to serological observations from blocks in which at least five households were surveyed , using the month and year of block data collection to anchor in time block-level densities to serology . Finally , longitudinal densities were calculated by averaging entomological data collected in the 12 months preceding the start of the seroconversion interval . Fig 2 illustrates how cross-sectional and longitudinal measures of vector abundance were calculated and linked to the 6-month seroconversion paired sample interval . The association between each Ae . aegypti indicator and the 6-month risk of DENV seroconversion was estimated using a log binomial generalized estimating equation ( GEE ) [35] , separately for each household-level and block-level indicator , and for both the cross-sectional and longitudinal scenarios . The log link with a binomial distribution allowed for estimation of risk ratio point estimates by exponentiating the beta coefficient for the indicator variable and calculation of 95% confidence intervals ( CI ) [36] . For models using household-level densities , the GEE accounted for clustering due to repeated individual measures and dependence due to household membership using an exchangeable correlation structure; models for block-level densities accounted for repeated observations from individuals and block level membership . A priori , we chose dengue transmission season , participant age and sex as confounding variables for use in all adjusted analyses of household-level indicators and season , participant age ( dichotomized at 18 years ) , and any reported use of larvacide by the head of household for adjustment of all block-level indicators . Dengue transmission season was defined as by the start of the seroconversion interval ( May-August ( reference group ) , September-December , January-April ) . All analyses were conducted in SAS/STAT software , version 9 . 4 of the SAS system for Windows ( SAS Institute , Cary , NC ) . Sensitivity analyses were conducted to account for possible bias resulting from construction of the dataset . The objective of the sensitivity analysis was to determine if the adjusted risk ratios were sensitive to decisions made to construct the analytical dataset . To implement these analyses , the same method as described in the main analysis was employed . First , different inclusion criteria for serological observations was used to test more restrictive or relaxed scenarios , as well as stratification by study years . Second , sensitivity analyses included alternate strategies for linking serology to entomology . Third , vector densities were calculated from entomological data 6 months prior to serology compared to 12 months prior to serology . Finally , the analysis was stratified by aspirator type used during data collection . Written informed consent ( and assent for children 8–17 years of age ) was obtained for all individuals providing serological data . Written informed consent was provided by parents or guardians for children under 18 years of age . Written consent ( 1999–2003 ) or oral consent ( 2008–2010 ) was obtained from an adult head of household for entomological surveys as approved by the institutional review boards . Oral consent for entomological surveys was documented upon obtaining access to the household and heads of households were provided information sheets describing the data collection procedures . Data collection procedures were approved by the University of California , Davis ( Protocols 2002–10788 and 2007–15244 ) , Instituto Nacional de Salud , and Naval Medical Research Center Institutional Review Boards ( Protocols NMRCD . 2001 . 0008 and NMRCD2007 . 0007 ) . This ancillary analysis was approved by the Institutional Review Board at the University of North Carolina at Chapel Hill ( Study # 14–3151 ) .
The adjusted RR point estimates and 95% CI are presented in Table 5 . The household-level point estimates ranged from 0 . 75 ( 95% CI: 0 . 48 , 1 . 34 ) to 1 . 05 ( 95% CI: 0 . 91 , 1 . 21 ) , suggesting no difference in the 6-month risk of DENV seroconversion based on Ae . aegypti density . At the block level , six indicators showed significant protective effects , which could be the result of higher background immunity , correlation with factors related to lower DENV risk , or chance . Compared to the adjusted RR estimates , crude risk ratio point estimates were similar for the household-level indicators and were slightly larger for block-level indicators ( S1 Table ) . Using the average of densities measured in the 12 months prior to the paired sample , the RR point estimate shifted above the null for categorical measures of adult density , adult female mosquitoes , and presence of adult mosquitoes indoors ( any adults as well as only females ) , ranging from 1 . 25 ( 95% CI: 1 . 12 , 1 . 39 ) to 1 . 30 ( 95% CI: 1 . 17 , 1 . 46 ) . This suggests that the observation of an adult female mosquito during a household survey performed during the 12 month period prior to collection of paired sera is associated with an approximately 25% increased risk in acquisition of DENV infection compared to the risk among individuals residing in households where no adult female was observed at any survey during the 12 months preceding the paired sera . In addition , four immature stage indicators suggested an elevated risk of DENV infection: any pupae observed; the Single Larval Method ( categorical ) ; Container Index ( categorical ) and the Stegomyia Index ( categorical ) . Analysis of block-level indicators that incorporated repeated measures demonstrated a similar trend in which all measures calculated based on adult mosquito data shifted in comparison to the cross-sectional analysis: the Adult Premise Index ( RR: 1 . 01; 95% CI: 1 . 01 , 1 . 02 when continuous and RR: 1 . 24; 95% CI: 1 . 01 , 1 . 48 as categorical ) and the Adult Density Index ( RR: 1 . 24; 95% CI: 1 . 02 , 1 . 50 as continuous and RR: 1 . 72; 95% CI: 1 . 22 , 2 . 43 as categorical ) . The Pupa Index ( categorical ) and the Infested Receptacle Index ( categorical ) were the only immature stage block-level indicators to demonstrate any association with DENV infection . Figs 3 and 4 compare the risk ratios calculated for cross-sectional to longitudinal densities for both household and block-level indicators . A number of sensitivity analyses were performed to determine if the construction of the analytical cohort introduced bias . Ae . aegypti densities calculated from 6 months prior to the start of a seroconversion interval followed a similar pattern as results presented in Table 5 ( S2 Table ) . Sensitivity analyses in which the inclusion of seroconversion events was relaxed and restricted did not alter interpretation of the main findings ( S1–S3 Figs , S3–S5 Tables ) . Future comparison of relaxed serological inclusion criteria were compared with the 6 and 12 month longitudinal entomological measures of Ae . aegypti ( S4–S6 Figs ) . When analyzed separately , use of different aspirators over the course of data collection did not result in substantially different results for adult stage measures ( S6 Table ) .
The major strengths of this analysis include the use of DENV infection ( not disease ) as an outcome , examination of longitudinal data , and its generalizability to similar settings in which routine , periodic entomological surveillance is conducted . While dengue disease is relevant from a public health perspective and easier to quantify , DENV infection , measured as seroconversion , is more important in terms of understanding patterns of transmission from mosquitoes to humans . Most prior studies of entomological indicators and dengue outcomes [7] used symptomatic disease as the outcome . Symptomatic cases represent the small fraction of all infections that were severe enough to seek medical evaluation , thus introducing selection bias . This analysis also benefitted from longitudinal serological data , which enabled exclusion of paired sample observations once an individual was determined to no longer be at risk of infection by circulating serotypes . Most prior studies used cross-sectional entomological data to test for an association with dengue outcomes . Longitudinal entomological monitoring allowed the use of multiple ( 1 to 3 per household ) mosquito measures per household . This may overcome some of the measurement error of entomological assessment and account for the temporal variability associated with entomological data collection , in which a household with lower levels of abundance could be misclassified as “unexposed” to Ae . aegypti . Our results comparing the RRs estimated from cross-sectional to longitudinal entomological measures of Ae . aegypti abundance suggest the possibility that in any single entomological survey , a household with low-levels of Ae . aegypti infestation may be misclassified as having no infestation , at least for adult stage measures of abundance , which would bias the RR downwards . The objective of this analysis was to evaluate the utility of periodic entomological monitoring as a proxy for DENV infection risk as it is typically implemented in the control setting . Under this monitoring framework , our findings are likely generalizable to similar dengue-endemic settings as the timing of serological and entomological collection employed are representative of the routine periodic monitoring used in dengue control programs . Our data have an added advantage given they were generated as part of a research study , and were subjected to rigorous monitoring of field collection procedures . Our results should be interpreted in light of several limitations . First , a large proportion ( 9 , 739 of 20 , 176 ) of serological data failed to meet the 6-month inclusion criteria , which could have resulted in bias due to their exclusion . Results from sensitivity analyses to include paired samples taken more than six-months apart did not qualitatively change our findings ( S1–S3 Figs ) . Second , the entomological and serological monitoring data relevant for DENV transmission did not perfectly coincide temporally , possibly leading to bias due to time of measurement . In sensitivity analyses , results were not sensitive to different approaches to link entomology and serology ( S1–S6 Figs ) . Our dataset contains more entomological monitoring data than would be available in most control settings . Even with a detailed longitudinal dataset of domestic vector density , we did not observe informative associations with DENV risk . Therefore , our study reveals the inherent limitations of using Aedes survey methods . The design of any entomological surveillance system should consider operational feasibility given the investment needed to generate sufficient data to describe temporal and spatial variability in vector density . While ovitraps were not used in the parent studies , we expect ovitrap-based sampling of Ae . aegypti to still be subject to the same limitations that apply to monitoring pupal , larval and adult mosquitoes . Alternatives that need to be evaluated include fixed trap methods , such as ovitraps or adult trapping methods that can monitor larger areas continuously overtime , better capturing temporal differences . Monitoring traps still presents operational challenges . Methods using fixed traps usually sample fewer houses than are possible with household-based survey methods [41] . Novel technologies that capture house-to-house temporal differences are yet to be developed . Second , even though the use of averages is not the most sophisticated method to incorporate temporal lags , it is implementable in basic statistical software and may be of utility to dengue program managers . Furthermore , in the control setting , it is highly unlikely that DENV infection outcomes would be well-resolved in time with entomological surveillance data as DENV infection cannot be monitored in real-time . Our study reports the relative risk of DENV seroconversion in a six-month period , an outcome similar in length to periods of increased dengue activity that occur seasonally in many endemic settings . While the temporal resolution between entomological and serological data collection in our study was not well resolved , our results provide quantitative evidence to challenge the use of periodic Ae . aegypti surveillance to generate suitable surrogates for DENV risk . Third , while PRNT70 is the most specific serological test for dengue infection , results from this assay may be biased due to cross-reactions from antibodies directed against multiple serotypes present in a single sample or with closely related viruses . The algorithm used to classify seroconversions was conservative , possibly underestimating the number of seroconversions , but this bias is likely non-differential with respect to mosquito density . We also acknowledge the possibility that block or neighborhood-level susceptibility to DENV may affect the performance of Ae . aegypti indicators , but it was not possible to address it in the analysis without full enumeration and serological testing of the entire study population , not just sampled individuals . Nevertheless , in the endemic setting , an assumption that herd immunity exists would only further undermine the utility of entomological monitoring endpoints to serve as correlates of infection . To ensure individuals in our analysis were susceptible to DENV infection , we reviewed longitudinal serological profiles to exclude those who were likely no longer at risk of the circulating serotype . We also tested a household-level variable estimating the proportion of susceptible individuals but this had no impact on the overall results ( data not presented ) . While our results may be generalizable to other areas with endemic transmission , this analysis should be repeated in a setting with a largely susceptible population to determine if household-based entomological surveillance is associated with DENV risk in such locations . Nevertheless , the majority of infections in our dataset were DENV3 and DENV4 , which were novel at the time of introduction in the community , so herd immunity may not have played a significant role for a large subset of our data . In this analysis , continuous indicators were tested as linear terms to maintain consistency with their definitions in the literature . It is possible that log-transformation or inclusion of polynomial terms could improve model fit , but such manipulation would reduce interpretability . For continuous indicators , the RRs measure the relative risk for a one-unit change in the indicator value; these measures are likely not informative for targeting interventions . From a public health perspective , categorical indicators are more useful to trigger vector control activities . In Iquitos , levels of infestation were heavily dispersed and binary classification ( any v . none ) was most informative . Finally , it is possible that vector control efforts could have reduced the vector population , making it difficult to detect an association between Ae . aegypti density and DENV risk . Over the period included in this analysis , there were large scale vector control interventions from October 2002-Jan 2003 and others in late 2003 [42] . Nevertheless , the majority of our data included periods where there was not extensive vector control . We did control for household larvacide use as a covariate in the analysis of block-level indicators to account for the possibility that some households implemented some form of vector control . Our results provide the first quantitative evidence of the limited utility of Ae . aegypti monitoring indicators as proxy measures of DENV infection . DENV transmission is complex and time-varying; the relationship between vector density and risk is not static nor adequately characterized through periodic entomological surveillance . None of the RRs presented in this analysis represent a causal relationship between household or block-level mosquito density and true exposure to DENV . It is logistically impossible to monitor human-vector contact to establish where and when mosquito-human interaction and infection occurs . Therefore , Ae . aegypti indicators serve as surrogates of true exposure , which will always remain unmeasured . Although adult measures that incorporated longitudinal data demonstrated an association with DENV seroconversion in our study , it is possible that some unmeasured variable associated with social network patterns , housing quality and day-time human movement further modifies dengue risk . Entomological monitoring indicators were not designed to account for the complexity of human-vector interaction , particularly given the role human movement may extend the boundaries of contact; it is likely that a substantial proportion of transmission occurs outside the home [18] . Technological advances in mosquito monitoring may eventually enable dengue control programs to quantify fluctuations in mosquito populations with greater precision across time and space . Nevertheless , DENV infection is difficulty to identify in real-time , especially given that most infections are inapparent in endemic settings . Without information on where and when individuals are infected , even detailed data of domestic vector density will require aggregation or categorization in order to attribute mosquito density to an interval-defined outcome ( such as the six-month seroconversion window , as in this analysis ) as DENV infection is measured at a coarse temporal interval . Globally , the incidence of dengue has continued to intensify and expand despite significant investments in vector control [1 , 43] . While vector control remains the only prevention strategy available to reduce DENV transmission in most settings , the persistence of DENV suggests transmission dynamics require a more complex understanding of human-vector interaction . Entomological monitoring will continue to serve a role in the evaluation of vector control interventions as it will be necessary to compare entomological measures of risk pre- and post-intervention as indicators of impact . Our analysis challenges the validity of most Ae . aegypti indicators as adequate proxies for true DENV exposure risk , and challenges the assumption that domestic vector data correlate with DENV transmission . In dengue-endemic settings such as Iquitos , single cross-sectional measures of adult mosquito density and the immature stage indicators commonly used by dengue control programs , such as the Breteau Index and Container Index , will likely fail to predict risk of DENV infection . Measuring adult mosquito density over multiple occasions may be the best option , but is difficult to implement . Our findings should be considered in the development and revision of enhanced DENV surveillance guidelines . Dengue control programs weighing the operational feasibility and cost of entomological monitoring against the limited utility of these indicators may wish to seek alternative monitoring frameworks that incorporate human dengue-related outcomes , such as passive case detection , and where possible , sero-surveys and active case detection . | In this study , we compared measures of entomological risk collected through routine household entomological monitoring by estimating an association with human DENV infection . Longitudinal entomological and human serology data from Iquitos , Peru , were used to test associations between Ae . aegypti indices and the 6-month risk of DENV seroconversion . Our analysis found no association between cross-sectional measures of Ae . aegypti abundance and the risk of DENV seroconversion . Longitudinal measures of Ae . aegypti were better proxies for DENV risk , primarily among adult stage mosquito indicators . DENV transmission is complex and time-varying; the relationship between vector density and risk is not static nor adequately characterized through periodic entomological surveillance . While entomological monitoring will continue to serve a role in the evaluation of vector control interventions ( e . g . , comparing pre- and post-intervention abundance ) , our analysis challenges the validity of most Ae . aegypti indicators as adequate proxies for true DENV exposure risk . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"animals",
"research",
"design",
"developmental",
"biology",
"pupae",
"surveys",
"infectious",
"disease",
"control",
"insect",
"vectors",
"zoology",
"research",
"and",
"analysis",
"methods",
"infectious",
"diseases",
"serology",
"aedes",
"aegypti",
"life",
"cycles",
"disease",
"vectors",
"insects",
"arthropoda",
"mosquitoes",
"survey",
"research",
"entomology",
"biology",
"and",
"life",
"sciences",
"species",
"interactions",
"larvae",
"organisms"
] | 2017 | The relationship between entomological indicators of Aedes aegypti abundance and dengue virus infection |
The analysis of the activity of neuronal cultures is considered to be a good proxy of the functional connectivity of in vivo neuronal tissues . Thus , the functional complex network inferred from activity patterns is a promising way to unravel the interplay between structure and functionality of neuronal systems . Here , we monitor the spontaneous self-sustained dynamics in neuronal cultures formed by interconnected aggregates of neurons ( clusters ) . Dynamics is characterized by the fast activation of groups of clusters in sequences termed bursts . The analysis of the time delays between clusters' activations within the bursts allows the reconstruction of the directed functional connectivity of the network . We propose a method to statistically infer this connectivity and analyze the resulting properties of the associated complex networks . Surprisingly enough , in contrast to what has been reported for many biological networks , the clustered neuronal cultures present assortative mixing connectivity values , meaning that there is a preference for clusters to link to other clusters that share similar functional connectivity , as well as a rich-club core , which shapes a ‘connectivity backbone’ in the network . These results point out that the grouping of neurons and the assortative connectivity between clusters are intrinsic survival mechanisms of the culture .
The theory of complex networks [1]–[5] has proven to be a useful framework for the study of the interplay between structure and functionality in social , technological , and biological systems . A complex network is no more than a specific representation of the interactions between the elements of the system in terms of nodes ( elements ) and links ( interactions ) in a graph . The analysis of such resulting abstraction of the system , the network , provides clues about regularities that can be connected with certain functionalities , or even be related to organization mechanisms that help to understand the rules behind the system's complexity . Particularly , in biological systems , the characterization of the emergent self-organization of their components is of utmost importance to comprehend the mechanisms of life [6]–[8] . One of the major challenges in biology and neuroscience is the ultimate understanding of the structure and function of neuronal systems , in particular the human brain , whose representation in terms of complex networks is especially appealing [9] , [10] . In this case , structural connectivity corresponds to the anatomical description of brain circuits whereas the functional connectivity is related to the statistical dependence between neuronal activity . Network theory and its mathematical framework have provided , through the analysis of the distribution of links , statistical measures that highlight key topological features of the network under study . These measures have facilitated the comprehension of processes as complex as brain development [11] , learning [12] and dysfunction [13] , [14] . Particularly , these measures have unfolded new relationships between brain dynamics and functionality . For instance , synchronization between neuronal assemblies in the developing hippocampus has been ascribed to the existence of super-connected nodes in a scale-free topology [15]; efficient information transfer has been associated to circuits with small-world features [16] , such as in the the nematode worm C . elegans [17] or the brain cortex [18] , [19]; and the coexistence of both segregated and integrated activity in the brain has been hypothesized to arise from a modular circuit architecture [20]–[22] . A network measure that has recently caught substantial attention is the assortativity coefficient , which quantifies the preference of a node to attach to another one with similar ( assortative mixing ) or dissimilar ( disassortative mixing ) number of connections [23] , [24] . Assortative networks have been observed in both structural [20] and functional [25] human brain networks . It has been proposed that assortative networks exhibit a modular organization [26] , display an efficient dynamics that is stable to noise [27] , and manifest resilience to node deletion ( either random or targeted ) [23] , [28] . Resilience is ascribed to the preferred interconnectivity of high-degree nodes , which shape a ‘connectivity backbone’ [29] that preserves network integrity . The existence of such a tightly interconnected community is generally known as the ‘rich-club’ phenomenon [19] , [22] , [30] . On the other hand , disassortative networks , such as the ones identified in the yeast's protein-protein interaction and the neuronal network of C . elegans [23] , are more vulnerable to targeted attacks . However , in these disassortative networks , the tendency of high degree nodes to connect with low degree ones results in a star-like topology that favors information processing across the network . The assortativty coefficient is usually calculated through the Pearson correlation coefficient between the unweighted degrees of each link in the network [23] . To account for effects associated to large networks , the Spearman assortativity measure was introduced [31] and , later , weighted assortativity measures were proposed to include the weight in degree-degree dependencies [32] . To better understand the importance of these network measures in describing neuronal networks , in vitro preparations in the form of neuronal cultures have been introduced given their accessibility and easy manipulation [6] , [33] . Two major types of cultured neuronal networks are of particular interest . In a first type , neurons are plated on a substrate that contains a layer of adhesive proteins . Neurons firmly adhere to the substrate , leading to cultures with a homogeneous distribution of neurons [34]–[37] . In a second type , neurons are plated without any facilitation for adhesion . Neurons then spontaneously group into small , compact assemblies termed clusters that connect to one another [38]–[41] . The formation of a clustered architecture from an initially isotropic configuration is an intriguing self-organization process [40] , [41] . This feature has made clustered networks attractive platforms to study the development of neuronal circuits as well as the interplay between structural and functional connectivity at intermediate , mesoscopic scales [39] , [42]–[45] . Moreover , the existence of a two-level network , one within a cluster and another between clusters , has made clustered cultures appealing to study dynamical and topological features of hierarchical [40] , [41] , [45] , [46] as well as modular networks [46]–[49] . In this work we use spontaneous activity measurements in clustered neuronal cultures to render the corresponding directed functional networks and study their topological properties . We introduce a novel theoretical framework that uses the propagation of activity between clusters as a measure of “causality” , although strictly speaking we should refer to as a sequence of delayed activations , giving rise to functional connections that are both directed and weighted . Based on this weighted nature of the network , we propose a new measure of assortativity that explicitly incorporates the weight of the links . We observed that all the studied functional networks derived from clustered cultures show a strong , positive assortative mixing that is maintained along different stages of development . On the contrary , homogeneous cultures tend to be weakly assortative , or neutral . Finally , in combination with experiments that measure the robustness of network activity to circuitry deterioration , we show that the strongly assortative , clustered networks are more resistant to damage compared to the weakly assortative , homogeneous ones . Our work provides a prominent example of the existence of assortativity in biological networks , and illustrates the utility of clustered neuronal cultures to investigate topological traits and the emergence of complex phenomena , such as self-organization and resilience , in living neuronal networks .
We used rat cortical neurons in all the experiments . As described in Methods , neurons were dissociated and seeded homogeneously on a glass substrate . Cultures were limited to circular areas in diameter for better control and full monitoring of network behavior . The lack of adhesive proteins in the substrate rapidly favored cell-to-cell attachment and aggregation , giving rise to clustered cultures that evolved quickly ( Figure 1A ) . By day in vitro ( DIV ) , cultures contained dozens of small aggregates , which coalesced and grew in size as the culture matured . Connections between clusters as well as initial traces of spontaneous activity were observed as early as DIV . Cultures comprised of interconnected clusters by DIV , and were sufficiently stable and rich in activity for measurements . Although the strength of the connections in the network and its dynamics evolved further , we observed that the size and position of the clusters remained stable . We therefore measured dynamics already at DIV , and studied cultures up to DIV . The example shown in Figure 1A corresponds to a culture at DIV . Clusters appear as circular objects with an average diameter of and a typical separation of . Connections between clusters are visible as straight filaments that contain several axons . We monitored spontaneous activity in the clustered network through fluorescence calcium imaging ( Figure 1B ) . Fluorescence images of the clustered network were acquired at a rate of frames per second , and with an image size and resolution that allowed the monitoring of all the clusters in the network with sufficient image quality ( see Movie S1 ) . Activity was recorded for typically hour , which provided sufficient statistics in firing events while minimizing culture degradation due to photo-damage . The analysis of the images at the end of the measurement provided the variations in fluorescence intensity for each cluster and the corresponding onset times of firing ( see Methods ) . As shown in the top panel of Figure 1C , the average fluorescence signal of the network is characterized by peaks of intense cluster activity combined with silent intervals . The accompanying raster plot reveals that this activity actually corresponds to the collective ignition of a small group of clusters , which fire sequentially in a short time window on the order of few hundred milliseconds . We denote by bursts these fast sequences of clusters' activations . Young cultures ( DIV ) exhibited an activity of about , while maturer cultures ( DIV ) displayed about ( see also Table 1 ) . We observed that the time spanned between two consecutively firing clusters typically ranged between and ( see Methods ) , as also observed by others [47] , [48] . These times are fairly large compared to the eventual scale of signal integration-propagation between single neurons ( ) , and is related to the large time scales associated to integration of the intra-clusters information . No consecutive activations were observed above , signaling the termination of a burst . We therefore use this value of as a cut-off to separate a given burst form the preceding one . Then , two clusters that fired above 200 ms cannot be influenced by one another and therefore are not causally connected . Bursts occurred every 30 s on average for the experiment shown in Figure 1C and , as illustrated by the yellow bands in this figure , each burst typically encompassed a subset of clusters rather than the entire network . In general , however , the number of participating clusters within a burst depended on the details of the culture . Although in a typical experiment the collective firing comprised between and clusters ( see Movie S1 ) , in some experiments the entire cluster population lighted up in a single bursting episode . The analysis of the onset times of firing provides the cluster's activation sequence within each burst . As an example , Figure 1D depicts a highly active region of the network shown in Figure 1B . This region contains clusters , and of them form a subset that regularly fires together . The series of frames show the progress in clusters' activation , revealed by the changes in fluorescence . Activity starts at the top-left cluster and progresses downwards . The time-line of sequence activation after image analysis is shown in Figure 1E , and the actual fluorescence traces are shown in Figure 1F . With our temporal resolution we could resolve well the propagation of activity from a cluster to its neighboring ones ( black arrows in Figure 1E ) . However , and for about of the cases , the time delay between clusters' activation was either too short for detection or activation occurred simultaneously . The clusters associated to these ‘simultaneous’ events are marked in yellow in Figure 1E , and their inter-relation was treated as a bi-directional link ( yellow arrow ) , since no causality can be inferred . A typical recording provided on the order of bursting episodes . Some of them included the same group of clusters , although the precise sequence of activation could vary . An illustrative example is shown in Figure 1F , which depicts the fluorescence traces of clusters along two consecutive bursts . The first sequence corresponds to the sketch of Figure 1E . The orange box at the bottom of the plot indicates the relative activation time of each cluster within the window , with two clusters treated as simultaneous . To introduce the construction of the directed functional network that is described later , we note that , intuitively , the firing of cluster #9 is most likely caused by #8 and therefore both clusters are ( functionally ) strongly coupled . At the other extreme , cluster #1 most likely did not trigger #9 , and therefore their mutual coupling is very weak . For the second burst , we note that the activation sequence is very similar , but the relative delay times differ , therefore modifying the cluster's coupling strengths . Indeed , cluster #1 and #9 are now functionally disconnected given their long temporal separation . We carried out measurements in different clustered networks , and labeled them with capital letters as networks ‘A’‘O’ . In order to compare their properties with the ones from cultures with a distinct structure , we applied the same measuring protocols and data analysis to cultures characterized with a homogeneous distribution of neurons ( see Methods and Figure S1 ) , and labeled them as networks ‘P’‘U’ . The above sequences of clusters' activations , extended to all the clusters and bursting episodes of the monitored culture , convey information on the degree of causal influence between any pair of clusters in the network . For instance , cluster #5 in Figures 1E-F can fire because of the first order influence of clusters #3 and #4 , but also because of the second and third order influences of clusters #2 and #1 , respectively . Hence , a realistic functional network construction should take into account these possible influences from the upstream connected clusters to build a network whose links are not only directed , but also weighted by the time delays in activation . This weighted treatment of the interaction between clusters is the major novelty of our work and the backbone of our model . More formally , the interaction between any two clusters follows the principle of causality , i . e . the firing of cluster immediately after cluster eventually implies that cluster has induced the activity of at that particular time . The likelihood of this relation between clusters is weighted according to its frequency along the full observational time , allowing to a statistical validation . Indeed , cluster could induce the activity of various clusters , if all of them activate in a physically plausible short time window after cluster . Such a construction is illustrated in Figure 2 . To construct the directed functional networks for each studied culture we proceed as follows . First of all , we divide the entire firing sequence into the bursts of clusters' activity ( Figure 2A ) using the cut-off of introduced in the previous section . Once the bursts have been detected , we compute the frequency distribution of time lags between pairs of consecutive firings ( Figure 2B ) . This frequency distribution informs about the characteristic times expected between two consecutive firings within the same burst , and hence it is a good proxy of the causal influence of a cluster on another . We will use this information to weight the causal influence of firing propagation . The frequency distribution presents a good fit to a universal Gaussian decay ( ) in all the analyzed cultures , although the variance is specific for each culture . We indeed observed ( Figure S2 ) that decreases with the culture age in vitro ( correlation coefficient , significance ) , and increases with the number of clusters present in the network ( , ) . The last step in the construction of the directed functional networks consists in linking the interactions within each burst , and weighting them according to the previous frequency distribution ( Figure 2B ) . The rationale behind this process is as follows: we hypothesize that every cluster influences other clusters ( posterior in time ) within a burst and , the larger the time after a cluster has fired the lower the influence we expect in the activation of another cluster ( simply because the signal fades out ) . Then , the weighting of the interaction by the expected frequency observed in the distribution conveys the functional influence between clusters . The weights are reinforced every time the same pair of clusters' sequence is observed . After processing the full sequence we obtain a peer-to-peer activation map that is our proxy of the functional network . We proceeded identically to construct the directed functional networks for homogeneous cultures ( see Methods ) , with the only difference that the cut-off time corresponds to . We tested for both clustered and homogeneous cultures that the obtained functional networks were stable upon variations of the cut-off times ( see Figure S3 and Discussion ) . We computed the functional networks of the ( ‘A’ to ‘O’ ) realizations of clustered cultures , as well as the ( ‘P’ to ‘U’ ) homogeneous ones , and analyzed some major topological traits . Firstly , for each culture we obtained the number of nodes , the number of edges , the average degree of the networks , and its average strength ( see Methods ) . The investigated networks and their topological measures are summarized in Table 1 . Although young cultures display a richer activity , in general all networks present a similar number of nodes and a comparable functional connectivity , which is described by the number of edges , the average degree and the average strength . Representative examples of the investigated functional networks for the clustered configuration are shown in Figure 3 ( see Figure S1 for an example of the homogeneous ones ) . The position of the nodes and their size are the same as the actual clusters for easier comparison . Edges in the directed network are both color and thickness coded to highlight their importance , with darker colors corresponding to the highest weights . This representation reveals those pairs of clusters that maintain a persistent causality relationship over time . Nodes are also color coded according to their strength , i . e . the total weight of the in- and out-edges . The functional networks exhibit some interesting features . First , there are groups of nodes that form tightly connected communities . These communities actually reflect the most frequent bursting sequences . Second , nodes preferentially connect to neighboring ones with some long-range connectivity , and often following paths that are not the major physical connections . This indicates that the structural connectivity of the network cannot be assessed from just an examination of the most perceivable processes . And third , as shown in Figure S4A , we observed that there is no correlation between the width of the physical connections and their weight ( , ) , or the size of the nodes and their strength ( , , Figure S4B ) , and indicates that the dynamical traits of the network cannot be inferred from its physical configuration , stressing the importance of the functional study . We also observed that the size of the clusters did not correlate with their average activity ( , , Figure S4C ) , i . e . small and big clusters displayed similar firing frequencies , and of on average . However , since some clusters are initiators of activity and others just followers , we also computed the relative contribution of a given cluster size to initiate activity in the network . We found no significant correlation between initiation and cluster size ( , , Figure S4D ) . These results strengthen the conclusion that one cannot predict the clusters that will initiate activity , or the most persistent sequences , by just a visual inspection of cluster sizes and their distribution over the network . These analyses are important in the context of the work by Shein-Idelson and coworkers [41] , who studied the dynamics of isolated clusters similar to ours , and observed that their firing rate increased from to as the clusters' radii escalated from to . This remarkable difference in the dynamics between ‘isolated’ and ‘networked’ clusters reflects the dominant role of the network circuitry in shaping its dynamics . Finally , to crosscheck that the results found for the functional networks presented here are robust to the inference method , we have also performed a classical mutual information analysis to construct functional networks for the same cultures ( see Methods ) . The results obtained with the mutual information analysis are totally in agreement with the constructed functional networks using time delays . We determined the values of the weighted formulation of assortativity , both for the Pearson and Spearman correlations , with values in ( see Methods for the generalization of assortativity to directed weighted networks ) . Positive values of the weighted assortativity indicate that nodes with similar strength tend to connect to one another , while negative values mean the preferred interconnectivity of nodes with different strength . In Table 1 we can observe that all clustered networks ( labeled ‘A’-‘O’ ) exhibit a positive weighted assortativity , in the range for the Pearson construction and for the Spearman one . Although the values fluctuate across different cultures , the two assortativity measures provide the same value within statistical error , and reflect that network size corrections provided by the Spearman's treatment have little influence in strongly assortative networks . To assess the importance of the measured assortativity values , we have also computed the weighted rich-club [50] . The rich-club phenomenon refers to the tendency of nodes with high degree to form tightly interconnected communities , compared to the connections that these nodes would have in a null model that preserves the node's degree but otherwise is totally random . Given the positive assortativity found , we analyzed whether this finding is also reinforced by the existence of rich-club structures . The weighted formulation for the rich-club takes into account the node's strength instead of the degree , and is particularly useful in situations in which the weights of the links can not be overlooked [51] . The evaluation of the rich-club is performed by computing the ratio between the connectivity strength of highly connected nodes and its randomized counterpart , and for gradually higher values of the strength threshold . The detailed calculation is described in the Methods section , and the results of the analysis for representative networks is shown in Figure S5 . Ratios larger than indicate that higher strength nodes are more interconnected to each other than what one would expect in a random configuration . On the contrary , a ratio less than reveals an opposite organizing principle that leads to a lack of interconnectivity among high-degree nodes . After the calculation of the ratios for all the studied clustered networks , we found a positive tendency towards the creation of rich-clubs in all of them ( Figure S5 ) , which is in good agreement with the observed values of assortativity . The above network measures were also analyzed in experiments with a homogeneous distribution of neurons ( labeled ‘P’‘U’ ) . The results are summarized in Table 1 . Interestingly , the assortativity values are much lower ( by an order of magnitude on average ) than the ones for clustered cultures , in the range for Pearson's and for Spearman's . Accordingly , the rich-club phenomenon for the homogeneous cultures vanishes ( Figure S5 ) . Several studies highlight the importance of assortative features for network resilience to damage . Given the strong assortativity of our clustered cultures , we carried out a new set of experiments to investigate the concurrent presence of resilient traits . As described in Methods , we considered two major ‘damaging’ actions to the network . In a first one , we gradually weakened the excitatory network connectivity by means of the AMPA-glutamate antagonist CNQX , and measured the decay in spontaneous activity as connectivity failed . In a second one , we continuously exposed a culture to strong fluorescence light , therefore inducing photo-damage to the neurons . This action resulted in random neuronal death across the network and hence a progressive failure of its spontaneous dynamics . The rate of activity decay upon radiation damage provided an estimation of the resistance of the network to node deletion . These investigations were carried out at the same time in clustered cultures ( strongly assortative ) and in homogeneous ones ( weakly assortative or neutral ) . Their comparison provided a first reference to relate assortativity , network topology and resistance to damage . Figure 4A shows the results for the application of CNQX to clustered cultures . We first monitored each cluster individually in the unperturbed case , and measured its average firing activity along . We then applied a given drug concentration , measured the firing activity for another , and computed the relative changes in activity respect to the unperturbed case , as . The protocol was repeated until activity ceased . Two illustrative examples of the action of CNQX on network activity are provided in Figure 4 . In a clustered cultured and for weak CNQX applications ( ) the activity in some clusters increases , while in some other decreases , and on average the network firing rate remains stable ( ) . As [CNQX] is increased to , we observe that most of the clusters have reduced their activity , although there are still some that maintain a high activity or even increase it . This different behavior from cluster to cluster suggests that clustered networks are highly flexible , and that they may have mechanisms to preserve activity even with strong weakening of the connectivity . Conversely , homogeneous cultures ( Figure 4B ) lose activity in a more regular and faster way . These networks are characterized by a highly coherent dynamics [36] , [37] , and therefore all neurons in the network reduce activity similarly as CNQX is applied . Interestingly , for the shown homogeneous culture has almost completely silenced ( ) , while the clustered culture is still highly active . We repeated this study on different realizations of each culture type and observed that , on average , the critical concentration at which activity complete stopped was for clustered and for homogeneous networks ( Figure 4C ) . Figure 4D shows the results for the resistance of the networks to node deletion as a consequence of direct photo-damage to the neurons . As can be observed , homogeneous cultures decay in activity much faster than the clustered ones , pinpointing the general resistance of clustered cultures to structural failure .
Clustered neuronal cultures have a unique self-organizing potential . An initially isotropic ensemble of individual neurons quickly group to one another to constitute a stable configuration of interconnected clusters of tightly packed neurons . The formation of the clustered network is primarily a passive process governed by the pulling forces exerted by the neurites . Interestingly , aggregation occurs even in the absence of glial cells and neuronal activity [40] , and is maintained up to the degradation of the culture [40] , [52] , [53] . Our work shows that this self-organizing process drives the network towards specific dynamic states , which shape a topology of the functional network that is distinctively assortative . We note that the number of clusters and their distribution are initially random . Therefore , a wide spectrum of physical circuitries and functional topologies are in principle attainable . However , in all the studied cultures , the network drives itself towards markedly assortative topologies with a ‘rich-club’ core . The emergence of these distinct topological traits , concurrently with a stronger network's resilience to activity deterioration , pictures a self-organizing mechanism that enhances network survival by procuring a robust architecture and dynamic stability . We remark that the link between assortativity and resilience is based on the comparison between the response of clustered and homogeneous cultures upon the same perturbation . To obtain conclusive evidences that assortativity confers resilience traits exclusively from topology , we would require an experimental protocol in which we could arbitrarily ‘rewire’ the connectivity between clusters , or shape in a control manner different circuitries while preserving the number of nodes in the network . Although these strategies are certainly enlightening , they are of difficult development and a major experimental challenge . We infer the functional connectivity maps of the clustered networks from their spontaneous dynamics . We considered small-sized networks to simultaneously access the entire population ( clusters ) . The approach that we have used to characterize this functional connectivity is based on the analysis of the time delays between consecutive clusters' activations . The uniqueness of our approach is to use these time delays to provide a direct measure of causality , giving rise to a functional network that is both directed and weighted , with the weights given by a decaying function that follows the frequency of the delay between pairs of clusters . Our formulation is simple and naturally derives from the intrinsic dynamics of the network . We used two main parameters to quantitatively construct the directed functional network , namely the cut-off time for causality , and the variance of the Gaussian-like weighting function . The cut-off time is set to , two times the maximum measured time delay between consecutive activations . The importance of the cut-off is first to discriminate two successive bursting episodes , and second to exclude individual firing events from an actual cascade of activations . Although these individual firings account for less than 2% of the total activations , they may occur in regions of the culture that are physically distant -though temporary close- from an actual sequence , and therefore they would add spurious , long-range functional connections to the network . On the other hand , the variance is obtained from a Gaussian fit of the distribution of consecutive activation delays within bursts . The value of is specific for each culture to take into account particular differences in the dynamics of the network , specifically the culture days in vitro or the number of clusters ( Figure S2 ) , parameters that could affect the delay times of activation . Young cultures for instance exhibit longer time delays between pairs of clusters , leading to a distribution shifted towards higher values and therefore a larger . We tested that the obtained functional networks were stable upon variation of the above parameters . In particular , to examine whether the choice of the cut-off does or does not substantially affect the features of the generated functional network , we performed a sensitivity analysis on this parameter . As the process of generating the network from the sets of bursts is deterministic , we analyzed the influence of the cut-off value on the formed groups of firings . To quantify the variation on the bursts generated for different values of the cut-off , we calculated the variation of information [54] between the grouping of bursts at a certain cut-off value and the previous one as a measure to assess their difference ( Figure S3 ) . In the case of clustered cultures , we found that for values of cut-off of the variation of information is , on average , on the order of . In the homogeneous case , for cut-off values of ms , this value is on the order of . This means that varying the cut-off values in these regions does not substantially change the grouping of the bursts , and therefore the generated networks are equivalent . To assess the goodness of our construction in inferring the functional connectivity of the clustered networks , we compared our connectivity maps with those procured by information theoretic measures , such as Mutual Information or Transfer Entropy , applied to the original fluorescence recordings . These approaches have been used to draw the topological properties of neuronal networks in vitro , both in electrode recordings [55] , [56] and calcium fluorescence imaging [57] , [58] . The comparison of our method with these theoretic measures showed that the identified functional links were fundamentally the same , with small quantitative differences associated to the particular weighting procedures . Our functional networks consistently maintained high assortativity values , and along a wide range of days in vitro . We also observed that , by contrast , the assortativity analysis in homogeneous cultures procured neutral or low assortativity values , a result that is supported by other studies in homogeneous networks similar to ours [55] . In our study , we have seen that the clustered , assortative networks exhibit a higher resilience of the network to damage compared to the homogeneous , non-assortative ones . Different studies also highlighted the importance of assortativity and the ‘rich-club’ phenomenon on higher-order structures of the network , in particular resilience , hierarchical ordering and specialization [10] , [30] . Several studies in brain networks advocate that the functional connectivity reflects the underlying structural organization [59]–[61] . To shed light on this interrelation in our cultures , we would need the identification of all the physical links between clusters . The top images of Figure 3 indeed suggest that some structural connections could be delineated by a simple visual inspection . However , we observed by green fluorescence protein ( GFP ) transfection that physical connections have long extensions and may easily link several clusters together , and not just in a first-neighbor manner as seen in the images . Since the images provide a very poor subset of the entire structural layout , a complete description of the physical circuitry must be carried out before comparing the structural and functional networks . Such a detailed identification is difficult , and requires the use of a number of connectivity-labeling techniques . Nevertheless , for the connections that we could visualize , we draw two major conclusions . First , that neither the width of the physical connections nor the size of the clusters were related to a particular trait of the functional links , such as the weight of the connections or the strength of the nodes ( Figure S4 ) . And , second , that our construction inferred strong functional links between clusters that were not directly connected in a physical manner , highlighting the importance of indirect paths as well as long-range coupling in the flow of activity . The identification of the full set of structural connections would certainly provide invaluable information to investigate the interplay between structure and function in our networks . In this context , the recent work by Santos-Sierra et al [52] is enlightening . They analyzed some major structural connectivity traits in clustered networks similar to ours , and observed that the networks were strongly assortative as well . Assortativity emerged at early stages of development , and was maintained throughout the life of the culture . Hence , in clustered cultures , the combined evidences of this study and ours hints at the existence of assortative properties in both structure and function . An interesting peculiarity of our experiments is that , in most of the studied clustered cultures , the spontaneous bursting episodes comprised of a small subset of clusters rather than the entire network . This activation in the form of groups or moduli is often referred as conditional activity . It contrasts with the coherent activity of homogeneous cultures , where the entire network lights up in a short time window during a bursting episode . Given the acute differences in assortativity between clustered and homogeneous cultures , we hypothesize that the modular dynamics by itself increases or reinforces assortative traits in the functional network . We finally remark that our neuronal cultures are spatial , i . e . embedded in a physical substrate , which imposes constraints to the layout of connections and , in turn , their assortative characteristics [52] , [62] . Spatial networks have caught substantial interest in the last years to understand the restrictions—or advantages—that metric correlations impose on the structure and dynamics of complex networks [63] , in particular brain circuits [64] . Vértes et al showed that spatial constraints delineate several topological properties of functional brain networks [65] , and Orlandi et al showed that the initiation mechanisms of spontaneous activity are governed by metric correlations inherited by the network during its formation [36] . Strong spatial constraints in clustered networks can be attained by anchoring the neuronal aggregates in specific locations , for instance through carbon nanotubes [39] . The comparison of the functional maps of such a forced organization with our free one is enlightening , and would shed light on the importance of structural constraints in shaping functional connectivity . To conclude , we have presented a simple yet powerful construction to draw the directed functional connectivity in clustered neuronal cultures . The construed networks present assortativity and ‘rich-club’ features , which are present concurrently with resilience traits . Our analysis has been based on spontaneous activity data , and may certainly vary from evoked activity . Hence , the combined experimental setup and functional construction can be viewed as a model system for complex networks studies , specially to understand the interplay between structure and function , and the emergence of key topological traits from network dynamics . Also , the spatial nature of our experiments may also procure invaluable data to understanding the role of short- and long-range connections in network dynamics; or to investigate the targeted deletion of the high degree nodes that shape the backbone of the network . The latter is a powerful concept that may assist in a detailed exploration of resilience in neuronal circuits , for instance to model the circuitry-activity interrelation in neurological pathologies .
All procedures were approved by the Ethical Committee for Animal Experimentation of the University of Barcelona , under order DMAH-5461 . In our experiments we used cortical neurons from day old Sprague-Dawley rat embryos . Following standard procedures [36] , [66] dissection was carried out in ice-cold L- medium enriched with glucose and gentamycin ( Sigma-Aldrich ) . Cortices were gently extracted and dissociated by repeated pipetting . Cortical neurons were plated onto glass coverslips ( Marienfeld-Superior ) that incorporated a poly-dimethylsiloxane ( PDMS ) mold . The PDMS restricted neuronal growth to isolated , circular cavities in diameter . Prior plating , glasses were washed in nitric acid for 2 h , rinsed with double-distilled water ( DDW ) , sonicated in ethanol and flamed . In parallel to glass cleaning , and following the procedure described by Orlandi et al . [36] , several diameter layers of PDMS thick were prepared and subsequently pierced with diameter biopsy punchers ( Integra-Miltex ) . Each pierced PDMS mold typically contained to cavities . The PDMS molds were then attached to the glasses and the combined structure autoclaved at , firmly adhering to one another . For each dissection we prepared identical glass-PDMS structures , giving rise to about cultures of in diameter . Neurons were plated in the PDMS cavities with a nominal density of , and incubated in plating medium at 37°C , CO2 and humidity . Plating medium consisted in of foetal calf serum ( FCS , Invitrogen ) , of horse serum ( HS , Inivtrogen ) , and 0 . 1% B27 ( Sigma ) in MEM Eagle's-L-glutamate ( Invitrogen ) . MEM was enriched with gentamicin ( Sigma ) , the neuronal activity promoter Glutamax ( Sigma ) and glucose . Upon plating , the absence of adhesive proteins in the glass substrate rapidly favored cell-cell attachment and , gradually , the formation of islands of highly compact neuronal assemblies or clusters that minimized the surface contact with the substrate . Clustered cultures formed quickly . By day in vitro ( DIV ) 2 the culture encompasses dozens of small aggregates that coalesce and grow in size as the culture matures . Spontaneous activity and connections between clusters were observed by DIV . Clusters at this stage of development also anchored at the surface of the glass and , although they continued growing and developing connections , their number and position remained practically stable along the next weeks . At the moment of measuring , each PDMS cavity contained an independent culture formed by interconnected clusters . Clustered cultures were maintained for about weeks , as follows . At DIV the medium was switched from plating to changing medium ( containing FUDR , Uridine , and HS in enriched MEM ) to limit glial cell division . Three days later , the medium was replaced to final medium ( enriched MEM with HS ) , which was then refreshed periodically every three days . Overnight exposure of the glass coverslips to poly-l-lisine ( PLL , Sigma ) provided a layer of adhesive proteins for the neurons to quickly anchor upon seeding , leading to cultures with a homogeneous distribution of neurons over the substrate . The remaining steps in the preparation and maintenance of the cultures were identical as the clustered ones , i . e . we used the same nominal neuronal density for plating , we included PDMS pierced molds to confine neuronal growth in cavities in diameter , and we refreshed the culture mediums in the same manner . The pharmacological protocols described below were used identically in clustered and homogeneous cultures . The acquired images ( recorded at a typical speed of fps ) were first analyzed with the Hokawo 2 . 5 software to extract the fluorescence intensity of each cluster as a function of time . The regions of interest ( ROIs ) were chosen manually and typically covered an area of pixels , each ROI corresponding to a single cluster . As illustrated in Figure 1C and Figure 1F , activity is characterized by a stable baseline ( resting state ) interrupted by peaks of fluorescence that correspond to clusters' firings . At the onset of firing , the fluorescence signal increases abruptly due to the fast intake of ions . Fluorescence then reaches a maximum , and slowly decays back to the baseline in s . The algorithm that we used to detect the onset of firing for each cluster was as follows . We first corrected the fluorescence signal from small drifts , and calculated the resting fluorescence level by discarding the data points with an amplitude two times above the standard deviation ( SD ) of the signal . The corrected signal was then expressed as . We next took and computed its derivative in order to detect fast changes in the fluorescence signal . Finally , the onset of ignition in cluster was defined as the time where a maximum in was accompanied by values of two times above the SD of the background signal , and for at least 5 frames . Recordings in homogeneous cultures provided the activity of neurons in an circular area in diameter . Neurons were marked individually as regions of interest in the images and the corresponding fluorescence time traces extracted using custom-made software . Ignition times for each neuron were next obtained by using the sub-frame resolution method described above ( detailed in Ref . [36] ) , and that consisted in fitting two straight lines to the fluorescence data , a first fit encompassing the points in the background region prior to firing , and a second fit including the points during the fast rise in fluorescence that follows ignition . The crossing point of the two lines provided the onset of firing . The extension of this analysis to all the active neurons within a burst , and along all the bursts , finally provided the entire set of ignition sequences . The construction of the directed functional networks for the homogeneous cultures was then carried out identically as the clustered ones . Recordings in clustered cultures typically lasted for h and contained between bursts in the quietest networks and bursts in the most active ones . To test whether bursts sufficed to draw the functional networks , we carried out a control experiment in which we monitored spontaneous activity along h in a standard clustered culture , measured at DIV and containing nodes ( Figure S6 ) . We then analyzed the data using two different procedures . In the first one we drew the functional connectivity using the data extracted from the entire recording , and determined its assortativity values . In the second procedure , we separated the recorded sequence in three blocks , each long , built the functional connectivity for each block , and computed the respective assortative values . The studied culture fired in a sustained manner at a rate of , and procured a total of bursts . Thus , each block typically contained about bursts . The results ( Figure S6 ) led to two major conclusions . First , that the functional connectivity is very similar among the blocks , and between any of the blocks and the entire recording , providing assortativity values that are compatible within statistical error . And second , that the first 40 min of recording ( with 45 bursts only ) sufficed to shape the major traits of the functional network , therefore validating our strategy of using h of acquisition to procure a reliable estimate of the functional connectivity of the network and its assortative traits . Here we describe the calculation of the assortativity coefficients , assortativity errors and the rich-club distributions . In the process , we have to define the assortativity for directed weighted networks . Mutual information [56] , [72] is a particular case of the Kullback-Leibler divergence [73] , an information-theoretic measure of the distance between two probability distributions . In fact , the mutual information between two stochastic variables and provides an estimation of the amount of information gained about when is known . Let us indicate by the time series corresponding to the -th cluster , with and the total number of time frames involved in the observation process . The time series adopted for the successive analysis are obtained by mapping the observed train of cluster activations to another time series termed walk , defined by ( 17 ) In the specific case of our analysis , the mutual information between two time series and , corresponding to two different clusters , is interpreted as the amount of correlation between the dynamics of cluster and . In general , the time scale of the correlation between two time series is not known a priori . Such a time scale corresponds to the time delay required to maximize the gain of information . Therefore , in the spirit of Fraser and Swinney [74] , we define the time delayed mutual cross information between and by ( 18 ) where and are indices running over some partition of the observed time series . In Eq . ( 18 ) , indicates the probability to find a value of time series in the -th interval , is the probability to find a value of time series in the -th interval , whereas denotes the joint probability to observe a firing from the -th cluster falling in the -th interval and a firing from the -th cluster falling in the -th interval exactly time frames later . For the sake of simplicity , in the following we will adopt the more concise notation to indicate the time delayed mutual cross information . Finally , in order to gain the highest amount of information about the dynamics of cluster by observing cluster , we consider only the maximum value of with respect to the time delay . We estimate the importance of the observed amount of correlation by performing the above analysis on surrogate data . Surrogates adopted in this study are time series generated by randomly reshuffling the temporal observations of the firing series , for each cluster separately . Such a procedure destroys any correlation between pairs of time series while preserving the empirical probability distribution , thus allowing to test the null hypothesis that the observed correlation is obtained by chance . We indicate by the walk corresponding to the surrogate obtained from time series and with the time delayed mutual cross information between and . We perform 200 independent random realizations of surrogates for each pair and we estimate the corresponding expected value of the maximum mutual cross-information , as well as the root mean square of the underlying distribution . Hence , we fix a priori the significance of the hypothesis testing and we estimate the -score corresponding to each pair by . Therefore , the observed correlation between cluster and is said to be statistically significant if , where is the standard error function . Finally , we obtain the functional network of clusters by building the weight matrix whose elements are defined by if , and if . | The architecture of neuronal cultures is the result of an intricate self-organization process that balances structural and dynamical demands . We observe that when the motility of neurons is allowed , these neurons organize into compact clusters . These neuronal assemblies have an intrinsic synchronous activity that makes the whole cluster firing at unison . Clusters connect to their neighbors to form a network with rich spontaneous dynamics . This dynamics ultimately shapes a directed functional network whose properties are investigated using network descriptors . We find that the networks are formed such that preference in connectivity between clusters is based on the similarity between their activity , a property that is called assortative mixing in networks' language . This particular choice of connectivity correlations must be rooted to basic survival mechanisms for the neurons constituting the culture . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"complex",
"systems",
"physics",
"computer",
"and",
"information",
"sciences",
"systems",
"science",
"mathematics",
"neural",
"networks",
"applied",
"mathematics",
"biophysics",
"theory",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"computational",
"biology",
"biophysics",
"neuroscience",
"biophysical",
"simulations"
] | 2014 | Emergence of Assortative Mixing between Clusters of Cultured Neurons |
Elimination of schistosomiasis as a public health problem and interruption of transmission in selected areas are key goals of the World Health Organization for 2025 . Conventional parasitological methods are insensitive for the detection of light-intensity infections . Techniques with high sensitivity and specificity are required for an accurate diagnosis in low-transmission settings and verification of elimination . We determined the accuracy of a urine-based up-converting phosphor-lateral flow circulating anodic antigen ( UCP-LF CAA ) assay for Schistosoma haematobium diagnosis in low-prevalence settings in Zanzibar , Tanzania . A total of 1 , 740 urine samples were collected in 2013 from children on Pemba Island , from schools where the S . haematobium prevalence was <2% , 2–5% , and 5–10% , based on a single urine filtration . On the day of collection , all samples were tested for microhematuria with reagent strips and for the presence of S . haematobium eggs with microscopy . Eight months later , 1 . 5 ml of urine from each of 1 , 200 samples stored at -20°C were analyzed by UCP-LF CAA assay , while urine filtration slides were subjected to quality control ( QCUF ) . In the absence of a true ‘gold’ standard , the diagnostic performance was calculated using latent class analyses ( LCA ) . The ‘empirical’ S . haematobium prevalence revealed by UCP-LF CAA , QCUF , and reagent strips was 14% , 5% , and 4% , respectively . LCA revealed a sensitivity of the UCP-LF CAA , QCUF , and reagent strips of 97% ( 95% confidence interval ( CI ) : 91–100% ) , 86% ( 95% CI: 72–99% ) , and 67% ( 95% CI: 52–81% ) , respectively . Test specificities were consistently above 90% . The UCP-LF CAA assay shows high sensitivity for the diagnosis of S . haematobium in low-endemicity settings . Empirically , it detects a considerably higher number of infections than microscopy . Hence , the UCP-LF CAA employed in combination with QCUF , is a promising tool for monitoring and surveillance of urogenital schistosomiasis in low-transmission settings targeted for elimination .
After many years of neglect , schistosomiasis and other parasitic worm infections are given considerable attention by the research community , non-governmental organizations , funding bodies , international organizations , policy makers , and disease control managers [1 , 2] . Indeed , drug donations , in conjunction with political and financial commitment for scaling up control interventions , have markedly expanded since the London Declaration on Neglected Tropical Diseases was launched in early 2012 [3 , 4] . For example , the World Health Organization ( WHO ) strategic plan 2012–2020 for schistosomiasis aims at morbidity control by 2020 , and elimination of schistosomiasis as a public health problem and interruption of transmission in selected areas by 2025 [5] . The mainstay of morbidity control is preventive chemotherapy that is the large-scale administration of praziquantel to at-risk groups ( e . g . , school-aged children ) . To achieve elimination of schistosomiasis as a public health problem ( i . e . , the reduction of heavy infection intensities in the at-risk population to below 1% ) , additional public health measures are recommended , alongside intensified interventions in pockets of high transmission [5 , 6] . For interruption of transmission ( i . e . , reducing the incidence of infection to zero ) , it is essential to increase the frequency of preventive chemotherapy and to supplement it with additional control measures , such as improved access to clean water and adequate sanitation , and controlling intermediate host snails [5–9] . Monitoring the progress of control programs and rigorous surveillance to identify remaining or reemerging transmission hotspots and individuals with high infection levels ( so-called superspreaders ) , will be relevant for tailoring an adequate response in close-to-elimination settings [5 , 6 , 10] . An accurate diagnosis is essential and should be adapted to the specific stage of a schistosomiasis control program [11 , 12] . Parasitological methods to detect Schistosoma eggs in stool ( e . g . , Kato-Katz thick smear [13] ) or urine ( e . g . , urine filtration [14] ) , or reagent strips to detect microhematuria in urine [15] are widely used in control programs . However , while these methods are reasonably accurate in diagnosing moderate and heavy infection intensities , they show low sensitivity for detecting light-intensity infections [16 , 17] . For appropriate monitoring and surveillance in areas approaching schistosomiasis elimination , other , highly sensitive tools are needed [4 , 11 , 18–22] . Moreover , as the prevalence of infection decreases over the course of a control program , specificity becomes more important and will be an absolute requirement for certification of elimination [11 , 19] . A promising diagnostic approach that might be suitable for highly sensitive and specific diagnosis of very light Schistosoma infections is the detection of Schistosoma adult worm circulating anodic antigen ( CAA ) in serum and urine using an up-converting phosphor-lateral flow ( UCP-LF ) assay [20 , 23–25] . Concentration systems allowing the recognition of very low worm numbers have been developed , and currently four urine-based assays are available in a robust dry-reagent format [26 , 27] . These UCP-LF tests can detect 30 pg/ml , 3 pg/ml , 0 . 3 pg/ml , and 0 . 1 pg/ml CAA using 10 μl ( UCAA10 ) , 250 μl ( UCAA250 ) , 2 ml ( UCAA2000 ) , and 7 . 5 ml urine ( UCAA7500 ) , respectively [27] . The UCAA2000 ( and perhaps the UCAA250 ) is currently considered to show the best trade-off between a high sensitivity and convenient field applicability [28] . However , the performance of different UCP-LF CAA assays as sensitive and specific diagnostic tools for non-invasive monitoring and surveillance remain to be determined in laboratories in Schistosoma low-endemic settings . Recently , the UCAA2000 , as well as a variant of the test conducted with 500 μl serum , have been successfully applied for the diagnosis of S . japonicum infections in low transmission settings in the People’s Republic of China [29] . First validation attempts have also been made with the UCAA250 for S . japonicum and S . mekongi detection in banked urine samples from the Philippines and Cambodia , respectively [30] . Here , we assess the accuracy of the UCAA2000 for the diagnosis of S . haematobium in three low-prevalence settings ( <2% , 2–5% , and 5–10% ) , as determined with a single urine filtration . In the absence of a true ‘gold’ standard , sensitivity and specificity were determined empirically and by means of latent class analysis ( LCA ) . Urine samples were collected from children attending primary schools on Pemba Island , United Republic of Tanzania , where prevalences ranged between 1% and 10% according to a single urine filtration in a recent parasitological survey conducted in the frame of the Zanzibar Elimination of Schistosomiasis Transmission ( ZEST ) project in early 2013 . The Zanzibar archipelago is deemed a setting where elimination of urogenital schistosomiasis is feasible and where a series of control measures , including biannual mass drug administration ( MDA ) , snail control , and behavioral change interventions , are currently being implemented to achieve this goal and to learn about which intervention works best [6 , 31–33] .
The study was approved by the Zanzibar Medical Research Ethics Committee ( ZAMREC ) in Zanzibar , United Republic of Tanzania ( reference no . ZAMREC 0003/Sept/011 ) , the ethics committee of Basel , Switzerland ( reference no . EKBB 236/11 ) , and the Institutional Review Board of the University of Georgia in the United States of America ( project no . 2012-10138-0 ) [31] . The study is registered with the International Standard Randomized Controlled Trial Number register ( identifier: ISRCTN48837681 ) . The purpose of collecting urine samples and potentially storing them for examination with newly developed and more sensitive diagnostic techniques was outlined in the participant information sheet that was explained in lay terms to the children in school and distributed to the parents for their information when asked to provide written informed consent on behalf of children’s participation in the study . All school-aged children were offered praziquantel ( 40 mg/kg ) against schistosomiasis and albendazole ( 400 mg ) against soil-transmitted helminthiasis free of charge in the frame of the island-wide MDA campaigns conducted in June 2013 and November 2013 as part of the elimination interventions . The required number of individuals per prevalence setting to detect significant differences in the diagnostic outcome was calculated with an equation given by Fleiss [34] . Based on preliminary laboratory findings , we estimated the ‘empirical’ prevalence outcomes with the UCAA2000 to be three times higher than with a single urine filtration in prevalence settings below 5% and at least two times higher in prevalence settings of at least 5% . Using a significance level of 5% and a power of 80% , the minimum required number of individuals that had to be examined per prevalence setting was 867 for settings <2% , 235 for settings 2–5% , and 303 for settings 5–10% , according to a single urine filtration . Hence , at least 1 , 405 urine samples were required from individuals stemming from the three respective prevalence settings . The Zanzibar archipelago consist of two main islands , Unguja and Pemba , which are located in the Indian Ocean , approximately 70 km East and 180 km North-East , respectively , from Dar es Salaam , the economic capital of the United Republic of Tanzania located on the mainland’s coast . According to the 2012 population and housing census , Unguja consists of 210 and Pemba of 121 administrative areas ( shehias ) with an approximate combined population of 1 . 3 million inhabitants [35] . Achieving elimination of urogenital schistosomiasis as public health problem on Pemba and interruption of transmission on Unguja are the goals of the Zanzibar president , the Ministry of Health , and an alliance of institutions , including the Schistosomiasis Consortium for Operational Research and Evaluation ( SCORE ) , WHO , the Schistosomiasis Control Initiative ( SCI ) , the Natural History Museum ( NHM ) , and the Swiss Tropical and Public Health Institute ( Swiss TPH ) [31 , 33] . Since early 2012 , biannual MDA on the whole islands and additionally snail control and behavior change interventions in selected communities have been implemented to achieve the primary goals and to learn lessons about which intervention combination works best for elimination . In the frame of a three-arm multi-year intervention trial funded by SCORE , annual parasitological surveys are carried out in 45 randomly selected shehias on both Unguja and Pemba to assess the extent of S . haematobium infections in school children and adults [31] . The urine samples used for the diagnostic investigations presented here were collected from children aged 9–12 years visiting primary schools in 16 shehias on Pemba Island between March and May 2013 , more than 3 months after the last round of MDA that had been carried out in early November 2012 . All urine samples were examined in the laboratories of the Public Health Laboratory-Ivo de Carneri ( PHL-IdC ) in Chake Chake , Pemba . In each primary school , the headmaster and teachers were informed about the aims of the study . Classes of standards 3 and 4 were visited by the field team of the PHL-IdC and the purpose of the study was explained in lay terms to the children . All children aged 9–12 years were asked to line up , stratified by boys and girls , and every third child was selected to participate in the study until 130 children were reached . The name , age , sex , and additional demographic information of these children were recorded and they received an information sheet and a consent form to bring to their parents . If the parents agreed that their child participated , the children were asked to return the signed consent form the following day . After collection of the signed consent forms by the field team , children received a urine collection container ( 120 ml ) and were asked to fill it with their own urine ( urine collection occurred between 10 a . m . and 12 a . m . ) and to give the filled containers to the field team . At the day of collection , between March and May 2013 , all urine samples of sufficient amount ( at least 10 ml ) were examined by trained laboratory technicians for microhematuria using reagent strips ( Hemastix; Siemens Healthcare Diagnostics GmbH , Eschborn , Germany ) , and for the presence and number of eggs detected under a microscope using the urine filtration method with polycarbonate filters ( Sterlitech , Kent , WA , United States of America ) . All urine filters were covered with hydrophilic cellophane soaked in glycerol solution and the slides were stored for a potential second reading for quality control . At the day of collection , if a sufficiently large amount of urine was submitted , 1 . 8 ml of urine was frozen and stored at -20°C from children with IDs 1–100 from each shehia for future examinations , before subjecting to reagent strip testing and urine filtration . The frozen samples from children from the 16 shehias selected for this study were examined with the UCAA2000 or UCAA250 assays in November 2013 at PHL-IdC . Four laboratory technicians received an in-depth training for preparation of samples and how to conduct the UCAA2000 and UCAA250 tests by two of the authors ( PLAMC and GJvD ) at PHL-IdC . Supervised by , and in collaboration with a trained post-doctoral fellow ( CIC ) , the technicians examined the samples as described elsewhere [26 , 27] blinded to the reagent strip and initial urine filtration reading results . In brief , for the UCAA2000 procedure , 1 . 5 ml of urine was mixed with 1 . 5 ml of 4% tri-chloro-acetic acid ( TCA ) in a 10-ml tube and centrifuged at 4 , 000 revolutions per min ( rpm ) using a table centrifuge ( Hettich Universal 320 centrifuge; Tuttlingen , Germany ) for 15 min . From each tube , 2 . 8 ml of the supernatant was added to an Amicon Ultra-4 Centrifugal Filter Device ( Merck Millipore ) and centrifuged at 4 , 000 rpm for 30–45 min . From the concentrate , 20 μl were added to a tube placed in a 96-wells rack , which contained dried UCP-conjugate that was resuspended in 100 μl of assay buffer . After incubation at 37°C at 900 rpm for 1 h on a microtiter-plate thermo-shaker ( L079; Kisker Biotech GmbH & Co; Steinfort , Germany ) , one lateral flow strip was placed into the solution . After the solution’s volume had been soaked up and run through the strip , it was left in the tube to dry for at least 2 h ( usually overnight ) . The following day , the strips were read in a portable strip reader ( UCP-Quant: QIAGEN Lake Constance GmbH; Hilden , Germany ) [27] . Scans were analyzed with Lateral Flow Studio version 3 . 03 . 05 ( QIAGEN Lake Constance GmbH; Hilden , Germany ) . A similar procedure was followed for the UCAA250 , except that the Amicon Ultra-0 . 5 Centrifugal Filter Devices were loaded two times , allowing a sample volume of 1 ml supernatant ( representing 500 μl urine ) to be tested . The devices were centrifuged in a benchtop microcentrifuge ( Eppendorf Mini Spin; Hamburg , Germany ) at 14 , 000 rpm for 2 x 15 min . The stored urine filtration slides from all individuals , whose urines were examined with a UCP-LF CAA test , were retrospectively re-read between November 2013 and January 2014 by a post-doctoral fellow ( CIC ) blinded to the reagent strip , initial urine filtration , and UCP-LF CAA results . This second reading is indicated as quality control urine filtration ( QCUF ) . The results from the reagent strip testing for microhematuria , urine filtration , and QCUF results were recorded on paper laboratory forms and subsequently double entered into a Microsoft Excel 2010 electronic database ( Microsoft Corporation 2010 ) and cleaned . Discrepant results in the double entry were traced back in the original paper record forms and corrected . Results of the UCAA2000 and UCAA250 tests were directly transferred from the UCP-Quant reader into an electronic format . Data were analyzed using STATA version 12 ( StataCorp . ; College Station , TX , United States of America ) and Mplus V7 [36] . Microhematuria was graded into negative , trace , 1+ , 2+ , and 3+ according to the color chart provided by the manufacturer . S . haematobium egg numbers were recorded per 10 ml of urine . The concentration of CAA in urine was calculated using standard curves derived from daily freshly prepared concentration series of partly purified antigen and expressed as pg/ml . High and low specificity cut-offs were determined as described elsewhere [24 , 27] . A sample was considered positive at CAA values of >0 . 4 pg/ml , as indecisive at 0 . 2–0 . 4 pg/ml , and as negative at <0 . 2 pg/ml for the UCAA2000 assay . Samples tested with the UCAA250 were considered as positive at CAA levels of >1 . 4 pg/ml , indecisive at 0 . 7–1 . 4 pg/ml , and as negative at <0 . 7 pg/ml . Of note , applied cut-off values are slightly different from those described by Corstjens et al . ( 2014 ) [27] , and directly related to the ( slightly smaller ) sample volume input and the concentration factor obtained with the Amicon concentration devices . The selection of schools with S . haematobium prevalences of <2% , 2–5% , and 5–10% for inclusion into the present study was based on results of the initial urine filtration examination performed on the day of sample collection and including all children with written informed consent , microhematuria , and urine filtration results . For assessing diagnostic accuracy , however , we only included data from individuals with complete diagnostic results on ( i ) reagent strip testing; ( ii ) urine filtration reading; ( iii ) UCAA2000 testing ( considering indecisive results either as positive ( UCAA2000+ ) or as negative ( UCAA2000- ) or as missing , depending on the approach applied to calculate diagnostic accuracy described below ) ; and ( iv ) QCUF reading into the final analysis . While urine samples stored for UCP-LF CAA examination were not selected fully at random ( i . e . , only urine samples of sufficient amount of the first 100 among 130 collected samples per school were stored ) , we nevertheless considered this approach as valid and assumed complete randomness of missing samples ( and that missing values are unrelated to the status of S . haematobium infection ) , since the overall percentage of positive individuals detected by the initial urine filtration did not differ between the initially sampled group ( 3 . 3%; Table 1 ) and the group included into the final analysis ( 3 . 4%; Table 2 ) . From this subsample , we calculated ‘empirical’ prevalences obtained by each diagnostic method , assuming 100% test specificity . Diagnostic accuracy parameters , including 95% confidence intervals ( CIs ) , were assessed using two different approaches . In the first approach , we considered the combined results of QCUF and UCAA2000+ as imperfect ‘gold’ standard and an individual was regarded as true-positive when the QCUF and/or UCAA2000+ indicated a S . haematobium infection . A test specificity of 100% was assumed for each method . The sensitivity of each individual test was determined by calculating the proportion of positives that were correctly identified by the test when compared to the imperfect ‘gold’ standard . The sensitivity of all diagnostic tests was calculated for ( i ) combined data from all individuals included into the final analysis and ( ii ) stratified data according to the originally selected different prevalence levels ( <2% , 2–5% , and 5–10% ) . To assess a correlation between CAA pg/ml levels and the number of eggs detected in 10 ml urines or microhematuria grading identified with reagent strips , we applied the non-parametric Spearman’s rank correlation test . In the second approach , in the absence of a true ‘gold’ standard , we used LCA to estimate the sensitivity , specificity , and model estimated prevalences for reagent strip , QCUF , and UCAA2000 [37–39] . Four LCA models were applied and validated . The exact procedure is presented in a supplementary file ( S1 Models ) and model details have been described by Ibironke and colleagues ( 2012 ) [38] . The four LCA models were fitted using MPlus V7 [36] with full information maximum likelihood estimation and assuming that data were missing at random . We included the indecisive results of the UCAA2000 in all LCA models by considering them as ‘missing’ and not forcing them in a positive or negative category [40] . The four LCA models were evaluated according to the lowest Bayesian information criterion ( BIC ) and Akaike information criterion ( AIC ) as indications of the best model fit and parsimony in combination with different biological plausible scenarios and tests of assumptions . In the results section , we present results from LCA model 1 ( Table S1 , Model 1 in S1 Models ) .
To meet the prevalence thresholds and sample size for the study , we selected eight primary schools with a prevalence of S . haematobium <2% , four schools with a prevalence of 2–5% , and four schools with a prevalence of 5–10% based on single urine filtration readings per child ( Table 1 ) . From the 16 selected schools , 2 , 067 children were randomly selected to participate in the annual parasitological survey in 2013 . Among them , 298 did not provide written informed consent from their parents and were therefore not asked to submit a urine sample ( Fig 1 ) . An additional 29 children did not submit a urine sample of a sufficiently large amount to perform reagent strip and urine filtration examinations . Hence , the initial S . haematobium prevalence at the unit of the school was calculated from single urine filtration results of 1 , 740 children . Among the 1 , 740 children examined with the initial urine filtration and reagent strip , 444 had no urine sample stored for future analysis , and 12 urine filtration slides were not available for reexamination by the quality control reader . Finally , UCP-LF CAA and QCUF readings were available from 1 , 284 children . The UCAA2000 and UCAA250 were applied on 1 , 200 and 84 urine samples , respectively . The sample sizes of the UCAA2000 tests for prevalence settings <2% , 2–5% , and 5–10% were 546 , 326 , and 328 , respectively . While the numbers for the latter two prevalence settings are in line with our initial sample size calculation , there were fewer UCAA2000 tests performed for the lowest prevalence settings . This is due to interim analyses , which revealed that the prevalence outcomes obtained with the UCAA2000 in these very low prevalence settings were seven times higher than those based on a single urine filtration , and hence the sample size could be lowered to around 500 UCAA2000 tests . The UCAA250 was performed to gather new information on a different approach , which is potentially less complicated to perform in resource-constrained settings . Fig 2 shows that there were considerable differences in the empirical S . haematobium prevalences at the unit of the school and according to endemicity level , depending on the diagnostic approach applied . The thresholds of <2% , 2–5% , and 5–10% were only met with the initial urine filtration reading , indicating an average prevalence of 1 . 5% , 3 . 4% , and 6 . 7% for the schools stratified to each endemicity level ( Table 2 ) . The QCUF reading revealed slightly higher average prevalences of 2 . 6% , 4 . 6% and 8 . 5% , respectively , and reagent strips of 1 . 8% , 5 . 8% , and 6 . 1% , respectively . Considerably higher prevalences for individual schools and average prevalences according to endemicity level were revealed by the UCAA2000 . Considering indecisive results as negative , the UCAA2000- indicated average prevalences of 10 . 1% , 12 . 0% and 19 . 8% , respectively , for the three endemicity levels . Considering indecisive results as positive , the UCAA2000+ indicated average prevalences of 14 . 8% , 17 . 2% and 24 . 7% , respectively . The numbers of positive , negative , and indecisive results for each diagnostic method combination are shown in 6-cell matrixes in Table 3 . Since QCUF revealed more S . haematobium-infected individuals than the initial urine filtration microscopy , only QCUF results were considered for the calculation of sensitivity and specificity . Applying a combination of the QCUF and UCAA2000+ as imperfect diagnostic ‘gold’ standard , the UCAA2000+ had the highest overall sensitivity of 95 . 2% , followed by the UCAA2000- with a sensitivity of 69 . 4% ( Table 4 ) . The QCUF and reagent strips showed very low sensitivities ( 24 . 9% and 16 . 6% , respectively ) . While the UCAA2000+ showed stable sensitivity across the three S . haematobium prevalence settings ( <2% , 2–5% , and 5–10% ) , a decreasing trend in sensitivity with lower prevalence was observed for the UCAA2000- and particularly for the QCUF . A considerable drop in the sensitivity of reagent strip results only occurred in the <2% prevalence setting . Changes in sensitivity were , however , not statistically significant . Noteworthy , the geometric mean egg count decreased significantly from highest to lowest prevalence settings from 0 . 22 eggs/10 ml urine to 0 . 05 eggs/10 ml urine . As shown in Fig 3 , we found a significant relationship between CAA pg/ml levels and S . haematobium egg counts ( Spearman’s rho = 0 . 24; p<0 . 001 ) , between CAA pg/ml levels and microhematuria grading ( Spearman’s rho = 0 . 23; p<0 . 001 ) , and between egg counts and microhematuria grading ( Spearman’s rho = 0 . 57; p<0 . 001 ) . Statistical information criteria ( i . e . , AIC and BIC ) indicated that no random effects were needed at the school level , suggesting that neither the diagnostic performance of reagent strip , QCUF , or UCAA2000 tests , nor the model estimated S . haematobium prevalence varied significantly between the surveyed schools . The assumption of conditional independence between the three diagnostic tests was considered as valid , since inspection of the standardized results from the final selected model ( S1 Model 1 ) did not show extreme values ( i . e . , residuals for all response patterns were between -2 and 2 ) . Furthermore , when we allowed for partial conditional independence between reagent strip and QCUF results , the model fit was not improved ( S1 Model 4 ) , which further strengthened the argument for conditional independence between the tests . Our final LCA model ( S1 Model 1 , with the lowest AIC and BIC ) revealed a sensitivity of 97 . 0% ( 95% CI: 90 . 5–100% ) , 85 . 5% ( 95% CI: 72 . 2–98 . 8% ) , and 66 . 7% ( 95% CI: 52 . 4–81 . 0% ) for UCAA2000 , QCUF , and reagent strip , respectively . The highest specificity was obtained for QCUF ( 99 . 1% , 95% CI: 98 . 5–99 . 7% ) , followed by reagent strip ( 98 . 9% , 95% CI: 98 . 3–99 . 5% ) , and UCAA2000 ( 90 . 1% , 95% CI: 88 . 3–91 . 9% ) . The model estimated S . haematobium prevalence including all schools was 4 . 5% .
Enhanced efforts to achieve the schistosomiasis control and elimination goals put forth by WHO for the years 2020 and 2025 will likely reduce the Schistosoma prevalence and infection intensities in targeted populations . To discover and investigate continuing transmission and to reliably confirm schistosomiasis elimination without missing very light infection intensities , diagnostic tools with high sensitivity and specificity are needed [11 , 19–22 , 41] . We assessed the accuracy of the UCAA2000 assay for S . haematobium diagnosis in low-endemicity settings on Pemba Island . Based on a single urine filtration , we selected schools with a prevalence of S . haematobium <2% , 2–5% , and 5–10% . LCA revealed an overall sensitivity and specificity of a single UCAA2000 of 97 . 0% and 90 . 1% , single QCUF of 85 . 5% and 99 . 1% , and single reagent strips of 66 . 7% and 98 . 9% , respectively , and a model estimated prevalence of 4 . 5% . No significant drop in the empirical sensitivity from highest to lowest investigated endemicity scenario was revealed , but we observed a clear tendency of decreasing sensitivity of particularly the QCUF and reagent strip test results with lower prevalence and geometric mean egg count levels . The overall S . haematobium prevalence empirically determined with the UCAA2000+ , UCAA2000- , urine filtration , and reagent strips were 18 . 2% , 13 . 3% , 4 . 8% , and 4 . 1% , respectively . Our results show that empirically , a single UCAA2000 test detects a considerably higher S . haematobium prevalence than microscopy or reagent strips . Even if indecisive results obtained with the UCAA2000 were considered as negative , the overall S . haematobium prevalence was almost three times higher than that elucidated by QCUF . Particularly evident were the differences in the lowest endemicity setting , where UCAA2000+ and UCAA2000- revealed prevalences of 14 . 8% and 10 . 1% , respectively , while QCUF and reagent strips found considerably lower prevalences of 2 . 6% and 1 . 8% , respectively . If indeed correct , this finding would have important ramifications for the schistosomiasis elimination program in Zanzibar . Since egg output and microhematuria were reasonably low , according to the current definitions , elimination of schistosomiasis as a public health problem had been reached and the setting was on the way toward interruption of transmission . The UCAA2000 revealed , however that more than 10% of the surveyed population excreted CAA and thus harbored living worms , which potentially could produce eggs at some point passed in urine . Given these persistent very low intensity infections in the presence of ongoing control interventions and the potential for disease recrudescence or the parasite’s reintroduction into parasite-free environments also by very modest external inputs , the situation would require an adequate response in terms of more effective locally targeted control strategies [42] . The model estimated overall S . haematobium prevalence , in contrast , was only 4 . 5% , taking into account an imperfect specificity of the UCAA2000 , which was estimated by LCA at 90 . 1% . Considering this imperfect specificity , a considerable amount of S . haematobium cases detected by the UCAA2000 might have been false-positives . However , a specificity of 90 . 1% is below the specificity of circulating antigen assays and the UCP-LF CAA test as postulated elsewhere [27 , 29 , 43 , 44] . Moreover , studies with the UCAA2000 in non-endemic African settings using samples from previous studies , banked at Leiden University Medical Center revealed that no false-positives were detected by the method . One also has to note that CAA is released by living worms that might or might not produce eggs . In case children were reached by the latest MDA conducted in November 2012 and knowing that CAA clears within a few days or weeks after successful treatment [45 , 46] , it might be that the CAA-positive results in the egg-negative urines collected between March and May 2013 indicate worms that survived but were sterilized by the previous praziquantel treatment , or worms that were schistosomula at the time of treatment and not affected by praziquantel , or new infections with schistosomes that were not yet producing eggs . Since we were working in an elimination setting where transmission is mostly low , it might also be that CAA-positive but egg-negative individuals were infected with single worms rather than worm pairs , and hence no eggs were produced . The sensitivity of the UCAA2000 of 97 . 0% determined with the LCA and of the UCAA2000+ of 95 . 2% determined with an imperfect ‘gold’ standard in our study , is in line with findings from the People’s Republic of China , where the UCAA2000+ sensitivity for S . japonicum detection was 93% [29] . A limitation of our study and approaches to estimate sensitivity and specificity is that the UCAA2000 was only compared to two diagnostic techniques with a limited accuracy , particularly for examining samples from a low-endemic area and when only a single urine sample was examined . Clearly , the UCAA2000 has a very high sensitivity and a sufficient specificity to serve as a tool for diagnosing urogenital schistosomiasis in low-endemicity settings targeting elimination . Whether its specificity is high enough to deserve also the title “confirmation of elimination tool” remains to be elucidated in future studies comparing its performance not only with microscopy and reagent strips but also with more accurate methods , for example with polymerase chain reaction ( PCR ) [47] , in a similar close-to-elimination setting . The considerable number of “potentially positive” individuals determined by “indecisive” UCAA2000 results is another limitation of our study . Indecisive results appeared due to the selection of a higher specificity and a lower specificity cut-off . As described in Corstjens et al . ( 2014 ) , the cut-off thresholds may be influenced by technical factors such as batch-to-batch variation , as well as ( immuno- ) epidemiological settings ( e . g . , co-infections , age , and geography ) [27] . In the current setting , suboptimal sample volume and centrifugation capacities were considered the main reason for the need of a higher cut-off threshold . Moreover , the definition of a precise cut-off for any UCP-LF CAA approach is pending and can only be developed for large reagent strip batches and by testing a large number of clearly uninfected people from endemic and non-endemic settings . Currently , to obtain a clear result for potentially positive individuals , their urine sample would either need to be repeatedly tested with the UCAA2000 , or larger amounts of their urines would need to be examined , for example with the UCAA7500 , which has a lower detection limit of 0 . 03 pg/ml [27] , or their CAA level could be retested after praziquantel treatment to investigate whether it decreased . It is worth mentioning that a rigorous second examination of the urine filtration slides by an experienced slide reader ( i . e . , QCUF ) revealed a higher number of S . haematobium egg-positive slides than the initial urine filtration reading of the same slides by local technicians . This finding confirms that ( i ) the glycerol soaked cellophane cover method we used to preserve our urine filtration slides for more than 6 months for quality control purposes works; and ( ii ) that conducting quality control of urine filtration slides is essential to achieve more reliable results for studies on diagnostic accuracy as well as for prevalence estimates in schistosomiasis control and elimination programs . Our study shows that the UCAA2000 is a highly sensitive diagnostic tool that is able to diagnose S . haematobium infections in very low endemicity settings . The dry format allows convenient transport of dry reagents without a cold chain to third-party laboratories [27] . The assay can be implemented by trained local technicians in laboratories in endemic settings , given they are adequately equipped such as the PHL-IdC in Pemba . When sufficient centrifugation capacities and a UCP-Quant reader are available , up to 100 samples can be processed by one technician per day , and hence , the test has a higher throughput than parasitological approaches requiring microscopy . However , in the current format , the UCAA2000 cannot be applied in field laboratories without centrifugation and pipetting capacities . Moreover , the costs for a single UCAA2000 are high and the test is not commercialized . Hence , at the time being , this test is out of reach for most control programs and large-scale studies in endemic areas , but is only used in collaborative projects . Efforts to develop a simple-to-use but still highly sensitive point-of-care ( POC ) -CAA rapid test for commercialization at an affordable price are underway . One also has to consider that the UCAA2000 is a highly sensitive test to reflect active infection , but that it does not encapsulate morbidity . In low endemic settings , where the primary aim is to assess transmission , this test holds particular potential . For indicating morbidity caused by urogenital schistosomiasis , rapid tests such as reagent strips showing the grade of hematuria , and detecting proteinuria and leukocyturia or urine albumin and creatinine have clear advantages [22 , 48 , 49] . Once standardized , commercialized , and widely available at reasonable costs , we consider the UCAA2000 as a suitable tool for large-scale monitoring of urogenital schistosomiasis in control programs in low-endemicity settings targeting elimination and for surveillance in areas that achieved elimination . For surveillance at a smaller scale , including testing of suspected cases in remote public health care centres without laboratory equipment , a simple-to-use but still highly sensitive POC-CAA rapid test is highly desirable . | The World Health Organization aspires to eliminate snail fever ( schistosomiasis ) as a public health problem and to interrupt the transmission of this disease in selected areas by 2025 . Efforts to achieve these goals are currently being intensified . As a result , the prevalence and intensity of infection will decline in many parts of the world . To detect light-intensity infections , diagnostic tools with a high sensitivity and specificity are needed . We assessed the accuracy of a method that is able to diagnose schistosomiasis via the detection of circulating anodic antigen ( CAA ) in urine . We examined 1 , 200 urine samples from children living on Pemba Island , Tanzania , a low-endemic area targeted for schistosomiasis elimination . We found that the CAA-test had a considerably higher sensitivity than conventional urine filtration microscopy and reagent strips that are widely used in schistosomiasis control programs . The empirical prevalence of infection with the parasite Schistosoma haematobium determined by the CAA-test was up to 10 times higher than that obtained by urine filtration . Our results suggest that the CAA-test—in combination with urine filtration—is a promising approach for the diagnosis of S . haematobium in low-transmission settings that are targeted for elimination . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [] | 2015 | Sensitivity and Specificity of a Urine Circulating Anodic Antigen Test for the Diagnosis of Schistosoma haematobium in Low Endemic Settings |
The natural evolution of rabies virus ( RABV ) provides a potent example of multiple host shifts and an important opportunity to determine the mechanisms that underpin viral emergence . Using 321 genome sequences spanning an unprecedented diversity of RABV , we compared evolutionary rates and selection pressures in viruses sampled from multiple primary host shifts that occurred on various continents . Two major phylogenetic groups , bat-related RABV and dog-related RABV , experiencing markedly different evolutionary dynamics were identified . While no correlation between time and genetic divergence was found in bat-related RABV , the evolution of dog-related RABV followed a generally clock-like structure , although with a relatively low evolutionary rate . Subsequent molecular clock dating indicated that dog-related RABV likely underwent a rapid global spread following the intensification of intercontinental trade starting in the 15th century . Strikingly , although dog RABV has jumped to various wildlife species from the order Carnivora , we found no clear evidence that these host-jumping events involved adaptive evolution , with RABV instead characterized by strong purifying selection , suggesting that ecological processes also play an important role in shaping patterns of emergence . However , specific amino acid changes were associated with the parallel emergence of RABV in ferret-badgers in Asia , and some host shifts were associated with increases in evolutionary rate , particularly in the ferret-badger and mongoose , implying that changes in host species can have important impacts on evolutionary dynamics .
Revealing how viruses jump species boundaries and establish productive infections in new hosts is key to understanding disease emergence . As most recent emerging and re-emerging viruses have RNA genomes [1] , it is of central importance to understand the drivers of RNA virus evolution , diversification and cross-species transmission . Clearly , successful virus emergence has diverse causes , likely involving anthropogenic , social and environmental factors [2] . However , the capacity of the viral genome to vary and generate advantageous mutations is also an important element , enabling RNA viruses to exploit new niches , including novel host species , often more rapidly than DNA-based organisms [1 , 3 , 4] . One important manifestation of RNA virus evolution and diversification is the rate of evolutionary change ( i . e . nucleotide substitution ) , with analyses of how this parameter varies by host species providing important information on the nature of virus-host interactions . Disease emergence results from complex mechanisms that shape the ability of a virus to be maintained within its primary host species , then be serially transmitted to a new host species and initiate a pathologic process to cause disease [5] . As such , lyssaviruses ( family Rhabdoviridae ) , the causative agents of rabies–an acute and almost invariably fatal encephalomyelitis in humans–represent an informative case study to examine the relationship between virus genetic diversity and disease emergence . In particular , the natural history of these zoonotic viruses provides an excellent model to study how replication in different host species alters the selection pressures that act on virus genomes . Lyssaviruses are single-stranded , negative-sense RNA viruses with a genome size of approximately 12 kb that encodes five proteins: the nucleoprotein ( N ) , the phosphoprotein ( P ) , the matrix protein ( M ) , the glycoprotein ( G ) and the Large protein or polymerase ( L ) . Currently , the lyssaviruses are classified into 14 species and one tentative species [6] . Like other RNA viruses , lyssaviruses exhibit high rates of mutation due to a lack of proofreading activity in the L protein [7] . Notably , although many mammalian species appear to be susceptible to lyssavirus infection , the virus is only able to establish sustained transmission networks in a relatively small number , indicating that there are major barriers to successful cross-species transmission [8–11] . One species of lyssavirus , rabies virus ( RABV ) , is present worldwide and circulates in a diverse set of reservoir hosts among the mammalian orders Chiroptera and Carnivora [12] . Its natural evolution provides an illustrative example of multiple host switches , in turn enabling comparative studies of the evolutionary patterns , processes and dynamics associated with host adaptation . Previous studies demonstrated that RABV isolates fall into two major phylogenetic groups; the bat- and the dog-related RABV groups [8 , 13 , 14] . The ‘bat-related’ RABV group is confined to New World viruses circulating mainly among bats , as well as in some terrestrial carnivores such as skunks and raccoons [14–17] . In contrast , the ‘dog-related’ RABV group contains viruses circulating worldwide in dogs , as well as in wildlife carnivores in specific geographic areas such as foxes and raccoon dogs in Europe , foxes in the Middle East , raccoon dogs and ferret-badgers in Asia , skunks , foxes , coyotes and mongooses in the Americas , and mongooses in Africa [14 , 16 , 18–22] . Importantly , dogs are responsible for more than 99% of the human rabies cases worldwide [23] and are likely the main vector for the inter-species transmission of dog-related RABV . Previous phylogenetic analyses have largely been performed on individual genes [13–19 , 21 , 24–29] with a few assessing the full-length viral genome [20 , 30 , 31] . In addition , most of these phylogenetic studies were performed on relatively small numbers of sequences originating from one specific geographical area and/or associated with a specific animal host [20 , 22 , 30 , 32 , 33] . Despite these limitations , these studies are consistent in showing that RABV is subject to strong purifying selection [10] coupled to geographical clustering that is occasionally disrupted by human mediated dispersion [13 , 34 , 35] . Recently , it was shown that nucleotide substitution rates in RABV vary markedly among those viruses infecting bats , such that rates in tropical and subtropical species were markedly higher than those from temporal bat species , perhaps reflecting a combination of host and environmental factors [36] . However , equivalent data for dog-related RABV are lacking . In addition , whether evolutionary rates in RABV vary among wild carnivores and domestic dogs is unknown , although studies in other systems have revealed that rates of RNA virus evolution may differ between wild and domestic animals [37] . Clearly , the large-scale analysis of RABV , particularly comprising full-length genome sequences , is needed to reveal the nature of the selection pressures associated with host switching . That the RABV genome encodes a limited number of proteins that necessarily have multifunctional roles [38] , and hence potentially large-scale epistasis , also means that these selection pressures may be complex . Herein we present the first phylogenetic study of RABV on a genome-wide and global scale , utilizing a data set of 321 whole-genome sequences sampled from 66 countries over a time period of 65 years , with the aim of inferring those evolutionary patterns and processes associated with host-switching . In particular , we compared RABV from wild carnivores and in domestic dogs with respect to selection pressures , evolutionary rates , and the time-scale of their evolutionary history . Importantly , the size of the data set allowed us to reveal any heterogeneity in evolutionary rates among RABV adapted to different primary hosts , and determine the complex evolutionary dynamics of RABV as it adapts to new hosts .
A phylogenetic analysis was performed on the ( 99% ) full-length genome sequences of 321 RABV sequences sampled from 66 countries ( S1 Fig , S1 Table ) . Of these viruses , 170 were newly sequenced as part of this study . As expected given the low levels of recombination in RABV , the topology of the maximum likelihood ( ML ) tree performed on the five concatenated RABV genes ( Fig 1 ) was similar to that obtained for each individual gene ( N , P , M , G and L genes ) and for the concatenated non-coding sequence ( S2 Fig ) . In particular , two major phylogenetic groups were apparent , corresponding to bat- and dog-related RABVs , each of which can be further subdivided into several major clades . This is consistent with previous analyses of smaller data sets and on individual RABV genes [13 , 14 , 16 , 29] . The bat-related group contained two major clades , one including the bat RABVs circulating in the Americas , and the other ( RAC-SK ) comprising viruses from American skunks and raccoons ( Fig 1 ) . In turn , the RAC-SK group contained a number of ‘subclades’ corresponding to Mexican skunks ( MeSK-1 ) , North American raccoons ( RAC ) and South-Central skunks ( SCSK ) as previously described ( S3 Fig ) [14 , 16 , 17 , 39] . Similarly , the dog-related group includes six major clades supported by high bootstrap values ( S2 Table and Fig 1 ) , and previously identified as the Africa-2 , Africa-3 , Arctic-related , Asian , Cosmopolitan and Indian subcontinent clades [13] . The phylogenetic analysis based on the five concatenated genes was particularly informative , allowing us to distinguish various subclades and lineages among these six major clades , with some of which are characterized for the first time here ( Fig 1 and supplementary text ) . Some of these clades and subclades are of particular interest . The SEA2 subclade contains viruses from China and is divided into two lineages , SEA2a and SEA2b , corresponding to isolates from dogs and ferret-badgers , respectively . Subclade SEA5 appears to be specific to RABV circulating in ferret-badgers in Taiwan , an epidemiological cycle that was only identified recently [21 , 40 , 41] . For the first time , we were also able to fully characterize full-length genome sequences of RABV isolates belonging to the Africa-3 clade ( n = 6 ) . These viruses circulating in Southern Africa are monophyletic and phylogenetically distinct from the other major RABV clades , particularly those circulating in Africa [19 , 42 , 43] . To determine the evolutionary dynamics of RABV , we first determined whether individual data sets contained sufficient temporal structure to undertake detailed molecular clock analyses by performing a regression of root-to-tip genetic distance against the year of sampling . Notably , no correlation between time and genetic divergence was found when the sequences of both bat- and dog-related RABV groups were analyzed together , indicating that there is extensive variation in the rate of RABV evolution among these taxa ( and hence that they should not be combined in molecular clock studies ) ( S4A Fig ) . In addition , no temporal structure was observed when the sequences of bat-related RABV were analyzed separately , indicating that this subset of viruses is not evolving uniformly ( S4B Fig ) as noted previously [36] . However , a clear association between genetic distance and time ( i . e . a molecular clock ) was observed for the dog-related group alone ( S4C Fig ) , allowing us to estimate substitution rates , and hence times to common ancestry , more precisely in this cluster using a Bayesian approach . The mean rate of evolutionary change in the dog-related RABV was estimated to be 2 . 44 x 10−4 subs/site/year ( 95% HPDs of 2 . 10–2 . 80 x 10−4 subs/site/year ) for the five concatenated genes . Importantly , we were also able to compare the substitution rate of each RABV gene and of the concatenated non-coding regions from the same genomic sequence data set . These estimates varied in the following ascending order: N , L , G , M and P ( Fig 2A ) . However , only the P gene had a nucleotide substitution rate considerably higher than those of N and L genes . As expected , the evolutionary rate in the non-coding regions was significantly higher than those of the coding regions , indicative of weaker selective constraints . The estimation of a reliable substitution rate allowed us to determine the mean times to common ancestry ( TMRCA ) for each RABV clade , subclade and lineage defined above ( Fig 3; estimates for different sub-clades shown in S3 Table and discussed in the Supplementary text ) . For this analysis we utilized the concatenated coding genes as these had the lowest variance . Notably , these TMRCA estimates exhibited less uncertainty than previous studies performed on N and/or G genes alone [13 , 25 , 27 , 29 , 43 , 44] . Briefly , the TMRCA of the dog-related RABV group was estimated to be approximately between 1308–1510 ( 95% HPD; mean of 1404 ) . Within this group , the Indian subcontinent clade branched basally and appeared to diversify between 1733–1840 ( mean of 1785 ) . The TMRCA of the Asian clade was estimated to be between 1535–1677 ( mean of 1604 ) , which is in accordance with other studies [44] . The emergence of the Africa 2 clade was estimated to be between 1750–1852 ( mean of 1802 ) , similar to the mean TMRCA found in a previous study conducted on complete N and G genes [27] . The Arctic-related clade appeared between 1725–1815 ( mean of 1770 ) , slightly earlier than previously estimated [25] , while the Africa-3 clade emerged between 1710–1815 ( mean of 1756 ) in accordance with another study [43] . Finally , for the first time , we estimate that the TMRCA for the Cosmopolitan clade existed between 1687–1773 ( mean of 1730 ) . The root-to-tip regression analysis also revealed that different groups of dog-related RABV have seemingly evolved at different rates ( S4C Fig ) , with a number appearing as distinct outliers . Interestingly , these outliers were confined to RABV circulating in mongooses in Africa ( Africa-3 clade ) and in ferret-badgers in Asia ( the SEA5 subclade and SEA2b lineage ) , suggesting that they might represent species-specific variation . To address this , we compared the evolutionary rates of these clusters to both the entire dog-related RABV group and to subsets of this group representing dog-related viruses circulating in Africa and Asia , and to mongoose viruses circulating in the Caribbean . For this analysis we focused on the N and G genes as they comprise the largest data sets . This analysis revealed that the N gene of those viruses circulating in ferret-badgers in Asia ( n = 81 ) and in mongooses in Africa ( n = 47 ) evolved between 2–4 times more rapidly than those of the whole dog-related group ( n = 248 ) , at rates of 7 . 82 x 10−4 subs/site/year ( 95% HPD 3 . 14–12 . 83 x 10−4 subs/site/year ) and 5 . 88 x 10−4 subs/site/year ( 95% HPD 3 . 67–8 . 11 x 10−4 subs/site/year ) , respectively ( Fig 2B ) . Importantly , these estimates and their associated uncertainty do not overlap with those for the dog-related group as a whole . This finding is confirmed using smaller subsets of dog-related RABV from more closely related geographically settings in Asia and in Africa ( Fig 2B ) . Interestingly , the rate of RABV evolution in mongooses in Africa is two times higher than that of RABVs from mongooses in the Caribbean ( i . e . Puerto Rico , Cuba and Grenada ) that belong to the Cosmopolitan clade ( Fig 2B ) . Although less rate variation was observed in the G gene , RABV associated with ferret-badgers in Asia still evolved considerably more rapidly than those obtained with the different subsets of dog-related RABV ( Fig 2B ) . These results were confirmed by using different nucleotide substitution models and a hierarchical phylogenetic model approach ( S4 Table ) [45 , 46] . To determine if the variation in rates of evolutionary change might result from differing selection pressures , we first compared the ratio of nonsynonymous ( dN ) to synonymous ( dS ) substitutions per site . This analysis was performed on each of the five RABV genes of the two major RABV groups . For each gene , the dN/dS ratios of the bat- and dog-related groups are very similar ( and very low ) and followed the same ascending order between genes: N , L , M , G and P genes ( Table 1 ) . Furthermore , we explored the number of positively selected sites using several different approaches ( SLAC , FUBAR and FEL ) [47 , 48] . In each of the two major RABV groups , one position was identified as positively selected by at least two of these methods: positions 496 and 484 in the G protein for the bat- and dog-related groups , respectively ( Table 1 ) . Interestingly , the dN/dS of the N and G genes for the branches leading to sequences found to be outliers in the analysis of evolutionary rates ( Africa-3 clade and ferret-badgers in Asia ) were 1 . 4 to 4 . 7 times higher than those of dog-related RABV data sets used as controls ( S4C Fig and S5 Table ) , but still relatively low . Together , these results are generally indicative of strong purifying selection among all sites and branches of the RABV phylogeny . To investigate selection pressures in greater detail we utilized a modified MEME analysis that considered internal branches of the tree only ( as external branches often contain transient deleterious nonsynonymous substitutions yet to be removed by purifying selection ) [49] . Using this MEME-internal analysis , we identified nine positions to be under positive selection ( N436 , P55 , P154 , P265 , G198 , G476 , L430 , L681 , L2091 ) . In addition , position G484 that was identified as positively selected using SLAC , FUBAR and FEL was not significant ( at the p<0 . 05 level ) in the MEME-internal analysis ( p-value = 0 . 084 ) . Finally , it was also clear that specific amino acid substitutions characterized RABV circulating in mongooses in Africa ( Africa-3 clade ) and in ferret-badgers in Asia ( SEA5 subclade and SEA2b lineage ) ( S6 Table ) . Four substitutions were specific ( i . e . not present in any other dog-related RABV sequences ) to mongoose RABV: two in the nucleoprotein , from Asp to Asn at codon position 88 ( Asp-N88-Asn ) and Leu-N108-Ile , and two in the glycoprotein–Ser-G223-Asn and Pro-G386-Ser . The case of the ferret-badger was more interesting as the host jump to this species from dogs has occurred independently in the SEA5 and SEA2b clades , allowing us to determine whether cross-species transmission in this case is associated with parallel viral evolution . This analysis revealed that two amino acid substitutions were common to all ferret-badger viruses across both clades: Leu-N374-Ser and Lys-L200-Arg . The Leu-N374-Ser substitution is particularly noteworthy as it only occurs in the ferret-badger , this residue is normally highly conserved in RABV , and Leu-to-Ser is a non-conservative amino acid change . Hence , we suspect that Leu-N374-Ser , and perhaps Lys-L200-Arg , facilitate RABV adaptation to ferret-badgers . Notably , neither of these sites was found to be subject to positive selection using the methods employed here ( Table 1 ) .
The central aim of this study was to determine whether the patterns and processes of RABV evolution vary between viruses sampled from different host species reflect the impact of cross-species transmission . To that end we present the largest phylogenomic analysis of RABVs circulating worldwide performed to date . Although the topology of the RABV phylogeny is similar to those obtained previously [13 , 14 , 16 , 29] , it clearly presents a more comprehensive and precise reconstruction of evolutionary history of this virus . In particular , the analysis of the five concatenated genes allowed us to obtain a finer-scale dating of the emergence of the major clades with narrower confidence intervals than obtained previously [13 , 25 , 27 , 29 , 43 , 44] . RABV undergoes relatively frequent cross-species transmission [8 , 11 , 13 , 18] , which provides an opportunity to determine whether host jumping impacts rates of evolutionary change . Notably , we found no correlation between root-to-tip genetic distance and sampling time in the bat-related RABV group , nor when combined with dog-related RABV group , indicating that these viruses have not evolved in a clock-like manner , with substantial rate variation already observed in bat-associated RABV [36] . In contrast , a strong association between genetic divergence and time ( i . e . a molecular clock ) was observed within the dog-related RABV group , with a mean evolutionary rate of 2 . 44 x 10−4 subs/site/year ( 95% HPDs of 2 . 10–2 . 80 x 10−4 subs/site/year ) for the five concatenated genes . This estimate is evidently more precise than those determined previously [13 , 25 , 27 , 30 , 44 , 50–53] . Despite the relative rate constancy in the dog-related RABV , it was striking that some of the clades or sub-clades have experienced substantially higher rates of nucleotide substitution . In particular , viruses circulating in ferret-badgers in Asia ( mainland China and Taiwan ) and in mongooses in Africa have evolved at least twice as rapidly as those of the dog-related group . Although there is some uncertainty in these rate estimates , they do not overlap with the estimates for the entire dog-related RABV group . Determining the evolutionary basis to this rate variation is more complex . Changes in evolutionary rate could only be driven either by changes in background mutation rate ( which we consider unlikely to differ between dog-related RABV ) or , more likely , by changes in the population size and/or incubation time that may vary among different animal hosts [36] . It is also possible that the evolutionary rates estimated here have been impacted by time-dependency , such that they are elevated toward the present ( i . e . in closely related sequences sampled recently ) due to the presence of transient deleterious mutations that have yet to be removed by purifying selection [54] . However , while this may in part explain the high rate in the recently sampled RABV from ferret-badgers , it is unlikely to explain the higher evolutionary rate in mongoose RABV whose evolutionary history sampled here covers a longer time period . In the case of the ferret-badgers , two amino acid changes ( Leu-N374-Ser and Lys-L200-Arg ) have evolved in parallel in the two clades associated which is compatible with the occurrence of adaptive evolution , and which have in turn elevated the nucleotide substitution rate . That these two sites were not detected in analyses of dN/dS suggests that these methods may have limitations when identifying adaptive evolution involving limited amounts of amino acid change . Our analysis also showed that the nucleotide substitution rate varied markedly according to the gene analyzed in the ascending order: N , L , G , M and P . As expected , the two proteins often described as more conserved for RABV—N and L—exhibited the lowest rates , as well as the lowest dN/dS ratios , indicating that they are subject to the strongest purifying selection . Notably , the highest substitution rate and dN/dS was observed in the P protein , perhaps reflecting the weak structural organization of the C-term part of this protein [55 , 56] . The presence of relatively constant molecular clock also enabled us to provide a more robust time-scale for the evolution of the principal geographical clusters of dog-related RABV ( Fig 3 , S3 Table ) . Accordingly , we estimate that the most recent ancestor of all dog-related RABV dates to between 1308 and 1510 . Consequently , any older canid RABV lineages , proposed to have circulated in the Middle-East more than 2000 years ago [57 , 58] , have not survived to be sampled in the current study . Interestingly , the timing of the most recent ancestor of all dog-related RABV circulating to date coincides with the development of the world's first truly global trade network following the explorations of Columbus , Vasco da Gama and Zheng He , commissioned by the Spanish , the Portuguese and the Chinese Ming Dynasty , respectively . This age of exploration and colonization contributed to the establishment of new long distance commercial practices and transoceanic shipping services between 1450 and 1750 [59] . The concomitant dissemination of RABV during this period , probably by dogs travelling by boats with their owners , therefore provides a powerful example of the early human-mediated dissemination of a zoonotic disease . In addition , all the ancestors of the major clades found circulating today in North and South America , Africa , Asia and Europe originated between 1687 and 1840 at the apogee of this international trade and colonization process [59] . This is further exemplified by the global spread of the Cosmopolitan clade . A fundamental question in evolutionary virology is how and why some viruses are seemingly better able to jump species boundaries than others . A compelling theory is that the more closely related the host species in questions , the greater the chance of successful transmission [9 , 60 , 61] . However , it is unclear how strictly this theory holds for RABV [11] , and our results confirm species jumps of RABV among animal species of the order Chiroptera , and from bats to striped skunks ( Mephitis mephitis ) [14 , 62 , 63] . In addition , there is also clearly a geographic component to cross-species transmission as bat-related RABVs are only found in the Americas . More notably , our study clearly confirms that although spill-over infections from wildlife species to dog take place , species jumps involving dog-related RABVs generally occur from dogs to wildlife species of the order Carnivora; not only to the family Canidae ( dog , red fox , raccoon dog ) , but also to more distant species belonging to the families Mustelidae ( ferret-badger ) , Herpestidae ( mongoose ) and Mephetidae ( skunk ) ( S5 Fig ) [13 , 18 , 42 , 43] . These changes in primary animal host species have occurred independently in different localities and at different times during RABV evolution . Further , some carnivore species , notably skunks , are infected by RABV of both dog and bat origin [14 , 16 , 39] . Revealing the respective roles of genetic drift and the selection of advantageous mutations in shaping the genetic diversity of RABV , particularly during host shifts , is a central evolutionary question . There is currently no definitive data on whether dog-related RABV emergence requires active adaptive evolution ( i . e . positive selection ) to in a new host species , or whether it is largely a chance process involving ecological factors facilitating the transmission of a viral strain with the pre-existing necessary genetic characteristics [64]; the latter has been proposed for the repeated outbreaks of bat-related RABV in striped skunks and gray foxes in Arizona [14] and of gray foxes due to skunk-associated RABV in California [65] . Our analysis showed that the dog-related RABV group is subject to strong purifying selection , and when positive selection did occur on internal branches of the phylogenetic tree it was not obviously associated with host jumping . As noted above , however , the failure to detect positive selection in the case of ferret-badger RABV despite the occurrence of parallel evolution suggests that these methods may suffer from false-negatives . Successful cross-species transmission is a complex ecological and evolutionary process , beginning with exposure and contact between the two species , followed by the successful infection of the new host species , and potentially host-adaptive evolution to enable long-term sustained transmission [66 , 67] . However , due to complex interactions among the five viral proteins and with their cellular counterparts , including epistasis [68] , it is often difficult to clearly determine which mutations are advantageous or fixed by genetic drift . Moreover , some mutations in the RABV P protein can improve the modulation of the innate immune response of the host but reduce replication efficiency [69] . That two amino acid changes have evolved in parallel in the ferret-badger alone suggests that they have played a role in host adaptation . Further , it is possible that some of the other amino acid substitutions that define individual viral clades associated with different host species represent host-adaptive sites that have not been identified as positively selected through simple analyses of dN/dS . Clearly , additional large-scale analyses of RABV based on full-length genome sequences , extending that presented here , followed by linked experimental studies including generation of mutant RABVs by reverse genetics and phenotypic testing , are needed to reveal the nature of complex evolutionary processes that occur during host switching . In conclusion , RABV is capable of infecting many mammals but paradoxically is maintained in distinct epidemiological cycles associated with animals almost exclusively from the orders Carnivora and Chiroptera . This strict association between RABV and host-species most likely arose from a combination of historical human-mediated spread of RABV and jumps into new primary host species . These data also suggest that the establishment of dog-related RABV in new carnivore hosts may only require subtle adaptive evolution as demonstrated by parallel evolution in the ferret-badger . Evidently , along with more defined analyses of individual mutations , additional studies are needed to determine the role played by the frequency of exposure , animal host behavior , density of the recipient species , duration of incubation and optimum infectious doses in cross-species transmission .
A total of 321 complete genome sequences of RABV isolates were analysed , originating from a wide variety of host species and collected in 66 countries between 1950 and 2015 . Details of these isolates are described in S1 Table and S1 Fig . Among these genome sequences , 170 came from the archives of the World Health Organization Collaborative Center for Reference and Research on Rabies , or from the National Reference Centre for Rabies , both located at Institut Pasteur , Paris , France . These samples were newly sequenced as part of this study . These data were combined with 151 full-length genome sequences extracted from GenBank and selected to be representative of the overall phylogenetic diversity of RABV . Total RNA was extracted using Trizol ( Ambion ) according to the manufacturer’s instructions from primary brain samples or after an amplification passage on suckling mouse brain . RNA was then reverse transcribed using Superscript III reverse transcriptase with random hexamers ( Invitrogen ) according to manufacturer’s instructions . The complete viral genome ( excluding the 3’ and 5’ extremities , corresponding to the leader and the trailer regions , respectively ) of 160 isolates was amplified with six overlapping PCR fragments by using the Phusion polymerase ( ThermoFisher ) . Details of primers are given in S7 Table . After electrophoresis , each PCR fragment was independently purified using the NucleoSpin Gel and PCR clean-up kit ( Macherey-Nagel ) and quantified using Picogreen dsDNA quantification kit ( Invitrogen ) . For each sample , all six PCR fragments were pooled with equimolar proportions to obtain 500 ng of dsDNA . Different protocols were used for the preparation of libraries and next-generation sequencing on Illumina platforms ( NextSeq 500 , HiSeq2000 , HiSeq2500 or MiSeq platforms ) , depending on the isolates considered ( details provided in S1 Table ) . Briefly , three different protocols were used: ( i ) dsDNA was fragmented by ultrasound with Bioruptor ( Diagenode ) , libraries were prepared using NEXTflex PCR-Free DNA-Seq kit ( Bioo Scientific ) , and then sequenced using an 100 or 150 nucleotides single-end strategy on the HiSeq2500 platform or a 2 x 300 nucleotides paired-end strategy on the MiSeq platform , ( ii ) dsDNA was fragmented by NEBNext dsDNA fragmentase ( New England Biolabs ) , libraries were prepared using NEBNext Ultra DNA Library Prep kit ( New England Biolabs ) and sequenced using an 100 nucleotides single-end strategy on the NextSeq500 platform , and ( iii ) dsDNA libraries were constructed using Nextera XT kit ( Illumina ) and sequenced using a 2 x 150 nucleotides paired-end strategy on the NextSeq500 platform . For nine remaining isolates ( S1 Table ) , the viral RNAs were reverse transcribed using Superscript III reverse transcriptase ( Invitrogen ) and then amplified using the whole-transcription amplification ( WTA ) protocol ( QuantiTect Whole Transcriptome kit; Qiagen ) as previously described [70] . dsDNA was fragmented by ultrasound , libraries were prepared using TruSeq protocol ( Illumina ) and sequenced using an 100 nucleotides single-end strategy on the HiSeq2000 platform . Finally , the sequence of 09035FRA was determined using a shotgun base approach [31] . All reads were pre-processed to remove low-quality or artifactual bases . Library adapters , PCR primers used for amplification of the genome , and base pairs occurring at 5’ and 3’ ends with a Phred quality score <25 were trimmed using AlienTrimmer as implemented in Galaxy [71–74] ( https://research . pasteur . fr/en/tool/pasteur-galaxy-platform/ ) . Reads with lengths of less than half of the original read after these pre-processing steps or those containing >20% of bp with a Phred score of <25 were discarded . The filtered reads were then mapped to complete genome sequences specific for each RABV clade obtained from GenBank using the CLC Genomics Assembly Cell ( http://www . clcbio . com/products/clc-assembly-cell/ ) implemented in Galaxy . The majority nucleotide ( >50% ) at each position with a minimum of coverage of 200 was used to generate the consensus sequence . All consensus sequences were manually inspected for accuracy , such as the presence of intact open reading frames , using BioEdit ( http://www . mbio . ncsu . edu/bioedit/bioedit . html ) . A sequence alignment of the 170 newly sequenced genomes combined with the 151 complete genome sequences from GenBank was constructed using ClustalW2 with default parameters [75] ( http://www . ebi . ac . uk/Tools/msa/clustalw2/ ) implemented in Galaxy and manually adjusted when necessary . Sequence alignments of individual RABV genes ( N , P , M , G and L genes ) and concatenated non-coding regions ( from the stop codon in N to the initiation codon of L ) were also generated . All the full-length genome sequences generated in the present study have been submitted to GenBank ( S1 Table ) . We used jModelTest2 [76 , 77] to determine the best-fit model of nucleotide substitution according to the Bayesian Information Criterion . This revealed that the general time reversible model with proportion of invariable sites plus gamma-distributed rate heterogeneity ( GTR+I+Γ4 ) was optimal for all the RABV data sets compiled here . Phylogenetic trees using the different data sets ( i . e . individual genes , concatenated genes or non-coding regions ) were then estimated using the maximum likelihood ( ML ) method available in PhyML 3 . 0 [78] utilizing SPR branch-swapping . The robustness of individual nodes on the phylogeny was estimated using 1 , 000 bootstrap replicates for the five concatenated gene data set , and using the approximate likelihood ratio test ( aLRT ) with SH-like supports for each individual RABV gene as well as the concatenated non-coding region data set [79] . To determine the degree of clock-like structure in each data set we employed root-to-tip linear regression as available in the TempEst program [80] . For those data sets with sufficient phylogenetic structure we then inferred a maximum clade credibility ( MCC ) tree using the Bayesian Markov chain Monte Carlo ( MCMC ) method available in the BEAST v1 . 8 package [81] by incorporating information on sampling time ( year ) of the dog-related RABV group ( isolates for which the date of sampling was unavailable and vaccine strains were excluded ) . Posterior probability values provided an assessment of the degree of support for each node on the tree . This analysis utilized the GTR+I+Γ4 model of nucleotide substitution , a relaxed ( uncorrelated log-normal ) molecular clock and the constant population size model as a coalescent prior . Ten independent MCMC analyses were run for 100 million steps and sampled every 10 , 000 states . The log and tree files of each MCMC chains were combined using Logcombiner v1 . 8 . 2 ( http://tree . bio . ed . ac . uk/software/beast/ ) , with a burn-in of 10% . The convergence of each parameter in this combined file was checked using TRACER v1 . 6 ( http://tree . bio . ed . ac . uk/software/tracer/ ) and indicated by an effective sample size >200 . The MCC tree was obtained using TreeAnnotator v1 . 8 . 2 ( http://tree . bio . ed . ac . uk/software/beast/ ) . Additional analyses were performed utilizing the GMRF Bayesian Skyride [82] and Bayesian SkyGrid [83] demographic models , and gave similar results . Based on the BEAST analysis , we also estimated the rate of nucleotide substitution per site , per year ( see below ) and the time of most recent common ancestor ( TRMCA ) for host-specific clusters of sequences . The degree of statistical uncertainty in each parameter estimate was given by the 95% highest posterior density ( HPD ) values . The root-to-tip regression analysis performed on the 248 sequences of the dog-related RABV group revealed a number of clear outlier taxa characterized by anomalously high evolutionary rates ( S4C Fig ) . These outliers belong to three clades or sub-clades: the Africa-3 clade that is specific to mongooses in Southern Africa , and the SEA5 and SEA2b subclades that are confined to viruses from ferret-badgers in Taiwan and China , respectively . To further assess if there are considerable differences in evolutionary rate in these clades , we performed additional analyses on the N and the G proteins for which relatively large numbers of sequences were available on GenBank . We therefore collected from GenBank an additional 41 N and 26 G sequences from the Africa-3 clade , and an additional 72 N and 71 G sequences from the ferret-badger in Taiwan and China ( S8 Table ) . These data sets were compared to the N and G sequences of the dog-related RABV group ( n = 248 ) and two RABV subsets corresponding to viruses circulating in dogs in Asia ( n = 51 ) and in Africa ( n = 46 ) . As the Africa-3 clade is specific to the mongoose , we also estimated the evolutionary rate of RABV circulating in mongooses in the Caribbean region , for which we constructed a data set of 64 N sequences ( no G sequences were available ) . As the ferret-badger data set was small , covered a relatively short time-range , and comprised two groups sampled during different time periods , it was unfortunately impossible to analyse the evolutionary dynamics in these two groups separately . Estimates of nucleotide substitution rate of each data set were performed using BEAST as described above . Preliminary analysis on the N and G gene data sets using different nucleotide substitutions models ( GTR+I+Γ4 or GTR+I ) , strict or relaxed ( uncorrelated log-normal ) molecular clocks , constant population size or Bayesian skyline coalescent priors gave similar results . Therefore , all analyses were performed using the GTR+I+Γ4 substitution model , a relaxed ( uncorrelated log-normal ) molecular clock , and a constant population size . Finally , to assess the robustness of our rate estimates we also utilized and hierarchical phylogenetic models [46] . This analysis considered the lineages of the N and G genes defined previously ( i . e . those viruses circulating in ferret-badgers , in mongooses in Southern Africa and the Caribbean , and in dogs in Africa and Asia ) which we treated as data partitions . To be as robust as possible we used two substitution models–SRD06 [45] and GTR+I+Γ4 . For the SRD06 model we specified hyperprior distributions to govern κ ( the relative rate of transitions to transversions ) and the shape parameter , α , of the Γ-distribution among the first and second , and third codon positions . In the case of the GTR+I+Γ4 model we linked each of the six rate parameters of the substitution matrix , α , and the proportion of invariable sites ( I ) . Importantly , for both substitution models we set separate uncorrelated lognormal relaxed clock models and constant-size coalescent tree priors for each of the partitions , which is appropriate because they involve different taxa . To reveal the selection pressures acting on the RABV genome we compared the numbers of nonsynonymous ( dN ) and synonymous ( dS ) substitutions per site for the different RABV genes and phylogenetic clusters using the Single Likelihood Ancestor Counting ( SLAC ) , Fixed Effect Likelihood ( FEL ) , the internal branch Mixed Effects of Model Evolution ( MEME-internal; Kosakovsky Pond SL , personal communication ) and the Fast Unbiased Bayesian Approximation ( FUBAR ) models [47–49] . Only codon positions with a p-value < 0 . 05 for the SLAC , FEL and MEME models and with a posterior of probability > 0 . 95 for the FUBAR method were considered as containing evidence for positive selection . For each data set and gene , the best-fit model of nucleotide substitution model was determined using the model selection tool available on the DATAMONKEY server [84 , 85] . | Zoonoses account for most recently emerged infectious diseases of humans , although little is known about the evolutionary mechanisms involved in cross-species virus transmission . Understanding the evolutionary patterns and processes that underpin such cross-species transmission is of importance for predicting the spread of zoonotic infections , and hence to their ultimate control . We present a large-scale and detailed reconstruction of the evolutionary history of rabies virus ( RABV ) in domestic and wildlife animal species . RABV is of particular interest as it is capable of infecting many mammals but , paradoxically , is only maintained in distinct epidemiological cycles associated with animal species from the orders Carnivora and Chiroptera . We show that bat-related RABV and dog-related RABV have experienced very different evolutionary dynamics , and that host jumps are sometimes characterized by significant increases in evolutionary rate . Among Carnivora , the association between RABV and particular host species most likely arose from a combination of the historical human-mediated spread of the virus and jumps into new primary host species . In addition , we show that changes in host species are associated with multiple evolutionary pathways including the occurrence of host-specific parallel evolution . Overall , our data indicate that the establishment of dog-related RABV in new carnivore hosts may only require subtle adaptive evolution . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"&",
"Methods"
] | [
"sequencing",
"techniques",
"taxonomy",
"organismal",
"evolution",
"microbiology",
"vertebrates",
"animals",
"mammals",
"dogs",
"animal",
"phylogenetics",
"phylogenetics",
"data",
"management",
"phylogenetic",
"analysis",
"microbial",
"evolution",
"molecular",
"biology",
"techniques",
"mammalian",
"genomics",
"zoology",
"research",
"and",
"analysis",
"methods",
"sequence",
"analysis",
"computer",
"and",
"information",
"sciences",
"evolutionary",
"rate",
"evolutionary",
"systematics",
"molecular",
"biology",
"evolutionary",
"genetics",
"viral",
"evolution",
"molecular",
"biology",
"assays",
"and",
"analysis",
"techniques",
"animal",
"genomics",
"virology",
"genetics",
"biology",
"and",
"life",
"sciences",
"genomics",
"evolutionary",
"biology",
"amniotes",
"evolutionary",
"processes",
"organisms"
] | 2016 | Large-Scale Phylogenomic Analysis Reveals the Complex Evolutionary History of Rabies Virus in Multiple Carnivore Hosts |
Virions are thought to contain all the essential proteins that govern virus egress from the host cell and initiation of replication in the target cell . It has been known for some time that influenza virions contain nine viral proteins; however , analyses of other enveloped viruses have revealed that proteins from the host cell can also be detected in virions . To address whether the same is true for influenza virus , we used two complementary mass spectrometry approaches to perform a comprehensive proteomic analysis of purified influenza virus particles . In addition to the aforementioned nine virus-encoded proteins , we detected the presence of 36 host-encoded proteins . These include both cytoplasmic and membrane-bound proteins that can be grouped into several functional categories , such as cytoskeletal proteins , annexins , glycolytic enzymes , and tetraspanins . Interestingly , a significant number of these have also been reported to be present in virions of other virus families . Protease treatment of virions combined with immunoblot analysis was used to verify the presence of the cellular protein and also to determine whether it is located in the core of the influenza virus particle . Immunogold labeling confirmed the presence of membrane-bound host proteins on the influenza virus envelope . The identification of cellular constituents of influenza virions has important implications for understanding the interactions of influenza virus with its host and brings us a step closer to defining the cellular requirements for influenza virus replication . While not all of the host proteins are necessarily incorporated specifically , those that are and are found to have an essential role represent novel targets for antiviral drugs and for attenuation of viruses for vaccine purposes .
Knowledge of the protein composition of a virus particle often serves as an initial guide in determining functional roles for viral proteins . Virion proteins are commonly termed “structural proteins” and broadly-speaking , include proteins that either form an integral part of the virus architecture or are required for the first round of genome replication . This view of a virion being a minimal package of genome and essential viral proteins is now being challenged due to enhanced proteomics techniques and the availability of annotated genomic sequences for several mammalian species . These advances have extended proteomic analyses of virions to include host proteins that may be packaged into the virus particle along with the viral components . Enveloped viruses in particular have the capability of incorporating numerous host proteins , both into the interior of the virus particle as well as into the lipid envelope [1] , [2] . Several proteomic studies on herpesviruses have been undertaken , the majority of which focused on correctly identifying the viral constituents of the virion but many also reported finding cellular proteins [3]–[9] . Similarly , host proteins have been detected in vaccinia virions [10] . For RNA viruses , extensive proteomic analysis has been performed on human immunodeficiency virus type 1 ( HIV-1 ) and Moloney murine leukemia virus ( MoMLV ) vector particles , and they too have been found to incorporate numerous cellular proteins [11]–[13] . For the most part the functional significance of these packaged host proteins has not yet been determined but some proteins are known to interact specifically with a viral protein and this has enabled the significance of their incorporation to be studied in more depth . These include Tsg101 , cyclophilin A and APOBEC3G , all of which are packaged into HIV-1 virions [11] , [12] , [14]–[17] . Tsg101 plays a crucial role in virus assembly [14] , [18] , cyclophilin A modulates HIV-1 infectivity [19] and APOBEC3G is an anti-viral factor that promotes hypermutation of the viral genome [20] . These three proteins alone have significantly added to the understanding of how HIV-1 interacts with its host and they serve as an example of what can be learned from studying virion-associated host proteins . Although there are descriptions of interactions between certain cellular proteins and individual influenza virus proteins , for the most part this has not been done in a comprehensive manner and comparatively little is known about the requirement for host cell factors during the different stages of the influenza virus life cycle . In an effort to discover host factors involved particularly in genome replication , proteomic analyses of native influenza virus ribonucleoprotein and polymerase complexes have been performed which resulted in the identification of 45 interacting cellular proteins [21] . It is anticipated that cellular proteins found within the influenza virus particle may provide clues as to the virus assembly pathway and also early events that govern virus infectivity . Of the eleven influenza A virus encoded proteins , nine have been identified in the virion [22] . The exceptions being NS1 and PB1-F2 , the latter of which is not encoded by all influenza A viruses . The glycoproteins hemagglutinin ( HA ) and neuraminidase ( NA ) are embedded into the lipid envelope of the virus particle and form the characteristic spikes visible under the electron microscope [23]–[25] . Another membrane protein , the ion channel protein M2 is also found within the virion but at significantly lower levels than HA or NA [26] . The matrix protein M1 lies beneath the viral membrane and surrounds the eight ribonucleoprotein ( RNP ) segments , which consist of viral RNA coated with the nucleoprotein ( NP ) and bound by the trimeric polymerase complex ( PB1 , PB2 , PA ) [25] , [27] . Finally the nuclear export protein ( NEP ) is also found within influenza virions [28] . The majority of these proteins were identified on the basis of size by polyacrylamide gel electrophoresis but because detection of proteins by this method is restricted to more highly abundant proteins , the presence of M2 and NEP proteins in the influenza virion was only discovered much later using specific antibodies [26] , [28] . Any cellular proteins that may be incorporated into viral particles are also likely to be present at very low levels and while antibody-mediated detection is extremely sensitive , it is not practical when analyzing complexes of unknown composition . Mass spectrometry of tryptic peptides combined with database searching for identification is now the preferred method for such proteomic studies . In this report we utilize two complementary mass spectrometry techniques to analyze the protein content of purified influenza virus particles and specifically , to identify incorporated cellular proteins . Our analysis resulted in the identification of 9 virus-encoded proteins and 36 host-encoded proteins .
Virion proteomic analysis requires a highly purified preparation of virus and the choice of host cell used for virus growth is also an important consideration . While MDCK ( Madin Darby canine kidney ) cells are the preferred cell line for growth of influenza virus in tissue culture , the dog genome is not yet fully annotated and this would restrict the identification of cellular proteins . For the same reason , virus grown in embryonated chicken eggs was also not the best option . As a compromise between cells that would support high levels of virus growth and cells that could be used to search the most extensive protein database ( i . e . human ) , Vero ( African green monkey kidney ) cells were selected as the host cell line . There are a growing number of non-human primate sequences in the NCBI database and because of significant homology between primate and human proteins the human protein database could be used to identify incorporated host proteins . For later comparison , smaller amounts of virus were also purified from infected A549 ( human carcinoma lung epithelial ) cells . Supernatant collected from Vero cells infected with influenza A/WSN/33 virus was first clarified and the virus was concentrated through a sucrose cushion before being purified over a 30–60% sucrose gradient . The purity of the virus preparation was assessed by electron microscopy following negative staining ( Fig . 1A ) . Both intact influenza virions and partially disrupted virions were observed but importantly , there was no obvious contamination with cellular material . The proteins in the purified virus preparation were separated by SDS-PAGE and stained with Coomassie blue , and for identification of the viral glycoproteins , a deglycosylated sample was compared to an untreated sample ( Fig . 1B ) . All major viral proteins were visible . The three polymerase proteins resolved as two bands , both uncleaved ( HA0 ) and cleaved ( HA1 and HA2 ) forms of HA were present , as were bands consistent with the molecular weights for NP , NA and M1 . There were also some much fainter bands visible that may represent cellular proteins . The ability to fractionate protein samples to enhance the dynamic range of detectable proteins is a key issue when identifying the components of a protein complex by mass spectrometry . For this study two complementary techniques were used , one of which is based on separation of proteins and the other on separation of peptides . For the first method , both glycosylated and deglycosylated virus preparations were separated by SDS-PAGE on an 8–16% gradient gel ( Fig . 1B ) . Deglycosylation is required for several reasons: Firstly , because trypsin does not always efficiently digest highly glycosylated proteins and , secondly , because unmodified peptides generally have higher electrospray ionization efficiencies than their glycosylated counterparts . Finally , because deglycosylation produces a more uniform set of peptides from a potentially diverse number of glycoprotein isoforms , the sensitivity is increased . That said , in this study we did not find that deglycosylation increased the number of proteins identified ( see Tables S1 and S2 for a comparison ) and therefore the reported identifications from the two approaches were combined . Following Coomassie blue staining , each lane was cut into successive slices from top to bottom and the individual slices were subjected to in-gel trypsin digestion . This procedure was repeated on a 20% gel and gel slices less than 25 kDa were excised , so as to maximize the chances of detecting small molecular weight proteins . The peptides in each gel slice were then analyzed by liquid chromatography tandem mass spectrometry ( LC-MS/MS ) and the resulting fragment ion spectra were searched against protein databases for identification . The second method employed in this study was multidimensional protein identification technology ( MudPIT ) . A deglycosylated purified virus preparation was digested with trypsin en masse and the peptides in the mixture were separated by two dimensional chromatography , first on the basis of charge and then on hydrophobicity . The second chromatography separation step was directly coupled to the mass spectrometer detector and the resulting spectra were searched against the database for protein identification . The disadvantage of MudPIT is that there is no information on the size of the proteins which is useful for confirmation of protein identity . However , the method allows for the detection of low abundance proteins and extremely small molecular weight proteins that are often lost during gel-separation or gel-extraction steps . All nine virus-encoded proteins previously described to be in the influenza virion were identified by both MS methods ( Table 1 ) . These are PB1 , PB2 , PA , HA , NP , NA , M1 , M2 and NEP . Peptides from NS1 or PB1-F2 were not detected . Table 1 lists the predicted mass of each protein , the gel slice in which it was detected , the number of observed peptides and the percent sequence coverage of the protein . The statistical score associated with the match is also noted . MASCOT scores are used for the SDS-PAGE and LC-MS/MS analysis , while protein prophet scores are used for the MudPIT analysis . The HA , NP , NA and M1 proteins were all found in multiple gel slices . For HA , this was expected due to the presence of uncleaved HA0 as well as the cleaved sub-units HA1 and HA2 . However for both HA and particularly NP and M1 , the proteins appear to be distributed over a wider-than-expected size range . This perhaps reflects the fact that they are predicted to be the most abundant proteins in the influenza virion [27] and these amounts may exceed the resolving capacity of the gel , causing them to smear . From their predicted size , PB1 and PB2 are expected to migrate together , however in fact we found that PB1 migrates slower than PB2 , which resolves together with PA . This is in agreement with the first mapping data for the assignment of protein products to RNA segments [29] but the reason for the different migration patterns of PB1 and PB2 is still not known . Generally , the sequence coverage for each protein , which represents the number of unique peptides identified , was greater with the gel-fractionation and LC-MS/MS analysis . The exceptions are HA and NA , where greater sequence coverage was obtained with the MudPIT analysis . In total , we identified 36 cellular proteins in the purified influenza virus preparation . Seventeen of these were identified by both MS methods ( Table 2 ) , another 13 were identified only with the MudPIT analysis ( Table 3 ) and 6 were identified only with the gel-fractionation and LC-MS/MS analysis ( Table 4 ) . Each table indicates the protein name , its predicted mass , the gel slice in which it was found ( where relevant ) , the number of observed peptides , the score associated with the match and the percent sequence coverage . In addition , the predicted cellular localization of the protein is shown along with its abundance at the transcript level . Abundance in the kidney is noted because of the use of Vero cells , while abundance in the lung is more biologically relevant for influenza virus . The final column lists other viruses that have been reported to incorporate the observed cellular protein into their virions . As with the viral proteins , comparison between the two MS methods reveals greater sequence coverage obtained with the gel-separated proteins , however in total more proteins were identified with the MudPIT analysis . Both cytoplasmic and membrane-bound proteins were identified and while several of these proteins are highly abundant according to their NCBI UniGene EST profiles , most do not fall into this category and are present at moderate or low abundance in the cell . It is also striking that the majority of the proteins , particularly those in Table 2 have been reported in other virus particles and that many proteins are related or can be grouped together in functional categories such as cytoskeletal components , glycolytic enzymes and annexins . Following identification of the cellular proteins by proteomic methods , their presence in the purified influenza virus preparation was verified by immunoblot analysis which provides the highest degree of specificity . Influenza virus preparations purified from both Vero and A549 cells were analyzed for the presence of HA , beta-actin , annexin A5 and cyclophilin A ( Fig . 2 ) . Extracts from uninfected Vero and A549 cells were included as a control for the reactivity of the antibodies and size of the cellular protein . Influenza virus purified from both cell lines showed the presence of these three cellular proteins , confirming that they are associated with the virus and that this can be demonstrated in virus grown in two different cell types . When analyzing the results of virion proteomic studies , the challenge is to prove that the cellular proteins are really an integral part of the virion and that they are not just attached non-specifically to the outside or are perhaps derived from a microvesicle or exosome that co-purified with the virus . To address this question , we used the subtilisin protease protection assay which has been shown to efficiently remove microvesicles from HIV-1 virion preparations [30] , [31] . Protease treatment of the purified virus preparation strips proteins off the outside of virus particles and off any contaminating microvesicles . In doing so , the microvesicles become lighter than the virions and therefore the virions can be isolated by density centrifugation . Proteins that are inside the virion are protected by the lipid envelope and therefore will remain after the protease treatment . This is illustrated by the presence and absence of NP and HA , respectively , after subtilisin treatment of influenza virions ( Fig . 3 ) . Immunoblot analysis of selected cellular proteins reveals that beta-actin , annexin A5 , tubulin , annexin A2 , cofilin , GAPDH and cyclophilin A are all still present following protease treatment and centrifugation ( Fig . 3 ) . This indicates that these proteins are inside the influenza virion , however it should be noted that these experiments do not absolutely exclude the possibility that some proteins may be derived from contaminants that were not efficiently removed by the protease treatment . In contrast , CD9 and CD59 are absent following treatment ( Fig . 3 ) . There are two possible interpretations of this finding: Firstly , their loss may be because they are associated with microvesicles rather than virions and secondly , these proteins may be exposed on the surface of the virion as is HA . Since CD9 and CD59 are both membrane-bound proteins found on cellular surfaces ( CD9 has two extracellular loops and CD59 is GPI-anchored ) , if they are incorporated into an influenza virus particle one would expect them to be in the viral envelope and thus sensitive to protease digestion . However , to further address the possibility that they are not part of the virion , we made use of an alternative gradient medium ( Optiprep ) which , unlike sucrose , maintains iso-osmotic conditions at high densities and is therefore particularly good at separating membranous organelles such as enveloped viruses and microvesicles . Influenza virus preparations were purified simultaneously over both sucrose and Optiprep gradients , which were then fractionated . Immunoblot analysis demonstrated that CD9 co-sediments precisely with influenza virus ( as detected by the presence of NP ) in both types of gradient ( Fig . 4 ) . We also examined the separation of MHC-I , which has been found in exosomes derived from a variety of cell types [32]–[34] but was not identified in the mass spectrometry analysis of purified influenza virus . In the sucrose gradient , the peak MHC-I staining overlaps partially with that of NP and CD9 but in the Optiprep gradient there is clear separation of MHC-I from virus and CD9 . This supports the idea that Optiprep allows for better separation and strongly suggests that CD9 is an integral part of the influenza virion . It should also be noted that despite partial co-purification of MHC-I in the sucrose gradient , this protein was not identified in the proteomic analysis , probably indicating that the level of sensitivity provided by these methods is not sufficient to detect very low levels of protein . To provide additional evidence that the membrane-bound cellular proteins identified by mass spectrometry are on the lipid envelope of influenza virus , immunogold labeling of Optiprep-purified influenza virions was performed . Virions were labeled with antibodies against HA , CD9 , CD81 ( Fig . 5 ) or CD59 ( data not shown ) and secondary gold antibodies , followed by negative staining . One or two gold particles located on the surface of a virion could be seen for CD9 , CD81 and CD59 . This was significantly less compared with the degree of HA labeling , however it is consistent with the fact that there is most likely far more HA present on the virions than there are molecules of CD9 , CD81 or CD59 .
The host cytoskeletal network is involved in the transport of viral components in the cell and particularly during the stages of virus entry and exit [46] , [47] . Several studies on RNA viruses have also indicated that cytoskeletal proteins such as tubulin and actin are required for viral gene expression [48]–[51] . For influenza virus , it has been shown that the virus requires an intact actin cytoskeleton for entry , specifically into polarized cells [52] and interactions between the cytoskeleton and lipid rafts has been proposed to facilitate budding of filamentous virus particles [53] . Furthermore , an association of M1 and NP with cytoskeletal elements has been reported [54] , [55] , and actin and tubulin were both identified as proteins that interact with influenza RNPs [21] . In the present study , protease treatment showed that actin , tubulin and cofilin ( which binds to actin ) were all present in the interior of influenza virions which most likely reflects their active participation in moving the viral components to the assembly site as well as cytoskeletal reorganization that occurs during bud formation . Other actin-binding proteins found to be associated with influenza virions are tropomyosin , annexin ( see below ) , WD repeat containing protein and destrin . Several annexin family members ( A1 , A2 , A4 , A5 and A11 ) were identified in influenza virus particles . Annexins are calcium-dependent phospholipid-binding proteins and are proposed to act as scaffolding proteins at certain membrane domains . Annexin A2 in particular has been shown to bind to actin and be involved in the assembly of actin at cellular membranes [56] . It is also required for the apical transport of vesicles in polarized cells and specifically vesicles that carry membrane raft-associated proteins [57] . This is intriguing since influenza virus also buds from raft domains at the apical surface of polarized cells . In fact , a role for annexin A2 in virus assembly has been proposed for HIV-1 [58] , and in HCMV , the presence of annexin A2 is thought to promote viral binding and fusion [59] . Interestingly , annexins A1 and A5 , which both interact with A2 , have the opposite effect of preventing fusion , perhaps indicating a potential regulatory role [60] . The calcium-binding protein S100A11 which is known to interact with annexin A1 [61] was also identified in the influenza virion , suggesting that they may be packaged as a complex . Two members of the tetraspanin family , CD9 and CD81 , were found to be associated with influenza virions and are most likely inserted into the viral envelope . Tetraspanins have four transmembrane domains and two extracellular loops and are involved in both homo- and heterotypic interactions in specialized membrane domains referred to as tetraspanin-enriched microdomains ( TEMs ) [62] . Despite some similarities to lipid rafts , proteomic analyses of TEMs and lipid rafts have shown that they have distinct compositions [63] , although they may interact with each other under certain conditions . Several tetraspanins have been reported to play a role during viral infections . Of these , CD81 is the best characterized in terms of its function as a co-receptor for hepatitis C virus [64] , [65] . Tetraspanins , including CD9 and CD81 , have also been implicated in both fusion and egress pathways for a number of viruses such as HIV-1 , feline immunodeficiency virus and canine distemper virus [66]–[70] . One such study also reported that in contrast to HIV , influenza virus does not assemble at domains rich in tetraspanins and does not incorporate either CD9 or CD63 into virus particles [71] . This finding is obviously contradictory to the present proteomic analysis of influenza virions in which CD9 was detected by mass spectrometry , immunoblot analysis and immunogold labeling of virions . The reason for the discrepancy may be technical as Khurana et al . [71] used HeLa cells to propagate the virus and detected incorporated proteins by immunofluorescent staining of concentrated virions . Integrin beta-1 was also identified in influenza virus particles and as integrins are well-characterized tetraspanin binding partners , it was possibly incorporated together with CD9 or CD81 . Cyclophilin A ( CypA ) , which was shown to be in the core of the influenza virion , is a peptidyl-prolyl isomerase and has been reported to be present in the virions of a number of different viruses . For HIV-1 , the specific incorporation of CypA is mediated by an interaction with the capsid portion of the Gag protein [15] , [16] . There is an abundant amount of literature concerning the requirement of CypA for HIV-1 infectivity but as it turns out , it is the CypA in the target cell that is more critical and therefore the precise role of the virion CypA is currently unclear [72] , [73] . Within the target cell , CypA is proposed to facilitate a conformational change in the capsid which enables the virus to evade detection by the host immune response [74] , [75] . CypA is incorporated into vesicular stomatitis virus ( VSV ) presumably via the described interaction with the nucleocapsid protein [76] . It has also been shown to be required for VSV replication , however this activity is serotype-specific [76] . A strong interaction between CypA and SARS coronavirus nucleocapsid protein has also been reported [77] and CypA relocalizes to sites of viral replication in vaccinia virus infected cells [78] . Another member of the cyclophilin family , cyclophilin B , is required for hepatitis C virus replication and acts by interacting with the viral polymerase and increasing its RNA binding activity [79] . Therefore , there is a strong precedent for the involvement of cyclophilin proteins in virus replication . CD59 is a complement regulatory protein that acts by inhibiting formation of the membrane attack complex ( MAC ) . It is a GPI-anchored protein and the experimental data confirm that it is associated with the influenza virus envelope . Enveloped viruses are susceptible to direct complement-mediated lysis by MAC and as a form of protection HIV-1 , vaccinia virus ( VV ) , human T cell lymphotropic virus ( HTLV ) and human cytomegalovirus ( HCMV ) all incorporate CD59 and other regulatory proteins such as DAF and CD46 into their lipid envelopes ( the latter two were not identified in influenza virions ) [80]–[82] . Complement control proteins are highly species specific and are only active against homologous complement . This has important implications for virus host-range as the virus produced and transmitted between one host species would be protected by incorporation of CD59/DAF/CD46 , however , virus transmitted to another host species would become susceptible to lysis by the complement system of that host . When one looks at the list of proteins associated with influenza virions , at first glance it is difficult to see an obvious role for some of these proteins in the virus life cycle . However , it is possible that some of these cellular proteins have functions other than their described major roles . For example , a number of proteins involved in the glycolytic pathway were identified ( pyruvate kinase , enolase 1 , GAPDH , phosphoglycerate kinase ) . Both enolase and phosphoglycerate kinase , in addition to tubulin , have been reported to stimulate transcription of the Sendai virus genome [83] , but it is unclear whether their glycolytic activities are required or whether this is an example of an alternative function for these proteins [84] . A role in RNA virus transcription has also been proposed for GAPDH . Phosphorylated forms of GAPDH have been shown to bind to the genomic cis-acting RNA of human parainfluenza virus type 3 ( hPIV3 ) and are also present in purified virions [85] , [86] . In vitro data indicate that GAPDH serves a negative regulatory role in hPIV3 transcription and that this is dependent on its phosphorylation [85] . Compared with the cellular proteins found to associate with the influenza RNP complex , the only ones also identified in influenza virions are alpha and beta tubulins , beta actin and ubiquitin carboxyl-terminal hydrolase [21] . This may indicate that these proteins are packaged with RNPs and that they interact with one of the RNP components i . e . NP , one of the polymerase proteins or the genomic RNA . The fact that there are not more proteins in common is probably because each viral protein associates with many different cellular proteins during the course of the viral life cycle and these interactions in most cases are transient . The proteins identified in this and other studies represent a snapshot of a particular point in the life cycle , but importantly they provide a foundation for further analysis of cellular requirements for influenza virus infection . Packaged cellular proteins have a unique importance as the virus literally transports them from one cell to the next . This is an ingenious way of ensuring that host cell activities required at or immediately after entry are instantly accessible to the virus . For viruses that can infect multiple species such as influenza virus , any host protein that is required for infection must be active in both species to allow for transmission to occur . Therefore , as discussed above for CD59 , virion-associated host proteins can be one of the determinants of virus host range due to their species-specific activity . It will also be interesting to compare the identity and abundance of host proteins in influenza viruses that produce virions with a filamentous morphology . One would assume that the increased volume and surface area of these particles would allow for greater levels of host protein incorporation but whether or not there is increased diversity may depend on specific versus non-specific incorporation . The presence of host proteins in influenza virions , whether they are incorporated specifically or non-specifically , could also be a concern for vaccine manufacturers as the vaccine is delivering more than just viral antigens . Although the relative amount of cellular protein compared to viral protein in the virus particle is expected to be extremely small , the choice of host cell for propagation of vaccines could be an important consideration , particularly for live-attenuated virus vaccines . Currently , all influenza vaccines are produced in embryonated chicken eggs but there is a move afoot to transition to cell culture systems , with Vero cells being one of the approved cell lines [87] , [88] . During the manufacturing process great care is taken to avoid the use of animal-derived products such as serum but the incorporation of non-human primate proteins into the vaccine virus will be unavoidable . Precise quantitation of these non-viral components will help to assess whether the levels present in each vaccine dose are high enough to risk inducing an allergic response .
Vero and A549 cells were maintained in Dulbecco's modified Eagle medium ( Gibco , San Diego , California ) supplemented with 10% fetal bovine serum ( HyClone , South Logan , Utah ) and Madin-Darby canine kidney ( MDCK ) cells were maintained in minimal essential medium supplemented with 10% fetal bovine serum . Influenza A/WSN/33 virus was propagated in MDCK cells in Minimal Essential Medium ( Gibco ) supplemented with 0 . 3% bovine serum albumin ( Sigma , St . Louis , Missouri ) and 0 . 1% fetal bovine serum . Viral titers were determined by plaque assay on MDCK cells . Antibodies against actin ( A4700 ) , annexin A5 ( A8604 ) , cofilin ( C8736 ) and tubulin ( T0198 ) were obtained from Sigma ( St . Louis , Missouri ) . Monoclonal antibody against annexin A2 ( sc-28385 ) was obtained from Santa Cruz Biotechnology ( Santa Cruz , California ) . Monoclonal antibodies against CD9 ( sc-13118 and 555370 ) were obtained from Santa Cruz Biotechnology and BD Pharmingen ( San Diego , California ) , respectively . Monoclonal antibody against CD59 ( MCA1054GA ) was obtained from Serotec ( Oxford , U . K . ) and monoclonal antibody against CD81 ( 555675 ) was obtained from BD Pharmingen . Rabbit polyclonal antibody against cyclophilin A ( SA-296 ) was obtained from Biomol ( Plymouth Meeting , Pennsylvania ) and monoclonal antibody against GAPDH ( RDI-TRK5G4-6C5 ) was obtained from RDI Research Diagnostics ( Concord , Massachusetts ) . Monoclonal antibodies against influenza virus NP ( HT103 ) and HA ( 2G9 ) were made by the Mount Sinai Hybridoma Center Shared Research Facility . The MHC-I antibody was kindly provided by Dr . Domenico Tortorella ( Mount Sinai School of Medicine , NY ) . Fifty 15cm dishes of 80% confluent Vero cells were infected with influenza A/WSN/33 virus at a multiplicity of 0 . 001 . At 65–70 hours post infection , the supernatant was harvested and clarified ( 2600×g , 5 min , 4°C , in a Sorvall RT6000D centrifuge ) . The clarified supernatant was layered over a 20% sucrose cushion in NTE buffer ( 100 mM NaCl , 10 mM Tris-Cl ( pH 7 . 4 ) , 1 mM EDTA ) and the virus concentrated by ultracentrifugation ( 112 , 600×g , 2 hrs , 4°C , in a SW28 rotor [Beckman Coulter , Fullerton , California] ) . The concentrated virus was purified over a 30–60% sucrose gradient ( 112 , 600×g , 3 hrs , 4°C ) and the banded virus collected , diluted with NTE buffer , pelleted ( 112 , 600×g , 90 min , 4°C ) and resuspended in approximately 1 ml of NTE buffer . Typical protein yields of 1–2 mg/ml were obtained . When using Optiprep medium ( Sigma , St . Louis , Missouri ) , a 10–30% gradient was made and fractions were taken from the top . Protein was precipitated from each fraction with 20% trichloroacetic acid ( TCA ) and subjected to western blot analysis . Purified virus equivalent to 100 ug of protein was denatured by heating at 100°C for 10 min in the presence of 0 . 5% SDS , 40 mM DTT and 1%NP40 . PNGase F ( New England Biolabs , Ipswich , Massachusetts ) was added in the presence of 50 mM sodium phosphate ( pH 7 . 5 ) and 1% NP40 and the reaction incubated at 37°C overnight . Purified virus equivalent to 50 ug of protein was incubated with 100ug of subtilisin protease ( Sigma , St . Louis , Missouri ) in 20 mM Tris-Cl ( pH 8 ) and 1 mM CaCl2 for 18 hours at 37°C . The treated virus was diluted to 1 ml with NTE buffer and 5ug of PMSF ( Sigma ) was added . The virus was concentrated through a 20% sucrose cushion by ultracentrifugation ( 222 , 030×g , 2 hr , 4°C in an SW41 rotor [Beckman Coulter , Fullerton , California] ) and then subjected to western blot analysis . Vero or A549 cells at 80% confluency were mock infected or infected with influenza A/WSN/33 virus at a multiplicity of 0 . 001 . At 65–70 hours post infection the cells were harvested and whole cell extracts were prepared by lysis in extract buffer ( 50 mM Tris [pH 7 . 5] , 280 mM NaCl , 1% Triton X-100 , 0 . 2 mM EDTA , 2 mM EGTA , 10% glycerol , 1 mM dithiothreitol , 0 . 1 mM sodium vanadate and protease inhibitors [Complete; Roche] ) on ice for 30 minutes . Extracts were centrifuged ( 15700×g , 15 min , 4°C in an Eppendorf 5415R microcentrifuge ) and the supernatants collected . Proteins from either purified virus ( 2 ug ) or whole cell extracts ( 10 ug ) were denatured by heating at 100°C for 10 min in 1× sodium dodecyl sulfate-polyacrylamide gel electrophoresis ( SDS-PAGE ) sample buffer and were then separated by SDS-PAGE . For western blot analysis the proteins were transferred to nitrocellulose membrane which was then probed with a specific primary antibody and a peroxidase-labeled secondary antibody . The blots were analyzed by chemiluminescence and exposed to x-ray film . For protein staining , gels were stained with SimplyBlue SafeStain ( Invitrogen , Carlsbad , California ) . Optiprep-purified virus was diluted 1∶20 with NTE buffer and adsorbed onto formvar/carbon-coated nickel grids ( Electron Microscopy Sciences , Hatfield , Pennsylvania ) . Following a 5 min wash with TBS buffer ( 50 mM Tris-Cl ( pH 7 . 5 ) , 150 mM NaCl ) , the sample was blocked with 3%BSA in TBS for 45 min . Primary antibody ( 10 ug/ml ) was diluted in 1%BSA/TBS and adsorbed to the grid for 1 hr at room temperature . Following three washes with TBS , secondary gold-conjugated antibody was added for 1 hr at room temperature . The grids were then washed twice with TBS , once with water and negatively stained with 1% sodium silicotungstate ( pH 7 ) for 15 sec . Images of stained virions were captured on a Hitachi H-7650 120 kV transmission electron microscope . For quantitation purposes , the number of virions and the number of gold particles were assessed in two representative images . These data were expressed as the number of gold particles per virion . Proteins separated in one dimensional polyacrylamide gels were cut sequentially and subjected to in situ tryptic digestion prior to mass spectrometric analysis . Digestion was performed robotically on the GE Healthcare Ettan Gel Digester in a 96 well plate . A 20 minute wash with 100 µl , 50 mM ammonium bicarbonate in 50% acetonitrile was followed with a 10 minute 75% acetonitrile wash . Gel bands were then air dried and 15 µl of 6 . 7 µg/µl sequencing grade trypsin ( Promega ) was added to each well . Digestion was carried out at 37°C for 16 hours . The protein digests were then analyzed using Waters/Micromass QTOF Ultima mass spectrometer equipped with a Waters CapLC liquid chromatography system . 10 µl of the digest supernatant was loaded into a capLC vial and 5 µl of the sample was directly injected onto a 100 µm i . d . ×150 mm long Atlantis C18 reversed phase column ( Waters ) running at 500 nl/min . Initial HPLC conditions were 95% buffer A and 5% buffer B with the following linear gradient: 3 min , 5% B; 43 min 37% B; 75 min 75% B; and 85 min 95% B . Buffer A consisted of 98% water , 2% acetonitrile , 0 . 1% acetic acid , and 0 . 01% TFA . Buffer B contained 80% acetonitrile , 20% water , 0 . 09% acetic acid , and 0 . 01% TFA . Data-dependent acquisition was performed so that the mass spectrometer switched automatically from MS to MS/MS modes when the total ion current increased above the 1 . 5 counts/second threshold set point . In order to obtain good fragmentation , a collision energy ramp was set for the different mass sizes and charge states , giving preference to double- and triple-charged species for fragmentation . All raw MS/MS spectral data were searched in-house using the MASCOT algorithm ( Matrixscience ) with the Mascot Distiller program utilized to generate Mascot compatible files . The Mascot Distiller program combines and centroids sequential MS/MS scans from profile data that have the same precursor ion . A charge state of +2 and +3 was preferentially located with a signal to noise ratio of 1 . 2 or greater and a peak list was generated for database searching . Using the Mascot database search algorithm , a protein was considered identified when Mascot listed it as a significant match/score ( p<0 . 05 ) with the proper enzymatic cleavage sites . Unlike the MudPIT analysis ( see below ) , the Peptide/Protein Prophet ( Institute for Systems Biology ) scoring system was not used here because this would have required either combining the data from all gel slices or treating each gel slice as an individual Peptide/Protein Prophet model . Combining the gel slices may allow for an effective PeptideProphet expectation maximization model to be built but would create false protein identifications in that a protein probability could be based on peptides present in separate bands on the gel . Applying Peptide/Protein Prophet to individual gel slices would result in a collection of small datasets ( 50–100 MS/MS queries ) that cannot be modeled accurately as there are not sufficient datapoints for the expectation maximization algorithm to assign correct versus incorrect peptides . The NCBInr protein database was chosen over other genome specific databases to allow a wider search match found based on homology to other species . Parameters used for searching were partial methionine oxidation and acrylamide modified cysteine , a peptide tolerance of ±0 . 6 Da , and a MS/MS fragment tolerance of ±0 . 4 Da . 100 µg of deglycosylated purified virus preparation was solubilized in 8 M Urea , 0 . 4 M NH4HCO3 ( pH 8 . 0 ) , reduced with 45 mM DTT , and alkylated with 100 mM iodoacetamide . Tryptic digestion was performed using a 1∶50 enzyme to substrate ratio at 37 degrees C for 18–24 hours ( sequencing grade trypsin , Promega ) . After digestion , off-line strong cation exchange chromatography ( SCX ) was performed on an Applied Biosystems Vision Workstation using a 2 . 1 mm×200 mm PolySulfoethyl A column , equilibrated with Buffer A ( 10 mM KH2PO4 , 25% Acetonitrile , pH = 3 . 0 ) . Peptides were separated into fractions using a 90 min linear salt gradient from 0–98% Buffer B ( 10 mM KH2PO4 , 25% Acetonitrile , 1 M KCl , pH = 3 . 0 ) All 22 collected fractions from the SCX chromatography were dried and reconstituted with 15 µl of 0 . 1 % TFA . A 5 µl aliquot of each of the samples was injected and desalted on a reversed phase C18 trap column ( Waters , Symmetry , Nanoease 0 . 180 mm i . d . ×23 . 5 mm , 5 micron ) and was separated on a C18 analytical column ( Waters , Atlantis , Nanoease 0 . 1 mm i . d . ×150 mm , 3 micron , 100 Å ) using the Dionex Ultimate chromatography system . On-line MS analysis was performed on the ABI QSTAR XL system . MS data was surveyed for 0 . 5 s , and MS/MS acquisition was performed on three highest peptide peaks . Each of the QSTAR XL mass spectrometer spectra files was processed with MASCOT Distiller version 2 . 1 and the resulting peak lists were database searched using MASCOT Server 2 . 1 . The search parameters included static carbamidomethyl modifications for cysteine and variable oxidation modifications for methionine amino acid residues . Data analysis on the resulting LC/MS and MS/MS datasets is accomplished using a dual processor Dell 650 Workstation . The search results for each fraction were analyzed using the NCBInr database . After MASCOT analysis , Peptide and ProteinProphet ( Institute for Systems Biology ) analysis was performed using the Trans-Proteomic Pipeline version 2 . 9 GALE rev . 1 , Build 200607201423 . Peptide and Protein Prophet computes the probabilities for both individually searched peptides and the resulting proteins . The 95% Protein Prophet probability cutoff corresponds to a 0 . 6% false positive error rate . Finally , TPP identifications are submitted to Yale Proteomics Expression Database ( YPED ) web site [89] for further user analysis . All data are publicly available through http://yped . med . yale . edu/repository . | Viruses are released from infected cells in the form of virions , which contain all the essential factors necessary for initiating infection in a new target cell . For influenza virus , it is known that virions contain the viral genome , a lipid envelope , and at least nine viral proteins . We performed a detailed proteomic analysis of purified influenza virus particles using mass spectrometry and database searching for protein identification , and in addition to the nine viral proteins , we identified 36 host proteins . These host proteins are present both inside the influenza virus particle and on the viral envelope . All viruses require host cell factors to complete their replication cycles , and they also have to contend with the antiviral defense mechanisms of the host . Virus–host interactions may therefore provide the key to understanding viral pathogenesis and may also present us with new targets for the design of antiviral drugs . For influenza virus , information on the requirement of cellular factors is limited , but the description of these 36 host proteins that are packaged into the virion provides a foundation for further analysis into the involvement of these cellular pathways in the influenza virus life cycle . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"virology/virion",
"structure,",
"assembly,",
"and",
"egress",
"virology/host",
"invasion",
"and",
"cell",
"entry",
"virology"
] | 2008 | Cellular Proteins in Influenza Virus Particles |
In most mammals , including mice and humans , meiotic recombination is determined by the meiosis specific histone methytransferase PRDM9 , which binds to specific DNA sequences and trimethylates histone 3 at lysine-4 and lysine-36 at the adjacent nucleosomes . These actions ensure successful DNA double strand break formation and repair that occur on the proteinaceous structure forming the chromosome axis . The process of hotspot association with the axis after their activation by PRDM9 is poorly understood . Previously , we and others have identified CXXC1 , an ortholog of S . cerevisiae Spp1 in mammals , as a PRDM9 interactor . In yeast , Spp1 is a histone methyl reader that links H3K4me3 sites with the recombination machinery , promoting DSB formation . Here , we investigated whether CXXC1 has a similar function in mouse meiosis . We created two Cxxc1 conditional knockout mouse models to deplete CXXC1 generally in germ cells , and before the onset of meiosis . Surprisingly , male knockout mice were fertile , and the loss of CXXC1 in spermatocytes had no effect on PRDM9 hotspot trimethylation , double strand break formation or repair . Our results demonstrate that CXXC1 is not an essential link between PRDM9-activated recombination hotspot sites and DSB machinery and that the hotspot recognition pathway in mouse is independent of CXXC1 .
Meiotic recombination ensures production of fertile gametes with a correct haploid chromosome number and genetic diversity [1] . In most of mammals , the meiotic recombination sites are restricted to 1–2 kb regions , termed recombination hotspots , locations of which are determined by the DNA binding histone methyltransferase PRDM9 [2–4] . Recombination initiates when PRDM9 binds to hotspots sequences with its zinc finger domain and trimethylates histone 3 at lysine 4 ( H3K4me3 ) and lysine 36 ( H3K36me3 ) resulting in formation of a nucleosome-depleted region [2–6] , probably by the action of nucleosome motors [7] . DNA double strand breaks ( DSB ) are created at the nucleosome-depleted regions of activated hotspots [8–11] , and eventually repaired as either crossovers or non-crossover conversions . Cytological staining for several proteins associated with DSB processing in early meiotic prophase show that , from the earliest time of their detection , DSB are associated with a proteinaceous structure known as chromosome axis [12–15] . We have previously shown that PRMD9 is associated , but not directly interacting , with chromosome axis elements such as phosphorylated REC8 ( pREC8 ) and SYCP3 in spermatocytes [16] . Efficient H3K4 trimethylation at hotspots is crucial for normal DSB formation and repair that occurs on the chromosome axis [17] , even though PRDM9 binding presumably occurs on the open chromatin loops [18] . However , we currently do not have detailed knowledge of the proteins and molecular mechanisms participating in hotspot association with the chromosome axis . In Saccharomyces cerevisiae , which has no PRDM9 , the PHD zinc finger protein Spp1 , a member of the COMPASS complex ( Complex associated with Set1 , the protein catalyzing trimethylation of histone 3 at lysine-4 ) , acts as a histone H3K4 methyl reader and promotes meiotic DSB formation at the existing H3K4me3 sites , such as promoters [19 , 20] . Spp1 is predominantly located on the chromosome axes and connects H3K4me3 sites with the axis protein Mer2 to stimulate Spo11 dependent DSB formation [19 , 20] . Recent study showed that Spp1 function in tethering DSB sites to chromosome axes and ensuring efficient DSB formation is independent of its function as a COMPASS complex member [21] . CxxC finger protein 1 ( CXXC1 , also known as CFP1 and CGBP ) is an ortholog of S . cerevisiae Spp1 in mammals [21] . In somatic cells , CXXC1 binds to both unmethylated CpGs and SETD1 , which is required for trimethylation of H3K4 at CpG islands [22] . CXXC1 is crucial for embryonic stem cell maintaining and development [23 , 24] . Knocking out Cxxc1 in mice results in lethality at the early embryonic stages [25] . We have reported that CXXC1 interacts with PRDM9 in yeast two-hybrid assay and in vitro [16] . This interaction has recently been confirmed by another group , which also reported that CXXC1 interacts with the chromosome axis element IHO1 by yeast two-hybrid assay [26] . IHO1 is considered to be the ortholog of yeast Mer2 and is known to be essential for ensuring efficient DSB formation [27] , therefore it is possible that CXXC1-IHO1 interaction serves the same function in mammalian meiosis as their orthologs Spp1-Mer2 in yeast . However , the function of CXXC1 in mammalian meiosis has not been characterized so far . It has been unclear whether CXXC1 binds to PRDM9 in germ cells and whether it participates in meiotic recombination initiation in any way , either as a partner of PRDM9 or as a methyl reader of H3K4me3/H3K36me3 marks that PRDM9 imposes at the nucleosomes surrounding the recombination hotspots . Here we confirmed that CXXC1 is co-expressed with PRDM9 and indeed interacts with it in spermatocytes . To address whether and how CXXC1 functions in meiotic recombination , we created two Cxxc1 conditional knockout mouse models and deleted Cxxc1 in all germ cells and in late spermatogonia just before the onset of meiosis . In both models , loss of CXXC1 did not affect normal meiotic recombination process . Our study demonstrates that the presence of CXXC1 in mouse meiosis is not essential , and unlike its S . cerevisiae ortholog Spp1 , CXXC1 does not appear to be a key factor for the DSB formation .
We tested whether CXXC1 interacts with PRDM9 in vivo by co-immunoprecipitation ( co-IP ) from spermatocytes isolated from 14 dpp B6 testis using antibody against PRDM9 [16] . We found that CXXC1 indeed interacts with PRDM9 in spermatocytes ( Fig 1A ) . However , the interaction was not as strong as with PRDM9’s predominant interactor EWSR1 ( Fig 1A ) [16] , which raised the possibility that the interaction between CXXC1 and PRDM9 could be mediated by stronger PRDM9 interactors . To test whether this is the case , we performed co-IP with EWSR1 and did not detect any interaction with CXXC1 in testicular extract ( Fig 1B ) . To further test the interactions between the three proteins , we co-expressed Myc-tagged mouse CXXC1 , HA-tagged EWSR1 and Flag-tagged PRDM9 proteins in human embryonal kidney 293 ( HEK293 ) cells , and performed co-IP with antibodies against HA or Myc tags . Both EWSR1 and CXXC1 immunoprecipitated PRDM9 under these conditions , but there was no interaction between CXXC1 and EWSR1 in the presence or absence of PRDM9 ( Fig 1C ) . Several reports have shown that Spp1 , the yeast ortholog of CXXC1 , binds to H3K4me3 and tethers H3K4-trimethylated recombination hotspots to the chromosome axis [19–21] . To test whether CXXC1 binds to H3K4me3 in mouse spermatocytes , we performed CXXC1 co-IP from 14-dpp B6 testicular extract . Indeed , we detected CXXC1 interaction with H3K4me3 but not with the closed chromatin mark H3K9me3 ( Fig 1D ) . These results indicate that CXXC1 and EWSR1 form separate complexes with PRDM9 . They also indicate that although CXXC1 interacts with PRDM9 in vivo , it is not a predominant interactor of PRDM9 , and that their interaction could be mediated by other proteins such as histone 3 trimethylated at lysine 4 . In seminiferous tubules of mouse testis , CXXC1 is expressed in both germ cells and Sertoli cells ( Fig 2A , top panel ) . CXXC1 showed high expression in spermatogonia , low expression in leptonema and zygonema , and then again high expression in pachynema and diplonema , decreasing to undetectable levels in spermatids ( Fig 2A , top panel ) . Previous reports showed that PRDM9 is present only in leptonema and zygonema during meiosis [28] . Double staining for PRDM9 and CXXC1 showed co-expression of these two proteins in nuclei from stage X seminiferous tubules ( Fig 2A , top panel ) and in 14-dpp ( days post partum ) testis ( Fig 2A , middle panel ) , when the majority of spermatocytes are at leptotene and zygotene stages . Since CXXC1 interacts with PRDM9 in vivo ( Fig 1A ) , we performed CXXC1 staining in Prdm9 knockout mouse testis ( Prdm9-/- ) to determine whether CXXC1 localization could be affected by the absence of PRDM9 . In this mutant , CXXC1 showed the same localization pattern as in controls ( Fig 2A , bottom panel ) . The pattern of PRDM9 and CXXC1 in leptonema and zygonema was further confirmed by chromosome spreads , where CXXC1 showed diffused signal over the entire nuclear region from leptonema through diplonema ( Fig 2B ) . These results show that CXXC1 and PRDM9 are both present in leptonema and zygonema nuclei , and that CXXC1 expression and localization are not affected by the presence or absence of PRDM9 . To test whether CXXC1 is involved in spermatogenesis , we generated a conditional knockout ( CKO ) model using CRISPR/Cas9 to insert loxP sites flanking exon 2 and 3 of Cxxc1 in C57BL/6J . The strategy of obtaining the CKO mutants is shown on S1 Fig . We bred late spermatogonia-specific knockout mice ( Cxxc1loxP/Δ;Stra8-iCre , hereafter Cxxc1 CKO ) by crossing the Cxxc1loxP/loxP mice with Stra8-iCre mice [29] and germ cell-specific knockout mice ( Cxxc1loxP/Δ;Ddx4-Cre , hereafter Cxxc1 CKODdx4-Cre ) with Ddx4-Cre mice [30] . Western blot confirmed that in knockout testes , the protein level of CXXC1 is reduced ( Fig 3A ) . CXXC1 was absent in spermatocytes of the CKO , but present in spermatogonia and Sertoli cells in CKO testes ( Fig 3B , short and long arrows ) and only in Sertoli cells in Cxxc1 CKODdx4-Cre ( S2A Fig , long arrows ) . We performed fertility test with two Cxxc1-deleted CKO mice of each model . To our surprise , in both models the male mice were fertile and produced similar number of viable progeny compared to the heterozygous ( het ) and B6 controls ( Figs 3C and S2B ) . Testis index ( testis weight/body weight ) was the same in CKO as in het and wild type B6 controls ( Fig 3D ) . Histology of testis and epididymis from CKO mice showed no detectable spermatogenesis defects ( Figs 3E and S2C ) . No increased apoptosis in germ cells was detected using TUNEL assay ( Figs 3F and S2D ) . In contrast , Cxxc1 germ cell-specific knockout female mice with Ddx4-Cre ( reduced protein level shown in S1E Fig ) were sterile—no viable pups were produced from homozygous knockout Cxxc1 CKODdx4-Cre mating test , while the heterozygous control ( Cxxc1loxP/+;Ddx4-Cre ) mating produced normal number of pups ( 5 . 3 ± 1 . 7 ) . However , the histology of CKO ovaries from 21 dpp and adult female mice both showed normal ovary morphology and follicle formation ( Figs 3E and S2F ) . Thus , their sterility is most likely due to early embryonic developmental deficiency caused by the lack of maternal genome activation at zygotic stage as reported before [31] and not meiotic defects . We further tested whether CXXC1 affects the localization , the expression pattern , or the function of PRDM9 . Localization of PRDM9 in seminiferous tubules was preserved in CKO ( Figs 4A and S3A , right panel ) . In addition , the expression pattern of PRDM9 in leptonema and zygonema was not affected in the absence of CXXC1 ( Fig 4B ) . To test whether lack of CXXC1 affects PRDM9 methyltransferase function , we first compared H3K4me3 patterns in control and Cxxc1 CKO mice . Both control and CKO chromosome spreads showed abundant H3K4me3 signal in leptonema and zygonema , lower signal in pachynema and increased signal in diplonema ( Fig 4C ) . In addition , the H3K4me3 staining on cross sections of CKO testis showed no decrease ( S3B Fig ) . These data indicate that the hotspot trimethylation and transcriptional activation in spermatocytes are not affected by the loss of CXXC1 . Second , we tested whether loss of CXXC1 affects PRDM9 binding and its methytransferase activity at individual hotspots by H3K4me3 ChIP-qPCR . We found that H3K4me3 enrichment at hotspots PbxI and Fcgr4 , which are regulated by Prdm9Dom2 , the Prdm9 allele present in B6 mice , was not different in B6 control , Cxxc1 heterozygous and CKO . We measured as a control the H3K4me3 enrichment at promoter regions of the housekeeping gene Actinb and the meiosis specific gene Sycp3 , which are not PRDM9-dependent . These were not changed as well ( Fig 4D ) . These results suggest that loss of CXXC1 does not affect PRDM9 expression , binding to hotspots , or its catalytic function . Therefore , CXXC1 is not required for PRDM9-dependent hotspot activation . To test whether lack of CXXC1 affects DSB formation process , we next determined the number , position and activities of DSBs in the Cxxc1 CKO . During DSB formation , the single stranded DNA tail is initially coated by the replication protein A ( RPA ) , and then RPA is gradually replaced by the RecA family members RAD51 and DMC1 [32–35] . When measured by the number of foci of DMC1 ( Figs 5A and S4A ) , RAD51 ( Figs 5B and S4B ) and RPA ( Figs 5C and S4C ) in early and late zygonema , these numbers in Cxxc1 CKO were not statistically different from control samples ( Fig 5A , 5B and 5C , lower panels ) . Also , we did not detect increased DMC1 , RAD51 or RPA foci in CKO pachynema ( Figs 5A , 5B and 5C and S4A , S4B and S4C ) . These data indicate that the number of DSB per meiosis and DSB repair process is not affected in the loss of CXXC1 . To determine whether the locations of DSB sites in Cxxc1 CKO are affected , we performed ChIP-seq for DMC1 [8 , 9] ( S5A Fig ) . We detected 8 , 233 DMC1 peaks in control spermatocytes and 8 , 569 DMC1 peaks in CKO spermatocytes , in which 7 , 501 peaks are shared in heterozygous control and CKO samples ( Fig 6A ) . We plotted the frequency distribution of DMC1 activity of the 732 unique peaks from control ( Fig 6B , left ) or 1068 CKO unique peaks ( Fig 6B , right ) , and found that these virtually unique peaks were not unique , but had low DMC1 activity which prevented them from being detected by the peak calling algorithm . We found that 91 . 8% of shared DMC1 peaks ( 6 , 886 peaks ) , 78 . 8% of control unique peaks ( 577 peaks ) and 84 . 3% of CKO unique peaks ( 900 peaks ) contain a PRDM9 binding site at their centers ( Fig 6C as an example ) . The aggregation plots also confirmed that DSBs contain PRDM9 binding motifs at their centers in both control and CKO spermatocytes ( Fig 6D ) . Also , in the 615 shared , 155 het unique and 168 CKO unique DMC1 peaks which do not contain detectable PRDM9 binding motifs , only 18 , 5 and 5 peaks , respectively , overlapped with transcription start sites . Therefore , unlike in Prdm9 knockout mice [9] , promoter sites are not predominantly used for DSB formation in Cxxc1 CKO . The activity of DMC1 signal in control and CKO is highly correlated ( r = 0 . 98 , Fig 6D ) , indicating the activity of DSB formation is not affected in the CKO spermatocytes . Furthermore , the activity of default DSB sites , which do not contain PRDM9 binding motif in the center , only contributes to 0 . 30% and 0 . 24% of total DMC1 activity in control and CKO samples , respectively ( Figs 6E and S5B ) , similar as reported default DSB activity in male germ cells [36] . These data suggest that loss of CXXC1 does not affect DSB number or positions . Therefore , CXXC1 is not essential for PRDM9 dependent DSB initiation pathway . To further investigate whether the DSB repair process and chromosomal synapsis are impaired in CKO , we used staining for phosphorylated H2AX ( γH2AX ) , which marks unrepaired DNA lesions and sex body in pachynema , to test for the processing of recombination repair . The pattern of γH2AX staining was not changed in CKO compared to the heterozygous control spermatocytes , showing γH2AX signal throughout the nucleus in leptonema when DSBs occur , which was then restricted to the sex body in pachynema when the autosomal breaks are repaired ( Figs 7A and S4D ) . We also measured spermatocyte proportion based on the staining , and did not detect difference in cell proportion ( Figs 7B and S4E ) . These results indicate that the sex body formation and DSB repair are not affected in the Cxxc1 CKO , and there is no major cell loss or arrest in CKO meiosis . Co-staining of SYCP1 and SYCP3 confirmed normal synapsis in all autosomes in CKO spermatocytes at pachynema ( Figs 7C and S4F ) . 98 . 5% of Cxxc1 CKO pachynema with Stra8-Cre and 98 . 23% of CKO pachynema with Ddx4-Cre showed full synapsis , compared with 97 . 5% of fully synapsed pachynema in control ( p = 0 . 15 and 0 . 91 , respectively ) . Therefore , there is no increased chromosome asynapsis in the CKO spermatocytes compared to controls . Finally , we examined whether loss of CXXC1 affects crossover resolution . Using MLH1 as a marker of crossover sites , we did not find any significant change of crossover number in the CKO spermatocytes compared to the het controls ( Fig 7D ) . Taken together , these results suggest that even though CXXC1 interacts with PRDM9 and H3K4me3 in spermatocytes , it is not required for PRDM9 binding at hotspots , their subsequent activation by PRDM9-dependent H3K4 trimethylation , DSB formation , repair , or crossover formation , and is therefore not essential for meiotic recombination events .
In this study , we demonstrate that CXXC1 interacts with PRDM9 in spermatocytes . However , this interaction does not seem to be important for any cell function necessary for germ cell development or recombination processes . The germ cell specific Cxxc1 knockout male mice are fertile . In the knockout spermatocytes , the expression and function of PRDM9 are unchanged . The loss of CXXC1 does not affect DSB formation and repair , chromosome pairing and synapsis , and crossover numbers . Together , these results convincingly show that CXXC1 is not essential for normal meiotic recombination events and generally for spermatogenesis and oogenesis . The yeast CXXC1 ortholog Spp1 is reported to be a key player in recombination by linking H3K4me3 sites with the chromosome axis and connecting them with the recombination protein Mer2 [19–21] . This suggested that CXXC1 might play similar function in mammalian meiosis . However , our results show that CXXC1 is not an essential player in mammalian recombination where PRDM9 controls the initial recognition and activation of recombination hotspots . In the absence of CXXC1 , hotspot activation , axis integrity , DSB formation and crossover resolution occur normally , showing that DSB formation and recombination determination in most of mammals , which use the PRDM9 dependent pathway , differs from that in the budding yeast . In species with functional PRDM9 , the function of the RMM complex consisting of orthologs of the yeast Rec114 , Mei4 and Mer2 ( REC114 , MEI4 and IHO1 in mice ) is still conserved [27 , 37–39] , and the association between hotspots and chromosome axis is crucial for efficient DSB formation [27 , 40 , 41] . However , the interaction between CXXC1 and IHO1 does not seem to play the same functional role as the one between Spp1 and Mer2 in yeast . One important difference is that in organisms that do not use PRDM9 , DSB occur at H3K4me3 sites , whereas in those that use PRDM9 , DSB occur at hotspots where surrounding nucleosomes are methylated at both H3K4 and H3K36 [5 , 6] . This raises the likelihood that proteins with H3K36me3 methyl-reading activity , such as PWWP domain containing proteins [42] , or with both H3K4me and H3K36me binding capability , such as Tudor domain containing proteins [43] , might be involved in hotspot recognition in these species . Alternatively , activated hotspots may be recruited to the chromosome axis and DSB machinery without assistance of an H3K4me3/H3K36me3 reader . A recent study demonstrated that randomized DSBs induced by radiation in Spo11 mutant spermatocytes are associated with chromosome axis and can successfully recruit DSB repair proteins such as DMC1/RAD51 complex [44] . Other direct PRDM9 interactors , such as EWSR1 , EHMT2 , CDYL [16] , PIH1D1 [26] and CTCF [45] , could also be involved in hotspot association with the chromosome axis . An alternative , PRDM9-independent pathway , can explain the fraction of DSB detected at promoters in wild type mice , and all DSB in PRDM9 mutant mice [9 , 46] . Although we did not detect any substantial reduction of PRDM9-independent hotspot activity in the absence of CXXC1 , this pathway could still , to some degree , involve CXXC1 as part of the SETD1 complex , known to bind H3K4me3 at promoters , in a way similar to Spp1-Mer2 role in yeast meiosis . It is not an essential pathway in most organisms using PRDM9 as hotspot determinant , but might play a major role in those lacking PRDM9 , such as canids [47–49] , where recombination hotspots are enriched in CpG-rich regions with a preference for unmethylated CpG islands [9 , 17 , 47 , 50] , similar feature as CXXC1 binding sites [22 , 51] . One recent report of a woman having no active PRDM9 but completely fertile suggests that this pathway may become activated and ensure proper recombination even in organisms using PRDM9 as a recombination regulator [52] .
The animal care rules used by The Jackson Laboratory are compatible with the regulations and standards of the U . S . Department of Agriculture and the National Institutes of Health . The protocols used in this study were approved by the Animal Care and Use Committee of The Jackson Laboratory ( Summary #04008 ) . Euthanasia for this study was done by cervical dislocation . All wild-type mice used in this study were in the C57BL/6J ( B6 ) background . The conditional-ready mutants were produced by flanking exons 2 and 3 of Cxxc1 with loxP sites using CRISPR-Cas9 . The Cxxc1 conditional knockout mice used in this study were produced by a two-step deletion scheme . Mice that harbor two conditional Cxxc1 alleles ( Cxxc1loxp/loxp ) were mated to Tg ( Sox2-cre ) 1Amc/J mice ( stock #004783 ) to generate one Cxxc1 allele deleted mice . The Cxxc1 hemizygous mice ( Cxxc1Δ/+ ) were mated to Tg ( Stra8-icre ) 1Reb/J ( stock #08208 ) and Tg ( Ddx4-cre ) 1Dcas/KnwJ ( stock #018980 ) to obtain Cxxc1Δ/+;Stra8-iCre and Cxxc1Δ/+;Ddx4-Cre mice . Those Ewsr1Δ/+;Stra8-iCre and Cxxc1Δ/+;Ddx4-Cre mice were then mated to homozygous Cxxc1 loxp mice to generate heterozygous control mice ( Cxxc1loxP/+;Stra8-iCreand Cxxc1loxP/+;Ddx4-Cre designated as Cxxc1 controls ) or conditional knockout mice ( Cxxc1loxP/Δ;Stra8-iCreand Cxxc1loxP/Δ;Ddx4-Cre designated as Cxxc1 CKO ) . B6;129P2-Prdm9tm1Ymat/J mice ( Prdm9-/- ) have been previously described [53] . All animal experiments were approved by the Animal Care and Use Committee of The Jackson Laboratory ( Summary #04008 ) . The co-immunoprecipitation assays for PRDM9 and EWSR1 with testicular extract were carried out using our reported protocol [16] . Total protein was extracted from testes of twenty 14-dpp B6 mice homogenized in 1 ml of Pierce IP buffer ( Thermo Fisher Scientific , 87787 ) . 10% of extract was set apart as input . Co- immunoprecipitation was performed by incubating extract with protein G Dynabeads conjugated with antibodies against PRDM9 [18 , 54] or guinea pig IgG overnight at 4°C . The beads were washed three times with 1 ml of Pierce IP buffer , eluted with 200 μl of GST buffer ( 0 . 2 M glycine , 0 . 1% SDS , 1% Tween 20 , pH 2 . 2 ) for 20 min at room temperature and neutralized with 40 μl of 1 M Tris-HCl , pH 8 . After heated at 95°C for 5 min , 10 μg of IP and input samples were then subjected to electrophoresis and western blotting for detection of PRDM9 ( 1:1000 , custom made ) , EWSR1 ( 1:1000 , Abcam , ab54708 ) and CXXC1 ( 1:1000 , Abcam , ab198977 ) . The co-IP experiment is performed in two replicates . The co-immunoprecipitation assays for PRDM9 , CXXC1 and EWSR1 in cell culture were carried out using our reported protocol [55] . The vector expressing the PRDM9 , CXXC1 and EWSR1 proteins were constructed by cloning mouse Prdm9 , Cxxc1 and Ewsr1 cDNA into pCEP4-Flag , pCMV-Myc and pCMV-HA vectors , respectively . 2 . 5 μg of plasmids were transfected into HEK293 cells by X-tremeGENE HP DNA Transfection Reagent ( Roche , 6366244001 ) in 6-well plates . At 2 days after transfection , cells were harvested and mixed with 600 μl Pierce IP buffer . 10% of extract was set apart as input . Co-immunoprecipitation was performed by incubating extract with protein G Dynabeads conjugated with antibodies against HA ( Sigma , H9658 ) or Myc ( Sigma , M5546 ) overnight at 4°C . After washing the beads and eluting with GST buffer , 10 μg of IP and input samples were then subjected to electrophoresis and western blotting for detection of HA , Myc and Flag ( Sigma , F1804 ) . All the blots were processed together with the same exposure . Co- immunoprecipitation for CXXC1 was performed similarly to those for PRDM9 and EWSR1 with the following changes . The seminiferous tubules were digested with liberase and the germ cells were isolated . Then , the nuclei were isolated by incubation germ cells in hypotonic lysis buffer ( 10 mM Tris-HCL pH 8 . 0 , 1 mM KCl , 1 . 5 mM MgCl2 ) for 30 min at 4°C and spinning down at 10 , 000 g for 10 min . The nuclear extract was obtained by incubation with the nuclear lysis buffer ( 50 mM HEPES , pH 7 . 8 , 3 mM MgCl2 , 300 mM NaCl , 1 mM DTT and 0 . 1 mM PMSF ) , 5 U/μl DNaseI and 2 U/ μl TurboNuclease for 30 min at 4°C . 10% of extract was saved as input . The co-IP was perform by incubating extract with protein G Dynabeads conjugated with antibodies against CXXC1 ( Abcam , ab198977 ) or guinea pig IgG overnight at 4°C . After wash and elution , the IP and input samples were then subjected to electrophoresis and western blotting for detection of CXXC1 ( 1:1000 , Abcam , ab198977 ) , H3K4me3 ( 1:1000 , Millipore , #07–473 ) and H3K9me3 ( 1:1000 , Active Motif , 39766 ) . Testicular weight and body weight of adult B6 ( n = 3 ) , Cxxc1 het ( n = 3 ) and CKO ( n = 4 ) mice were measured . Testis index was calculated as testis weight/body weight . Student’s t-test was used to determine the statistical significance . Male fertility test was performed with 3 Cxxc1 het control and 5 CKO male mice . Each mouse was mated with at least two B6 females for at least two to five month period . Female fertility test was performed with 2 Cxxc1 control and 2 CKO female mice . Each one was mated with one B6 male for 3 month period . Litter size and viable pup number were recorded . Testis , epididymis , ovaries from adult or 21 dpp Cxxc1 het control or CKO mice were dissected out . Testis and epididymis were fixed with Bouin’s solution , and ovaries were fixed in 2% PFA . and the tissues were embedded in paraffin wax , and sectioned at 5 μm . Sections of testis were stained with Periodic acid–Schiff–diastase ( PAS ) , and section of epididymis and ovaries were stained with haematoxylin and eosin ( H&E ) using standard techniques . The drying-down technique [56] was used for preparation of chromosome spreads from spermatocytes of 14-dpp and 8-weeks B6 , Cxxc1 control or CKO mice . Chromosome spread slides were immunolabeled with anti-PRDM9 ( 1:200 ) , CXXC1 ( 1:1000 ) , SYCP1 ( 1:300 , Novus , NB300-229 ) , SYCP3 ( 1:400 , Novus , NB300-231 ) , γH2AX ( 1:1000 , Abcam , ab26350 ) , DMC1 ( 1:200 , Santa Cruz , sc-8973 ) , RAD51 ( 1:200 , Santa Cruz , H-92 ) , RPA ( 1:300 , Abcam , ab87272 ) or MLH1 ( 1:100 , BD Pharmingen , 550838 ) antibodies . For protein immunolocalization on tissue sections , testicular tissues from 8 week old B6 , Prdm9-/- , Cxxc1 control and CKO mice were dissected out , fixed with 4% paraformaldehyde solution overnight , embedded in paraffin wax . 5-μm sections were prepared . For antigen retrieval , sections were heated in a microwave in Tris-EDTA buffer ( 10mM Tris , 1mM EDTA and 0 . 05% Tween 20 , pH 9 . 0 ) for 10 min and cooled down to room temperature . Then , sections were treated with PBS containing 0 . 1% Triton X-100 , blocked with 10% normal donkey serum , and stained with antibodies against PRDM9 ( 1:200 ) , CXXC1 ( 1:1000 ) or H3K4me3 ( 1:1000 , Millipore , #07–473 ) . Chromatin immunoprecipitation ( ChIP ) was performed as previously described [57] . Briefly , spermatocytes were isolated from 14-dpp B6 , Cxxc1 het and CKO spermatocytes , and crosslinked using 1% formaldehyde . Nuclei were isolated using hypotonic lysis buffer ( 10 mM Tris-HCL pH 8 . 0 , 1 mM KCl , 1 . 5 mM MgCl2 ) and digested by MNase . The ChIP was done using antibody against H3K4me3 . Real-time PCR was performed with purified ChIP DNA using Quantifast SYBR Green PCR Kit ( Qiagen ) Primer sequences used for real-time PCR are: Pbx1_F: AGAAACTGACATATGAAGGCTCA; Pbx1_R: GCTTTTGCTCCCTTAAACTGG; Fcgr4_F: CAAGGTGCATTCTTAGGAGAGA; Fcgr4_R: TTAATGCTTGCCTCACGTTC; Hlx1_F: GGTCGGTGTGAGTATTAGACG; Hlx1_R: GGCTACTATACCTTATGCTCTG; Actinb_promoter_F: GCCATAAAAGGCAACTTTCG; Actinb_promoter_R: TTTCAAAAGGAGGGGAGAGG; Sycp3_promoter_F: AAGGCGCCACAACCAAGG; Sycp3_promoter_F: TGCCTGGATGCCCAACTC . DMC1 ChIP was performed from spermatocytes of 8 weeks old Cxxc1 het and CKO ( Stra8-Cre ) using an established method [58] in two replicates . The testes were cross-linked with 1% paraformaldehyde solution for 10 min , and then homogenized . After that , the nuclei were isolated , the chromatin was sheared to ~1000 bp by sonication and incubated with antibody against DMC1 overnight at 4°C , and then with protein G Dynabeads ( Thermo Fisher Scientific , 10004D ) for 4 hrs at 4°C . The beads were washed once with wash buffer 1 ( 0 . 1% SDS , 1% Triton X-100 , 2 mM EDTA , 20 mM Tris-HCl , pH 8 . 0 , 150 mM NaCl ) , wash buffer 2 ( 0 . 1% SDS , 1% Triton X-100 , 2mM EDTA , 20mMTris-HCl , pH 8 . 0 , 500 mM NaCl ) , then wash buffer 3 ( 0 . 25 M LiCl , 1% NP-40 , 1mM EDTA , 10mMTris-HCl , pH 8 . 0 , 1% Deoxycholic acid ) , and finally twice with TE buffer . The chromatin was eluted with dilution buffer ( 1% SDS , 0 . 1 M NaHCO3 pH 9 . 0 ) at 65°C for 30 min and then reverse-crosslinked by adding 200 mM NaCl and incubation overnight at 65°C . The libraries were then prepared according to the currently established method [58] , and sequenced on an Illumina HiSeq 2500 platform , with 75 bp paired-end reads . Fastq files for sequenced DMC1 libraries were trimmed using Trimmomatic ( v0 . 32 ) and subsequently parsed for detection and selection of paired reads having homology at the 5’ and 3’ ends [9 , 59] as established by protocols for single strand DNA enrichment to generate the files that contain only the detectable single strand reads , and then , these files were aligned to mm10 mouse genome using BWA ( v . 0 . 5 . 10-tpx ) . Bam files were parsed for detection and selection of reads containing true genomic sequence versus fill-in sequence at the homologous region . These reads were selected from the original paired-end fastq files , and then single-end fastq files were created that contained only the true genomic sequences of single strand DNA reads . All genomic data are available at NCBI Gene Expression Omnibus ( GEO; http://www . ncbi . nlm . nih . gov/geo ) under accession number GSE116336 ( https://www . ncbi . nlm . nih . gov/geo/query/acc . cgi ? acc=GSE116336 ) . 1 , 688 , 630 and 1 , 349 , 516 aligned DMC1 reads in Cxxc1 het controls , 1 , 590 , 925 and 1 , 961 , 645 aligned DMC1 reads in Cxxc1 CKO spermatocytes were obtained from the two replicate libraries . The correlation between the two biological replicates in each experiment was high ( r = 0 . 96 in Cxxc1 het controls; r = 0 . 99 in Cxxc1 CKO . S5A Fig ) ; thus , the data from each pair of replicates were merged . The DMC1 activity was normalized to reads per million ( rpm ) . Peak calling was performed using MACS ( v . 2 . 0 . 9 ) with a FDR value 0 . 01 . PRDM9 dependent or default sites were determined using bedtools ( v2 . 22 . 0 ) intersects compared with unknown PRDM9 binding sites ( GEO number: GSE61613 ) [18] . Analyses for the aggregation plots were carried out using the ACT [60] , of which parameters were: nbins = 500 , mbins = 0 , radius = 1500 . | Meiotic recombination increases genetic diversity by ensuring novel combination of alleles passing correctly to the next generation . In most mammals , the meiotic recombination sites are determined by histone methyltransferase PRDM9 . These sites are proposed to become associated with the chromosome axis with the participation of additional proteins and undergo double strand breaks , which are repaired by homologous recombination . In budding yeast , Spp1 ( ortholog of CXXC1 ) binds to methylated H3K4 and connects these sites with the chromosome axis promoting DSB formation . However , our data suggest that even though CXXC1 interacts with PRDM9 in male germ cells , it does not play a crucial role in mouse meiotic recombination . These results indicate that , unlike in yeast , a recombination initiation pathway that includes CXXC1 could only serve as a non-essential pathway in mouse meiosis . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"meiosis",
"chromosome",
"staining",
"medicine",
"and",
"health",
"sciences",
"reproductive",
"system",
"spermatocytes",
"cell",
"cycle",
"and",
"cell",
"division",
"cell",
"processes",
"germ",
"cells",
"animal",
"models",
"fungi",
"model",
"organisms",
"experimental",
"organism",
"systems",
"dna",
"sperm",
"homologous",
"recombination",
"research",
"and",
"analysis",
"methods",
"saccharomyces",
"specimen",
"preparation",
"and",
"treatment",
"staining",
"animal",
"cells",
"animal",
"studies",
"chromosome",
"biology",
"mouse",
"models",
"yeast",
"biochemistry",
"eukaryota",
"cell",
"biology",
"nucleic",
"acids",
"anatomy",
"genetics",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"saccharomyces",
"cerevisiae",
"dna",
"recombination",
"yeast",
"and",
"fungal",
"models",
"seminiferous",
"tubules",
"organisms",
"genital",
"anatomy"
] | 2018 | CXXC1 is not essential for normal DNA double-strand break formation and meiotic recombination in mouse |
In a previous study , we found that Trichinella spiralis muscle larva excretory and secretory proteins ( ES-P ) most likely activate collagen synthesis via TGF-β/Smad signaling , and this event could influence collagen capsule formation . In order to identify the specific collagen inducing factor , ES-P was fractionated by a Superdex 200 10/300 GL column . We obtained three large fractions , F1 , F2 , and F3 , but only F3 had collagen gene inducing ability . After immunoscreening , 10 collagen inducing factor candidates were identified . Among them , TS 15–1 and TS 15–2 were identical to the putative trypsin of T . spiralis . The deduced TS 15–1 ( M . W . = 72 kDa ) had two conserved catalytic motifs , an N-terminal Tryp_SPc domain ( TS 15-1n ) and a C-terminal Tryp_SPc domain ( TS 15-1c ) . To determine their collagen inducing ability , recombinant proteins ( rTS 15-1n and rTS 15-1c ) were produced using the pET-28a expression system . TS 15–1 is highly expressed during the muscle larval stage and has strong antigenicity . We determined that rTS 15-1c could elevate collagen I via activation of the TGF-β1 signaling pathway in vitro and in vivo . In conclusion , we identified a host collagen inducing factor from T . spiralis ES-P using immunoscreening and demonstrated its molecular characteristics and functions .
Trichinella spiralis can make collagen capsules in host muscles during their life cycle that surround muscle stage larvae and might protect the larvae from the host immune system . This phenomenon can be understood as the parasite creating a simple structure to protect itself , but when examined closely , numerous different mechanisms are involved in this stage of the parasite’s life . Division of the host muscle cell nucleus , regulation of host cell cycling , huge elevation of host collagen gene expression , and generation of new blood vessels around the collagen capsule are observed during nurse cell formation by T . spiralis [1–4] . The process of nurse cell formation induces de-differentiation , cell cycle re-entry , arrest of infected muscle cells , and activation , proliferation , and differentiation of satellite cells . These events are very similar to those occurring during muscle cell regeneration and repair [2] . In a previous study , we found that T . spiralis excretory and secretory proteins ( ES-P ) most likely activate collagen synthesis via TGF-β/Smad signaling , and this event could influence collagen capsule formation [5] . These events were closely related with protease activated receptor 2 ( PAR2 ) , which was activated by various serine proteases [5] . However , the question of which protease in T . spiralis ES-P has a role in collagen gene expression of host muscle cells is still unanswered . The identification of a specific collagen gene inducer from T . spiralis could be exploited as a therapeutic and/or cosmetic agent . In this study , we isolated and characterized the collagen gene expression inducer from T . spiralis ES-P by immunoscreening and investigated the candidate for its usefulness as a wound healing therapeutic agent .
The T . spiralis strain ( isolate code ISS623 ) used in this study has been maintained in our laboratory via serial passage in rats . For acquisition of muscle larva , eviscerated mouse carcasses were cut into pieces , followed by digestion in 1% pepsin 1% hydrochloride digestion fluid ( artificial gastric juice ) for 1 hr at 37°C with stirring . Larvae were collected manually from muscle digested solution under microscopy and washed 6 times with sterile PBS containing 100 μg/ml ampicillin , 5 μg/ml kanamycin and 50 μg/ml tetracyclin . After collection , in order to prevent contamination with the host material , worms were thoroughly and carefully washed several 3 times with PBS . Whole parasite proteins ( total extract; TE ) was obtained from muscle larva according to previous study [6] . In brief , muscle larva were rinsed in PBS and homogenized in 50 mM Tris–HCl ( pH 7 . 5 ) with a glass homogenizer . The homogenates were briefly sonicated and then centrifuged for 30 min at 12 , 000 × g and 4°C . The supernatant ( TE ) was stored at -20°C . Small intestines were removed on the day 6 after infection from infected rat , opened , sliced by 2 cm , washed with PBS , and incubated for 1 hr at 37°C in PBS containing antibiotics . Adult worms were collected on a PBS , washed 3 times with PBS containing antibiotics , and incubated for 24 hrs in serum-free RPMI 1640 medium containing antibiotics . After incubation , NBL were passed through 40 μl nylon mesh ( BC falcon , USA ) to be separated from adult worms . Muscle larvae were isolated from T . spiralis infected mice ( 4 weeks after infection ) and ES-P from cultured muscle larvae was obtained according to the previously reported method [5] . The ES-P was fractionated using gel filtration chromatography . ES-P ( 5 mg ) in 10 ml PBS was applied to a Superdex 200 10/300 GL column ( GE Healthcare , Uppsala , Sweden ) . The flow rate was 0 . 25 ml/min . Each 0 . 5 ml fraction was collected and protein quantity was measured by UV detection at 260 nm . Three big fractions , F1 , F2 , and F3 , were acquired and used for collagen gene inducing experiments ( Fig 3A ) . Twenty female C57BL/6 mice at the age of 6 weeks and twenty female 14 week-old mice were purchased from Samtako Co . ( Gyeonggi-do , Korea ) . The skin of the left ear of each mouse was treated with T . spiralis ES-P ( 30 μg ) or rTS 15-1c ( 30 μg ) in PBS ( total volume 50 μl ) every day for 14 days , and that of the right ear was treated with PBS ( Figs 1A and 7A ) . The mice were housed in a specific pathogen-free facility at the Institute for Laboratory Animals of Pusan National University . In order to compare the expression level of type I collagen and their signal pathway related genes , mouse fibroblast ( MEF ) cells were used in this study because type I collagen was preferentially synthesized by two cell types , the osteoblast and the fibroblast [7] . MEF cells were isolated from C57BL/6 mouse fetuses 10 days after fertilization [8] . MEF cells were incubated in DMEM ( Difco ) with 5% FBS and 5 × 105 cells were plated in 24-well plates and incubated overnight at 37°C in 5% CO2 . The cells were treated with ES-P , boiled- ES-P , F1 , F2 , F3 , boiled F3 , TS 15-1c , and TS 15-1n ( final conc . 1 μg/ml ) ; ES-P with PMSF ( serine protease inhibitor , final conc . 1 mM; Sigma-Aldrich , USA ) , F3 with PMSF for 2 hrs . Gelatin-gel containing 0 . 2% gelatin was prepared from gelatin-stock solution . The proteins , T . spiralis ES proteins , TS 15-1c , and TS 15-1n were mixed with 2 × sample buffer ( 1 M Tris pH 6 . 8 , 1% bromphenol blue , glycerol , β-mercaptoethanol ) , and the gel loaded with these proteins was run with 1 × Tris-Glycine SDS running buffer on 125 V for 2 hrs at 4°C . After running , the gel was washed to remove the SDS and re-natured proteinase activity with zymogram renaturing buffer ( 2 . 5% Triton X-100 ) . The gel was developed with zymogram developing buffer ( 0 . 5 M Tris-HCl pH 7 . 6 , 0 . 02 M NaCl , 0 . 5 mM CaCl2 ) for 30 min at room temperature . The gel was incubated with developing buffer at 37°C for 8 hrs . The gel was stained with Coomassie Blue R-250 for 30 min and distained with an appropriate destaining solution ( Bio-Rad laboratories , Inc . , USA ) . Homogenized ear tissues were mixed with TRIzol ( Invitrogen , Germany ) , and RNA extraction and cDNA synthesis ( Invitrogen , Germany ) was performed in accordance with the manufacturer’s protocols . Expression levels of several genes were determined with real-time RT-PCR using the iCycler ( Bio-Rad laboratories Inc . , USA ) real-time PCR machine . Primer sequences for collagen I , TGF-β , smad2 , smad3 , and GAPDH , and PCR conditions were identical to those mentioned in the previous study [5] . To evaluate variation of Ts-15-1 gene expression during T . spiralis life cycle , total RNAs were extracted from new born larva , adult worm , muscle larva and T . spiralis infected mouse muscle ( 1 , 2 , and 4 weeks after infection ) using TRIzol ( Invitrogen , Germany ) , and cDNA synthesis ( Invitrogen , Germany ) was performed in accordance with the manufacturer’s protocols . Expression levels of several genes were determined with real-time RT-PCR using the iCycler ( Bio-Rad laboratories Inc . , USA ) real-time PCR machine . The primer sequences for the putative trypsin ( TS 15-1c ) , and T . spiralis GAPDH were 5′- TTG GAA TGA CGC TGA TTG -3′ , 5′- GTG GCT TAT GAT GGT AGG AGA AT -3′ and 5′- CAG GTG CTG ATT ACG CTG TT -3′ , R—5′- ACG CCA ATG CTT ACC AGA T -3′ respectively . Amplification of two genes was performed under the following conditions: 1 min 30 sec host start at 95°C , followed by denaturation at 95°C for 25 sec , primer annealing at 50 ~ 55°C for 20 sec , and elongation at 72°C and 30 sec for 40 cycles . Fluorescent DNA-binding dye SYBR was monitored after each cycle at 50 ~ 55°C . An iCycler multi-color real-time PCR detection system ( Bio-Rad Laboratories ) was used for estimation of expression levels . Then , using the Gene-x program ( Bio-Rad Laboratories ) , relative expression of the gene was calculated as the ratio to a T . spiralis GAPDH gene . A cDNA library generated from 60 , 000 plaques forming units of T . spiralis muscle larvae was screened with the α-TS F3 antibody . Immunoscreening was performed using the SMART cDNA Library Construction Kit ( Clontech , USA ) in accordance with the manufacturer’s protocols . Briefly , after primary and secondary screening , positive plaques were picked and the phagemids were prepared by in vivo excision . The phagemids were transformed into XL1-Blue MRF cells . Clones were selected based on blue-white color selection of the colonies grown on LB-ampicillin agar plates . The plasmid harboring the cDNA inserts were then extracted using a plasmid DNA purification system ( Cosmogenetech , Seoul , Korea ) . The cDNA inserts were then sequenced using the primer for T3 promotor ( Cosmogenetech , DNA sequencing service , Seoul , Korea ) and compared against the GenBank database . Following confirmation of the PCR product sequences , TS 15–1 , The TS 15-1c ( C-terminal serine protease domain ) and TS 15-1n ( N-terminal serine protease domain ) , the genes were ligated with pET-28a vector ( Novagen , USA ) . After gene ligation , the constructed plasmids were expressed in Escherichia coli BL21 ( DE3 , Novagen , USA ) . Pre-cultured cells were inoculated into Luria-Bertani broth containing kanamycin , and the cells were grown at 37°C until an OD600 of 0 . 5–0 . 6 was reached . Recombinant TS 15-1N and TS 15-1C expressions were induced addition of 0 . 5 mM isopropyl β-D-1-thiogalactopyranoside ( IPTG ) at 25°C for 16 h . The cells were harvested by centrifugation and the cell pellets were resuspended in buffer . A consisting of 50 mM Tris–HCl pH 7 . 5 and 200 mM NaCl . Cell disruptions were lysed by sonication on ice and the crude extracts were centrifuged to remove the cell debris . Ts15-1 N and C pellets were then sonicated in buffer including 50 mM Tris–HCl pH 7 . 5 , 200 mM NaCl , and 6 M Urea on ice . The clear supernatant of the lysate was subjected onto the Ni–NTA column which had been pre-equilibrated with buffer A . The column was subsequently washed with buffer A containing imidazole , after which the bound proteins were eluted by varying the imidazole concentration ( 20–400 mM ) . The eluted proteins were analyzed using 10% SDS-PAGE . However , we could not get recombinant protein of TS 15–1 because very poor expression level . Female four-week-old Wistar rats were purchased from Samtako Co . ( Gyeonggi-do , Korea ) . Rats were immunized subcutaneously with a 1:1 mixture of the 250 μg F3 fraction of TS 15-1c protein ( in 0 . 5 ml PBS ) and 0 . 5 ml Freund’s complete adjuvant ( #F5881 , Sigma-Aldrich , USA ) at 0 week . At 2 weeks the rat was given additional infections of the 250 μg F3 fraction or TS 15-1c protein with Freund’s incomplete adjuvant ( #F5506 , Sigma-Aldrich , USA ) . One week after their final booster , rats were sacrificed and serum was obtained . Homogenized ear tissues were mixed with TRIzol ( Invitrogen , Germany ) , and total RNA extraction and cDNA synthesis ( Invitrogen , Germany ) was performed in accordance with the manufacturer’s protocols . Expression levels of several genes were determined with real-time RT-PCR using the iCycler ( Bio-Rad laboratories Inc . , USA ) real-time PCR machine . Primer sequences for collagen I , TGF-β , smad2 , smad3 , and GAPDH , and PCR conditions were identical to those mentioned in the previous study [5] . Each gene expression levels were normalized with GAPDH gene expression . Mice were killed at 0 , 1 , 2 , and 4 weeks after T . spiralis infection and serum was obtained . Sera were stored at -20°C until used . Ten μg each of ES-P and F3 fraction ( Fig 3 ) or 10 μg ES-P and total extract from T . spiralis , TS 15-1c ( Fig 5B ) or 10 μg of purified TS 15-1c antibody ( Fig 5C ) or 15 μg of each ear tissue samples ( Fig 7 ) were separated on 10% acrylamide SDS-PAGE gel at 100 V for 90 min . Sweden ) , The loaded proteins were transferred onto a nitrocellulose membrane ( Amersham Biosciences , Little Chalfont , UK ) and blocked with 5% skim milk in TBST at 4°C overnight . Then , the membrane was incubated with primary antibody ( polyclonal α-F3 , α-TS 15-1c ( 1:500 ) ; time-course sera ( 1:1 , 000 ) , α-TGF-β1 ( 1:1000; abcam , Carlsbad , CA , USA ) , ;p-Smad2/3 ( 1:1000; Thermofisher science , Waltham , MC , USA ) , ;α-mouse type I collagen ( 1:1000; abcam ) , and actin ( 1:5000 , abcam ) ) in 5% skim milk in TBST for 2 hrs at room temperature . The secondary antibody , α-mouse or α-rat IgG-HRP conjugate ( Sigma , Seoul , Korea ) was used at 1:5 , 000 dilution for 1 hr at room temperature . HRP was detected using an ECL substrate ( Amersham Biosciences , Uppsala , Sweden ) , analyzed using the LAS 3000 machine . ( Areas of the detected bands were determined and compared by Image J software ) . Paraffin-embedded T . spiralis infected or non-infected mouse muscle tissue were de-paraffinized and hydrated . For antigen retrieval , slides were immersed in citrate buffer ( 0 . 01 M , pH 6 . 0 ) and heated twice in a microwave ( 700 W or ‘high’ ) for 5 min . Slides were then quenched with endogenous peroxidase by incubation in a 3% hydrogen peroxide solution for 5 min and were washed three times in PBS for 5 min each . Slides were immuno-stained with primary antibody ( α-TS 15-1c antibody that was produced according to the polyclonal antisera method; 1:500 dilution ) at 4°C overnight . After primary antibody incubation , slides were washed three times in PBS for 5 min each and were incubated with secondary antibody , the Alexa Fluor 594 goat anti-rat IgG secondary antibody ( 1:500; Invitrogen , USA ) was applied for 1 h at 24°C . The slides were washed in PBS and mounted with Permount ( Fisher Scientific , Pittsburgh , PA , USA ) . Confocal images of stained muscle tissue were examined under an inverted fluorescence microscope . All experiments were performed three times for confirmation of statistical significance . Mean ± standard deviation ( SD ) was calculated from data collected from individual mice . Significant differences were determined using one-way or two-way analysis of variance . Statistical analysis was performed with GraphPad Prism 5 . 0 software ( GraphPad Software Inc . , CA , USA ) . The study was performed with approval from the Pusan National University Animal Care and Use Committee ( IACUC protocol approval; PNU-2016-1175 ) , in compliance with ‘‘The Act for the Care and Use of Laboratory Animals” of the Ministry of Food and Drug Safety , Korea . All animal procedures were conducted in a specific pathogen-free facility at the Institute for Laboratory Animals of Pusan National University .
In order to understand the collagen gene inducing effect of ES-P , transcription and protein expression levels of type I collagen and TGF-β1 signaling related proteins were compared in ear tissues of 6 and 14 week-old mice that had or had not received ES-P treatment ( Fig 1A ) . The expression levels of the collagen I and TGF-β1 genes of the 14 week-old mice were significantly decreased compared to those of the 6 weeks mice . However , those of the ES-P treated 14 week-old mouse group were significantly increased compared to un-treated 14 week-old mice ( Fig 1B ) . To identify type I collagen inducing factors from the T . spiralis ES-P , the ES-P was fractionated to several fractions including three big fractions by gel filtration chromatography . The three major fractions obtained were named F1 ( about 100 kDa– 140 kDa ) , F2 ( about 80 kDa—120 kDa ) , and F3 ( about 50 kDa -90 kDa ) respectively ( Fig 2A ) . In order to determine which major fraction had collagen gene expression inducing factors , each fraction was used to treat MEF cells , and expression levels of the type I collagen gene were measured . After F1 , F2 , and F3 treatment , only F3 treated MEF cells had significantly increased type I collagen gene expression ( Fig 2B ) . Moreover , expression levels of collagen I in F3 treated MEFs was higher those of ES-P treated MEFs ( Fig 2B ) . In order to determine whether the collagen inducing ability of F3 is related with serine protease activity , we evaluated the collagen inducing ability of F3 following pre-treatment with a serine protease inhibitor , PMSF , on MEF cells . The collagen I gene expression levels were significantly decreased in MEF cells pre-treated with PMSF , the serine protease inhibitor . In addition , the gene expression levels were also not increased when treated with boiled F3 ( Fig 2C ) . In order to confirm the existence and expression levels of F3 in ES-P , α-F3 polyclonal antibody was used against ES-P and F3 in a western blot analysis . The presence of a strong band was observed at 60–80 kDa in both the ES-P and F3 samples ( Fig 2D ) . In order to identify the collagen gene inducing factors from T . spiralis ES-P , immunoscreening was conducted against the T . spiralis muscle larvae Express cDNA library with the α-F3 antibody . Thirty-five positive plaques were detected in primary screening , among them , 10 plaques were confirmed by second screening ( S1 Fig ) . These plaques were amplified and processed in an in vivo excision step . All the insert DNA from the 10 positive clones were sequenced and their amino acid sequences were determined . Two insert DNA fragments ( TS 15–1 and TS 15–2 ) were similar to a putative trypsin of T . spiralis with 90% identity . Another clone ( TS 15–3 ) was matched to a nuclear receptor-binding protein of T . spiralis with 35% of identity . Another clone ( TS-16-1 ) was matched to a putative BTB/POZ domain protein of T . spiralis with 63% identity . The remaining 6 insert clones were not matched with any previously known genes . Collagen inducing factors in ES-P and F3 had serine protease activity and were measured to be about 60–72 kDa in size . After evaluation of the size and serine protease activity of positive clone matched genes , the TS 15–1 gene was selected for downstream identification of the collagen inducing factor . The TS 15–1 fragment was 2 , 004 bp long and encoded a 667 amino acid protein , and the molecular weight and pI was calculated as 71 . 6 kDa and 8 . 83 . The deduced TS 15–1 protein has two conserved catalytic motifs , an N-terminal Tryp_SPc domain ( TS 15-1n ) and a C-terminal Tryp_SPc domain ( TS 15-1c ) ( Fig 3A ) . The TS 15-1n and TS 15-1c peptides were composed of 238 amino acids and 239 amino acids respectively , and the molecular weight was calculated to be 26 . 1 kDa and 26 . 2 kDa respectively . Unfortunately , collection of recombinant full length of TS-15 protein was very difficult because its expression level was very low . We conducted recombinant protein expression of the N and C terminal domains ( about 26 kDa , Fig 3B ) and evaluated their collagen gene inducing ability . collagen I expression levels in the recombinant TS 15-1c protein treated cells were significantly increased compared to a media control . However , recombinant TS 15-1n protein treated cells were not significantly changed compared to those of medium treated cells ( Fig 3C ) . In order to know the protease activity of both recombinant proteins , zymogram analysis was conducted . A collagen digested clear zone was detected around ~26 kDa in the recombinant TS 15-1c protein lane , but the clear zone was not detected with the recombinant TS 15-1n protein ( Fig 3D ) . In this study , we determined a molecular model by homology modeling based on the structure of another serine protease ( PDB ID: 1KYN , 1–235 ) . In TS 15-1c ( G342—T580 ) , predictions of the active sites ( H389 , D444 , and S533 ) and the substrate binding sites ( G527 , S553 , and G555 ) are shown in blue and red letters , respectively ( Fig 4A ) . Interestingly , these results indicated that these sites interact with inhibitors and ligands . Most residues in these regions had negative charges in a globular fold ( Fig 4B ) . Surprisingly , both TS 15-1n and granzyme B structures were shown to have very similar folding patterns ( S2 Fig ) . The TS 15-1n is approximately 35% homologous to granzyme B ( 21–247 ) . We found TS-15-1c and TS 15-1n molecular 3D model was quite different each other . In order to determine when the Ts 15–1 mRNA was the most highly expressed in the T . spiralis life cycle , real-time PCR was performed on new born larvae , adult worms , muscle stage larvae of T . spiralis , and during the T . spiralis infection period ( at 0 , 7 , 14 , 28 days after infection ) . As the results show , the Ts 15–1 gene was the most highly expressed in muscle stage larvae , and its expression is also highly elevated 28 days after infection ( Fig 5A ) . In order to know whether TS 15–1 was secreted from parasites , an α-TS 15-1c antibody was produced and was reacted with T . spiralis ES-P and total extract . The TS 15-1c antibody strongly reacted with proteins around 72 kDa in ES-P and slightly reacted with a total extract at the same size ( Fig 5B ) . Furthermore , to know whether TS 15-1c has antigenicity or not , T . spiralis infected mice sera ( 0 , 1 , 2 , and 4 weeks after infection ) were reacted with recombinant TS 15-1c protein ( Fig 5C ) . rTS 15-1c most strongly reacted with mouse serum collected 4 weeks after infection . To know where TS 15–1 is secreted in the parasite , α-TS 15-1c antibody was reacted against serial sections of the T . spiralis infected muscle using immunohistochemical methods . α-TS 15-1c antibody strongly reacted with only the ladder shapes structure around the esophagus in muscle stage larvae that appear to be stichocytes ( Fig 6 ) . In order to know whether TS 15-1c had type I collagen elevating ability , we applied TS 15-1c to the ear skin of 6 week- and 14 week-old mice and evaluated the expression levels of collagen I , Smad2/3 , and TGF-β1 ( Fig 7A ) . All of the tested genes’ expression levels , including the collagen I gene of the 14 week-old mice , were significantly lower than those in the 6 week-old mice . However , after 14 TS 15-1c treatments on 14 week-old mice , collagen I , Smad2/3 , and TGF-β1 gene expression levels in these mice were significantly increased compared with non-treated mice of the same age ( Fig 7B ) . We investigated protein levels of type I collagen , the phosphorylation form of Smad2/3 , and levels of TGF-β1 in TS 15-1c treated 14 week-old mice and the protein levels were considerably recovered relative to those of the non-treated 14 week-old mice ( Fig 7C ) .
In this study , we identified the host collagen inducing factor from T . spiralis , named it TS 15–1 , and confirmed its serine protease activity and ability to elevate type I collagen , TGF-βI , and related signal proteins ( Smad2/3 ) on a transcriptional and protein level . In addition , we found that it was secreted outside the parasite and elicited specific antibody production from the host immune system . In a previous study , we revealed that the ES-P of T . spiralis could induce collagen production of host muscle tissue during the infection period , and that it was closely related with serine protease activity [5] . We can carefully suggest that TS 15–1 is one of the key collagen inducing factors in ES-P revealed in our previous study . Although we could not demonstrate that TS 15–1 is one of the key molecules in the collagen capsules around the nurse cell formation step , it might be one of the central factors for collagen capsule formation . This is because TS 15–1 gene expression level was the highest during the T . spiralis muscle larva stage and its specific antibodies could be detected in mouse serum from 1 week up to 4 weeks after a T . spiralis infection ( Fig 5 ) . During the nurse cell formation period ( 1 week–4 weeks ) , T . spiralis might strongly secrete TS 15–1 to induce collagen capsule synthesis by the host muscle cell . Parasite secretory proteases might have important functions in modulating the interactions between parasites and hosts because of their particular roles in the invasion of host tissues , parasite nutrition , and evasion of host immune responses [9–12] . A trypsin-like serine protease of parasites could be involved in host immune regulation [13–15] . Serine proteases in nematodes are known to be involved in invasion into host cells and tissues and are likely to be important in molting . TS 15–1 was revealed to be a serine protease , trypsin like protein , because its activity was inhibited by PMSF ( Fig 2C ) and it was composed two domains which were very similar but not identical to each other ( Fig 3A ) . Several secreted serine proteases have been identified among T . spiralis ES proteins , including the 69 kDa putative serine protease TsSerP ( two trypsin-like domains ) , the 45 kDa serine protease TspSP-1 , and a 35 . 5 kDa serine protease [9 , 11 , 16–18] . Most of these have strong antigenicity , specific antibodies for them are easily detected experimentally in infected animal sera , and they have one or two trypsin like domains [9 , 11 , 16] . Most secreted proteases could elevate their specific antibodies during nematodiasis [16 , 19] . Trap et al . , reported the identification of the putative serine protease , TsSerP , isolated from the T . spiralis adult-newborn larvae stage . It has two trypsin-like serine protease domains flanking a hydrophilic domain , which is the same structure as TS 15–1 . Immunohistochemistry analysis revealed that TsSerP was located on the inner layer of the cuticle and esophagus of the parasite , TS 15–1 was also detected on the inner layer of the cuticle and stichocytes in this study ( Fig 6 ) . These two serine proteases of T . spiralis might have similar functions , although the function of TsSerP was not clearly revealed [9] . In this study , it was revealed that TS 15–1 could elevate collagen expression via the TGF-β1 signaling pathway in host tissue of normal aged mice . Because , the recombinant TS 15-1c protein could elevate collagen I production and the TGF-βI signaling pathway related to Smad2/3 proteins ( Fig 7 ) . Type I collagen expression is closely related with the TGF-βI/Smad2/Smad3 signaling pathway [20–22] . This characteristic could be used for therapeutic effects including wound healing and cosmetic usefulness with wrinkle reduction . The various serine proteases may participate in physiological or pathological processes , like tissue repair , vascular remodeling , and wound healing , which depend on cell proliferation and migration [23 , 24] . Type I collagen is the major structural protein in the skin . Collagen destruction is thought to underlie the appearance of aged skin and changes resulting from chronic sun exposure [25] . Ultraviolet irradiation from the sun has deleterious effects on human skin including cancer , photo-aging , and intrinsic aging [26] . TGF-β/Smad pathway is the major regulator of collagen homeostasis and plays a crucial role in dermal fibrosis [27 , 28] . TGF-β is the most potent direct stimulator of collagen production . Moreover , TGF-β is central to the process of wound healing and fibrosis formation in skin [29 , 30] . It is well understood that activation of TGF-β signaling pathways stimulus the Smad family downstream via phosphorylation . Wound healing is a well-orchestrated process , where numerous factors are activated or inhibited in a sequence of steps [31] . Numerous signaling pathways are involved , among of them , the TGF-β1/Smad pathway is representative and well known to participate in the wound healing process [31] . Hozzein et al . , suggested that topical application of propolis would promote the wound healing process by promoting TGF-β/Smad signaling , leading to increased expression of collagen type I [32] . The gradual loss of collagen in skin with aging results in wrinkles and other signs of skin aging [33] . The content of type I collagen , the major collagen in the skin and a marker of collagen synthesis , is deceased by 68% in old skin versus young skin , and cultured young fibroblasts synthesize more type I collagen than old cells [33] . In addition , a possible influence of collagen membrane on extracellular matrix synthesis was addressed using analysis of TGF-β1 and Smad2/3 complex [34 , 35] . In conclusion , we identified a host collagen inducing factor from ES-P using immune screening methods and demonstrated the molecular/genetic characteristics and function of TS 15-1c . Further study will be required to understand the detailed mechanisms for receptors in the host cells , and to identify the minimal structure that can induce collagen for cosmetic and medical purposes . | Trichinella spiralis can make collagen capsules in host muscle cells during its life cycle , which encapsulates muscle stage larvae . Many investigators have tried to reveal the complex mechanism behind this collagen capsule architecture , and it has been suggested that several serine proteases in excretory-secretory proteins of the parasite are potential collagen capsule inducing factors . In addition , collagen synthesis is activated through the TGF-β/Smad signaling pathway and these events are closely related with protease activated receptor 2 which was activated by various serine proteases . In this study , we isolated and characterized a collagen gene expression inducer from T . spiralis ES-P using immunoscreening and investigated the candidate protein for its usefulness as a wound healing therapeutic agent . | [
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion"
] | [
"enzymes",
"enzymology",
"collagens",
"serine",
"proteases",
"developmental",
"biology",
"muscle",
"proteins",
"proteins",
"gene",
"expression",
"life",
"cycles",
"recombinant",
"proteins",
"biochemistry",
"signal",
"transduction",
"cell",
"biology",
"genetics",
"tgf-beta",
"signaling",
"cascade",
"biology",
"and",
"life",
"sciences",
"proteases",
"cell",
"signaling",
"larvae",
"signaling",
"cascades"
] | 2018 | Identification of a host collagen inducing factor from the excretory secretory proteins of Trichinella spiralis |
Control of infection with Mycobacterium tuberculosis ( Mtb ) requires Th1-type immunity , of which CD8+ T cells play a unique role . High frequency Mtb-reactive CD8+ T cells are present in both Mtb-infected and uninfected humans . We show by limiting dilution analysis that nonclassically restricted CD8+ T cells are universally present , but predominate in Mtb-uninfected individuals . Interestingly , these Mtb-reactive cells expressed the Vα7 . 2 T-cell receptor ( TCR ) , were restricted by the nonclassical MHC ( HLA-Ib ) molecule MR1 , and were activated in a transporter associated with antigen processing and presentation ( TAP ) independent manner . These properties are all characteristics of mucosal associated invariant T cells ( MAIT ) , an “innate” T-cell population of previously unknown function . These MAIT cells also detect cells infected with other bacteria . Direct ex vivo analysis demonstrates that Mtb-reactive MAIT cells are decreased in peripheral blood mononuclear cells ( PBMCs ) from individuals with active tuberculosis , are enriched in human lung , and respond to Mtb-infected MR1-expressing lung epithelial cells . Overall , these findings suggest a generalized role for MAIT cells in the detection of bacterially infected cells , and potentially in the control of bacterial infection .
Mycobacterium tuberculosis ( Mtb ) , which causes tuberculosis ( TB ) , remains a leading cause of infectious disease mortality worldwide [1] . The majority of TB cases are exclusively pulmonary , suggesting a need for mucosal immunity in the control of Mtb . Th1-type immunity , including strong CD4+ Th1 cell and CD8+ T-cell responses , mediates control of Mtb infection [2] . Though many functions of CD4+ Th1 cells and CD8+ T cells are redundant , CD8+ T cells contrast with CD4+ cells in their ability to recognize MHC class II-negative cells and preferentially recognize cells heavily infected with Mtb [3] . In humans , Mtb-specific CD8+ T cells are present at high frequencies in both Mtb-infected and uninfected individuals [4] , [5] . The presentation of peptide antigen bound to HLA-A , B , or C to CD8+ T cells is well characterized [4] , [6] and has been termed HLA-Ia or classical antigen presentation . Several nonclassical MHC-Ib ( HLA-Ib ) systems have been described as well . In general , these systems utilize molecules of limited polymorphism to present antigens uniquely characteristic of an infectious pathogen . Examples include presentation of short formylated peptides by mouse H2-M3 [7] , presentation of lipids and glycolipids by human group 1 CD1 ( CD1a–c ) molecules [8]–[11] , and the presentation of bacterial glycolipids by CD1d [12] , [13] . In some cases these nonclassically restricted T cells have been found at high frequency prior to pathogen exposure , suggesting an innate role . In our previous studies we have determined that human neonates have high frequencies of innate Mtb-reactive thymocytes that are not restricted by classical HLA-I molecules [14] . Functionally , such cells could either provide a direct role in the control of intracellular infection or could facilitate the acquisition of adaptive immunity . In humans , Mtb-reactive group 1 CD1 [15] and HLA–E restricted CD8+ T cells [16] have been described . We have observed that all individuals regardless of exposure to TB have CD8+ T cells capable of recognizing Mtb-infected cells [4] , [5] , [14] . Moreover , a proportion of these CD8+ T cells can be defined as nonclassically restricted [5] . Therefore , to define the relative contribution of classically versus nonclassically ( NC ) restricted CD8+ T cells , we used limiting dilution analysis ( LDA ) to characterize human , Mtb-specific CD8+ T cells in those with TB , those with latent TB infection ( LTBI ) , and those with no evidence of prior exposure to Mtb . We show that CD8+ T-cell clones from individuals infected with Mtb are primarily HLA-Ia restricted . In contrast , NC restricted CD8+ T-cell clones that are neither HLA-Ia nor CD1-restricted , predominate in Mtb-uninfected donors but are nevertheless present in all donors . Furthermore , we demonstrate that these NC restricted CD8+ T-cell clones are restricted by MHC-related molecule 1 ( MR1 ) , an HLA-Ib molecule that displays striking evolutionary conservation in mammals [17] . These human Mtb-reactive T cells recognize Mtb-infected dendritic cells ( DCs ) and lung epithelial cells . Moreover , we show that Mtb-reactive mucosal associated invariant T ( MAIT ) clones cross react with cells infected with other mycobacteria as well as other bacteria such as Escherichia coli , Salmonella typhimurium , and Staphylococcus aureus . These clones express the semi-invariant Vα7 . 2 T cell receptor , are activated in a manner independent of the transporter associated with antigen processing and presentation ( TAP ) , and have a mucosal homing phenotype . These phenotypic data lead us to designate these cells as MAIT cells [18] , [19] , a cell type with no previously known physiological function . Additionally , we demonstrate that infection with Mtb induces cell surface expression of MR1 on lung epithelium . Furthermore , Mtb-reactive MAIT cells are enriched in human lung and respond to Mtb-infected lung epithelial cells . Finally , we have performed direct ex vivo analysis of Mtb-reactive MR1-restricted MAIT cells , and find they are present at lower frequencies in the blood of those with active TB . These findings suggest that MAIT cells could play a direct role in the control of bacterial infection .
In humans , direct ex vivo analysis of Mtb-specific CD8+ T cells reveals a strong association of HLA-Ia–restricted responses and infection with Mtb [4] , [20] . Nonetheless , NC HLA-I–restricted CD8+ T cells comprise a substantial proportion of the overall response to Mtb in Mtb-infected individuals [4] , [5] , [20] . In individuals with no evidence of infection , we have consistently found high frequency CD8+ T cell responses against Mtb-infected DCs . To address the hypothesis that NC restricted CD8+ T cells comprise the dominant response in those without Mtb infection we performed LDA [5] using CD8+ T cells stimulated with Mtb-infected DCs . LDA was performed on individuals with no evidence of Mtb infection ( uninfected controls , n = 5 ) , individuals with evidence of latent infection with Mtb ( LTBI , n = 5 ) , and individuals with clinical TB ( active TB , n = 6 ) . From each of the 16 individuals , we screened an average of 128 clones per donor ( Table 1 ) for their ability to specifically release interferon-γ ( IFN-γ ) in response to a panel of Mtb-infected but not uninfected targets . The antigen presenting cell ( APC ) target groups were: autologous DCs , HLA-mismatched DCs , or HLA-mismatched macrophages . HLA-Ia restricted clones were defined as those responding only to Mtb-infected HLA matched DCs . DCs were grown in X-Vivo media to ensure expression of cell surface CD1 . NC-restricted clones were defined as those responding to all three Mtb-infected APC types . As macrophages do not express CD1 , NC CD1-restricted T clones were defined as those responding only to infected DCs . Using this method , we have not observed CD1-restricted T-cell clones , resulting in categorization of all the non-HLA-Ia–restricted clones as NC-restricted T cells . None of the T-cell clones were stimulated by uninfected HLA mismatched targets ruling out responses due to alloreactivity . The results from the LDA analysis are presented in Table 1 and Figure 1 . The proportion of NC-restricted T-cell clones from each group of donors is presented in Figure 1 . As expected , HLA-Ia–restricted CD8+ T-cell clones were strongly associated with TB ( p = 0 . 009 ) ( Figure 1 ) . Nonetheless , consistent with prior observation [5] , a significant proportion of CD8+ T-cell clones from infected individuals were NC-restricted . Furthermore , CD8+ T-cell clones isolated from uninfected donors predominantly displayed a NC phenotype ( Figure 1 ) . To facilitate further analysis a representative subset of the NC-restricted T-cell clones from each donor was expanded and further characterized . Phenotypic analysis of expanded clones ( n = 120 ) revealed uniform expression of CD8α and the αβ TCR ( unpublished data ) . Additionally , we excluded potential activation by a soluble mediator by successfully using Mtb-infected paraformaldehyde-fixed DCs as stimulators . As a result , we have isolated 120 stable NC Mtb-reactive CD8+ αβ TCR+ T-cell clones . To explain the high proportion of NC-restricted CD8+ T cells , we considered three hypotheses: presentation by an HLA-Ib molecule , natural killer ( NK ) -receptor mediated activation , and Toll-like receptor ( TLR ) -mediated activation . To exclude the possibility that TLR stimulation of DCs would be sufficient to activate the NC clones , we stimulated DCs with agonists to TLR2 ( lipoteichoic acid ) or to TLR4 ( lipopolysaccharide ) [14] , as both TLRs have been associated with Mtb infection [21] . TLR stimulation of DCs did not result in T-cell activation ( Figure 2A ) . To further evaluate the possibility that TLR2 and/or TLR4 stimulation was required for the recognition of Mtb-infected targets , antibody blockade was performed ( Figure 2B ) . Neither TLR2 nor TLR4 blockade prevented Mtb-dependent T-cell activation . However , the TLR2 and TLR4 antibodies blocked 100% and 80% of interleukin-6 ( IL-6 ) production by DCs treated with TLR2 and TLR4 agonists respectively ( unpublished data ) . NK cells do not utilize a TCR but instead are regulated through opposing signals triggered through inhibitory or activating receptors . Mtb is known to induce the cell surface expression of stress molecules such as ULBP1 [22] and MICA [23] , which are ligands for the activating NK receptor NKG2D . Antibody blockade of NKG2D/CD94 and the ligands ULBP1 and MICA did not alter recognition of Mtb-infected DCs by any of the T-cell clones ( Figure 2C ) . We next tested the hypothesis that an HLA-Ib molecule was restricting the NC CD8+ T-cell clones . We previously isolated Mtb-specific human CD8+ T-cell clones restricted by the molecule HLA–E [16] . Human Mtb-specific CD8+ T cells restricted by the HLA-Ib molecules CD1a , CD1b , and CD1c [24] have been extensively characterized . To assess if these HLA-Ib molecules were restricting the NC T cells , we performed antibody blockade experiments . We previously showed that addition of the pan HLA-I blocking antibody W6/32 effectively blocks the HLA–E restricted clone ( D160 1–23 ) [16] . While addition of W6/32 blocked recognition of Mtb-infected targets by the HLA–E-restricted clone , three NC-restricted Mtb-reactive CD8+ T-cell clones derived from different donors were unaffected ( Figure 2D ) . In addition to blocking all HLA-Ia molecules and HLA–E , W6/32 also blocks the HLA-Ib molecule HLA–G . As expected , the addition of blocking antibodies previously shown to block responses to CD1a , CD1b , CD1c , or CD1d also had no effect on the clones ( Figure 2D ) . We have extended these findings to all 120 NC CD8+ T-cell clones isolated from the 16 donors listed in Table 2 . None of the 120 clones were blocked by the addition of W6/32 or CD1 blocking antibodies ( unpublished data ) . These results suggest that neither CD1a , CD1b , CD1c , CD1d , HLA–E , HLA–G , nor HLA-Ia molecules restrict the panel of 120 Mtb-reactive CD8+ NC-restricted T-cell clones . These data suggest that a common Mtb-reactive CD8+ T subset is present in all individuals regardless of prior exposure to Mtb . We next postulated that the HLA-Ib molecule MR1 was the restricting allele for the NC clones . MR1 is a nonpolymorphic HLA-Ib molecule genetically linked with the CD1 locus in humans [25] and is the most evolutionarily conserved HLA–I molecule among mammals [26] . MR1 is required for the selection of a subset of T cells found primarily in the gut of mammals and thus named mucosal associated invariant T ( MAIT ) cells . The expansion of MAIT cells is dependent on the presence of gut flora suggesting that a bacterial derived or induced ligand is required for MR1-restricted T-cell expansion and activation [19] . Nevertheless , no bacterial or endogenous MR1-restricted antigen has been identified although considerable evidence supports an antigen presentation function by MR1 [27]–[29] . Furthermore , the biological role of MR1-restricted T cells is unknown , even though several parallels suggest a Natural Killer T- ( NKT ) cell–like regulatory role [30] , [31] . As demonstrated in Figure 2E , addition of an anti-MR1 blocking antibody ( 26 . 5 ) [28] prior to the addition of NC clones abolished IFN-γ production by three different Mtb-reactive NC CD8+ T-cell clones and an additional 11 clones listed in Table 2 . The addition of a different anti-MR1 blocking antibody ( 8F2 . F9 ) resulted in similar blocking ( unpublished data ) . In contrast , CD8+ T-cell clones restricted by HLA–E ( D160 1–23 ) or HLA-B08 ( D480F6 ) , or a CD4+ HLA-II restricted clone ( D454E12 ) were unaffected by the addition of the anti-MR1 blocking antibody ( Figure 2E ) . We next performed phenotypic analyses of Mtb-reactive MR1-restricted T-cell clones to determine if they shared properties of previously characterized MR1-restricted MAIT cells . We selected a subset of clones representative of TB exposed ( active , n = 5; LTBI , n = 4 ) and uninfected donors ( n = 5 ) ( Table 2 ) . One defining feature of both mouse and human MR1-restricted MAIT cells is the expression of a semi-invariant TCR Vα chain: Vα7 . 2/Jα33 for humans and the highly homologous Vα19/Jα33 for mice , respectively . Using an antibody that labels all T cells containing the Vα7 . 2 chain including those that pair with the Jα33 region [32] , the Vα7 . 2 chain was detected by flow cytometry on all 14 MR1-restricted Mtb-specific T-cell clones as well as on an HLA-B08–restricted clone , but not on the HLA–E , or HLA-II–restricted clones ( Figure 2F; Table 2 ) . Given that 14 of 14 randomly selected clones were restricted by MR1 , binomial analysis suggests a high prevalence of MR1 restriction among our panel of clones ( >95% ) . Furthermore , we performed an analysis of Vα7 . 2 TCR expression on an additional 28 NC clones . Here all 28 clones expressed the Vα7 . 2 TCR suggesting that the remaining clones are MR1-restricted . The TCR of Vα7 . 2-expressing MAIT cells from the gut has been associated with the expression of the Vβ2 or Vβ13 TCR β chains [31] . However , we found at least 10% of the Mtb-specific MR1-restricted T-cell clones did not express either Vβ2 or Vβ13 ( Table 2 and unpublished data ) . To determine if Vα7 . 2+ Mtb-reactive MR1-restricted T-cell clones expressed the canonical Vα7 . 2/Jα33 CDR3 region , the TCR alpha encoding cDNA was cloned from six representative Mtb-reactive clones chosen on the basis of their distinct patterns of Vβ TCR usage . All six T-cell clones were found to express the hAV72 segment as expected , and five of six expressed the hAJ33 segment ( Table 3 ) . Further , all six Mtb-reactive TCRs were found to have VJ junctional heterogeneity with two N additions , as previously reported for Vα7 . 2/Jα33 TCRs [33] , [34] . Importantly all six TCRs from Mtb-reactive T-cell clones were found to encode CDR3α loops of the same length , which is highly conserved among all mammalian Vα7 . 2/Jα33+ cells studied thus far . And finally , each of the sequences of the CDR3α loops of the five Vα7 . 2/Jα33+ Mtb-reactive T cells matched a sequence from a previously reported Vα7 . 2/Jα33+ cell of undefined restriction and antigen specificity ( Table 3 ) [33] , [34] . In humans , gut-derived MR1-restricted MAIT cells have been shown to express the CD8αα form of the CD8 co-receptor or lack CD4 or CD8 coreceptor expression [19] , [34] . In peripheral blood , invariant TCR Vα7 . 2+ T cells were originally identified from and found to be overrepresented in the CD4−CD8− fraction of T cells [33] . More recently , MAIT CD8αβ T cells have also been described [32] , [35] . As shown in Figure 2G ( and unpublished data ) , all of the Mtb-reactive MR1-restricted clones tested ( n = 14; Table 2 ) coexpressed CD8α and CD8β chains . Human and mouse MR1 lack residues associated with CD8 interaction [25] , such that the functional significance of coreceptor expression on Mtb-reactive MR1-restricted T cells remains to be determined . In a recent analysis of MAIT cells from blood , the canonical Vα7 . 2+ cells were associated with expression of the NK receptor CD161 [32] . We found that Mtb-reactive MR1-restricted T-cell clones cells varied in their CD161 expression ( Figure 2H; Table 2 ) , although all cells expressed the mucosal homing integrin α4β7 , CD45RO , and lacked CD45RA as previously described for MAIT cells ( unpublished data ) [32] . Prior work with mouse and human MR1-restricted MAIT cells has demonstrated that neither HLA-II nor TAP are required for thymic selection nor for antigen processing and presentation [18] , [34] . To determine if TAP transport is required for presentation to Mtb-reactive MAIT cells we used an adenoviral vector expressing the TAP inhibitor ICP47 ( Figure 3 ) [36] , [37] . When ICP47-expressing DCs were subsequently infected with Mtb , neither representative MR1-restricted clones , nor an HLA-II restricted clone were affected by TAP inhibition . In contrast , TAP blockade resulted in over 85% inhibition of the response by the CFP-103–11 HLA-B08-restricted CD8+ T-cell clone . Hence , human Mtb-reactive MR1-restricted T cells , like previously described MAIT cells , do not require TAP for antigen processing and presentation . To determine if an antigen from Mtb could stimulate the NC-restricted T-cell clones , we initially screened the panel of 120 stable NC clones for their ability to recognize autologous DCs loaded with the cell wall ( CW ) fraction from Mtb . In contrast to HLA-Ia restricted Mtb-specific T-cell clones , we found that all of the NC clones were stimulated by DCs loaded with Mtb CW ( unpublished data ) . To delineate the antigen recognized by clones we compared the ability of CW to culture filtrate protein ( CFP ) from Mtb strain H37Rv ( courtesy of K . Dobos ) to induce a response by a panel of NC-restricted CD8+ T-cell clones ( n = 21 and representative of the 16 donors ) . As expected , the CW fraction derived from Mtb induced robust responses by all the T-cell clones ( Figure 4A ) , whereas the CFP was not stimulatory . To determine if the presentation of Mtb CW was dependent on MR1 we performed antibody blockade . Figure 4B shows that the CW response by three distinct MR1-restricted CD8+ T-cell clones was dependent on MR1 . To further characterize the antigen associated with the CW fraction , we subjected the CW to a variety of treatments and tested the ability of the treated fractions to induce a response by the 21 NC-restricted CD8+ T-cell clones ( Figure 4C ) . We have found that delipidated CW ( dCW ) , compared to untreated CW , is strongly antigenic . To determine whether or not the MR1 antigen was proteinaceous , dCW was subjected to proteolytic digestion with a panel of proteases . With all but three NC T-cell clones , protease treatment of the dCW abrogated the antigenic activity . The mean and standard error for the respective treatment groups were: ( dCW , 298 . 7+/−31 . 72 ) ; ( subtilisin , 69 . 38+/−20 . 05 ) ; ( trypsin , 79 . 57+/−17 . 25 ) ; ( chymotrypsin , 74+/−17 . 93 ) ; ( pronase , 33 . 38+/−10 . 86 ) ; ( Glu-C , 113 . 5+/−27 . 48 ) . In each case the Dunn's Multiple Comparison test showed significant differences between dCW and each of the protease-treated fractions ( p<0 . 05 ) . To determine if Mtb-reactive MAIT cells were specific for Mtb we screened known MR1-restricted clones for their ability to recognize M . smegmatis and E . coli . We found that all clones recognized DCs infected with either M . smegmatis or E . coli ( unpublished data ) . Further evaluation of Mtb-reactive MAIT clones showed that neither adenovirus ( Figure 3 ) , nor vaccinia-infected cells ( unpublished data ) elicited a response by the clones . To further define the cross-reactivity of Mtb-reactive MAIT cells , three independent clones were tested for their ability to recognize DCs infected with S . typhimurium , S . aureus , and Listeria monocytogenes . As shown in Figure 4D , Mtb-reactive MAIT cells recognize DCs infected with S . typhimurim and S . aureus , but not L . monocytogenes . To confirm the lack of response by L . monocytogenes we infected cells at multiplicity of infection ( moi ) in excess of 60 and did not observe a response ( unpublished data ) . Research on TB has traditionally focused on myeloid derived APCs such as macrophages and DCs . However , TB has the capacity to infect a variety of other cell types including epithelial cells [38] . Moreover , Mtb DNA has been detected in a variety of cell types in the lung including epithelial cells [39] . Furthermore , HLA-Ib molecules are expressed on mucosal epithelial cells [40] . Because MAIT cells are located in the gut and lung mucosa we hypothesized that MR1-restricted T cells could play a role in the detection of Mtb in lung epithelium . As shown by flow cytometry ( Figure 5A ) and microscopy ( Figure 5B ) , Mtb infects the human lung epithelial cell line A549 [38] . Mtb-infection of A549 cells resulted in robust IFN-γ production by NC-restricted HLA-E ( Figure 5C ) and MR1-restricted T cells ( Figure 5D ) , in a manner that was dependent on the HLA-Ib molecules HLA-E and MR1 , respectively . Similarly to results shown with DCs , Mtb-infected A549 cells activated MR1-restricted , but not an HLA–E–restricted T cell [37] , independently of TAP ( Figure 5E ) . The ability of anti-MR1 antibodies to block recognition of the Mtb-infected cells implies that T-cell activation is occurring via the cell-surface expression of MR1 . Our ability to successfully block DC recognition by MR1-restricted T cells with antibody concentrations lower than 0 . 5 µg/ml ( unpublished data ) suggests that low levels of MR1 are sufficient for MAIT cell recognition . Although MR1 mRNA is ubiquitous in all human cell types [25] and MR1 protein detectable in all mouse tissues [18] , cell surface expression of MR1 has not been demonstrated . Mtb infection of A549 cells resulted in the detectable cell surface expression of MR1 while HLA-I expression was unaltered by infection with Mtb ( Figure 5F ) . We have not detected surface MR1 on DCs . We speculate that high levels of Fc receptor expression have made it difficult to discern low-level MR1 expression . To determine whether or not primary human lung epithelial cells could act as APCs to Mtb-reactive MR1-restricted T cells , we generated human primary large airway epithelial cells from tracheal brushings [41] . As shown in Figure 5G , Mtb-infected large airway epithelial cells elicited a robust response by the MR1-restricted clone D426B1 . Furthermore , this response was blocked by the addition of anti-MR1 blocking antibody ( 26 . 5 ) but not the isotype control ( Figure 5G ) . To determine if Mtb-reactive MR1-restricted T-cell responses are correlated with exposure to Mtb , we performed flow-cytometric ex vivo analyses of MR1-restricted , Mtb-reactive MAIT cells from subjects from all three Mtb exposure groups ( uninfected , n = 6; LTBI , n = 5 , active TB , n = 8 ) . To enumerate these cells ex vivo , Mtb-infected A549 cells were used as APCs . A549 were chosen as APCs because they do not produce tumor necrosis factor-α ( TNF-α ) in response to infection with Mtb , nor elicit an allogeneic response by polyclonal CD8+ T cells isolated from the periphery and lung ( Figures 6 , 7 , and unpublished data ) . Furthermore , in addition to IFN-γ , we have found that Mtb-reactive MAIT clones produced TNF-α in response to infected A549 cells in a manner dependent on MR1 ( unpublished data ) . To evaluate the nonclassical response to Mtb-infected APCs , we enriched CD3+ T cells from peripheral blood mononuclear cells ( PBMCs ) by negative selection and then depleted CD4+ T cells . The remaining CD8+ and CD4−CD8− T cells , as the source of responding T cells , were incubated overnight with Mtb-infected A549 cells in the presence or absence of MR1-blocking antibody . Cells were surface stained to detect the Vα7 . 2 TCR and the CD8α coreceptor , and intracellular staining was performed to detect TNF-α production . We found that Mtb-reactive MAIT cells were uniquely CD8 single positive and CD161 negative . Therefore , subsequently , T cells were selected on the basis of forward and side scatter distribution and then selected on CD8 ( Figure 6A ) , such that the numbers presented in Figure 6A–6E represent frequencies of total CD8+ cells . Figure 6B shows a representative analysis of Mtb-dependent TNF-α production from Vα7 . 2+ and Vα7 . 2− CD8+ T cells . Mtb-infected A549 cells , but not uninfected A549 cells , induced TNF-α production by CD8+ T cells ( Figure 6B and 6C ) . Although , Mtb-reactive responses could be detected in both the Vα7 . 2+ and negative subsets only Vα7 . 2+ CD8+ T cells were blocked by the addition of the anti-MR1 antibody ( Figure 6B and 6C ) , consistent with the observation that MR1-restricted cells express the Vα7 . 2 TCR . On average , addition of anti-MR1 resulted in a 37% reduction in TNF-α production by Vα7 . 2+ T cells , whereas no blocking was observed from the Vα7 . 2− T cells ( p<0 . 0001 ) . We note that in all donors a proportion of Vα7 . 2− cells expressed the γδTCR and were activated by Mtb-infected APCs as expected from studies performed by De Libero et al . ( unpublished data ) [42] . Of the 19 donors , four shared at least one HLA-Ia allele with A549 cells [43] . However , three of these were from the uninfected group that were previously shown to have no detectable Mtb-specific HLA-Ia restricted T-cell responses [4] . To define the relationship of Mtb-reactive MAIT cells and host infection status , we determined the frequency of CD8+ T cells that expressed the Vα7 . 2 TCR as well as the frequency of Mtb-reactive Vα7 . 2+ T cells recognizing Mtb-infected APCs in an MR1-dependent manner ( Figure 6D ) . The frequency of Vα7 . 2+ cells ranged from 0 . 039% to 10 . 5% of CD8+ T cells demonstrating considerable heterogeneity among these 19 donors . Nonetheless , a lower proportion of Vα7 . 2+ cells was present in those with active TB compared to LTBI . Figure 6E demonstrates the frequency CD8+ T cells that were Vα7 . 2+ MR1-dependent responses analyzed by donor infection status . When compared with uninfected subjects ( mean = 0 . 092 ) , those with active TB ( mean = 0 . 011 ) had markedly diminished responses ( p = 0 . 0025 ) , whereas comparison with those with LTBI ( mean = 0 . 185 ) revealed a less dramatic decrease ( p = 0 . 0611 ) . While MAIT cells have been reported in the gut lamina propria , they have also been described in the lung [19] . To test the hypothesis that MAIT cells resident in the human lung would be reactive to Mtb , the lung and adjacent lymph node ( LN ) were obtained from two individual organ donors whose lungs were not suitable for transplantation . Single cell suspensions from the LN and lung parenchyma were prepared and T cells were then enriched via magnetic bead positive selection . Figure 7 represents the results from intracellular cytokine staining assays performed as described in Figure 6 . The frequencies of Vα7 . 2+ CD8+ cells in the lungs were similar to those detected in blood . However , the proportion of Vα7 . 2+ cells producing TNF-α in response to Mtb-infected APCs was notably higher ( donor A , 13%; donor B , 22% ) than that seen in the adjacent LN or from the blood of the 19 donors described in Figure 6 ( range , 0%–10 . 7%; mean , 3 . 03; median , 2 . 26 ) . These data therefore demonstrate that a substantial proportion of lung-resident MAIT cells are Mtb-reactive , and that these cells are enriched relative to the adjacent LN and peripheral blood .
Humans both infected and uninfected with Mtb have high frequencies of Mtb-reactive CD8+ T cells [4] , [5] . In this study we demonstrate , by both LDA and direct ex vivo analysis that NC-restricted T cells predominate in TB-uninfected individuals . Moreover , our data suggest that MR1-restricted T cells make up a substantial proportion of the Mtb-reactive CD8+ NC-restricted T-cell response . Furthermore , we demonstrate that a large panel of Mtb-reactive MAIT clones are broadly reactive with mycobacterial species as well as E . coli . Using LDA , we find that MR1-specific Mtb-reactive clones predominate in individuals without infection with Mtb . Our previous experience with LDA cloning has resulted in the identification of immunodominant HLA-Ia antigens and epitopes [4] . However , LDA cannot distinguish a proportionate versus absolute reduction in NC responses in individuals with active TB . Here , direct ex vivo analysis confirmed a dramatic reduction in the absolute frequency of Mtb-reactive MAIT cells in individuals with active TB . CD8+ T cells restricted by both CD1 and HLA-E have been previously described but were noticeably absent from the LDA and from the resulting CD8+ T-cell clones . In this regard , it is possible that LDA based on positive selection of CD8+ T cells in conjunction with Mtb-infected DCs is not optimal for the selection of these cells . For example , CD1 molecules are downregulated by infection with Mtb [44] . Furthermore , CD1-restricted T cells may be less frequent in the CD8+ population [15] . Similarly , we did not isolate γδT cells [42] . As a result , the relative contribution of different HLA-Ib molecules in the host response to infection with Mtb remains to be determined . Our findings highlight differences between MAIT cells originally characterized phenotypically , on the basis solely of TCR usage versus those characterized functionally on the basis of reactivity with Mtb . MAIT cells , as defined by the expression of the canonical Vα7 . 2/Jα33 TCR , appear to be universally present in humans and range in frequencies from 1% to 4% of peripheral blood T cells [32] . Here , we confirm that MR1-dependent , Mtb-reactive MAIT cells are present in the Vα7 . 2+ but not the Vα7 . 2− population . However , Mtb-reactive MAIT cells represent a relatively small proportion of cells previously defined as MAIT cells . The observation that Mtb-reactive MAIT cells are found exclusively in the CD8 single positive CD161 negative population further defines Mtb-reactive MAIT cells as a subset of all MAIT cells . Our studies do not allow us to distinguish whether or not Mtb-reactive MAIT cells are representative of a uniform population of bacterially reactive MAIT cells or if alternate bacterial specificities exist in the Vα7 . 2 population . Detailed genotypic characterization of several Mtb-reactive MR1-restricted T-cell clones reveals diversity of TCR usage . Of the six Mtb-reactive Vα7 . 2+ T cells from which the TCRα sequence was characterized , one does not express the canonical Vα7 . 2/Jα33 segment . Furthermore , at least 10% of the Mtb-reactive , MR1-restricted T-cell clones do not express Vβ chains previously found in preferential association with Vα7 . 2/Jα33 expressing PBMC of undefined restriction and antigen specificity . Whether or not differences in the TCR reflect antigenic discrimination remains to be addressed . We speculate that environmental bacteria play a role in the selection and maintenance of human MAIT cells analogous to results previously shown in the mouse model [19] . In this regard , TCR heterogeneity could be the result of antigenic selection . We have found that DCs infected with viable mycobacteria ( M . smegmatis and M . bovis bacille Calmette-Guérin [BCG] ) can stimulate MR1-restricted Mtb-reactive T-cell clones ( unpublished data ) . Environmental mycobacteria are ubiquitous and therefore may affect MAIT cell selection and maintenance . Nonetheless , we find that cells infected with nonmycobacterial microorganisms such as E . coli , S . typhimurium , and S . aureus , also can elicit a response by the Mtb-reactive MR1-restricted CD8+ T-cell clones tested thus far . At present , the molecular basis for this cross-reactivity is not known . However , recent studies using limited numbers of mouse MAIT cell hybridomas have implicated antigen presentation in MR1-restricted MAIT cell activation . This conclusion was supported by the fact that an acid eluate of purified mouse MR1 enhanced MAIT cell activation in an MR1-restricted manner [29] . Importantly , these results were obtained using uninfected cells , suggesting presentation of an endogenous antigen . Thus it is possible that MAIT cell detection of cells infected with various bacteria results from the presentation of an endogenous MR1 ligand induced by infection with various bacteria . Alternatively , it is possible that MR1 presents an exogenous antigen shared by bacteria . However , we note that cell lines overexpressing human MR1 ( courtesy Ted Hansen , unpublished data ) do not stimulate Mtb-reactive MAIT cells . Based on analogy with iNKT cells , there is a precedent for recognition of either exogenous antigen or endogenous ligands by CD1d-restricted T cells [45] . However , regardless of whether an endogenous or exogenous antigen is being presented by MR1 , it seems likely that MAIT cells have only a limited ability to discriminate ligands bound to MR1 . Indeed the high level of activation of mouse or human MAIT cells by MR1 of different mammalian species is highly suggestive that all three components ( ligand , MR1 , and MAIT TCR ) were highly conserved in evolution [29] . Again from analogy with CD1d-restricted presentation to iNKT cells , recent structural studies suggest they also have only limited antigen discrimination [46] , [47] . The identification of physiological MR1 ligands and how they are detected by MAIT cells will clearly benefit from further studies of the extensive panel of human Mtb reactive T cells reported here . We demonstrate here that infection with Mtb results in a modest induction of surface expression of MR1 on epithelial cells . Failure of past studies to detect surface expression of endogenous MR1 is enigmatic , since MR1 message and ER luminal MR1 protein is ubiquitously expressed in different tissues [29] . Based on these observations , it is attractive to speculate that constitutive expression of MR1 may be deleterious because of inappropriate MAIT cell activation . Such a model would be consistent with studies of induced expression of MICA/B [48] . In any case , it is also clear that very little MR1 is likely sufficient for MAIT cell activation based on our antibody blocking and cytofluorometric studies . Alternatively , it is possible that bacterial infection may alter the intracellular trafficking of MR1 and consequently determine which self or bacterial antigens are loaded and presented at the cell surface to MAIT cells . In this regard , it is important to note that in both our studies here and previous mouse studies MAIT cells are activated in a TAP-independent manner . Indeed in the mouse studies trafficking of MR1 to endosomal compartments enhanced MAIT cell activation [18] . In combination , these findings support the model that MR1 trafficking and ligand acquisition are likely altered by bacterial infection . A surprising finding of this study has been the observation that primary human large airway epithelial cells infected with Mtb can induce a robust response by MR1-restricted MAIT cells . Following inhalation , Mtb is far more likely to encounter airway epithelium than alveolar macrophages . As a result , the capacity of lung resident MAIT cells to respond directly ex vivo to Mtb-infected lung epithelial cells suggests these cells could play a physiological role in directly controlling Mtb in the lung early in infection . Mtb-reactive MAIT cells not only produced IFN-γ but also TNF-α and granzyme ( unpublished data ) in response to infected targets . These effector functions could directly inhibit mycobacterial growth . Mtb-reactive MAIT cells , by IFN-γ conditioning of DCs , could also facilitate optimal priming of Mtb-specific CD8+ and CD4+ Th1 responses that are essential to control the disease in TB-exposed individuals . Ex vivo analyses of circulating , MR1-restricted , Mtb-reactive MAIT cells demonstrate that subjects with active TB have substantially lower frequencies than those without evidence of infection with Mtb . These data suggest that Mtb-reactive MAIT cells participate in the host response to infection with Mtb . It is also possible that a similar observation would be made in bacterial pneumonia . With regard to Mtb the precise role of MAIT cells remains to be determined . We found that individuals with active TB had reduced Mtb-reactive MAIT cells . One explanation may be a genetic predisposition towards lesser expression of MR1 and/or diminished capacity to process and present bacterially derived ligands . For example , the very limited polymorphisms noted in the MR1 gene , 28 single nucleotide polymorphisms ( SNPs ) over a region of more than 2 MB , might allow for the delineation of SNPs associated with disease susceptibility . Alternately , it is possible that mycobacterial exposure can elicit and maintain Mtb-reactive MAIT cells . In this regard , it would be interesting to delineate the effect of BCG and/or environmental mycobacterial exposures to the prevalence of Mtb-reactive MAIT cells . Conversely , it is possible that the diminished frequencies reflect either selective migration of Mtb-reactive MAIT cells to disease sites , or their selective depletion through activation-induced cell death . In conclusion , we have demonstrated that MAIT cells , with no previously known in vivo function , recognize bacterially infected cells . Furthermore , we demonstrate an association with Mtb exposure and/or disease status and the prevalence of Mtb-reactive MAIT cells ex vivo . Given these findings and the observation that MAIT cells are broadly reactive to bacterial infection , we postulate that MAIT cells likely play a role in the direct control of bacterial infection and/or in the subsequent acquisition of adaptive immunity to bacterial infections . By virtue of their prevalence , location , and effector functions , MAIT cells are poised to play a significant role in the control of bacterial infection .
Study participants , protocols , and consent forms were approved by the Oregon Health & Science University institutional review board . Informed consent was obtained from all participants . Uninfected individuals and individuals with LTBI were recruited from employees at Oregon Health & Science University as previously described [5] . Uninfected individuals were defined as healthy individuals with a negative tuberculin skin test and no known risk factors for infection with Mtb . Individuals with LTBI were defined as healthy persons with a positive tuberculin skin test , and no symptoms and signs of active TB . Individuals with active TB were recruited via institutional review board-approved advertisement and were self-referred from the Multnomah County TB Clinic , Portland , Oregon , US , or from the Washington County TB Clinic , Hillsboro , Oregon , US . In all active TB cases , pulmonary TB was diagnosed by the TB controller of these counties and confirmed by positive sputum culture for Mtb . Those with active TB were under the care of the local TB controller . At the time of apheresis , subjects were required to be smear and culture negative . PBMCs were isolated from whole blood obtained by venipuncture or apheresis . De-identified lung and LNs were obtained from the Pacific Northwest Transplant Bank ( PNTB ) . The H37Rv strain of M . tuberculosis was used for all live Mtb infections ( ATCC ) , prepared as previously described [20] , and infected at moi of 30 unless stated otherwise [3] . H37RvDsRED Mtb was kindly provided by David Sherman . Fractions of the Mtb CW were obtained from K . Dobos ( Mycobacteria Research Laboratories at Colorado State University , Fort Collins ) . Adenoviral vectors [36] were kindly provided by David Johnson ( OHSU ) . To generate delipidated Mtb CW ( dCW ) 1 g of lyophilized Mtb CW was extracted at 22°C with agitation twice for 2 h with chloroform∶methanol ( 2∶1 v/v ) ( 30 ml/g of CW ) followed by one 18-h extraction . The 2∶1 extracted CW material was collected ( 27 , 000 g for 30 min ) and dried under N2 and further extracted twice for 2 h followed by one 18-h extraction with chloroform∶methanol∶water ( 10∶10∶3 v/v/v ) . The resulting delipidated cell wall was dried under N2 , resuspended in PBS ( pH 7 . 4 ) , and protein concentration determined by BCA assay ( Pierce ) [49] . S . typhimurium and L . monocytogenes were kindly provided by Fred Heffron and David Hinrichs , respectively . S . aureus was obtained from ATCC . A549 cells were obtained from ATCC ( CCL-185 ) . Primary large airway lung epithelial cells were derived from the trachea as previously described [41] . Limiting dilution cloning methodology was performed as previously described with minor modifications [5] . DCs generated for use in cloning and screening T-cell clones were prepared as above with the exception that X-Vivo medium was used ( BioWhittaker ) . Macrophages were generated using a monocyte-isolation kit ( Miltenyi ) and then grown for 5 d in IMDM ( Invitrogen ) serum-free medium . When prepared in X-Vivo medium , DCs were CD1a positive and CD14 negative , whereas macrophages grown in IMDM were CD14 positive and CD1a negative . To generate T-cell clones DCs were infected with Mtb ( moi 30 ) overnight . CD8+ T cells were sorted to high purity ( >99% ) by FACS and were added over a range of dilutions to the infected DCs ( 20 , 000/well ) in the presence of irradiated autologous feeder PBMC ( 1−e5/well ) and rhIL-2 ( 10 ng/ml ) . T cells were screened by ELISPOT 10–14 d later . All donors from which APCs and T cells were used in these assays were genetically haplotyped ( Blood System Laboratory ) , thereby ensuring a complete mismatch of HLA-Ia alleles when necessary for screening . T-cell clones that retained Mtb specificity were subsequently expanded in the presence of irradiated allogeneic PBMC ( 25×106 ) , irradiated allogeneic lymphoblastoid cell line ( 5×106 ) , and anti-CD3 mAb ( 30 ng/ml ) in RPMI 1640 media with 10% HS in a T-25 upright flask in a total volume of 30 ml . The cultures were supplemented with IL-2 ( 0 . 5 ng/ml ) on days 1 , 4 , 7 , and 10 of culture . The cell cultures were washed on day 4 to remove remaining soluble anti-CD3 mAb [51] and used no earlier than day 11 . Antibodies to the following molecules were used: CD1a , CD1b , CD8a CD161 TCR αβ ( BD Pharmingen ) ; CD1c ( MCA694 ) , pan HLA-I antibody ( W6/32 ) ( Serotec ) ; anti-TNF-α ( Beckman Coulter ) ; TCR Vβ usage was determined using the IOTest Beta Mark Kit ( Beckman Coulter ) ; TCRgd ( 5A6 . E9-Endogen ) ; CD1d ( CD1d51 , kindly provided by Steven Porcelli ) ; CD8β ( GenWay ) ; CD49d ( 9F10 ) , LEAF ms IgG1 , LEAF msIgG2a , Integrin B7 ( FIB504 ) ( Biolegend ) ; MR1 ( 26 . 5 ) [28]; Va7 . 2 ( 3C10 ) [32]; CD94 ( MAB1058 ) NKG2D ( MAB139 ) , ULBP1 ( MAB1380 ) , MICA ( MAB1300 ) TLR2 ( MAB2616 ) , TLR ( AF1478 ) ( R&D ) , Lamp1 ( H5G11 , SCBT ) ; Tubulin ( E1332Y , Abcam ) ; TLR agonists: Lipoteichoic Acid ( Sigma ) ; LPS ( Sigma ) , Pam3CysK4 ( InVivo Gen ) ; Fluoromount G ( Southern Biotech ) . | About one-third of the world's population is infected with Mycobacterium tuberculosis ( Mtb ) , yet thanks to a robust immune response most infected people remain healthy . CD8 T cells are unique in detecting intracellular infections . Surprisingly , Mtb-reactive CD8 T cells are found in humans with no prior exposure to Mtb . We show that mucosal associated invariant T ( MAIT ) cells , which have no previously known in vivo function , make up a proportion of these Mtb-reactive CD8 T cells and detect Mtb-infected cells via a specific major histocompatibility molecule called MHC-related molecule 1 , which is evolutionarily conserved among mammals . Mtb-reactive MAIT cells are enriched in lung and detect primary Mtb-infected lung epithelial cells from the airway where initial exposure to Mtb occurs . We go on to show that MAIT cells are not specific for Mtb since they can detect cells infected with a variety of other bacteria . Curiously , Mtb-reactive MAIT cells are absent in the blood of individuals with active tuberculosis . We postulate that MAIT cells are innate detectors of bacterial infection poised to play a role in control of intracellular infection . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"immunology/antigen",
"processing",
"and",
"recognition",
"microbiology/immunity",
"to",
"infections",
"respiratory",
"medicine/respiratory",
"infections",
"immunology/immune",
"response",
"microbiology/innate",
"immunity",
"immunology/innate",
"immunity",
"immunology",
"infectious",
"diseases/respiratory",
"infections",
"infectious",
"diseases/bacterial",
"infections",
"immunology/immunity",
"to",
"infections"
] | 2010 | Human Mucosal Associated Invariant T Cells Detect Bacterially Infected Cells |
Although heterotypic secondary infection with dengue virus ( DENV ) is associated with severe disease , the majority of secondary infections are mild or asymptomatic . The mechanisms of antibody-mediated protection are poorly understood . In 2010 , 108 DENV3-positive cases were enrolled in a pediatric hospital-based study in Managua , Nicaragua , with 61 primary and 47 secondary infections . We analyzed DENV-specific neutralization titers ( NT50 ) , IgM and IgG avidity , and antibody titer in serum samples collected during acute and convalescent phases and 3 , 6 , and 18 months post-infection . NT50 titers peaked at convalescence and decreased thereafter . IgG avidity to DENV3 significantly increased between convalescent and 3-month time-points in primary DENV infections and between the acute and convalescent phase in secondary DENV infections . While avidity to DENV2 , a likely previous infecting serotype , was initially higher than avidity to DENV3 in secondary DENV infections , the opposite relation was observed 3–18 months post-infection . We found significant correlations between IgM avidity and NT50 in acute primary cases and between IgG avidity and NT50 in secondary DENV infections . In summary , our findings indicate that IgM antibodies likely play a role in early control of DENV infections . IgG serum avidity to DENV , analyzed for the first time in longitudinal samples , switches from targeting mainly cross-reactive serotype ( s ) to the current infecting serotype over time . Finally , serum avidity correlates with neutralization capacity .
The four serotypes of the flavivirus dengue virus ( DENV1–4 ) cause the most common mosquito-borne viral disease in humans worldwide , with 50–100 million people infected annually and over 3 billion people at risk [1] . DENV infection can be asymptomatic or cause a spectrum of disease ranging from classical dengue fever ( DF ) to more severe , life-threatening forms termed dengue hemorrhagic fever ( DHF ) and dengue shock syndrome ( DSS ) [2] . Approximately 500 , 000 dengue patients require hospitalization annually , of whom a large proportion are children [3] . Although several antiviral and vaccine candidates are in various phases of preclinical and clinical evaluation , current treatment remains supportive care [4] . The immune response to primary ( 1° ) DENV infection is characterized by an early IgM response followed by an IgG response with predominantly IgG1 and IgG3 subclasses [5] . Naïve B cells are stimulated and develop into DENV-specific B cells , which either differentiate into memory B cells ( MBCs ) residing in the secondary lymphoid organs or into plasma cells ( PCs ) secreting antigen-specific antibodies ( Abs ) . Short-lived PCs are active during acute infection , while long-lived PCs ( LLPCs ) migrate to the bone marrow and are responsible for long-term humoral immunity [6] , [7] . MBCs , which retain antigen-specific Abs at their surface , and LLPCs , which secrete antigen-specific Abs , undergo affinity maturation , and only clones bearing Abs with the highest affinity survive long-term [8] . This process takes several weeks after acute infection and continues despite the absence of circulating antigen . During a secondary ( 2° ) DENV infection , MBCs are rapidly activated [9] , [10] . Prior DENV infection provides robust immunity against the homotypic DENV serotype [11] , [12] . In contrast , 2° heterotypic infections are associated with a higher incidence of DHF/DSS , possibly attributable in part to antibody-dependent enhancement ( ADE ) , where pre-formed Abs to the 1° infecting serotype bind but do not neutralize the 2° infecting serotype , instead facilitating an increase in viral uptake by Fcγ-receptor bearing cells [13] . In addition to ADE , cross-reactive T cells formed during the 1° DENV infection can be over-activated , potentially contributing to dengue pathogenesis [14] , [15] . However , the vast majority of 2° DENV infections are asymptomatic or only result in mild disease [16] , suggesting a protective immune response [17] . DENV neutralization requires sufficient levels of neutralizing Abs and the number of Abs bound to a single virion to exceed the threshold of enhancement , which depends on antibody avidity and the accessibility of epitopes on the virus particle [18] . The avidity of anti-flavivirus monoclonal Abs ( MAbs ) was shown to positively correlate with neutralization activity in vitro [19] , consistent with lower neutralizing activity observed with MAbs with lower affinity against variant genotypes within a single DENV serotype [20] , [21] . However , the relation between neutralizing activity and polyclonal serum avidity is still unclear [9] , [22] . Measurement of IgG avidity has been shown to discriminate between 1° and 2° DENV infection in polyclonal serum [23]–[25] but has not been followed longitudinally . In this study , we analyzed the DENV-specific neutralization capacity and IgM and IgG avidity to DENV of serum samples from our hospital-based study in Managua , Nicaragua . In Nicaragua , one DENV serotype tends to dominate for several years , while other DENV serotypes co-circulate at lower levels . DENV3 was the dominant serotype in 2008–2011 [26] . Prior to this , DENV2 was the predominant serotype between 1999 and 2002 and between 2005 and 2007 [27]–[29] , while DENV1 predominated between 2002 and 2005 [30] . DENV4 only circulates at a low level in Nicaragua [31] . We studied both 1° and 2° DENV infections from the acute phase until 18 months ( m ) post-infection . In 2° infections , we measured neutralizing Ab titers and avidity to DENV3 , the currently infecting serotype , and to DENV2 , the most prevalent previously circulating serotype and our prototype to assess cross-reactive responses and cross-protection [9] , [28] . As both Ab concentration and avidity can play a significant role in virus neutralization [32] , we tested possible correlations between IgM and IgG avidity to DENV , total Ab titer and DENV-specific neutralization titer .
The protocol for this study was reviewed and approved by the Institutional Review Boards ( IRB ) of the University of California , Berkeley , and the Nicaraguan Ministry of Health . Parents or legal guardians of all subjects provided written informed consent , and subjects 6 years of age and older provided assent . Design and execution of the study , inclusion criteria for the study population , and laboratory tests for confirmation of DENV infection in patients have been previously described [9] , [33] . Briefly , study enrollment occurred in the Nicaraguan National Pediatric Reference Hospital , Hospital Infantil Manuel de Jesús Rivera ( HIMJR ) , in Managua from August 1 , 2010 , to January 31 , 2011 , during the peak dengue season . Inclusion criteria included age between 6 m and 15 years of age . Samples were collected for 3 consecutive days after enrollment ( acute ) , 14–28 days after onset of symptoms ( convalescent ) , and at 3 m , 6 m and 18 m post-illness . DENV infection was confirmed by reverse transcription–polymerase chain reaction ( RT-PCR ) amplification of viral RNA [34]; isolation of DENV in C6/36 Aedes albopictus cells [27]; seroconversion of DENV-specific IgM Abs as measured by IgM capture enzyme-linked immunosorbent assay ( ELISA ) between acute-phase and convalescent-phase serum samples [35]; and/or a ≥4-fold increase in total Ab titer , as measured by Inhibition ELISA [30] , [36] , between paired acute- and convalescent-phase serum samples as previously described [26] . 1° and 2° DENV infections were defined by an Ab titer by Inhibition ELISA of <10 or ≥10 in acute-phase samples , respectively , and/or <2 , 560 or ≥2 , 560 in convalescent phase samples , respectively [31] , [37] . The total Ab titer was measured only during the acute and convalescent phases of the infection , as it decreases substantially thereafter . DENV was propagated in Aedes albopictus C6/36 cells as previously described [9] . Cell supernatants were concentrated by centrifugation through Amicon filters ( 50 kDa , 3 , 250×g for 20 min at 4°C ) or by ultracentrifugation ( 90 , 000×g for 2 h at 4°C , Beckman SW28 ) and resuspended in PBS . DENV2 ( strain N172 , passage 3 ) and DENV3 ( strain N7236 , passage 3 ) are clinical strains from Nicaraguan patients isolated in the National Virology Laboratory in Managua and passaged minimally . Raji-DC-SIGN-R cells ( gift from B . Doranz , Integral Molecular , Philadelphia , PA ) were grown in RPMI-1640 medium ( Invitrogen ) with 5% FBS at 37°C in 5% CO2 for use in neutralization assays [38] , [39] . Serum samples were heat-inactivated at 56°C for 20 min and then diluted in RPMI-1640 with 10% FBS at pH 8 . 0 using eight 3-fold dilutions ( 1∶10–1∶21 , 870 ) . Neutralization was assessed by flow cytometry using GFP-expressing DENV reporter virus particles ( RVPs ) as previously described [38] , [39] . The percent infection for each serum dilution was calculated in relation to the no-serum control . Data were expressed as percent infection versus log10 of the reciprocal serum dilution and fitted to a sigmoidal dose-response curve using GraphPad Prism 5 software ( La Jolla , CA ) to determine the titer of Ab that achieved a 50% reduction in infection ( 50% neutralization titer , NT50 ) , which is expressed as the reciprocal of the serum dilution . Serum avidity was measured using a modified ELISA protocol with urea washes [22] , [23] . DENV3 N7236 was used at a 1∶300 dilution ( 3 . 4×104 pfu/ml ) , which yielded an OD of 1 . 0 using WHO polyvalent serum . To standardize the amount of DENV2 N172 to the amount of DENV3 , we used a pan-DENV MAb , 2H12 ( gift from G . Screaton , Imperial College , UK ) [40] . Serial dilutions of DENV2 and the 1∶300 dilution of DENV3 were coated and incubated with 1 µg/mL of 2H12 MAb . The 1∶300 dilution of DENV2 ( 3 . 6×104 pfu/ml ) yielded the same OD as the 1∶300 DENV3 and was used thereafter . To determine serum IgG avidity , plates were coated with whole virus and incubated with heat-inactivated diluted serum samples ( 1∶100 ) in triplicate , then washed with urea or PBS for 10 min before adding the secondary biotin-conjugated Ab ( donkey anti-human IgG , 1∶1 , 000 ) , streptavidin-AP conjugate ( 1∶1 , 000 ) , and PnPP substrate [9] . Using samples from the same study population , we previously optimized the amount of urea to be used , and 6M urea for 1° DENV infections and 9M urea for 2° DENV infections yielded the best results for analyzing serum avidity to DENV over time [9] . To measure serum IgM avidity , a donkey anti-human IgM , Fc-fragment-specific MAb ( 1∶1 , 000 , Jackson ImmunoResearch ) was used as secondary Ab . For acute-phase samples , days 4–6 post-onset of symptoms were chosen . Background levels were measured with dengue-negative human sera . For each sample , avidity was calculated as percentage of IgG or IgM bound by dividing the background-adjusted OD after urea washes by the adjusted OD after PBS washes . Quality control criteria included: background <0 . 2 OD , WHO polyvalent serum positive control >5X background OD , and WHO positive control of each plate within the mean +/−1 SD of control plate . The control plate was coated with either DENV2 or DENV3 and incubated with WHO polyvalent serum; half the plate was washed with urea ( 6M or 9M ) , while the second half was washed with PBS . The mean and SD for the WHO polyvalent positive control was calculated for each urea concentration and each virus as follows: The data were stratified by 1° and 2° infection status for analysis , and gender was evaluated as a possible modifier . Geometric mean total Ab titer , percentage avidity , and NT50 were compared using the two-sided Wilcoxon Signed Rank test to detect differences between the following time-points: acute and convalescent; convalescent and 3 m; 3 m and 6 m; and 6 m and 18 m . Geometric mean Ab titer , avidity , and NT50 were also compared by infection status at each time-point using the Wilcoxon Signed Rank test . Bivariate correlations between NT50 and Ab titer or NT50 and avidity were estimated using the Spearman correlation coefficient at each time-point . To test for differences in avidity , NT50 , and Ab titer by clinical signs of dengue severity ( vascular leak , hypotensive shock , compensated shock , cutaneous bleeding , hemoconcentration and mucosal bleeding [37] ) , two-sided Wilcoxon Signed Rank tests were used to compare the geometric mean NT50 , total Ab titer and avidity at each time-point , stratifying by immune status . An alpha of p<0 . 05 was used for statistical significance testing . Calculations were performed in SAS Version 9 . 2 ( The SAS Institute , Cary , NC ) .
Between August 1 , 2010 , and January 31 , 2011 , 216 patients were enrolled for suspected dengue at the National Pediatric Reference Hospital , HIJMR . Twelve patients were excluded from analysis: one patient dropped out of the study after enrollment and 11 patients had an undetermined dengue diagnostic result . Of the 204 patients who were followed , 108 patients ( 52 . 9% ) were laboratory-confirmed as DENV3-positive by RT-PCR and/or virus isolation and were included in this analysis ( Table S1 ) . Of these , 61 ( 56 . 5% ) were 1° and 47 ( 43 . 5% ) were 2° DENV infections ( Table 1 ) . Of note , disease severity was relatively low in the 2010–2011 season , with 27 ( 25 . 0% ) DHF/DSS cases ( Table 1 ) [2] . In the absence of a pre-infection sample , and while we cannot exclude previous DENV1 infections , we hypothesized that most children with 2° DENV infections were previously infected with DENV2 , which was the predominant serotype circulating in previous years [28] , [29] . Thus , we used DENV2 as a representative previously infecting serotype in our analysis . We measured the DENV-specific neutralization capacity of patient sera from 1° and 2° DENV3 infections against DENV3 , the current infecting serotype , and against DENV2 , a representative previously infecting serotype , in 2° infections . Using a flow cytometry-based assay , NT50 was determined at acute , convalescent , 3 m , 6 m , and 18 m time-points ( Table S2 ) . The NT50 against DENV3 peaked at convalescence in both 1° and 2° cases , with significantly higher titers detected in 2° cases ( Figure 1A–B ) . In 1° infections , we observed a significant increase in NT50 from the acute to convalescent phase ( p<0 . 001 ) followed by a significant drop from convalescence to 3 m post-onset of symptoms ( p<0 . 0001 ) , further decreasing at 6 m ( p<0 . 0001 ) ( Figure 1A ) . For 2° infections , we noted a similar pattern ( p<0 . 0001 for each pair of adjacent time-points ) ( Figure 1B ) . In addition , for 2° infections , we measured the DENV2-specific neutralization capacity of the serum , which showed a similar trend ( Figure 1C ) . We measured DENV-specific IgG serum avidity to DENV3 and DENV2 and DENV-specific IgM serum avidity to DENV3 using a modified ELISA with urea washes [22] , [23] . Avidity was defined as the percentage ( % ) of IgG or IgM that remained bound after the urea washes . In 1° DENV3 infections , we observed a large increase in IgG serum avidity to DENV3 between convalescence and 3 m post-onset of symptoms ( p<0 . 0001 ) and a further rise between 3 and 6 months ( p = 0 . 017 ) and between 6 and 18 months ( p = 0 . 003 ) post-onset of symptoms ( Figure 2A , Table S2 ) . IgM serum avidity to DENV3 in 1° DENV3 acute samples revealed a mean % IgM bound of 47 . 9 ( SD±12 . 3 ) . In 2° DENV infections , we observed a significant increase in IgG serum avidity to DENV3 between the acute and convalescent phase ( p<0 . 0001 ) , but no further rise at later time-points ( Figure 2B , Table S2 ) . Measuring the avidity of the same 2° infections to DENV2 , a likely previously infecting serotype , we observed significantly higher levels of % IgG bound to DENV2 than DENV3 in the acute phase ( p = 0 . 0004 ) ( Figure 2C , 3A ) . However , due to the increase in avidity to DENV3 in the convalescent phase , no significant difference in avidity was detectable between the two serotypes ( p = 0 . 22 ) ( Figure 3B ) . Subsequently , at the 3 m , 6 m , and 18 m time-points , a shift occurred to significantly higher avidity against DENV3 , the currently infecting serotype , than against DENV2 ( p<0 . 0001 , p = 0 . 0004 , and p = 0 . 001 , respectively ) ( Figure 3C–E ) . Of note , the IgG serum avidity remained high for 1° DENV cases , while it declined for 2° cases after the convalescent time-point ( Figure 2 ) . The ability of serum to neutralize a virus can depend on Ab concentration and/or serum avidity [32] , among other parameters . During 1° DENV infections , a positive correlation between IgM serum avidity to DENV3 and DENV3-specific NT50 in the acute phase was observed ( Figure 4 ) , suggesting contribution of DENV-specific IgM Abs to early virus neutralization . We also observed a positive correlation between IgG serum avidity to DENV3 and NT50 to DENV3 in the acute phase of 2° DENV3 cases and at the 3 m time-point ( Figure 4 , Table 2 ) . Due to the slow kinetics of affinity maturation , these highly avid IgG are most likely secreted by pre-existing MBCs , suggesting the contribution of DENV cross-reactive Ab to virus neutralization . Moreover , a correlation was observed between DENV-specific total Ab titers and NT50 against DENV3 in the acute phase of 2° DENV3 infections ( Table 2 ) . Interestingly , positive correlations were noted between DENV2-specific NT50 and % IgG bound to DENV2 from the acute phase until 18 m post-infection ( Figure S1 , Table 2 ) . Lastly , a positive correlation was observed between DENV2-specific NT50 and IgG serum avidity to DENV3 at acute and convalescent time-points ( Figure S2 , Table 2 ) . NT50 , total Ab titer , and % IgG bound were compared between patients with and without the following clinical signs , stratified by immune status: compensated shock , hypotensive shock , vascular leak , cutaneous bleeding , mucosal bleeding , and hemoconcentration ( defined as in [33] ) . No difference was observed in the mean NT50 , total Ab titer , or % IgG at any time-point examined among 1° or 2° infections . When using the WHO classification ( DF vs . DHF/DSS ) , patients classified as DF did not demonstrate a statistically significant difference in mean NT50 or % IgG bound compared to those classified as DHF/DSS ( stratified by immune status ) , except for convalescent % IgG bound to DENV3 in 1° DENV3 infections , with 41 . 2±8 . 3 for DF and 53 . 4±14 . 2 for DHF/DSS ( p = 0 . 005 ) . A significant difference in total Ab titer was observed between 1° DF and DHF/DSS cases at the convalescent time-point ( p = 0 . 035 ) with a higher Ab titer observed in more severe cases ( 248 . 7±79 . 9 vs . 109 . 5±30 . 9 ) .
A better understanding of protective immune responses to natural DENV infections is critical for the development of safe and effective vaccines and for defining robust correlates of protection for vaccine trials . Protection from DENV infection and/or disease may depend on DENV-specific serum neutralization and thus may be associated with DENV-specific Ab titer and DENV-specific serum avidity . However , several questions remained unanswered , such as the role of IgM Abs and whether serum avidity correlates with serum neutralization . Using our longitudinal sample series from our hospital-based study , we show for the first time that DENV-specific neutralization titers peak in the convalescent phase and then decrease over time in both 1° and 2° DENV3 infections . We also observed increasing DENV3-specific avidity between the convalescent phase and 3 m post-infection in 1° DENV3 cases and between acute and convalescent phases in 2° DENV3 infections . In addition , we detected higher avidity against a heterologous , potentially previously infecting serotype ( DENV2 ) in the acute phase of 2° DENV3 infections , while avidity was higher against the current infecting serotype at later time-points ( 3–18 m ) . Finally , we show for the first time a correlation between serum avidity and serum neutralization titers in the context of DENV infection , and specifically demonstrate that serotype-specific neutralizing titers correlate with serum IgM avidity in 1° acute DENV infections and with serum IgG avidity in 2° DENV infections . In this study , we show that the DENV-specific NT50 increases from the acute to the convalescent phase in both 1° and 2° infections , with a higher NT50 in 2° cases . Whereas in 1° infections , naïve B cells are stimulated and IgG Abs are detected only in the convalescent phase , cross-reactive memory B cells are reactivated during 2° cases , inducing a rapid increase in DENV-specific IgG Ab in serum [41] . Neutralization is achieved when enough Abs bind to accessible epitopes of DENV , preventing binding to target cells or fusion of the viral and endosomal membranes and subsequent release of viral RNA into the cytoplasm of susceptible cells [42] . At later time-points ( 3–18 m ) , a decrease in NT50 was observed , as previously shown with DENV and other viruses [41] , [43] . In 1° DENV infections , we observed an increase in serum IgG avidity to DENV3 between convalescent phase and 3 m , reflecting affinity maturation of Abs . MBCs and LLPCs develop from naïve B cells and undergo affinity maturation and selection during the first 3 m after infection . As only clones with the highest avidity survive long-term , this leads to a higher mean IgG avidity over time [7] , [8] . In contrast , in 2° DENV infections , we observed an increase in serum IgG avidity to DENV3 earlier , between the acute and convalescent phase , which cannot be explained by affinity maturation of newly activated DENV3-specific B cells . Rather , this suggests that cross-reactive Abs secreted by pre-existing MBCs contribute to the early increase in avidity during 2° heterotypic DENV infections . In previous studies , we found significantly higher serum IgG avidity directed to a possible previously infecting DENV serotype ( DENV2 ) as opposed to the current infecting serotype ( DENV3 ) in the acute phase of 2° DENV3 infections [9] . Here , we confirm these findings and show that subsequently , serum IgG avidity against the current infecting serotype increases over time such that avidity to the current infecting serotype is greater than that to the previous infecting serotype 3–18 m post-infection . Serum IgG avidity to DENV2 decreases at 18 m time-point , possibly as DENV2-specific LLPCs are displaced from the bone marrow and replaced by newly generated DENV3-specific LLPCs . LLPC niches are limited in the bone marrow , and newly formed LLPCs can replace LLPCs formed during earlier infections [44] , [45] . The decrease in serum IgG avidity to DENV3 at the 18 m time-point could be explained by a strong extra-follicular response induced by the presence of pre-existing anti-DENV Abs in 2° infections , inhibiting the germinal center response and thus inhibiting a strong long-term immunity to the current infecting serotype , DENV3 [46] . Correlation between serum avidity and serum neutralization has been reported for measles , HIV and cytomegalovirus infections [47]–[49] . We analyzed here whether DENV-specific serum avidity and total Ab titer correlate with DENV-specific neutralization . During 1° DENV infections , neither IgG serum avidity to DENV3 nor DENV-specific total Ab titer correlated with DENV3-specific NT50 . Rather , the NT50 correlated with IgM serum avidity to DENV3 in the acute phase of 1° DENV3 infections , showing that highly avid IgM Abs are also highly neutralizing and suggesting that IgM Abs play an important role in DENV-specific serum neutralization . In contrast , in the acute phase of 2° DENV infections , DENV3-specific NT50 correlated with IgG serum avidity to DENV3 . Because of the rapid kinetics of appearance of IgG Abs in a 2° DENV infection and the slow affinity maturation process , these highly avid IgG Abs are most likely secreted by MBCs formed during the previous infection rather than by newly-activated naïve B cells . This suggests that cross-reactive Abs contribute to neutralization of acute 2° DENV infections . This is consistent with the isolation of strongly neutralizing cross-reactive MAbs from 2° DENV infections ( S . Smith , J . Crowe , R . de Alwis , A . de Silva & E . Harris , unpublished data ) . IgG serum avidity to DENV2 and DENV2-specific NT50 correlated at all time-points , suggesting that affinity maturation after the 1° infection contributes to strengthening the neutralizing activity of serum Abs . We also observed a correlation between IgG serum avidity to DENV3 and DENV2-specific NT50 , further supporting the relation between cross-reactive Abs and neutralization . Furthermore , a positive correlation between DENV-specific total Ab titer and DENV3-specific NT50 in acute 2° DENV infections was found , showing that a greater amount of Abs correlates with neutralization activity . Since dengue disease severity may be associated with sub-neutralizing ( enhancing ) concentrations of DENV-specific Abs [50] , [51] , we analyzed NT50 titers and IgG serum avidity among DF vs . DHF/DSS patients and among cases with different clinical signs of severity but did not find many significant differences . However , this could be due to the relatively low disease severity in the 2010–2011 epidemic in Nicaragua and thus the small sample size of DHF/DSS . Finally , pre-infection samples are not available through this hospital-based study; thus , we chose one likely previous infecting serotype ( DENV2 ) to analyze in this group of patients . Analysis of samples collected from documented sequential infections in our long-term prospective dengue cohort study will enable more precise investigation of cross-reactive Abs in 2° DENV infections . Overall , this study showed that IgM and cross-reactive IgG contribute to neutralization during acute DENV infections . To further support these results , we are conducting additional analyses of the DENV-specific serum neutralization capacity , including: 1 ) the use of β-mercaptoethanol to chemically deplete IgM Abs and 2 ) the use of DENV virions to deplete heterologous serotype cross-reactive Abs . While cross-reactive serum avidity dominates the acute 2° DENV response , avidity to the current infecting serotype becomes dominant over time . Significant correlations were observed between neutralizing Ab titers and serum avidity to both the current and a heterotypic serotype . Future studies will address the relation of avidity and NT50 to infection outcome ( symptomatic vs . inapparent ) and to disease severity . The unexpected results from the first proof-of-concept dengue live attenuated vaccine efficacy trial ( Phase 2b ) that were recently published [52] highlight the critical need to better understand the immune response to natural DENV infections and vaccine candidates and to identify robust correlates of protection . We believe that measurement of DENV-specific serum avidity should be integrated into evaluation of future vaccine trials and applied more broadly to the study of the immune response to DENV after natural infections . | Dengue is the most common mosquito-borne viral illness in humans , with 3 billion people at risk for infection . Four different dengue virus serotypes ( DENV1–4 ) cause the disease , which can be either inapparent or present with flu-like symptoms ( Dengue Fever , aka “breakbone fever” ) . The disease can be more severe and sometimes fatal , with signs of bleeding and vascular leakage leading to shock ( Dengue Hemorrhagic Fever/Dengue Shock Syndrome ) . No specific treatment or vaccine is available . Understanding how the human immune response develops during a natural infection can be beneficial for future vaccine studies and trials . DENV-specific serum neutralizing capacity may play a role in protection . The neutralization capacity of the serum may depend on the serum avidity against DENV , the amount ( or titer ) of the anti-DENV antibodies , and the accessibility of the epitopes targeted by these antibodies . Here we show that DENV-specific IgM antibodies likely play a role in neutralization during primary DENV infections and show a correlation between serum avidity and neutralization capacity in secondary DENV infections , with greater avidity to a previously infecting DENV serotype as compared to the current infecting DENV serotype in the early phases of infection , switching over time to the opposite situation . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"humoral",
"immunity",
"medicine",
"infectious",
"diseases",
"immunity",
"virology",
"emerging",
"viral",
"diseases",
"immunity",
"to",
"infections",
"immunology",
"dengue",
"biology",
"microbiology",
"viral",
"diseases",
"immune",
"response"
] | 2013 | Correlation between Dengue-Specific Neutralizing Antibodies and Serum Avidity in Primary and Secondary Dengue Virus 3 Natural Infections in Humans |
Tuberculosis ( TB ) is a devastating disease to mankind that has killed more people than any other infectious disease . Despite many efforts and successes from the scientific and health communities , the prospect of TB elimination remains distant . On the one hand , sustainable public health programs with affordable and broad implementation of anti-TB measures are needed . On the other hand , achieving TB elimination requires critical advances in three areas: vaccination , diagnosis , and treatment . It is also well accepted that succeeding in advancing these areas requires a deeper knowledge of host—pathogen interactions during infection , and for that , better experimental models are needed . Here , we review the potential and limitations of different experimental approaches used in TB research , focusing on animal and human-based cell culture models . We highlight the most recent advances in developing in vitro 3D models and introduce the potential of lung organoids as a new tool to study Mycobacterium tuberculosis infection .
Tuberculosis ( TB ) kills over 1 . 8 million people every year and thus remains the leading cause of death by an infectious agent [1] . Additionally , TB afflicts over 10 . 4 million new individuals per year and is estimated to exist in a latent form in nearly 2 billion people worldwide [1] . In addition to the human toll , TB imposes a significant economic burden , corresponding to 0 . 52% of the global gross national product , with a cost of over 500 million euros per year in the European Union alone [2] . Tackling TB is therefore a matter of urgency , as reflected in the current WHO End TB Strategy , which targets a 90% reduction in the incidence of TB to less than 100 cases per million people by 2035 [3] . Achieving this target requires a much quicker decline in TB incidence , from the current annual reduction of 2% to a 20% decrease per year [4 , 5] . For this , 3 areas in TB research are generally accepted as critical: development of novel vaccines , improved diagnostic tools , and better treatment options [5 , 6] . Succeeding in advancing these areas requires fresh approaches and ways of thinking , notably the development of better experimental models to study TB [7] . In this review , we discuss the different experimental approaches used in TB research , from in vivo models to human-based cell culture ones ( Table 1 ) . We also propose a road map of the available experimental approaches to study TB and of alternatives that are envisaged in a near future ( Fig 1 ) . We highlight the most recent advances in developing in vitro 3D models and introduce the potential of lung organoids as a new tool to study host—pathogen interactions during Mycobacterium tuberculosis infection . The development of such models requires a deep understanding of the disease pathogenesis and of the immune players , which are not the focus of this review and have been extensively reviewed elsewhere [8–10] .
Several animal models are used in TB research ( Fig 2 ) , ranging from zebrafish to nonhuman primates ( NHPs ) [11 , 12] . Mice are preferred model animals for a number of practical reasons , such as availability of immunological-based tools for mice , the existence of genetically modified mouse strains , and the small size and cost-effectiveness of maintaining mice in the laboratory [13–15] . Whereas many important aspects of the immune system are indeed conserved , there are also important differences that hamper the use of the mouse model of infection in our understanding of TB pathogenesis . The mouse is not a natural host for M . tuberculosis , and lung cavitation , a key characteristic for the disease transmission in humans [16] , is not observed for the 2 most-used mouse strains ( Balb/c and BL6 ) [13] . Necrotizing responses to M . tuberculosis occur in other mouse strains [17] , indicating the impact of genetic variability on the outcome of infection . A recent study illustrates this fact by demonstrating that the susceptibility to TB infection and the efficacy of Bacillus Calmette-Guerin ( BCG ) vaccination varied greatly when genetically different mouse strains were used [18] . It is thus not surprising that , depending on the mouse strain used , different studies report different data . Furthermore , variability in the reported results is enhanced by different experimental end points used [19 , 20] . The route and dose of M . tuberculosis administration and the mouse microbiome are also thought to contribute to variable findings . Since currently used mouse models fail to fully reflect human immunity to TB , several studies were performed using humanized mice . Humanized mice can be generated through the reconstitution of immunocompromised mice with human hematopoietic cells of different origins [21] . Infection of humanized mice with M . tuberculosis reproduced important hallmark features of human TB disease pathology , such as the formation of organized granulomatous lesions , caseous necrosis , and bronchial obstruction [22 , 23] . However , abnormal T-cell responses and an impaired bacterial control were also observed [23] . In line with this , humanized mice generated by engraftment of human leukocyte antigen ( HLA ) -restricted cells showed partial function of innate and adaptive immune systems , culminating in antigen-specific T-cell responses to mycobacterial infection but also in lack of protection [24] . Other approaches consist in infecting transgenic mice expressing human-specific molecules such as , for example , the human cluster of differentiation group 1 CD1 , which allows for the study of a humanized immune system using the mouse model of infection [25] . In all , humanized mice are a good tool to study TB , being particularly relevant for the study of HIV/TB , as recently shown [26] . However , this model requires further improvement to reach its full potential for TB research . To address some of these limitations , other animal models have been used . For example , guinea pigs and rabbits may be considered better models to study the humanlike granuloma formation , a hallmark of M . tuberculosis infection in the lung [14 , 27 , 28] , although they still fail to display other characteristics of the human disease . Additionally , they are much more difficult to maintain in the lab and a lot less immunology tools are available for these 2 species , which greatly limits their use . Infection of zebrafish ( Danio rerio ) embryos with the natural fish pathogen M . marinum is also used as a model for the study of granuloma formation [29–31] . Several similarities were found in the cellular and molecular events presiding M . marinum and M . tuberculosis infections [32–34] , despite the many differences between these 2 diseases . Research on zebrafish embryos benefits from the similarities between M . marinum and M . tuberculosis , i . e . , from the optical transparency of the embryos , which facilitates the use of advanced imaging techniques , and from the easy genetic manipulation of zebrafish , which allows for deep mechanistic molecular studies . Because zebrafish embryos lack a fully developed immune system , the study of later stages of infection requires the use of adult fish , thus abrogating the advantages of using embryos . Furthermore , the physiological differences between zebrafish and humans are enormous , which inevitably imposes some limitations to the use of this model . As for the other animal models , specific facilities for housing zebrafish are required . NHPs are so far considered as the best animal model for TB research [35 , 36] , as the disease pathogenesis parallels that observed in humans [37] . NHPs present lung cavitation [38]; show a spectrum of disease overlapping that of humans , namely , with the establishment of latent TB infection [38]; display a susceptibility to TB in the presence of comorbidities such as HIV and anti—tumor necrosis factor ( TNF ) treatment similar to that reported in humans [39 , 40]; and present a transcriptomic signature of disease comparable to the human one [41] . However , the ethical , practical , and economic problems that are inherent to NHP research [36 , 42] , exacerbated when the animals are made to develop a potentially fatal infection , hinder the generalized use of this animal model , which in fact accounts for only 1% of the papers published in TB ( Fig 2 ) . In conclusion , important advances in our understanding of TB have been made through the use of different animal models . However , in addition to each model’s specific limitations , all animal-model research into human diseases is ultimately restricted by the need to translate findings across species . This calls for the wider use of human-based models to complement and reduce the use of experimental in vivo research .
Owing to the central role of the macrophage as host and effector cell during M . tuberculosis infection [43 , 44] , many studies have been centered in macrophage cell cultures . In terms of human-based systems , monocyte-derived macrophages are the most widely used culture . Among these is the human monocytic leukemia cell line , THP-1 , which is easy to culture , yielding a nearly unlimited amount of cells for experimental purposes . THP-1 cells are typically differentiated to macrophages through the stimulation with phorbol 12-myristate 13-acetate ( PMA ) for 3 days , although different protocols are found in the literature [45 , 46] , which may contribute to some variable findings . Macrophages can alternatively be freshly derived by extracting and culturing human peripheral blood mononuclear cells ( PBMCs ) in the presence of differentiating factors , namely , granulocyte-macrophage colony stimulating factor ( GM-CSF ) or macrophage colony stimulating factor ( M-CSF ) [47] , or of human serum [48] . In these cases , the macrophages are of primary origin , but because of the in vitro differentiation process , their properties are most likely different from tissue-resident cells . Although alveolar macrophages would be ideally used , access to these cells is a costly procedure that requires lengthy ethical approvals , which limits their use . In vivo , M . tuberculosis is found in foamy macrophages . These cells result from pathogen-induced dysregulation of host lipid synthesis and sequestration and play a key role in both sustaining persistent bacteria and contributing to the tissue pathology [49] . Therefore , in vitro differentiation of foamy macrophages is an excellent tool for the study of macrophage-pathogen interactions . A protocol to convert cultured macrophages ( THP-1 or primary ) into foamy cells has been developed by incubating these cells under hypoxia [50] . Other alternatives for the differentiation of foamy cells include the exposure of cell cultures to palmitic acid , oleic acid , or lipoproteins [51] or to surfactant lipids [52] . Given the importance of working with primary , unmanipulated cells , many studies have been performed using freshly isolated human PBMCs [53] . Human PBMCs are easily accessible , cost-effective , and readily infected with M . tuberculosis , responding to the infection with the production of relevant immune mediators such as TNF and other interleukins as well as chemokines [53] . Furthermore , the PBMC response captures interactions between different immune cell types , such as monocytes , T cells , and B cells , which are in fact interacting during natural immune responses . However , these cells still differ from the tissue-resident ones and when used in in vitro cultures lack the environmental stimuli that ultimately shape cellular responses to infection . In addition to the standard monolayer cultures , PBMCs have been used to develop in vitro models of human mycobacterial granulomas . In 1 study [54] , a sequential recruitment of human monocytes and lymphocytes towards mycobacterial antigen-coated artificial beads or live mycobacteria was observed . This recruitment culminated with the formation of a cellular structure reminiscent of natural mycobacterial granulomas in terms of morphology and cell differentiation [54] . This or similar/improved models have been used in several studies [55–57] . A different approach based on the culture of human PBMCs in a collagen matrix with a low dose of M . tuberculosis was used to develop an in vitro model of human TB granuloma with dormant bacteria [58] . This model recapitulated important characteristics of the mycobacterial granuloma , such as the aggregation of lymphocytes surrounding infected macrophages , the formation of multinucleated giant cells , the presence of secreted cytokines and chemokines in the culture supernatants , and the reactivation of M . tuberculosis upon immune suppression caused by TNF blockade [58] . These models offer the possibility of studying the infection by M . tuberculosis in a more physiological environment , resembling the structure of the infected human tissue . They constitute valuable approaches for the study of cell—cell interactions , cell differentiation , and bacterial control . To further reflect the complex environment and structure of the human lung , a growing body of studies are resorting to the use of new technologies in the tissue-engineering field to advance human-based TB research models into the 3D era ( Fig 3 ) [59] . Tissue bilayer systems consisting of epithelial and endothelial cell layers were initially developed to study the early events of alveolar infection [60 , 61] . More recently , through the use of these systems , microfold ( M ) cells were shown to play a critical role in translocating M . tuberculosis to initiate lung infection [62] . A study combining lung-derived epithelial cells and fibroblasts with peripheral blood primary macrophages reported the establishment of a lung tissue model that upon infection led to the clustering of macrophages reminiscent of early TB granuloma formation [63] . Similarly , another report showed the implementation of an in vitro human 3D lung tissue model to study M . tuberculosis infection that allowed the analysis of human granuloma formation and resembled some features of TB [64] . A novel bioengineering approach utilized bioelectrospray technology to generate microspheres of M . tuberculosis–infected human PBMCs in a 3D extracellular matrix [59 , 65] . This model takes advantage of the high throughput potential of the bioelectrospray system and allows the interrogation of host—pathogen interactions in 3D in the context of an extracellular matrix [59 , 66] . When combined with a microfluidic system to enable pharmacokinetic modeling , this model also showed great potential to monitor the efficacy of new antibiotic regimens or anti–M . tuberculosis drugs [65] . Although these experimental systems facilitate the discovery of the interactions between mycobacteria and host cells in a more physiological environment , they still bear some limitations , namely , the lack of vasculature and absence of other immune cells ( e . g . , neutrophils ) that play a role in the multifaceted response in TB infection . Also , not all the models include epithelial and stromal cells , which are known to play important roles during infection [67 , 68] . Finally , the spatial organization of the lung is mostly lost , and so is the role of the anatomical constraints during infection . The advances made in the development of all these models will certainly contribute to moving the field forward into novel strategies that overcome current limitations . In this context , other 3D and tissue-chip models are being explored .
Organoids are in vitro 3D cell cultures generated from embryonic stem cells ( ESCs ) , induced pluripotent stem cells ( iPSCs ) , or adult stem cells ( aSCs ) that functionally and structurally mimic the organ they model [69 , 70] . This technology is emerging as a promising tool to study organ development and disease “in a dish” [69 , 70] . The potential of organoids to study infectious processes has been increasingly demonstrated in many original papers and recently reviewed by Mills and Estes [71] , with most examples coming from human gastric [72] , brain [73 , 74] , and gut [75 , 76] organoids . So far , lung organoids have not been explored as a model to study infection .
Human lung organoids have been generated through different technologies [77] . The most advanced studies involve the differentiation of human embryonic stem cells into endoderm cells , anterior foregut endoderm cells , lung progenitor cells , and , finally , various types of airway epithelial cells . If this procedure is performed in a 3D structure , a human lung organoid is formed , as initially described by Dye et al . [78] and Konishi et al . [79] . These relatively immature organoids may be transplanted into mice to complete their differentiation in an in vivo environment , into adultlike airways [80] . Despite some limitations , lung organoids recapitulate important features of the lung , such as heterogeneous cell composition , spatial organization , and retention of a stem cell population harboring the capacity for both self-renewal and differentiation [70] . There is increasing evidence that human lung organoids may be used to investigate the cellular and molecular pathways implicated in lung development and lung diseases as well as screening platforms for drugs directed at respiratory diseases [77] . At the disease level , the application of lung organoids to cancer development , cystic fibrosis , and infection is envisaged although still is unexplored in TB research . The obvious advantage of lung organoids over 2D and 3D cultures relies on their spatial organization and heterogeneity of the cellular components . As compared to the animal model , infection of lung organoids allows the inclusion of very early time points , which are difficult to follow in in vivo infections , whilst at the same time overcoming species differences and reducing the use of animals in research . Thus , as the lung organoid technology stands , human-derived lung organoids could be explored to study the early events of infection , namely , the initial interactions of M . tuberculosis with the lung epithelium [67] . Of the aforementioned experimental models , both 2D and 3D cultures based on PBMCs may also be explored to investigate the early immune events during infection , although to a lesser complexity than organoids . Although there are indeed exciting perspectives for the use of lung organoids as a model for TB research , some important challenges remain before they can be more systematically used as experimental models . Chief among these is the introduction of immune cells in the structure of lung organoids . Only then will lung organoids cover the complexity of immune response and of the stromal-immune cells’ cross talk upon in vitro infection . Also , the introduction of the vasculature would be an important improvement to create a more dynamic model in which the microenvironment of an airway could be experimentally controlled . This dynamic lung organoid would be an interesting model for drug screening . In this context , microfluidic cell culture devices called “organs on a chip” have also generated airway epithelium from human adult airway cells grown on an air—liquid interface platform [81 , 82] . Another important step forward would be the development of lung organoids from iPSCs instead of ESCs , as this will offer the possibility of including in the disease modelling individual variability , either genetic or caused by extrinsic conditions . In the context of TB research , this would allow for the study of host–M . tuberculosis–microenvironment interactions at an individual level by infecting lung organoids generated from individuals with HIV or diabetes versus controls or from smokers versus nonsmokers . This would be of utmost importance as the molecular mechanisms underlying the impact of comorbidities and life habits on the course of infection remain incompletely understood . Additionally , comorbidities are very difficult to incorporate in the other complex experimental system—the animal . Generation of personalized lung organoids would also open new avenues for the study of individual responses to therapies and thus for the implementation of personalized medicine .
TB remains a devastating disease to mankind and a huge challenge for the scientific community . From many epidemiological studies , it is clear that the progression of the disease is highly related to the host immune status , and as such , a deep understanding of the immune response to M . tuberculosis is critical for the development of novel preventive and therapeutic strategies . However , the lack of experimental systems that parallel the complexity of the human disease remains a major gap hindering the in-depth study of the immune response in TB . Critical species differences mean that the widely used animal models only partly recapitulate the human disease . NHP models are the most representative ones but bear high operational and maintenance costs . Traditional human cellular systems overcome the interspecies translation problem but are limited by their low level of complexity and the abnormal characteristics of cell lines . Recent development of human-based tissue models is promising real alternatives for the experimental study of human TB . State-of-the art in vitro models have now incorporated several important characteristics of “real-life” tissues , namely , the presence of different cell types and of the extracellular matrix . Technological advances coupled to these models allow for the experimental manipulation of different variables , which is critical in studies of host—pathogen interactions or in drug-screening processes . A key next step will be to introduce in these models the anatomical constraint associated with the lung tissue . Albeit at very early days , lung organoids hold a great promise here . The road from lung organoids to complete lungs “in a dish” is still a long one , but creating a lung structure composed of different stromal cells and coupled with a competent immune system would unquestionably provide a major leap forward in TB research . Being able to use as starting points cells from different individuals ( TB patients or latently infected people with different genetic backgrounds and comorbidities ) would constitute a revolutionary way of studying TB . This would open many new avenues to investigate long-standing questions and put us in a privileged position to effectively tackle TB . In sum , recent advances in tissue engineering and future steps in this area will certainly play an important role in the development of new tools for the study of infectious diseases . Such tools hold the potential to replace some animal experiments and overall lead to a reduction of the number of animals used in TB research . Most importantly , these tools will allow for a series of key questions to be answered in a more precise way by including individual variability at the single-cell and tissue levels . | Tuberculosis ( TB ) is the number 1 killer in the world due to a bacterial infection . The study of this disease through clinical and epidemiological data and through the use of different experimental models has provided important knowledge on the role of the immune response generated during infection . This is critical for the development of novel vaccines and therapeutic strategies . However , in spite of the advances made , it is well accepted that better models are needed to study TB . This review discusses the different models used to study TB , highlighting the advantages and disadvantages of the available animal and cellular models and introducing recently developed state-of-the-art approaches based on human-based cell culture systems . These new advances are integrated in a road map for future study of TB , converging for the potential of lung organoids in TB research . | [
"Abstract",
"General",
"introduction",
"In",
"vivo",
"models",
"in",
"TB",
"research",
"Human-based",
"in",
"vitro",
"models",
"in",
"TB",
"research",
"Organoids",
"as",
"infection",
"models",
"Lung",
"organoids",
"in",
"TB",
"research",
"Conclusions"
] | [
"organoids",
"medicine",
"and",
"health",
"sciences",
"animal",
"models",
"of",
"disease",
"biological",
"cultures",
"tropical",
"diseases",
"microbiology",
"animal",
"models",
"bacterial",
"diseases",
"developmental",
"biology",
"review",
"model",
"organisms",
"organism",
"development",
"experimental",
"organism",
"systems",
"lung",
"development",
"organ",
"cultures",
"bacteria",
"research",
"and",
"analysis",
"methods",
"animal",
"models",
"of",
"infection",
"infectious",
"diseases",
"animal",
"studies",
"tuberculosis",
"organogenesis",
"mouse",
"models",
"actinobacteria",
"mycobacterium",
"tuberculosis",
"biology",
"and",
"life",
"sciences",
"organisms"
] | 2017 | Experimental study of tuberculosis: From animal models to complex cell systems and organoids |
A recent trend in drug development is to identify drug combinations or multi-target agents that effectively modify multiple nodes of disease-associated networks . Such polypharmacological effects may reduce the risk of emerging drug resistance by means of attacking the disease networks through synergistic and synthetic lethal interactions . However , due to the exponentially increasing number of potential drug and target combinations , systematic approaches are needed for prioritizing the most potent multi-target alternatives on a global network level . We took a functional systems pharmacology approach toward the identification of selective target combinations for specific cancer cells by combining large-scale screening data on drug treatment efficacies and drug-target binding affinities . Our model-based prediction approach , named TIMMA , takes advantage of the polypharmacological effects of drugs and infers combinatorial drug efficacies through system-level target inhibition networks . Case studies in MCF-7 and MDA-MB-231 breast cancer and BxPC-3 pancreatic cancer cells demonstrated how the target inhibition modeling allows systematic exploration of functional interactions between drugs and their targets to maximally inhibit multiple survival pathways in a given cancer type . The TIMMA prediction results were experimentally validated by means of systematic siRNA-mediated silencing of the selected targets and their pairwise combinations , showing increased ability to identify not only such druggable kinase targets that are essential for cancer survival either individually or in combination , but also synergistic interactions indicative of non-additive drug efficacies . These system-level analyses were enabled by a novel model construction method utilizing maximization and minimization rules , as well as a model selection algorithm based on sequential forward floating search . Compared with an existing computational solution , TIMMA showed both enhanced prediction accuracies in cross validation as well as significant reduction in computation times . Such cost-effective computational-experimental design strategies have the potential to greatly speed-up the drug testing efforts by prioritizing those interventions and interactions warranting further study in individual cancer cases .
Over the past decade , there has been a significant increase in the R&D cost when translating new cancer drug candidates into effective therapies in the clinic . The two single most important reasons are ( i ) lack of efficacy and ( ii ) clinical safety of the candidate drug compounds [1] . This decline in productivity of the pharmaceutical industry has greatly challenged the dominant paradigm in drug discovery , where such ‘magic bullet’ compounds are being searched that could produce dramatic treatment outcomes at a population-level by modulating one specific target only . The shortcomings of such ‘one-size-fits-all’ treatment strategies are well reflected in the disappointing outcome of the current anticancer drug development , where agents directed at an individual target often show limited efficacy and safety due to factors such as off-target activities , network robustness , bypass mechanisms and cross-talk across compensatory escape pathways [2]–[4] . Most cancers develop from specific combinations of genetic alterations accumulated in tumor cells , which are often distinct between different cancer types and result in different treatment responses to the same therapy . Moreover , the extensive mutational heterogeneity results in alterations within multiple molecular pathways , making most advanced tumors readily resistant to single-targeted agents . Therefore , rational strategies to develop multi-targeted therapies for specific cancer types are needed to attack the resistance problem and to provide more effective and personalized treatment strategies [5] . Targeted drug combinations may also overcome the side effects associated with high doses of single drugs by countering pathway compensation and thereby increasing cancer cell killing while minimizing overlapping toxicity and allowing reduced dosage of each compound [6] . Even though it is widely acknowledged that effective cancer treatments need to go beyond the traditional ‘one disease , one drug , one target’ paradigm , the major bottleneck hindering the development of combinatorial therapies is the lack of such systematic experimental-computational approaches that could pinpoint the most effective combinations [7]–[9] . While efforts based on next-generation sequencing are very successful at systematically characterizing the structural basis of cancers , by identifying the genomic mutations associated with each cancer type [10] , these findings often do not lead to clinically actionable therapeutic strategies and rarely to rational targeted combinations . The large number of genetic alterations present in tumor cells makes the discrimination of the cancer-specific driver mutations and pathways highly challenging , and even when genetic aberrations with pathogenetic importance can be identified , these may not be pharmaceutically actionable . Moreover , genes not altered at the genomic level may also play essential roles in the cancer progression , hence providing additional therapeutic opportunities [11] . In contrast , systematic assessment of genes for their contribution to tumor addictions can provide functional insight into the molecular mechanisms and pathways behind specific cancer types , hence highlighting their vulnerabilities associated with driver genes , synthetic lethal interactions and other tumor dependencies [12]–[14] , which are complementary to the structural information obtained from the cancer mutational landscape . Advances in high-throughput chemical and RNAi screening have now made it possible to carry out comprehensive functional screening in cancer cells , providing novel targets for the next generation of anticancer therapies for patients sharing a common genetic background [15]–[18] . However , despite the emerging possibilities for perturbing gene functions with a wide spectrum of shRNA/siRNA libraries or using diverse drug and compound collections , functional interactions between genes and/or drugs have remained extremely difficult to predict on a global scale [18] . The complex genotype-phenotype relationships behind such interactions pose modeling challenges beyond the reach of the classical linear approaches . Moreover , polypharmacologic compounds elicit their bioactivities by modulating multiple targets , which leads to a combinatorial explosion both in the pharmacological and molecular spaces . Taken together , the exponentially increasing number of possible RNAi , chemical , target and dose combinations poses great experimental challenges , and exhaustive experimentation with all the possible combinations is impossible in practice , making the pure experimental approach quickly unfeasible [19] . To meet these computational and experimental challenges , novel modeling frameworks and efficient computational algorithms are needed to effectively reduce the search space for determining the most promising combinations and prioritizing their experimental evaluation . Ideally , the experimental setup should be both economical and practical , utilizing such functional measurements and phenotypic readouts that are readily available in typical drug screening experiments . Moreover , the experimental and computational platforms should also be compatible with the eventual clinical translation in the sense that the measurements and their analysis can be made in each patient individually , and that the modeling and algorithmic predictions can be calculated in a reasonable time . A number of computational algorithms have been developed for predicting drug combinations in silico [5] , [9] , [20] . Most of the approaches are based on detailed mathematical modeling , utilizing a priori knowledge extracted from databases , such as those focusing on established cancer pathways , metabolic network constructions or literature-curated models [21]–[23] . A limitation of such detailed models is that global kinetic information for many cancer-related systems are still rarely available , and reduced subsystem models are often biased toward what is already known about the cancer processes . For instance , pathway-specific models may miss important novel features , such as pathway cross-talks or novel cancer dependencies . Accordingly , although major canonical pathways involved in different cancer types are increasingly well established , individual pathway models cannot capture the complex and context-dependent cellular wiring patterns behind distinct cancer phenotypes [5] . There are also approaches that take the cell context into account by means of global gene expression or targeted phosphoproteomics profiling [24]–[27] . However , such molecular phenotypes are not routinely profiled in a typical high-throughput drug testing approaches , especially in clinical settings . Moreover , downstream changes in the expression patterns are shown to be suboptimal in distinguishing mechanism of action between different compounds [28] , [29] . Perhaps more importantly , targets identified by means of genomic profiling may not be pharmaceutically actionable in clinical practice . For instance , many genes identified through expression profiling or genomic sequencing are either not druggable at all , or druggable , but not actionable , as there are no approved drugs available in the clinic . In this article , we present an efficient model construction algorithm , named TIMMA ( Target Inhibition inference using Maximization and Minimization Averaging ) , which makes the use of partly overlapping target subsets and supersets of promiscuous drug-target binding profiles in the estimation of anticancer efficacies for novel drug target combinations . The model construction and target combination predictions are based on functional data on drugs and their targets that are available from comprehensive target binding assays and from high-throughput drug sensitivity screens . We implemented a modified sequential forward floating search algorithm for model selection , which enables scaling-up to proteome-wide evaluation of the targets in terms of their relevance to cancer survival . Both simulation studies and an application to a canine osteosarcoma cell line data showed that TIMMA achieved improved prediction accuracy , when compared to a published algorithm [30] , at significantly lower computational cost . Importantly , application case studies in MCF-7 and MDA-MB-231 breast cancer and BxPC-3 pancreatic cancer cells confirmed that TIMMA-predicted kinase targets are essential for tumor survival , either individually or in combination , as validated by independent single and pairwise target knockdowns with siRNA screening . Our model predictions , visualized as a target inhibition network , provide insights into such druggable cancer cell addictions , the inhibition of which can jointly block the survival pathways . With the increasing interest in drug combination screens , our modeling strategy can be readily used as an efficient prioritization procedure to pinpoint the most potential drug combinations based merely on their selectivity profiles and individual responses in given cancer samples .
Consider a set of drugs where the single-drug treatment efficacy on a given cancer sample is measured as a phenotypic response in a high-throughput drug screen . The drug's treatment efficacy to kill cancer cells is conventionally scored using response parameters , such as the drug concentration at which the cancer cell growth is inhibited by a certain percentage ( e . g . half-maximal inhibitory concentration IC50 ) . A drug with a smaller inhibitory concentration is usually considered as more potent . Drug treatment efficacy and potency can be also quantified based on the area under the dose-response curve , such as the activity area ( AA ) [31] or the drug sensitivity score ( DSS ) , which provide summary information about the complex dose-response relationships . We denote the drug treatment efficacy data by a vector with length and scale it into the interval of [0 , 1] , with the minimum and maximum efficacies being 0 and 1 , respectively . To relate a drug's treatment efficacy with its underlying mechanism of action , the cellular targets of the drug need to be mapped into a drug-target inhibition profile . Let the potential target set be , where refers to the total number of targets that bind to at least one of the drugs . A target inhibition profile of a drug i can be binarized from drug-target binding affinities as a binary vector , where 0 and 1 is a result of classification of low and high binding affinities , respectively . The target inhibition profile for all the drugs is abbreviated as . An example of such binarized target annotations can be derived from quantitative binding assay measurements collected from the ChEMBL database [32] , provided that knowledge of relevant binding affinity cutoffs is applicable . Given the single drug efficacy and target inhibition profiles , our aim is to predict the treatment efficacy for novel drug combinations . We consider the target inhibition profile of a drug combination as a union of the target inhibition profiles of each component drug in the cocktail ( Figure 1 ) . However , not all the targets in the profile are essential in explaining the treatment efficacy . Ideally , an effective drug combination should affect signaling pathways involved in cell proliferation and growth of the particular cancer type . In searching for a rational design in polypharmacology , one needs to first identify a set of targets whose interactions play critical roles in delivering the anticancer efficacy [9] , [33] . Therefore , a fundamental computational problem is to identify a subset of therapeutic targets whose combinatorial interaction effects can be predicted in relation to cancer survival phenotypes . Note that in an individualized experimental setting , where different cancer types are tested for drug efficacy , the therapeutic targets should be also cancer-specific . Let denote such a cancer-specific therapeutic target set . Identification of corresponds to a partition of the potential target set into two non-empty and non-overlapping groups . Let the space of distinct partitions for be denoted by . We will learn an optimal partition from such that the cancer-specific targets can be separated from the remaining ones in . We assume that the drug target inhibition profiles and the drug treatment efficacy data can be used for evaluation of target set relevance provided that is a treatment outcome of drug perturbations on cancer survival pathways by multi-target inhibition in . A plausible assumption is that the targets of more effective drugs are more likely to be involved in cancer survival pathways than those of less effective drugs . Therefore , targets that are predictive of drug efficacy are , in general , functionally important for cancer survival and should be selected with a higher probability for drug target combinations as well . More formally , the learning procedure for identifying such a cancer-specific target set is to find a model that gives the best prediction performance . We are especially modeling multiple interactions among the target set for the prediction of drug efficacies and therefore capturing the synergistic combination effects that cannot be revealed by inhibiting any of the targets individually . Let denote the model prediction error for a drug or drug combination in a testing set . In its most basic form , the prediction error is calculated as the absolute difference between the predicted and the actual treatment efficacy: ( 1 ) where refers to the predicted efficacy for drug by a model that takes and as training data . We take here a formal model-based strategy to estimate by formulating a predictive modeling framework for any training data ; the model construction and model selection algorithms will be proposed in the sequel . In an earlier work by Pal and Berlow [30] , two fundamental set theoretic rules were exploited for predicting the drug efficacy according to its target profiles: Construction of a TIMMA model for predicting drug efficacy requires a selection of cancer-specific target set as the model parameter . Usually is a priori unknown and need to be inferred from the potential target set . In our model-based learning framework , the likelihood of a proposed target set being composed of cancer-specific targets can be evaluated using the prediction accuracy of the corresponding TIMMA model that takes as its parameter . More formally , we consider an objective function for model selection as the average leave-one-out ( LOO ) TIMMA prediction error: ( 8 ) where the leave-one-out prediction error for drug is given by Eq . 1 and Eq . 3–7 . Given that the combinatorial space for is huge for even a modest number of targets , it is not possible to calculate the objective function for all the possible target subsets using exhaustive enumeration . We consider a Sequential forward floating search ( SFFS ) algorithm modified from [34] for minimizing Eq . 8 in a computationally efficient manner . The modified SFFS algorithm learns the optimal cancer-specific target set by aggregating and subtracting targets in at different steps , as defined in the following , with the aim of minimizing the prediction error , where is the cardinality of set , i . e . : We have further improved the scalability of the TIMMA algorithm to large and complex data in MATLAB by exploiting its matrix computation architecture . Briefly , the TIMMA model was represented as a 3- dimensional array , where each drug's contribution to the estimate of is calculated independently of each other . This multi-dimensional data structure has enhanced the computation efficiency significantly as most of the iteration loops can be avoided . Meanwhile , independent computing enables parallel distribution of the model prediction on separate processors , e . g . one processor for one drug , which will further decrease the computation time . For the SFFS target selection , the multi-dimensional data structure also facilitates the aggregation and comparison of prediction error at the Inclusion step when the target is added to , as can be incrementally derived based on that has been obtained in the previous iterations . The TIMMA implementation code is freely available at http://timma . googlecode . com/ . For the optimal target set selected by the SFFS algorithm , the result of the TIMMA model prediction is summarized in the predicted efficacy matrix , which enumerates the treatment efficacy for each of the combinatorial target inhibition in ( Figure 1C ) . Here , we considered the predictions for the single and pairwise target inhibitions only , and derived a synergy score for the target pair ( A , B ) based on the multiplicative null model: ( 9 ) where and denote the predicted efficacies for the target pair and its individual targets , respectively . The multiplicative model is widely being used in the gene knock-out studies in model organisms to score quantitative genetic interactions between gene deletions [35] , [36] . It has also been recently applied to investigate genetic interactions in human cancer cells using combinatorial RNAi screening [37] , as well as to characterize drug synergy effects according to the Bliss independence model [38] , [39] . Using the model predictions , we can calculate the synergy score also for those drug pairs ( d1 , d2 ) whose targets are included in . If one or both of the drugs are inhibiting multiple targets , e . g . and , then we assign a drug synergy score for the drug pair using the mean of its corresponding target pair synergy scores defined by the multiplicative model ( Eq . 9 ) , i . e . ( 10 ) In the given example , A deviation of from zero provides evidence for a non-additive interaction between the two drugs , where indicates synergy and indicates antagonism . When the target set size is fixed at two , the TIMMA model construction algorithm evaluates the pairwise target inhibitions without considering any higher-order interactions . This enables the TIMMA modeling strategy to systematically predict target pairs with synthetic lethality effect . By definition , synthetic lethality among a target pair states that: ( i ) inhibition of either of the single targets will result in incomplete cancer killing; and ( ii ) inhibition of both of the targets simultaneously will block the complete cancer survival sub-network . Therefore , the target inhibition network for the synthetic lethal target pair can ideally be represented as two nodes in parallel , similar to the topology of or shown in Figure 1E . In comparison , there are two competing models: one with no links connecting the target nodes ( referred to as a singleton model ) , and the other with two nodes linked in a sequence ( referred to as a series model ) . Under the series model , no synthetic lethality effect is expected since the inhibition of a single target is already sufficient to block the cancer survival pathway . Therefore , from the model fitness perspective , we are expecting higher prediction accuracy for a synthetic lethal target pair under the parallel model , compared to both the series model as well as the singleton model . To evaluate the likelihood of a parallel model against the competing models for a given target pair ( A , B ) , we defined a synthetic lethality score as the ratio of the fitness function of these two models , given by the total sum of squares ( TSS ) of the predictions: ( 11 ) The synthetic lethality score is conceptually different from the multiplicative synergy score as they are addressing different questions . The synthetic lethality score evaluates the pairwise target interactions by comparing the likelihood of three competing model structures , whereas the synergy score is derived based on the model averaging by combining all the possible models . Synthetic lethality corresponds to a special case of synergy , which requires minimal individual effects that are not considered explicitly in the multiplicative synergy score . Further , the higher-order target interactions , which are evaluated during the sequential forward search for the TIMMA model , are not considered when calculating the pairwise synthetic lethality score .
We started by evaluating the relative performance of TIMMA and PKIM in terms of their accuracy in predicting the treatment efficacies for new drugs on the simulated dataset . It was found out that TIMMA systematically improved the average leave-one-out ( LOO ) prediction accuracy , compared to PKIM , at each predefined drug-target threshold ( Figure 2A , paired t-test , p = 5 . 0024×10−10 ) . Since TIMMA combines the information from a drug's subsets and supersets simultaneously , its predictions are more robust to data noise and other technical factors that are inconsistent with the model assumptions , compared to PKIM , which does not consider model averaging . In particular , TIMMA gains on average 22 . 4% increase in the prediction accuracy especially for affinity thresholds lower than 0 . 8 , which correspond to the promiscuous cases with , on average , more than two targets per drug ( Figure 2B ) . These results demonstrate the importance of the improvements provided by the TIMMA algorithm , which make it applicable also to more challenging and practical cases , where target promiscuity is common and knowledge about all the cellular targets of drugs is rarely available . Another important consideration in the large-scale drug screens is the computational complexity of the prediction algorithms . The computation times for TIMMA and PKIM model construction algorithms , SFFS and greedy search , respectively , were compared on a standard 2 . 6 GHz desktop computer . In contrast to the exponentially increasing time that is needed for the PKIM model construction , TIMMA takes approximately linear increase in time with the number of targets ( Figure 3A ) . Even though the SFFS is computationally more demanding than greedy search in model selection , TIMMA achieved marked speed-up due to the optimization techniques using multi-dimensional matrix computations ( Figure 3B ) . Notably , with 20 targets and 10 drugs , for example , the greedy search will take 10 days , while the TIMMA takes on average 30 minutes to complete , and thus saves up 99% of the computation time . The enhancement in the computation speed facilitates the analyses of larger and more complex datasets with increasing number of drugs and their target information . We next tested whether TIMMA can lead to improvements in the real dataset used in the PKIM work [25] , first by fixing the threshold at 0 . 9 . From the set of 317 kinases , the PKIM model identified 8 kinases with a mean LOO error of 0 . 1314 , while TIMMA identified a different set of 8 kinases with a decreased LOO error of 0 . 0574 ( Dataset S2 ) . When varying the threshold , the average LOO prediction accuracy of TIMMA was significantly better than that of PKIM ( Figure 4A , paired t-test , p = 1 . 3910×10−5 ) . Similarly as in simulated dataset ( Figure 2A ) , the improvement in the prediction accuracy varied with the selected cut-off threshold ( Figure 4A ) . As expected , when the threshold is close to 1 , the two models performed equally well , as the drug-target information is too few to make any reliable predictions; while TIMMA again systematically outperformed PKIM at the smaller thresholds . As revealed in many kinome-wide drug binding assays , most drugs , albeit considered previously specific to single or double targets , have shown a relatively wide range of binding affinities to multiple off-target kinases [47] . Our model can also make use of such promiscuous drug-target interactions that are informative for predicting drug cancer killing efficacies . This was further investigated in a receiver operating characteristics ( ROC ) analysis of the prediction performance , where the problem was to distinguish the 12 most sensitive drugs with positive efficacy values ( Figure 4B ) . In this analysis , the area under the ROC curve ( AUC ) for TIMMA was 0 . 9679 and for PKIM 0 . 7144 , further demonstrating the improved predictive power of the TIMMA model . To test whether the SFFS model selection algorithm can find solutions close to the global optimal target sets , we performed an exact analysis for maximally 12 kinases , where exhaustive search can be performed at a reasonable running time . More specifically , kinases from the full set of 317 kinases were randomly selected , where , and an exhaustive search was run to determine the optimal subsets of the kinases . We applied here a fixed cut-off threshold of , which equals to the average of all the values over the drug-target pairs . The optimal sets determined by the SFFS algorithm in TIMMA and by the greedy search algorithm in PKIM were compared with the global optimum in terms of prediction accuracy . The SFFS algorithm gave significantly better results than the greedy search for ( Figure 5 , paired t-test , p = 3 . 3397×10−6 ) . This demonstrates that the computationally efficient SFFS algorithm can find solutions that are not too far from the globally optimal solution . After confirming the appropriate performance of the TIMMA model , we applied it to two practical case studies . In the first one , we systematically evaluated the predictions of the TIMMA MCF-7 model against the experimental results from an independent kinome-wide siRNA study in the MCF-7 breast cancer cells [48] . The knock-down data were generated using a Methylene blue assays to assess cancer cell density in order to evaluate the quality of their siRNA screen ( Figure S2 and Table S2 in [48] ) . The siRNA screen was designed to target 712 kinases in the human kinome , with three distinct siRNAs per kinase . The data was analyzed using the R package cellHTS2 [49] , where a mean Z-score scaled by the per-plate median of the intensities of the negative controls was calculated for each kinase . A large positive Z-score indicates a strong inhibition effect and thus indicates high essentiality of the kinase for the cancer cell survival . Here , we tested the essentiality of the kinases in the cancer-specific target set predicted by TIMMA using the 15 drugs targeting a total of 384 kinases . In other words , we asked the question: are the kinases selected by TIMMA as the most predictive of anticancer efficacy also highly essential individually for the cancer cell viability ? The optimal target set found by TIMMA included 12 kinases {ZAK , CSF1R , GAK , MEK5 , ABL2/EPHA8 , ALK/LTK/PLK4/ROS1 and MEK1/MEK2} , with a mean LOO prediction error of 0 . 1392 ( Dataset S5 ) . The/symbol stands for the targets that are inhibited by the same set of drugs in the data and thus are indistinguishable by the model . The mean Z-score for these 12 kinases was 0 . 926 , which is significantly higher than the average Z-score for random sets of 9 kinases selected from the 712 kinases ( Figure 6A , permutation test , p = 0 . 0015 ) . This shows that TIMMA tends to choose , in general , such kinases that are also individually more effective in blocking cancer cell growth . Among these kinases , ALK had the highest predicted single-kinase efficacy . ALK was also identified in the independent siRNA screen as the top essential kinase . However , our model does not assume that all the kinases in the optimal target set are essential individually . For instance , GAK and ROS1 had a relatively low Z-score , but still these were considered to have an important role in the cancer survival and/or proliferation process when combined with the other selected kinases ( Figure 6B ) . On the basis of the predicted efficacy matrix based on the selected kinase targets ( Dataset S5 ) , we derived the multiplicative synergy score ( Eq . 9 ) for the drug pairs that are pairwise inhibiting the selected targets ( Supplementary Table S1 ) . We found that the top synergistic drug pairs are mainly GAK and ALK/LTK/PLK4/ROS1 inhibitors , some of which have been reported in the recent literature . For example , crizotinib combined with erlotinib has recently been shown to cause a complete and genotype-specific inhibition of tumor growth in non-small cell lung cancer ( NSCLC ) adenocarcinoma patient-derived pre-clinical treatment models in vivo [50] . Crizotinib-erlotinib combination was also ranked as the top one among the 12 drugs that are available in the MCF-7 model analysis , indicating that such a combination might also be effective for the treatment of specific resistive subtypes of breast cancer . Similarly , TAE-684 , a potent ALK inhibitor has been found to provide selective activity against those mutations that conferred crizotinib resistance in cancer patients [51] , suggesting a mechanistic insights into the crizotinib-TAE-684 combination , which was ranked as the second most synergistic pair by our model predictions . In general , the top-predicted synergistic drug pairs are not necessarily the individually most sensitive drugs , as their individual efficacies do not correlate with the multiplicative synergy score ( Supplementary Table S1 ) . To visualize the combinatorial effect of the selected kinase targets , a target inhibition network was constructed by applying a threshold of 0 . 318 to binarize the predicted efficacy ( Figure 7 , Dataset S5 ) . The threshold 0 . 318 was the scaled drug efficacy for crizotinib that inhibits ALK , which is the most essential kinase according to the siRNA screen and thus considered as effective in treating MCF7 cancer cells . The target inhibition network suggested that two parallel MEK1/2-dependent pathways as most important for the MCF-7 cancer cell survival . For example , simultaneous targeting of CSF1R and ALK/LTK/PLK4/ROS1 was predicted to enable blocking the two redundant pathways and result in a complete inhibition of the MEK1/2-dependent cell proliferation . Notably , CSF1R has been shown to act upstream of MEK1 and to induce Cyclin D2 expression via the Ras/Raf/MAPK pathway [52] . Similarly , ALK has been suggested to directly activate MEK1/2 , independent of c-Raf [53] . Also , LTK has been implicated in cell growth via MAPK signaling [54] . Taken together , these findings support the idea that inhibition of both CSF1R and ALK/LTK/PLK4/ROS1 should have a synergistic effect on the cell survival . Indeed , the combination of sorafenib and crizotinib , inhibitors of CSF1R and ALK/LTK , respectively , has been considered for a clinical trial for treating advanced solid tumors ( Pfizer , ClinicalTrials . gov , Identifier: NCT01441388 ) . To further show the applicability of TIMMA to such cases where combinatorial effects of kinase inhibition are considered , we utilized the results from a kinome-wide drug sensitization screen , in which the kinase siRNA-silencing was combined with the treatment of Aurora kinase inhibitors in BxPC-3 pancreatic cancer cell line [55] . Aurora kinases ( Aurora A , Aurora B , and Aurora C ) are serine/threonine kinases that are frequently overexpressed in many tumors . Accordingly , Aurora kinase inhibition has been proposed as potential cancer therapy to disrupt cancer cell division . The purpose of the study was to identify those kinases that when silenced would sensitize pancreatic cancer cells to the Aurora kinase inhibitor treatments . The RNAi screen was done using the Human Validated Kinase Set ( HVKS ) siRNA library from Qiagen , with two siRNAs per kinase . A total of 17 kinases were identified and confirmed in a validation screen to have at least 2 out of 4 siRNA sequences showing greater than 1 . 5-fold decreases in EC50 or EC30 values of the Aurora kinase inhibitor AKI-1 in dose-response curves [55] . We wanted to evaluate here the TIMMA model performance in predicting the experimental results in [56] , especially the kinases that would sensitize the pancreatic cancer cells to the AKI-1 treatment . This question can be addressed in TIMMA by determining the synthetic lethality score for such kinases paired with the targets of AKI-1 . The synthetic lethality score ( Eq . 11 ) was calculated for the kinase pairs using the data of 15 drugs and 384 kinases and the drug efficacy in BxPC-3 cells [31] . The higher the score , the stronger the synthetic lethality effect for the kinase pair . Of these 15 drugs , 3 drugs ( CHIR-265/RAF-265 , nilotinib and PD0332991 ) were not tested for BxPC-3 and thus were removed ( Dataset S3 ) . Since none of the 12 compounds effectively targeted the two Aurora kinases , Aurora A and Aurora C , we considered here the Aurora B kinase as the only effective target of AKI-1 . The TIMMA model was therefore tested on all those kinase pairs which contain Aurora B , and those kinase pairs whose synthetic lethality scores were higher than that of {Aurora B , Aurora B} pair were considered as synthetic lethal partners of Aurora B . The TIMMA analysis based on Eq . 11 identified 19 kinases ( multiple kinases are ranked the same as they are targeted by the same drug set ) , which showed stronger synthetic lethality interactions with Aurora B than with itself ( Figure 8 ) . Two ( MET , PDGFRA ) out of the three targets ( MET , PDGFRA and PYK2 ) were experimentally validated as sensitizing targets of AKI-1 in the pancreatic cancer , representing a highly significant enrichment ( hypergeometric test , p = 0 . 0046 ) ( Figure S4 in [55] ) . In addition , the model predicted that PDGFRB might also be a potential sensitizer of AKI-1 treatment . Similar to the result in the MCF-7 cells , ZAK ( ranked 3rd ) , MEK5 ( ranked 7th ) and GAK ( ranked 9 th ) were again found in the cancer-specific target set for BxPC-3 cells , suggesting that the synergy patterns of these kinases is common across these cancer types . In contrast , the model predicted that the combination of MEK1/MEK2 and AURKB inhibitors has least synthetic lethal capacity ( Dataset S6 ) , because individual essentiality of these two factors favors the series connection model rather than the parallel model in the synthetic lethality score [56] , [57] . The final application case study was the human triple-negative breast carcinoma , where we experimentally validated the TIMMA target combination predictions using single and pairwise siRNA knock-downs on the MDA-MB-231 cells . The TIMMA model selected 20 optimal kinase targets {PLK1 , AURKB , CDKL2 , ZAK , ERBB4 , TEK , TXT/BMX/CSK/EPHA5/EPHB1/EPHB4 , CAMKK1/MAK/VRK2/TNNI3K/CDC2L6/DYRK1B/DYRK1A/TYK2} with an average LOO error of 0 . 11 ( Dataset S7 ) . These kinases and their functional interactions were mapped to the target inhibition network , which contained a total of 8 target nodes ( Figure 9 ) . The kinases belonging to the same node are inhibited by a common set of drugs , and therefore these drug targets are indistinguishable in terms of drug inhibition and their predicted efficacy values . Two of the selected kinase targets , PLK1 and AURKB , are known to be essential for cell growth , therefore serving here as positive controls for the model target predictions . However , due to their known role in cell growth , we excluded these two kinases from the experimental evaluation , and focused on the synergistic combinations between the remaining 18 kinases targets among the 6 target nodes . In general , there were significant differences between the TIMMA-selected kinase targets , when these were silenced either individually or in combination in the siRNA screens , especially after their ranking according to the predicted efficacy ( Figure 10A , Kruskall-Wallis rank sum test , p<10−15 ) . Even after excluding the two essential kinases ( PLK1 and AURKB ) , the 18 TIMMA-selected kinases showed higher cancer cell growth inhibition power in the single knock-down experiments ( 22% increase in cell inhibition ) , compared to the inhibition observed in the kinome-wide single-siRNA screen ( Wilcoxon rank sum test , p = 0 . 28 , Supplementary Table S2 ) . Importantly , the 153 TIMMA-selected kinase pairs resulted in highly significant cancer cell killing improvement in the pairwise knock-down experiments ( 38% increase ) , compared to their single kinase inhibition efficacy ( p = 0 . 0089 , Bonferroni adjustment ) , indicating that TIMMA could select such kinase targets that , in general , are important for cancer cell survival , and especially when combined . Notably , when categorizing the selected target pairs as High and Low efficacy groups , according to their predicted treatment efficacies above or below the average of 0 . 6 , there was a significant increase in the cancer cell growth inhibition percentages ( 23% , 48% and 80% ) , when comparing the High efficacy group to either the Low efficacy group , the single selected kinases or the kinome-wide background ( p = 0 . 031 , p = 0 . 013 , p<10−15 , Bonferroni adjustment , Supplementary Table S2 ) . Taken together , these results indicate that the TIMMA model can effectively select and prioritize among the massive number of possible combinations those target combinations that are most potential for experimental testing or eventual clinical translation . To investigate whether the model can select also such drug target combinations that individually show relatively low drugs efficacies , but will lead to increased drug synergy when combined , we focused on the set of 15 kinase pairs among the 6 target nodes ( {CDKL2 , ZAK , ERBB4 , TEK , TXT/BMX/CSK/EPHA5/EPHB1/EPHB4 , CAMKK1/MAK/VRK2/TNNI3K/CDC2L6/DYRK1B/DYRK1A/TYK2} , Figure 9 ) that are unique in terms of their drug profiles and thus distinguishable based on their TIMMA-predicted efficacy . We took an average of the synergy scores for those kinas pairs that are represented by the same target node pair . The synergy score calculated on the basis of the TIMMA-predictions correlated significantly with the synergy calculated on the basis of the single and pairwise siRNA measurements ( Kendall correlation 0 . 39 , p = 0 . 0463 ) . When mapping the selected kinase target pairs to the available kinase inhibitor pairs , i . e . using Eq . 10 , the correlation between the predicted and measured synergies improved further ( Figure 10B , p = 0 . 0002 ) . In particular , when using a cut-off predicted synergy of 0 . 36 ( the dotted vertical line ) , the likelihood of obtaining a high measured synergy increased significantly ( Wilcoxon rank sum test , p<5 . 9−7 , Bonferroni adjustment ) . Among these top-20 most synergistic drug combinations for the MDA-MB-231 cells , there were a number of examples , such as the two top pairs , where the efficacy of one of the drugs in the combination was relatively low , or even zero , yet the predicted and measured synergy for the drug combination was high ( Table 1 ) . This demonstrates that our model is able to predict not only those pairs that are essential either individually or in combination , but also a number of synergistic combinations , where the predicted efficacy cannot be explained by the efficacy of the two single compounds when used alone ( Supplementary Figure S4 ) .
In this study , we utilized the principles of polypharmacological target inhibition modeling as a generic framework for pinpointing cancer-specific targets and predicting the effect of putative drug combinations . The main contribution of the present work was to introduce a novel model construction model , called TIMMA , and to demonstrate its feasibility in systematic investigation of the model predictions using kinome-wide single and pairwise siRNA knock-down experiments . We also showed that our enhanced model construction algorithm resulted in significantly better predictive accuracy and computational efficiency , compared with an existing algorithmic solution . With such improvements , the number of targets that can be included in the minimal set can go up to 20 , which corresponds to maximally 20 drugs in a combination . In the three case studies , where we combined large-scale drug sensitivity screening and comprehensive drug-target data , we were able to identify a number of potential drug combinations for breast and pancreatic cancers . In more general terms , the optimized experimental-computational approach , empowered by the target inhibition network , allowed us to systematically explore how the kinase inhibitors and their cellular targets interact to modulate cancer growth phenotype on a global network-level , with the aim to identify molecular pathways behind drug action , as well as to suggest combinatorial treatment strategies that can block the cancer escape pathways and therefore tackle the resistance problem of the many current treatments approaches . Network-based strategies , such as the one developed in the current work , provide a principled approach to systematically identify the key set of druggable vulnerabilities of cancer networks . Such efforts create a solid foundation towards implementing the emerging paradigm in drug discovery , the so-called ‘network pharmacology’ [3] , which provides a more global understanding of the mechanism behind drug action and resistance by considering drugs and targets in their context of cellular networks and pathways . The current work also support the detection of synthetic lethal interactions , which is another conceptual framework recently proposed toward developing more effective therapeutic strategies [12] , [13] , [15]–[19] . More specifically , targeted perturbation or inhibition of a gene that has a synthetic lethal relationship with a driving cancer mutation holds great promise for being a highly specific and selective means to kill cancer cells without severe side-effects to normal cells . Compared to the conventional cytotoxic drugs , that affect both normal and cancerous cells , synthetic lethality can therefore address the fundamental challenges of anticancer therapy by optimally targeting differential features in each cancer type while sparing normal cells . However , despite the advances in siRNA and compound screening , synthetic lethal interactions between genes and/or drugs have remained extremely difficult to predict on a global scale [13] , [18] . Network-based methods provide a convenient platform to finding functional interactions in disease networks , toward enabling identification of such effective drug targets and their combinations that tailored for more effective and personalized cancer medicine . We focused here on the kinase targets because of their importance in many multi-target cancer treatment developments . This is also why we experimentally validated the model predictions using kinome-wide single siRNA and TIMMA-predicted pairwise siRNA screens , where the selected kinase targets were knocked down individually or in pairs in the given cell type to experimentally evaluate their essentiality either alone or in combination for the cancer cell survival . However , the same modeling principles could be applied also to other target families , such as enzymes or G protein coupled receptor ( GPCR ) targets , provided there will be enough target and drug promiscuity to allow for construction of the target inhibition networks . Moreover , while the siRNA silencing screens are convenient for the drug target investigation , the perturbation effects from the siRNAs cannot fully mimic the phenotypic effects of drug treatments . RNAi has also potential limitations due to potential off-target silencing effects and variable reagent efficacy , which may also partly explain the observed discrepancies between the drug treatment-based model predictions and their siRNA-based experimental validations . Therefore , one of our future aims is to apply the TIMMA model predictions to designing potential drug combination treatments , initially in various cancer cell models in vitro , and later also in primary samples from cancer patients ex-vivo . The drug treatments are also closer to the eventual translation of the model predictions in a clinical setup , at least until the RNAi-mediated target silencing has become safe and efficacious enough for clinical applications . In an effective combinatorial setting , one needs to modulate a set of targets to achieve maximal efficacy , while avoiding others to reduce the risk of side effects . The current TIMMA algorithm addressed the first challenge: the optimal efficacy by multi-target modulation . The different model parameters and thresholds lead to a multiple candidate target inhibition networks for combinatorial treatments . From those candidate models , clinician could then ideally choose the combination that is most feasible and results in less known adverse effects , based on prior knowledge . Although there are information sources on drug side effects scattered around in databases , such as SIDER [58] , ChEMBL [32] , and PROMISCUOUS [59] , we chose not to try to incorporate the side effect information in the current model building , because such information is still missing for many targeted drugs and the initial aim was to find effective target combinations . However , incorporating known side effect or toxicity information of drugs and their targets will be an important topic of future research . Possible approaches for such modifications include , for instance , usage of metabolic networks and pathways that are targeted by drugs [60] , or combining multiple databases that contain a collection of drug features , such as medical indications , molecular targets , toxicity profiles or anatomical therapeutic and chemical classifications [61] . Further , rather than using a single response readout for drug efficacy , such as IC50 , AA or DSS , the gene expression or metabolomic changes after a treatment could also be included as part of the drug response profiles , perhaps leading to be more comprehensive drug-disease networks in the future . | Selective inhibition of specific panels of multiple protein targets provides an unprecedented potential for improving therapeutic efficacy of anticancer agents . We introduce a computational systems pharmacology strategy , which uses the concept of target inhibition networks to predict effective multi-target combinations for treating specific cancer types . The strategy is based on integration of two complementary information sources , drug treatment efficacies and drug-target binding affinities , which are readily available in drug screening labs . Compared to the cancer sequencing efforts , which often result in a huge number of non-targetable genetic alterations , the target combinations from our strategy are druggable , by definition , hence enabling more straightforward translation toward clinically actionable treatment strategies . The model predictions were experimentally validated using siRNA-mediated target silencing screens in three case studies involving MDA-MB-231 and MCF-7 breast cancer and BxPC-3 pancreatic cancer cells . In more general terms , the cancer cell-specific target inhibition networks provided additional insights into the drugs' mechanisms of action , for instance , how the cancer cell survival pathways can be targeted by synergistic and synthetic lethal interactions through multi–target perturbations . These results demonstrate that the principles introduced here offer the possibilities to move toward more systematic prediction and evaluation of the most effective drug target combinations . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [] | 2013 | Target Inhibition Networks: Predicting Selective Combinations of Druggable Targets to Block Cancer Survival Pathways |
Low adherence to multidrug therapy against leprosy ( MDT ) is still an important obstacle of disease control , and may lead to remaining sources of infection , incomplete cure , irreversible complications , and multidrug resistance . We performed a population-based study in 78 municipalities in Tocantins State , central Brazil , and applied structured questionnaires on leprosy-affected individuals . We used two outcomes for assessment of risk factors: defaulting ( not presenting to health care center for supervised treatment for >12 months ) ; and interruption of MDT . In total , 28/936 ( 3 . 0% ) patients defaulted , and 147/806 ( 18 . 2% ) interrupted MDT . Defaulting was significantly associated with: low number of rooms per household ( OR = 3 . 43; 0 . 98–9 . 69; p = 0 . 03 ) ; moving to another residence after diagnosis ( OR = 2 . 90; 0 . 95–5 . 28; p = 0 . 04 ) ; and low family income ( OR = 2 . 42; 1 . 02–5 . 63: p = 0 . 04 ) . Interruption of treatment was associated with: low number of rooms per household ( OR = 1 . 95; 0 . 98–3 . 70; p = 0 . 04 ) ; difficulty in swallowing MDT drugs ( OR = 1 . 66; 1 . 03–2 . 63; p = 0 . 02 ) ; temporal non-availability of MDT at the health center ( OR = 1 . 67; 1 . 11–2 . 46; p = 0 . 01 ) ; and moving to another residence ( OR = 1 . 58; 95% confidence interval: 1 . 03–2 . 40; p = 0 . 03 ) . Logistic regression identified temporal non-availability of MDT as an independent risk factor for treatment interruption ( adjusted OR = 1 . 56; 1 . 05–2 . 33; p = 0 . 03 ) , and residence size as a protective factor ( adjusted OR = 0 . 89 per additional number of rooms; 0 . 80–0 . 99; p = 0 . 03 ) . Residence size was also independently associated with defaulting ( adjusted OR = 0 . 67; 0 . 52–0 . 88; p = 0 . 003 ) . Defaulting and interruption of MDT are associated with some poverty-related variables such as family income , household size , and migration . Intermittent problems of drug supply need to be resolved , mainly on the municipality level . MDT producers should consider oral drug formulations that may be more easily accepted by patients . Thus , an integrated approach is needed for further improving control , focusing on vulnerable population groups and the local health system .
Leprosy control is based on early diagnosis , treatment , and cure , aiming at the elimination of sources of infection and of sequels in affected individuals . Similar to other countries , in Brazil leprosy control measures are integrated into general public health care , thus facilitating access to affected individuals and reduction of disease-related stigma [1] . Interruption and defaulting of multidrug therapy against leprosy ( MDT ) are still important obstacles of disease control in many endemic countries , with consequences for both patients and the control programs: low adherence is responsible for potentially remaining sources of infection , incomplete cure , and irreversible complications , and in addition may lead to multidrug resistance [2] . In Brazil , the number of patients defaulting treatment was reduced from 3 , 148 individuals in 2002 to 529 in 2009 ( with approximately 49 , 000 and 37 , 500 new cases , respectively ) [3] . The causes leading to low adherence and non-compliance to MDT are diverse and may include socio-economical , cultural , psychosocial , behavioral , drug-related and disease-related factors , as well as health service-related aspects [2] , [4]–[9] . For example , a recent study from India identified stigma as the most common reason given by defaulters , but failed to detail data and to compare these factors with non-defaulters [4] . In Paraíba State in the northeast of Brazil , defaulting of MDT was associated with regular alcohol use , but not with clinical characteristics [5] . However , that study involved only 13 patients who defaulted , as compared to 28 patients finishing treatment regularly . Here we present - as part of a major epidemiological investigation in 78 municipalities in Brazil - population-based data to further investigate factors associated with interruption and defaulting of MDT in a hyperendemic area .
Tocantins State is located in the central savannah region of Brazil ( Figure 1 ) . The state has been created in 1988 and has a total population of 1 , 3 million ( 2009 ) , distributed throughout 139 municipalities; 83% of the municipalities have less than 10 , 000 inhabitants . Tocantins is hyperendemic for leprosy: in 2009 , a total of 1 , 345 new cases were notified , and the detection rate was 88 . 54/100 . 000 inhabitants . The present study is part of a major epidemiological investigation performed in 79 municipalities of northern Tocantins . These municipalities are at highest risk for leprosy transmission , according to a recent cluster analysis performed by the Brazilian Ministry of Health ( Figure 1 ) [10] , [11] . The target population included all individuals newly diagnosed with leprosy from 2006–2008 , living and notified as leprosy cases in these municipalities . We excluded the municipality of Araguaína from the present analysis , the biggest city in the region with about 120 thousand inhabitants . Araguaína has a leprosy reference clinic and shows different characteristics , as compared to the other smaller municipalities that share mainly rural characteristics . These results will be published elsewhere . We also excluded patients who moved to municipalities outside the endemic cluster , suffered from mental disability or who have shown other characteristics that impeded an interview , such as individuals under the influence of alcohol . Relapsed leprosy cases were also excluded . Individuals who had died after diagnosis were not included in data analysis . The 78 Municipal Health Secretariats were informed by the Tocantins' State Health Secretariat about the study and the timeframe when the team would perform field visits for data collection . Previous to field visits , the target population was identified in the database of the National Information System for Notifiable Diseases ( Sistema de Informação de Agravos de Notificação – SINAN ) . In the municipalities , the patients' charts and the local notification records were first reviewed regarding clinical variables ( clinical form , operational classification , disability grade at diagnosis , mode of case detection , date of diagnosis , date of release from treatment and date of last appearance at health center for treatment ) . If in the local records patients were identified that had not been notified , we included them in the target population . Then , affected individuals were invited by community health agents to be interviewed at the local health care center . If individuals did not present at the health care center , we performed home visits accompanied by local community health agents . Data were obtained at this occasion according to a previously defined framework , using pre-tested structured questionnaires . The framework comprised of four blocks of independent variables possibly associated with the outcomes: 1 . Socio-demographic block ( gender , age , marital status , education , residence area , number of rooms , number of persons per household , household income , migration ) ; 2 . Disease-related block ( clinical form of disease , operational classification , disability grade , leprosy reaction , adverse events to MDT , difficulty swallowing MDT drug ) ; 3 . Health service-related block ( mode of case detection , non-availability of MDT drugs , distance to health care center , perceived difficult access to health care center ) ; 4 . Knowledge , attitudes and practices block ( alcohol consumption , information of peer persons regarding disease , knowledge on leprosy and cure ) . Data were collected from September to December 2009 . To reduce inter-observer bias , all questionnaires where applied by two previously trained field investigators ( OAC , ARO ) who were supervised during the entire study . Data from patients' charts were collected by another two investigators ( KH , FW ) . Extensive pre-tests were performed under supervision . Data were entered twice , using Epi Info software version 3 . 5 . 1 ( Centers for Disease Control and Prevention , Atlanta , USA ) and cross-checked for entry-related errors . Answers to open-ended questions were grouped according to similarities and categorized for bivariate analysis . Open-ended questions included information on clinical characteristics for definition of leprosy reaction and adverse events; and questions on knowledge , attitudes and practices . Data analysis was done using STATA version 9 ( Stata Corporation , College Station , USA ) . As the number of individuals defaulting MDT was relatively low , two separate bivariate analyses were performed , with two different outcomes based on the non-attendance of patients at treatment centers: Variables were first analyzed and presented in a bivariate manner . Odds ratios and their respective 95% confidence intervals are given . We applied Fisher's exact test to estimate significance of the difference of relative frequencies . Continuous and discrete variables were not normally distributed and thus compared applying the Wilcoxon rank sum test for unmatched data . Unconditional logistic regression analysis using backward elimination was then performed to calculate adjusted odds ratios for the independent association between 1 ) interruption of; and 2 ) defaulting MDT , and the respective explanatory variables . Results of both analyses are presented separately . In addition to sex , age and leprosy form ( PB/MB ) which we used as adjusting variables throughout multivariate analysis , variables with a p value<0 . 25 in the Fisher's exact test were entered into the initial regression models , and then backward elimination was run . To remain in the model , a significance of p<0 . 05 was required . Variables were checked for collinearity . Confounding and interaction between variables were also investigated by stratification and by constructing 2×2 tables . All variables that remained in the final models are presented , and odds ratios were adjusted for all other variables in the respective model . The study was approved by the Ethical Review Board of the Federal University of Ceará ( Fortaleza , Brazil ) and by the Ethical Review Board of Lutheran University of Palmas ( Tocantins , Brazil ) . Permission to perform the study was also obtained by the Tocantins State Health Secretariat , the State Leprosy Control Program and the municipalities involved . Informed written consent was obtained from all study participants after explaining the objectives of the study . In the case of minors , consent was obtained from a caretaker . Interviews were always performed separately to guarantee strict privacy , and the diagnosis of leprosy was not given to family members or other community members , in case the patient had not revealed the diagnosis . If any leprosy-associated pathology was observed during the interview or during clinical examination ( data of clinical examination to be published elsewhere ) , participants were referenced to the responsible health care service .
Of the target population of 1635 individuals from 78 municipalities , 936 ( 57 . 2% ) from 74 municipalities were included in data analysis; one municipality did not diagnose a single case of leprosy in the study period , and another three municipalities had few cases , but no participants were included ( non-consent or not encountered ) . Twelve patients refused to participate in the study . We excluded another 13 ( five under of influence of alcohol that impeded an interview; four convicted; three severely sick who were hospitalized; and one due to advanced age ) . In addition , 674 were not encountered even after home visits , were not known at the local health centers , or had moved to another city outside the cluster . For the analysis of interruption of MDT 130 individuals were excluded ( 92 did not have information about date of the beginning of treatment or last date of supervised monthly dose in the health care center , and 38 were classified as MB leprosy with treatment started <13 months before data collection ) . Thus , data analysis regarding defaulting included 936 , and regarding interruption 806 individuals . Information from patients' charts was available in 894 of cases . Of the total of 936 individuals , 491 ( 52 . 5% ) were males; the age ranged from 5 to 98 years ( mean = 42 . 1 years; standard deviation: 18 . 8 years ) . Two-hundred and twenty-five ( 24 . 0% ) were illiterate . Median monthly family income was R$ 465 ( about 270 USD at the time of the study; interquartile range: R$ 300–R$ 900 ) . In total , 497 ( 55 . 6% ) were classified as PB leprosy , and 395 ( 44 . 1% ) as MB . We identified 28 ( 3 . 0% ) patients who defaulted MDT; 16 defaulters were included by reviewing the SINAN data information system , and an additional 12 locally in the patients' charts . Only 5 individuals were in the both databases . In total , 147/806 ( 18 . 2% ) interrupted MDT . Factors associated with interruption of MDT are detailed in Table 1 . Moving to another residence after diagnosis and living in a small residence were significantly associated with interruption . In addition , disease- and health service-related variables ( difficulty in swallowing MDT drug; temporal non-availability of MDT drugs ) were significantly associated with an increased chance of interruption of treatment ( Table 1 ) . Interestingly , disease-related factors such as the clinical form , presence of leprosy reactions or occurrence of adverse events to MDT did not play a significant role . Figure 2 depicts the frequency of interruption of MDT , stratified by age groups and gender . In general , the 16–30 year-olds showed the highest chance of interruption , as compared to all other age groups together ( OR = 1 . 84; 95% confidence interval: 1 . 20–2 . 77; p = 0 . 04 ) . This effect could be mainly attributed to the 16–30 year-old males , who showed the highest frequency of interruption ( 34 . 4% ) , roughly a two-fold difference to females of the same age group ( 17 . 6%; p = 0 . 01; Figure 2 ) . Logistic regression analysis identified temporal non-availability of MDT drugs at the health care center as an independent risk factor for treatment interruption ( Table 2 ) . An increased number of rooms per household ( as an indicator for wealth ) was identified as an independent protective factor . Bivariate analysis of factors associated with defaulting MDT is depicted in Table 1 . Several socio-economic variables ( number of rooms per household; moving to another residence after diagnosis; family income ) were significantly associated with defaulting ( Table 1 ) . Similar to interruption of MDT , disease-related factors did not play a significant role . Health service variables did also not show any significant association . In logistic regression analysis , we identified the number of rooms per residence as a factor independently associated with defaulting , with a protective odds ratio of 0 . 67 for each additional room in the household ( Table 2 ) , but no other factors .
Low adherence to drugs is in general a major obstacle in the control of infectious diseases that require prolonged treatment , such as leprosy and tuberculosis . Our comprehensive population-based study shows that poverty , behavior , drug-related and service-related factors were associated with adherence to MDT , hampering leprosy control in a hyperendemic area in Brazil , and suggest evidence-based actions for improving control measures . It is widely believed that understanding and behavior of patients in relation to drug compliance are largely influenced by their socio-economic condition and level of knowledge; socio-economic factors were previously suggested to influence adherence to MDT [5] , [7] , [13] . Even though family income as a direct indicator of poverty was not significantly associated with low adherence ( but with defaulting ) , number of rooms was identified as an independent risk factor in both bivariate and multivariable analyses . Poverty and its consequences , similar to other neglected tropical diseases , has been shown to be associated with leprosy in general [14] , and our results reflect this complex interaction of causation leading to higher risk of disease in underprivileged populations . In addition , population movements are usually associated with socio-economic conditions in Brazil . In our study , people who had moved to another residence were more vulnerable for low adherence . These people may lose their bonds with community health workers and other health professionals of the primary health care centers , besides other factors that change in life when moving to another place . Similar findings have been made in India and southeast Brazil , where treatment interruption due to migration has been reported [15] , [16] . In the case of tuberculosis , moving to another district with subsequent change of health unit was also shown to increase the risk of defaulting treatment in Uganda [17] . On the other hand , changing residence due to leprosy was clearly not a factor that played a role in our study ( data not shown ) . Interestingly , the frequency of defaulting MDT was relatively low , as compared to other settings [2] , [4] , [13] , [18] , [19] , with a rate of only 3% . In Tocantins , the defaulting rate was 47% in 2005 , but was reduced drastically in subsequent years [20] . This may reflect the success of efforts made in the last years by Tocantins's health services . In fact , the Brazilian national and state leprosy control programs have put a major effort in improving the decentralized primary health care services , with 90% population coverage of the Family Health Program in Tocantins . As another consequence , variables related to health services seemed to play a minor role for defaulting in our study , despite the identification of temporary shortage of drugs as a significant risk factor for interruption of MDT . We have shown previously that the patients of this area answered most commonly to an open-ended question about the reason for interrupting MDT with temporary shortage of drugs at the health care center , but median time of interruption was only 15 days which indicates that this operational issue was usually resolved quickly [21] . In fact , these logistical problems occurred mainly on the municipality level , as MDT provided by the State Leprosy Control Program to the municipalities did not suffer any shortage in the study period ( A . C . F . , unpublished observation ) . In other countries and settings , where leprosy control programs are not yet well established , such as in northern Mozambique , Nigeria and Sudan , health-service related factors play a more crucial role [4] , [7] , [18] , [19] , [22] . Our data also indicate that in a setting with an established leprosy control program , clinical variables are of minor importance for low adherence to MDT . In case of leprosy reactions , for example , the primary health care services and the reference centers seem to be prepared to cope with the situation . Similarly , previous studies from northeast Brazil , the Philippines and Nepal suggested that clinical data such as type of leprosy , occurrence of reactions or disability grading at diagnosis would not play a significant role in the given context [2] , [5] , [23] . Difficulty in swallowing drugs was previously suggested as a factor associated with low adherence to MDT [2] . Considering also the long course of treatment , this shows the need for the search of new formulations that may be better accepted by patients . Studies from other parts of the world , mainly from the South Asian and Southeast Asian sub-regions , identified other risk factors for low adherence . For example , in the Philippines adverse events were given by the patients as the most important reason ( 40% ) for defaulting [2] . People in Assam ( India ) who defaulted treatment mentioned loss of occupational hours when going to the health care center ( 33 , 1% ) , adverse events ( 26 , 0% ) and social stigma ( 18 , 1% ) as the most common reasons [13] . About 10 years ago , these factors were identified in a qualitative study from Espírito Santo State in Brazil [16] . Since then , Brazilian control programs have improved considerably , e . g . by performing health education on adverse events and leprosy reactions , by training health care professionals and by improved access of the users to the primary health care system . The results of our study reflect these efforts and highlight the differing situation in other countries . Available evidence on the influence of demographic variables on adherence to treatment is contradictory . Similar to the study from the Philippines [2] , demographic data such as gender , age and civil status were not associated with low adherence in our study population . In contrast , in endemic regions of Nepal and India , more males than females completed treatment , and illiteracy was also significantly associated with low treatment compliance [9] , [13] . However , both studies had some methodological problems , and analysis of data is limited . Interestingly , our study showed highest interruption rates in young males , when data were stratified by gender . This indicates that factors are multifaceted and that in this case , young males , who are generally known to show insufficient health care behavior , should be considered a vulnerable group for low adherence . In fact , the Brazilian Ministry of Health has taken into consideration the special needs of the male population and recently launched an integrative program focusing on male gender issues [24] . Similar to leprosy , tuberculosis needs prolonged treatment and has also shown to reveal problems regarding adherence . Improving adherence to treatment against leprosy can thus be expected to have positive impact also on other diseases , such as tuberculosis . In fact , the factors associated with low adherence to tuberculosis are similar . For example , in Ethiopia , the occurrence of adverse events to tuberculosis treatment was found to be a significant risk factor for defaulting , whereas knowledge about duration of treatment was protective and increased the odds of terminating treatment [25] . A study from Nepal identified distance to health care services and low knowledge on disease and its treatment as risk factors for non-adherence to tuberculosis directly observed short-course ( DOTS ) [26] . An ancillary finding was the detection of incomplete patients' charts and registries in many cases . We detected in total 128 leprosy cases that were not included in the national SINAN database for notifiable diseases , and a considerable number of cases of abandonment from treatment , which had not been registered as such in SINAN . In addition , only in 72 . 1% ( 645/894 ) information on degree of disability at diagnosis was available in the patients' charts . The quality of patients' records and datasets has improved in the past years , but there is still a clear need for more complete data sets and patient charts , as suggested recently in a study performed in northeast Brazil [27] . Though being a population-based study performed in a considerable number of municipalities in a leprosy hyperendemic region , our study is subject to limitations . First , the number of defaulters , as a result of the ongoing leprosy control measures , has been reduced significantly in the past years , and we included only 28 patients who defaulted treatment . This hampered statistical analysis to some degree . Second , non-participation bias , mainly of those who abandoned treatment , may have played a role . Thus , we performed an additional analysis using a less stringent criterion for compliance: interruption of treatment , based on the duration of treatment . However , this analysis did not take into account adherence to drugs taken at home , but was based on appearance at the health care centers for the monthly supervised dose , which should be taken into account in the interpretation of results . Finally , incomplete patients' charts and subsequent missing data hampered analysis regarding clinical variables in some cases . On the other hand , integration of local primary health care professionals and of the State and Municipal Leprosy Control Programs reduced non-participation bias . We conclude that in an area in Brazil where leprosy control actions are well established , adherence to MDT is a result of a complex interaction between different socio-cultural , service-related , drug-related and economical factors . Intermittent problems of drug supply need to be resolved , mainly on the municipality level . MDT producers should consider oral drug formulations that may be more easily accepted by patients . An integrated approach is needed to further improve adherence and other aspects of leprosy control , such as early diagnosis , including the stakeholders involved: patients and their families , health care professionals , and policy makers [6] , [28] , [29] . Improved adherence to MDT will further improve the leprosy control programs and in addition minimize the risk of possibly upcoming drug resistance . | Leprosy is still a public health problem in Brazil , and low adherence to multidrug therapy against leprosy ( MDT ) is an important obstacle of disease control . This may lead to remaining sources of infection , incomplete cure , complications , and multidrug resistance . We performed a study in 78 municipalities in central Brazil , and interviewed leprosy-affected individuals . In total , 3% of patients defaulted , and 18 . 2% interrupted MDT . Risk factors for interruption of treatment include: reduced number of rooms per household ( OR = 1 . 95; p = 0 . 04 ) ; difficulty in swallowing MDT drugs ( OR = 1 . 66; p = 0 . 02 ) ; temporal non-availability of MDT drugs at health center ( OR = 1 . 67; p = 0 . 01 ) ; and moving residence after diagnosis ( OR = 1 . 58; p = 0 . 03 ) . Defaulting MDT was significantly associated with: reduced number of rooms per household ( OR = 3 . 43; p = 0 . 03 ) ; moving to another residence ( OR = 2 . 90; p = 0 . 04 ) ; and low family income ( OR = 2 . 42; p = 0 . 04 ) . Our study shows that defaulting and interruption of MDT against leprosy are associated with some poverty-related variables such as family income , household size , and migration . Intermittent problems of drug supply need to be resolved , mainly on the municipality level . MDT producers should consider drug formulations that are more easily accepted by patients . An integrated approach is needed for further improving control , focusing on most vulnerable population groups and the local health system . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"infectious",
"diseases",
"public",
"health",
"and",
"epidemiology",
"epidemiology",
"infectious",
"disease",
"epidemiology",
"neglected",
"tropical",
"diseases",
"leprosy",
"infectious",
"disease",
"control",
"public",
"health"
] | 2011 | Interruption and Defaulting of Multidrug Therapy against Leprosy: Population-Based Study in Brazil's Savannah Region |
Despite considerable research efforts , little is yet known about key epidemiological parameters of H5N1 highly pathogenic influenza viruses in their avian hosts . Here we show how these parameters can be estimated using a limited number of birds in experimental transmission studies . Our quantitative estimates , based on Bayesian methods of inference , reveal that ( i ) the period of latency of H5N1 influenza virus in unvaccinated chickens is short ( mean: 0 . 24 days; 95% credible interval: 0 . 099–0 . 48 days ) ; ( ii ) the infectious period of H5N1 virus in unvaccinated chickens is approximately 2 days ( mean: 2 . 1 days; 95%CI: 1 . 8–2 . 3 days ) ; ( iii ) the reproduction number of H5N1 virus in unvaccinated chickens need not be high ( mean: 1 . 6; 95%CI: 0 . 90–2 . 5 ) , although the virus is expected to spread rapidly because it has a short generation interval in unvaccinated chickens ( mean: 1 . 3 days; 95%CI: 1 . 0–1 . 5 days ) ; and ( iv ) vaccination with genetically and antigenically distant H5N2 vaccines can effectively halt transmission . Simulations based on the estimated parameters indicate that herd immunity may be obtained if at least 80% of chickens in a flock are vaccinated . We discuss the implications for the control of H5N1 avian influenza virus in areas where it is endemic .
Highly pathogenic avian influenza virus strains of the H5 or H7 subtypes are noted for being highly contagious among various bird species and inducing high mortality rates in poultry . Although outbreaks of highly pathogenic avian influenza have been reported since the 1950s the current focus is on the H5N1 subtype . The first outbreaks of H5N1 were reported in Hong Kong in 1997 [1]–[2] . Since then the virus has spread to South East Asia , Africa , and Europe . The outbreaks in Europe were controlled by rapid depopulation of infected premises , pre-emptive culling of neighbouring farms , movement restrictions , and zoo-sanitary measures [3]–[5] . In Asia , however , the disease has become endemic , and control by means of culling in conjunction with movement restrictions and zoo-sanitary measures is both infeasible socio-economically and unlikely to result in elimination [6]–[10] . Therefore , vaccination is the most widely used containment strategy . For instance , in Indonesia alone more than 400 million of vaccine doses have been administered since 2004 . Despite the fact that aspects of H5N1 avian influenza biology have been studied in detail , ranging from molecular studies of host range factors , phylogenetic analyses aimed at unravelling the virus' evolutionary pathways , surveillance of H5N1 in wild birds , studies into the clinical course of H5N1 infections in humans , and vaccine efficacy and safety studies , there is scant information of the basic epidemiological characteristics of H5N1 viruses in their avian hosts . Specifically , little is known about the infectious period of H5N1 in various host species , the duration of the latent period , and the transmissibility of the virus from bird to bird . For a proper understanding of the transmission dynamics of the virus and to be able to assess the potential impact of control measures such as vaccination , however , this information is crucial . For instance , it is well-known that both the invasion prospects of the virus as well as the number of individuals ultimately infected are critically affected by the ( distribution of the ) infectious period and transmission parameter . The ( distribution of the ) period of latency is also of importance since it is a key factor affecting the initial growth rate and duration of an epidemic [11]–[13] . Here we present and analyze experimental transmission studies with highly pathogenic H5N1 avian influenza virus ( A/Chicken/Legok/2003 ) in chickens to obtain quantitative estimates of key epidemiological parameters . Specifically , we performed experiments in which an artificially infected chicken was placed in a cage with a susceptible contact bird , and in which the transmission chain was monitored by taking daily samples from the trachea and cloaca [14]–[16] . The samples were subsequently tested for the presence of virus by egg-culture . In addition , blood samples were taken weekly to determine the antibody response to infection . In all , two experiments , each containing 11 trials , were carried out with unvaccinated chickens , two experiments of 11 trials were performed using an H5N1 inactivated oil emulsion vaccine which contains a strain that is identical to the challenge virus ( A/Chicken/Legok/2003 ) , and four experiments of 11 trials were carried out with two heterologous H5N2 inactivated oil emulsion vaccines ( A/Turkey/England/N28/73 and A/Chicken/Mexico/232/94/CPA ) that are both genetically and antigenically distant from the challenge virus . The experiments are analyzed by tailored statistical methods based on a SEIR ( susceptible-exposed-infectious-removed ) epidemiological model . In this way all estimated parameters have a clear-cut biological interpretation ( mean and variance of the latent and infectious period , transmission rate , reproduction number ) . Here we use two different methods of analysis . The first uses final size data i . e . the number of birds that are ultimately infected , and is aimed at estimation of the reproduction number [17] . The second approach uses all available information and is based on Bayesian inference that relies on Markov Chain Monte Carlo ( MCMC ) techniques [18]–[21] . This allows one to estimate not only the reproduction number , but also other epidemiological parameters of interest . The main advantage of our controlled experimental setup over field studies [6] , [7] , [22] is that the parameters of interest can be estimated with high precision using a limited number of birds . In addition , our controlled experimental setup makes it possible to ascribe differences between control and treatment groups directly to the treatment without having to take into account the potential effect of confounding variables ( e . g . , age and size of the birds , stocking density , feeding status ) .
All inoculated unvaccinated birds ( Tables 1 and 2 ) showed signs of infection ( depression , labored breathing ) , shed virus from both the trachea and cloaca ( apart from a single bird in Table 2 ) , and died within a few days after infection ( range: day 2–day 3 ) . Furthermore , all contact birds died on day 4 or on day 5 after infection of the inoculated bird , indicating rapid infection as well as rapid progression of the disease towards death . In the experiments with unvaccinated birds 8 out of 22 birds escaped infection . These birds did not show signs of disease , did not shed detectable virus , and remained serologically negative when tested in the HI assay at days 7 and 14 . In the experiments with vaccinated birds no contact birds were infected and only a few of the inoculated birds shed virus on just a few days . In fact , only 7 out of 66 inoculated birds shed virus for a total of 12 days . Of these , virus was isolated from the trachea only on 11 days and from the trachea and cloaca on a single day . None of the vaccinated birds died in the course of the experiments , and no signs of disease were observed in any of the vaccinated birds . Details of the vaccination experiments are given in Tables S2 , S3 , S4 , S5 , S6 , and S7 . With regard to transmission , the final size analyses indicate that there are significant differences between the experiments with unvaccinated and vaccinated birds . Table 3 summarizes the results . For the low-dose experiment with unvaccinated birds estimates of the reproduction number are 9 . 0 ( 95% confidence interval ( CI ) : 1 . 9–86 ) in case of an exponentially distributed infectious period and 3 . 4 ( 95%CI: 1 . 3–7 . 6 ) in case of a fixed infectious period . For the high-dose experiment the estimates of the reproduction number are 1 . 7 ( 95%CI: 0 . 40–6 . 6 ) in case of an exponentially distributed infectious period and 1 . 2 ( 95%CI: 0 . 37–2 . 9 ) in case of a fixed infectious period . Although the difference between the two experiments is not statistically significant ( p = 0 . 23 ) [23] , it does hint at the possibility of a role of the inoculation dose in impacting on the transmission dynamics . If all experiments with unvaccinated birds are combined the outcome is that 14 out of 22 initially susceptible contact birds are infected . In this case the estimates of the reproduction number are 3 . 5 ( 95%CI: 1 . 4–9 . 6 ) and 2 . 0 ( 95%CI: 1 . 0–3 . 5 ) , assuming exponentially distributed and fixed infectious periods respectively , indicating that the virus is able to spread epidemically in unvaccinated populations . No transmission was observed in all six experiments with vaccinated birds , resulting in a maximum likelihood estimate of the reproduction number of 0 . The ( two-sided ) 95% confidence interval ranges from 0–0 . 80 or 0–0 . 67 , depending on the assumptions regarding the distribution of the infectious period . Furthermore , the null-hypothesis that the reproduction number is larger than the threshold value 1 can safely be rejected ( p = 0 . 011 in case of an exponentially distributed infectious period , and p = 0 . 0041 in case of a fixed infectious period ) . Hence , it is unlikely that an epidemic can occur in vaccinated populations . The experiments with unvaccinated birds are analyzed using Bayesian methods to obtain estimates of the transmissibility of the virus and the distributions of the latent and infectious periods . Table 4 and Figures 1–3 , S1 , S2 , S3 , S4 , and S5 summarize the main findings . In the low-dose experiment ( Table 1 ) as well as the high-dose experiment ( Table 2 ) the estimated mean of the latent period ( or , more precisely , the median of the marginal posterior distribution of the parameter determining the mean ) is small ( 0 . 20 ( day ) or 0 . 44 ( day ) ) , as is the variance of the latent period ( 0 . 044 or 0 . 078 ) ( Figures S1 and S2 ) . The estimated means of the transmission parameter are also comparable in the two experiments , ranging from 0 . 74 ( day−1 ) in the high-dose experiment to 0 . 80 ( day−1 ) in the low-dose experiment . With regard to the infectious period , however , there appear to be differences between the low- and high-dose experiments , with the birds in the low-dose experiment having a substantially longer infectious period . In fact , the estimated mean of the infectious period is 2 . 5 ( day ) ( 95%CI: 2 . 2–2 . 8 ( day ) ) for the low-dose experiment , and 1 . 3 ( day ) ( 95%CI: 0 . 92–1 . 8 ( day ) ) for the high-dose experiment . In both experiments the estimated variance of the infectious period is small ( 0 . 16 or 0 . 13 ) , indicating that the infectious period distribution is narrowly centered around the mean . We then analyzed the data of the low- and high-dose experiments simultaneously to obtain more precise estimates of the parameters of interest . We considered three scenarios ( labeled by B , C , and D ) that differ with regard to assumptions on the latent and infectious periods ( see Methods ) . Analysis of the pooled data ( scenario B ) verifies the earlier indications ( scenarios A1–A2 ) that both the estimated mean and variance of the latent period are small , while the estimated mean of the infectious period ( 2 . 1 ( day ) ) lies between the estimated means of the infectious period in the analyses of the low- and high-dose experiments ( Figure 1 ) . In comparison with the separate analyses of the low- and high-dose experiments the estimated variance of the infectious period increases ( 0 . 16 and 0 . 13 in the low- and high-dose experiments versus 0 . 33 in scenario B ) , probably because of the need to accommodate both short ( ∼1 . 5 ( day ) ) and long ( ∼2 . 5 ( day ) ) infectious periods . Alternatively , if the infectious period distributions are allowed to differ between the low- and high-dose experiments ( scenario C ) , then the estimated mean infectious periods as well as the corresponding variance estimates revert to values close to those in the separate analyses of the low- and high-dose experiments ( Figure 2 ) . Based on Bayes factor ( see Methods ) the model that allowed for differences in the infectious periods has substantially higher support ( BF = 21 for the pair of simulations of competing models with the smallest difference in marginal likelihoods ) than the model in which the infectious period distributions in the low- and high-dose experiments are assumed to be equal . Finally , if the latent and infectious period distributions are allowed to differ between inoculated and contact birds ( scenario D; Figure 3 ) there is some evidence that , overall , the infectious period of the contact infected birds was somewhat longer than that of the artificially infected birds ( mean 2 . 5 ( day ) ( 95%CI: 1 . 9–3 . 3 ) versus mean 1 . 7 ( day ) ( 95%CI: 1 . 4–2 . 1 ) ) . Figure 3 furthermore shows that the variances of the latent and infectious period distributions of the contact infected birds could not be estimated with precision . An extended analysis including alternative informative prior distributions and an artificially extended dataset indicate that this is indeed the case , and that the experiments of Table 1 and 2 do not contain sufficient information to estimate the variance of the latent and infectious periods of the contact infected birds ( unless substantial prior information is added ) ( results not shown ) . Two derived epidemiological measures of interest are the reproduction number R and the generation interval Tg [11]–[13] . In our setting the reproduction number is given by the product of the infectious period and the transmission rate , while the generation interval is defined as the moment of infection of the contact bird , relative to the time at which the inoculated bird was returned to the cage following inoculation . Overall , the generation interval ranges from an estimated mean of 1 . 2 ( day ) in scenario A1 and scenario D to 1 . 8 ( day ) in scenario A2 , with limited variation around these estimates . This indicates that the generation interval is short , and lies in the range of 1–2 days . With regard to the reproduction number , we find substantial differences in the reproduction number between the low- and high-dose experiments . In fact , the estimated reproduction number is 2 . 0 ( 95%CI: 0 . 96–3 . 6 ) in the low-dose experiment , and 0 . 99 ( 95%CI: 0 . 38–2 . 1 ) in the high-dose experiment . This difference can be ascribed to differences in the mean infectious period in the low- versus high-dose experiments ( Table 4 ) . If the data of the low- and high-dose experiments are pooled and assumed to have the same infectious period distribution ( scenario B ) , the estimated reproduction number lies between the above extremes ( 1 . 6; 95%CI: 0 . 90–2 . 5 ) . Alternatively , if the data are pooled but the infectious period distributions are allowed to vary between the low- and high-dose experiments , the ( infection-type specific ) estimated reproduction numbers are 1 . 8 ( 95%CI: 1 . 1–3 . 0 ) and 1 . 2 ( 95%CI: 0 . 71–2 . 0 ) in the low- and high-dose experiments , respectively . To explore the implications of the parameter estimates for the dynamics of H5N1 avian influenza in large populations of poultry we have performed stochastic simulations of an SEIR model using the parameter estimates presented in Table 4 . The parameters determining the latent and infectious periods can directly be plugged into the model , but some care should be taken with the transmission parameter as it is not obvious how the parameter determining transmission between two individuals should be extrapolated to larger populations . The two common assumptions are that each individual makes a fixed number of contacts per unit of time regardless of population size ( the frequency dependent transmission assumption ) , or that each individual makes a fixed number of contacts with each of the other individuals in the population per unit of time ( the density dependent transmission assumption ) [24]–[25] . Under the frequency dependent transmission assumption the total number of contacts that an individual makes per unit of time does not depend on total population size , while under the density dependent transmission assumption the number of contacts that an individual makes per unit of time increases linearly with total population size [24] . It is plausible that for small to moderately sized populations the transmission rate increases monotonically with increasing population size and that this increase flattens off as population size becomes large ( birds cannot increase their activity levels indefinitely ) . Here we perform simulations of populations of 10 , 000 birds . In the simulations we first use the transmissibility estimates presented in Table 4 , which we subsequently multiply by a factor 2 . This implies that in our simulations birds in a population of 10 , 000 are either as active as birds that are kept in pairs , or twice as active as birds in pairs . Figure 4 shows two representative simulations of an epidemic in a population of 10 , 000 individuals using the parameter estimates of the low-dose experiment ( Table 4 ) . The top panel shows the time course of the epidemic in case of low transmissibility ( leading to a reproduction number of R = 2 . 0 ) , while the bottom panel shows the dynamics if the transmission rate parameter is increased twofold ( implying a reproduction number of R = 4 . 0 ) . The figure shows that the epidemic unfolds in about a month ( top panel ) to approximately two weeks ( bottom panel ) , depending on whether the transmission parameter is small or large . Furthermore , the figure shows that the peak prevalence is about 25% of total population size if transmissibility is low , and approaches 65% if transmissibility is high . Increasing the virus' transmissibility from twofold to , say , tenfold leads to minor changes in the infection dynamics as every susceptible individual is already very quickly ( within a time span of a week ) infected in the high transmissibility scenario ( results not shown ) . It is of note that in comparison with standard stochastic models that assume exponentially distributed latent and infectious periods the epidemics in Figure 4 are considerably more peaked , while their durations are substantially shorter ( results not shown ) [12] . Rapid detection of outbreaks of H5N1 highly pathogenic avian influenza virus in poultry is of paramount importance for efficient control within poultry flocks and to be able to minimize the opportunities of virus transmission between flocks [26]–[28] . If we assume that avian influenza can be detected with high specificity if mortality is at least 0 . 5% on two consecutive days [26] , [29] , then an outbreak will be detected in our simulations between days 11 and 12 after introduction if transmissibility is low , and between days 7 and 8 if transmissibility is high ( see the blue arrows in Figure 4 ) . In case of low transmissibility , this gives a window of opportunity of at most ten days to reduce the infectious output of the flock ( Figure 4A ) . If , however , transmissibility is high , circulation of the virus will only be detected near the moment of peak infectivity , and there is a window of opportunity of at most five days for control measures to be effective in reducing the infectious output of infected flocks once they are detected ( Figure 4B ) . Overall , our simulations indicate that control of H5N1 avian influenza in poultry flocks once an outbreak has been detected may be more difficult than hitherto thought [22] , [26]–[27] . To further investigate the potential for control by vaccination we have carried out simulations using estimates of the epidemiological parameters ( Table 4 ) and efficacy of vaccination ( Table S1 ) . Because of the reasons discussed above , it is highly unlikely that an outbreak can be controlled by vaccination once it has been detected . Adding to this is the fact that it may take 7–10 days for vaccination to become effective in interfering with transmission [14]–[16] . However , it may still be possible to prevent or curb outbreaks by preventive vaccination . Figure 5 gives an overview of the fraction of outbreaks that yield a major outbreak ( numbers near circles ) , the size of the major outbreaks ( circles ) , and the duration of the epidemics ( squares ) as a function of the fraction of birds that is vaccinated prior to introduction of the virus . If transmissibility is low ( cf . Figure 4A ) ( blue lines ) , the probability of a major outbreak as well as the size of the major outbreaks decrease with increasing vaccination coverage . The duration of major outbreaks , however , increases with increasing vaccination coverage [11]–[12] . Major outbreaks cannot occur for the parameters presented in Table 4 if coverage is at least 60% . If , on the other hand , pathogen transmissibility is high ( cf . Figure 4B ) ( red lines ) , then the probability of a major epidemic and final size of the epidemics increase in comparison with the low-transmissibility scenario , while the duration of the epidemics decreases [11] . Still , both the probability of a major outbreak as well as the size of the outbreak decrease with increasing vaccination coverage , and major outbreak cannot occur if vaccination coverage is at least 80% . Summarizing , our simulations indicate that it is possible to attain a state of herd immunity by incompletely vaccinating flocks of chickens even if birds are assumed to make twice as many contacts per unit of time as estimated in our transmission experiments .
In this study we have attempted to fill the remarkable void of quantitative information on key epidemiological parameters of H5N1 highly pathogenic avian influenza in chickens . Our results indicate that H5N1 virus induces a short period of latency and a short infectious period . In fact , our estimate of the mean of the latent period varies from 0 . 20 days ( 95%CI: 0 . 049–0 . 43 days ) in scenario A1 to 0 . 44 days ( 95%CI: 0 . 14–0 . 87 days ) in scenario A2 ( Table 4 ) . Likewise , the mean infectious period varies from 1 . 3 days ( 95%CI: 0 . 92–1 . 8 days ) in scenario A2 to 2 . 5 days ( 95%CI: 2 . 2–2 . 8 days ) in scenario A1 . Estimates of the variance of the infectious period are generally low , much lower than the corresponding means ( Table 4 ) . This implies that the distributions of the infectious periods are fairly narrow . Similar results were reported by Carrat and colleagues [30] who found that shedding of human influenza viruses increased sharply 0 . 5–1 day after infection , while the infectious period was centered narrowly around five days . Our estimates of the transmission parameter are remarkably similar across the different datasets and model scenarios . The estimate of the transmission parameter is lowest if the data of all experiments are combined ( median: 0 . 73 per day; 95%CI: 0 . 43–1 . 2 per day ) and highest if the analysis allows for differences between inoculated and contact birds ( median: 0 . 81 per day; 95%CI: 0 . 44–1 . 3 per day ) . In combination with the estimates of the mean infectious period these estimates yield estimates of the reproduction number varying from 0 . 99 ( 95%CI: 0 . 38–2 . 1 ) in the high-dose experiment ( scenario A2 ) to 2 . 0 ( 95%CI: 0 . 96–3 . 6 ) in the low-dose experiment ( scenario A1 ) . In view of the generally held belief that highly pathogenic avian influenza viruses spread easily and rapidly among chickens [14]–[16] , [22] , [26]–[27] our estimates of the reproduction number may seem low . In this respect a number of points are worth of discussion . First , we have assumed frequency dependent transmission , which assumes that each bird makes a fixed number of contacts per unit of time , regardless of the size of the population [24] . This is convenient since it allows one to directly extrapolate from small to large populations . The reason is that under this assumption the reproduction number does not depend on total population size . There is , moreover , evidence that a frequency dependent transmission model provides a better description of the pathogen dynamics than a density dependent model in farm animals that are generally held at a constant stocking density [31] . Still , some uncertainty remains as to how our estimates of the transmission parameter and infectious period should be combined into an estimate of the reproduction number . To address this potential problem we have in our simulations included a high transmissibility scenario ( Figures 4 and 5 ) that in essence assumes that birds in large populations are twice as active as birds in our transmission experiments with pairs of birds . Second , it is not straightforward to extrapolate our results that were obtained in an experimental setting to the situation in the field . This is especially so for estimates of the transmission parameter , which are the result not only of an autonomous process of viral replication and interaction of the pathogen with the immune system within a single host , but also of an interaction between different individuals . Ambient temperature , stocking density , feeding status of the birds , etcetera could all impact on this interaction and critically affect estimates of the transmission parameter . To counter this we have tried to match the conditions in our experiments to those in commercial laying chicken farms . Reassuringly , a recent analysis of transmission of H5N1 in the field [7] also indicates that the reproduction number of H5N1 virus among chickens is fairly low , ranging from 2 . 0 to 3 . 5 . This suggests that our estimates of the reproduction number obtained using pairs of birds are low but not unreasonable . A third point that deserves attention is the fact that housing systems of layer flocks vary from floor systems in which birds can mingle freely to caged systems in which no direct contact between ( groups of ) birds is possible . In principle , our study is aimed at quantifying transmission in a situation where there is direct contact between birds , corresponding to a floor system . However , the lone study that focused on within-flock transmission ( mostly backyard flocks ) did not find differences between different housing systems , suggesting that if there are differences in the transmission dynamics they cannot be large [7] . Nevertheless , more information on the infection dynamics in the field would be highly welcome to help bridging the gap between findings obtained in experimental studies and the situation in the field . While it is not straightforward to extrapolate from our experimental setting to the field situation , experimental transmission studies also have distinct advantages over field studies . In particular , while field studies often suffer from various sources of bias and confounding , this is not the case in an experimental setting . This allows one to directly ascribe differences between control and treatment groups directly to the treatment ( e . g . , vaccination ) since all other animal and environmental conditions are held constant . Moreover , an experiment has the added advantage over a field study that far fewer birds are needed and that the birds can be sampled more often and efficiently than in a field study . This has allowed us to obtain precise estimates of the key epidemiological parameters of H5N1 highly pathogenic avian influenza in unvaccinated chickens using no more than 50 birds . Our results show remarkable differences between experiments in which the inoculated bird received a low infection dose ( 0 . 2*105 EID50 ) and experiments in which the inoculated bird received a high dose ( 0 . 2*106 EID50 ) . Specifically , while 9 out of 11 birds were infected in case of a low infection dose ( Table 1 ) , only 5 out of 11 were infected in case of a high infection dose ( Table 2 ) . This is an interesting and counterintuitive result , which is likely to result from the fact that the infectious period in the experiments in which the inoculated bird received a high inoculation dose is significantly smaller than in the experiments in which the inoculation dose was low or in which the infectious period of the naturally infected birds was estimated separately ( low dose: mean 2 . 5 days ( 95%CI: 2 . 2–2 . 8 ) ; high dose: mean 1 . 3 days ( 95%CI: 0 . 92–1 . 8 ) ; contact birds only: mean 2 . 5 days ( 95%CI: 1 . 9–3 . 3 ) ) . Earlier experimental transmission studies with H7N7 highly pathogenic avian influenza virus ( A/Chicken/Netherlands/621557/03 ) in a variety of birds and H5N1 highly pathogenic avian influenza virus ( A/Chicken/China/1204/04 , also designated A/Chicken/GxLA/1204/04 ) in ducks used an infection dose of 0 . 2*106 EID50 since this yielded comparable infections in inoculated and naturally infected animals [14]–[16] . The finding that the infection dose is of importance in determining the duration of infection is of both theoretical and practical relevance as it suggests that the infection pressure in the population may not only determine the incidence of infection but also the course of infection . If it is typical that a low infection dose is associated with a long infectious period while a high infection dose generally leads to infections that are of short duration , then this would necessitate a rethinking of the critical determinants of H5N1 avian influenza transmission in populations of birds , and it could potentially have profound implications for optimal control and containment strategies . To investigate the implications of our parameter estimates for the dynamics of H5N1 avian influenza virus in large groups of chickens we have carried out stochastic simulations . Since it is not obvious how the transmission parameter as estimated between pairs of chickens can be extrapolated to large populations , we considered a low and high transmissibility scenario ( Figure 4 ) . The simulations indicate that , even if we assume that the transmission parameter is small , the epidemic usually unfolds in about a month , and that once the epidemic has taken off it only takes about two weeks to come to an end . If , as appears more likely , the transmission rate is larger in large population than in populations of two birds , then the epidemic takes off more quickly after a primary introduction , and also comes to an end more quickly . For control purposes this implies that it will be very difficult , if not impossible , to effectively control an outbreak once it has been detected . It may even prove difficult to reduce transmission opportunities from an infected population ( a farm , say ) to susceptible populations , as the number of dead birds may start to rise just before peak infectivity ( Figure 4 ) . This suggests that perhaps other indicators of infection , such as lethargy , reduced feed or water intake should be added to the mortality indicator to obtain a sensitive syndrome-reporting system [32] . While H5N1 virus spreads rapidly among unvaccinated chickens , no transmission was observed at all in the experiments with inactivated oil emulsion vaccines ( Tables S1 , S2 , S3 , S4 , S5 , S6 , and S7 ) . This was true not only for an H5N1 vaccine virus which had 100% homology to the challenge virus , but also for genetically distant heterologous viruses that contained inactivated H5N2 viruses . These findings indicate that it is possible , at least in principle , to reduce transmission by vaccination to the extent that no epidemics can occur . This suggestion is corroborated by our simulations which indicate that a vaccination coverage as low as 60%–80% may already be sufficient to obtain herd immunity ( Figure 5 ) . Of course , it should be borne in mind that in our experiments all birds received two vaccination doses , that the timing of challenge ( two weeks after the last vaccination bout ) was probably ideal , and that in the field there are various factors that may interfere with vaccination ( concurrent infections , immune depression by various causes ) [33] . Still , our results and those of others [34]–[36] provide a proof-of-principle that herd immunity can be obtained with currently available inactivated oil emulsion vaccines . The finding that H5N1 avian influenza virus has a lower transmissibility than hitherto believed [26] also implies that outbreaks may be easier to prevent than previously thought , since the reproduction number is already relatively close to the threshold value of 1 .
All experiments were carried out in PT Medion laboratories in Bandung , Indonesia , which have high containment facilities ( BSL3 ) . In all experiments , specific pathogen-free ( SPF ) layer chickens from the animal unit of Medion were used . The birds were hatched and housed in one group until 4 weeks of age . At that age , pairs of birds were housed in cages . Three rooms were available to house the various vaccinated and unvaccinated pairs of birds . Two rows with three levels of cages on top of each other were available in each room . The rows with cages were separated by a corridor of approximately 1 m width . The various rooms as well as the rows with the cages had separate ventilation systems . Each cage had a separate feeding and drinking system . The floor and walls of each cage were covered with plastic to prevent spread of manure or other materials between cages . When sampling the birds , animal caretakers used a new pair of gloves for each cage . Unvaccinated sentinel birds were placed at regular distances between the cages used in the experiments to ensure that no transmission had taken place between cages . All sentinels survived and remained seronegative during the course of the experiments . The challenge strain used in the experiments was A/Chicken/Legok/2003 H5N1 , a highly pathogenic H5N1 strain isolated in Indonesia in 2003 which is genetically very close to strains that circulate in Indonesia in 2008 . The strain has been used in experiments carried out at Medion and is able to induce infection , typical signs of disease , and high mortality rates in chickens . Inactivated oil emulsion vaccines were available from three different manufactures: PT Medion ( Bandung , Indonesia ) , PT Vaksindo ( Bogor , Indonesia ) and Intervet ( Mexico ) . The vaccines contained either an H5N1 or H5N2 virus strain . The H5N1 vaccines contained A/Chicken/Legok/2003 H5N1 , i . e . the vaccine and challenge strains were identical . The H5N2 vaccines contained either A/Turkey/England/N28/73 H5N2 or A/Chicken/Mexico/232/94/CPA H5N2 . The protein homologies of the antigenic part of the hemagglutinin ( HA1 ) of the challenge strain to the H5N2 A/Turkey/England/N28/73 and H5N2 A/Chicken/Mexico/232/94/CPA vaccine strains are 92% and 86% , respectively . All vaccines were re-vialed in coded bottles , and the identity of the vaccines was not known to the staff involved in the experiment . In this manner the experiments were double blinded . Because the size of a natural infection dose is unknown the inoculum consisted of diluted allantoic fluid containing either 105 EID50 per ml ( low inoculation dose ) or 106 EID50 per ml ( high inoculation dose ) . The birds were inoculated both intranasally ( 0 . 1 ml ) and intratracheally ( 0 . 1 ml ) . Virus titres were confirmed before and after inoculation by titration on embryonated SPF eggs . Each experiment consisted of a set of 11 trials . In each of the trials an inoculated bird was placed in a cage with an uninfected contact bird , and the transmission chain was monitored daily by virus isolation performed on swabs taken from the trachea and cloaca . In all , a total of eight experiments were carried out . Unvaccinated birds were used in two experiments . In the first of these the inoculated birds received a low infection dose , while in the second the inoculated birds received a high infection dose . The remaining six experiments with vaccinated birds differed with respect to the vaccine used , the manufacturer , and the inoculation dose . Tables 1 and 2 show the data of experiments with unvaccinated birds , and Tables S1 , S2 , S3 , S4 , S5 , S6 , and S7 give an overview of the experiments with vaccinated birds . At 4 weeks of age all birds of the vaccination experiments received their first vaccination dose . A second vaccination was carried out at 7 weeks of age . At 10 weeks of age ( day 0 ) one bird was chosen at random per cage , taken from the cage , and infected intratracheally and intranasally . To avoid direct infection of the contact bird by the inoculum the artificially infected birds were placed back in their cages only after a delay of 8 hours . Tracheal and cloacal swabs were taken daily for 10 days after challenge from all birds . Swabs were incubated for 1 h in one ml of PBS medium containing antibiotics . The medium was subsequently stored at −70°C until testing . Three embryonated SPF chicken eggs were injected with 0 . 2 ml of the swab medium per egg . After culture for 4 days or when embryos died , the allantoic fluid was harvested and a hemagglutination ( HA ) assay was performed following standard procedures ( www . oie . int ) . When at least one of the eggs was positive in the hemagglutination assay the swab was considered to be positive . The serological status of the birds was determined just before vaccination , at the start of the experiments just before inoculation ( day 0 ) and , for birds that survived , at the end of the experiments ( day 14 ) . Serum blood samples were taken from all birds by puncturing the wing vein . Blood samples were centrifuged and serum was stored at −20°C until tested . The sera were tested in the hemagglutination inhibition ( HI ) test according procedures described in the Manual of Diagnostic Tests and Vaccines for Terrestrial Animals of the OIE ( www . oie . int ) using 4 HA units ( HAU ) of A/Chicken/Legok/03 H5N1 as antigen . Titres were expressed as 2log of the serum dilution that caused complete inhibition of agglutination , as specified by OIE guidelines . Clinical signs of disease were recorded daily for a period of up to 10 days after challenge . As a first step we estimated the reproduction number R by final size methods [14]–[17] . Since each trial contains only one inoculated bird and one susceptible contact bird , the likelihood function takes the following simple form: ( 1 ) In this equation N and n are the number of trials per experiment and the number of infected contact birds , while represents the Laplace transform of the infectious period probability distribution when the mean infectious period is scaled to 1 . Hence , in case of an exponentially distributed infectious period , and in case of a fixed infectious period . Table 3 provides estimates of the reproduction number with corresponding 95% confidence intervals , as well as p-values of the null-hypothesis that the reproduction number is greater than or equal to the threshold value of 1 [23] . In a second step , we estimated all parameters of interest by Bayesian methods [18]–[21] . In the following we denote by the transmission rate parameter , by and the parameters determining the latent period probability distribution , and by and the parameters of the infectious period probability distribution . We assume that the latent and infectious periods are gamma distributed , and that and , and and represent the means and variances of these distributions . The corresponding probability densities are denoted by and . Further , , , and are N-dimensional vectors which contain the time points of the S→E , E→I , and I→R transitions for inoculated ( ) and contact ( ) birds in the N trials . Hence , we have by definition , while all other transition times are unknown . The unknown transitions are added in the analyses by Bayesian imputation . We adopt the convention that denotes the exact time at which the contact bird in experiment j is infected , that denotes the exact time that the inoculated bird in experiment j became infectious , etcetera . With these notational conventions , the contribution of trial j to the likelihood is given by ( 2 ) In the above equation and denote the infection hazard in trial j at time t and the probability that the contact bird in trial j remains uninfected up to time t , respectively . If we let […] denote the indicator function , the infection hazard is given by ( 3 ) where the parameter represents the delay between the moment of inoculation and the placing back of the inoculated birds in their cages , and the function marks the beginning of the at-risk period for the contact bird . In all trials and experiments , the delay is 8 hours , i . e . ( day ) . The probability that the contact bird in trial j remains uninfected up to time t can be expressed in terms of the infection hazard as follows ( 4 ) Using Equations ( 2 ) – ( 4 ) the likelihood function is given by the product of the contributions of the individual trials: ( 5 ) where P represents the set of trials . Equations ( 2 ) – ( 5 ) form the basis of the analyses in Figures 1 , S1 , S2 , and S3 . The likelihood contribution in Equation ( 2 ) assumes that the latent and infectious periods of inoculated and infectious birds are identically distributed . To investigate the validity of these assumptions we also considered a model which allows for differences between the inoculated and contact birds . In this case , the likelihood contribution becomes ( 6 ) where and are the probability density functions of the latent and infectious periods of the inoculated birds ( ) and contact birds ( ) . The results of the analyses based on Equation ( 6 ) are given in Figure 3 and Figure S5 . In a similar manner , the likelihood contribution in Equation ( 2 ) is adapted to allow for differences in the infectious period in the low- versus high-dose experiments . The results of these analyses are summarized in Figure 2 and Figure S4 . Notice that , since the transmission rate in Equations ( 1 ) , ( 2 ) , and ( 5 ) is divided by the total size of the population ( i . e . 2 ) , the above model assumes frequency dependent transmission ( as opposed to density dependent transmission ) [24] . For the present experimental setup with one inoculated bird and one contact bird , the value of the transmission parameter of the density dependent model is simply given by the transmission rate parameter of the frequency dependent transmission model divided by 2 ( the size of the population ) . In case of a frequency dependent transmission model the ( basic ) reproduction number is given by the product of the transmission rate parameter and the mean infectious period: . In case of a density dependent transmission model the reproduction number is a function of population size , and it is given by , where denotes the transmission parameter of the frequency dependent model with two birds [25] . As in earlier papers [18]–[21] the epidemiological parameters of interest ( , , , , and ) were estimated by Bayesian methods of inference using Markov Chain Monte Carlo . Throughout , all prior distributions of the parameters were uniformly distributed on the interval ( 0 . 001–5 ) . As an alternative we also considered vague gamma prior distributions , and obtained comparable results ( results not shown ) . In our simulations the epidemiological parameters and unobserved transitions were updated by a random-walk Metropolis algorithm . We used Normal proposal distributions with the current value as mean , and a standard deviation of 0 . 025 , 0 . 05 , or 0 . 1 . The transmission parameters and unobserved transitions were updated in blocks , in the order , , , , , and . Notice that updating of the individual transition vectors needs to take into account the infection data of Tables 1 and 2 and the information contained in the other transition vectors , as these specify the admissible intervals of the various transitions . The above updating scheme yielded chains that converged quickly and showed satisfactory mixing . In all analyses we took a burn-in of 25 , 000 cycles and a simulation length of 200 , 000 cycles . Thinning was applied by taking only each 100th cycle as a sample from the posterior distribution . We performed four replicate simulations to check the precision of the parameter estimates obtained by the above procedures . These simulations yielded parameter estimates and 95% credible intervals that were close to those given in Table 4 . To choose between models of different complexity we made use of Bayes factors ( BF ) [37] . To this end the marginal likelihoods of competing models were estimated by importance sampling using the harmonic means of the posterior likelihood values [37] . The BF converged slowly , possibly because of the high dimensionality of the model ( 86 unobserved transition events plus 5–9 epidemiological parameters ) , the mutual dependencies of the unobserved transitions , and the fact that the likelihood is strongly affected by the parameters defining the variances of the latent and infectious periods if those are small . However , this did not appear to be a major practical problem as differences between competing models were usually large . When reported in the text we calculated the BF of the pair of simulations that had the smallest difference in marginal likelihoods . A suite of Bayesian analyses were performed for the experiments with unvaccinated birds . First , we analyzed the low- and high-dose experiments of Tables 1 and 2 separately ( scenarios A1 and A2 ) . Second , we pooled the data of the low- and high-dose experiments ( scenario B ) . We then considered an integrated analysis of the two experiments that allowed for differences in the infectious periods in the low- versus high-dose experiments ( scenario C ) . Finally , we considered a scenario which allowed for differences in the epidemiological characteristics of the inoculated and contact birds ( scenario D ) . To explore the implications of the parameters estimated by the above procedures for the pathogen dynamics in large groups of birds , we performed simulations of the stochastic SEIR model using the Sellke construction [16] . In the simulations we assumed gamma distributed latent and infectious periods , and used the medians of the parameter estimates of Table 4 as input values . The programs for the MCMC analyses and simulated epidemics were written in Mathematica 6 . 0 ( www . wolfram . com ) . | Outbreaks of highly pathogenic H5N1 avian influenza in poultry first occurred in China in 1996 . Since that time , the virus has become endemic in Asia , and has been the cause of outbreaks in Africa and Europe . Although many aspects of H5N1 virus biology have been studied in detail , surprisingly little is known about the key epidemiological parameters of the virus in its avian hosts ( the length of time from infection until a bird becomes infectious , the duration of infectiousness , how many birds each infectious bird will infect ) . In this paper we show , using experimental transmission studies with unvaccinated and vaccinated chickens , that H5N1 avian influenza induces a short duration of infectiousness ( ∼2 days ) and a very short period of time from infection until infectiousness ( ∼0 . 25 day ) in unvaccinated chickens . Furthermore , while transmission was efficient among unvaccinated birds , no bird-to-bird transmission was observed in vaccinated chickens . Our results indicate that it may be difficult to curb outbreaks by vaccination after an introduction in a flock has been detected . On the other hand , preventive vaccination could be effective in preventing virus introductions and limiting the size of outbreaks . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"virology/vaccines",
"virology/animal",
"models",
"of",
"infection",
"virology/emerging",
"viral",
"diseases",
"infectious",
"diseases/viral",
"infections",
"public",
"health",
"and",
"epidemiology/infectious",
"diseases",
"infectious",
"diseases/epidemiology",
"and",
"control",
"of",
"infectious",
"diseases"
] | 2009 | Estimation of Transmission Parameters of H5N1 Avian Influenza Virus in Chickens |
Adenosine-5’-triphosphate ( ATP ) is generally regarded as a substrate for energy currency and protein modification . Recent findings uncovered the allosteric function of ATP in cellular signal transduction but little is understood about this critical behavior of ATP . Through extensive analysis of ATP in solution and proteins , we found that the free ATP can exist in the compact and extended conformations in solution , and the two different conformational characteristics may be responsible for ATP to exert distinct biological functions: ATP molecules adopt both compact and extended conformations in the allosteric binding sites but conserve extended conformations in the substrate binding sites . Nudged elastic band simulations unveiled the distinct dynamic processes of ATP binding to the corresponding allosteric and substrate binding sites of uridine monophosphate kinase , and suggested that in solution ATP preferentially binds to the substrate binding sites of proteins . When the ATP molecules occupy the allosteric binding sites , the allosteric trigger from ATP to fuel allosteric communication between allosteric and functional sites is stemmed mainly from the triphosphate part of ATP , with a small number from the adenine part of ATP . Taken together , our results provide overall understanding of ATP allosteric functions responsible for regulation in biological systems .
Adenosine-5’-triphosphate ( ATP ) , the naturally occurring nucleotide used in cells as a coenzyme , serves as the “energy currency of the cell” in intracellular energy metabolism [1] . In eukaryotes , the majority of ATP is manufactured in the chloroplasts ( for plants ) or in the mitochondria ( for both plants and animals ) as a result of cellular processes including fermentation , respiration and photosynthesis [2] , [3] . The well-known functional role of ATP in cells is the hydrolysis of its high-energy phosphate bond to provide energy for biological processes . In addition , intracellular protein modification via ATP as a substrate , known as phosphorylation , orchestrates a fine-tuned network for the cell to switch on diverse signal transduction processes , such as proliferation , gene transcription , metabolism , kinase cascade activation , and membrane transport [4] . Although ATP is one of the most productive endogenous ligands in cells , striking evidence from clinical practice demonstrates that intravenous or oral administration of ATP may have pivotal effects on the treatment of pathological processes , including the regulation of diverse processes ( coronary blood flow [5]–[7] , neurotransmission [8] , muscle contraction [9] , and inflammation [10] , [11] ) , the protection of the myocardium and the maintenance of the homeostasis of the cardiovascular system [12] , [13] . The usage of ATP indicates that ATP could play even more roles in physiological/pathological regulation . Recently , the determination of allosteric ATP-binding sites provided new evidences that ATP can act as an allosteric modulator regulating protein function [14] , [15] . For instance , in the mismatch repair initiation protein MutS , ATP binding to the nucleotide-binding site triggers long-range interactions between the ATPase and DNA-binding domains [16] . In the chaperonin GroEL , binding of ATP to a unique pocket in the equatorial domains of GroEL allosterically initiates a series of conformational changes triggering the association of the cochaperonin GroES [17] . As a extracellular messager , ATP was confirmed to allosterically involve in activation mechanism of transmembrane protein P2X receptors [18] , [19] . The role of ATP in cell signaling is known to serve not only as a substrate but also as an allosteric modulator . However , little is understood about this critical behavior of ATP in allosteric binding sites . The dynamic process of ATP access to the allosteric and substrate binding sites and the allosteric mechanism for ATP molecules are also unclear . Here , we addressed these questions by combining bioinformatic analyses , molecular dynamics ( MD ) and nudged elastic band ( NEB ) simulations . The results suggest that allosteric ATP-binding sites have evolved along different evolutionary pathways compared to the extremely conserved substrate ATP-binding sites . ATP molecules adopt both compact and extended conformations in the allosteric binding sites but conserves extended conformations in the substrate binding sites , and both compact and extended conformations of ATP are dominant in solution . We follow the preference of ATP access to the allosteric and substrate binding sites at atomic resolution and explore the mechanism of ATP-triggered allostery . Understanding the regulatory mechanism of ATP in biological systems as a substrate and an allosteric modulator in cell signaling is of fundamental importance .
To elucidate differences in the evolutionary characteristics of allosteric and substrate ATP-binding sites , we systematically explored the sequence conservation of ATP-binding sites in the ATP allosteric and substrate datasets . Sequence conservation scores for residues in the allosteric and substrate ATP-binding sites as well as in the surface of proteins were assessed by means of multiple sequence alignments of homologous sequences ( see Materials and Methods for details ) . As shown in Figure 2A , residues in the allosteric ATP-binding sites , with the average conservation score of 0 . 63 , were less conserved compared to residues in the substrate ATP-binding sites , with the average conservation score of 0 . 83 ( P = 1 . 2×10−3 ) . Yet , the residues in the allosteric and substrate ATP-binding sites are markedly more conserved compared to residues in the rest of the surface; the average conservation score of residues in the latter is 0 . 31 ( P = 9 . 9×10−7 ) . Furthermore , the average conservation score distributions of residues in the allosteric and substrate ATP-binding sites as well as in the surface for each protein sequence were analyzed . As shown in Figure 2B , remarkably , the average conservation score of residues in the allosteric ATP-binding sites show a wide distribution in a range from 0 . 4 to 0 . 8 . These results are consistent with a previous study , as done by Yang et al . [21] , that allosteric sites are evolutionarily variable against catalytic sites . Taken together , these data indicate that allosteric ATP-binding sites have evolved along different evolutionary pathways compared to the conserved substrate ATP-binding sites . Given the significant structural differences between the allosteric and substrate ATP-binding sites that may accommodate diverse ATP conformations , we first simulated the ATP in solution to explore its conformational preferences in the unbound state . The landscapes with the reaction coordinates of distance ( defined by the distance from the Pγ atom to the centroid of adenine moiety ) and angle ( defined by the centroids of triphosphate , ribose and adenine moieties ) were calculated on ATP conformations ( 200 , 000 snapshots ) from the MD trajectory . As shown in Figure 3 , two major conformation regions exist in solution , one region with distance values from 6 . 5 Å to 8 . 5 Å and angles between 100° and 130° , the other with distance values from 9 . 3 Å to 12 . 8 Å and angles ranging between 90° and 140° . An analysis of the average structures of ATP corresponding to the two conformation regions indicates that the former region represents the compact conformation of ATP , whereas the latter represents the extended conformation . These data suggest the pre-existence of the compact and extended conformations of ATP in solution . Next , we investigated the conformational properties of ATP molecules in the allosteric and substrate binding sites by comparing the mean structural differences ( RMSD ) between each ATP molecule and the remaining ATP molecules in the ATP allosteric and substrate datasets . As shown in Figure S1 , the average RMSD value for the 13 allosteric ATP molecules is 1 . 83 Å . By contrast , the average RMSD value for the 24 substrate ATP molecules is 1 . 37 Å ( P = 7 . 1×10−5 ) , revealing smaller conformational variations of ATP in the substrate binding sites . This distinction could be because of the differences in the evolutionary conservation of the allosteric and substrate ATP-binding sites . To further elucidate the structural propensities of ATP molecules in the allosteric and substrate binding sites , we mapped the bound ATP molecules in the allosteric and substrate datasets to the MD-generated ATP conformations . As shown in Figure 3 , ATP molecules in the allosteric binding sites show a wide distribution with the distance values ranging from 5 . 7 Å to 12 . 2 Å . Inspection of these ATP molecules reveals that seven cases adopt compact conformations ( PDB: 1FA9 , 1I2D , 1PFK , 3HWS , 2XCW , 3R1R , and 4DW1 ) , while the remaining six cases adopt the extended conformations ( PDB: 1W7A , 4AT1 , 4GFH , 2HCB , 2JJX , and 1KP8 ) . These data suggest that allosteric ATP molecules can adopt both the compact and extended conformations in the allosteric binding sites of proteins . However , as for the substrate ATP molecules , the analysis reveals that they adopt the extended conformations in the substrate binding sites of proteins . The functions of ATP in vivo are triggered by the protein transitioning from the free unbound state to the bound state . The uridine monophosphate ( UMP ) kinase of Bacillus anthracis was chosen as a good model to investigate the dynamic process of ATP binding . This kinase not only binds ATP as a substrate in its active site but also has allosteric ATP-binding sites near the center of the hexameric structure [22] . This property of the kinase permits exploration of which site ATP preferentially binds to in solution . The UMP kinase catalyzes the transfer of the γ-phosphoryl group from ATP to UMP in the active site , which is essential for cellular growth in bacteria; the catalytic reaction is regulated by ATP in the allosteric site . We first examined the sequence conservation of residues in the allosteric and substrate ATP-binding sites of UMP kinase family . As shown in Figure S2 and Table S4 , allosteric ATP-binding sites ( average conservation score = 0 . 44 ) are significantly less conserved compared to substrate ATP-binding sites ( average conservation score = 0 . 77 ) ( P = 6 . 1×10−4 ) , but more conserved than the rest of surface residues ( average conservation score = 0 . 25 ) ( P = 2 . 2×10−3 ) . The structural characteristics of ATP binding at the allosteric and active sites of UMP kinase were then explored using MD simulations . Although the above analyses elucidated the binding sites and the related binding process of allosteric ATP molecules , the mechanism of allostery for these identical ATP molecules in different allosteric sites of proteins remains unclear . According to the structural view of allostery [30] , [31] , allostery works by the propagation of strain energy created at the allosteric site by ligand binding , post-translation modifications , or mutations of the functional site . Guided by this principle , we investigated which part of ATP ( the adenine , ribose , or triphosphate ) triggers allostery . Therefore , we first analyzed the detailed interactions between each part of ATP and the corresponding protein . As shown in Table 2 , the analysis revealed that the contributions of the interactions stem mainly from the adenine and triphosphate parts of ATP . Subsequently , a comprehensive structural analysis of the identical protein between the allosteric ATP-bound and unbound structures was performed [32] . Ten out of thirteen allosteric ATP unbound structures were retrieved from the PDB . A protein 3D structural alignment of the allosteric ATP-bound and unbound structures ( Figure 7 ) displayed that the local structure of proteins proximal to the adenine portion of ATP shows significant conformational change with limited conformational change proximal to the ribose and triphosphate parts of ATP in two cases ( Figure 7A–B ) . This finding is indicative of an allosteric trigger by the adenine part of ATP . By contrast , the local structure of the proteins shows significant conformational change in the proximity of the triphosphate portion of ATP with limited conformational change in the proximity of the adenine and ribose portions of ATP in the remaining eight cases ( Figure 7C–J ) . This finding is indicative of a trigger by the triphosphate part of ATP . For example , in the P2X4 ion channel ( Figure S4A ) , residues Asn296 , Arg298 , and Lys316 from both the unbound ( PDB: 4DW0 ) and ATP-bound ( PDB: 4DW1 ) structures engage in hydrogen bonding/electrostatic interactions with the triphosphate in ATP , revealing marginal conformational changes proximal to the triphosphate in ATP . However , the conformation of Arg143 in the two structures shows prominent variation , giving rise to the markedly local conformational change between the head and dorsal-fin domains proximal to the adenine portion of ATP . This local conformational change is transmitted from the lower body domain to transmembrane domains 1 ( TM1 ) and TM2 , and leads to the opening of the ion channel pore because of the outward flexing of TM1 and TM2 [19] . In addition , biochemical experiments demonstrated that GTP and UTP are unable to activate P2X4 receptors [33] , further supporting the notion that the adenine portion of ATP is the trigger of allostery . In the case of human liver glycogen phosphorylase ( Figure S4B ) , Tyr75 from both the unbound ( PDB: 1FC0 , inactive state ) and ATP-bound ( PDB: 1FA9 , active state ) docked structures forms hydrogen bonding interactions with the adenine in ATP , revealing marginal conformational changes in the proximity of the adenine in ATP . However , residues Glu195 , Phe196 , and Arg309 proximal to the triphosphate in ATP show pronounced conformational changes , which propagate to the active site to switch the gate ( residues 280–289 ) from the “closed” position to the “open” position , allowing glycogen and pyridoxal phosphate access to the catalytic site [34] .
ATP is generally utilized by protein kinases or enzymes as the source of phosphate groups to phosphorylate their substrates . Thus , ATP commonly acts as a substrate by virtue of its hydrolysis . Conversely , compared with the overwhelming number of protein-substrate ATP complexes , the few proteins capitalize on ATP as an allosteric modulator to regulate their functional activity . This may be ascribed to the dearth of ATP allosteric sites in protein surfaces to accommodate ATP binding . Simultaneously , the question arises as to why the co-evolution of ATP and optimized allosteric sites did not occur on a more extensive scale . Evolution seemingly selected for other allosteric regulatory mechanisms such as allosteric post-translational modifications [35] . The commonality of the allosteric phosphorylation mechanism may indicate the advantages of a covalent linkage which also allows a higher residence time . According to ASD v2 . 0 [15] , allosteric proteins in which ATP has been biologically confirmed to serve as an allosteric modulator are 78 ( Table S6 ) regardless of species . Taking into account the species , the available co-crystal structures of proteins-allosteric ATP complexes , and the structural diversity of co-crystal structures , only thirteen distinct co-crystal structures of protein-allosteric ATP or its derivative complexes are available . In humans , all of these allosteric proteins are located inside the cells with the exception of the P2X4 receptor which is expressed on the cell surface . These data suggest that ATP can also act as an allosteric modulator to regulate protein function inside cells . As to extracellular regulation , all P2X subtypes ( P2X1-7 ) are membrane ion channels that allosterically respond to the binding of extracellular ATP ( only the crystal structure of the P2X4 receptor-allosteric ATP complex has been described ) [19] . The physiological functions of the seven P2X receptor subtypes include modulation of synaptic transmission , contraction of smooth muscle , secretion of chemical transmitters and the regulation of the immune response [18] . ATP drugs , mostly as adenosine disodium triphosphate tablets , are frequently used in the treatment of progressive muscular atrophy , myocardial diseases , hepatitis ( as an adjuvant therapy ) and heart failure . Based on the present study , we suggest that extracellular ATP molecules , such as ATP drugs , serve as allosteric modulators in extracellular signaling by activating the P2X receptors , a suggestion supported by the functional roles of ATP drugs corresponding to the pharmacology of P2X receptors [33] . We systematically analyzed the amino acid composition of allosteric and substrate ATP-binding sites . In the substrate ATP-binding sites , the amino acid composition is conserved as a consequence of high sequence-similarities of enzymes . However , the composition of allosteric ATP-binding sites varies . The sequence variability in the allosteric ATP-binding sites is likely to be linked to the fine-tuning of allosteric regulation commensurate with function [36]–[40] . The conformational diversity of allosteric and substrate ATP molecules may reflect the properties of binding sites . Overall , substrate ATP molecules show smaller conformational changes , indicative of the conserved substrate ATP-binding sites of enzymes . In contrast , allosteric ATP molecules show larger conformational changes , indicative of the structural variations of allosteric ATP-binding sites . These differences in structural propensities of ATP may be attributed to the different functional roles of ATP in the allosteric and substrate sites . The structural analysis reveals that all ATP molecules in the substrate binding sites conserve extended conformations whereas ATP molecules in the allosteric binding sites may adopt compact conformations ( Figure 3 ) . Among the thirteen allosteric cases , more than half of ATP molecules regulate protein functions with compact conformations encapsulated in deep hydrophobic pockets ( ribonucleotide reductase , P2X4 ion channel , cytosolic 5’-nucleotidase II , glycogen phosphorylase , phosphofructokinase 1 , ATP sulfurylase and ClpX ) . However , the remaining six ATP molecules in GroEL , MutS , DnaA , DNA Topoisomerase II , UMP kinase and Aspartate carbamoyltransferase function as allosteric modulators with extended conformations , which is similar to ATP conformations in the substrate binding sites . Remarkably , the allosteric binding sites of GroEL , MutS , DnaA and DNA Topoisomerase II have dual effects on allosteric regulation and ATPase activities , ATP molecules in the sites allosterically induce protein functions and then are hydrolyzed to ADP when the allosteric regulations are complete [16] , [17] , [41] , [42] . To carry out both allosteric and substrate functions , ATP molecules thus adopt extended conformations in the sites . In UMP kinase , ATP molecules allosterically bind at the protein interfaces between neighboring subunits . The subunits interfaces are narrow , leading to the bound allosteric ATP molecules in extended conformations . Despite the overall extended conformations , the triphosphate moiety of ATP is in the curved conformation in the allosteric binding site , resulting in the formation of internal hydrogen bonds between the γ-phosphate group of triphosphate and the hydroxyl group of ribose . In aspartate carbamoyltransferase , the allosteric ATP binding site is rather flat in the surface of enzyme , with the adenine and ribose moieties of ATP occupying the cavity and the triphosphate moiety of ATP drifting towards the solvent ( Figure S5 ) . The feature of allosteric site and the interaction mode between the enzyme and ATP render the allosteric ATP molecule in extended conformation . Overall , the features of allosteric ATP-binding sites in allosteric proteins and the functional roles of allosteric ATP molecules in allosteric proteins determine the conformations of ATP , compact or extended conformations , in allosteric sites . An MD simulation of ATP in solution indicates that the conformational ensemble of ATP in its unbound state is primarily characterized by two states: the compact conformation and the extended conformation . These two states agree with the structural analysis . The UMP kinase was selected to explore the dynamic process of the access of ATP to the corresponding allosteric and substrate binding sites because this enzyme possesses both the allosteric and substrate ATP-binding sites [22] . NEB simulations revealed markedly different pathways . Notably , the conformational rollover of ATP was observed in the process of ATP binding to the allosteric site but not to the substrate site . This may reflect the differences in the volume and shape of the allosteric and substrate binding sites of the UMP kinase . As shown in Figure 4B , the UMP kinase substrate ATP-binding sites point toward the solvent , and the cavity is spacious which may facilitate the access of ATP to the UMP kinase substrate binding sites . Conversely , the allosteric ATP-binding sites are located at the subunit interface . As shown in Figure 4B , the γ-phosphate moiety of ATP points toward the solvent , whereas the nucleoside moiety points toward the cavity interior . As a consequence , the cavity of the allosteric binding sites is narrow , hindering the access of ATP to the allosteric binding sites . The calculated barrier of ATP binding to its substrate site is lower than that to the allosteric site ( 22 . 8 kcal/mol versus 30 . 2 kcal/mol , respectively ) , and the total interaction energy between the UMP kinase and the substrate ATP molecule is also lower than that between the UMP kinase and the allosteric ATP molecule ( −59 . 3 kcal/mol versus −50 . 8 kcal/mol , respectively ) . Collectively , these data suggest that , in solution , ATP preferentially binds to substrate sites of the UMP kinase . Once an allosteric ATP molecule occupies the allosteric binding site of a specific protein , the source of the allosteric trigger from ATP to fuel allosteric communication between allosteric and active sites may differ . Our structural analysis revealed that a majority of proteins ( 80% ) select the triphosphate in ATP as an allosteric trigger , whereas a smaller number of proteins ( 20% ) opt for the adenine in ATP as an allosteric trigger . The significant difference between the contributions of the allosteric trigger from ATP indicates that the conformational flexibility of the triphosphate in ATP endows ATP with various regulatory mechanisms and may play a crucial role in the initiation of allostery .
We built two types of ATP datasets: allosteric and substrate . First , we constructed an annotated ATP-allosteric dataset collected from a hand-curated dataset . ASD v2 . 0 has manually curated allosteric proteins and allosteric modulators with at least three cases with experimental evidence , crystal structure of the complex or biochemical data [15] . The non-redundant 3D-structures of allosteric proteins in complex with allosteric ATP and its derivatives were considered . We retrieved 13 allosteric proteins deposited in ASD ( Table 1 ) . In the 3D-structures of glycogen phosphorylase ( PDB: 1FA9 ) [34] , ATP sulfurylase ( PDB: 1I2D ) [43] , phosphofructokinase 1 ( PDB: 1PFK ) [44] , and ClpX ( PDB: 3HWS ) [45] , the solved allosteric effectors in their ATP binding sites are the ATP derivatives , which are most likely because of the hydrolysis of ATP during crystallization . Therefore , we manually docked the ATP molecule into the allosteric ATP binding site in the aforementioned proteins . Second , we constructed an ATP substrate dataset . Twenty-four co-crystal structures of proteins in complex with substrate ATP molecules ( Table S3 ) , which exhibit structural diversity , were extracted from the PDB database [21] . The binding site residues for the allosteric and substrate ATP molecules were identified from those within 6 Å of ATP using a fpocket-based pocket detection algorithm [46] . Surface residues were identified as those with high solvent accessible surface areas ( SASAs ) ( >50% of SASA values for corresponding residues in the natural state ) , which were calculated by POPS [47] . To recapitulate the family profile of allosteric proteins , a phylogenetic tree was built from the ATP allosteric database . A multiple sequence alignment ( MSA ) was performed by a ‘progressive algorithm’ using ClustalX [48] and BLOSUM 30 matrix [49] . MEGA 5 . 0 [50] was used to build the phylogenetic tree from the MSA results . The sequence alignments and the calculation of amino acid conservation scores for substrate , allosteric , and surface residues were carried out via the ConSurf server [51] using default parameters . ConSurf retrieved homologous sequences to calculate amino acid conservation scores from the UniProtKB/SwissProt database [52] . The high sequence identity ( >95% ) and short sequence length ( <60% ) to the query sequence were eradicated , with the resulting homologous sequences to calculate amino acid conservation scores . The percentile normalization method to normalize the conservation scores , as done by Yang et al . [21] , was performed to compare the conserved degree of substrate , allosteric , and surface sites . Molecular docking of ATP to the active site of 1FA9 , 1I2D , 1PFK , and 3HWS was performed using the AutoDock 4 . 2 program [53] . Polar hydrogen atoms were added to the proteins . Kollman united partial atomic charges were then assigned and the AutoDock atom types were defined for the proteins using AutoDock Tools ( ADT ) . ATP geometry was minimized using the AM1 Hamiltonian as implemented in the program Gaussian09 [54] . Gasteiger charges were then added to ATP , with the default root , rotatable bonds , and torsion setting for ATP using the TORSDOF module in ADT . The grid center was defined at the centroid of the ATP derivatives in the aforementioned proteins , and the number of grid points in the x , y , and z directions were set to 60 , 60 , and 60 with a spacing value of 0 . 375 Å using AutoGrid . The distance-dependent function of the dielectric constant was used to calculate the energetic maps . The Lamarckian genetic algorithm was employed for the ATP conformational search with identical docking parameters used previously [55] , [56] . Fifty independent docking runs were conducted , and the binding energy was used to rank the docked ATP in order of fitness . In the simulation of ATP in aqueous solution , the structure of ATP in complex with Mg2+ , [ATP:Mg]2− , was extracted from the UMP kinase of Bacillus anthracis ( PDB: 2JJX ) [22] as previously suggested by Li et al . [57] in the simulation of ATP in water solvent . The polyphosphate parameters developed by Carlson et al . were adopted for ATP [58] . ATP was explicitly solvated by TIP3P [59] water molecules in a truncated octahedral box . The distance to the edge of the solvent box from the ATP atoms was set to be 15 Å . Counterions were added to maintain electroneutrality in the system . The final system contains ∼7 . 5×103 atoms . To perform unbiased simulations of the allosteric and substrate ATP-bound UMP kinase , the bound UMP-ATP complex was modeled on the basis of the crystal structure of the Bacillus anthracis UMP kinase ( PDB: 2JJX ) [22] . In the crystal structure of the Bacillus anthracis UMP kinase , the ATP and UMP in the active sites are not clearly visible in the electron density map . Therefore , ATP and UMP were manually docked into the active sites of the Bacillus anthracis UMP kinase after superposition with the Pyrococcus furiosus UMP kinase structure that was solved in complex with UMP and AMP-PCP ( PDB: 2BMU ) [28] . The unbound state of the Bacillus anthracis UMP kinase was obtained by removing both allosteric and substrate ATP molecules from the crystal structure . Prior to hydrogen atom placement , the program PROPKA [60] was used to perform pKa calculations to aid the assignment of side chain protonation states of all His residues . The AMBER ff03 force field [61] was assigned for proteins . ATP parameters used with the AMBER force field were included [58] , and UMP parameters were calculated with the RESP HF/6-31G* method using the Antechamber encoded in AMBER11 [62] and Gaussian09 [54] . Both systems were then solvated in ∼158×159×106 Å water boxes with TIP3P water molecules to ensure the minimum distance between any protein atom and the side of the box is 25 Å and neutralized with 25 ( 3 ) Na+ ions for the bound ( unbound ) complex . The final systems contained ∼2 . 4×105 atoms . To remove bad contacts in the solvated systems for the MD simulations , the steepest descent and conjugate gradient algorithm energy minimization methods were used . Energy minimization of the water molecules and counterions with a positional restraint of 500 kcal mol−1 Å−2 in the complex was first performed; the steepest descent method was applied for the first 2 , 000 steps , and then the conjugated gradient method was used for the subsequent 3 , 000 steps . Afterward , the entire system was minimized without any restraints; the steepest descent method was used for the first 4 , 000 steps , and then the conjugated gradient method was used for the subsequent 6 , 000 steps . After minimization , each system was heated gradually from 0 K to 300 K within 200 ps . This was followed by constant temperature equilibration at 300 K for 500 ps , with a positional restraint of 10 kcal mol−1 Å−2 in the complex in a canonical ensemble ( NVT ) . The structures from the final stage of equilibration were chosen as the initial conformations for subsequent MD simulations . A total of 2 microsecond ( µs ) MD simulations were performed for the simulation of ATP in a fully solvated water environment , and each of 100 ns MD simulations was conducted for the ATP bound and unbound UMP kinase ( Table S7 ) . All simulations were performed with periodic boundary conditions using the NPT ensemble . Langevin dynamics [63] was used to maintain the temperature at 300 K with a collision frequency of 1 ps−1 , and a Langevin piston was assigned to maintain the pressure at 1 atm . An integration step of 2 fs was set for the MD simulations . The long-range electrostatic interactions were incorporated by using the particle mesh Ewald method [64] with a cubic fourth-order B-spline interpolation and by setting the direct sum tolerance to 10−5 . A cut-off equal to 10 Å was used for short-range electrostatics and van der Waals interactions . The SHAKE method [65] , with a tolerance of 10−5 Å , was applied to constrain all covalent bonds that involve hydrogen atoms . After MD simulations , the analysis of RMSD of UMP kinase relative to the initial structure , together with the calculations of the temperature , total energy , mass density , and volume during MD simulations ( Figure S3 ) , suggests that the 100 ns MD simulations are sufficient for obtaining the stability of UMP-ATP complexes . To generate representative structural ensembles for the ATP unbound UMP state , a RMSD conformational clustering was performed . The cluster analysis was undertaken to group coordinate snapshots from the trajectory into distinct sets by virtue of the PTRAJ module in AMBER 11 . The clustering was performed with the average-linkage algorithm that has been described previously [66] . Structures were selected in 200 ps intervals over the simulation trajectory of the unbound state . The resulting 500 structures were superimposed using all Ca atoms to remove overall rotation and transition . Then , pairwise Cα atoms RMSD comparisons were performed between any snapshot and the average coordinate after rigid-body alignment using a threshold of 1 . 5 Å . Two main clusters were obtained . One snapshot was chosen from each cluster for NEB calculations . The NEB method [67]–[70] is a powerful algorithm to investigate the transition pathway for the binding of ATP to its allosteric and substrate binding sites . In NEB , a string of replicas ( or ‘images’ ) of the system are created and connected together with springs to form a discrete representation of a path from the start to end configuration . Minimization of the entire system with the start and the end point structures fixed provides a minimum energy path . Each image between the two point structures is connected to the previous and next image by ‘springs’ along the path that maintain each image from sliding down the energy landscape onto adjacent images . In NEB , the total force F on each image , i , is decoupled as a perpendicular and parallel force by a tangent vector [Eq . ( 1 ) ] . The perpendicular component of the force is obtained by subtracting out the parallel component of the force [Eq . ( 2 ) ] , where ∇V ( Pi ) is the gradient of the energy with respect to the atomic coordinates in the system at image , i , and τ is the 3N dimensional tangent unit vector that describes the path . The parallel component of the force accounts for the artificial springs linking each image together [Eq . ( 3 ) ] , where ki is equal to the spring constant between images Pi and Pi-1 , and P is the 3N dimensional position vector of image i . ( 1 ) ( 2 ) ( 3 ) In all NEB calculations , the end point was selected from the equilibrated ATP bound UMP kinase . The starting point was the source ATP unbound UMP kinase . We placed ATP in the bulk in which the minimum distance between ATP and UMP kinase was larger than 15 Å with different initial positions . Ten different sets of ATP configurations were chosen from the ATP clusters ( half from the compact and half from the extended clustering ) . To further enhance the sampling , two distinct unbound UMP kinase structures were used based on the conformational diversity of the unbound UMP kinase state in the MD simulations for each ATP configuration . Therefore , all twenty NEB calculations were performed to explore the dynamics process of ATP binding to the allosteric and substrate sites of the UMP kinase . The simulated annealing version of NEB from AMBER 11 was applied in these simulations . The initial NEB pathway consisted of eleven staring-points followed by eleven end-points . The initial path was heated from 0 K to 300 K in 100 ps with a Langevin dynamics of frequency of 1000 ps−1 and a spring force of 10 kcal mol−1 Å−2 . Then , the path was equilibrated at 300 K in 200 ps . After that , a total of 600 ps simulated annealing protocol ( Table S8 ) involved quickly heating the path to 500 K , followed by slow cooling and finally quenched the dynamics to remove any remaining kinetic energy from the path with the spring force of 50 kcal/mol [71] , [72] . The random number generator was seeded differently for each calculation . | The endogenous ATP can be regarded as a substrate and an allosteric modulator in cellular signal transduction . We analyzed the properties of allosteric and substrate ATP-binding sites and found that the allosteric ATP-binding sites are less conserved than the substrate ATP-binding sites . Allosteric ATP molecules adopt both compact and extended conformations in the allosteric binding sites , while substrate ATP molecules adopt extended conformations in the substrate binding sites . The two different conformational characteristics may be responsible for ATP to exert distinct biological functions in cell signaling . In addition , to our knowledge , this study illuminates the first comprehensive atomic level investigations of ATP access to the corresponding allosteric and substrate ATP-binding sites . Specially , both the adenine and triphosphate parts of ATP could be an allosteric trigger to propagate the signal from the allosteric to functional sites . The detailed mechanism presented in this study may apply to other enzymes in complex with allosteric or substrate ATP molecules , and provide important insights for the molecular basis of ATP acting as a substrate and an allosteric modulator in cell signaling . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"biology",
"and",
"life",
"sciences",
"computational",
"biology"
] | 2014 | The Structural Basis of ATP as an Allosteric Modulator |
Knowing brain connectivity is of great importance both in basic research and for clinical applications . We are proposing a method to infer directed connectivity from zero-lag covariances of neuronal activity recorded at multiple sites . This allows us to identify causal relations that are reflected in neuronal population activity . To derive our strategy , we assume a generic linear model of interacting continuous variables , the components of which represent the activity of local neuronal populations . The suggested method for inferring connectivity from recorded signals exploits the fact that the covariance matrix derived from the observed activity contains information about the existence , the direction and the sign of connections . Assuming a sparsely coupled network , we disambiguate the underlying causal structure via L1-minimization , which is known to prefer sparse solutions . In general , this method is suited to infer effective connectivity from resting state data of various types . We show that our method is applicable over a broad range of structural parameters regarding network size and connection probability of the network . We also explored parameters affecting its activity dynamics , like the eigenvalue spectrum . Also , based on the simulation of suitable Ornstein-Uhlenbeck processes to model BOLD dynamics , we show that with our method it is possible to estimate directed connectivity from zero-lag covariances derived from such signals . In this study , we consider measurement noise and unobserved nodes as additional confounding factors . Furthermore , we investigate the amount of data required for a reliable estimate . Additionally , we apply the proposed method on full-brain resting-state fast fMRI datasets . The resulting network exhibits a tendency for close-by areas being connected as well as inter-hemispheric connections between corresponding areas . In addition , we found that a surprisingly large fraction of more than one third of all identified connections were of inhibitory nature .
The networks of the brain are key to understanding its function and dysfunction [1] . Depending on the methods employed to assess structure and to record activity , networks may be defined at different levels of resolution . Their nodes may be individual neurons , linked by chemical or electrical synapses . Alternatively , nodes may also be conceived as populations of neurons , with links represented by the net effect of all synaptic connections that exist between two populations . In any case , this defines the structural substrate of brain connectivity , representing the physical ( causal ) basis of neuronal interactions . Nodes in a brain network influence each other by sending signals . For example , the activities of nodes in a network are generally not independent , and neuronal dynamics are characterized by correlations among the nodes involved in the network . This suggests an alternative perspective on active brain networks: Functional connectivity assigns a link to a pair of nodes to the degree to which their activities are correlated . It has been argued that this concept emphasizes connections that “matter” , including the possibility that the same substrate may give rise to different networks , depending on how they are used . As a consequence , functional connectivity and structural connectivity are not equivalent . A well-known phenomenon is that two nodes may be correlated , even if there is no direct anatomical link between them . For example , a shared source of input to both nodes may generate such a correlation , which does not correspond to a direct interaction between the two nodes . Apart from that , correlation is a symmetric relation between two nodes , whereas a physical connection implies a cause-effect relation that is directed . There have , in fact , been multiple attempts to overcome the shortcomings of functional connectivity , especially the lack of directed interaction . The term effective connectivity has been suggested for this [2] . The idea is to bring the networks , inferred from activity measurements , closer to structural connectivity , which can only be inferred with anatomical methods . The dichotomy between structural and functional aspects of connectivity raises the general question whether it is possible to infer brain networks from recorded activity . We are only beginning to understand the forward link between structural connectivity and functional connectivity . As a consequence , it is possible to compute correlations from connectivity in certain simplified network scenarios [3] . The correspondence between connectivity and correlation , however , is not one-to-one . Networks with different connectivity may lead to exactly the same correlations between nodes . As a consequence , the inverse problem of inferring connectivity from correlation is generally ill-defined . As we will demonstrate in this paper , additional assumptions about the connectivity can help to resolve the ambiguity . Specifically , we search for the network with the lowest number of nonzero edges ( via L1-minimization ) to disambiguate the problem . Structural , functional and effective connectivity are not equally well accessible . Some aspects of the anatomical structure can be assessed post mortem by invasive tracing methods , or non-invasively by Diffusion Tensor Imaging , DTI . In contrast , functional connectivity is based on statistical relationships between the activity of neuronal populations and can be easily estimated from recorded signals . For estimating effective connectivity there are methods like Dynamic Causal Modelling , DCM [4 , 5] , Granger causality [6] and others [7–13] . Only few methods to infer effective connectivity , however , can deal with large numbers of nodes ( 40 or more ) based on zero-lag correlation only . However , they are either limited to small networks [14] , or to directed acyclic graphs [15] . Here , we are proposing a new method for the estimation of effective connectivity from population activity in the brain , especially BOLD-related signals . The new method is a variant of the procedure described in [16] , based on a L1-minimization . For the method proposed here it is sufficient to use zero-lag covariances to estimate directed effective connectivity .
The main idea of our estimation method is inspired by the finding , “that the key to determining the direction of the causal relationship between X and Y lies in ‘the presence of a third variable Z that correlates with Y but not with X , ’ as in the collider X → Y ← Z…” [17 , 18] . Similarly , assuming a linear interaction model , the presence of a collider structure in a network ( see Fig 1 ) produces specific entries in the corresponding inverse covariance ( precision ) matrix . Fig 1 shows a disconnected network in the left column , and a network which induces the same covariance matrix if all links have opposite direction in the middle column . In the latter case an estimation of the direction is impossible , because there is simply no information about it in the covariance matrix . Whenever a collider structure is present , however , the entry in the inverse covariance matrix for the two source nodes ( here , 2 and 3 ) is non-zero . This is due to the fact that in a linear model the entry in the inverse covariance matrix depends not only on the connections of the nodes 2 and 3 , but also whether these nodes have a common target . This means the presence of a collider structure allows us to disambiguate the direction of this particular connection . We consider here a scenario , where the interaction between nodes is described by a generic linear model . Assuming stationarity , let the neural activity x ( t ) be implicitly defined by the consistency equation x ( t ) = ( G * x ) ( t ) + v ( t ) ( 1 ) where G ( t ) is a matrix of causal interaction kernels and v ( t ) denotes fluctuating external inputs ( “driving noise” ) . All variables are also listed in Table 1 . Fourier transformation of Eq ( 1 ) yields x ^ ( f ) = G ^ ( f ) x ^ ( f ) + v ^ ( f ) and simple rearrangement leads to x ^ ( f ) = ( 1 - G ^ ( f ) ) - 1 v ^ ( f ) where x ^ denotes the Fourier transform of x . The cross spectral density of the signals is then given by C ^ ( f ) = ( 1 - G ^ ( f ) ) - 1 Z ^ ( f ) ( 1 - G ^ * ( f ) ) - 1 where Z ^ ( f ) is the cross-spectral density of the external inputs . It follows C ^ - 1 ( f ) = ( 1 - G ^ * ( f ) ) Z ^ - 1 ( f ) ( 1 - G ^ ( f ) ) = B * ( f ) B ( f ) ( 2 ) with B ( f ) = Z ^ - 1 ( f ) ( 1 - G ^ ( f ) ) . In our model , we assume that the components of the external fluctuating input are pairwise stochastically independent for all nodes . Then , Z ^ is a diagonal matrix , and we make the additional assumption that Z ^ = 1 . For the linear model considered here , there is a relation between covariance and connectivity , which can be exploited for the estimation of connectivity from correlation . In the case Z ^ = 1 it is given by C ^ - 1 ( f ) = ( 1 - G ^ * ( f ) ) ( 1 - G ^ ( f ) ) = 1 - G ^ ( f ) - G ^ * ( f ) + G ^ * ( f ) G ^ ( f ) where the last term contributes the information of the collider structures . If the matrix product G ^ * ( f ) G ^ ( f ) has a non-zero off-diagonal entry the corresponding nodes have outgoing connections terminating at the same node , which means these nodes form a collider . It is clear that for any unitary matrix U ∈ U ( n ) the product UB is still a solution of Eq ( 2 ) , as U * U = 1 . We will resolve this ambiguity with an L1 minimization which is known to prefer sparse solutions under certain conditions [19] . In order to find G from a given C we first fix an initial matrix B , and then search for a unitary matrix U ∈ U ( n ) such that ‖UB‖1 is minimal , so we are minimizing the function Γ : U ( n ) ⟶ R U ⟼ ∥ U B ∥ 1 . ( 3 ) When estimating connectivity from simulations with known underlying network structure ( ground truth ) , one can quantify the performance of the estimation . For measuring the accuracy of our estimation we employ three different methods . First , we use the area under the ROC-curve ( AUC ) . The ROC ( receiver operating characteristic ) curve is obtained as following: For each possible parameter value ( in our case the threshold for the existence of a connection ) , the number of true-positives ( TP ) and false-positives ( FP ) is used to calculate the true-positive rate ( or recall ) TP/ ( TP+ FN ) and the false-positive rate FP/ ( FP+ TN ) . The ROC curve is then obtained by plotting the true-positive-rate against the false-positive rate . Secondly , we use the average precision score ( PRS ) which is the area under the precision-recall curve . This also includes the false-negatives ( FN ) ( precision: TP TP + FP ) . If both AUC and PRS are equal to 1 , the connections in the network are perfectly estimated . Sample curves are shown in Fig 2D . Thirdly , we calculate the Pearson Correlation Coefficient ( PCC ) which in contrast to the measures defined before also take the strength and the sign of the interactions into account . This also means that this measure is less suited to assess whether a connection exists or not . It rather measures whether the estimated connections have the same strength as the original ones . We consider all three performance measures simultaneously to establish the quality of our estimates . Seven healthy subjects underwent a 20-minute resting-state fMRI experiment on a 3 T Siemens Prisma scanner . The data was acquired using the MREG sequence [28] , yielding a high temporal resolution ( TR = 0 . 1 s , 12000 time points ) that facilitates functional connectivity analyses [29] . The other sequence parameters were TE = 36 ms , FA = 25° , 64 × 64 × 50 matrix and 3mm isotropic voxel size . Additionally , cardiac and respiratory signals were recorded with the ECG and abdominal breathing band from the scanner’s physiological monitoring unit . Motion correction was done with FSL and physiological noise correction was performed with RETROICOR [30] . Average CSF and white matter signals were regressed out , but no global signal regression was performed . Following image normalization to MNI space , voxels were parcellated according to the AAL atlas ( excluding the cerebellum ) , and the mean activity within each atlas region was calculated . The connectivity was then estimated using zero-lag covariances of the standardized signals .
Intrinsic properties of our new estimation procedure can be identified by studying the performance of the method for perfectly estimated ( noise-free ) covariance matrices . This way we address properties that do not depend on any particular feature of the underlying data , and that are not due to the success of the measurement process . In particular , we show for which types of networks our estimation procedure gives good results on technical grounds , with a wide range of networks hopefully including those arising in applications . We used random Erdős-Rényi connectivity profiles for all simulations . The macro-connectivity between neuronal populations has to satisfy certain conditions in order to be tractable by our methods . Two of these conditions concern the dynamic stability of the network and the strength of the interactions . There is a trade-off between the number of physical links and the resulting strength of macro-connections , and the dynamic stability of the network . To study the performance of our method in these various regimes , we separately varied the network size N , the connection probability p , and the absolute strength of connections |J| in the connectivity matrix G , while the fraction of inhibitory couplings was kept at 50% . The spectral radius ρ of the bulk eigenvalue spectrum is approximately given by ρ 2 = J 2 p ( 1 - p ) N . ( 8 ) The default values of the parameters used in our study were N = 100 , p = 0 . 1 and ρ = 0 . 7 , where only one of them at a time was systematically varied . Low values of the spectral radius ρ correspond to networks with weak recurrent interaction and high values to networks with strong interaction , respectively . According to the model of network interaction assumed here , the networks need to have a spectral radius ρ > 0 for network interaction to be present and ρ < 1 for the dynamics to be stable . First , our results in Fig 3A indicate that a certain minimal level of interaction is necessary to be able to estimate the connections reliably . Above a value of ρmin = 0 . 2 , the influence of the spectral radius on the performance of the estimation is weak , but the larger the spectral radius is the better the estimation gets . Secondly , the connection probability of the network influences the quality of the estimation . For all connection probabilities tested here the network size was kept constant at N = 100 nodes . The networks were constructed such that the strength |J| of all connections was the same and such that the spectral radius ρ was constant according to Eq ( 8 ) . Fig 3B shows that the estimation works very well for sparse matrices with a connection probability in the range between 5% and 15% . For networks with higher connection probability and equally strong connections , the performance decreases as expected , due to the bias associated with L1-minimization . But even for a connection probability of p = 0 . 21 , a fraction of 14 . 2% of the estimated connections are false negative , and 3 . 3% are false positive . More than 90% of the correctly estimated connections have the correct sign . In applications , the focus of the estimation often lies on the strongest connections in the network . In networks with a background of weak connections and a sparse skeleton of stronger connections , it is possible to selectively estimate these strong links although , strictly , the assumption of a sparse network is violated . Fig 3D shows the performance of our method for such networks: the networks consist of a skeleton of strong connections with connection probability p = 10% and a connection strength derived from Eq ( 8 ) for ρ = 0 . 7 . Additionally , we created a second network with weaker connections for various connection probabilities q . The two networks were combined by adding the connectivity matrices . The connection strength of this weaker connections is also derived from Eq ( 8 ) , with a spectral radius of the background network being 20% of the spectral radius of the skeleton network . Then the performance of the estimation is calculated with respect to the skeleton of strong connections . Thirdly , to be applicable to a broad range of data types , a method of connectivity estimation should perform stable for different network sizes N . For most common types of non-invasive recordings of population activity the number of nodes considered is in the range between 30 and 150 . It is , of course , possible to consider larger networks , although the estimation becomes computationally more expensive . The runtime of the algorithm for networks with 200 nodes still in the range of seconds on a state-of-the-art desktop computer , but even networks with 1000 nodes or more are tractable . The strength of the connections |J| are set such that the spectral radius ρ of G is constant; the connection probability is constant at p = 0 . 1 . Fig 3C shows that our method performs better for bigger networks . We have observed that the L1 cost landscape becomes smoother for larger networks . In order to create surrogate data which fit fast fMRI data [28] , we simulated interacting stochastic processes known as Ornstein-Uhlenbeck processes . In this case , the performance of the network inference depends on how well the inverse covariance matrix , which is the basis of the estimation , can be derived from the data . In addition to finite size effects , we studied the impact of observation noise on the performance , see Fig 4 . We used N = 100 , p = 0 . 1 , dt = 0 . 1 s , ρ = 0 . 74 and τ = 0 . 1 s as default values of the parameters . Generally , it seems natural to use Welch’s method to calculate cross-spectral densities directly , and then to estimate the connectivity for each frequency band separately . For the data described here , however , we can estimate the connectivity from zero-lag sample covariances in the time domain . This is possible when the mass of the covariance function is concentrated very close around lag 0 . Then lag 0 is the only one contributing to the integral of the covariance function , which corresponds to the cross-spectral density C ^ ( 0 ) . As shown in Fig 4A , with noisy data the AUC is still good , but the PRS is lower than in the case , where the covariance is known without error . However , for a signal-to-noise ratio above 1 the performance improves very quickly . In the case of fMRI usually the whole brain is scanned , and there are no unobserved nodes in the network . However , for other data types ( e . g . fNIRS ) only parts of the brain can be observed . The question then is , whether this sub-network can nevertheless be reconstructed from the recorded signals . To model this scenario , we took simulated data and removed randomly a certain subset of components from the dataset . The interaction of the removed nodes is then not part of the covariance matrix of the reduced dataset , although the unobserved nodes of course still exert their influence on the observed ones . The performance of the estimation of the sub-network based on the reduced dataset is shown in Fig 4B . Our analysis shows very clearly that the estimation still leads to reasonable results under these conditions . In fact , we can demonstrate that we are inferring causal connections only: For unconnected observed nodes X , Y and a latent node L connected to both X and Y , our method does not erroneously indicate a link between X and Y . One key factor for a reliable estimation of the covariance matrix is the amount of data available . This depends on the length of the measurement or simulation , and on the sampling rate . Since fast fMRI time series are obtained by measuring the BOLD response as a proxy of neuronal activity , the time scale of the measured data is relatively slow compared to the time scale of the underlying neuronal activity . Fig 5 shows the performance of network inference depending on the amount of data available , and on the time-scale of the neuronal activity . Not surprisingly , the more voluminous the dataset is , the better the estimation gets . On the other hand , it shows that the estimation generally leads to better results for slower temporal dynamics . Also , for data of sufficient length with a fairly good signal-to-noise ratio , the estimation of the connectivity is possible even when only a part of the network is observed . To allow comparison of our new method with other known methods for network inference [31–33] , we applied it to the NetSim dataset provided by [34] . For details on the result of this , please see Fig A in S1 Text . We estimated connectivity from seven fast fMRI datasets , for details see the methods section . The resulting networks , after a threshold of 10% was applied , consist of 810 connections for each dataset . The threshold of 10% was chosen arbitrarily . In Fig B in S1 Text we show the histogram of estimated connection strengths for all seven reconstructed networks before thresholding . The threshold is derived from the 10% strongest connections , disregarding their signs . As there is generally no full ground truth for the connectivity inferred from human fMRI recordings available [31 , 32 , 35] , we cannot definitely assess the degree to which the result of our inference are correct . We can , however , establish whether they are plausible . One representative connectivity matrix is shown in Fig 6 . On average , 34% of the connections were inhibitory , with negligible variability across subjects . Of all connections found , 301 ( 37% ) were found in four subjects or more , and 4872 out of 8100 possible connections were absent in all subjects . On average , 245 of the connections were bi-directional and 565 connections were identified only for one direction . In general , close-by areas are more likely to be connected than more distant ones . This fact is ( approximately ) represented by a concentration of connections along secondary diagonals in the within-hemisphere blocks . Also , there are frequent inter-hemispheric connections between corresponding areas . This fact is represented by the diagonal entries in the across-hemisphere blocks . As mentioned above , different heuristics have been suggested to reconstruct networks from neuronal signals . In Fig 7 we compare the performance of the new method we propose here and the established method of Regularized Inverse Covariance [34] , based on the implementation provided at https://fsl . fmrib . ox . ac . uk/fsl/fslwiki/FSLNets . Our comparison clearly shows that our new method performs significantly better than the Regularized Inverse Covariance method , mainly , because the latter cannot establish the direction of connections . The superior performance of the new method is reflected in higher values for all three performance measures , in particular PRS and PCC . As regularization parameter required by the software toolbox , we used λ = 5 . Furthermore , we applied the RIC method on all seven MREG datasets described before . A threshold was applied , such that only the 10% strongest connections are retained . To compare the outcome of both methods , we only condidered the existence of connections ( binary and symmetric connectivity ) and disregarded weights and directions ( weighted nonsymmetric connectivity ) . One representative example of the comparison of both methods is shown in Fig 8 . For RIC , 376 out of 810 possible connections where identified in four subjects or more out of seven , the corresponding number for our method is 392 out of 810 possible connections . If any method produced directed networks with 10% connection probability at random , this would yield an average count of less than 25 connections ( 3% of 810 connections ) that agree for least four out of seven independently generated networks . On average , 290 . 5 out of 4050 possible connections ( undirected ) are identified by both methods , 3530 . 5 connections were found by neither of the methods . This means that both methods agree on 3821 out of 4050 connections on average . The two methods disagreed on the remaining 229 connections .
With the presented method we can estimate directed effective connectivity on a whole-brain scale . Also we are able to detect whether connections are excitatory or inhibitory . The estimation is possible based on zero-lag covariances , but can also be applied to frequency-resolved cross spectral densities . | Changes in brain connectivity are considered an important biomarker for certain brain diseases . This directly raises the question of accessibility of connectivity from measured brain signals . Here we show how directed effective connectivity can be inferred from continuous brain signals , like fMRI . The main idea is to extract the connectivity from the inverse zero-lag covariance matrix of the measured signals . This is done using L1-minimization via gradient descent algorithm on the manifold of unitary matrices . This ensures that the resulting network always fits the same covariance structure as the measured data , assuming a canonical linear model . Applying the estimation method on noise-free covariance matrices shows that the method works nicely on sparsely coupled networks with more than 40 nodes , provided network interaction is strong enough . Applying the estimation on simulated Ornstein-Uhlenbeck processes supposed to model BOLD signals demonstrates robustness against observation noise and unobserved nodes . In general , the proposed method can be applied to time-resolved covariance matrices in the frequency domain ( cross-spectral densities ) , leading to frequency-resolved networks . We are able to demonstrate that our method leads to reliable results , if the sampled signals are long enough . | [
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion"
] | [
"medicine",
"and",
"health",
"sciences",
"diagnostic",
"radiology",
"functional",
"magnetic",
"resonance",
"imaging",
"neural",
"networks",
"particle",
"physics",
"random",
"variables",
"neuroscience",
"covariance",
"colliders",
"magnetic",
"resonance",
"imaging",
"simulation",
"and",
"modeling",
"mathematics",
"algebra",
"network",
"analysis",
"brain",
"mapping",
"neuroimaging",
"research",
"and",
"analysis",
"methods",
"computer",
"and",
"information",
"sciences",
"operator",
"theory",
"imaging",
"techniques",
"probability",
"theory",
"physics",
"radiology",
"and",
"imaging",
"diagnostic",
"medicine",
"linear",
"algebra",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"eigenvalues"
] | 2018 | From correlation to causation: Estimating effective connectivity from zero-lag covariances of brain signals |
Mutant dwarf and calorie-restricted mice benefit from healthy aging and unusually long lifespan . In contrast , mouse models for DNA repair-deficient progeroid syndromes age and die prematurely . To identify mechanisms that regulate mammalian longevity , we quantified the parallels between the genome-wide liver expression profiles of mice with those two extremes of lifespan . Contrary to expectation , we find significant , genome-wide expression associations between the progeroid and long-lived mice . Subsequent analysis of significantly over-represented biological processes revealed suppression of the endocrine and energy pathways with increased stress responses in both delayed and premature aging . To test the relevance of these processes in natural aging , we compared the transcriptomes of liver , lung , kidney , and spleen over the entire murine adult lifespan and subsequently confirmed these findings on an independent aging cohort . The majority of genes showed similar expression changes in all four organs , indicating a systemic transcriptional response with aging . This systemic response included the same biological processes that are triggered in progeroid and long-lived mice . However , on a genome-wide scale , transcriptomes of naturally aged mice showed a strong association to progeroid but not to long-lived mice . Thus , endocrine and metabolic changes are indicative of “survival” responses to genotoxic stress or starvation , whereas genome-wide associations in gene expression with natural aging are indicative of biological age , which may thus delineate pro- and anti-aging effects of treatments aimed at health-span extension .
The complexity of the aging process , as well as the conspicuous lack of tools to study it , has hindered hypothesis-driven reductionist approaches to identifying the molecular mechanism of aging in mammals . In mice , recent progress has revealed that several aspects of aging could be accelerated or delayed by single gene mutations . Mouse models for progeroid syndromes are invaluable for studying accelerated aging [1] . In humans , defects in the genome maintenance mechanisms can lead to a variety of progeroid disorders [2] suggesting a causative role of DNA damage in aging [3] , [4] , [5] . Prime examples are Cockayne syndrome ( CS; affected proteins: CSB , CSA ) , XPF-ERCC1 syndrome ( XFE; affected proteins: XPF , ERCC1 ) or trichothiodystrophy ( TTD; affected proteins: XPB , XPD , TTDA ) that are caused by defects in the transcription-coupled subpathway of nucleotide excision repair ( TC-NER ) [6] , [7] . NER removes a wide range of helix-distorting DNA damage such as UV lesions , and is divided into global genome ( GG-NER ) that recognizes helical distortions throughout the genome and TC-NER that removes transcription-blocking lesions on the transcribed strand of active genes [2] , [8] . Defects in TC-NER lead to progeria but are not associated with increased cancer predisposition , whereas defects in GG-NER lead to high susceptibility to skin cancer but not to significant accelerated aging . Recently , we applied genome-wide expression profiling to characterize the severe progeroid Csbm/m;Xpa−/− and Ercc1−/− [7] , [9] and the phenotypically milder progeroid Ercc1−/Δ−7 mouse models ( GAG/LJN unpublished data ) , and uncovered a systemic attenuation of the growth hormone/insulin-like growth factor 1 ( GH/IGF1 ) somatotropic axis along with the thyrotropic ( e . g . deiodinases I and II; thyroid hormone receptor ) and lactotropic ( e . g . prolactin receptor ) axes , suppression of oxidative metabolism ( glycolysis and the Krebs cycle ) and upregulation of anti-oxidant and detoxification defenses early in life [10] . On the other side , long-lived mouse models provide valuable insights into the biology of delayed aging [11] , [12] , [13] and point to an important role of the GH/IGF1 axis in determining lifespan . Ames and Snell dwarf mice [14] , [15] , GH releasing hormone ( GHRH ) defective little mice ( Ghrhrlit/lit ) [16] , GH receptor/binding protein null mice ( Ghr/bp−/− ) [17] and IGF1 receptor heterozygous mice ( Igf1r+/− ) [18] all show suppressed GH/IGF1 signaling , dwarfism and a significantly increased lifespan compared to wild type ( wt ) control mice . Likewise , calorie restriction ( CR ) results in decreased insulin/IGF1 signaling , prolonged lifespan and delays several age-associated diseases in mammals [19] . Intriguingly , attenuation of the somatotropic axis and oxidative metabolism also occur in naturally aged mice [7] , [9] . Defective DNA damage repair mechanisms can lead to lifespan shortening , whereas suppression of the somatotropic axis can lead to lifespan extension . The relation between these two distinct aspects of longevity , however , is unclear . The suppression of the GH/IGF1 axis in both progeroid and long-lived mice lies in contrast to the opposing nature of accelerated and delayed aging . We , therefore , sought to determine whether , and to what extent , these two extremes of murine lifespan are inter-related and further , how both relate to natural aging . Importantly , comparing genome-wide expression profiles across accelerated , delayed and normal aging could allow delineating markers indicative of biological age .
To analyze genome-wide expression profiles across the extremes of murine lifespan , we compared independent liver microarray datasets from a series of short-lived DNA repair-deficient mouse mutants with severe ( Csbm/m;Xpa−/− , Ercc1−/− ) , intermediate ( Ercc1−/Δ−7 ) , mild ( Csbm/m ) or no significant progeria ( Xpa−/− ) [7] , [9] with mice that show lifespan extension either due to genetic alteration ( Ames and Snell dwarf , growth hormone receptor knockout Ghr−/− mice ) [20] , [21] , [22] , calorie restriction ( CR ) [21] or a combination of both ( Ames-CR ) [21] ( Figure 1A ) . In addition , we generated extensive microarray datasets of lung , liver , kidney and spleen of 13- and 130-week old naturally aged wt mice . Information on the groups of mice , the wt control mice , genetic background , number , gender , age and tissue for which the expression profiles were generated is summarized in Figure S1A . Using these datasets , we asked whether the genes that have significantly altered expression in the livers of short-lived Csbm/m;Xpa−/− , Ercc1−/− and Ercc1−/Δ−7 progeroid mouse groups as compared to their corresponding wt control mouse groups ( Figure S1A ) have the same or opposite direction of expression change in the liver of long-lived mouse groups relative to their respective wt control mouse groups ( expression changes relative to corresponding wt controls; ≥1 . 2 fold change , two-tailed t-test p≤0 . 01 , Tables S1–S2 ) . To do this , we first classified all significantly differentially expressed genes derived from the short-lived Csbm/m;Xpa−/− mice relative to their respective wt controls in terms of having increased or decreased expression . We then asked how many of those genes show the same or opposite direction in expression in the long-lived mouse groups relative to their own wt control mouse groups using the non-parametric Spearman's rank correlation coefficient . This correlation method tests the direction and strength of the relationship between two variables resulting in values ranging from perfect similarity ( r = +1 . 0 ) to no correlation ( r = 0 . 0 ) or dissimilarity ( r = −1 . 0 ) ( Figure S1B and Methods ) . As depicted in Figure 1A , unlike the non-progeroid wt and Xpa−/− mice , the significant expression changes identified in the progeroid Csbm/m;Xpa−/− mice relative to their respective wt control mice showed a positive correlation with the expression changes in all long-lived mouse groups ( Ghr−/− , Ames , CR , Ames-CR and Snell; r = 0 . 20 to 0 . 50 , p≤10−4 ) as compared to their corresponding wt control groups ( Figure S1A ) , despite the difference in age , genetic background and gender . This was also confirmed by comparing the significant expression changes of 2- and 16-week old Ercc1−/− and Ercc1−/Δ−7 progeroid mouse groups relative to their corresponding wt control mice to those of long-lived mouse groups ( Figure S2A–B; Table S3 ) . As the expression profiles of NER progeroid mice are more comparable to the group of calorie restricted Ames dwarf mice ( Ames-CR ) -previously known to result in synergistic lifespan extension- than to the groups of Ames dwarf or CR mice alone ( Figure 1A and Figure S2 ) , our findings suggest that NER progeroid mice ( Csbm/m;Xpa−/− , Ercc1−/− and Ercc1−/Δ−7 ) share similar gene expression changes with both the somatotropic mutants ( Ghr−/− , Ames , Ames-CR and Snell ) and dietary restricted mice ( CR ) [21] . We next asked whether the correlations between the expression profiles were reciprocal . Following the same approach , the expression profiles from the significantly differentially transcribed genes of Ames-CR mice ( as compared to their wt controls ) also correlated significantly with those of progeroid Csbm/m;Xpa−/− , Ercc1−/− and Ercc1−/Δ−7 mice ( r: 0 . 15 to 0 . 34 , p≤10−4 , Figure 1B ) but not with those of mild progeroid ( Csbm/m ) , or non-progeroid ( Xpa−/− ) mice relative to their respective wt controls ( Figure S1A ) . To further challenge the strength of correlation between NER progeroid and long-lived mice , we then compared the transcriptomes of individual mice to each other on the basis of a predefined gene set . To do this , we first pooled all progeroid mice together ( Csbm/m;Xpa−/− , Ercc1−/− and Ercc1−/Δ−7 ) and derived the genes whose expression changed significantly as compared to the group of all their respective controls ( i . e . all the wt mice of Csbm/m;Xpa−/− , Ercc1−/− and Ercc1−/Δ−7; Figure S1A , two-sided t-test , p<0 . 01 ) . Next , we pooled all the long-lived mice together ( Ghr−/− , Ames , CR , Ames-CR and Snell ) and derived the genes whose expression changed significantly relative to the group of all their corresponding wt control mice ( i . e . all the wt mice of Ghr−/− , Ames , CR , Ames-CR and Snell; Figure S1A , two-sided t-test , p<0 . 01 ) . This approach generated two sets of genes representing the progeroid and long-lived gene sets . We then used all the significantly differentially expressed genes of both progeroid and long-lived gene sets to ask whether the significantly differentially expressed “progeroid” or “long-lived” genes have the same or opposite direction of expression change between the livers of any two mice employed in this study when each mouse was compared to the group of its respective wt controls ( Figure S1A ) . In Figure 1C , the deeper color of each cell indicates the positive ( red ) , negative ( blue ) or lack of correlation ( white ) between the transcriptomes of two mice . Unlike wt controls and non-NER progeroid mutants , long-lived and NER progeroid expression profiles correlated strongly to one another ( r: 0 . 40–0 . 95 ) forming a red triangle at the top of the plot ( Figure 1C ) . Thus , failure to repair DNA damage causes gene expression changes that are associated with hyposomatotropism and CR , either of which in itself promotes longevity . The seemingly paradoxical genome-wide associations prompted us to explore the biological processes that underlie the gene expression parallels between accelerated and delayed aging . We first grouped all genes according to their known or predicted biological function into gene ontology ( GO ) categories . Next , we asked which GO terms are significantly over-represented among the significantly differentially expressed genes in either the NER progeroid or long-lived mouse groups as compared to the group of their respective control mice ( Figure S1A; Methods ) . This approach revealed seven common biological processes that were significantly over-represented in each of the short-lived and the long-lived mice relative to their respective control mice ( Figure 2A ) . These processes were ranked by their relative enrichment score ( see Methods ) and included: ( i ) macromolecular biosynthesis , ( ii ) lipid metabolism ( iii ) hormonal regulation , ( iv ) carbohydrate metabolism , ( v ) response to oxidative stress , ( vi ) membrane metabolism and ( vii ) co-enzyme and co-factor metabolism . The identification of shared biological processes does not necessarily imply that expression of the same genes is altered ( e . g . , distinct sets of genes may elicit the same biological outcome ) or that the direction of the change in expression is the same . However , when we analyzed the expression changes of genes associated with the previously identified shared biological processes , we found a substantial suppression of the GH/IGF1 axis , oxidative metabolism ( suppression of glycolysis and Krebs cycle with upregulation of glycogen synthesis ) and peroxisomal biogenesis , coupled with a widespread up-regulation of a response to ( oxidative ) stress in long-lived and short-lived mice ( Figure 2C ) . Whereas the majority of genes associated with the somatotropic , thyrotropic and lactotropic axes , electron transport and oxidative metabolism were downregulated , genes associated with the response to oxidative stress , DNA repair and apoptosis were upregulated . Therefore , processes related to growth , stress and metabolism might be most responsible for the genome-wide expression parallels between NER progeroid and long-lived animals . Interestingly , reduced protein biosynthesis as well as enhanced detoxification mechanisms have been recently identified as evolutionary conserved mechanisms controlling lifespan in multiple animal species [23] . Few processes were only enriched in either the progeroid ( amino acid , purine metabolism and cell cycle; Figure 3B ) or long-lived animals ( innate immune response ) . There was no significantly over-represented process enriched in both mouse groups with opposite direction in gene expression . Importantly , the suppression of the GH/IGF1 axis seen already in the extremely short-lived NER progeroid mice ( Csbm/m;Xpa−/− , Ercc1−/− ) is not likely the result of a developmental defect as 16-week old adult Ercc1−/Δ−7 mice showed a similar response at a later stage in life ( Figure 2C and D ) . We next sought to examine whether biological processes altered in NER progeria , pituitary dwarfism or CR are also similarly altered in naturally aged mice . As the age-related functional decline is ubiquitously manifested in all tissues and stochastic DNA damage likely affects the function of most organs , we also asked whether there might be similar , systemic gene expression changes with aging . We analyzed the kidney , lung , and spleen transcriptomes in addition to liver , in a cohort of adult 13-week old and naturally aged 2 . 5 year-old wt mice ( C57BL/6J; n = 3 per age group per organ ) . We included all ∼45 , 000 probe sets to avoid any potential introduction of bias ( Table S4 ) . The transcriptome of 2 . 5 year-old mouse liver , kidney , lung and spleen tissues was compared to that of 13 week-old mice and the results compiled in a circular heat map ( red and green color indicate up- and downregulated genes , respectively; Figure 3A ) . The average fold change for each gene across all four organs was used to sort the expression profiles in a clockwise direction from the most negative ( deep green ) to the most positive ( deep red ) fold change . Subsequently , all genes were sorted by their consistency of expression changes across all four organs and plotted in a clockwise direction in five sections beginning with those downregulated in all organs ( Section I ) to those consistently upregulated ( Section V ) . Sections II and IV include genes for which expression was different in one of the four tissues , whereas section III includes genes for which the direction of expression changes was different in two of the four organs in aged mice . Approximately 70% of the total transcriptome represented genes with identical ( ∼20%; section I and V ) or similar ( ∼50%; section II and IV ) direction of expression in all four organs . The remaining ∼30% of the genome showed greater variance , likely reflecting previously described tissue-specific expression changes with age [24] or expression changes with no apparent relevance to aging . These data reveal a remarkably homogeneous expression with aging across mouse tissues with distinct physiology and marked differences in age-related pathology . To delineate which biological processes are significantly over-represented in each of the five sections of the circular map , we identified in each section , those genes whose expression was significantly altered in all organs examined ( spleen , kidney , lung and liver ) of 2 . 5 year-old relative to 13 week-old mice ( two-sided t-test; p≤0 . 01 ) . Then , using the set of significant genes in each section of the circular map , we identified all GO terms with an unequal distribution between this gene set and the remainder of the genome . Processes related to energy utilization and oxidative metabolism , growth , ubiquitin cycle and ATP synthesis were significantly over-represented in sections I and II , which include genes that are systemically downregulated ( Figure 3B ) . Immune and stress responses as well as programmed cell death ( apoptosis ) were significantly enriched in sections V and IV , containing genes that are upregulated with age . Biological processes such as cellular differentiation and tissue development were significantly over-represented in section III of the circular map , which contains genes with variable expression changes in the different organs . Interestingly , a large fraction of the common responses previously identified in accelerated and delayed aging also occur with normal aging . Genes that are associated with immune , stress and defense responses were overrepresented in the upregulated genes , whereas genes associated with growth , energy utilization , lipid and carbohydrate metabolism were overrepresented in the downregulated genes in aged liver , kidney , spleen and lung . A subsequent analysis to identify over-represented gene networks , representing pathways rather than biological processes , revealed genes related to the WNT , NOTCH , TGF-β and AKT/mTOR signaling pathways to be significantly enriched in section I , indicating significant down-regulation with aging ( Figure 3C ) . Interestingly , both WNT and AKT ( a downstream target of IGF-1 ) signaling have been previously implicated in longevity regulation [18] , [25] , [26] . In contrast , genes related to signaling in response to cytokines and activation of T-cell receptors were significantly enriched in section V , indicating significant up-regulation with aging . These data reveal a similar down-regulation of genes associated with growth , energy utilization and metabolism in aged mice as observed in progeroid NER mutant and long-lived mice and a similar up-regulation of genes associated with stress and defense responses . To independently assess the relevance of these processes to natural aging , we examined the expression levels of several genes relevant to immune responses ( Ccr2 , Tnfsf13 , Saa1 , Saa3 , Fcgr3 , Ccl6 , C1qb , C1qc ) , apoptosis ( Cd5l , Siva , Tnfrsf21 , Tnfrsf1a , Casp4 ) , carbohydrate and lipid metabolism ( Impa1 , Gyk , Phkb , Crot , Dhrs8 , Akr1d1 ) and ATP and protein biosynthesis ( Atp5k , Harsl , Rsl1d1 , Atp5k , Rpl37 ) in livers derived from an independent aging cohort of male mice ( n = 6 ) by means of quantitative real time PCR ( Figure 4A–D ) . Confirming the gene expression changes of the first aging cohort of female mice ( Figure S3 ) , we found that all examined genes associated with immune and apoptotic responses were significantly upregulated in the 130-week mouse livers ( compared to 13-week old mouse livers ) whereas the expression of those genes associated with energy utilization and metabolism was substantially downregulated . The identification of overlapping biological processes between NER progeroid , long-lived and naturally aged mice , prompted us to measure the extent to which each of the progeroid and long-lived transcriptomes relate to those of natural aging . To facilitate this , we pooled all progeroid mice together ( Csbm/m;Xpa−/− , Ercc1−/− and Ercc1−/Δ−7 ) and derived the genes whose expression changed significantly as compared to the group of all their respective controls ( i . e . all the wt mice of Csbm/m;Xpa−/− , Ercc1−/− and Ercc1−/Δ−7; Figure S1A , two-sided t-test , p<0 . 01 ) . Next , we also pooled all the long-lived mice together ( Ghr−/− , Ames , CR , Ames-CR and Snell ) and derived the genes whose expression changed significantly relative to the group of all their corresponding wt control mice ( i . e . all the wt mice of Ghr−/− , Ames , CR , Ames-CR and Snell; Figure S1A , two-sided t-test , p<0 . 01 ) . This approach generated two sets of genes representing the progeroid and long-lived gene sets respectively . Then , using the non-parametric Spearman's rank correlation coefficient , we asked whether the “progeroid” or “long-lived” genes have the same or opposite direction of expression change with those seen in the livers of 130-week old as compared to 13-week old mice . This approach revealed a significant positive correlation between the significant expression profiles of short-lived NER-deficient mutants but not those of long-lived mice with aged mice ( Figure 5A ) . Importantly , implementing the reverse approach to compare the significant expression profiles of naturally aged 130-week old mice relative to 13-week old mice to those of the short- or long-lived mouse groups showed that this correlation was bi-directional ( Figure S4 ) . Thus , despite a number of shared biological processes , the gene expression profiles of naturally aging mice correlate , on a genome-wide scale , with the progeroid but not with the long-lived mutant dwarfs or CR mice .
Genome instability promotes aging and shortens lifespan and , therefore , a priori is not anticipated to be associated with lifespan extension . Contrary to these expectations , we found significant genome-wide , reciprocal associations between mice with accelerated aging due to DNA repair defects and those with delayed aging . The genome-wide expression responses in the NER progeroid mouse models ( Csbm/m;Xpa−/− , Ercc1−/− and Ercc1−/Δ−7 mice ) included the suppression of the GH/IGF1 somatotropic axis and oxidative metabolism along with upregulated stress responses . Paradoxically however , the systemic suppression of the GH/IGF1 axis and energy metabolism ( i . e . glycolysis , tricarboxylic acid cycle and oxidative respiration ) , together with the upregulation of antioxidant and detoxification defense genes are all associated with increased longevity rather than with the short lifespan of progeroid mice . For instance , Ghr−/− mice , as well as Ames and Snell dwarf mice that have a primary , congenital defect in the pituitary gland responsible for producing GH , thyroid stimulating and prolactin hormones , benefit from delayed age-related pathology and substantially prolonged lifespan [27] , [28] . Conversely , transgenic mice over-expressing GH show pathology early in life , increased tumor incidence and die prematurely [29] . However , CR mice show a similar longevity response , including suppression of the GH/IGF1 axis , but carry neither a pituitary defect ( as in the dwarf mutants ) nor are they exposed to rapid accumulation of irreparable DNA damage ( as in the NER progeroid mutants ) . Thus , whereas pituitary dwarf mutants reveal the biological significance of these processes in lifespan extension , the similar expression patterns in NER progeroid mice and CR mice suggest that intrinsic and environmental stressors , such as genome instability or food shortage can trigger similar transcriptional changes . Suppression of genes associated with the somatotropic , lactotropic and thyrotropic axes together with the upregulation of genes associated with stress and defense responses are indicative of a shift from growth to somatic maintenance [30] . This reallocation of the organism's resources from growth to somatic preservation might have evolved as a mechanism to overcome crises such as food shortage , infection or other disease states ( Figure 5B ) . For instance , the C . elegans long-lived dauer larvae are formed during starvation periods by suppressing Insulin/IGF1-like signaling [31] . In mammals , a shift from growth to somatic preservation may also function as a tumor suppressor mechanism [32] , [33] and accordingly may explain the diverse outcomes of distinct DNA repair deficiencies in human diseases . Whereas irreparable DNA damage accumulates in both TC-NER and GG-NER deficiencies , only TC-NER leads to progeroid syndromes though without associated cancer predisposition . Indeed , not only CR or mutant dwarf mice but also progeroid XpdTTD mice , which have a milder TC-NER defect than Csbm/m;Xpa−/− , Ercc1−/− and Ercc1−/Δ−7 mice , are protected from tumorigenesis and despite showing segmental progeria in a variety of tissues they show a number of paradoxically improved histopathologic changes related to a CR-like phenotype in other tissues such as lower incidence and/or severity of de-myelination of the peripheral nerve , cataract , thyroid follicular distension , pituitary adenomas and ulcerative dermatitis [34] . Interestingly , a similar shift in gene expression towards somatic preservation as seen in NER progeria , mutant dwarfism and CR also occurs with natural aging . For example , the majority of upregulated genes was associated with immune , stress and defense responses as well as programmed cell death , whereas the majority of downregulated genes was associated with growth , energy utilization , lipid and carbohydrate metabolism in aged liver , kidney , spleen and lung . A number of recent consonant findings suggest evolutionary conservation of our findings . For instance , Tower et al . , have found that immune genes are induced with age and in fact are predictive of remaining lifespan in Drosophila [35] whereas others have recently showed that NFkappaB activity , a key driver of immune and stress response genes , is induced with age in multiple murine and human tissues [36] . Importantly , however , a genome-wide correlation analysis , which extends to far broader expression changes than those linked to longevity-associated biological processes , revealed a strong association between naturally aged and progeroid but not long-lived mice . The correlations we found between certain groups of mice are most likely due to distinct groups of differentially expressed genes , i . e . there might be one large set of genes similarly affected in short-lived and long-lived mice and a separate large group of genes similarly affected in progeroid and naturally aged mice . This appears indeed to be the case . Nonetheless , there are also groups of genes , such as genes of the somatotropic axis that are similarly affected in accelerated , delayed and natural aging . However , these are smaller sets of genes that cannot influence genome-wide correlations in gene expression . In terms of the expression similarities , the resemblance of progeroid mice to both naturally aged and long-lived dwarfs suggests that progeroid mice harbor “expression signatures” associated with either age-related pathology or lifespan assurance mechanisms that likely attempt to counteract it . However , as age-related pathology is most likely absent in long-lived mice at this age ( 6 months ) , it might also explain the lack of any substantial positive correlation between the expression profiles of long-lived and naturally aged mice . It also suggests that both accelerated and natural aging trigger a common gene expression response , likely in response to accumulating ( DNA ) damage [37] , [38] . However , the NER progeroid and long-lived animals employed in this study had a biological age of ∼50% and 10–15% of their lifespan respectively . Thus , these findings also indicate that a genome-wide correlation analysis may serve as a powerful tool to determine the biological age of animals and might hence allow prognosis of longevity . In contrast , the expression of particular genes ( i . e . specific biomarkers ) or groups of genes associated with e . g . somatic growth or oxidative metabolism are not indicative of the biological age as they are similarly affected in natural , accelerated and delayed aging . Determination of biological age is indispensable for the assessment of anti-aging treatments . Although reliable biomarkers of aging are long sought after , they have yet remained elusive [39] . To this end , single genes or limited sets of genes used as biomarkers of aging may poorly reflect a true biological age; instead these markers likely indicate survival responses that can be beneficial upon intrinsic or extrinsic challenges ( e . g . macromolecular damage or limited food availability ) but futile when the detrimental effects of DNA damage accumulation become too severe , as in the case of progeroid syndromes . In diagnostic terms , a CR treatment might , therefore , equally induce a similar age-related biomarker , as treatment with a DNA damaging agent does . We , therefore , propose the facilitation of comprehensive genome-wide correlation analyses to evaluate pro- and anti-aging effects of treatments aimed at health-span extension . It will also be of utmost importance to identify biological parameters based on e . g . the previously obtained genome-wide expression profiles , which may also be applicable to easily accessible samples such as sera .
The generation and characterization of NER-deficient Csbm/m , Xpa−/− , Csbm/m;Xpa−/− , Ercc1−/− and Ercc1−/Δ−7 mice has been previously described [7] , [9] , [40] , [41] , [42] . With the exception of Ercc1−/− and Ercc1−/Δ−7 mice which were generated in an FVB∶C57BL/6J ( 50∶50 ) genetic background , all mice were in a C57BL/6J genetic background . Animals were kept on a regular diet and housed at the Animal Resource Center ( Erasmus University Medical Center ) and the National Institute of Public Health and the Environment ( RIVM ) , which operate in compliance with the “Animal Welfare Act” of the Dutch government , using the “Guide for the Care and Use of Laboratory Animals” as its standard . As required by Dutch law , formal permission to generate and use genetically modified animals was obtained from the responsible local and national authorities . All animal studies were approved by an independent Animal Ethical Committee ( Dutch equivalent of the IACUC ) . Standard procedures were used to obtain total RNA ( Qiagen ) from the liver , kidney , spleen and lung of naturally aged wt mice ( 3 animals per age group per organ ) at 13 and 130 weeks of age as well as from the liver of 16-week old Ercc1−/Δ−7 mice and wt control mice ( 6 mice per group ) . Synthesis of double-strand cDNA and biotin-labeled cRNA was performed using the GeneChip Expression 3′-Amplification IVT Labeling kit according to the manufacturer's instructions ( Affymetrix , USA ) . Fragmented cRNA preparations were hybridized to full mouse genome oligonucleotide arrays ( Affymetrix , mouse 430 V2 . 0 arrays ) , using Affymetrix hybridization Oven 640 , washed , and subsequently scanned on a GeneChip Scanner 3000 ( Affymetrix , USA ) . Initial data extraction and normalization within each array was performed by means of the GCOS software ( Affymetrix ) . Expression intensities were log transformed and normalized within and between arrays with the quantile normalization method using the R open statistical package ( http://www . r-project . org/ ) . Figure S1A provides detailed information on the different mouse groups , the number of animals , their gender , genetic background , age and tissue of each mouse group . Data were collected from cited sources or generated in this study: To facilitate the comparison , analysis was restricted to the 8 , 524 probe sets ( Table S3 ) that were present in both Affymetrix microarray platforms used in this and previous studies with long-lived mice [20] , [21] , [22] ( Affymetrix Mouse Genome 430 Av2 and Affymetrix Murine genome UV74 sets ) . When the comparison did not include the long-lived mice , the analysis was extended to include the full mouse transcriptome covering all known and predicted genes in the Affymetrix Mouse Genome 430 Av2 platform ( Table S4 ) . All microarray experiments complied with the standards set by the “minimum information about microarray experiments; ( MIAME ) ” and are available through ArrayExpress , a public repository for microarray experiments . The accession codes are: E-MEXP-835 for Csbm/m;Xpa−/− , Csbm/m , Xpa−/− and wt , E-MEXP-834 for Ercc1−/− and wt , E-MEXP-1503 for Ercc1−/Δ−7 and wt and E-MEXP-1504 for 13- and 130-week old wt mouse liver , spleen , kidney and lung . Two-tail , pair-wise analysis or a two-way analysis of variance was used to extract the statistically significant data from each group of mice by means of the Spotfire Decision Site software package 7 . 2 v10 . 0 ( Spotfire Inc . , MA , USA ) . The criteria for significance were set at p≤0 . 010 and a ≥±1 . 2-fold change in gene expression . We used the bivariate correlations procedure to compute Spearman's rho , and Pearson's correlation coefficient with their two-tailed significance levels by means of the statistical package SPSS 12 . 0 . 1 . ( SPSS Inc . IL , USA ) . All correlations reported were calculated by Spearman's rank correlation ( except the heatmap visualization where correlations were calculated by Pearson's r correlation; Figure 1C ) . The Spearman's rank correlation coefficient ( rho ) is a non-parametric measure of correlation that assesses how well an arbitrary monotonic function describes the relationship between two variables without making any assumptions about the frequency distribution of the variables . The Spearman rank correlation coefficient is defined as: where di = the difference between each rank of corresponding values of x and y . The correlation coefficients were derived from comparisons of expression profiles between two mouse genotypes ( as in Figure 1 ) . These coefficients were calculated on coordinates assigned to genes in each of the following categories: upregulated in both mouse groups ( 1 , 1 ) , upregulated in one mouse group and down in another ( [1 , −1] or [−1 , 1] ) , or downregulated in both mouse groups ( −1 , −1 ) . Genes having no direction of change ( +1 . 2>fold change>−1 . 2 ) when all mice of the same genotype were compared against their own wt controls were discarded , because these genes lack any information about changes associated with progeria , long-lived dwarfism , calorie restriction or aging . In addition , by scoring for qualitative rather than quantitative similarities , this approach disregards variations in the magnitude of gene expression that might originate from , for example , differences in genetic background , sex or animal housing conditions . In addition , when examining the p-values derived from the Spearman rank correlation analysis , one should note that such an analysis is based on the assumption that all probe sets represent statistically independent pairs of variables: each pair has a value derived from either the long- or short-lived mice and there is one pair per probe . However , for a number of genes , their expression changes can be highly correlated with each other because genes are functionally interrelated such as being part of the same pathway . As a result , the data used in the correlation application may not be as statistically independent as it was originally assumed . In fact , these values could be mutually correlated because “correlated” genes in a small group of mice might drive them all . The Pearson correlation is defined as: All significant gene entries were subjected to GO classification ( http://www . geneontology . org ) . Significant overrepresentation of GO-classified biological processes was determined by comparing the number of genes in a given biological process that were significantly differentially expressed in a particular mouse strain to the total number of the genes relevant to that biological process printed on the array ( Fisher exact test , p≤0 . 01 False discovery rate ( FDR ) ≤0 . 1 ) using the publicly accessible software Ease and/or DAVID ( http://david . abcc . ncifcrf . gov/summary . jsp ) . Due to the redundant nature of GO annotations , we employed kappa statistics to measure the degree of the common genes between two annotations and heuristic clustering to classify the groups of similar annotations according to kappa values ( http://david . abcc . ncifcrf . gov/summary . jsp ) . Significant overrepresentation of pathways and gene networks was determined by DAVID ( http://david . abcc . ncifcrf . gov/summary . jsp; through BBID , BIOCARTA and KEGG annotations ) as well as by means of the ingenuity pathway analysis software ( www . ingenuity . com ) . The expression data from naturally aged mice is summarized using a visualization that sorts all probe sets present on the Affymetrix GeneChip™ by their pattern of expression across all four tissues ( Figure 3A ) . It was created as PNG file using a combination of Perl , the GD . pm graphics module and the gdlib graphics library ( http://www . boutell . com/gd/; http://search . cpan . org/dist/GD/ ) . The circular map maximizes the display area by plotting data around a series of concentric circles . Probe sets were sorted by their consistency of expression across all four tissues and plotted in a clockwise direction: those downregulated in all tissues towards the top right; those upregulated in all tissues towards the top left and those with mixed expression states towards the middle . This results in five sectors . Probe sets within each sector are ordered by the most extreme average fold change observed for that probe set in each of the four tissues . Due to the high density of the Affymetrix GeneChip ( ∼45 , 000 Probe sets ) , it was necessary to combine data from individual probe sets prior to plotting . The direction of the probe sets were counted in the minimum length of the arc that was practical to plot , then colored red if the majority of probe sets were upregulated and green if the majority were downregulated . The innermost circle denotes the average fold change across all tissues . The color ramp ranges from green to white to red representing -4-fold change to no change to +4-fold change . A detailed representation of all expression changes depicted in the circular map is shown in Table S4 . Total RNA was isolated from liver , heart , kidney , spleen and lung of 13- and 130-week old mice as well as the livers of 2-week old Ercc1−/− and 16-week old day Ercc1−/Δ−7 mice using a Total RNA isolation kit ( Qiagen ) as described by the manufacturer . Quantitative PCR ( Q-PCR ) was performed with a DNA Engine Opticon device according to the instructions of the manufacturer ( MJ Research ) . Primer pair designed to generate intron-spanning products of 180–210 bp were as follows: Ghr: 5′-ATTCACCAAGTGTCGTTCCC-3′ and 5′-TCCATTCCTGGGTCCATTCA-3′; Igf1: 5′-TGCTTGCTCACCTTCACCA-3′ and 5′-CAACACTCATCCACAATGCC-3′; Prlr: 5′-GCATCTTTCCACCAGTTCCG-3′ and 5′-GCTCGTCCTCATTGTCATCC-3′; Dio1: 5′-CCCTGGTGTTGAACTTTGGC-3′ and 5′-TGAGGAAATCGGCTGTGGA-3′; Saa1: 5′-CATTTGTTCACGAGGCTTTCC-3′ and 5′-TGTCTGTTGGCTTCCTGGT-3′; Saa3: 5′-AGCCTTCCATTGCCATCATT-3′ and 5′-CTTCTGAACAGCCTCTCTGG-3′; Fcgr3: 5′-TGATGTGCCTCCTGTTTGC-3′ and 5′-GAGCCTGGTGCTTTCTGATT-3′; C1qb: 5′-TGTCCAACAGCAAGCAGGTC-3′ and 5′-TCAGGAAAGAGCAGGAAGCC-3′; C1qc: 5′-AGCACACAGTCAGGACCAA-3′ and 5′-AGTCAGGGAAGAGCAGGAAG-3′; Tnfrsf21: 5′-TGTGAACAAGACCCTCCCGA-3′ and 5′-ACACCACGATGACCACCAA-3′ Tnfrsf1a: 5′-AAAGTGTGGAGATGGGCAAA-3′ and 5′-CTGGCTGACATTTATCGCAC-3′; Impa1: 5′-CCAGAGCACCAGAGACTGTA-3′ and 5′-CCCACCTGTCACATCCATT-3′; Ccr2: 5′-ATTCTCCACACCCTGTTTCG-3′ and 5′-CCTTCGGAACTTCTCTCCAAC-3′; Ccl6: 5′-ATGAGAAACTCCAAGACTGCC-3′ and 5′-TGCTGATAAAGATGATGCCCG-3′; Tnfsf13: 5′-ATCTAAGGAGAGAGGTGGCTC-3′ and 5′-ACCGAGTGCTTCTTCTTCTGT-3′; Cd5l: 5′- CGACACAACAGCAGCAGAA-3′ and 5′-CTGGAAACCCACATACGACTC-3′; Casp6: 5′-ACATCAGACAGCACATTCCTG-3′ and 5′-GTAGACCTGGACAGTGGCAA-3′; Siva: 5′-CGCTCCAACTCAAAGTCCA-3′ and 5′-GCCATCAGGTCCAATCAACA-3′; Gyk: 5′-GAGGGAGGAATAGGTTGGAGA-3′ and 5′-GACAAGGGATAGCAATGACCA-3′; Phkb: 5′- ACATTCTCCAGCCTCAACAGA-3′ and 5′-ACCATTAGGTGTGCGTTCCA-3′; Crot: 5′-ATGTATCCCAAGCCAAAGCC-3′ and 5′-AAGGTATCAGGGTGAAGGGC-3′; Dhrs8: 5′-CTTCTTGCTGGCTTACTGCT-3′ and 5′-TGGTGCTTGGGTTCTTGATG-3′; Akr1d1: 5′-TTTCAACATCCAGCGAGGG-3′ and 5′-AGCAACTCCACATAGCGGA-3′; Atp5k: 5′-TTCAGGTCTCTCCACTCATCA-3′ and 5′-TATTCTCCTCTCCTCCTCTGC-3′; Harsl: 5′- CTATCCCAGAACAAGCAGGC-3′ and 5′-CAGGCTGAGGTCAAAGGAGA-3′; Rsl1d1: 5′-AATGCGGGCTCAAGACATC-3′ and 5′-CTGACTTCCCAGTTTCCACAA-3′; Rpl37: 5′-GGTCGGATGAGGCACCTAAA-3′ and 5′-AAGAAGAACTGGATGCTGCG-3′; Adcy1: 5′-TTACTGGTCACAGCCGCCTT-3′ and 5′-ATCCGCACGAAGACGCCATA-3′ . The generation of specific PCR products was confirmed by melting curve analysis ( which measures product specificity by the decrease in fluorescence signal when the PCR product is denatured ) and gel electrophoresis ( using Roche Agarose MS for analyzing small PCR products ) . Each primer pair was tested with a logarithmic dilution of a cDNA mix to generate a linear standard curve ( crossing point ( CP ) plotted versus log of template concentration ) , which was used to calculate the primer pair efficiency ( E = 10 ( −1/slope ) ) . Hypoxanthine guanine phosphoribosyltransferase1 ( Hprt-1 ) mRNA was used as an external standard . For data analysis , the second derivative maximum method was applied: ( E1gene of interest ΔCP ( cDNA of wt mice - cDNA of genetically modified or treated mice ) gene of interest ) / ( Ehprt-1 ΔCP ( cDNA wt mice- cDNA of genetically modified or treated mice ) hprt-1 ) . | To identify mechanisms that regulate mammalian longevity , we have quantified the expression parallels of a number of long-lived mice that show delayed aging and DNA repair mutants that age and die prematurely . Unexpectedly , we found significant , genome-wide similarities and a widespread overlap of over-represented biological processes in the transcriptomes of these disparate mouse strains . Subsequent analysis revealed that similar responses are triggered constitutively in a number of organs in aged mice . Thus , both intrinsic and environmental stressors ( e . g . , aging , genome instability , or food shortage ) induce survival responses aimed at overcoming crisis and extending lifespan . Such survival responses are likely to allow assessment of biological age as well as provide valuable targets for therapies aimed at health-span extension . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"molecular",
"biology/bioinformatics",
"genetics",
"and",
"genomics/animal",
"genetics",
"genetics",
"and",
"genomics/bioinformatics",
"molecular",
"biology/dna",
"repair"
] | 2008 | Delayed and Accelerated Aging Share Common Longevity Assurance Mechanisms |
Mycobacterium abscessus is considered the most common respiratory pathogen among the rapidly growing non-tuberculous mycobacteria . Infections with M . abscessus are increasingly found in patients with chronic lung diseases , especially cystic fibrosis , and are often refractory to antibiotic therapy . M . abscessus has two morphotypes with distinct effects on host cells and biological responses . The smooth ( S ) variant is recognized as the initial airway colonizer while the rough ( R ) is known to be a potent inflammatory inducer associated with invasive disease , but the underlying immunopathological mechanisms of the infection remain unsolved . We conducted a comparative stepwise dissection of the inflammatory response in S and R pathogenesis by monitoring infected transparent zebrafish embryos . Loss of TNFR1 function resulted in increased mortality with both variants , and was associated with unrestricted intramacrophage bacterial growth and decreased bactericidal activity . The use of transgenic zebrafish lines harboring fluorescent macrophages and neutrophils revealed that neutrophils , like macrophages , interact with M . abscessus at the initial infection sites . Impaired TNF signaling disrupted the IL8-dependent neutrophil mobilization , and the defect in neutrophil trafficking led to the formation of aberrant granulomas , extensive mycobacterial cording , unrestricted bacterial growth and subsequent larval death . Our findings emphasize the central role of neutrophils for the establishment and maintenance of the protective M . abscessus granulomas . These results also suggest that the TNF/IL8 inflammatory axis is necessary for protective immunity against M . abscessus and may be of clinical relevance to explain why immunosuppressive TNF therapy leads to the exacerbation of M . abscessus infections .
The rapidly-growing non-tuberculous mycobacteria ( NTM ) , Mycobacterium abscessus ( Mabs ) , is an emerging pathogen that causes a wide clinical spectrum of syndromes , including skin and soft tissues infections and pseudotuberculous pulmonary infections , especially in patients with underlying lung disorders [1–3] . Mabs is considered to be the most pathogenic NTM affecting cystic fibrosis ( CF ) patients , and is often associated with a dramatic decline in lung function and even death in these patients [4] . This organism is notorious for being intrinsically resistant to most antibiotics and disinfectants [5 , 6 , 7] and unsuccessful eradication of Mabs is a contraindication to lung transplantation in many CF centers , leaving patients without any therapeutic option . Despite being a rapid grower , Mabs can persist for years or decades within the lungs of infected patients [8 , 9] where it forms organized granulomas [10] . These hallmarks of pathogenic mycobacteria are composed of infected macrophages surrounded by additional macrophages , neutrophils and lymphocytes , and the centers of these tightly aggregated structures can develop caseous necrosis [10] . The pathways leading to Mabs granuloma formation and maintenance , however , have been poorly characterized . Mabs exhibit rough ( R ) and smooth ( S ) morphotypes that , as a consequence of alterations within the GPL biosynthetic/transport gene cluster , differ in the amounts of surface-associated glycopeptidolipids ( GPL ) [11] . Pulmonary Mabs infections in CF patients that are characterized by chronic airway colonization and poor outcomes have been linked to a genetic conversion allowing the S variant to become R [8 , 12 , 13] . This is also supported by results in mice and in cultured macrophages , emphasizing the hyper-virulent phenotype of the R compared to the S form [14–16] . MmpL4b is involved in translocation of GPL to the bacterial surface and its absence correlates with the lack of GPL and the highly virulent phenotype of the R variant [11 , 17 , 18] . A plausible explanation for the enhanced virulence of the Mabs R morphotype is that loss of GPL unmasks cell wall inflammatory-provoking lipoproteins and/or phosphatidyl-myo-inositol mannosides known to be TLR2 agonists [19 , 20] . Despite the demonstration that Mabs R induces a stronger TLR2-mediated TNF response than Mabs S [15 , 20] , there is little information regarding the events leading to the inflammatory response in Mabs infection and how the inflammation impacts on the outcome of the disease . By exploiting the optical transparency of the zebrafish embryo model , we confirmed the hyper-virulence of Mabs R and revealed the ability of both Mabs morphotypes to induce granulomas [17] . The increased virulence of the R variant correlated with a massive production of extracellular serpentine cords of bacteria which , due to their size , prevent phagocytosis , thus suggesting cording as a mechanism of immune evasion . Cords also initiated abscess formation , particularly in the central nervous system ( CNS ) of the infected animal , with subsequent tissue damage presumably caused by the induction of a potent inflammatory response . Lung injury in CF patients is caused by an intense and persistent pulmonary infection associated with a massive influx of neutrophils into the airways [21] . Thus , scrutinizing the inflammatory and neutrophilic response to Mabs infection could provide important advances in our understanding of the Mabs immunopathology and the unexplained innate susceptibility of CF patients to these infections . Herein , through the use of loss- and gain-of-function approaches coupled with fluorescent reporter zebrafish lines and high resolution imaging , we have dissected the TNF/IL8-mediated signaling pathway that contributes to immuno-protection against Mabs infection . Our findings unraveled the crucial and dual role of TNF in activating the macrophage bactericidal activity , in restricting intracellular bacterial growth and , importantly , in neutrophil recruitment for the generation and maintenance of protective granulomas .
To identify key effectors of the Mabs-induced inflammation , the pro-inflammatory profile in R- and S-systemic infected zebrafish embryos was determined and compared at various time points post-infection . Quantitative RT-PCR revealed an induction of tnf-α ( Fig 1A ) , il1-β and ifn-γ2 ( S1 Fig ) . Expression of tnf-α is induced by both variants at 2 days post-infection ( dpi ) , corresponding to the early appearance of granulomas , and prior to abscess formation [17] . This response was further increased at 5 dpi , with higher levels for R , consistent with the increased TNF release in murine macrophages [15] . This indicates that Mabs , notably the R variant , induces a robust TNF response in zebrafish . Tg ( tnfα:eGFP-F ) zebrafish larvae , which express farnesylated eGFP under the control of the tnf-α promoter [22] , were infected with Mabs to investigate the nature and the spatiotemporal distribution of the TNF-producing cells . While the control PBS injection in the otic cavity failed to induce eGFP expression , all animals injected with Mabs exhibited green fluorescent cells that were recruited and clustered around the injection site as early as 2 hours post-infection ( hpi ) ( Fig 1B and 1C ) , with equal numbers of eGFP-positive cells recruited in response to S and R variants ( Fig 1D ) . In intravenously ( iv ) infected embryos , eGFP expression was detected from the earliest hours post-infection ( S2A Fig ) , increased over time and peaked at 5 dpi , in an expanding bacterial-dependent manner , usually in close vicinity of the infection foci , particularly after R infection ( Fig 1E ) . Both Mabs R and S were found close to or within eGFP-positive cells near to the injection site at 1 dpi ( S2B Fig ) . At later time points , in both S- and R-infected embryos , a strong eGFP signal was detected in Mabs-containing granulomas ( Fig 1F and 1G ) but not around or within R-abscesses ( Fig 1G ) , the latter consisting essentially of highly replicating extracellular mycobacteria associated with tissue damage [17] . eGFP-positive cells were also surrounding Mabs serpentine cords ( Fig 1H ) . Because TNF-α can be produced by numerous cells [23] , we attempted to identify the TNF-producing cells in response to Mabs , with the double transgenic Tg ( tnfα:eGFP-F/mpeg1:mCherry-F ) or Tg ( tnfα:eGFP-F/LysC:DsRed ) embryos . It was possible to visualize green tnf-α expression concomitantly with red macrophages or red neutrophils , respectively . For both R and S infections , cells producing TNF-α co-localized with Mabs-containing macrophages , either isolated or within granulomas ( S2B Fig and Fig 1I ) . In sharp contrast , the eGFP signal failed to co-localize with neutrophils ( Fig 1J and S2C Fig ) . The total lack of TNF-positive cells in macrophage-depleted Tg ( tnfα:eGFP-F ) larvae ( S2D Fig ) , generated after lipo-clodronate injection [17] , confirmed macrophages as the primary source of TNF in response to Mabs infection . Taken collectively , these results indicate that TNF is principally produced by macrophages following infection with both Mabs variants , from very early phagocytosis after infection to later time points when the characteristic granulomas have appeared . TNF-α is a multifunctional cytokine playing a pivotal role in the regulation of inflammation and infection via the stimulation and engagement of the specific cell surface receptor 1 ( TNFR1 , ZDB-GENE-040426-2252 ) . To address the role of TNF signaling in Mabs infections , loss-of-function experiments for TNFR1 were carried out using a specific morpholino , leading to complete abrogation of the native tnfr1 mRNA ( S3A Fig ) and thereby subsequent TNF production ( S3B–S3D Fig ) . Although showing morphological defects ( S3A Fig ) , uninfected tnfr1 ( tnfr ) morphants exhibited survival rates similar to those of wild type ( WT ) embryos ( Fig 2A ) . Importantly , TNFR1 impairment led to an increase in the severity of the infection and hyper-susceptibility to R and S variants ( Fig 2A ) . This correlated with higher bacterial burdens as demonstrated by whole embryo imaging ( Fig 2B ) and fluorescent pixel counts ( FPC ) ( Fig 2C ) . Imaging of the R- and S-infected tnfr morphants showed that the increased bacteremia coincided with the presence of highly-replicating extracellular bacteria , resulting in the rapid development of abscesses ( Fig 2B and 2D ) . Whereas abscesses remain the exclusive attribute of R infections in WT embryos , 60% of S-infected tnfr morphants developed abscesses ( Fig 2D ) . Similarly , after R-infection , rapid and massive cord formation occurred in nearly all tnfr morphants within 1 dpi ( Fig 2E and 2F ) . At 1 dpi , WT embryos had fewer cords ( <5 ) while 70% of the morphants had high cord numbers ( >5 ) ( Fig 2G ) . Whereas cords developed essentially within the CNS in WT fish , all tnfr morphants exhibited widespread cording in the vasculature , in addition to the CNS ( Fig 2F and 2H ) . Thus , hyper-cording of Mabs R occurs very rapidly in the absence of TNF-mediated immunity , leading to early larval death , as has been described for Mycobacterium marinum in embryos lacking TNF signaling [24] . These results demonstrate the crucial and protective role of the TNFR1-dependent pathway in response to Mabs S and R by restricting extracellular multiplication and pathogenesis . To define how TNF orchestrates the events leading to the Mabs pathophysiology , we addressed whether the increased mortality and unrestricted mycobacterial growth in tnfr morphants are linked to a defect in macrophage recruitment , phagocytosis and/or bactericidal activity . The ability of macrophages to traffic across the epithelial and endothelial barriers was evaluated following S and R injection into the muscle ( S4A Fig ) and the otic cavity ( S4B and S4C Fig ) of Tg ( mpeg1:mCherry-F ) larvae . Irrespective of the infection site , the number of early recruited macrophages is comparable in both tnfr morphants and WT embryos at 2 hpi , implying that TNF signaling is not required for the early trafficking of macrophages , consistent with previous studies with M . marinum [24] . Although bacteria were rapidly engulfed by macrophages after infection of tnfr morphants ( S4A Fig ) , the number of infected macrophages harboring either R or S bacteria was lower in morphants than in WT embryos at 4 hpi ( Fig 3A ) . Knocking-down TNFR1 expression altered TNF expression ( S3B–S3D Fig ) and a defective TNF-positive feedback loop seems to be required for the efficient recruitment of macrophages at later stages ( S4B Fig ) and subsequent phagocytosis . The ability of tnfr morphants to develop abscesses in the CNS at later time points after intravenous infection suggests that macrophages remain efficient in transporting and disseminating the bacteria from the bloodstream to deeper tissues ( Fig 2 ) . To assess the contribution of TNF signaling in modulating the mycobactericidal activity of macrophages , the number of intracellular bacteria in individual infected macrophages was determined . The proportion of slightly infected ( <5 bacteria ) , moderately infected ( 5–10 bacteria ) or heavily infected ( >10 bacteria ) macrophages was enumerated at 1 dpi ( Fig 3B ) . Compared to the WT embryos , the tnfr morphants displayed a greater percentage of macrophages in the high burden category . This was true for both R and S variants and suggests that TNF restricts intracellular growth by stimulating the macrophage bactericidal activity . Further confirmation was obtained following staining of the infected embryos with a probe that detects reactive oxygen species ( ROS ) ( CellROX ) ( Fig 3C ) . At 1 dpi , ROS-labeled infected macrophages were found in WT embryos ( Fig 3D ) , in agreement with previous reports showing that macrophages can produce ROS to control Mabs infections [25 , 26] . No differences in the proportion of ROS-positive macrophages containing either Mabs S or R were noticed ( Fig 3D ) . However , fewer ROS-positive infected macrophages were found in morphants as compared to WT embryos ( Fig 3D ) , and ROS-positive macrophages in granulomas were only seen in WT animals ( Fig 3E ) . Aggregating cells positive for CellROX staining were occasionally observed , but only in WT larvae ( S5A and S5B Fig ) . However , the relatively low numbers of ROS-labeled infected macrophages , reflecting a reduced bactericidal activity in tnfr morphants , is unlikely to explain the extreme susceptibility of the tnfr morphants to Mabs and the very high extracellular bacterial loads . Because macrophage death has been previously shown to release extracellular Mabs , we examined the extent of macrophage death in infected larvae [17] . Imaging of the acridine orange ( AO ) -infected larvae ( S6A Fig ) and quantitative determination of AO-positive macrophages ( S6B Fig ) showed the significantly higher numbers of dead infected-macrophages in the tnfr morphants at 2 dpi as compared to the WT embryos . As expected , the basal levels of dead macrophages were very low in PBS-injected WT embryos or uninfected tnfr morphants ( S6A Fig ) . Overall , these results indicate that TNF signaling is pivotal in establishing the initial innate immune response by: i ) triggering the early bactericidal activity in macrophages and granulomas to restrict intracellular growth; ii ) reducing uncontrolled extracellular bacterial growth by preventing macrophage death; and iii ) promoting the formation of inflammatory cell aggregates which , in turn , may contribute to amplifying the local inflammation and recruitment of other immune cells [27] . Neutrophils are the first line of defense against pathogenic microorganisms and are rapidly recruited to infection sites where they engulf microorganisms and excrete their granule contents [28] . Mabs-containing neutrophils were previously identified in granulomas [17] but how these cells are recruited and contribute the immunity against Mabs remains unknown . Time-lapse microscopy of neutrophil mobilization in the caudal vein ( Fig 4A ) or in the muscle ( Fig 4B and S1 Movie ) of Tg ( mpx:eGFP ) embryos revealed a massive influx of neutrophils at the infection site starting at 10–20 min post-infection ( mpi ) . Isolated Mabs were rapidly engulfed by neutrophils ( Fig 4C and 4D ) , however the number of neutrophils harboring either R or S bacteria remained lower than the number of infected macrophages at 4 hpi ( Fig 4E ) . Time-lapse microscopy showed a massive mobilization of neutrophils within the deeper CNS infection foci ( S7A Fig ) , especially around abscesses ( Fig 4F ) . Interestingly , the R-abscesses continued to expand , concomitant with a time-dependent disappearance of the neutrophils , reminiscent of the neutropenia ( S7A Fig ) , and high bacteremia occurs prior to larval death as reported in embryos infected with Shigella [29] . IL8 is a central chemokine in neutrophil mobilization from hematopoietic tissues to infection sites . qRT-PCR revealed up-regulation of il8 early after infection , coinciding with granuloma formation and peaking at 5dpi , with higher levels for R infection than for S infection ( Fig 5A ) . Injection of a morpholino targeting il8 expression [30] in WT embryos strongly inhibited early neutrophil mobilization into infected muscle , hindbrain or otic cavity ( Fig 5B ) while leaving the baseline number of neutrophils unchanged , as reported earlier [30] . At later stages of infection , no neutrophils were associated to the CNS abscesses of il8 morphants , a phenomenon unrelated to neutropenia ( Fig 5C ) . In sharp contrast , while infection of il8 morphants resulted in strongly impaired neutrophil recruitment in the muscle , the hindbrain and the otic cavity ( Fig 5B ) , mobilization ( S8A Fig ) and phagocytosis ( S8B Fig ) of macrophages were unaffected by the il8 morpholino injection . Importantly , IL8 ablation correlated with reduced larval survival ( Fig 5D ) and with increased S and R loads ( Fig 5E ) . The lack of neutrophil recruitment seen with IL8 ablation paralleled the pronounced increase in the bacterial loads , as evidenced by the numerous large abscesses ( Fig 5E ) , suggesting that the absence of neutrophils at the site of infection may be deleterious for the host . To test this hypothesis , specific neutrophil depletion was carried out though injection of csf3r morpholino [31] , which at the concentration used did not affect the macrophage recruitment ( S9A Fig ) or macrophage phagocytic activity ( S9B Fig ) . As seen in the il8 morphants , the csf3r morphants showed hyper-susceptibility to both R and S infections with 100% larval mortality at 6 dpi ( Fig 5F ) and massive extracellular bacteremia ( Fig 5G ) . Overall , these findings indicate that IL8 mediates neutrophil mobilization to the infection sites and plays a critical role in host defense against Mabs . TNF orchestrates the early regulation of chemokine induction , including IL8 , essential for neutrophil activation and recruitment to inflamed tissues [32] . To address the role of TNF in neutrophil mobilization , local infections were done in the otic ( Fig 6A and 6B ) and hindbrain ( Fig 6C and 6D ) cavities of Tg ( mpx:eGFP ) tnfr morphants . While neutrophils were rapidly recruited to the infected ear or hindbrain in WT animals , their recruitment was severely reduced in the tnfr morphant for both S and R variants ( Fig 6A–6D ) , supporting a major role of TNF in early neutrophil mobilization . Whilst confocal microscopy of the cord/abscess-containing environments demonstrated massive mobilization of neutrophils around cords ( Fig 6E and S2 Movie ) and abscesses ( Fig 6F and S4 Movie ) in WT embryos , this was not true in tnfr morphants ( Fig 6E and S3 Movie , Fig 6F and S5 Movie ) . The reduced number of neutrophils in tnfr morphants ( S10A and S10B Fig ) , was consistent with a study reporting the influence of TNF on hematopoietic stem cell formation [33] . To inquire whether the decreased neutrophil recruitment in tnfr morphants is linked to a possible alteration in a basal neutrophil number , we performed a neutrophil mobilization assay using fMLP , a synthetic neutrophil chemoattractant [34] . After injection of fMLP into the otic cavity of tnfr morphants , neutrophils were recruited to the injection site to the same extent as in the WT embryos ( S10C Fig ) . Comparable results were also obtained following injection of recombinant IL8 into the tnfr morphants ( S10D Fig ) . Thus , the decreased neutrophil recruitment in tnfr morphants ( Fig 6 ) is due neither to the general reduction of the neutrophilic population nor to nonspecific effects of the tnfr morpholino used . Together , these results further position TNF as a critical mediator in initiating the early and late phases of neutrophil recruitment and substantiate the crucial role of neutrophils in controlling Mabs infections . Despite the co-existence of macrophages and neutrophils in Mabs-induced granulomas [17] , the importance of neutrophils in the maintenance and/or integrity of this organized cellular structure remains elusive . Monitoring the kinetics of granuloma development revealed that granulomas induced by both R- and S-variants appeared at 2 dpi and expanded in most WT embryos at 5 dpi ( Fig 7A and 7B and [17] ) . In sharp contrast , the granuloma-like structures in the tnfr morphants appeared as poorly delimited loose cellular aggregates ( Fig 7A–7E and S11 Fig ) , similar to those previously documented in TNF defective M . marinum [24] , and highlighting the absolute requirement of a functional TNF pathway for granuloma formation . Confocal microscopy revealed a more open and disjointed structure with the presence of high numbers of extracellular bacterial aggregates ( Fig 7C ) , prompting us to examine whether the increased proportion of dead macrophages in infected tnfr morphants ( S6A and S6B Fig ) may contribute to the morphologically altered granulomas and the reduced granuloma numbers . As shown in S6C Fig , despite the higher proportion of dead macrophages in the tnfr morphants as compared to the WT embryos , there was no correlation with the number of “defective” granulomas found in the tnfr morphants . While granulomas in WT embryos contained numerous neutrophils , they were nearly absent in the corresponding structures in Tg ( mpx:eGFP ) tnfr morphants , which were characterized by substantial numbers of extracellular bacteria/cords ( Fig 7D , S6 and S7 Movies ) . Time-lapse monitoring ( Fig 7E ) and determination of the number of infected neutrophils recruited to the granulomas ( Fig 7F ) established a linear relationship between the number of recruited neutrophils and the granuloma volume in WT embryos . In contrast , tnfr morphants exhibited an important lack of neutrophils ( Fig 7E ) and no linear correlation with the size of the granuloma-like structures could be established ( Fig 7F ) , suggesting a direct participation of neutrophils in elaborating and shaping Mabs granulomas . That TNF modulates the early mobilization of neutrophils into granulomas led us to determine whether TNF depletion influences IL8 expression . Quantitative RT-PCR revealed that , while Mabs stimulated il8 expression levels in WT larvae , its expression was severely decreased in the TNFR1-depleted larvae ( Fig 8A ) and was even further reduced in macrophage-depleted larvae following injection of lipoclodronate ( Fig 8A ) , indicating that macrophages are part of the pathway that triggers IL8 release in response to Mabs infections . This suggests that in WT embryos , TNF production by macrophages ( Fig 1I ) governs IL8-driven chemotaxis , and thus the absence of TNF signaling profoundly restricts neutrophil mobility . Having confirmed that macrophages are required for IL8 production , we next examined whether neutrophil mobilization to the infection site is dependent on macrophages . Neutrophil recruitment was dramatically reduced in the macrophage-depleted larvae ( Fig 8B ) , demonstrating that macrophages are key players in neutrophil mobilization in response to Mabs infection . In addition , while neutrophil recruitment is strongly impaired in the absence of either macrophage or TNFR signaling , the injection of exogenous IL8 fully rescued neutrophil mobility ( Fig 8B and 8C ) and restored survival of tnfr morphants infected with R or S in the otic cavity ( Fig 8D ) . This correlated with decreased bacterial loads compared to untreated tnfr morphants , as evidence by the determination of the FPC ( Fig 8E ) and fluorescence microscopy ( Fig 8F ) . These findings highlight the immune-protective role of IL8-dependent neutrophil mobilization during Mabs infections . Furthermore , granuloma formation was severely impaired in il8 morphants and , consistent with these findings , granuloma formation was abrogated in the neutrophil-depleted csf3r morphants ( Fig 8G and 8H ) . Similarly to tnfr morphants ( Fig 7F ) , il8 morphants exhibited neutrophil-poor granulomas ( Fig 8I ) , further supporting the direct participation of neutrophils in elaborating and shaping Mabs granulomas . Overall , these results emphasize the absolute requirement of an IL8/TNF-dependent neutrophil mobilization for granuloma formation and control of Mabs infections .
Herein , we report the first stepwise dissection study of the immune control of Mabs using a non-invasive visualization approach with special emphasis on the inflammatory response . The spatiotemporal immunopathological events ( Fig 9 ) can be summarized as follows: i ) rapid engulfment of Mabs by macrophages; ii ) TNF release by activated macrophages , leading to ROS production and intracellular killing of Mabs , and IL8-driven chemotaxis that guides neutrophils to the infection site; iii ) proficient granulomatogenesis and development of chronic infections . In contrast , defective TNF signaling correlates with i ) impaired macrophage activation with increasing intramacrophage bacterial loads and disruption of IL8 production , resulting in impaired neutrophil recruitment; ii ) absence of bona fide granulomas . Additionally , the increased macrophage death releases free bacilli that multiply extracellularly in an uncontrolled manner , resulting in mycobacterial cords that prevent phagocytosis by macrophages and neutrophils [17] , acute infections and larval killing . Mabs infections are characterized by growth in highly inflamed tissues , suggesting a role for neutrophils in the host response . Supporting this hypothesis , patients with CF , a disease that is dominated by persistent neutrophil-mediated inflammation , are particularly susceptible to Mabs in addition to other extracellular pathogens [4 , 35] . However , despite the prevalence of neutrophils in Mabs infections , information regarding the neutrophil response is scarce . Previous work suggested that human neutrophils mediate killing of Mabs , but phagocytosis was reduced when compared to Staphylococcus aureus , another important CF pathogen [36] . In experimental models of Mabs infection , the presence of increased numbers of neutrophils is associated with a worse response to Mabs [37] . We assessed here the in vivo capacity of neutrophils to migrate in response to Mabs , to engulf the bacilli and to participate directly to granuloma formation in infected zebrafish . Our results are distinct from those with Mycobacterium tuberculosis or M . marinum where neutrophils appear to be less mobilized [34 , 38] . Ex vivo infections of human lung tissues indicated that neutrophils had a greater tendency to phagocytize Mabs than M . tuberculosis and that Mabs has a higher capacity to induce the migration of neutrophils than other mycobacterial species [38] . Additionally , while neutrophils are unable to kill virulent strains of M . tuberculosis or M . marinum [39 , 40] , they appear to be important for controlling virulent and avirulent Mabs because neutrophil-depleted embryos are extremely susceptible to Mabs infection . The mobilization of neutrophils toward the large cords and abscesses is also in agreement with recent studies reporting that neutrophils , through a microbial size-sensing mechanism , tailor their antimicrobial responses to pathogens based on microbial size [41] . Recent studies have indicated that M . tuberculosis-infected neutrophils can be considered as biomarkers for poorly controlled mycobacterial replication and are associated with severity in human tuberculosis [42] . Our data show that infected neutrophil-depleted embryos exhibit increased CNS pathology , high mycobacterial loads , decreased survival rates and reduced granuloma numbers . They also support the crucial role of IL8 in the control of Mabs at the infection site due to its function as the main mediator of neutrophil mobilization to the infection site , and through this recruitment of neutrophils , the formation of granulomas . Indeed , the absence of granulomas caused by il8 or tnfr knock-down was associated with compromised survival and reduced bacterial clearance of Mabs , phenotypes that were restored upon injection of recombinant IL8 . Despite the important role of macrophages in IL8 signaling , production of IL8 was not completely abrogated in the macrophage-depleted embryos , suggesting that other cell types may also contribute to IL8 production . Our results thus shed new light on the role of neutrophils in early granuloma formation , integrity and maintenance , and reveals striking differences with the dynamics of granuloma formation in the M . marinum zebrafish model , where granulomas contribute to early bacterial growth and expansion of the infection [43] . Although one cannot exclude the possibility that Mabs exploits the granuloma to manipulate host immune responses for its own benefit as suggested for M . marinum , our conclusions support the primary role of granulomas in preventing a widespread expansion of Mabs in the extracellular milieu . Granuloma-defective embryos ( tnfr , il8 or csf3r morphants ) were all hyper-susceptible to S and R infections with pronounced larval mortality and unrestricted extracellular bacterial growth . Both R and S strains simulated granuloma formation at comparable levels [17] , indicating that granuloma formation is not correlated with virulence of Mabs . In contrast , granuloma formation with M . marinum is linked to virulence , as demonstrated using the attenuated RD1-deficient mutant [44] . Apparently , the distinctive kinetics and functions of granulomas are species-specific and conclusions drawn from one mycobacterial species cannot be extrapolated to others . Furthermore , while M . marinum-infected embryos deficient in TNF signaling showed increased granuloma formation in early stages of infection [24] , after Mabs infection of tnfr morphants the granuloma formation was almost completely abrogated . Instead of compact organized granuloma , Mabs infection of tnfr morphants elicited only the formation of disorganized cellular structures that consisted of aggregated macrophages . A similar result was seen in TNF-α KO mice where impaired Mabs control was associated with profound alteration of granuloma formation [16] . Why would TNF exert distinct responses when stimulated with different mycobacterial species ? It is well known that the mycobacterial cell envelope possesses a large panel of surface-exposed glycolipids , some of which are particularly granulomatogenic . Although the overall architecture of the cell wall is conserved among mycobacteria , each species can be typified by a specific lipid/glycolipid signature , and subtle variations in the lipid composition , structure and size can affect granulomatogenesis . Other effectors , such as the ESX-1 are also required for efficient granuloma formation with M . tuberculosis and M . marinum [44 , 45] . While the loss of RD1 in a M . marinum is associated with altered early aggregation of infected macrophages and a delay in granuloma formation [44] , Mabs , which naturally lacks RD1 , induces “normal” granuloma formation in embryos [17] , adult zebrafish [46] and in mice [16] . M . marinum mutants with reduced granuloma formation were also found to be defective in EspL , located in the ESX-1 cluster [47] . Mabs presents two ESX-like loci [48] but whether these clusters participate in Mabs-induced granulomas and pathogenesis awaits further investigation . The identification of the Mabs-specific effectors participating in granuloma formation would contribute to our knowledge of Mabs pathogenicity and could foster the development of needed therapeutic interventions , as Mabs is refractory to most antimicrobials . In conclusion , we report here new and unexpected insights into several aspects of Mabs immunopathogenesis by demonstrating the crucial role of TNF signaling in a continuum of effects starting from limiting intracellular bacterial multiplication to the induction of granulomas that together exert a protective effect by limiting Mabs dissemination . Through its orchestration of the inflammatory cytokine/chemokine network , dominated by IL8 production , TNF modulates the engagement of neutrophils to the infection site and their subsequent recruitment to granulomas , which are essential for control of the early and later stages of Mabs infection , respectively . This explains why immunosuppressive TNF therapy increases the risk of Mabs infections [49] . Airways of CF patients are characterized by a severe inflammation [50] but whether this directly impacts on subsequent Mabs colonization is difficult to predict . The airways of these patients are chronically colonized with complex , polymicrobial infections [51] and these polymicrobial communities promote intricate inter-microbial and host-pathogen interactions which alter the lung environment , impact the response to treatment , and drive the course of the disease . In addition , the relationship between abnormal CFTR expression and the predisposition of CF patients to chronic Mabs infections remains elusive , but the results presented here suggest that it would be interesting to determine whether the CFTR defect has a detrimental effect on neutrophil-mediated immunity .
Zebrafish experiments were conducted in accordance with the guidelines from the European Union for handling of laboratory animals ( http://ec . europa . eu/environment/chemicals/lab_animals/home_en . htm ) and approved by the Direction Sanitaire et Vétérinaire de l'Hérault et Comité d'Ethique pour l'Expérimentation Animale de la région Languedoc Roussillon under the reference CEEA-LR-1145 . For zebrafish anaesthesia procedures , embryos are immersed in a 270 mg/L Tricaine solution in osmotic water . When required , larvae were cryo-anesthetized by incubation on ice for 10 min and then euthanized using an overdose of Tricaine ( 500 mg/L ) . Experimental procedures were performed using the golden zebrafish mutant [52] , along with transgenic lines: Tg ( mpx:eGFP ) i114 [53] or Tg ( LysC:DsRed ) nz5 [54] , harboring either green- or red-fluorescent neutrophils , respectively; Tg ( mpeg1:mCherry-F ) ump2 [17] , harboring red-fluorescent macrophages; and Tg ( tnfα:eGFP-F ) ump5 [22] , which allows visualization in green of the transcriptomic expression of tnf-α under appropriate stimulatory conditions . R and S variants of M . abscessus sensu stricto strain CIP104536T ( ATCC19977T ) carrying pTEC15 ( Addgene , plasmid 30174 ) , pTEC27 ( Addgene , plasmid 30182 ) or pTEC19 ( Addgene , plasmid 30178 ) that express green fluorescent protein ( Wasabi ) , red fluorescent protein ( tdTomato ) or bright far-red fluorescent protein ( E2-Crimson ) , respectively , were prepared and microinjected in zebrafish embryos , according to procedures described earlier [17 , 55] . Briefly , systemic infections were carried by the injection of 100–200 CFU into the caudal vein of 30 hours post-fertilization ( hpf ) embryos . For phagocyte recruitment assays , 100 CFU were injected locally into the otic vesicle , the muscle or the hindbrain compartments of 3 days post-fertilization ( dpf ) larvae or into the caudal vein of 2 dpf embryos . Neutrophil recruitment was elicited in zebrafish embryos through injection of the f-Met-Leu-Phe ( fMLP , Sigma-Aldrich ) or recombinant human IL-8 ( rhIL-8 , R&D Systems , Inc . ) , chemoattractants previously described [34] . IL-8 ( 15–25 pg ) or 3 nl of 300 nM fMLP were injected into the otic cavity of 3 dpf larvae . The quantity of recruited neutrophils was determined at the injection site at 3 hpi using fluorescence microscopy . Morpholinos purchased from Gene Tools were injected into 1–4 cell stage embryos . tnfrsf1a splice-blocking morpholinos targeting TNF receptor 1 ( 5’-GGAAGCATGAGGAACTTACAGTTCT-3’ ) were used at a concentration of 0 . 5 mM . The efficiency of gene knockdown was confirmed by RT-PCR with the following primers for both sides of the morpholino target sequence: tnfrsf1a , CCCGCATGCTCCACGTCTCC and TTATAGCGGCCGCCCGACTCTCAAGCTTCA . The zcxcl8 splice blocking morpholino for IL8 knock-down ( 5’-TATTTATGCTTACTTGACAATGATC-3’ ) was prepared and injected as described earlier [30] . To generate neutrophil depleted-embryos , csf3r translation morpholino ( 5’-GAAGCACAAGCGAGACGGATGCCAT-3’ ) targeting the csf3r gene was used [31] . For the selective depletion of macrophages into embryos , lipo-clodronate ( Lipo-C ) [56] was injected into the caudal vein of 24 hpf embryos as previously reported [17 , 55] . Dead cells in living zebrafish embryos were detected using Acridine Orange ( AO ) , as described [17] . For detecting ROS , living embryos were soaked in 5 μM CellROX Green Reagent ( Invitrogen ) in Hanks’s Balanced Salt Solution ( HBSS ) for 30 min at 28 . 5°C , followed by two washes in HBSS , then transferred into a dish for fluorescent microscopic observation and analyses . To determine cytokine/chemokine expression levels , total RNA from a pool of 10–15 larvae per biological experiment was isolated using the Nucleospin RNAII kit ( Macherey-Nagel ) . cDNA synthesis was performed with M-MLV reverse transcriptase ( Invitrogen ) and then quantitative RT-PCR was performed using a homemade SYBR Green mix on a LightCycler 480 instrument ( Roche ) as described [57] with the following pairs of primers ( sense and antisense ) : ef1α , TTCTGTTACCTGGCAAAGGG and TTCAGTTTGTCCAACACCCA; tnfα , TTCACGCTCCATAAGACCCA and CCGTAGGATTCAGAAAAGCG; il8 , CCTGGCATTTCTGACCATCAT and GATCTCCTGTCCAGTTGTCAT; ifnγ2 , TGCACACCCCATCTTCCTGCGAA and GTGTTGCTTCTCTATAGACACGCTT; il1β , TGGACTTCGCAGCACAAAATG and GTTCACTTCACGCTCTTGGATG . Each experiment was run in triplicate . qRT-PCR datas were calculated using the ΔCt or ΔΔCt method and normalized to the housekeeping gene ef1α . To quantify bacterial loads , granulomas , cords and neutrophil recruitment , infected larvae were tricaine-anesthetized , positioned on 35-mm dishes , immobilized in 1% low-melting-point agarose and covered with water containing tricaine . Bright-field and fluorescence pictures of live infected embryos were taken with a Zeiss microscope equipped with a Zeiss Plan Neo Fluor Z 1x/0 . 25 FWD objective and a Axiocam503 monochrome ( Zeiss ) camera , with acquisition and processing using ZEN 2 ( blue edition ) . Evaluation of intracellular mycobacterial growth , enumeration of macrophages recruitment , phagocytosis , mortality and granuloma organization , infected embryo were prepared for fixed microscopy . Animals were tricaine-anesthetized and fixed overnight at 4°C in 4% ( vol/vol ) paraformaldehyde in PBS , then washed twice in PBS , and transferred gradually from PBS to 50% ( vol/vol ) glycerol for microscopic observation . Confocal microscopy was performed using a Leica SPE upright microscope with 40x ACS APO 1 . 15 oil objective . Images were captured by LAS-AF software ( Leica Microsystems ) . Overlays of fluorescent and bright-field images and 2D reconstructions of images stacks were produced , assembled and adjusted using LAS-AF software or GIMP 2 . 6 freeware . Three-dimensional volume reconstitutions and movies were performed using Imaris 7 . 0 software ( Bitplan AG ) . Statistical analyses were performed using Prism 4 . 0 ( Graphpad , Inc ) or R 3 . 0 . 3 and detailed in each figure legend . *p< 0 . 05; **p< 0 . 01; ***p<0 . 001; ****p< 0 . 0001; ns , not significant ( p≥ 0 . 05 ) . | The incidence of non-tuberculous mycobacterial infections has recently increased and has even surpassed tuberculosis as a public health concern in many developed countries . These infections require long treatment regimens that are often unsuccessful . Among these , Mycobacterium abscessus has emerged as perhaps the most difficult-to-manage pathogen , especially in cystic fibrosis patients . Unfortunately , very little is known regarding the contributions of the pro-inflammatory and innate immune responses during M . abscessus infection . Here , we exploited the transparency of zebrafish embryos to study , at high resolution , the interactions of M . abscessus with macrophages and neutrophils , and found that both cell types are required to control the infection . We also describe the dramatic consequences of impaired TNF/IL8 immunity on the outcome of the infection . Most importantly , by tracking the dynamics of neutrophil mobilization , we demonstrated the crucial role of these cells in the formation and integrity of protective granulomas . Together , our data provide a significant advance in deciphering the immunopathology of M . abscessus infection , which is particularly relevant for understanding the exquisite vulnerability of cystic fibrosis patients to this bacterium . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"blood",
"cells",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"granulomas",
"immunology",
"vertebrates",
"light",
"microscopy",
"animals",
"animal",
"models",
"osteichthyes",
"developmental",
"biology",
"model",
"organisms",
"microscopy",
"confocal",
"microscopy",
"embryos",
"neutrophils",
"research",
"and",
"analysis",
"methods",
"embryology",
"white",
"blood",
"cells",
"fishes",
"animal",
"cells",
"fluorescence",
"microscopy",
"life",
"cycles",
"zebrafish",
"cell",
"biology",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"macrophages",
"larvae",
"organisms"
] | 2016 | Mycobacterium abscessus-Induced Granuloma Formation Is Strictly Dependent on TNF Signaling and Neutrophil Trafficking |
Plasticity in the oculomotor system ensures that saccadic eye movements reliably meet their visual goals—to bring regions of interest into foveal , high-acuity vision . Here , we present a comprehensive description of sensorimotor learning in saccades . We induced continuous adaptation of saccade amplitudes using a double-step paradigm , in which participants saccade to a peripheral target stimulus , which then undergoes a surreptitious , intra-saccadic shift ( ISS ) as the eyes are in flight . In our experiments , the ISS followed a systematic variation , increasing or decreasing from one saccade to the next as a sinusoidal function of the trial number . Over a large range of frequencies , we confirm that adaptation gain shows ( 1 ) a periodic response , reflecting the frequency of the ISS with a delay of a number of trials , and ( 2 ) a simultaneous drift towards lower saccade gains . We then show that state-space-based linear time-invariant systems ( LTIS ) represent suitable generative models for this evolution of saccade gain over time . This state-equation algorithm computes the prediction of an internal ( or hidden state- ) variable by learning from recent feedback errors , and it can be compared to experimentally observed adaptation gain . The algorithm also includes a forgetting rate that quantifies per-trial leaks in the adaptation gain , as well as a systematic , non-error-based bias . Finally , we study how the parameters of the generative models depend on features of the ISS . Driven by a sinusoidal disturbance , the state-equation admits an exact analytical solution that expresses the parameters of the phenomenological description as functions of those of the generative model . Together with statistical model selection criteria , we use these correspondences to characterize and refine the structure of compatible state-equation models . We discuss the relation of these findings to established results and suggest that they may guide further design of experimental research across domains of sensorimotor adaptation .
The accuracy of saccadic eye movement is maintained through mechanisms of saccade adaptation , which adjust the amplitude [1–3] or direction [4–6] of subsequent movements in response to targeting errors . As online visual feedback cannot be used to correct the ongoing movement , saccadic eye movements need to be preprogrammed and adaptation must largely rely on past experience and active predictions [7 , 8] rather than closed-loop sensory information . To induce saccade adaptation in the laboratory [1] , participants are instructed to follow a step of a target stimulus with their eyes and this visual cue is then displaced further during the saccade eye movement . Typically , this second , intra-saccadic step ( ISS ) is constant across trials and directed along the initial target vector towards smaller or larger saccade amplitudes . Although the ISS is visually imperceptible [9] , saccades adjust their amplitude to compensate for the induced error . In phenomenological analyses of such saccade adaptation data , the amount of adaptation is usually quantified by comparing saccade gain values before and after the adapting block and interpolating an exponential fit in between [1–3 , 10] . We recently presented a version of this paradigm in which the ISS ( the disturbance responsible for inducing adaptation ) follows a sinusoidal variation as a function of trial number ( [11 , 12]; see also [4 , 13 , 14] ) . We reported that gain changes were well described by a parametric functional form consisting of two additive components . One component was a periodic response reflecting the frequency of the ISS that was adequately fitted with a lagged but otherwise undistorted sinusoid . The second component constituted a drift of the baseline toward lower saccade gain ( larger hypometria ) that was appropriately accounted for using an exponential dependence . Here , we investigate whether a generative algorithm that models saccade gain modifications on a trial-by-trial basis by learning from errors made on previous trials can account for this response . To this end , we implemented and fit a series of state-space models in which a modified delta-rule algorithm updates a hidden or latent variable ( for which the experimentally observed adaptation gain is a proxy ) by weighting the last experienced visual error , in addition to other error-based and non-error based learning components [8 , 15–23] . We adopt the approach that these algorithms are linear time-invariant systems ( LTIS ) , in that their coefficients are time and trial-independent . LTIS models , also known as linear dynamical systems ( LDS ) have been successfully used in a number of motor adaptation studies [8 , 19–22 , 24–27] . Applied to saccade adaptation , they may predict the dynamics of the saccade amplitude itself as well as various forms of movement gain typically used in describing adaptation [2 , 3 , 10 , 11] . Our first goal was to establish empirically whether LTIS models could fit the data recorded with a sinusoidal adaptation paradigm , as efficiently as when using a constant ( fixed ) ISS . Once we have established this point , we will explore the relation between the predicted phenomenological parameters [11 , 12] and the learning parameters of the underlying generative model , as well as their potential dependence on the perturbation dynamics . We first analyze the ability of a family of generative models to describe experimental recordings of saccade adaptation by fitting the relevant learning parameters . We then perform statistical model-selection analysis to determine those that best fitted the same data in the various experimental conditions . We fitted models to two data sets , a previously published one [11] and a variation of that paradigm that extended the range of frequencies of the sinusoidal variation of the ISS . Both data sets contrasted two established saccadic adaptation protocols [11 , 12]: Two-way adaptation ( i . e . , bidirectional adaptation along the saccade vector of saccades executed along the horizontal meridian ) and Global adaptation ( i . e . , adaptation along the saccade vector of saccades executed in random directions ) . We then explore consequences for current models of motor learning and suggest possible modifications that may be required to generate a suitable description of sensorimotor learning during sinusoidal saccadic adaptation . In conducting this selection , we confirm that a single learning parameter model ( a state-equation with just an error-based learning term; cf . [19] ) does not suffice to fit the data . We then demonstrate that including an extra term that weights the next-to-last trial’s error provides a better fit for the Two-way type of adaptation . This learning rate has the intriguing feature that it has negative values for all frequencies , suggesting an active unlearning of the next-to-last trial’s feedback error , close , but not equal in magnitude to the learning rate of the last trial’s error . We discuss possible functional roles of these processes for oculomotor adaptation in natural situations , where saccadic accuracy is expected to exhibit slow dynamic changes across time .
The Ethics Committee of the German Society for Psychology ( DGPs ) approved our protocols . We obtained written informed consent from all participants prior to the inclusion in the study . The present study conformed to the Declaration of Helsinki ( 2008 ) . We re-analyzed the data we recently collected using a fast-paced saccade adaptation paradigm with a sinusoidal disturbance . We had previously described these data by fitting a phenomenological model that we identified using statistical model selection . For details on the experimental procedures pertaining to this original data set ( henceforth , ORIG ) and to the selection of the functional form of this phenomenological model , please refer to our former communication [11] . We applied the same experimental procedure in collecting further data with an enhanced range of frequencies . In this case , thirteen participants ran two sessions with similar Two-way and Global adaptation protocols as used in previous reports [11 , 12] . In short , Two-way adaptation refers to bidirectional adaptation along the saccade vector of saccades executed along the horizontal meridian . In turn , Global adaptation refers to adaptation along the saccade vector of saccades executed in random directions . In collecting this dataset ( henceforth , FREQ ) , each session had 2370 trials divided in 11 blocks . Odd numbered blocks had 75 no-adaptation trials ( zero ISS ) . The five even-numbered blocks consisted of 384 trials each with a sinusoidal disturbance similar to that used before but with frequencies of 1 , 3 , 6 , 12 and 24 cycles per block ( i . e . , 384 , 128 , 64 , 32 , and 16 saccades per cycle , respectively ) . The order of adaptation blocks was randomly interleaved for each observer and type of adaptation . The program was paused after each adaptation block , giving participants some resting time , and we calibrated eye position routinely at the beginning of each non-adapting ( odd-numbered ) block . In each trial , the pre-saccadic target step was fixed at 8 degrees of visual angle ( dva ) . The subsequent second step ( ISS ) then ranged between –25% and +25% of the first step , changing size according to a sine function of trial number . To investigate generative models , we adopt the following rationale . In each trial , the oculomotor system must generate a motor command to produce the impending saccade . This needs to be calibrated against the actual physical size of the required movement [15 , 20 , 22 , 24 , 30 , 31] . If the saccade fails to land on target , the motor command needs to be recalibrated based on preexisting calibrations , and we will hypothesize that those changes take place in an obligatory manner ( cf . [19] ) through additive , error-based modifications attempting to ameliorate post-saccadic mismatches between the eyes’ landing position and target location . We model the underlying sensorimotor learning using linear time-invariant systems ( LTIS ) . The model parameters ( or the learning coefficients ) are time independent in each experimental block , although they can vary across experimental conditions or phases [32] . These models are closely related to linear dynamical systems ( LDS; cf . [20–22] ) , except that here we only address noise-free models . Because saccades are extremely rapid movements that do not admit reprogramming in mid-flight , it is assumed that all gain changes take place in between saccades . In our models , therefore , the error-based correction terms weight errors that were experienced in previous saccades . As a consequence , in the estimation of the forthcoming event , the post-saccadic stimulus gain is not compared against the adaptation gain measured for that trial but against the previous estimate of the gain . To justify these assumptions , it is usually assumed that the motor system sends an efference copy of the motor command to the sensory areas , which enables prediction of the sensory consequences of the movement and therefore avails comparison to experienced post-saccadic feedback [7 , 19 , 20 , 31 , 33–35] . We will assume that the values of saccade and adaptation gains observed and extracted from the recorded data ( i . e . , SG ( n ) =SA ( n ) preTP ( n ) and g ( n ) =SA ( n ) -preTP ( n ) ∥ISS∥ ) are adequate proxies of that motor calibration process . Yet the calibration itself is an internal feature of the brain and therefore the adaptation gain that enters the generative algorithm ( the state-equation ) that we intend to study is a hidden variable representing the internal state of the system . A model providing its temporal evolution can then be fitted to the data; yet the variable itself is not experimentally accessible . We denote the internal variable associated to the saccade gain by z ( n ) . To describe the evolution of this state variable we introduce the state-equation: z ( n+1 ) =A∙z ( n ) +K∙ ( t ( n ) -z ( n ) ) +M+D∙ ( t ( n-1 ) -z ( n-1 ) ) , ( 4 ) supplemented with an initial condition that sets the initial value z ( 1 ) = G · t ( 1 ) . Here , the target gain t ( n ) is available from recordings in each trial and we shall assess how well the prediction of the saccade gain ( z ( n + 1 ) , provided by Eq 4 ) fits the recorded data SG ( n ) . The first term on the RHS of the equation is a persistence term . The persistence rate A determines how much of the estimate of the state variable at trial n is transferred to the estimate at the next trial [8 , 25 , 36] . Therefore , its magnitude is expected to be typically slightly smaller than 1 and it is set to be 1 in the models that do not include its effect . The second term weights the discrepancy between the gain of the target at trial n and the predicted gain of the movement under the underlying assumption that the size of the state variable is an adequate proxy for the ( sensory ) consequences of the movement . The weighting coefficient K is called learning rate . M embodies any systematic effect ( drift or bias ) that takes place in each trial but it is not directly determined by the sensory feedback [37]; we shall call it a drift parameter . The last term is a second error-based correction term that weights the discrepancy between the gain of the target and the estimate of the movement at a trial other than the last error with an additional ( distal ) learning rate D . For concreteness we shall assume that this correction is based on the sensory feedback arising from the next-to-last trial . However , we shall return to this specific assumption further in the Discussion . Note that with the inclusion of this hypothetical double error sampling the full model of Eq 4 ( and Eq 5 below ) becomes an algorithm that coherently uses two delayed feedbacks to estimate the state of a single internal variable that models the sensory consequences of the intended motion . We conducted our analyses using the full form of Eq 5 . We were interested in determining which model suffices to account for the data with the least number of parameters . The magnitude being learned is x , the internal representation of the adaptation gain of the imminent saccade . This gain has value zero upon the ideal outcome of perfect movement accuracy and in that respect , it can be interpreted as the gain of an internal prediction error . Using Eq 5 , we generated the predicted values of x ( n ) in each condition and for each participant , and then fitted a number of models that differed from each other in which parameters were estimated . When a parameter among K , m , or D was not present , the corresponding term was removed from Eq 5 . Note however , that when the parameter A was not included as a fitting parameter , its value was set to unity ( i . e . , A = 1 ) . In the case of the initial value G , we obtained an estimate by taking the average of the first five values of the gain . We proceeded in this way because the initial value of the state of the system is unknown and , while the first recorded value of the gain could be considered a proxy for such initial state , execution and motor errors could yield a value of the gain significantly different than the actual initial state of the system; we averaged over 5 trials to alleviate this problem . In models where the initial value of the gain was left free to become a fitting parameter , this average over the first five saccades was used as an initial value for the fitting routine for that particular parameter . Improvements can be achieved by letting the initial condition become an extra parameter . We discuss below the interpretation of using the initial condition as a fitting parameter of the model . In view of these features of the generative model , a natural classification of the models tested arises as follows: given the parameters K , A , m , D1 , ⋯ , Dw , G , we will 1 ) include K in every model because we are modeling intrinsic learning where we assume that learning from the last experienced feedback is always present as well as obligatory [19 , 20 , 22 , 31]; 2 ) models will be generated by adding successively the parameters A , m , and D , of which one or more could be present but in this study we restrict ourselves to learning possibly from only one extra feedback in the past; 3 ) G is an optional parameter that is included in an attempt to alleviate extreme effects of the initial condition ( s ) as explained above . By applying points 1 ) through 3 ) , sixteen different models can be generated . For reasons to become clear below we would group them in four families according to whether or not they contain the bias term ( m ) and the additional error term ( with learning rate D ) : K only ( although with A = 1 when omitted ) , KA , KG , KAG feature zero bias and a single error term; Km , KAm , KmG , KAmG are models with a single error term that allow bias; KD , KAD , KDG , KADG have no bias term but sample two errors , and KmD , KAmD , KmDG , KAmDG feature both a bias term and learn based on double error sampling . Therefore , the simplest model had a single fitting parameter ( the learning rate K , cf . [19] ) and was obtained by setting A = 1 , removing the terms that involved m and D , and setting the initial value G to be the mean of the first five values of the gain in the block . The full model had all five as fitting parameters . All parameters of the generative models were estimated by fitting the model to the experimental data using MATLAB function nlinfit; 95% confidence intervals for the fitted parameters were computed using MATLAB function nlparci and predicted responses for the hidden variable x with its corresponding 95% confidence intervals were obtained from MATLAB function nlpredci . All 16 models were fitted to data from each individual participant parameters were extracted for each model , and models were compared using the Akaike information criterion ( AIC; [38–41] ) by computing Akaike weights across models for each participant . Finally , these weights were averaged across participants for each model in each condition . The adaptation gain of the oculomotor response to a sinusoidal disturbance is best described by a phenomenological function consisting of a decaying exponential added to a lagged but otherwise undistorted sinusoid [11] . The sinusoidal component of the response onsets at the beginning of the adaptation block but all fittings include the pre-adaptation block as well . The frequency of the stimulus disturbance is matched closely by the gain . To fully describe the response , five extra phenomenological parameters are required: amplitude ( a ) and lag ( ϕ ) of the periodic part of the error gain complete the description of the periodic part . The exponential decaying component that describes the baseline on which the periodic response rides requires other three: an asymptotic value ( B0 ) where the baseline stabilizes at large trial number , a timescale ( λ ) in which the baseline reaches 1/e of the full decay , and the amplitude of the decay ( B ) : g ( n ) =a∙sin ( ωn-ϕ ) +Be-λn+B0 , withω=2πνN . ( 6 ) We use here the same denominations used in our previous report [11] , except for changing the name of the timescale to λ to prevent confusion with the amplitude of the periodic component a . To estimate parameters of the phenomenological functional form that best fits the data we used the same general procedure and parameter estimation algorithm implemented in our earlier contributions [11 , 13] . Solving the state-equation via iteration in the simpler case where the system learns only from the last experienced feedback ( cf . S1 Appendix ) , or borrowing techniques from the theory of LTIS reveals a correspondence between these phenomenological parameters and the coefficients of the generative model of Eq 5 . ( A complete derivation of the phenomenological parameters as functions of the generative ones is not presented here due to space limitations; details about the analytical procedures adopted can be found in [42] ) . Depending on the parameters that each generative model includes , the functional form and value of the phenomenological coefficients may change . Here we are interested in assessing which theoretical prediction of the relation among phenomenological and generative model parameters matches the data best as a way to validate the underlying sensorimotor learning algorithm . The lag of the periodic response of the error gain derived from the ( full version of the ) generative model of Eq 5 including the next-to-last feedback-error term is given by: cos ( ϕ ) =cosω- ( A-K-Dcosω ) [cosω- ( A-K-Dcosω ) ]2+[ ( 1-D ) sinω]2sin ( ϕ ) = ( 1-D ) sinω[cosω- ( A-K-Dcosω ) ]2+[ ( 1-D ) sinω]2 . ( 7 ) In models without next-to-last feedback term D should be set to zero; in models that do not have A as a fitting parameter , its value should be set to 1 in Eq 7 . The periodic component of the response to a sinusoidal disturbance in models where the next-to-last feedback is included can be written as: h ( n ) =KRsin ( ωn-ϕ ) +DRsin ( ω ( n-1 ) -ϕ ) , ( 8 ) where R=[cosω- ( A-K-Dcosω ) ]2+[ ( 1-D ) sinω]2 . Eq 8 shows that if D = 0 we recover the solution expected by iteration when there is learning from the last error only . Then the amplitude of the periodic component ( a ) in Eq 6 can be read out directly to be a=KR . When D ≠ 0 we need to re-write Eq 8 so that it matches the periodic part of Eq 6 . After some algebra Eq 8 can be recast as: h ( n ) =QR∙sin ( ωn- ( ϕ+φ ) ) ( 9 ) where Q= ( K+Dcosω ) 2+ ( Dsinω ) 2 , cosφ=K+DcosωQ , sinφ=DsinωQ . ( 10 ) Eqs 7 to 10 clarify the effect of the presence of the next-to-last error learning rate D . Eq 7 shows how the bare lag ϕ changes when D is present . Yet , it would be incorrect to compare the fitted values of the phenomenological lag to Eq 7 . The reason is that the second contribution in Eq 8 modifies not only the amplitude of the periodic component to the new value a=QR , but it also adds the shift φ to the lag . Therefore , if there were also learning from the next-to-last error , the observed ( behavioral ) lag should be compared to ϕ + φ . Following a sinusoidal disturbance , the baseline of the error gain will approach an asymptote at large trial number that can be written as a function of parameters of Eq 5 as ( see also S1 Appendix ) : B0=m1- ( A- ( K+D ) ) ( 11 ) The timescale λ for the decay of the baseline , has units of 1/trials and it is defined by: e-λ=12{ ( A-K ) ±[ ( A-K ) 2-4D]12} ( 12 ) Eq 12 provides the weights of the impulse response that generates the integral solution by convolving the stimulus ( i . e . , s ( n ) ; cf . S1 Appendix , [42] ) . The inverse of the timescale parameter λ gives the number of trials over which the stimulus is integrated . Beyond this window of integration , the weighting of the stimulus would have reduced enough to ignore further contributions . When D = 0 , the integration weight becomes e−λ = ( A − K ) , which is positive and smaller than 1 , provided that the learning rate K < 1 and A ~ 1 . When D ≠ 0 , Eq 12 provides timescales for two modes that compose the integral solution of the state-equation . These result from the addition or subtraction of the second term in braces . If the parameter D is negative , the second term inside the braces becomes slightly larger than the first . The timescale resulting from the addition is positive and can be expressed as a decaying exponential . The subtraction solution is negative and of small magnitude and , therefore , it will decay much faster when raised to the trial number . It introduces small additive fluctuations to the exponential decay of the addition solution without changing its overall behavior . Critically , diverse sizes of the learning parameters may result in smaller or larger timescales in models with D ≠ 0 compared to models where D = 0 ( cf . Results section and S1 Appendix ) . To recap , Eq 6 has four phenomenological parameters that we shall explore in further detail: B0 , λ , a , and ϕ . The former two parameters are already familiar from phenomenological descriptions of data in paradigms using fixed-sized second-step for the target . The latter are new , arising in paradigms with sinusoidal disturbances . The amplitude of the decay of the baseline also bears dependence on the learning rates as well as on the initial condition . Because of the strong influence of the initial condition on this parameter , we refrain from a comparison of the behavioral fittings to the predictions from the generative model for this case . Part of the material discussed in this contribution have been presented in the form of posters or slide presentations [43 , 44] .
To obtain a general idea of patterns present in the data , we first collapsed the data for each stimulus frequency and adaptation type across participants ( group data ) . We fit these data using a piecewise continuous function given by the addition of a monotonic ( exponential ) decay of the baseline–spanning both pre-adaptation and adapting trials- and a periodic entraining of the oculomotor response to the sinusoidal stimulus that begins at the onset of the adaptation block . This choice was supported by the fact that we had confirmed using statistical model selection criteria ( i . e . , AIC and BIC , [38–41 , 45] ) that this functional dependence was the best descriptor of the oculomotor response among the set of models tested in Cassanello et al . [11] . For illustration purposes only , Fig 1 shows the group data in each dataset , along with the fits resulting from the parameter estimation based on the phenomenological model of Eq 6 . The same parameterization was used to fit each participant’s run . Figs 2 and 3 summarize the estimation of the phenomenological parameters entering Eq 6 . Fig 2 shows the values of mean ± SEM of the parameters estimated from every individual dataset for each frequency and adaptation type . Some features are readily apparent from these plots . First , the frequency of the ISS is reliably estimated ( cf . Fig 3b and 3d ) . Second , the amplitude and the lag of the periodic components of the adaptation gain decay with increasing frequency of the stimulus ( Fig 2a and 2b ) . The amplitudes of the periodic component are systematically larger in Two-way adaptation , while the lags observed in global adaptation are systematically larger than in the Two-way case . The systematic decay of the values of the lag with increasing frequency does not seem to extend to the smallest frequency ( 1 cpb in the new dataset ) . This may be related to the fact that at such low frequency the stimulus resembles more the behavior of a ramp that then turns rather than a truly periodic disturbance . The parameters that affect the observed drift in the baseline ( i . e . , asymptote and timescale , Fig 2c and 2d ) remain rather independent of the experimental condition . This feature is more apparent in the ORIG dataset , but it still seems to hold in the FREQ dataset . An exception arises at the lower frequency ( 1 cpb ) tested in the FREQ dataset . However , the case of frequency one is rather special and should possibly be considered as transitional between periodic and non-periodic stimuli . Fig 3 provides an idea of the quality of the fits by showing the evidence of the data in favor of the models tested ( cf . [11 , 13 , 28] ) . Upper and lower rows correspond to Two-way and Global adaptation type respectively . For dataset ORIG , Fig 3a shows the logs of the odds ratio of the parametric model of Eq 6 against a noise only model consisting of the block mean with variance similar to that of the data . Each bar is split into the log of the odds ratio of the full model to a drift only model that lacks the sinusoidal component ( darker tone of the bars ) added to the log of the odds ratio of the drift only model to the noise only model described above ( lighter tone of the bars ) . This separation is possible because the models are nested so that the simpler models can be obtained from the full model by eliminating parameters . The evidence then compares the density of models likely to fit the data . Fig 3b shows the estimation of the frequency of the oculomotor response against the actual frequency of the stimulus for the three frequency values tested in dataset ORIG ( 3 , 4 , and 6 cpb ) . Fig 3c and 3d shows the evidence and the agreement of the response with the five frequencies used in dataset FREQ ( 1 , 3 , 6 , 12 , and 24 cpb ) . To assess the generative model , we fit Eq 5 to all data available . For illustration puposes only , we first show that the model provides a reasonable overall fit to the group data . Fig 4 shows fits of the oculomotor response predicted by the full form of the generative model given by Eq 5 with all five parameters described in the Methods section: K , A , m , D as well as the initial condition G . As before the qualitative agreement of the fits and the data is evident in both datasets . As we did with the phenomenological fits , we included the pre-adaptation blocks in each condition in each dataset . For all subsequent analyses , we fitted models to individual data . In particular , we compared 16 different models that differed from each other depending on which parameters were fitted ( see Methods for details ) . We used Akaike’s information criterion ( AIC ) to explore statistical selection among these models . Akaike weights ( cf . section II of [40] ) are shown in Fig 5 segregated by model and condition , for datasets ORIG and FREQ , respectively . In each condition ( identified by adaptation type and stimulus frequency ) , we computed a matrix of weights in the following way . Because the best fitted model may differ between individuals , we first computed the AIC weights among the 16 models for each participant and condition . Then we averaged the resulting individual weights across participants . Results from this procedure are shown in Fig 5 . Inspection of Fig 5 suggests clear overall preference for models in groups II ( which include m but not D ) and IV ( featuring both m and D ) . We discuss below why this is expected on theoretical grounds given the features of the data . Models from group IV that learn based on two error samples , are preferred in Two-way adaptation , specifically the full model ( KAmDG ) and the model in which A was set to unity ( KmDG ) . Models in group II that feature a single learning rate ( error-correcting based only on the last experienced feedback ) , specifically KAm and KAmG , have an edge in Global adaptation . In what follows , we will focus on a comparison of these four models . Fig 6 shows the values of the generative parameters ( Mean ± SEM , N = 10 for dataset ORIG , N = 13 for dataset FREQ ) of the best models that learn only from the last experienced feedback error ( KAm , KAmG ) . Left and right columns correspond to datasets ORIG and FREQ respectively . Learning rate K , persistence rate A and drift parameter m are shown in Fig 6a , 6b and 6c . Fig 7 reports the parameters of the best models that update their hidden variable based on double error sampling . Those models are KmDG and KAmDG . Fig 7a and 7d show respectively the learning rates K and D that weight the contributions of last and next-to-last feedback , Fig 7b and 7c show the persistence rate A and the drift parameter m . Note that all models include the drift parameter m as a fitting parameter . We shall explain below why this should be expected . Again , several features are readily apparent from these plots . The learning rates ( K and D ) obtained from ORIG [11] , show a rather clear segregation between Two-way adaptation and Global adaptation: K and D are larger for Two-way ( blue colors ) than for the Global case ( red colors ) suggesting that the extra variability brought upon by the random directions of the subsequent saccades characteristic of Global adaptation has a detrimental effect on all learning rates . They do not show a strong dependence on the frequency but the range of values used in that experiment was rather narrow , ranging from 3 to 6 cpb . This segregation in the learning rates between Two-way and Global adaptation is also clearly present in the best models fitted to dataset FREQ . A feature observed in all cases is that in models that learn only from the last experienced error , the ( single ) learning rate ( K ) shows a mild increase with the frequency ( cf . Fig 6a ) . This changes substantially if learning from the next-to-last feedback is included . In all of these models , the following features are observed . First , the magnitude of K , the learning rate of the last-feedback error-correction term increases by about an order of magnitude with respect to the models that do not have next-to-last error-correction ( compare the scales of Figs 6a and 7a ) . Second , the magnitude of the next-to-last error learning rate ( D ) is similar to that of the last error ( K ) but with opposite sign . This seems to suggest that the next-to-last error is weighted negatively ( or actively attempted to be forgotten ) in the algorithm . Third , the discrepancy in magnitude between K and D is consistently larger for Two-way than for Global adaptation ( compare the separation between corresponding blue and red lines in Fig 7a and 7d ) . Fourth , the learning rate K reverses its dependence with the frequency of the stimulus with respect to the models without D , and now decreases monotonically as the frequency of the stimulus increases . At the same time , the magnitude of D also decreases with the frequency . As a consequence , the discrepancy in magnitude between K and D is such that the addition of both learning rates approximately matches the range of the values of K fitted in the models that learn only from the last feedback ( compare the values plotted in Figs 6a and 7e ) . This suggests that when the additional error learning is not part of the model , the only learning rate fitted may represent an average across subprocesses . The values of the parameters fitted with the best four models are shown in S1 Table ( Mean ± SEM , N = 10 for ORIG , N = 13 for FREQ ) . To assess dependence of the generative parameters on the experimental conditions we run 2 X 3 ( ORIG ) and 2 X 5 ( FREQ ) repeated-measures ANOVA on the fitted values using as regressors type of adaptation ( Two-way and Global ) and ISS frequency . Results are shown in S2 Table for the parameters given in S1 Table . We regard as more representative the results from dataset FREQ due to the more extended range of frequencies tested . Consistent with the qualitative observations mentioned above , while type of adaptation is highly significant for the learning rates in every model , frequency shows significance for K and D only in the models that feature double error sampling ( KmDG , for both datasets , KAmDG only for FREQ ) but not in those learning just from the last feedback ( KAm , KAmG ) . As for the persistence rate , frequency is never significant suggesting that it can be kept fixed as in model KmDG . Type of adaptation is significant in KAm and KAmG but such significance disappears in KAmDG . The iteration of state-equations that learn from the last feedback already qualitatively predicts both components of the phenomenological response . In general , the complete response can be interpreted as a convolution of the stimulus with a response function . This response function integrates the stimulus by weighting the disturbance over a temporal window , the size of which depends on the magnitude of the learning and persistence rates that combine to assemble the weights ( cf . S1 Appendix ) . Contributions from constant components of the disturbance that arise either from constant features in the stimulus ( as in the traditional fixed-ISS paradigm [1] ) or from intrinsic biases that may not be strictly error-based in nature ( e . g . , in our case represented by the drift parameter; cf . [37] ) accumulate across trials , changing saccade gain in a monotonic fashion akin to a drift of the baseline towards an asymptote . Iteration of the systematically varying part of the disturbance results in its convolution with similar weights but the trial-by-trial variation usually prevents finding a closed form for the series re-summation . However , a sinusoidal disturbance avails a closed analytical integral solution , it is periodic with the same frequency , lagging the stimulus by a number of trials . Two new phenomenological parameters of this periodic response—its amplitude and lag—bare characteristic dependences on the learning parameters . Above , we fitted the extended version of Eq 6 to the data and obtained and reported estimates for its phenomenological parameters ( i . e . , frequency ν , amplitude a , lag ϕ , asymptote B0 , timescale λ and decay amplitude B; cf . Fig 2 above ) . Similarly , we fitted the generative parameters for all generative models using the corresponding versions of Eq 5 . Figs 6 and 7 display those estimates for the four models that provided the best fits ( excluding D: KAm and KAmG and including D: KAmDG and KmDG respectively ) . When the learning algorithm includes several error-based terms , Eq 5 can be integrated using techniques standard within the theory of LTIS [42] . This integration provides analytical predictions of the phenomenological parameters as functions of the learning parameters fitted with the generative models ( Eqs 7 through 12 ) . We attempt matching these predictions to the values fitted using the phenomenological parameter estimation implemented before ( see Fig 2 ) . It should be pointed out , however , that the phenomenological parameter values have also been obtained from fits to the data and therefore should only be regarded as indicative reference values to guide intuition , not as ground truth . Validation of the actual underlying structure of the learning model relies ultimately on statistical model selection . Yet , a direct comparison between the fitted phenomenological parameters and analytical predictions evaluated on the fitted generative parameters is informative because a given value of a phenomenological parameter has to be compared to diverse combinations of the generative parameters that in turn depend on the specific structure of the learning model . We start with Eq 11 that provides a relationship between the expected asymptote of the adaptation gain at large trial number and the generative model parameters . A first significant observation about this expression is that in order to observe a drift in the baseline of the adaptation gain ( i . e . , in order to have an asymptote B0 ≠ 0 ) , a finite value of the drift parameter m is strictly necessary . If m vanishes , the adaptation gain would maintain a baseline pinned at zero regardless of the values of K , A or D . In addition , in a situation where A ~ 1 , B0≈mK+D or mK in models where D = 0 . Note that these are all signed magnitudes , not absolute values . In other words , a small learning rate K or a small number resulting from the addition K + D will modulate the size of the asymptote and will determine its sign ( i . e . , will modify the degree of hypometria or hypermetria ) . Still a finite value for m is strictly needed to have non-zero asymptote . Recall that when m is not a fitting parameter , its value is set to zero . Due to the pervasive baseline drift across all of our data , all models favored under statistical model selection contain m as a fitting parameter . This is why model groups II and IV ( cf . Fig 5 ) are preferred , as pointed out above and in the Discussion . Note , in addition , that the smaller the learning rate ( K or K + D ) , the larger the size of the asymptote B0 . Experimentally , we observed drifts towards higher hypometria in all averages and in most of the individual data . Note that formatting the data in terms of adaptation gain instead of saccade gain allows us to remove confounds coming from constant contributions from the stimulus and therefore the parameter m should be regarded as intrinsic to the system . In other words , m characterizes or quantifies learning that would occur in absence of stimulus disturbance ( i . e . , with zero ISS ) , as if the system has an intrinsic propensity to modify its gain by virtue of environmental or experimental conditions not necessarily linked to an error . Fig 8a displays the matching of the analytical predictions of the asymptotes computed by inserting the fitted values of m , A , K and D into Eq 11 for each participant’s data , to the phenomenological estimation of B0 obtained from Eq 6 and the parameter estimation of the phenomenological fits of the data for both datasets and both adaptation types . A second parameter characteristic of the baseline drift is given by the timescale . Fig 8b shows predicted values for the timescales that result when values of m , A , K and D fitted with the state-equation are inserted in Eq 12 . The first two rows of Fig 8b show a clear overestimation of the baseline timescale in models that do not feature double error sampling ( i . e . , KAm and KAmG ) as several individual data points fall outside the boundaries of the plot . Yet , models that include corrections based on the next-to-last error term , seem to underestimate the timescale ( in particular model KmDG ) . When introducing Eq 12 , we pointed out that if the second error learning rate D is negative , the dominant mode in the solution still features a monotonic decay that can fit the phenomenologically observed exponential baseline drift of the gain . This is indeed the case in the majority of fits to the individual participants’ runs: Across models in group IV , D was non-negative in only 13% of the individual runs; 6% for Two-way adaptation and 21% for Global adaptation data . For model KmDG , D was non-negative in 7% of all runs; with only 1% ( 1 run out of 95 ) for Two-way adaptation and 14% for Global adaptation . Furthermore , when estimating the timescales of models that include double error-correction , Eq 12 consistently gives smaller values than for models without the second error term ( compare subplots of Fig 8b for the corresponding models , cf . Fig A , S1 Appendix ) . This ordering relation between the timescales of models with and without D was unknown before conducting the fits . Thus , data collected using a sinusoidal adaptation paradigm suggests that including a second error-correction term yields a significant decrease in the timescale with respect to models featuring a single error-correction term . Therefore , the integration windows ( i . e . , the inverse of the timescale ) of models with double error-correction grow significantly larger compared to those that lack the second error sampling . Asymptote and timescale are parameters traditionally investigated and reported in adaptation to fixed-step disturbances . Sinusoidal adaptation paradigms provide two additional parameters associated to the periodic component of the adaptation gain observed in these protocols . Fig 8c and 8d compare predictions for the amplitude and the lag of the periodic component of the gain obtained by using Eqs 7 through 10 above . Data from both datasets suggest that models that do not feature double error sampling underestimate the magnitude of the amplitude of the periodic component of the oculomotor response ( cf . predictions from these models in Fig 8c ) . This feature in fact is common to all models that learn from a single feedback and include m ( besides models KAm and KAmG ) , but the inclusion of D helps mitigating misestimation of this amplitude . The last comparison is provided by the lag of the periodic component . Fig 8d compares predictions based on the state-equation learner ( Eqs 7 and 10 furnish predictions for the components of the lag ϕ and φ after inserting the parameter values fitted with Eq 5 ) and the phenomenology ( parameter ϕ in Eq 6 ) . From Fig 8c and 8d it is apparent that the models that include both m and D as fitting parameters provide better predictions , also displaying less variability across participants , in particular for the Two-way adaptation type . Among models with D = 0 , again models KAm and KAmG fit best . Fig 8d shows , however , that these models appear to overestimate the lag ( cf . compare corresponding subplots in the figure ) , while models that have a second learning rate D match better the empirically observed lag . In addition , all models fail the estimation of the lag for a disturbance of frequency one as they all significantly overestimate the lag observed . Even though the predictions of the other phenomenological parameters are reasonable ( the amplitude of the periodic component , timescale and asymptote ) , predictions for the 1cpb condition for both Two-way and Global adaptation have been omitted altogether in Fig 8 . This mismatch between the direct phenomenological estimation of the lag from the data and the analytical predictions arising from the integration of the state-equation for the case of the 1cpb condition , may be rooted in the fact that the functional dependence of the phenomenological parameters on the generative ones is determined by the specific sinusoidal dynamics of the driving stimulus , while the case of a 1-cpb frequency is the least periodic condition among all tested .
In mathematical terms , the functional form in Eq 6 is the integral solution of a family of LTISs of which Eq 5 is a particular example . It is referred to as a state-equation or state-space model because the internal variable x characterizes the gain or state of adaptation of the system . This algorithm is generative because it estimates the value of x at trial n + 1 by modifying its estimate at the previous trial including possible effects of systematic biases and correcting the former value by weighting sensory feedback resulting from movement execution [21 , 26 , 47 , 48] ( see also [25 , 32]; for further details on our specific use see the Methods section ) . Here we limit our discussion to noise-free generative models in that Eq 5 does not include any noise term . Yet , Fig 1 together with Fig 4 suggest that integral solutions as well as numerical outcomes of noise-free generative models survive ensuing variability , at least for the paradigm , type of stimulus and within the ranges of the conditions tested . We analyzed 16 models that differed in the specific parameters that were fitted and then used Akaike’s information criteria to attempt model selection . Since we were primarily modeling intrinsic error-based sensorimotor learning , the learning rate K—that weights the impact of the last feedback error on the state of adaptation—was present in every model . Second , we included the initial condition G as a fitting parameter in half of the models . This parameter is not part of the trial-by-trial learning algorithm and its effect should decay as the trial number increases ( cf . S1 Appendix ) . However , the initial condition affects the amplitude of decay of the baseline drift ( cf . B in Eq 6 ) . Because the argument of Eq 5 is an internal variable not directly experimentally accessible , a proxy for its initial value can only be approximated ( for example , by averaging the first five gain values in the block ) or included as a fitting parameter . Third , we included a persistence rate A that weighted how much of the estimate from the previous movement remained in the subsequent one . The fourth parameter , m , captured systematic effects , that are not error-based in nature , and gave origin to drifts in the baseline that were pervasive across all conditions . Finally , we considered the plausibility and study the effects of a second learning rate D that tracks errors other than the most recent ( here , the next-to-last feedback error ) . To further discuss the effect of the generative parameters , we split the 16 models into four groups: We recall that in models where A is not a fitting parameter , A = 1 . The groups are listed on the left side of Fig 5 . Models within group I consistently fitted worst . Moreover , models that do not include m ( groups I and III ) cannot capture an evolution of the gain into a stationary asymptotic value because the state equation does not admit a solution featuring that behavior ( that is , if the stimulus has no constant term ) . These models , however , may be useful in experimental paradigms where a stable state of adaptation is not clearly reached either because the length of the adaptation block used may be too short or because the driving disturbance is unbounded ( for example a linear ramp ) . On the other hand , models that include sampling from two errors ( cf . groups III and IV ) will likely be better suited to extract correlations built into the stimulus as it is the case of a sinusoidal ISS . The fits of the phenomenological model ( Fig 1; Eq 6 ) suggest that asymptotic behavior of the baseline and reflection of the stimulus self-correlation ( entraining ) were clear structural properties of the oculomotor response . The analytical solutions of models in both groups II and IV are consistent with this phenomenology . Fig 5 summarizes the AIC weights emerging from the fits to the individual participants’ data . The weights shown in the horizontal bars are averages over individual participants’ weights for each condition and color coded by the frequency of the stimulus . Data from Two-way adaptation is depicted with blue tones in bars increasing towards the right . Global adaptation is shown with bars spanning to the left in red tones . The average weight for each model family is shown by the gray background behind the corresponding group . While models in group II already generate responses in qualitatively good agreement with the evolution of the adaptation gain , it remains to be decided whether corrections based on the memory of more than a single error provide for a better fit . AIC weights show that group IV clearly outperforms all others in Two-way adaptation in both datasets , suggesting that the best generative model to describe this type of adaptation includes all four parameters K , A , m and D . In Global adaptation , models from group II either match or slightly outperform those of group IV . Model comparison showed that a state-equation including a single parameter or any combination of only two of the four parameters K , A , m and D could not adequately account for our data ( cf . Fig 5 ) . In addition , an inspection of actual values of the parameters fitted across the population suggests that the parameters A and m may be set to constant values , that is , to almost one for the former and to a very small and negative number for the latter ( cf . Fig 7c and 7d ) , at least within the range of frequencies tested in these experiments . Overall , the drift parameter m and the second learning rate D proved useful and necessary to account for systematic effects in our data , suggesting ( 1 ) that some changes in the adaptation state are not error-based and ( 2 ) that—at least under specific circumstances—the brain keeps track of at least one extra occurrence of the error besides the last experienced one . Three-parameter models that did not involve D ( specifically KAm ) were most successful in Global adaptation and in the high frequencies of Two-way adaptation . This could be simply a reflection of increased levels of measurement noise in these conditions giving an upper hand to models with fewer parameters . More interestingly , it could point to an architecture that samples two errors only under certain conditions , for instance , when errors are repeatedly experienced for the same saccadic vector , or , when the variation of the feedback error has a high signal-to-noise ratio . We speculate that overtraining along a given direction , understood as the repetitive experience of consistent error along similar saccade vectors in Two-way adaptation ( note that in our paradigm Two-way adaptation stimulates only two retinal locations ) may give rise to vector specificity and , consequently , to the adaptation fields typically observed with fixed-step paradigms . Indeed , Rolfs and collaborators [18] suggested that Global adaptation , featuring apparent full transfer across random directions , appears to onset ahead of the development of vector-specific adaptation fields . This appears consistent with the present finding that models that rely on a single error-correction show timescales corresponding to faster evolution of the baseline drift ( although with longer lags in the sinusoidal component ) as compared to those of Two-way adaptation ( featuring shorter lags in the sinusoidal component consistent with tracking the stimulus more closely due to the repetitive training in a specific direction ) . The persistent drift of the baseline towards higher hypometria is a distinctive feature in our data that cannot be accounted for on the basis of motor adaptation [49] . We included an extra parameter m to account for this drift in mean adaptation gain towards an asymptote differing from the mean of the stimulus ( cf . Eq 11 ) . This parameter is conceptually novel , distinct and independent of the persistence rate A , and determines the presence of a non-zero asymptote via Eq 11 . Because in our paradigm the goal of the task was to land on the target as close as possible , and because the sinusoidal ISS introduced a continuously changing prediction error , the best expected outcome would be to track the disturbance within the levels of error typical of trials without disturbance . With respect to that goal , the presence of a baseline drift introduces an additional discrepancy that does not , however , hinder successful adaptation to the disturbance . Saccadic eye movements slightly undershoot their target on average [50] and this systematic offset corresponds to the internally predicted visual outcome of a saccadic eye movement [51 , 52] . We surmise that our paradigm may have yielded a re-calibration of this desired offset [53] over the course of an experimental run . This recalibration towards a larger undershoot may result from the high probability of a quick return saccade after every eye movement in our fast-pace paradigm , reducing the utility of maintaining a saccade gain close to one . We note that this systematic decrease in saccade gain may in general—albeit to different degrees—pervade the study of saccadic adaptation ( but see [7 , 54] ) . In fixed-step paradigms ( as opposed to the sinusoidal paradigm employed here ) it would have been obscured as the error-based correction for the surreptitious target displacement undergoes similar dynamics as the drift reported here . On the other hand , from the point of view of the internal model of the movement that the brain may implement [33–35] , this bias parameter m may hint to a discrepancy between the experimental coordinate system where measurements are acquired and the coordinate system in which the internal model is represented . On a neurophysiological level , the small systematic bias that gives rise to the drift of the baseline may originate from the dynamics of the responses in the neuronal substrates involved with saccade adaptation ( [55–60] , R . Shadmehr , personal communication , July 12 , 2018 ) . It is also possible that the fast-pacing used in our paradigm exacerbates effects that generate a small and negative bias parameter , m , which appeared to onset already at the pre-adaptation block . That would further suggests that the magnitude of m may depend on the inter-saccade interval as well as on the precise timing of the ISS onset , which should be addressed in future studies . The models that best explained the data featured a double error sampling , learning not only from the feedback experienced after the last saccade but also from the movement that occurred in a trial before that . Hence , the best models used a feedback reaching further back in time through the K- and D-terms of Eq 5 . Yet , does the oculomotor system actually implement this double error sampling that may coherently participate in a single internal model prediction ? We suggest that the brain may attempt to approximate the performance achieved by the double-error-sampling algorithm by using two single-feedback learners operating on appropriate combinations of the stimulus sampled at two different times . To understand that , we return to Eq 5 . For simplicity , we will assume that m = 0 . x ( n+1 ) = ( A-K ) x ( n ) -Dx ( n-1 ) +Ks ( n ) +Ds ( n-1 ) , ( 13 ) and write a transformation among state variables sampled at two different trials as , X+ ( n ) =12{x ( n ) +x ( n-1 ) }andX- ( n ) =12{x ( n ) -x ( n-1 ) } , that can be substituted in the RHS of Eq 13 using the inverse relations: x ( n ) =X+ ( n ) +X- ( n ) , andx ( n-1 ) =X+ ( n ) -X- ( n ) . We can re-write Eq 13 in terms of these alternative state variables X+ and X−: X+ ( n+1 ) +X- ( n+1 ) = ( A-κ ) X+ ( n ) + ( A-η ) X- ( n ) +κS+ ( n ) +ηS- ( n ) , ( 14 ) where we adopted the definitions of κ = K + D , η = K − D , S+ ( n ) =12 ( s ( n ) +s ( n-1 ) ) and S- ( n ) =12 ( s ( n ) -s ( n-1 ) ) . Eq 14 avails the interpretation of the generative model as selectively learning into two component channels that learn from a single feedback error taken from different sources . The source for the learner X+ is the mean of the two samplings of the stimulus , i . e . , S+ ( n ) =12 ( s ( n ) +s ( n-1 ) ) . The source for the second learner is the rate of change of the stimulus across the sampling events given by S- ( n ) =12 ( s ( n ) -s ( n-1 ) ) which , when the samplings occur on successive trials , it could be interpreted as the discrete time derivative of the stimulus taking the elementary timestep as the ( average ) inter-trial interval . Note that the representation in terms of these alternative internal variables would significantly alter the underlying structure of the noise-free learning model . But if we insist on keeping a close connection to the parameters extracted using the double-error-sampling algorithm , we would expect that the learning rate for learner X+ would be the addition of the rates for the two errors , κ = K + D , while for learner X− it would be η = K − D ( cf . Fig 7e and 7f ) . In all our fittings using the double error sampling , K and D were very close in magnitude but carried opposite sign . Furthermore , κ was small and similar in magnitude to the learning rate K obtained for models that learned only from the last error . Because D was negative , the learning rate η for the second learner became also positive but much larger than κ , in fact about an order of magnitude larger ( Fig 7e and 7f ) effectively enhancing the overall gain of the process without driving the system unstable [61–63] . As a consequence , ( A − κ ) , which can be thought of as an effective A+ will be much closer to unity than A− = ( A − η ) . Therefore , X+ will learn and forget much slower than X− . Using this double error sampling , the oculomotor system could track the rate of change of the stimulus from one saccade to the next , besides just its last change in size and it would approximate the learning efficiency of the double-error-sampling algorithm . The new internal learning variables ( X+ and X− ) would learn from smoothed-out versions of the disturbance resulting from the average sum and difference of the two sampled inputs . Whether this constitutes an advantage over learning exclusively from the last feedback depends on the nature of stimulus . If the disturbance is constant or fully random there would be very little advantage in performing the double error sampling . In the former case , the inter-sampling variation is zero leaving nothing to learn . In the latter , the inter-sampling variation would be another random magnitude and there would be little advantage in learning from the variation in the feedback . However , if the mean of the disturbance varies in a systematic way—as it does during sinusoidal adaptation , and presumably in natural scenarios—learning from its rate of variation would be advantageous and could well justify a large learning rate . In the representation of the double-error-sampling model , unlearning actively the next-to-last sampled feedback error ( i . e . , with a large and negative D subtracted from an enhanced K ) would materialize this advantage with little extra investment . However , a negative learning rate feels counter-intuitive as learning is believed to follow the direction of the correction suggested by the feedback . Segregation of the learning underlying motor ( or saccade ) adaptation into two learners displaying similar characteristics to those suggested here have indeed been proposed in other contexts [8 , 25 , 64 , 65] . The argument presented above suggests a mechanistic way to construct a two-learner system , in which the components X+ and X− can be considered statistics in counterphase . To approximate the double-error-sampling learner , the system may hold in memory both samples , compute mean sum and differences between the samples and implement two learners based on those statistics rather than from bare values of errors or stimulus occurrences . To achieve that , the oculomotor system would need to keep memory and weight prediction errors from a former time scale besides the last feedback [65] . An important point to notice is that , even if there is double error sampling , it does not need to be strictly the next-to-last error . It would be enough that the brain keeps a correlation of errors over two different trials ( cf . [66] ) although it would be reasonable that they are spaced only by a short delay [61] . This is a reasonable generalization since the inter-trial interval is rather arbitrarily set by the pacing of the task that may or may not match a possible internal sampling frequency by the brain . The frequency of the stimulus then determines to what degree differences in the stimulus can be sampled , which may explain the dependence of the amplitude and lag of the periodic component of the response with the frequency as well as the fact that the evidence for the full model seems to peak at intermediate frequencies . In other words , it may be easier to learn at certain frequencies ( for a fixed amplitude ) or at certain effective rates of change of the stimulus . We further explored whether the values of the generative parameters exhibited dependence on the experimental condition , specifically with the type of adaptation and the frequency of the disturbance . The parameters of our models remained time-invariant across pre-adaptation and adaptation blocks . However , we did not rule out that these parameters may change with adaptation type and stimulus frequency . In fact , LTIS models with parameters not strictly time-invariant have been invoked to model ( meta-learning in ) savings in adaptation to visuomotor rotations [32] . Strict LTIS models with two learners had been able to successfully account for savings in long-term saccade adaptation [8 , 25 , 64 , 67] ) but were not able to fit differences in the dynamics of the adaptation , extinction and re-adaptation phases observed using counter-adaptation and wash-out paradigms in adaptation to visuomotor rotations without letting the rates change across the phases [32] . We limit our discussion to the best four generative models selected in the Results section . In models KAm and KAmG ( and in general in all models of groups I and II ) , the ( only ) learning rate K remained relatively independent of , or exhibited a tendency to grow with the frequency of the stimulus ( Fig 6a ) . Learning rates for Two-way adaptation roughly ranged between 0 . 01 and 0 . 035 fraction of the error across the frequencies tested . The same parameter in Global adaptation was smaller and remained within the range 0 . 005 to 0 . 015 ( cf . Fig 6a , S1 Table ) . These observations were confirmed by ANOVAs run on the fitted values of the parameter K in models KAm and KAmG in that type of adaptation was always a significant factor while ISS frequency never was ( S2 Table ) . These values of K compare reasonably well with the magnitude of learning rates previously reported in the literature ( cf . [8 , 19] ) . The dependence of the learning rate on the frequency of the disturbance seems in qualitative agreement with results from reaching experiments in which subjects learned to track a target undergoing surreptitious displacement that followed a random walk [30 , 47] . Using a Kalman filter to estimate corrections to the learning rate due to various types of variability Burge and collaborators [30] argued that the learning rate increased as the drift of the walker increased . In the sinusoidal adaptation paradigm where the amplitude of the sine function that produces the ISS is of fixed amplitude , this situation occurs when the frequency increases because its size from one trial to the next changes faster . However , this suggestion seems at odds with the intuition that a more consistent stimulus should drive more efficient adaptation [68 , 69] . In particular , it has been reported that a smooth gradual variation results in more efficient adaptation [3 , 70] . If this were the case and reflected onto the model parameters , the learning rate should be higher for smaller frequencies . However , the dependence of the learning rate ( s ) on the frequency described above changed rather dramatically when double error sampling was included ( cf . Fig 7a and 7d ) . Interestingly , in models that feature double error sampling , the learning rate of the most-recent error-term ( K ) reversed its tendency and decreased as the frequency increases , achieving its highest values in the conditions of lower frequency , this is , in situations where the stimulus displayed higher consistency . Concurrently , the learning rate for the next-to-last feedback ( D ) achieved its most negative values at lower frequencies and grew less negative as the stimulus frequency increased . In the alternative scenario of two additive learners with single error correcting terms that learned respectively from the half-sum and the half-difference of the two sampled errors suggested in the previous sub-section , the learning rates κ and η also showed a distinct dependency on the ISS frequency . The slow-learner ( with learning rate κ ) would only have corrected up to 1% of the average of the two errors sampled while the fast-learner ( with rate η ) would have produced corrections of up to 40% of the change experienced between the two sampled errors ( cf . Fig 7e and 7f ) . Note that this massive change in the dependence of the learning rates on the frequency was a consequence of changing the hypothesized structure of the model and not of correcting the magnitude of the rates for effects of variability . Once again , ANOVAs confirmed that not only the type of adaptation but also the stimulus frequency had significant impact on the learning rates ( K and D , as well as κ and η ) in models KmDG and KAmDG as well as all models of group IV ( cf . S2 Table ) . In contradistinction , the retention rate A ( Figs 6b and 7b ) and the bias parameter m ( Figs 6c and 7c ) remained relatively independent of the frequency under such changes , although their overall variability was clearly reduced in the models featuring double error sampling ( contrast the value ranges of m and A in Fig 6 , against the corresponding ones in Fig 7 , aside from model KmDG in which A = 1; see also corresponding entries in S1 Table ) . ANOVAs run over these parameters further confirmed non-significance of the frequency except for model KmDG on m in dataset ORIG ( S2 Table ) . Type of adaptation occasionally modulated A in dataset FREQ in models with a single error term . Taken altogether this suggests that both A and m may be largely frequency independent and can be modeled as constant values maybe differing in value for Two-way and Global types . In summary , introducing a second error term increased the magnitude of both learning rates ( K and D ) by an order of magnitude with opposing signs . The learning rates of these models showed a clear dependence on the frequency of the disturbance: higher stimulus consistency ( i . e . , lower stimulus frequencies ) correlated with higher adaptation efficiency . At the same time , the inclusion of the double error sampling reduced variability in the estimates of the persistence rate A and the drift parameter m , indicating that their estimates were not affected by the ISS frequency , and could thus be set to appropriate constant values . Multiple distinct learning processes contribute to sensorimotor adaptation [8 , 25 , 64–66 , 71] . Recent research conducted primarily within adaptation to visuomotor rotations or in reaching movements , suggests that adaptation can be decomposed into two fundamental processes that may operate in parallel: one that would be responsible for implicit learning that progresses slowly and can be described mechanistically by a state-equation [49] . This slow learning process is relatively stable over breaks , takes place with automatic , reflex-like behavior and its properties tend to be sturdy and do not change fast with recent experience . A second , parallel process , in turn , learns explicitly , is faster although it may require longer reaction time and possibly voluntary behavior to be engaged . This faster process would exhibit longer term memory of prior learning [71–74] . We believe that our paradigm taps only the first , implicit component . Yet , we suggest that our analyses provide evidence for two separable subcomponents , although both would be intrinsic in nature [75] . In fact , a key difference between our oculomotor learning and learning that occurs in adaptation to visuomotor rotations and during reaching in force fields is that our participants were primarily unaware of the inducing disturbance . In contrast , in the aforementioned paradigms , participants immediately notice a disturbance even when they may not be fully aware of the exact effect . In this sense , our paradigm could be considered qualitatively closer to that used by Cheng and Sabes [22] who studied calibration of visually guided reaching in participants fully unaware of the stimulus manipulation . Their paradigm used a random , time-varying sequence of feedback shifts . They found that a linear dynamical system ( LDS ) with a single error term and trial-by-trial state update for variability implemented with an estimation-maximization algorithm successfully described mean reach point and the temporal correlation between reaching errors and visual shifts . They further argued that the learning taking place under a random stimulus generalizes to a situation of constant shifts in a block paradigm and , therefore , that adaptation dynamics does not rely on the sequence ( or correlation ) of feedback shifts but can be generally described with the LDS model . In contrast to random or block constant ISS , our paradigm featured a disturbance that was fully self-correlated since it followed a sine variation with the trial number . Therefore , it may prove advantageous for the oculomotor system to extract correlations embedded in the disturbance because they would help tracking the target . As pointed out , including double error sampling would serve this purpose . We believe that the presence of a systematically varying disturbance enables a further decomposition of the implicit component of adaptation , perhaps into a primary one , that attempts to mitigate the end-point discrepancy regardless of self-correlations in the disturbance , and a second one that attempts to extract ( and use ) such correlations . It remains an open question how these putative subprocesses may map on distinct or overlapping anatomical structures , such as cerebellar cortices , deep cerebellar nuclei and extracerebellar structures [55 , 57 , 59 , 60 , 64 , 76–80] . A recent study suggested that learning in dynamic environments may be adequately modeled with an algorithm popular in industrial process control , the proportional-integral-derivative ( PID ) controller [81] . The algorithm generates a control signal adding three error-related contributions: a term proportional to the current error that resembles a usual delta-rule ( the P-term ) , a term that integrates over a history of errors experienced before the current one , and a derivative term estimated from the difference between the last two errors . The model shares some features with ours , in particular that the learning rate for the next-to-last error needs to be negative to approximate the derivative term . The PID controller acts on the actual recorded errors ( the equivalent of the visual errors observed after each saccade is executed ) and contains no internal state estimation , whereas our model operates on an internal variable that contains the state estimation of the prediction error that would result from the movement execution . Our state variable in fact accumulates and retains a substantial portion of the history of previous error ( the persistence term in Eq 5 , see also the example given in S1 Appendix ) , which is updated on a trial-by-trial basis by the term proportional to the latest prediction error ( the delta-rule term ) . The inclusion of an extra error in our state-equation ( specifically that of the previous to last one ) effectively brings into play a contribution similar to the derivative term of the PID model . In short , our D-term enables a systematic correction to the integral term ( our A-term ) that otherwise would be determined rigidly by the iteration of the equation . In that respect , keeping track of former errors enables a structural correction that acts at a global level even when it is introduced on a trial-by-trial basis , lending both robustness and flexibility to the algorithm . Ritz and collaborators [81] further compared the performance of the PID model to a Kalman filter used to update a state variable in presence of noise applied on the single error structure of the usual delta-rule and found that the PID controller performs better . A further similarity with the aforementioned work lies in their observation that models with a derivative term are usually not readily selected under statistical model selection even when they may display significant improvement in the description of the behavior ( see [81] for a longer discussion on this point ) .
Having adequate generative models that describe eye movements have been stressed before [80 , 82–86] as an important tool to assess , at the single patient level , a variety of movement abnormalities that have been identified as markers of neurological conditions or disorders at a group level . In this study , elaborating on the idea of tracking a memory of errors [65] , we attempted to identify and constrain a relatively minimal set of requirements that would suffice to model saccade adaptation data collected under the paradigm and stimulus that we recently implemented [11] but that would also include previous accounts of the phenomenon under other known paradigms . While certainly many refinements are still due , we unveiled features of an algorithm that seems suitable to account for the sensorimotor learning observed in our data . We hope it can be generalized , extended and adapted for use in future research . | Constant adjustments of saccade metrics maintain oculomotor accuracy under changing environments . This error-driven learning can be induced experimentally by manipulating the targeting error of eye movements . Here , we investigate oculomotor learning in healthy participants in response to a sinusoidally evolving error . We then fit a class of generative models to the observed dynamics of oculomotor adaptation under this new learning regime . Formal model comparison suggests a richer model parameterization for such a sinusoidal error variation than proposed so far in the context of classical , step-like disturbances . We identify and fit the parameters of a generative model as underlying those of a phenomenological description of adaptation dynamics and provide an explicit link of this generative model to more established state equations for motor learning . The joint use of the sinusoidal adaption regime and consecutive model fit may provide a powerful approach to assess interindividual differences in adaptation across healthy individuals and to evaluate changes in learning dynamics in altered brain states , such as sustained by injuries , diseases , or aging . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion",
"Conclusions"
] | [
"learning",
"medicine",
"and",
"health",
"sciences",
"statistics",
"applied",
"mathematics",
"social",
"sciences",
"neuroscience",
"learning",
"and",
"memory",
"simulation",
"and",
"modeling",
"algorithms",
"cognitive",
"psychology",
"mathematics",
"systems",
"science",
"evolutionary",
"adaptation",
"eyes",
"research",
"and",
"analysis",
"methods",
"sensory",
"physiology",
"computer",
"and",
"information",
"sciences",
"mathematical",
"functions",
"mathematical",
"and",
"statistical",
"techniques",
"head",
"dynamical",
"systems",
"visual",
"system",
"statistical",
"models",
"psychology",
"eye",
"movements",
"anatomy",
"physiology",
"biology",
"and",
"life",
"sciences",
"sensory",
"systems",
"physical",
"sciences",
"ocular",
"system",
"evolutionary",
"biology",
"cognitive",
"science",
"evolutionary",
"processes",
"sine",
"waves"
] | 2019 | A generative learning model for saccade adaptation |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.