Search is not available for this dataset
article
stringlengths 4.36k
149k
| summary
stringlengths 32
3.35k
| section_headings
listlengths 1
91
| keywords
listlengths 0
141
| year
stringclasses 13
values | title
stringlengths 20
281
|
|---|---|---|---|---|---|
Genetic similarity of spouses can reflect factors influencing mate choice , such as physical/behavioral characteristics , and patterns of social endogamy . Spouse correlations for both genetic ancestry and measured traits may impact genotype distributions ( Hardy Weinberg and linkage equilibrium ) , and therefore genetic association studies . Here we evaluate white spouse-pairs from the Framingham Heart Study ( FHS ) original and offspring cohorts ( N = 124 and 755 , respectively ) to explore spousal genetic similarity and its consequences . Two principal components ( PCs ) of the genome-wide association ( GWA ) data were identified , with the first ( PC1 ) delineating clines of Northern/Western to Southern European ancestry and the second ( PC2 ) delineating clines of Ashkenazi Jewish ancestry . In the original ( older ) cohort , there was a striking positive correlation between the spouses in PC1 ( r = 0 . 73 , P = 3x10-22 ) and also for PC2 ( r = 0 . 80 , P = 7x10-29 ) . In the offspring cohort , the spouse correlations were lower but still highly significant for PC1 ( r = 0 . 38 , P = 7x10-28 ) and for PC2 ( r = 0 . 45 , P = 2x10-39 ) . We observed significant Hardy-Weinberg disequilibrium for single nucleotide polymorphisms ( SNPs ) loading heavily on PC1 and PC2 across 3 generations , and also significant linkage disequilibrium between unlinked SNPs; both decreased with time , consistent with reduced ancestral endogamy over generations and congruent with theoretical calculations . Ignoring ancestry , estimates of spouse kinship have a mean significantly greater than 0 , and more so in the earlier generations . Adjusting kinship estimates for genetic ancestry through the use of PCs led to a mean spouse kinship not different from 0 , demonstrating that spouse genetic similarity could be fully attributed to ancestral assortative mating . These findings also have significance for studies of heritability that are based on distantly related individuals ( kinship less than 0 . 05 ) , as we also demonstrate the poor correlation of kinship estimates in that range when ancestry is or is not taken into account .
The mating pattern determines the genetic structure of a population [1 , 2 , 3] . The genetic structure of a population refers to the distribution of alleles and genotypes in the population . Characterizing the genetic structure of a study population is important because ignoring it can lead to undetected biases , including false positive findings in genetic association studies [4 , 5 , 6] and inaccurate estimation of kinship and heritability [7 , 8] . Population substructure can arise in a study population from geographic stratification and/or non-random mating . Geographic stratification occurs when the study participants are recruited from different geographic areas , and the genotype data is aggregated for analysis . In this setting , although there may be random mating within each geographic subgroup , there is non-random mating within the aggregated study population . Population substructure can also occur when the study participants are recruited from the same geographic area but individuals are more likely to choose mates with similar genotypes to themselves–which can be a reflection of trait-based or ancestry-related assortative mating [9 , 10] . Assortative mating occurs when spouse choice is influenced by some observable phenotype or social characteristic [11] . Phenotype-based assortative mating has been well documented in humans for several traits including age [12 , 13 , 14] , height [12 , 15 , 16 , 17] , weight [12 , 16] and other physical characteristics such as skin pigmentation [18] , and eye and hair color [12 , 14] . In addition , there are other behavioral and social factors that are highly correlated between spouse-pairs and are thought to affect mate selection such as educational level [15 , 16 , 19] , occupation [15] , socioeconomic status [12 , 13] , religion [12 , 16] , smoking [20] , alcohol consumption [21 , 22] , language [23] and culture [23] . Phenotypic assortative mating generally does not change allele frequencies in the genes related to the trait for which the assortment occurs [24] . However , phenotypic assortative mating leads to both within-locus correlation and between-locus correlation specifically for alleles in the genes underlying the trait or traits that are the basis for the assortment [11] ( but not at other alleles ) . The within-locus correlation changes the distribution of the genotype frequencies for the alleles involved in the assortment towards an excess of homozygosity and Hardy-Weinberg disequilibrium ( HWD ) . The between-locus correlation leads to linkage disequilibrium ( LD ) between alleles at different loci involved in the assortment , even ones on different chromosomes [2 , 24] . By contrast , when assortative mating is related to ancestry ( ancestry-related assortative mating ) , the same phenomena are seen , but in this case , at all loci that differ in allele frequency between ancestral populations , whether they have a phenotypic impact or not [9 , 10 , 25] . After a single generation of random mating , the within-locus correlation disappears and Hardy-Weinberg Equilibrium ( HWE ) is achieved [24] . However , the between-locus correlation decays at a slower rate [9] . The correlation ( LD ) between unlinked loci is halved after a single generation of random mating; more generally , in the case of persistent ancestry-related assortative mating , the rate of decay in each subsequent generation is proportional to ½ ( 1+ ρ ) , where ρ is the correlation in ancestry between spouse-pairs [9] . While observed Hardy-Weinberg deviations and linkage disequilibrium can be due to a number of factors ( e . g . genotyping error , inbreeding ) , there is a very specific pattern of Hardy-Weinberg deviation and linkage disequilibrium caused by assortative mating . As opposed to inbreeding , which affects all loci equally and creates no linkage disequilibrium , and genotyping error , which presumably is based on technical aspects of genotyping and also has no population pattern , assortative mating only influences the univariate and joint genotype distributions of the loci that are related to the trait for which assortment occurs , and not others . And , there is also a direct relationship between the strength of the association of particular SNPs with the trait undergoing assortment and the degree of HWD and LD expected for them . This relationship was first described for a general , observable trait [24] . We previously extended those formulas to apply to the situation of assortative mating related to genetic ancestry [9] in populations of Latino race/ethnicity , and now apply them here to a three-generation European ancestry population from Framingham , Massachusetts . The Framingham Heart Study ( FHS ) is an epidemiologic longitudinal study of several cohorts of participants who reside in Framingham , Massachusetts [26] . Three generations of participants , i . e . the original cohort , their offspring and the offspring’s spouses ( offspring cohort ) , and the original cohort’s grandchildren ( third generation cohort ) have been studied [26–28] . Some participants of each FHS cohort have been genotyped , first on the Affymetrix 100K single nucleotide polymorphism ( SNP ) array and more recently on the Affymetrix 500K SNP array [29] . The FHS participants reside in the same geographic area ( Framingham , Massachusetts ) . Therefore any non-random mating seen in the FHS likely reflects true mate preference and social endogamy , which may also be related to local neighborhood demography . This multigenerational study provides us with the unique opportunity to investigate generational changes in the mating pattern . Here , we analyze all three cohorts of the FHS , to investigate evidence of phenotypic assortative mating between spouses for the physical traits of height , weight , body mass index ( BMI ) and blood pressure . We also assess the FHS original and offspring cohorts for evidence of ancestry-related assortative mating , evaluate the key genetic factors affecting mate selection and assess how the change in mating patterns over time has affected HWE and LD in subsequent generations . For comparison , we also examine the genetic consequences of assortative mating for height , a highly heritable trait . However , height is correlated with European genetic ancestry . We investigate the degree to which the genotypic consequences of spouse assortment for height is primarily due to the effect of the contributing SNPs on height versus their correlation with genetic ancestry in the FHS original , offspring and third generation cohorts . Estimates of kinship between individuals using genome-scale genotype data has become a topic of recent interest . For example , reliably estimating the proportion of complex trait variation that is explained by genome-wide association study ( GWAS ) SNPs has been a challenging problem in the genetics community . The discrepancy between the heritability of the trait estimated from family-based studies and the proportion of the trait variation explained by common SNPs is termed the “missing heritability” [30] . Recent methods for estimating genetic variance explained by common SNPs employ “unrelated” individuals–i . e . those whose kinship is typically less than . 05 [31 , 32] Thus , at the heart of estimating the proportion of trait variation explained by all the common SNPs on a genotyping array is kinship estimation . Spouse kinship is also of interest in assessing the impact of assortative mating patterns . Estimates of kinship are influenced by population stratification , because such estimates using SNP data are influenced by population allele frequencies , which may not apply on an individual level [7 , 33] . Thus , we evaluate and compare spouse kinship estimates before and after adjusting for genetic ancestry .
All participants provided written informed consent for genetic analysis , and the study was approved by the Institutional Review Board of Boston University Medical Center . Third party data were used and therefore no participant consent was required for this study . Original ethical approval is available for review from previously published articles [26 , 27 , 28 , 34] . The participants are from the Framingham Heart Study ( FHS ) original , offspring , and third generation cohorts . The original cohort was ascertained from the population of Framingham , Massachusetts [26] , and subsequent generations of the offspring and the spouses of the offspring ( offspring cohort ) as well as their offspring ( grandchildren of the original cohort ) have also been studied [27 , 28] . These participants have undergone numerous serial clinical examinations and completed several medical questionnaires . The FHS cohorts have been genotyped using the Affymetrix 500K SNP array [29] , and this data was used for the genetic analyses . The FHS cohorts are primarily white and contain a mix of different ethnic white subpopulations characteristic of the greater Boston metropolitan area , including substantial proportions of Northern/Western Europeans ( Irish , English ) , Southern Europeans ( Italians ) , and individuals of Ashkenazi ancestry [29 , 35] . Here spouse-pairs were identified by having at least one offspring in the subsequent generation cohort . If an individual had children with more than one partner then we selected the partner who was genotyped . If there was more than one genotyped partner then we randomly selected a single partner . There are numerous spouse-pairs in the original and offspring cohort , which allows for calculation of spouse-pair ancestry correlations as well as calculation of phenotypic correlations for several traits ( age , height , weight , BMI , systolic blood pressure ( SBP ) , diastolic blood pressure ( DBP ) ) in the original and offspring cohorts . This study includes 8 , 507 participants who had good quality genotypes ( less than 3% missing data ) from the Affymetrix arrays ( described below ) , with 962 ( 124 white spouse-pairs ) from the original cohort ( Gen1 ) ; 3 , 576 ( 755 white spouse-pairs ) from the offspring cohort ( Gen2 ) ; and 3 , 872 from the third generation cohort ( Gen3 ) , plus an additional 97 spouses of the FHS offspring ( Gen2 ) participants who provided DNA but did not have trait values available and hence did not contribute to the correlation analyses . Fig 1 is a flow diagram illustrating exclusion and selection criteria for the FHS participants for each analysis . Genotyping was conducted by the FHS SHARe ( SNP Health Association Resource ) project using the Affymetrix 500K mapping array plus Affymetrix 50K supplemental array . A total of 549 , 781 SNPs were originally available for analysis , but several quality control ( QC ) exclusion measures were implemented to insure the highest quality data were used . The SNPs that were excluded were: SNPs that were duplicates , not available in HapMap or not reliably mapped to the HapMap SNPs ( 118 , 381 ) ; SNPs on the sex chromosomes ( 8 , 725 ) ; SNPs with ambiguous strand information ( A/T or C/G ) ( 68 , 799 ) ; SNPs with minor allele frequency ( MAF ) < 1% ( 5 , 346 ) ; SNPs with HW p-values < 0 . 0001 ( 18 , 241 ) ; and SNPs with genotype call-rates < 99% ( 84 , 099 ) . The 246 , 190 SNPs that passed these quality-control measures were used for further analyses , including the initial principal components ( PC ) analysis used to detect global ancestry . Fig 2 is a flow diagram illustrating the SNP selection and exclusion criteria for each analysis . To interpret the principal components ( PCs ) , we used SNP data from the HapMap [36] and from the Human Genome Diversity Panel ( HGDP ) [37] for comparison . The HGDP individuals have been genotyped with the Illumina 650Y array [38] , whereas the FHS participants had been genotyped with the Affymetrix 500K + 50 K arrays . We identified a common set of 70 , 735 SNPs that were not duplicates and passed the same quality control measures described above in the FHS , the HapMap data and the HGDP data . These common SNPs were used in the PC analyses to characterize European/West Asian ancestry . The HapMap and HGDP data are available in the National Center for Biotechnology Information ( NCBI ) Database of Genotypes and Phenotypes ( dbGAP ) . As mentioned above , the majority of the FHS participants are white . Because our analyses focused on the white participants , we first identified participants with evidence of other ( e . g . African or Asian ) continental genetic ancestry . We selected the maximum number of unrelated FHS participants for this analysis ( 521 ) to ensure the PCs were not confounded or exaggerated by the correlation between relatives . We performed an initial PC analysis of these 521 unrelated FHS participants along with the unrelated individuals from the HapMap ( CEPH individuals with Northern and Western European ancestry ( CEU ) , Yoruba from Ibadan , Nigeria ( YRI ) , Han Chinese from Beijing ( CHB ) and Japanese from Tokyo ( JPT ) ) . The PC analysis was performed using the smartpca program from the EIGENSOFT package [39 , 40] . The remaining FHS participants were projected onto the first two PCs derived from that analysis . There were no outliers detected when computing PCs with HapMap samples , so outlier exclusion was not performed . The results revealed that the first PC clearly separated the HapMap participants with European ancestry ( CEU ) from the participants with African ancestry ( YRI ) , and the second PC clearly separated the participants with Asian ancestry ( CHB+JPT ) from the participants with European ancestry ( CEU ) . The large majority of FHS participants clustered with the CEU individuals suggesting that these individuals had European ancestry . However , there were a few FHS participants seen to cluster between the CEU and the YRI , consistent with African/European admixture as would be seen in African-Americans . There were also a few participants identified who clustered between the HapMap CHB + JPT and CEU individuals , likely representing admixed participants with European and Asian ( or Native American ) ancestries . A total of 27 admixed and/or non-white participants were identified and excluded from the subsequent analyses . A second PC analysis was used to delineate European/Middle Eastern ancestry in the FHS . To help provide geographic orientation to the results , HGDP individuals were projected onto the results of these PC analyses . Because we were investigating the genetic ancestry and phenotypic correlation between spouse-pairs and to eliminate any bias from inclusion of relatives and both members of each spouse-pair in the PC analysis , we selected the maximum number of unrelated individuals ( 339 ) including 333 FHS participants ( 111 from Gen1 , 222 from Gen2 ) and 6 spouses of the FHS participants who provided DNA but were not part of the FHS . When we selected unrelated individuals from the 97 spouses of the FHS participants who provided DNA but were not part of the FHS , only 6 of the spouses were selected . For the other 91 spouses , a relative ( child ) was already selected for inclusion so these 91 spouses were omitted . Again , the PC analysis was performed using smartpca in the software package EIGENSOFT [39 , 40] . There were no outliers identified , so the PC analysis was performed without outlier exclusion . The remaining FHS participants ( including members of spouse-pairs ) were projected onto the first two PCs from this analysis . These first two PCs were used as measures of the FHS participants’ genetic ancestries . Plots of PC2 versus PC1 were generated for each cohort to assess changes over time/generations . Price et al . [39] recommended not filtering SNPs to remove SNPs in high LD with each other , so this PC analysis was based on 70 , 735 markers that were not filtered to remove nearby SNPs with high LD . However , as an additional quality control check , we performed LD pruning and removed 20 , 129 markers with LD r2> . 30 so that among the remaining 50 , 606 markers , no two had an r2 greater than . 30 . The PC scores for the two top PCs derived with the reduced set of markers ( 50 , 606 ) gave virtually identical results to the analysis with the full marker set ( 70 , 735 ) ( correlation , r = . 998 for PC1 and correlation , r = . 978 for PC2 ) . FHS spouse-pair demographics were collected at the time of participant recruitment into the FHS ( baseline exam ) including age , weight , height , BMI , SBP and DBP . The total number of white spouse-pairs in this analysis was 879 , with 124 from the original cohort ( Gen1 ) and 755 from the offspring cohort ( Gen2 ) . No information on spouse-pairs was available from the third generation ( Gen3 ) cohort . Spouse correlations for anthropometric measures ( height , weight , BMI ) , age , SBP and DBP were calculated . The eigenvalues/scores from the first and second PCs from the PC analysis were used as measures of genetic ancestry . Quantitative evidence of ancestry-related assortative mating was obtained by examining spouse correlations for PC scores in the original and offspring cohorts . In addition , scatterplots of PC scores for husbands versus wives for PC1 and PC2 , in both the original cohort and the offspring cohort were assessed to determine trends in the mating pattern . To assess the impact of ancestry-related assortative mating on the genetic structure of the FHS , we calculated the standardized homozygote excess parameter F [9] and the unstandardized linkage disequilibrium ( LD ) parameter D [41] for pairs of unlinked markers across three generations of FHS . For a locus with alleles A and a , the formula for F is given by: F= ( 4NAANaa−NAa2 ) /[ ( 2NAA+NAa ) ( 2Naa+NAa ) ] where NAA , Naa and NAa are the individual counts of genotypes AA , aa and Aa , respectively [8] . F is positive when there is ancestry-related assortative mating and F increases with the ancestry information of the marker . Previously [9] , we considered the situation of an ancestrally admixed population and showed that the expected value of F , E ( F ) for a genetic marker is given by: E ( F ) =σ2ρδ2/p*q* where σ2 is the variance of genetic ancestry ( in this case the proportion of ancestry from one of the progenitor populations ) , ρ is the correlation in ancestry between spouse-pairs , δ2 is the squared marker allele frequency difference between the two ancestral populations and p* and q* = 1-p* are the allele frequencies in the admixed population . This formula shows directly the relationship between F and the squared marker allele frequency difference , which in this case is the relevant measure of strength of association of the marker with ancestry . In fact , defining X as genetic ancestry of an individual and S as the presence of an allele at a marker locus , it can be shown that the correlation of X and S is given by: Correl ( X , S ) =σδ/ ( p*q* ) 1/2 and [Correl ( X , S ) ]2=σ2δ2/p*q* so that E ( F ) =ρ[Correl ( X , S ) ]2 ( 1 ) For the case we consider here , genetic ancestry is continuous and not assumed to represent admixture from ancestral populations . Rather , it is represented by principal components of the genetic marker data . However , there is a direct analogy based on formula ( 1 ) above . Specifically , the correlation of genetic ancestry X and SNP S is directly related to the loading of that SNP on the PC score . Hence , we expect the same linear relationship between F and the square of the PC loading for that SNP . Therefore , to demonstrate the impact of ancestry-related assortative mating , we performed a linear regression analysis , using the F value for each SNP as the dependent variable and the square of the PC loading for that SNP as the independent variable . The F values for the 246 , 190 SNPs were calculated , and the regression of F on the square of the PC loading for each SNP was performed using a random sample of 100 unrelated participants from each generational cohort . These 100 unrelated participants were a subset of the 333 unrelated participants in the original and offspring cohorts used for deriving the PCs from the principal component analysis . The 100 unrelated participants in the third generation cohort were a random unrelated sample from that cohort . We also previously showed [9] that under ancestry-related assortative mating , the expected value of the LD parameter , D , for two markers , is given by: E ( D ) =δφσ2 ( 2 ) where δ and φ are the ancestral allele frequency differences for the two markers . Again define X as an individual’s genetic ancestry , S as the presence of a particular allele at the first marker and T the presence of a particular allele at the second marker . From the derivation of formula ( 1 ) above we can write: δ=Correl ( X , S ) ( p*q* ) 1/2/σ and φ=Correl ( X , T ) ( r*s* ) 1/2/σ where r* and s* = 1-r* are the allele frequencies at the second marker . Formula ( 2 ) can then be rewritten as: E ( D ) =[Correl ( X , S ) ][Correl ( X , T ) ] ( p*q*r*s* ) 1/2 ( 3 ) As we described above , in the case where genetic ancestry X is represented by a principal component , Correl ( X , S ) reflects the PC loading for that marker; hence , we also expect a linear relationship between D and the product of the PC loadings for the pair of markers . Therefore , to demonstrate the effect of assortative mating on LD , we performed a linear regression analysis with D as the dependent variable and the product of PC loadings as the independent variable . For computational efficiency in the regression analysis , a subset of 1000 SNPs was randomly selected , and D values were computed between SNPs on different chromosomes ( N = 469 , 461 ) for the same individuals from each generation . We calculated D values between SNPs on different chromosomes , because these SNPs are unlinked . The D values were calculated and regressed on the product of the loading on PC1 for the first SNP and the loading on PC1 for the second SNP in the random sample of 100 unrelated participants from each generation . The same subset of participants used in the F regression described above was utilized for this analysis . This analysis was repeated regressing D values on the product of the loading on PC2 for the first SNP and the loading on PC2 for the second SNP . Height is a highly heritable trait for which we also observed a significant spouse correlation . Large scale GWA studies of height have identified a substantial number of height quantitative trait locus ( QTL ) SNPs [42] . For comparison to our results for ancestry , we therefore decided to also examine evidence of HWD and LD for these height SNPs . Among 154 genome-wide significant SNPs reported in [42] , 36 were genotyped on the Affymetrix arrays; an additional 118 were successfully imputed ( r2> . 80 ) from HapMap release 22 . For the HWD and LD analyses , for the imputed SNPs we assigned the most likely genotype for each SNP/individual . For regression analyses of F and D , instead of PC-loadings for the independent variables , we used the SNP regression coefficients ( betas ) as reported in [42] as the measure of strength of relationship of the SNP to the trait ( height ) . Specifically , in the analysis of F , we used the squares of the betas , and for the regression on D we used the product of the betas for the two SNPs . For the height SNPs that were genotyped and included in the PC analysis , we also had PC-loadings . However , for the height SNPs that were imputed , we did not . To create “quasi PC-loadings” for these imputed SNPs , we performed a regression analysis of SNP genotype on PC1 and used the regression coefficients . We did the same for the genotyped SNPs for consistency . We evaluated spouse-pair kinship estimates adjusting for genetic ancestry using PC-Relate software [33] and compared these estimates to those obtained without adjusting for genetic ancestry using GCTA software applying the default options with LD pruning at r-squared>0 . 3 and requiring minor allele frequency ( MAF ) > 0 . 05[31 , 32] . Probability density functions for kinship estimates in the original and offspring cohorts were generated separately .
PC analysis based on 339 FHS participants was performed and the remaining FHS and the HGDP individuals were projected onto the first two principal components ( Fig 3A ) . From the HGDP individuals , the first PC approximates the usual northwest-southeast cline in Europe , with Orcadians towards the left of the figure and Italians and Sardinians towards the right . The other prominent feature is a flow of points forming a discrete cluster along PC2 into the upper right corner of the diagram , which from other studies would indicate increasing Ashkenazi ancestry [43–45] . Separating the PC results by generation demonstrates a highly clustered distribution for the first generation ( Fig 3B ) . In addition to the participants clustered along PC2 ( Y-axis ) , there are two separated clusters along PC1 ( X-axis ) . The two major clusters along PC1 likely reflect highly endogamous mating within the Northern/Western and Southern European subgroups . Based on historical demography and recent US Census data of Framingham , the Northern/Western European subgroup is likely comprised primarily of individuals of Irish , English , French and German ancestries , and the Southern European subgroup is likely comprised of primarily Italian ancestry [29 , 46] . The two , almost discrete clusters seen along PC1 for the first generation ( original cohort ) start to merge into a single cluster in the second ( offspring cohort ) ( Fig 3C ) and more strongly in the third ( Fig 3D ) FHS generations , reflecting increased exogamy between Northern/Western and Southern Europeans over time . By contrast , the Y-axis clustering ( PC2 ) appears to remain into the third generation . This suggests a slower rate of exogamy between the Ashkenazi and European subgroups . In Table 1 we provide summary statistics for the various traits considered , including age , height , weight , BMI , SBP , and DBP . The age at enrollment of the offspring cohort was comparable to the original cohort because recruitment occurred several decades after enrollment of the original cohort . The offspring were slightly taller ( P<0 . 05 ) and heavier ( P<0 . 05 ) at entry into the cohort than their parents . The offspring had higher average BMI ( P<0 . 05 ) and in particular the male offspring had higher BMI than their counterparts in the original cohort ( P<0 . 05 ) . SBP and DBP were both significantly lower in the offspring cohort ( females and males ) . This may reflect the fact that 3 . 3% of the offspring cohort participants were on hypertension medication at the baseline exam , while none of the original cohort participants were treated at the baseline examination . Interclass correlations for spouse-pairs calculated separately for the first two generations ( N = 124 and 755 , respectively ) were generated for age , the various anthropometric , and blood pressure measures along with the PC scores ( Table 2 ) . As expected , spouse correlations for age are high in both generations , but lower in the parent generation ( r = 0 . 64 , P<0 . 001 ) than the offspring generation ( r = 0 . 91 , P<0 . 001 ) . Also as expected , there are moderate correlations for height between spouses , higher in the original cohort ( r = 0 . 45 ( P<0 . 001 ) than in the offspring cohort ( r = 0 . 27 , P<0 . 001 ) . Correlations for weight are more modest , but still significant and again higher in the original cohort ( r = 0 . 28 , P = 0 . 002 ) than the offspring cohort ( r = 0 . 16 , P<0 . 001 ) . Correlations for the two blood pressure measures were not different from 0 in the original cohort ( r = -0 . 06 , P = 0 . 54 for SBP and r = -0 . 03 , P = 0 . 75 for DBP ) , but slightly positive and significant in the offspring cohort ( r = 0 . 16 , P<0 . 001 for SBP and r = 0 . 14 , P<0 . 001 for DBP ) . On the other hand , the spouse correlations for ancestry , as represented by PC1 and PC2 , are strikingly positive and significant , especially in the first FHS generation , with values of r = 0 . 73 ( P<10−21 ) and r = 0 . 80 ( P<10−27 ) for PC1 and PC2 , respectively . The pattern is similar in the offspring generation , although the correlations are attenuated compared to the prior generation , r = 0 . 38 ( P = 10−27 ) for PC1 and r = 0 . 45 ( P = 10−38 ) for PC2 . In both generations of the FHS , the ancestry correlation of spouses far exceeds the correlations observed for the anthropometric and clinical traits , indicating that ancestry has been the most significant factor related to spouse choice apart from age . The reduced spouse ancestry correlations in the offspring generation reflect decreased endogamy over time . Fig 4 illustrates the ancestry-related assortative mating that has been present in Framingham over several generations . Each sub-figure is a scatter plot of PC scores for husbands versus wives for either PC1 or PC2 , for the original cohort or the offspring cohort . The high correlation of PC scores between spouses as well as the clustering is quite apparent in the original cohort for both PC1 ( Fig 4A ) and PC2 ( Fig 4B ) . In the original cohort , the clustering of spouses into Northwestern and Southern European subgroups is quite strong , with only a few spouse-pairings occurring between the two subgroups ( with somewhat more unions between Southern European men and Northwestern European women than the opposite ) . The scatter diagram for PC2 in the original cohort also reveals strong endogamy , with few , if any , unions between clusters . The second set of plots in Fig 4 , characterizing the spouse-pairs of the offspring cohort , reveals similar patterns to the original cohort , but with less striking clustering , in particular for PC1 . While again we see evidence of assortative mating based on ancestry , we also see a higher rate of exogamy compared to the original cohort , with a greater number of unions between Northwestern and Southern Europeans ( Fig 4C ) . As in the top corresponding figure , we again see a higher rate of unions between Southern European men and Northwestern European women than the opposite ( Fig 4C ) . Fig 4D , depicting PC2 , once again illustrates a relatively higher rate of endogamy for the Ashkenazim . As expected , the ancestry-related positive assortative mating resulted in both within-locus ( HWD due to homozygote excess ) and between-locus ( LD ) observable effects . We examined HW deviations in each of the three generations of the FHS by calculating the standardized homozygote excess measured by F for each SNP , and performed regression analysis with F as the dependent variable and squared PC loading as the independent variable ( Table 3 ) . In all three generations , the regressions are significant , both for PC1 and PC2 . As expected , the magnitude of the regression coefficient and strength of the association is greatest for the participants in the original cohort and decreases continuously in the second ( offspring cohort ) and third generations ( Table 3 ) , although the decrease is much more dramatic between the last two generations . This result is consistent with the reduction of endogamous mating over the past 60 years in Framingham as we observed in the spouse PC correlations in the first two ( original and offspring ) FHS cohorts . HW deviations reflect assortative mating patterns from the previous generation . Hence , the high level of HW deviation observed in the original cohort suggests that spouse ancestry correlations were even stronger in the generation that gave rise to the original cohort ( parents of the original cohort participants ) , although a more rapid decline in endogamy is inferred to have occurred between the original and offspring generations than in earlier generations . Similarly , we examined the LD parameter D for pairs of unlinked SNPs as a function of the product of PC SNP loadings via regression analysis ( Table 4 ) . Again we see a pattern very consistent with those in Table 3 for HW deviations . With the exception of PC2 in generation 3 , the regressions are significant across each of the 3 generations of the FHS , with attenuation of the regression coefficients in each succeeding generation , particularly between the second and third generations . As opposed to HW deviations , which depend only on mating patterns of the prior generation , LD can persist over many generations , depending on the linkage relationship between markers , even under random mating . Nonetheless , we did see a reduction of the regression of the LD parameter D on the product of the PC loadings across generations for both PC1 and PC2 , again consistent with the relaxation of endogamy over the same time period . Height is a trait with high heritability and also showed significant spouse correlations in our sample ( Table 2: r = 0 . 45 and r = 0 . 27 in original and offspring generations , respectively ) . Direct assessment of HWD and LD for the 154 height SNPs was complicated by the fact that height is also correlated with genetic ancestry ( PC1 ) . Specifically , we calculated a weighted genetic risk score ( GRS ) from the 154 SNPs based on the betas ( regression coefficients ) as reported in [42] . Measured height is positively correlated with PC1 ( r = 0 . 21 and 0 . 11 in the original and offspring cohorts , respectively ) ; the height GRS shows even higher correlations with PC1 ( r = 0 . 29 and 0 . 25 in the original and offspring cohorts , respectively ) . This means that at least some of the height SNPs could show HWD and/or LD due to their correlation with ancestry ( PC1 ) . Therefore , we constructed a regression model with F as the dependent variable and both height beta-squared and quasi PC-loading squared included as independent variables . Similarly , we included products of both betas for each pair of SNPs and products of quasi PC1-loading for each SNP pair as independent variables in the regression model for D . For the regression model on F , neither the beta-squared nor quasi PC1-loading-squared terms are statistically significant . This is not surprising given the limited number of SNPs for this analysis and the fact that assortative mating has less impact on HWD than on LD . By contrast , there is a highly significant effect of the quasi PC1-loading products on D ( t = 11 . 63 , P<2 . 2x10-16; t = 5 . 21 , P = 1 . 9x10-7; t = 4 . 23 , P = 2 . 3x10-5 for original , offspring and third generation cohorts , respectively ) ; however there is no significant effect of the beta products ( t = 0 . 091 , P = 0 . 93; t = -0 . 019 , P = 0 . 98; t = 0 . 058 , P = 0 . 95 for the three cohorts , respectively ) . Thus , the impact of assortative mating for height on population structure ( i . e . LD ) appears to be dramatically less than assortative mating for ancestry . We first compared the spouse kinship coefficients estimated using GCTA ( without ancestry adjustment ) to those using PC-Relate ( with ancestry adjustment ) . We categorized individuals into three groups based on scores for PC1 and PC2 –those representing an Ashkenazi cluster , those representing a Northwestern European cluster , and those representing a Southern European cluster . It is important to note that there are individuals who are mixed between these groups , so to some degree the clustering is imprecise; however it is still useful to demonstrate the impact of ancestry adjustment . A clear pattern of difference between ancestry adjusted and ancestry unadjusted kinship estimates is observed ( Fig 5 ) , both in the FHS original cohort and offspring cohort . While the mean kinship varies little among the various ancestry groupings , the mean kinship is clearly increased for the ancestry endogamous pairs–with the highest mean for the Ashkenazi pairs , followed by the Southern European pairs; the Northwestern European pairs appear to have the smallest increase , likely because the majority of the sample is of Northwestern European ancestry . The figure clearly shows the upward bias in estimation of kinship when ancestry is not taken into account . These results are also reflected in histograms of kinship coefficients for the original and offspring cohorts ( Fig 6 ) . The unadjusted kinship estimates show a large right tail , likely due to the endogamy seen in the Ashkenazim in both the original and offspring cohorts , as observed in Fig 5 . In both cohorts , mean kinship estimates as well their standard deviations ( SD ) are higher when not adjusting for genetic ancestry ( original cohort: mean = 0 . 001653 , SD = 0 . 00470; offspring cohort: mean = 0 . 00070 , SD = 0 . 00367 ) than when adjusting for genetic ancestry ( original cohort: mean = 0 . 00033 , SD = 0 . 00286; offspring cohort: mean = -0 . 0001 , SD = 0 . 00241 ) . For the original and offspring cohorts , the ancestry unadjusted means are significantly greater than 0 ( P<10−4 and P<10−7 , respectively ) , while the ancestry adjusted means are not significantly greater than 0 .
We have conducted the first multi-generational analysis of mating patterns for a U . S . population , based on FHS participants from Framingham , Massachusetts . Our findings have significance for a variety of reasons: they reflect demographic patterns that have occurred over the past 60 years and reveal changes in those patterns over time; document the genetic implications of mating patterns and changes in mating patterns over time; and provide a warning about potentially simplistic assumptions in the genetic modeling of human populations . Soon after World War II , when the FHS was initiated , the demography of Framingham was largely white and middle class , with recent growth from the baby boom . It largely reflects the ancestral make-up of Massachusetts , which is both Northern and Southern European . According to the U . S . Census Bureau 2010 American Community Survey , among whites , the predominant ancestries in Framingham are: Irish ( 23% ) , English ( 13% ) , French ( 10% ) , other Northern European ( 27% ) , Italian ( 18% ) , and other Southern European/Mediterranean ( 8% ) . Similarly , in the FHS , the ancestries are predominantly Irish ( 15% ) or other British ( 20% ) , Italian ( 19% ) , and other Western European ( 32% ) [29] . There is also a sizeable Ashkenazi population in Framingham constituting perhaps 5% of the total population [35] . Mate choice reflects a large number of factors including local geodemographics , social class , nationality , ethnicity , religion , anthropometric traits such as height and weight , as well as behavioral characteristics . Our findings clearly document the strong endogamy that existed in Framingham prior to World War II . These patterns may also have reflected neighborhood characteristics , and the tendency for unions to occur locally . Intermixing between participants with Northwestern and Southern European ancestries was relatively uncommon in the original cohort , but increased in subsequent generations . Similarly , there was a strong tendency towards endogamy in the participants that clustered along PC2 , implicitly those with Ashkenazi ancestry . The strong endogamy was also clearly evident in the large deviations from HW of SNPs correlated with either of the first two PCs , as well as the dramatic evidence of LD between unlinked markers correlated with those same PCs . It is the strength of the regressions of F and D on these PC loadings that demonstrates directly the impact of the ancestry-related assortative mating , as other factors creating HW deviations and LD ( e . g . inbreeding and genotyping error ) would not produce such a pattern [10] . The spouse genetic ancestry correlations observed in the original cohort were remarkably high , even greater than spouse correlation in age . This assortment was also evident in the strong HW deviations and LD observed in the subsequent generation ( offspring cohort ) . In fact , the regression analyses on HW and LD parameters revealed evidence of even stronger spouse ancestry correlations in the generation that gave rise to the original Framingham cohort . We also observed a consistent decay in endogamy in Framingham over 6 decades , reflected both in the spouse correlations for genetic ancestry and in the HW and LD deviations . While strong clustering was observed in the original FHS cohort , one can see a clear decay in such clustering in subsequent generations , and also an increase in the number of unions between the clusters in the offspring cohort compared to the original cohort . Although examination of genetic ancestry correlation in spouse-pairs provides direct evidence of assortative mating , indirect evidence can also be obtained by examination of HW and LD patterns of SNPs as a function of their correlation with genetic ancestry , as we have also demonstrated here and previously [9 , 10] . Indeed , for large population cohorts where spouses have not been identified , one can examine both the evidence of population structure by PC analysis and correlation of HW disequilibrium and LD statistics with ancestry-informative SNPs as a function of age . As we have seen in the Framingham population , genetic structure can vary significantly by age as a consequence of changes in social demography and mating patterns . Also , we have only analyzed LD patterns for pairs of SNPs located on different chromosomes . Because LD persists over time , especially when SNPs are linked , we anticipate considerable residual LD for many nearby linked markers as a consequence of strong assortative mating in the past . Although we have provided clear evidence of HW deviations and LD in the FHS , it is still important to note that the strength of such deviations depends on the degree of allelic differentiation ( for example , as measured by Wright’s fixation index FST ) among the ancestral groups contributing to the current population . Because the large majority of white ethnic groups have only modest FST values on average , it is expected that most SNPs typed in a sample similar to the FHS would not demonstrate significant homozygote excess , especially in a sample that is not large . However , for some SNPs , i . e . those with greater European differentiation , the spouse genotype correlation and HW deviation can be substantial . Genome-wide association studies , while often correcting for the inclusion of close relatives , generally do not correct for inclusion of spouses . Close relatives may be correlated for genotypes across the entire genome . By contrast , the genotype correlation between spouses may be proportional to how correlated a marker is with genetic ancestry , and therefore marker specific . So , for ancestry-related SNPs , ignoring the genotype dependence between spouses could also lead to inflated statistical significance , as it does when closely related individuals are included . The power of the ancestry-related assortative mating analysis presented here is due to the strong relationship of some SNPs with genetic ancestry in Europeans ( i . e . those with the strongest impact on the PC scores ) . By contrast , a similar analysis of SNPs underlying a phenotypic trait undergoing assortative mating would not typically show such evidence of HWD and LD because the individual SNP effects on the trait are too modest . For example , we also examined height , a highly heritable trait with a significant spouse correlation , in the same fashion , but found no evidence of correlation of the LD measure D among the pairs of 154 height-associated SNPs with the product of their regression coefficients on height , but did find a significant correlation of LD for the same pairs of SNPs with their PC1-loading products . On the other hand , genotyping error can also have a strong influence on HWD ( but not LD ) . For example , in our original SNP QC analysis , we examined the proportion of SNPs with a significant HW test ( P < . 0001 ) as a function of call rate . For SNPs with call rate of at least 99% , 0 . 2% had a significant HW test . For SNPs with call rate of 98–99% , 1 . 2% had a significant HW test . For SNPs with call rate <97% , 5 . 9% had a significant HW test . However , the HWD of these SNPs is unlikely to be correlated with PC1 or PC2 . We also demonstrated that ancestry adjustment can influence kinship estimates , particularly those for pairs of individuals less than close ( e . g . third degree ) relatives . Because modern heritability estimation approaches depend on kinship estimates in this range , caution needs to be exerted in interpreting those heritability estimates if ancestry is not considered . In other words , heritability estimates obtained using kinship estimates without ancestry adjustment may be biased , depending on the degree to which genetic ancestry is related to the trait being examined . A remaining and important question is the degree to which our observations in Framingham generalize to other populations , both within the U . S . and elsewhere . The issue becomes an even greater concern as populations from different geographic locations are combined in meta-analyses . The 2010 US census collected ancestry information , and maps created from that information reveal patterns of national diversity as well as changes over time due to migrations . While unions historically have been preferentially local , increased movements of the population over past decades are contributing to the decay of local endogamy , as seen in Framingham . In addition to geography , which itself is correlated with ancestry , other social factors may have influenced mate selection such as race/ethnicity , nationality , education , socioeconomic status and religion . While the exact patterns we have seen in the FHS may not replicate elsewhere , it is likely that similarly complex mating patterns , including spouse correlations related to genetic ancestry , are a common feature , and need to be addressed in any population study .
|
We analyzed three generations of whites from the Framingham Heart Study ( FHS ) using genome-wide genotype data to characterize their genetic ancestry . By examination of spouse-pairs , we observed that individuals of Northern/Western European , Southern European and Ashkenazi ancestry preferentially chose spouses of the same ancestry , however , the degree of endogamy decreased in each successive generation , especially between Northern/Western and Southern Europeans . We then showed that the mating pattern results in Hardy-Weinberg disequilibrium ( HWD ) at ancestrally-informative SNPs , and also results in linkage disequilibrium ( LD ) between unlinked loci . The HWD and LD decrease as theoretically expected with the decrease in endogamy noted in each generation . In the FHS sample , spouse genetic similarity can be explained by ancestry-related assortative mating .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"genome-wide",
"association",
"studies",
"medicine",
"and",
"health",
"sciences",
"variant",
"genotypes",
"population",
"genetics",
"genetic",
"mapping",
"regression",
"analysis",
"linkage",
"disequilibrium",
"mathematics",
"statistics",
"(mathematics)",
"genome",
"analysis",
"population",
"biology",
"research",
"and",
"analysis",
"methods",
"mathematical",
"and",
"statistical",
"techniques",
"statistical",
"methods",
"blood",
"pressure",
"genetic",
"loci",
"phenotypes",
"heredity",
"genetics",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"genomics",
"evolutionary",
"biology",
"vascular",
"medicine",
"computational",
"biology",
"human",
"genetics"
] |
2017
|
Structured mating: Patterns and implications
|
We consider the problem of how to detect cognate pairs of proteins that bind when each belongs to a large family of paralogs . To illustrate the problem , we have undertaken a genomewide analysis of interactions of members of the PE and PPE protein families of Mycobacterium tuberculosis . Our computational method uses structural information , operon organization , and protein coevolution to infer the interaction of PE and PPE proteins . Some 289 PE/PPE complexes were predicted out of a possible 5 , 590 PE/PPE pairs genomewide . Thirty-five of these predicted complexes were also found to have correlated mRNA expression , providing additional evidence for these interactions . We show that our method is applicable to other protein families , by analyzing interactions of the Esx family of proteins . Our resulting set of predictions is a starting point for genomewide experimental interaction screens of the PE and PPE families , and our method may be generally useful for detecting interactions of proteins within families having many paralogs .
The PE and PPE gene families in Mtb make up nearly 10% of the bacterium's coding DNA [2] . The two families combined have about 150 members , amounting to 4% of the open reading frames ( ORFs ) in Mtb . The PE and PPE gene families account for much of the genomic difference between Mtb and other ( nonpathogenic ) mycobacterial genomes [3] , [4] . Therefore it is thought that they may have a role in Mtb's virulence and host-specificity . A subset of PE proteins is displayed on the bacterium's cell surface [5] , can elicit an immune response [6] , and may be a source of antigenic diversity for Mtb [7] . PPE proteins have also been found on the cell surface [8] , [9] , may be secreted [10] , and can confer virulence [11] . These studies indicate the likely importance of the PE and PPE gene families in pathogenesis . More extensive characterization of their function , interactions , and roles in infection are therefore important areas for investigation . Genome analysis suggests that the PE and PPE families are functionally linked [12]–[14] . Pairs of PE and PPE genes are frequently found adjacent on the Mtb genome , and the structure of a complex of one such PE/PPE protein pair was recently characterized [13] . These results indicate that there may be many other instances of interactions between PE and PPE proteins . However , with only one complex characterized so far , it remains unclear which specific members of the two families interact . The 87 PE and 65 PPE proteins ( depending on similarity threshold ) in the Mtb H37Rv genome generate ∼6 , 000 possible pairwise combinations . It may be that dozens of biologically relevant PE/PPE complexes remain to be characterized . Because the PE and PPE families can interact with the host immune system [5] , [6] , [11] , combinatorial formation of complexes might enable immune evasion during tuberculosis infection . Mapping the PE/PPE interaction network is therefore of critical importance for accelerating drug discovery . Because PE and PPE proteins are difficult to express and purify experimentally [13] , new computational methods are needed to detect likely PE/PPE complexes and efficiently prioritize experiments . Perhaps the most straightforward bioinformatic approach for detecting PE/PPE complexes is to simply predict interaction of the PE/PPE pairs found in the same operon [15]–[18] . Some 14 pairs of PE and PPE genes , including the one complex that has been structurally characterized to date [13] , are found adjacent on the genome , in the same orientation , with minimal intergenic distance , and with the PE 5′ to ( upstream of ) the PPE ( the PE proteins in such pairs do not include any of the repeat-containing PE_PGRS proteins ) . Because of this recurring genome organization motif , such pairs are likely expressed in the same operon [19] . However , these same-operon PE/PPE pairs comprise less than 10% of the total number of PE and PPE genes in Mtb . The majority of PE and PPE proteins are found unpaired in the genome , and it is possible that some of these interact despite not having genomic proximity . Computationally detecting PE/PPE complexes not found by the operon method is therefore an important challenge . The tendency for proteins to coevolve with their interaction partners has been described [20]–[23] , and bioinformatic methods to detect protein coevolution have therefore been proposed for predicting protein interactions [24] . The idea is to exploit the correlation of the phylogenetic distance matrices of two protein families whose members are known from experiments to interact . Known interacting proteins tend to be found at analogous regions of their respective phylogenetic trees [20] , [24] , [25] ( which can also be represented as distance matrices ) . Such methods can accurately pair ligands with receptors [24] , and could potentially be used to infer interactions between the PE and PPE families . However , a difficulty of applying these methods in our case is that benchmarking predictions requires a set of experimentally determined interactions , and currently only a single known example of a PE/PPE complex exists [13] . Our challenge , therefore , for the computational prediction of PE/PPE interactions is the evaluation of predictions given the currently limited number of known PE/PPE interactions . We combined the operon method , coevolution analysis , and structural knowledge of interacting domains to develop a coevolution-based strategy to predict PE/PPE complexes in the Mtb H37Rv strain . Some 289 predicted complexes resulted from the application of our method . To validate the predictions , we used several published mRNA expression datasets from Mtb to assess PE/PPE coexpression in vivo . A significant overlap was seen between coevolved and coexpressed PE/PPE gene pairs , supporting the coevolution-based predictions , and resulting in a high-confidence list of possible complexes . To demonstrate the extensibility of our method to other protein families , we performed a similar analysis of interactions of the ESAT-6/CFP-10 ( Esx ) family of proteins . Our results are a starting point for experimental genomewide screens of PE/PPE and Esx complexes , and our method may be applicable to other functionally linked protein families in Mtb and other microbial pathogens .
We assumed that each interacting pair of PE/PPE proteins must have complementary interfaces , and that the residues in these interfaces may coevolve due to positive selective pressure on the interaction . Although we currently do not have sufficient data from PE/PPE complexes to accurately predict residue-residue interactions from sequence using correlated mutations analysis [26]–[29] , we can delineate the likely interacting regions by their similarity to the structurally characterized PE/PPE interacting domains [13] . We assumed that PE/PPE gene pairs adjacent on the genome , and in the same orientation , are in expression operons , as has been shown for Rv2431c/Rv2430c [13] . The components of protein complexes and metabolic pathways in prokaryotes are often located together on the genome in operons [19] . These operons are transcribed as a single , polycistronic mRNA . Genes located on an operon usually function together , and often form protein complexes . We predict thirteen other PE/PPE gene pairs lie in operons ( Figure 1A ) based on their short intergenic distance ( <100 bp ) and same transcription direction . These pairs have a high degree of coexpression ( average mRNA correlation 0 . 59 for operon-paired , 0 . 05 for genomewide PE/PPE gene pairs , see Materials and methods ) , suggesting that these PE/PPE pairs are indeed in operons . Finally , we assumed that PE/PPE pairs in operons are likely to interact in a manner similar to the structurally characterized , operon-coded , PE/PPE complex of Rv2431c/Rv2430c [13] . To support our assumption that bacterial operons tend to code protein complexes , we analyzed the tendency for annotated E . coli protein complexes to reside in operons in the EcoCyc database [30] . We extracted 280 complexes , involving 692 proteins , from EcoCyc . We asked what fraction of protein pairs found in complexes also had their genes in the same operon , and found this to be 49% ( 942 protein pairs in operons out of 1918 protein pairs in complexes ) . To assess the significance of this result , we shuffled the identity of the genes in complexes by replacing each with a random E . coli gene , and re-assessing the overlap . One thousand shufflings were performed and an overlap of 49% was never achieved; in fact , the highest overlap obtained was 2% . We conclude there is a significant tendency for bacterial protein complexes to be coded in the same operon . While this does not guarantee that proteins coded in operons interact , given a known example of an operon-coded PE/PPE complex , we might expect PE/PPE pairs similarly organized in operons to interact . Figure 1 illustrates our method for detecting pairs of coevolved PE and PPE genes ( and thus , possible interacting proteins ) . Figure 1A shows all PE and PPE gene pairs that lie in the same orientation of 5′ PE → PPE 3′ with no more than 100 bp separation between the PE and PPE genes . These PE/PPE pairs are likely within the same operon [15]–[18] , and are summarized in Table 1 . We refer to these as the ‘operon pairs’; they form the training data for our method . PE and PPE protein sequences coded by the operon pairs are aligned to the sequence of the appropriate subunit of the PE/PPE complex of Rv2431c/Rv2430c ( Figure 1B ) . Next , the structure-based multiple alignment is used to generate phylogenetic distance matrices , which contain pairwise protein similarity relationships ( Figure 1C ) . Notice that each equivalent row in the matrix is an operon-paired ( and hence assumed interacting ) PE/PPE pair . These are called the ‘reference matrices’ . For all ( operon-paired or otherwise ) PE and PPE protein sequences in the Mtb genome , distance vectors to the reference matrices are generated ( Figure 1D ) . The correlation between these vectors , Cij , is a measure of the PE/PPE pair's possible coevolution . Next , Cij scores are further processed ( Figure 1E ) to yield Sij , the paralog matching score for the predicted complex of PEi and PPEj . The probable interacting regions of all PE and PPE proteins in the Mtb genome were delineated . This was done using the ClustalW program [31] to perform a multiple sequence alignment of each protein family to a secondary structure profile derived using the DSSP program [32] on the appropriate subunit of the known PE/PPE complex structure . The secondary structural alignment was motivated by the observation that the PE and PPE proteins of known structure are composed of long α-helices interspersed with turns and loops , and our intuition that insertions and deletions would preferentially occur in regions outside the helices . The alignment was visually inspected to remove outlying or poorly-aligned sequences . All remaining PE and PPE sequences in the alignment were truncated to eliminate regions not aligned to the structure . In many cases both the PE and PPE proteins contained additional domains in their C-terminal regions , including the PGRS repeats in PE_PGRS proteins . All subsequent sequence analysis in this work was performed with these truncated sequences . We reasoned that limiting our analysis to homologous interacting domains would facilitate detection of coevolution relevant to protein interaction , and would therefore not be confused by spurious coevolution signals from regions not involved in PE/PPE interface . This additional truncation step was performed because we observed that most PPE , and some PE , domains have additional domains , low complexity regions , or membrane helices C-terminal to the conserved interacting domains . The resulting alignments are provided in the Supporting Information ( Datasets S1 and S2 ) . Phylogenetic distance matrices for the subset of PE and PPE proteins linked in the 14 operons ( Table 1 ) were constructed using the ClustalW program [31] . For each of the PE and PPE families , the 14 sequences in operon pairs were manually extracted from the full-family alignment . The 14-sequence subalignments were then loaded into ClustalW to generate 14×14 distance matrices . Phylogenetic distance matrices represent the pairwise distance between protein sequences . In ClustalW , pairwise distances between sequences are measured by the fraction of mismatches in ungapped positions of an alignment of two sequences . If our assumption that operon-paired PE/PPE genes code complexes were correct , we reasoned that there would be a correlation of the two distance matrices when the genes of both matrices are respectively ordered by the genomic position of the operon in which they occur ( Figure 1C ) . Such a correlation between matrices would be consistent with previous analyses demonstrating correlation of distance matrices for known interacting proteins [20] . We indeed found that the PE and PPE matrices were correlated , with a Pearson correlation coefficient of 0 . 84 . To assess the significance of the correlation of the PE and PPE matrices , we performed random shuffling of the matrices' gene order , thus removing any mapping of paired genomic position between the matrices . One million shuffling steps were performed , and the frequency with which the shuffled correlation exceeded the correlation from the operon-ordered matrices was recorded . The correlation of 0 . 84 was never exceeded in 106 matrix shuffling steps ( the maximum correlation in any shuffling was 0 . 20 ) . These results suggest that the PE and PPE matrices , ordered by operon position , may represent an optimal pairing of PE/PPE proteins , and , in light of previous findings of correlated distance matrices of interacting proteins [20] , support the hypothesis that PE/PPE operons code complexes . The correlation of the operon-ordered distance matrices can be visualized using phylogenetic trees to provide an intuitive feeling for the results . To illustrate this , we generated trees from distance matrices in the ClustalW program ( Figure 2 ) . In the two trees , operon-paired PE and PPE proteins are in the same-colored shaded region , illustrating similar topologies of the trees . This qualitative tree similarity illustrates the notion of coevolution of the PE and PPE families . We next used the correlated 14×14 PE and PPE distance matrices as reference matrices to evaluate pairwise correlations between the 86 PE proteins and 65 PPE proteins in the Mtb genome , excluding those present in operon pairs . This was done by generating a distance vector of length 14 for each protein in a PE/PPE pair . The vector contained the distance between the protein being tested and the 14 members of the appropriate reference matrix ( PE or PPE ) . The Pearson correlation for the two vectors was calculated to obtain a measure of the coevolution of the test PE/PPE pair ( Figure 1E ) . Notice that here we are taking the correlation of two vectors of length 14 , as opposed to our earlier calculation of the correlation of the two 14×14 matrices . The coevolution of 5590 PE/PPE pairs was evaluated using this approach . We define the coefficient Cij , as the correlation which measures the coevolution of PEi with PPEj . Pairwise correlations between all PEi and PPEj ( Cij ) were further processed using a reciprocal ranking procedure , to produce a predictive paralog matching score , Sij . This was done because we noticed that many PE/PPE pairs had high Cij values ( average Cij = 0 . 54 ) . The distribution of Cij is shown in Figure 3A ( blue bars ) . From the histogram , it is clear that a great number of PE/PPE pairs have a high Cij . This may reflect the overall coevolution of the two families , but is not of use in a prediction scheme , as nearly all pairs have a high score . Such a result is inconsistent with our intuition that in a large collection of proteins , only a relatively small number of the possible pairs should interact . Further , we found that the Cij distributions for operon pairs and all PE/PPE pairs do not differ significantly ( Kolmogorov–Smirnoff ( KS ) test , α = 0 . 05 , k = 0 . 29 , P = 0 . 16 ) . In the reciprocal ranking procedure , the predicted complex of PEi and PPEj was assigned a high Sij only if PEi and PPEj were mutually at the top of each protein's list of interaction partners ranked by Cij . In other words , PEi and PPEj were required to be reciprocally the most coevolved partners in order to get a high Sij ( see Materials and methods ) . Figure 3A , shows the distribution of Sij scores ( red bars ) . The distribution of Sij suggests it is a more useful measure than Cij for complex prediction , as the bulk of the Sij scores are low ( reflecting that in a large dataset , most protein pairs do not form complexes ) . The operon pairs have a significantly higher Sij than PE/PPE pairs overall ( KS test , α = 0 . 05 , k = 0 . 97 , P–value< . 0001 ) , a result illustrated in Figure 3B . We conclude that the reciprocally ranked coevolution score , Sij , performs better than Cij for predicting protein interactions . To evaluate the predictive value of the two pairwise PE/PPE scores , Cij and Sij , we assessed recovery of the 14 operon linked PE/PPE pairs used to generate the reference matrices . Of the 14 pairs , all were given Sij scores in the top 5% implying the method could be used to detect complexes with reasonably high accuracy . In contrast , only 3/14 ( 20% ) of the operon pairs were given Cij scores in the top 5% . A mere 8/14 pairs ( 60% ) had Cij scores above the median , implying poor recovery by the raw distance vector correlations . The relationship between Cij and Sij is illustrated in Figure 3B . The distribution of scores shows that Sij as a predictor of PE/PPE complexes gives much better recovery of operon-paired PE/PPE proteins than Cij , and therefore is likely a better indicator of interaction . To further evaluate prediction accuracy , and to determine a prediction score threshold , we compared the sensitivity ( also called true positive rate or TPR ) and 1-specificity ( also called the false positive rate or FPR ) for Sij and Cij . Sets of positive and negative interactions were defined , and an Sij threshold of 0 . 75 was found to capture the best balance between sensitivity and specificity ( Materials and methods ) . Applying a prediction criterion of Sij≥0 . 75 gave 289 PE/PPE pairs or roughly 5% of the possible 5 , 590 PE/PPE pairs genomewide . We therefore proceeded with our analysis by taking the predictions with Sij scores in the top 5% . To see if the coevolution-based predictions were biologically sensible , we analyzed correlations in mRNA expression ( which we call ‘coexpression’ ) of possible interacting PE/PPE pairs . We reasoned that interacting proteins would tend to be expressed at similar times to perform their biological functions together [33]–[37] , and that if some of our predicted interactions were correct , we should see non-random enrichment in coexpression of the genes encoding the predicted complexes . Gene microarray data from Mtb was compiled from nine published datasets covering a broad range of experimental conditions ( Table S1 ) in the Gene Expression Omnibus ( GEO ) database [38] . Vectors of expression values for each PE or PPE gene were generated , and used to derive a correlation , Rij , for the expression of each PE/PPE gene pair ( see Materials and methods ) . To confirm our intuition that predictions of PE/PPE coexpression and coevolution should overlap significantly , we analyzed the distribution of the two coevolution scores ( Cij and Sij ) combined with the coexpression score , Rij . The resulting distributions of the two score combinations are shown in Figure 4 . In the figure , coevolution ( Cij or Sij ) is shown on the x axes; coexpression ( Rij ) is shown on the y axes . Red dots represent the 14 operon pairs; green dots represent the 182 inter-operon pairs ( in which the PE and PPE are from different operon pairs ) ; blue dots represent the other 5 , 394 genomewide PE/PPE pairs . Dashed lines are drawn to represent the upper 5% threshold for each score . Figure 4A illustrates our earlier assessment that Cij is not a useful prediction score due to the operon pairs not having a significantly different distribution from all PE/PPE pairs . Also notable in Figure 4A , only a minority of the operon pairs is found in the top 5% by both methods ( 3/14 operon pairs or 21% ) . Figure 4B shows better recovery of operon pairs in the top 5% by both methods ( 12/14 operon pairs or 86% ) . In light of these results , we concluded that the paralog matching score , Sij , is superior to Cij for predicting PE/PPE complexes . We therefore chose to combine Sij and Rij for subsequent predictions , and Cij was not used further in this study . The Sij and Rij scores for all PE/PPE pairs are provided in the Supporting Information ( Dataset S3 ) . To assess the statistical significance of the overlap between predictions from coevolution and coexpression , we again employed a KS test . We asked whether the coexpression values of the top 5% coevolved PE/PPE genes ( excluding the operon pairs ) were higher than coexpression values of PE/PPE gene pairs overall . We found this to be the case ( KS test , P = 0 . 02 , α = 0 . 05 , k = 0 . 09 ) . From this we conclude that the PE and PPE proteins we predict to interact tend to be coexpressed , which we take as additional evidence for their possible interaction . To assess the specificity of interaction in PE/PPE operon pairs , we analyzed coevolution and coexpression of inter-operon pairs , the PE/PPE pairs in which both proteins are from different operon pairs . A KS test showed that inter-operon pairs ( Figure 4 , green crosses ) had significantly lower score distributions than all other pairs by both coevolution scores , Cij and Sij , and the coexpression score , Rij ( P≪0 . 0001 in all tests ) . We noticed a bimodal distribution of Cij ( Figure 4A ) . Using a k-means algorithm we identified two clusters: a larger ( 4 , 208 protein pairs ) , positive-valued cluster with mean Cij = 0 . 78 and a smaller ( 1 , 382 pairs ) , negative-valued cluster with mean Cij = −0 . 21 . The negative cluster contained no operon pairs , and was more than twice as likely to contain inter-operon pairs than the higher group ( 7% of the negative cluster; 3% of the positive one ) . We interpret these results as evidence of negative selection , both at the amino acid and gene expression levels , against cross-reactivity of operon paired PE and PPE proteins , and conclude that PE/PPE operon pairs , in general , interact specifically . To demonstrate that our method is extensible to protein families other than PE and PPE we studied the ESAT-6/CFP-10 ( Esx ) family of proteins , which include some secreted antigens [39] . We chose the Esx family because they , like PE and PPE , tend to be found in operon pairs , some of which are known to code interacting proteins [40]–[42] . We applied our method to the 22 Esx proteins in Mtb H37Rv , in an identical manner to our analysis of the PE/PPE pairs , and found that known Esx interacting pairs and operon pairs , were given a high Sij ( coevolution ) by our method , and that many of these were supported by a high Rij ( coexpression ) ( Figure S1 ) . The Sij and Rij scores for the Esx analysis are given in the Supporting Information ( Dataset 4 ) . We conclude that our method has the potential to predict interactions in protein families beyond PE and PPE .
To generate a high-confidence set of predicted PE/PPE complexes , we took the overlap of the top 5% by both coevolution ( Sij ) and coexpression ( Rij ) , yielding the 35 pairs shown in Table 2 . The same predicted interactions are shown in a network representation in Figure 5 . Panel A shows that 6 of the 12 operon pairs in the top 35 predicted complexes are predicted to interact specifically . That is , the PEs in this group do not appear to interact with PPEs other than their operon partner , and vice versa . The specificity of interaction in operon pairs is also suggested by the tendency for inter-operon pairs to have low scores ( Figure 4 , green crosses ) . Figure 5B shows predicted cross-reaction of inter-operon pairs Rv3872/Rv1387 , Rv1195/Rv3478 , Rv3477/Rv1196 , and Rv2769c/Rv1039c . Notice that the inter-operon interactions in Figure 5 are between pairs of proteins in the same colored regions in the phylogenetic trees in Figure 2 , suggesting that pairs of paralogs with sufficiently similar sequences ( nearby on a tree ) could also cross-react . A high degree of mRNA coexpression ( Table 2 ) provides additional evidence that there could be some cross reactivity between these PE/PPE operons . Cross-reactivity between genome-paired Esx proteins has been noted previously in Mtb [40] , and it may be that the subunits of closely related PE/PPE complexes can similarly cross-react to confer functional flexibility as with the Esx family . However , our finding of negative coevolution and coexpression of the majority of the 196 possible inter-operon pairs ( Figure 4 ) suggest that the four interactions we predict are exceptions rather than the rule . Figure 5C shows possible cell surface-associated PE/PPE complexes . Four of the 6 PE proteins are PE_PGRS proteins ( Rv0109 , Rv0754 , Rv1803c , and Rv2487c , ) , which are thought to be variable surface antigens displayed on the exterior of Mtb cells [5] . The other two PE proteins ( Rv0151c and Rv0160c ) were predicted in a previous study to contain membrane beta barrels [43] , and thus are also likely localized to the cell surface . The PE and PPE proteins here appear to have multiple interaction specificities , particularly PE proteins Rv0160c and Rv0754 , each of which is linked to several PPEs . Rv0160c and Rv0754 also have overlapping patterns of interaction as they share three predicted PPE partners: Rv0355c , Rv0442c , and Rv1801 . However , in contrast to the example of Rv1195 and Rv3477 ( Figure 5B ) , Rv0160c and Rv0754 do not have remarkably high sequence similarity ( 58%; both proteins are more similar to many other PEs ) or close distance in the PE phylogenetic distance matrix . We would therefore conclude that the patterns of cross-reactivity shown in Figure 5C cannot be explained simply by sequence similarity . Future structural studies could reveal the detailed residue interactions responsible for complex formation among these proteins . Because of the possible cell surface localization of the proteins in Figure 5C , it may be that they are part of multiprotein cell-surface complexes involving varying combinations of PPE proteins interacting with surface-localized PE proteins . We truncated all proteins to include only the PE or PPE domains; therefore our method predicts that PE_PGRS and membrane-associated PE proteins interact with PPEs through their PE domains . All of these interpretations await experimental confirmation . Two of the 14 Mtb H37Rv operon pairs ( Rv1040c/Rv1039c and Rv3622c/Rv3621c ) were not among the 35 putatively interacting PE/PPE pairs identified by our procedure . Both of these operon pairs exceeded our coevolution ( Sij ) threshold , but were slightly below our coexpression ( Rij ) threshold of 0 . 34 ( Rij = 0 . 26 and 0 . 25 , respectively; see also Figure 4 ) . Employing a lower Rij threshold , for example the 85th percentile ( Rij = 0 . 24 ) , would result in the predicted interaction of all 14 operon pairs . To predict new PE/PPE interactions we analyzed sequences homologous to the interacting domains in the known PE/PPE complex [13] , without explicitly limiting our analysis to defined interface residues between the subunits as in [44] . We reasoned that , because both PE and PPE are helical , that in different paralogous complexes the registry of the helices could change , bringing different residues into the interface . This may be especially true for PE and PPE , which are highly elongated in shape ( the characterized PE/PPE complex is about 108 Å long by 26 Å wide [13] ) , and , as a result , most residues in either subunit are near ( within 10 Å ) a residue in the other subunit . Computational and experimental studies [29] , [45] have found evidence for energetic coupling of distant residues ( >10 Å ) in a number of protein families . While these studies focused on protein monomers , it is possible that the finding of long-range residue-residue interactions could also apply to complexes . With this in mind , we were cautious about excluding regions homologous to either subunit in which a mutation might not be ‘felt’ in the other subunit . To test whether excluding residues distant from the complex interface would influence our results , we inspected the PE/PPE structure [13] and found a contiguous region spanning residues 101–148 of the PPE with C-α distances greater than 10 Å from the nearest residue of the PE . We constructed a modified PPE alignment omitting these residues ( and homologous regions of aligned PPE proteins ) and repeated our analysis , and found no significant change in the distance matrix made from the unmodified PPE alignment ( correlation of PE and modified PPE distance matrices = 0 . 84; no improved correlations in 106 matrix shufflings ) . We conclude that including some non-interface residues , at least in the case of PE/PPE , did not significantly bias our results . The large expansion of the PE and PPE genes in Mtb [12] allowed us to obtain results using the genome sequence of Mtb strain H37Rv alone , without bringing in data from other genomes . Our method could therefore in principle be applied to large families of interacting paralogs in microbial genomes without necessarily having any closely related genome sequences from other microbes . To explore whether adding genome data from other mycobacteria would improve our results , we searched an additional 14 mycobacterial genomes for PE/PPE operon pairs . Orthologs to H37Rv PE/PPE operon pairs are summarized in Figure S2 . Ninety additional operon pairs were found , adding to the 14 of Mtb H37Rv to give 104 operon pairs in total . However , only 24 of these operon pairs had PE or PPE domains with amino acid sequences different ( usually just by an amino acid or two ) from the 14 H37Rv operon pairs . The 38-protein reference matrices derived from multiple genomes showed a reasonable increase in correlation over the 14-protein reference matrices from H37Rv ( 0 . 91 up from 0 . 84; no higher correlations yielded by shuffling the matrices' gene order 106 times ) . No clear improvement in the results was evident from using the multi-genome set of operon pairs ( data not shown ) , perhaps because each of the added 24 sequences was highly similar to one already in the 14 H37Rv operon pairs . However , it is likely that as still more mycobacterial genomes are sequenced , new sequence variants of PE and PPE domains will be discovered , and this may improve our results . In the five Mtb strains analyzed , 65/70 or 93% of the operon pairs were conserved ( Figure S2 ) . Five operon pairs were missing either the PE or PPE protein . In particular , the Mtb C strain has single-gene deletions in three operon pairs ( Rv1040c/Rv1039c , Rv1806/Rv1807 , and Rv2107/2108 . It is possible that in Mtb strains with ‘broken’ operon pairs , another interacting partner is able to interact with the orphaned gene , possibly restoring the PE/PPE complex's function , or introducing new complexes that help these strains survive in their environmental niches . It is possible that a single Mtb strain ( in this case H37Rv ) , in which the PE and PPE families are more expanded relative to other mycobacteria [12] , provides a broad sampling of the tolerated residue variations in these families , as proteins with many paralogs are thought to be under negative selective pressure on their interactions with paralogs other than their cognate partner [46] . Thus , the more specifically interacting PE/PPE protein pairs there are in a genome , the more residue variation we might see , due to positive selective pressure on interaction specificity . The extent of interaction promiscuity between the PE and PPE families is unknown , but our observations are consistent with negative selection on promiscuous interaction . This suggests that there may be some advantage to Mtb in maintaining interaction specificity of PE and PPE proteins , at least in those that are operon-paired ( and which we also predicted to interact with some degree of specificity , Figure 5A ) . We conclude that our analysis was not significantly affected by the inclusion of only a single genome , and that this could be a useful approach for mining the interactions of newly sequenced genomes for which there are initially no ( or just a few ) related genomes to compare to . Related prediction methods [20] , [21] , [24] that compared phylogenetic distance matrices relied on a training set of experimentally determined protein-protein interactions to build the reference matrices , or large sets of orthologs of a few known interacting proteins [25] . At the time of this study , there is only a single experimentally characterized PE/PPE interaction [13] . We resolved this impasse by employing the operon method [15]–[18] to define a high-confidence set of predicted interactions ( including the known complex ) to build phylogenetic distance matrices capable of capturing some of the residue covariation patterns in PE/PPE complexes . The validity of our results depends on future verification of this assumption . Even so , the high degree of coevolution and coexpression seen in operon-paired PE/PPE genes , combined with definitive experimental characterization of at least one PE/PPE complex [13] , implies that our assumption that operon-paired PE/PPE genes code complexes is fair . We envision testing the PE/PPE complexes predicted by our approach with a scalable high-throughput strategy . Rapid cloning methods such as ligation-independent cloning ( LIC ) [47]–[49] could be employed to rapidly build up PE/PPE co-expression plasmids like that described for the Rv2431c/Rv2430c complex [13] . These strategies allow the experimentor to avoid time-consuming restriction and ligation steps during cloning . Using LIC , putative PE/PPE complexes could rapidly be screened to assess whether further study , including structural characterization , would be worthwhile . Using our set of predicted interactions to prioritize experiments would likely reduce the required number of assays to successfully characterize complexes . As all PE/PPE pairs in our set of operon pairs were found in the top 5% of predictions it is possible that the success rate of such prioritized assays , relative to all-vs-all screening , could be increased by an order of magnitude or more . A method for predicting the interactions of the PE and PPE families of proteins in M . tuberculosis , beyond those simply linked by the operon method , is proposed . The method combines known interacting domain structure , genomic operon organization , and protein coevolution , and predicts that 35 pairs of PE and PPE proteins interact . Our method can be applied to a single genome if sufficient numbers of paralogs are present , or could be used in a multi-genome framework . A subset of the predictions from our coevolution-based method is confirmed by high mRNA coexpression , suggesting their biological relevance , and likely weeding out false positives . Our results may be a useful starting point for experimentally probing the interactions of PE , PPE , and other microbial protein families .
Annotations for the Mtb H37Rv genome were downloaded from the NCBI FTP site ( ftp://ftp . ncbi . nih . gov ) . PE and PPE genes were identified from these annotations . The gene coordinates and orientation information provided in the annotations were used to compile a list of adjacent PE/PPE pairs in the same orientation , with the PE protein located 5′ ( upstream ) to the PPE protein , and with no more than 100 base pairs intergenic separation . Increasing the intergenic distance cutoff to 500 base pairs did not result in any additional PE/PPE pairs . 280 multiprotein complexes involving 692 proteins , and 2 , 909 unique operons involving 4 , 510 genes , were extracted from the EcoCyc database [30] . A total of 1918 pairwise protein interactions were found in the complexes . Of these pairs , 942 or 49% of the pairs were also found together in an operon . To assess the significance of the overlap , the identities of the proteins in the complexes were randomized ( each protein was replaced with a unique , random E . coli protein ) , and the co-occurrence in operons was reassessed . One thousand shuffling trials were performed and the overlap of 49% was not met or exceeded in any of the trials . The maximum overlap achieved in any trial was 2% . For each of the PE and PPE families , family member sequences were selected from the SwissProt database [50] by two criteria: 1 . the sequence was annotated as belonging to either the PE or PPE protein families in Pfam , or 2 . the protein was otherwise annotated as belonging to one of these families . Multiple alignment of the protein sequences was performed using the ClustalW program using default parameters , and a secondary structure profile generated by the DSSP program [32] and the known structure [13] . Alignments were visually inspected and hand-edited to omit sequences with obvious low homology . Rv3893c ( PE36 ) , though classified by SwissProt as a PE protein , appeared to be an outlier and was therefore removed from the multiple alignment . Rv3892c ( PPE69 ) , its genome-paired neighbor , was kept in the PPE multiple alignment . Because of the omission of Rv3893c , the PE/PPE pair of Rv3893c/Rv3892c was not included in the genome-paired reference set . Rv3892c was included in subsequent predictions . The resulting structure-based alignments had 87 proteins in the PE alignment , and 65 proteins in the PPE alignment . Multiple alignments were truncated to include only those positions homologous to proteins in the known PE/PPE complex of Rv2431c/Rv2430c . Rv2431c and Rv2430c are among the shortest members of the PE and PPE families , respectively , and inspection of the complex structure suggests that these sequences may represent the minimal interface regions necessary to form a complex . Many PE and PPE proteins have other domains , low-complexity regions , and transmembrane domains C-terminal to their PE or PPE domain . These regions are unlikely to participate directly in the PE/PPE interaction; this was our reasoning for removing these regions from the alignment . From the full-family multiple alignments , Phylip distance matrices were generated in the ClustalW program [31] . This resulted in an 87×87 matrix for the PE family and a 65×65 matrix for the PPE family . These matrices would be used subsequently in predictions to give us the distance between any pair of PE sequences and any pair of PPE sequences . Next , phylogenetic distance matrices were made for just the 14 pairs of operon-paired PE and PPE proteins . This was done by extracting the 14 operon-paired sequences from each of the full-family multiple alignments . The resulting subalignments were used to generate a 14×14 PE distance matrix and a 14×14 PPE matrix . Importantly , the two 14×14 matrices were ordered so that the ith protein in the PE matrix was the operon partner of the ith protein in the PPE matrix . We would later use these matrices as ‘reference’ matrices for prediction of non-operon-paired PE/PPE complexes . Phylogenetic trees were generated from the PE and PPE 14-sequence subalignments using the ClustalW program [31] . The correction for multiple substitutions was not used to generate the trees . Bootstrapping of the trees was done within the ClustalW program . Let X be the 14×14 distance matrix of PE proteins , and Y be the 14×14 distance matrix of PPE proteins . Xij is the percent divergence of PEi and PEj; likewise Yij is the percent divergence of PPEi and PPEj . Xi is the vector of length 14 for the distances of PEi from all PEj ( including for i = j , in which case Yij = 0 . 0 ) ; likewise Yi . To determine the correlation between the ordered 14×14 distance matrices , X and Y , the Pearson correlation is taken . To avoid counting protein distances twice , only the unique elements in the matrices are taken ( that is , we count i , j pairs but not j , i ) . L = 14 , the number of operons with paired PE and PPE genes . rXY is a measure of the coevolution of the operon-paired subsets of the PE and PPE protein families . Next we derive Cij , a measure of the coevolution of PEi with PPEj , for all PE and PPE proteins in the Mtb genome , including but not limited to proteins in the operon-paired set . For PEi , we generate Ai , a distance vector of length 14 , containing the distances from PEi to each of the 14 PEk in the 14×14 reference matrix . Bj is equivalently generated for PPEj . To get Cij , a measure of the coevolution of PEi with PPEj , the Pearson correlation of Ai and Bj is calculated . Note that here we are taking the correlation of two vectors with length 14 . To generate the paralog matching score , Sij , a reciprocal ranking procedure was used . For each PE , a ranked list of the most coevolved PPEs was produced . The same was done for PPEs . Then , for each PEi , we recorded the position of each PPEj protein in the PE's list of PPEs ranked by Cij to give ri→j , the number of PPEs ranking below PPEj . The reciprocal procedure was performed to yield rj→i . N is the total number of PE/PPE pairs . Using the above formula , a high score is assigned only if both pairs were high on each other's list of most coevolved potential interacting partners . Positive examples of PE/PPE complexes were defined as all of the 14 operon-paired proteins . A dataset of negative PE/PPE interactions are not currently available , so we made the assumption that operon-paired PE/PPE interactions were specific , and therefore a PE in one operon would not interact with a PPE in a different operon . 182 inter-operon PE/PPE pairs resulted , and were used as a negative set . Next , for a range of prediction thresholds from 0 . 0 to 1 . 0 , true positive ( TP ) , true negative ( TN ) , false positive ( FP ) , and false negative ( FN ) rates were determined . The sensitivity , or true positive rate ( TPR ) , was calculated as1-specificity , or the false positive rate ( FPR ) , was calculated as In a receiver operator characteristic curve ( not shown ) , the prediction threshold corresponding to the upper left-most portion of the curve represents the optimum compromise between TPR and FPR . For prediction with Sij , this threshold was roughly 0 . 75 . Taking all PE/PPE pairs with Sij≥0 . 75 gives roughly 5% of the total 5 , 590 possible PE/PPE pairs , in which all 14 of the operon pairs were included . Nine Mtb gene expression datasets ( Table S1 ) in the Gene Expression Omnibus ( GEO ) [38] were downloaded . All available Mtb datasets in GEO were used excluding for consistency those that studied deletion mutants or attenuated strains . Also for consistency , only datasets that reported gene expression changes as a ratio of a sample and a reference were used . Gene expression data from the nine studies were represented as a matrix where the rows were genes and the columns were experiments . To construct an expression vector for a gene , the data from each of the nine studies were concatenated . Combined expression vectors were made up from the field labeled ‘VALUE’ in the data files . In all data sets , this value represents the measured expression level of a gene under experimental conditions versus that gene in a reference sample . Various normalization schemes were applied by the authors of the individual datasets to correct for scale differences due to differing intensities among genes in response to different experiments . Because of the difficulty in combining these schemes to make a self-consistent combined dataset , we chose not to further normalize the expression data . Correlation coefficients of gene expression vectors were calculated for all possible pairs of genes . To obtain a correlation coefficient for genes i and j over N experiments , the Pearson correlation coefficient , Rij , was used . where gix and gjx are the expression values reported in the GEO data file for genes i and j , respectively , in experiment x . For each pair of genes analyzed , combined expression vectors were truncated by deleting experiments in which either or both genes had missing values . Thus N varied for each pair of genes assessed . In all , 734 experiments were used for the inference of pairwise coexpression relationships between pairs of PE and PPE genes . The Kolmogorov–Smirnoff ( KS ) test asks whether two collections of random samples come from the same distribution . We want to know if the coexpression scores for a group of PE/PPE gene pairs predicted to code interacting proteins has a different distribution ( with a higher mean ) than PE/PPE gene pairs overall . Because we expect the linked proteins to have a higher-valued mean than the unlinked proteins , we used the one-tailed version of the KS test . An α significance criteria of 0 . 05 was used for hypothesis acceptance/rejection . The PE/PPE complex described in [13] was analyzed using the RasMol program [51] . The structure was visually inspected to identify a contiguous region spanning residues 101–148 of the PPE with C-α distances greater than 10 Å from the nearest residue of the PE . The PPE multiple alignment was modified by deleting all columns that aligned to this contiguous region . The 22 Esx family genes were identified in Mtb H37Rv from NCBI annotations . The genes were divided into two groups: ESAT-6 paralogs ( 12 proteins ) and CFP-10 paralogs ( 10 proteins ) . We based this categorization on the observation that 20 of the 22 the Esx genes are organized into 10 adjacent ( operon ) pairs on the genome , with a gene similar to CFP-10 lying upstream from a gene similar to ESAT-6 . We therefore categorized upstream genes as CFP-10 paralogs and downstream genes as ESAT-6 paralogs . The two annotated Esx genes not in operon pairs , Rv1793 and Rv3017c , were both judged to be ESAT-6 paralogs from visual inspection of a multiple alignment . The 10 Esx operon pairs were used to build reference matrices as with PE and PPE . Coevolution ( Sij ) and coexpression ( Rij ) scores were derived using the same procedure as for PE and PPE . For each PE or PPE gene in an operon pair , orthologs were manually extracted from the TB Database ( http://www . tbdb . org ) . The results are summarized in a tab-delimited file in the Supporting Information ( Dataset 5 ) . The ORF identifiers , gene names , and SwissProt accession codes of the PE and PPE proteins analyzed in this study are listed in Text S1 .
|
We consider the problem of detecting protein interactions from genome sequences when the potential interacting partners belong to large families of similar ( homologous ) proteins . Many computational methods for predicting protein interactions rely on similarity to a pair of known interacting proteins . When the proteins in question are members of large groups of similar proteins within the same organism ( paralogs ) , the problem of inferring the correct interactions becomes difficult . To illustrate the problem , we undertook prediction of interactions of some highly expanded protein families of Mycobacterium tuberculosis ( Mtb ) , which are believed to contribute to the bacterium's ability to infect human beings . To generate predictions , we analyzed patterns of coevolution in a small subset of likely interacting proteins , and extended these patterns to predict additional interactions throughout the genome . Our results provide a map for experimental probes of the Mtb interaction network , for the benefit of drug and vaccine discovery . More generally , our procedure is applicable to detecting interactions of proteins that belong to large families of paralogs in any organism with a sequenced genome .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"evolutionary",
"biology/microbial",
"evolution",
"and",
"genomics",
"genetics",
"and",
"genomics/functional",
"genomics",
"microbiology/microbial",
"evolution",
"and",
"genomics",
"computational",
"biology/genomics",
"infectious",
"diseases/bacterial",
"infections",
"computational",
"biology/systems",
"biology"
] |
2008
|
Identifying Cognate Binding Pairs among a Large Set of Paralogs: The Case of PE/PPE Proteins of Mycobacterium tuberculosis
|
Small molecule signaling promotes the communication between bacteria as well as between bacteria and eukaryotes . The opportunistic pathogenic bacterium Legionella pneumophila employs LAI-1 ( 3-hydroxypentadecane-4-one ) for bacterial cell-cell communication . LAI-1 is produced and detected by the Lqs ( Legionella quorum sensing ) system , which regulates a variety of processes including natural competence for DNA uptake and pathogen-host cell interactions . In this study , we analyze the role of LAI-1 in inter-kingdom signaling . L . pneumophila lacking the autoinducer synthase LqsA no longer impeded the migration of infected cells , and the defect was complemented by plasmid-borne lqsA . Synthetic LAI-1 dose-dependently inhibited cell migration , without affecting bacterial uptake or cytotoxicity . The forward migration index but not the velocity of LAI-1-treated cells was reduced , and the cell cytoskeleton appeared destabilized . LAI-1-dependent inhibition of cell migration involved the scaffold protein IQGAP1 , the small GTPase Cdc42 as well as the Cdc42-specific guanine nucleotide exchange factor ARHGEF9 , but not other modulators of Cdc42 , or RhoA , Rac1 or Ran GTPase . Upon treatment with LAI-1 , Cdc42 was inactivated and IQGAP1 redistributed to the cell cortex regardless of whether Cdc42 was present or not . Furthermore , LAI-1 reversed the inhibition of cell migration by L . pneumophila , suggesting that the compound and the bacteria antagonistically target host signaling pathway ( s ) . Collectively , the results indicate that the L . pneumophila quorum sensing compound LAI-1 modulates migration of eukaryotic cells through a signaling pathway involving IQGAP1 , Cdc42 and ARHGEF9 .
Bacteria accomplish intra-species and inter-species communication through the production , secretion and detection of low molecular weight compounds [1 , 2] . Many of these compounds , termed “autoinducers” , trigger above a certain concentration threshold transmembrane phosphorylation signaling and ultimately gene regulation . The bacterial signaling compounds belong to a variety of chemical classes , including the furanosyl borate ester autoinducer-2 ( AI-2 ) , cis-2-dodecenoic acid , alkylhydroxyquinolines ( e . g . Pseudomonas aeruginosa quinolone signal , PQS ) , N-acylhomoserinelactones ( AHLs ) , or α-hydroxyketones ( AHKs ) [1–5] . The AHKs CAI-1 ( Cholerae autoinducer-1; 3-hydroxytridecane-4-one ) and LAI-1 ( Legionella autoinducer-1; 3-hydroxypentadecane-4-one ) have been identified in Vibrio cholerae [6] or Legionella pneumophila [7] and are produced by the homologous autoinducer synthases CqsA or LqsA , respectively . Moreover , Janthinobacterium sp . HH01 [8] and Photobacterium angustum [9] harbor CqsA/LqsA orthologues and appear to employ AHK-dependent quorum sensing . The signaling molecule LAI-1 is produced and sensed by the lqs ( Legionella quorum sensing ) genes [10] , which are clustered and divergently transcribed from individual promoters [11] . The lqs cluster encodes the autoinducer synthase LqsA , the putative cognate sensor kinase LqsS and the prototypic response regulator LqsR [3] . The production of LqsR is dependent on the alternative sigma factor RpoS ( σ38/σS ) , and therefore , LqsR is an element of the stationary-phase regulatory network of L . pneumophila [10] . In addition , the putative sensor kinase LqsT represents an orphan LqsS homolog , which is also a component of the LAI-1 circuit [12] . LqsS and LqsT act as antagonists , as 90% of the genes up-regulated in absence of lqsS are down-regulated in absence of lqsT . Recent biochemical experiments revealed that LqsS and LqsT are indeed sensor histidine kinases , the auto-phosphorylation of which is regulated by LqsR [13] . In turn , the sensor kinases phosphorylate a conserved aspartate in LqsR , leading to dimerization of the putative response regulator . Synthetic LAI-1 reduces auto-phosphorylation of LqsS and LqsT , regulates gene expression and promotes the motility of L . pneumophila in the micromolar range [14] . The Lqs system controls pathogen-host cell interactions and production of virulence factors [10 , 15] . While L . pneumophila lacking lqsA is only slightly impaired for intracellular replication [16] , the lqsA mutant strain and all other lqs mutants are outcompeted by wild-type bacteria upon co-infection of Acanthamoeba castellanii [12] . L . pneumophila lacking lqsR [10] , lqsS [16] , lqsT [12] or the whole lqs cluster ( lqsA-lqsR-hdeD-lqsS ) [15] are defective for host cell uptake and intracellular replication . The ΔlqsR and ΔlqsS mutant strains produce a network of extracellular filaments , and therefore , sediment more slowly than wild-type bacteria [16] . Furthermore , in absence of lqsS , a 133 kb genomic “fitness island” is up-regulated [16] , and all lqs mutant strains show much higher natural competence for DNA acquisition [12] . L . pneumophila is an amoebae-resistant environmental bacterium that can cause a severe pneumonia termed Legionnaires’ disease [17 , 18] . The opportunistic pathogen employs the Icm/Dot type IV secretion system ( T4SS ) and the remarkable number of about 300 different translocated effector proteins to form a replication niche , the Legionella-containing vacuole ( LCV ) and to define other interactions with host cells [19–24] . Accordingly , L . pneumophila impedes the migration of infected Dictyostelium discoideum amoebae and mammalian cells in an Icm/Dot-dependent manner [25] . The Icm/Dot-translocated effector protein LegG1 , a Ran GTPase activator [26] , antagonizes migration inhibition by Ran-dependent microtubule stabilization . The small GTPases RhoA , Rac1 and Cdc42 promote directional migration , proper microtubule assembly and actin cytoskeleton organization in the cell , in concert with the scaffold protein IQGAP1 , which represents a key node within the small GTPase network [27] . In the present study , we show that the L . pneumophila quorum sensing compound LAI-1 inhibits cell migration through a signaling pathway involving IQGAP1 , Cdc42 and the Cdc42 activator ARHGEF9 .
Wild-type L . pneumophila , but not mutant bacteria lacking a functional Icm/Dot T4SS , inhibit cell migration of free-living amoebae and mammalian cells [25] . To further define the bacterial factors implicated in inhibition of cell migration , we infected D . discoideum amoebae or RAW 264 . 7 macrophages with L . pneumophila mutant strains lacking components of the Lqs quorum sensing system and tested the effects on cell migration in under-agarose chemotaxis assays ( Fig 1A ) . L . pneumophila strains lacking lqsR , lqsS and/or lqsT inhibited the chemotactic migration of D . discoideum ( Fig 1B ) or macrophages ( Fig 1C ) to a similar extent as wild-type bacteria , suggesting that these Lqs components , albeit implicated in bacterial virulence , play a minor role for cell migration inhibition . In contrast , however , the migration of D . discoideum or macrophages infected with L . pneumophila ΔlqsA was not inhibited ( similar to cells infected with ΔicmT mutant bacteria ) . Next , we tested in the D . discoideum under-agarose assay the effects of over-expressing lqsA in the ΔlqsA or ΔicmT mutant strains or in wild-type L . pneumophila ( Fig 1D ) . The overexpression of plasmid-borne lqsA partially complemented the defect in migration inhibition of D . discoideum by the ΔlqsA mutant strain ( Fig 1E ) . To analyze the effects of over-expressing lqsA in another cell migration system , we used A549 lung epithelial carcinoma cells and a scratch wound healing assay [25] ( Fig 1F ) . Under these conditions , the overexpression of lqsA in the ΔlqsA mutant strain completely restored the inhibition of cell migration ( Fig 1G ) . Moreover , the overexpression of lqsA in the ΔicmT mutant strain significantly inhibited A549 cell migration , suggesting that LqsA ( and in consequence , LAI-1 ) inhibit the migration of eukaryotic cells . Taken together , while most lqs genes do not appear to play a major role for cell migration inhibition by L . pneumophila , lqsA is required for inhibiting the migration of amoebae , macrophages and epithelial cells . These findings suggest that the signaling molecule LAI-1 produced by the autoinducer synthase LqsA might directly or indirectly affect cell migration . Based on the genetic results , we tested whether synthetic LAI-1 modulates chemotaxis and migration of eukaryotic cells . To this end , D . discoideum amoebae were treated for 1 h with different concentrations of LAI-1 and cell migration towards folate was monitored in under-agarose assays for 4 h ( Fig 2A ) . We observed that 0 . 5–10 μM racemic LAI-1 inhibited migration of the amoebae in a dose-dependent manner . The biosynthetic pathway and the stereochemistry of LAI-1 are currently unknown . Given the efficacy of LAI-1-meditated inhibition of cell migration , we sought to employ this readout to assess the biological activity of enantiomers of LAI-1 and its putative precursor amino-LAI-1 [28 , 29] . We treated D . discoideum with 10 μM ( R ) - or ( S ) -LAI-1 , and with 10 μM ( R ) - or ( S ) -amino-LAI-1 . The ( R ) -enantiomers of LAI-1 or amino-LAI-1 inhibited the migration of the amoebae more efficiently than the ( S ) -enantiomers ( Fig 2B ) . Furthermore , the V . cholerae signaling molecule CAI-1 also impeded the migration of D . discoideum in under-agarose chemotaxis assays: 0 . 5–10 μM CAI-1 inhibited migration of the amoebae in a dose-dependent manner ( Fig 2C ) . Neither LAI-1 nor CAI-1 acted as chemo-attractants , and up to 50 μM LAI-1 did not affect uptake or cytotoxicity of L . pneumophila in D . discoideum ( S1 Fig ) . We also investigated as motility parameters the forward migration index and the velocity of D . discoideum amoebae treated with 10 μM LAI-1 ( Fig 2D ) . Single cell tracking analysis using the ImageJ manual tracker and Ibidi chemotaxis software revealed that upon treatment with LAI-1 the forward migration index was reduced by approximately 50% , while the velocity of the amoebae was not affected . These results indicate that the directionality but not the speed of the phagocytes was impaired by the bacterial quorum sensing signal . Taken together , the chemotactic migration of D . discoideum amoebae was found to be inhibited in a dose-dependent manner by LAI-1 or CAI-1 , and the ( R ) -enantiomers of the α-hydroxyketones or α-aminoketones are biologically more active with respect to inhibition of cell migration . To assess whether the migration of macrophages is also affected by LAI-1 , we treated murine macrophage-like RAW 264 . 7 cells with different concentrations of racemic LAI-1 and monitored cell migration towards the chemokine CCL5 in under-agarose assays for 4 h ( Fig 2E ) . Similar to amoebae , migration of macrophages was also inhibited in a dose-dependent manner upon treatment with 1–10 μM LAI-1 . Also , up to 10 μM LAI-1 did not affect uptake or cytotoxicity of L . pneumophila in macrophages ( S1 Fig ) . Next , we tested the effects of LAI-1 on the cellular microtubule and actin cytoskeleton . RAW 264 . 7 macrophages were treated with 10 μM LAI-1 , the cells were then immuno-labeled for α-tubulin , and microtubule polymerization was quantified by counting the number of microtubule fibers along cross-sections ( Figs 2F and S2A ) . These experiments revealed that LAI-1 reduced the number of microtubule fibers per cell by approximately 50% , while the overall amount of α-tubulin was not affected . As a control for microtubule disintegration , the cells were treated with 30 μM of the microtubule destabilizing compound nocodazole . Analogously , we visualized the actin network by fluorescence microscopy in macrophages exposed to 10 μM LAI-1 ( Fig 2G ) . The treatment with LAI-1 markedly altered the architecture of the actin cytoskeleton . We found that the cortical actin nearly completely disappeared upon treatment with the L . pneumophila signaling molecule . At the same time , treatment with LAI-1 did not affect the overall amount of actin in the cell , as revealed by Western blot ( Fig 2G ) . Taken together , LAI-1 inhibits cell migration by profoundly affecting microtubule polymerization as well as the F-actin network architecture . LAI-1 inhibits the chemotactic migration of phagocytes towards folate or CCL5 ( Fig 2 ) . To test whether LAI-1 also inhibits migration of eukaryotic cells in absence of an exogenously added chemo-attractant , we used the A549 lung epithelial cells and the scratch wound healing assay ( Figs 3 and S2B ) . Confluent layers of the epithelial cells were treated or not with 10 μM LAI-1 , scratched and let migrate for 24 h . Within this period of time , untreated cells repopulated the scratch area and thus formed a confluent layer again . In contrast , cells treated with 10 μM LAI-1 were severely impaired for migration ( Fig 3A ) , and the area of the scratch was closed to only 25% ( Fig 3B ) . Thus , LAI-1 not only inhibits directed migration towards an exogenously added chemo-attractant but also towards a scratch wound . A549 cells are readily amenable to RNA interference ( RNAi ) , thus allowing the assessment of host factors during L . pneumophila infection [25 , 26 , 30] . Analogously , factors implicated in LAI-1-dependent signaling can be investigated . To assess possible eukaryotic factors involved in LAI-1-dependent signaling , we studied the roles of the scaffold protein IQGAP1 and small GTPases of the Rho/Rac/Cdc42 family implicated in cytoskeletal dynamics [31 , 32] . IQGAP1 , Cdc42 , RhoA or Rac1 were depleted by RNAi for 2 days in confluent layers of A549 cells , which were then treated or not with 10 μM LAI-1 , scratched and let migrate for another 24 h ( Fig 3A ) . Western blot analysis revealed that after the RNAi treatment the proteins were not detectable anymore ( S3 Fig ) . Upon depletion of IQGAP1 or Cdc42 ( but not RhoA or Rac1 ) LAI-1 no longer prevented scratch closure compared to cells treated only with the corresponding siRNA ( Fig 3B ) . Therefore , IQGAP1 and Cdc42 , but not RhoA or Rac1 , promote the transmission of LAI-1-mediated inter-kingdom signaling . The depletion of IQGAP1 , Cdc42 or Rac1 did not significantly affect the scratch closure of untreated cells , yet the depletion of RhoA reduced scratch closure by approximately 50% , regardless of whether LAI-1 was present or not ( Fig 3B ) . The Icm/Dot-translocated Ran activator LegG1 antagonizes the inhibition of cell migration by L . pneumophila in a Ran GTPase-dependent manner [25] . To assess a role for Ran and its effector RanBP1 in LAI-1-mediated inhibition of cell migration , we depleted the small GTPase or its effector in A549 cells and performed scratch assays upon treatment with LAI-1 ( S4A Fig ) . In these experiments , neither Ran nor RanBP1 were found to play a significant role in LAI-1-mediated inhibition of cell migration . To investigate whether LAI-1 modulates the activation of Cdc42 , a pulldown assay was performed . A549 cells treated or not with 10 μM LAI-1 for 1 h were lysed , and the lysate was incubated with GST-PBDPak1 , a fusion protein specifically binding activated Cdc42 . After pulldown with glutathione resin , the amount of active Cdc42 ( GTP ) was quantified by densitometry of Western blots using an antibody that specifically recognizes Cdc42 ( GTP ) ( Fig 4A ) . The analysis revealed that upon treatment of the cells with LAI-1 the concentration of active Cdc42 ( GTP ) decreased approximately 10-fold . As an input control , the amount of GAPDH in the cell lysates was determined by Western blot . These findings were confirmed by an analogous approach using an anti-Cdc42 ( GTP/GDP ) antibody instead of GST-PBDPak1 , followed by Western blot with an antibody recognizing Cdc42 ( GTP ) ( S5A Fig ) . Using an antibody that recognizes Cdc42/Rac1-phospho-Ser71 , we also assessed the phosphorylation state of Cdc42 in response to LAI-1 by Western blot ( S5B Fig ) or fluorescence microscopy ( S5C Fig ) . Yet , we did not observe changes in the phosphorylation pattern or intensity upon treatment of the cells with LAI-1 . Taken together , LAI-1 signaling promotes the inactivation of Cdc42 , without affecting the phosphorylation of the small GTPase . To analyze whether LAI-1 alters the spatial distribution of the scaffold protein IQGAP1 or the small GTPase Cdc42 , we incubated A549 cells for 1 h with 10 μM of the quorum sensing compound and stained the cells with an antibody against IQGAP1 ( Fig 4B ) or Cdc42 ( Fig 4C ) . Upon treatment of the cells with LAI-1 , IQGAP1 redistributed from the cytoplasm to the cell cortex . The re-localization was very efficient , as nearly 100% of the cells observed by microscopy showed a cortical accumulation of IQGAP1 after exposure to LAI-1 . In contrast , the cytoplasmic localization of Cdc42 remained unaffected by LAI-1 treatment . The treatment with LAI-1 did not affect the overall amount of IQGAP1 , Cdc42 or Rac1 ( S5D Fig ) . To test whether in the LAI-1 signal transduction pathway the redistribution of IQGAP1 to the cell cortex requires Cdc42 , we depleted Cdc42 in A549 cells , followed by exposure to 10 μM LAI-1 and an analysis of the cellular localization of IQGAP1 by fluorescence microscopy ( Fig 5A ) . Treatment of the cells with siRNA against Cdc42 efficiently depleted the cells of the small GTPase ( Figs 5B and S3 ) . Under these conditions , IQGAP1 still re-distributed to the cell cortex upon treatment with LAI-1 , indicating that Cdc42 is dispensable for the re-localization of the scaffold protein ( Fig 5C ) . Thus , treatment of A549 cells with LAI-1 leads to the deactivation of the small GTPase Cdc42 , as well as to the redistribution to the cell cortex of the scaffold protein IQGAP1 , which is located upstream of Cdc42 in the LAI-1 signaling cascade . In order to obtain further insight into the LAI-1-dependent inactivation of Cdc42 , we depleted by RNAi different guanine nucleotide exchange factors ( GEFs ) or GTPase activating proteins ( GAPs ) specific for Cdc42 . Confluent cell layers of A549 cells were treated with siRNA against the Cdc42-specific GEFs ARHGEF9 , FGD1 or DOCK11 , or against GAPs with more relaxed specificity , ARHGAP1 or ARHGAP17 ( Fig 6A ) . Western blot analysis revealed that after the RNAi treatment the proteins were not detectable anymore ( S3 Fig ) . Among these GTPase modulators , only ARHGEF9 appeared to be significantly involved in LAI-1-mediated cell migration inhibition , as the depletion of this GEF abolished the inhibitory action of LAI-1 ( Fig 6B ) . In an attempt to identify other host factors possibly implicated in LAI-1-mediated inhibition of cell migration , the cellular transcriptome of D . discoideum was analyzed in response to the bacterial signaling molecule . We exposed exponentially growing D . discoideum to 20 μM of synthetic LAI-1 for 3 h and compared gene regulation by transcriptome analysis to untreated amoebae . Under these conditions , LAI-1 was found to up- or down-regulate 115 or 144 genes respectively , at least 1 . 5-fold ( S6A Fig , S1 and S2 Tables ) . This number of genes constitutes approximately 5% of the 5 , 400 genes on the array [33] . 74 up- and 113 down-regulated genes , respectively , could be functionally categorized based on the yeast classification scheme which was adapted for Dictyostelium [34] . Notably , in the categories “protein destination” ( including vesicle trafficking ) , and in “signal transduction” most differentially regulated genes were up-regulated . In contrast , in the categories “translation” , “cell proliferation” and “movement” most of the genes were down-regulated ( S6A Fig ) . The latter result is in agreement with the notion that treatment with LAI-1 directly or indirectly may modulate ( reduce ) cell movement . Several genes of the ubiquitin proteasome system , the “core” autophagy genes atg8 and atg16 as well as the autophagy adaptor sequestosome-1 were up-regulated . In addition , we noted three members of the ABC transporter G family , a gene named iliA ( induced after Legionella infection ) and the gene DDB_G0274423 which encodes a Src homology 3 ( SH3 ) domain-containing protein ( S1 and S2 Tables ) . The latter gene is homologous to CD2AP , a scaffold protein modulating actin dynamics and cell migration [35] . Down-regulated genes include the aldehyde reductases alrA and alrE , several genes of the counting factor complex , the putative metallophosphoesterase dduA , rliA ( repressed after Legionella infection ) encoding a putative 12 transmembrane domain protein of the major facilitator family , as well as several peptidase encoding genes ( S1 and S2 Tables ) . The regulation of six of these genes ( three up- and three down-regulated ) was validated by quantitative real time ( RT ) -PCR ( S6B Fig ) . In summary , whole-genome transcriptome analysis revealed that synthetic LAI-1 in the micromolar concentration range regulates the expression of a number of eukaryotic genes involved in different processes , including protein homeostasis , vesicle trafficking , and cell migration . Based on the finding that LAI-1 increased the expression of D . discoideum DDB_G0274423 encoding a Src homology 3 ( SH3 ) domain-containing protein , we sought to analyze its potential role in LAI-1-mediated inhibition of cell migration . We depleted by RNAi for 2 days its mammalian homologue , CD2AP , in confluent layers of A549 cells , which were then treated or not with 10 μM LAI-1 , scratched and let migrate for another 24 h ( S4B Fig ) . Western blot analysis revealed that after the RNAi treatment CD2AP was not detectable anymore ( S3 Fig ) . Yet , compared to untreated cells or cells treated with scrambled siRNA oligonucleotides or mock-treated cells , the depletion of CD2AP had no effect on LAI-1-mediated inhibition of cell migration ( S4C Fig ) . Therefore , CD2AP appears to be dispensable for LAI-1-dependent signaling . L . pneumophila inhibits the migration of free-living amoebae as well as mammalian cells dependent on the Icm/Dot T4SS [25] as well as through LAI-1 ( Fig 2 ) . To test , whether synthetic LAI-1 affects Icm/Dot-dependent inhibition of cell migration , we infected D . discoideum amoebae or RAW 264 . 7 macrophages with wild-type L . pneumophila or ΔicmT mutant bacteria and treated the infected phagocytes with different concentrations of LAI-1 ( Figs 7 and S7A ) . In under-agarose assays LAI-1 dose-dependently reversed the inhibitory effect of wild-type L . pneumophila on the migration of D . discoideum ( Fig 7A ) or macrophages ( Fig 7B ) . In contrast , LAI-1 dose-dependently inhibited cell migration in phagocytes infected with ΔicmT mutant bacteria similar to uninfected cells . Next , we investigated the effects of LAI-1 on the motility parameters ( forward migration index and velocity ) of single D . discoideum cells infected with wild-type or ΔicmT L . pneumophila . This single cell tracking analysis revealed that 10 μM LAI-1 completely reversed the inhibitory effect of wild-type L . pneumophila on the forward migration index of infected D . discoideum ( Fig 7C ) or macrophages ( Fig 7D ) , but had no effect on the velocity of infected amoebae or macrophages ( S7B Fig ) . 10 μM LAI-1 reduced the forward migration index of uninfected or ΔicmT-infected phagocytes to a similar extent . Taken together , LAI-1 dose-dependently reverted the effects on migration of wild-type or ΔicmT-infected phagocytes by affecting the forward migration index but not the velocity of the cells . We also tested whether the inhibition of scratch wound closure by wild-type L . pneumophila is reversed by LAI-1 . Confluent layers of A549 cells were infected with L . pneumophila wild-type or ΔicmT mutant bacteria , treated or not with 10 μM LAI-1 , scratched and let migrate for 24 h ( Fig 7E ) . As above , LAI-1 reversed the migration inhibition of wild-type bacteria , but prevented scratch closure of cells infected with ΔicmT L . pneumophila ( Fig 7F ) . In summary , these results suggest that the exogenously added quorum sensing compound LAI-1 and intracellular L . pneumophila antagonistically target a signaling pathway to inhibit the migration of eukaryotic cells . To test the hypothesis that migration inhibition by LAI-1 or L . pneumophila converges on common host factors , we depleted the small GTPases Cdc42 or Rac1 in A549 cells prior to an infection with wild-type or ΔicmT mutant L . pneumophila ( Fig 8A ) . The depletion of Cdc42 ( but not Rac1 ) markedly further augmented the inhibition of cell migration by L . pneumophila ( Fig 8B ) . In contrast , the depletion of Cdc42 ( or Rac1 ) had no effect on A549 cells infected with the ΔicmT mutant strain . A549 cells depleted for Cdc42 and concomitantly infected with wild-type L . pneumophila appeared normal , and no increased cytotoxicity was observed . Moreover , confocal immuno-fluorescence microscopy revealed that IQGAP1 as well as Cdc42 co-localize with L . pneumophila wild-type or ΔicmT mutant bacteria in infected A549 cells ( Fig 8C ) . Yet , LAI-1 did not affect the co-localization of the bacteria with either the scaffold protein or the small GTPases ( S8 Fig ) . Finally , neither Cdc42 nor IQGAP1 appeared to play a role for intracellular replication of L . pneumophila ( S9 Fig ) . Collectively , these results indicate that the small GTPase Cdc42 is an essential component of the migration signal transduction pathway inhibited either by LAI-1 or by wild-type L . pneumophila , and the bacteria co-localize with Cdc42 as well as with IQGAP1 in infected cells .
Bacterial quorum sensing signals are not only implicated in population density-dependent signaling and gene regulation of prokaryotes , but also have an impact on eukaryotic cells in a process called inter-kingdom signaling [36] . However , little is known on a molecular level about the response of mammalian and protozoan cells to prokaryotic quorum sensing signals . The results presented here indicate that the L . pneumophila quorum sensing compound LAI-1 inhibits the directed migration of eukaryotic cells through a signaling pathway involving IQGAP1 , Cdc42 and ARHGEF9 . To our knowledge this is the first analysis of host factors comprising a signaling pathway implicated in inter-kingdom signaling of a bacterial AHK compound . The compounds ( R ) -LAI-1 and also ( R ) -amino-LAI-are biologically more active as inhibitors of eukaryotic cell migration than the corresponding ( S ) -enantiomers ( Fig 2B ) . In contrast , the ( S ) -enantiomer of the V . cholerae AHK autoinducer CAI-1 and its derivatives are more active than the corresponding ( R ) -enantiomers–at least for bacterial cell-cell communication [6 , 37] . It is presently unclear , whether this observation reflects inter-kingdom versus inter-bacterial signaling . The structural determinants of CAI-1 and derivatives but not the stereochemistry of the compounds have been tested for inter-kingdom signaling responses of Caenorhabditis elegans [38] . Upon depletion of IQGAP1 by RNAi , treatment with LAI-1 no longer abolished the migration of A549 epithelial cells ( Fig 3 ) . IQGAP1 is a 189 kDa multi-domain scaffold protein , which harbors among others a calmodulin-binding IQ domain and a GRD Ras GAP-related domain [31 , 32] . IQGAP1 is widely expressed and conserved among eukaryotes; and IQGAP-like proteins ( Dgap1 , GapA , RgaA ) are also present in D . discoideum [39 , 40] as well as on LCVs purified from the infected amoebae [30] . The scaffold protein IQGAP1 integrates multiple signaling pathways and coordinates a plethora of cellular activities , including chemokine- and growth factor-dependent cell proliferation , adhesion , migration and phagocytosis [31 , 32] . To date , over 90 interacting proteins have been identified , including the small GTPases Cdc42 , Rac1 and RhoA . The GRD domain does not function as a GAP , but rather binds to and stabilizes activated ( GTP-bound ) Cdc42 or Rac1 by inhibiting the intrinsic GTPase activity [41 , 42] . Accordingly , overexpression of IQGAP1 increases the amount of active Cdc42 and Rac1 in cells , whereas depleting endogenous IQGAP1 substantially decreases the activity of both small GTPases [42] . Given the pivotal role of IQGAP1 in modulating the actin cytoskeleton through Cdc42 and Rac1 , we depleted these and other small GTPases and assayed LAI-1-mediated inhibition of cell migration . The depletion of Cdc42 abrogated LAI-1-mediated cell migration inhibition , while depletion of Rac1 did not affect migration inhibition ( Fig 3 ) . Therefore , Cdc42 is clearly involved in LAI-1-mediated inter-kingdom signaling , while Rac1 is dispensable . The depletion of RhoA alone already significantly reduced the migration of A549 cells , which was not further inhibited by LAI-1 . While this observation might argue for a role of RhoA in LAI-1-mediated inter-kingdom signaling , there is no evidence that IQGAP1 binds RhoA , and therefore , this small GTPase is probably not involved in the pathway . However , at this point we cannot rule out that RhoA promotes LAI-1 inter-kingdom signaling in an IQGAP1-independent way . Treatment with LAI-1 caused the inactivation of Cdc42 , i . e . lower amounts of GTP-bound Cdc42 were present in the cells ( Fig 4A ) . Interestingly , the depletion of the Cdc42 activator ARHGEF9 abolished LAI-1-mediated cell migration inhibition , whereas the depletion of several other Cdc42 modulators had no effect ( Fig 6 ) . The Cdc42-specific GEF ARHGEF9 is up-regulated by the oncogene transcription factor CHD1L and has been previously implicated in tumor cell migration , invasion and metastasis by increasing cell motility and inducing filopodia formation [43 , 44] . In the LAI-1-affected signaling pathway , ARHGEF9 seems indispensable to activate Cdc42 and promote cell migration . Collectively , these results suggest that LAI-1 ( directly or indirectly ) inactivates ARHGEF9 or prevents its activation and thus reduces the amount of GTP-bound Cdc42 , thereby impeding cell migration . The scaffold protein IQGAP1 and the small GTPase Cdc42 are conserved among eukaryotes; e . g . there are human and D . discoideum putative orthologues of similar size , which are overall more than 30% or 60% identical , respectively . Moreover , the proteins are also found in other amoebazoa , including Acanthamoeba or Entamoeba spp . In contrast , an ARHGEF9 homologue is apparently not present in D . discoideum or other protozoa , and therefore , in these cells the LAI-1-dependent activation of Cdc42 apparently proceeds through another GEF . Upon depletion of Cdc42 , IQGAP1 still redistributed to the cell cortex ( Fig 5 ) . This finding suggests that in the LAI-1-dependent signaling pathway IQGAP1 is functioning “upstream” or at the level of Cdc42 . Thus , in this pathway and in the A549 cells used , IQGAP1 is a regulator of Cdc42 . It is discussed quite controversially in the literature , whether IQGAP1 functions as a regulator of Cdc42 ( and Rac1 ) , an effector of the small GTPase , or both , and whether the function is specific for a given cell-type or signaling pathway [31] . Collectively , our data suggest the following mechanism of LAI-1-mediated cell migration inhibition: LAI-1 ( directly or indirectly ) inhibits or prevents the activation of the Cdc42-specific GEF ARHGEF9 , which in turn prevents the IQGAP1-dependent activation of Cdc42 and perhaps also its stabilization by IQGAP1 ( Fig 9 ) . The scaffold protein IQGAP1 has been implicated in the interactions of a number of intracellular pathogens or their products with host cells . While intracellular replication of L . pneumophila in A549 cells is not affected by the depletion of IQGAP1 ( S9 Fig ) , invasion by Salmonella enterica Typhimurium ( S . typhimurium ) of embryonic fibroblasts from mice lacking IQGAP1 is decreased , and Cdc42 as well as Rac1 activation is abrogated [45] . S . typhimurium also subverts host cell motility and migration . The vacuolar pathogen promotes chronic infection by translocating through the SPI2 type III secretion system ( T3SS ) the effector SseI ( alias SrfH ) , which directly binds to IQGAP1 and inhibits the motility of immune phagocytes such as primary macrophages and dendritic cells [46 , 47] . Analogously , L . pneumophila abolishes cell migration in a manner dependent on the Icm/Dot T4SS . However , whereas the effector LegG1 , a Ran activator , was found to antagonize Icm/Dot-dependent cell migration inhibition , an effector inhibiting cell migration has not been identified yet [25] . In any case , L . pneumophila abolishes cell migration likely by an Icm/Dot-translocated effector protein as well as by the low molecular weight signaling molecule LAI-1 , and migration inhibition triggered by L . pneumophila or LAI-1 proceeds through a common host factor , Cdc42 ( Fig 8 ) . IQGAP1 is also involved in modulation of cell migration by obligate intracellular bacteria of the genus Chlamydia . Chlamydia pneumoniae promotes cell adhesion and migration of vascular smooth muscle cells through IQGAP1 [48] , while Chlamydia trachomatis impairs the migration of HeLa epithelial cells [49] and endosymbiotic environmental Chlamydia spp . control the motility of their host , Acanthamoeba sp . [50] . Finally , the AHL autoinducer N-3-oxo-dodecanoyl-L-homoserine lactone from P . aeruginosa has been implicated in migration inhibition of epithelial cells through an IQGAP1-dependent pathway . The AHL quorum sensing molecule alters cell migration by apparently interacting with IQGAP1 , which upon phosphorylation of the small GTPases Rac1 and Cdc42 alters its cellular localization [51] . Yet , in this study the autoinducer was effective only at very high concentrations ( 200 μM ) and further factors comprising the signaling pathway have not been identified . It is currently unknown under what physiological conditions and at which concentrations LAI-1 is produced by L . pneumophila or how the compound is secreted . Yet , the findings that overproduction of lqsA complements the defect in cell inhibition of an ΔlqsA mutant strain and leads to inhibition of cell migration by a ΔicmT mutant suggest that under these conditions LqsA ( and in consequence , LAI-1 ) are produced . Possibly , L . pneumophila also produces LAI-1 in biofilms . In this setting , the adherent cells could secrete the signaling compound , thus inhibiting protozoa migration and increasing the likelihood that the pathogen is taken up by a potential host cell . At this point , we do not have any evidence suggesting that LAI-1 increases infectivity . Rather , treatment of D . discoideum or RAW 264 . 7 macrophages with up to 10 μM LAI-1 did not affect the uptake of wild-type or ΔicmT mutant L . pneumophila ( S1 Fig ) . In summary , our study reveals that the L . pneumophila autoinducer LAI-1 inhibits the migration of eukaryotic cells in the low micromolar range through a signaling pathway involving the host factors IQGAP1 , Cdc42 and ARHGEF9 . These findings provide the basis for a further mechanistic analysis of how L . pneumophila impedes cell migration and benefits from this strategy .
The chemical synthesis of ( S ) -LAI-1 ( 5 ) ( Fig 10 ) started with commercially available ( S ) -2-hydroxybutyric acid 1 that was protected as a TBDPS ether by a two-step protocol to yield carboxylic acid 2 , analogous to the synthesis of 3-hydroxytridecane-4-one ( CAI-1 ) [6] . Formation of the Weinreb amide 3 followed by reaction with undecanemagnesium bromide provided ketone 4 . Final deprotection with TBAF and HPLC purification gave LAI-1 ( 5 ) in pure form ( Fig 10A ) . ( R ) -LAI-1 was synthesized analogously starting from ( R ) -2-hydroxybutyric acid . Enantiopure ( S ) -amino-LAI-1 ( 9 ) as its HCl-salt was synthesized as follows . ( S ) -N-Boc-Abu 6 was converted into the corresponding Weinreb amide 7 via HBTU activation , which was treated with undecanemagnesium bromide to provide the corresponding ketone 8 after silica gel chromatography . Next , the N-Boc-ketone derivative 8 was dissolved in anhydrous diethyl ether and deprotected via the treatment with 2 M HCl in diethyl ether . After evaporation to dryness , the resulting residue was recrystallized from anhydrous chloroform to give enantiopure crystalline ( S ) -amino-LAI-1 ( 9 ) as its HCl salt ( Fig 10B ) . ( R ) -amino-LAI-1 was synthesized analogously starting from ( R ) -N-Boc-Abu carboxylic acid . LAI-1 and its derivatives were dissolved in 100% dimethylsulfoxide ( DMSO ) as a stock solution for the experiments . For details of the chemical syntheses see section Supporting Information . L . pneumophila strains ( Table 1 ) were grown for 3 days on CYE agar plates [52] containing charcoal yeast extract , buffered with N- ( 2-acetamido ) -2-amino-ethanesulfonic acid ( ACES ) . Liquid cultures were inoculated in AYE medium—supplemented with chloramphenicol ( Cm 5μg ml-1 ) , if necessary—at an OD600 of 0 . 1 and grown at 37°C to an OD600 of 3 . 0 ( 21–22 h ) . D . discoideum strains were grown as described [53] . Murine RAW 264 . 7 macrophages and A549 lung epithelial carcinoma cells were cultivated in RPMI 1640 medium amended with 10% heat-inactivated fetal bovine serum and 1% glutamine ( all from Life Technologies ) . The infection of phagocytes with L . pneumophila was analyzed as described using D . discoideum , murine RAW 264 . 7 macrophages or A549 cells [10 , 53–56] . Briefly , cells were infected ( MOI 10 ) with L . pneumophila grown for 21–22 hours in AYE broth , the infection was synchronized by centrifugation [450 × g , 10 min , room temperature ( RT ) ] , and the infected phagocytes were incubated at 23°C ( D . discoideum ) or at 37°C/5% CO2 ( mammalian cells ) for the indicated time . Under-agarose assays using D . discoideum Ax3 pSW102 ( GFP ) were performed as described [25 , 57] . Briefly , microscopy dishes ( μ-Dish , 35 mm , Ibidi ) were filled with a mixture of melted agarose in SM medium [10 g bacteriological peptone ( Oxoid ) , 1 g Bacto yeast extract ( BD Biosciences ) , 1 . 9 g KH2PO4 , 0 . 6 g K2HPO4 , 0 . 43 g MgSO4 , 10 g glucose per liter , pH 6 . 5] . After solidification , 3 parallel slots of 2 × 4 mm ( for cells and chemo-attractant solution ) were cut 5 mm apart into the agarose ( Fig 1A ) . The chemo-attractant solution , 1 mM folic acid ( Sigma-Aldrich ) in SM medium , was filled into the central slot 30 min before the cell suspensions were filled into the neighboring slots . Prior to the experiments , 106 D . discoideum Ax3 pSW102 ( GFP ) cells were seeded onto 6-well plates overnight in HL-5 medium . The amoebae were washed once with MB medium [14 g bacteriological peptone ( Oxoid ) , 7 g Bacto yeast extract ( BD Biosciences ) , 4 . 26 g MES ( Sigma-Aldrich ) per liter , pH 6 . 9] and kept for 1 h in 3 ml MB medium . During this period LAI-1 , CAI-1 or derivatives were added at the concentrations indicated and DMSO was used as a negative control . If indicated , infections with L . pneumophila were performed in parallel at an MOI of 10 , for 1 h at 23°C . After 2 washing steps with MB medium ( 10 min centrifugation , 1500 rpm ) , the amoebae were detached by scratching into 500 μl MB , and 30 μl of the cell suspension was filled into the slots . The dishes were incubated in a humid chamber and the cells let migrate for 4 h at 23°C . Under-agarose assays using murine RAW 264 . 7 macrophages were performed as described [25 , 58] . The microscopy dishes ( μ-Dish , 35 mm; Ibidi ) were incubated with 10% FCS , for 30 min at RT . After washing twice with PBS , the dish was filled with 1% UltraPure agarose in a 1:1 mixture of RPMI/HBSS ( Life Technologies ) . Three parallel slots ( 5 mm part ) were formed using a template , and the chemo-attractant solution ( CCL5 , 100 ng ml-1 , Invitrogen ) was placed in the middle for 1 h min , before the cell suspensions were filled into the neighboring slots . Prior to the experiments 106 macrophages were seeded onto 6-well plates in RPMI and incubated overnight at 37°C . LAI-1 , CAI-1 or derivatives were added at the concentrations indicated for 1 h . If indicated , infections with L . pneumophila were performed in parallel at an MOI of 10 , for 1 h at 37°C . After two washing steps with RPMI , the cells were incubated for 45 min with 1 μM CellTracker BODIPY ( green ) , washed again twice and kept in 3 ml RPMI . The cells were detached by scratching into 500 μl RPMI , 150 μl was placed into the slots and let migrate for 4 h in a humid chamber at 37°C . Under-agarose cell migration of macrophages ( labelled with BODIPY ) or D . discoideum ( producing GFP ) was analyzed by fluorescence microscopy using a Leica TCS SP5 confocal microscope ( HCX PL APO CS 10×/0 . 40 dry UV objective , Leica Microsystems ) . The tile scan function of the Leica software allowed obtaining merged overview pictures . Cell migration was quantified using ImageJ software ( version 1 . 45 , function ‘plot profile’ ) . The fluorescence intensities of infected cells relative to uninfected cells were plotted against the migration distance . 100% fluorescence is defined as the maximum fluorescence intensity of the untreated control sample , i . e . uninfected cells or cells infected with wild-type or ΔicmT mutant L . pneumophila . Individual phagocytes ( D . discoideum or macrophages ) were tracked in the under-agarose assay using a SP5 confocal microscope ( HCX PL APO CS 40×/1 . 25 oil UV objective ) essentially and the ImageJ manual tracking plugin ( ‘chemotaxis and migration tool 2 . 0’ , Ibidi ) as described [25] . D . discoideum cells were filled into the slots , and monitored after 1 h for 15 min at 23°C within a 2 h time window by taking 1 frame per 25 s . RAW 264 . 7 macrophages were tracked at 2 h post infection at 37°C during another 1 h with 1 frame per 35 s . In vitro scratch assays were performed as described [25 , 49 , 59] . Briefly , A549 cells were seeded into 35 mm μ-Dishes ( Ibidi ) at a density of 1 . 5 × 105 cells ml−1 ( 3 × 105 cells/dish ) and incubated for 24 h . Confluent cell layers were washed with fresh medium and infected for 1 . 5 h with L . pneumophila strains ( MOI 10 ) and/or treated with 10 μM LAI-1 . After the infection and/or compound treatment , the cell layer was scratched with a sterile pipette tip and washed with fresh medium to remove detached cells . Images of the scratched positions were taken at time point zero and after 24 h using a Leica SP5 confocal microscope ( HCX PL APO CS 10×/ 0 . 40 dry UV objective ) . The percentage of ‘scratch closure’ was quantified using ImageJ software ( function ‘analyze particles’ ) by comparing the remaining scratch area with the initial cell-free scratch area ( S2 Fig ) . For RNAi experiments in scratch assays , A549 cells were grown in 35 mm μ-Dishes ( Ibidi ) and treated for 2 d with a final concentration of 10 nM of a mixture of 4 different siRNA oligonucleotides ( S3 Table ) . To this end , the siRNA stock ( 10 μM ) was diluted 1:15 in RNAse-free water , and 22 . 5 μl of diluted siRNA was added per well . Allstars siRNA ( Qiagen ) was used as a negative control . Subsequently , 181 . 9 μl RPMI medium without FCS was mixed with 5 . 6 μl HiPerFect transfection reagent ( Qiagen ) , added to the well , mixed and incubated for 5–10 min at RT . In the meantime , cells were diluted in RPMI medium with 10% FCS , 1 . 312 ml of the diluted cells ( 1 . 5 × 105 cells ml-1 ) were added on top of each siRNA-HiPerFect transfection complex and incubated for 48 h . After a washing step with fresh RPMI medium , cells were infected with L . pneumophila strains and/or treated with 10 μM LAI-1 , and the scratch assay was performed as described above . The depletion efficiency of all siRNA oligonucleotides was assessed by Western blot ( S3 Fig ) using antibodies against IQGAP1 , Cdc42 , RhoA , Rac1 , ARHGEF9 , ARHGAP17 , DOCK11 , FGD1 , GAP1 , CD2AP or GAPDH ( Abcam , 1:1 , 000 ) . For RNAi experiments in growth assays , A549 cells were grown in 96-well plates and treated for 2 d with a final concentration of 10 nM of the siRNA oligonucleotides indicated ( S3 Table ) . To this end , the siRNA stock ( 10 μM ) was diluted 1:15 in RNAse-free water , and 3 μl of diluted siRNA was added per well . Allstars siRNA ( Qiagen ) was used as a negative control . Subsequently , 24 . 25 μl RPMI medium without FCS was mixed with 0 . 75 μl HiPerFect transfection reagent ( Qiagen ) , added to the well , mixed and incubated for 5–10 min at RT . In the meantime , cells were diluted in RPMI medium with 10% FCS , 175 μl of the diluted cells ( 2 × 104 cells ) were added on top of each siRNA-HiPerFect transfection complex and incubated for 48 h . The cells were then infected ( MOI 10 ) with GFP-producing L . pneumophila grown for 21 h , diluted in RPMI , centrifuged and incubated for 1 h . The infected cells were washed 3 times with pre-warmed medium containing 10% FCS and incubated for 24 h ( the plate was kept moist with water in extra wells ) . Intracellular bacterial growth was measured by fluorescence using a plate reader ( FluoStar Optima , BMG Labtech ) . Exponentially growing D . discoideum Ax2 wild-type cells ( 1–4 × 106 cells/ml ) were either treated with 20 μM of synthetic LAI-1 in DMSO or mock-treated with DMSO only . After 3 h 1 × 107 cells were harvested , washed twice with Soerensen phosphate buffer ( 2 mM Na2HPO4 , 15 mM KH2PO4 , pH 6 . 0 ) , and total RNA was isolated ( Qiagen RNeasy mini kit ) using the protocol for isolation of cytoplasmic RNA . In total six microarrays with dye swaps for each isolation were analyzed with labelled cDNAs derived from three independent RNA isolations . Microarray analysis was essentially carried out as described [33 , 60] . Oligonucleotide primers for quantitative RT-PCR ( S4 Table ) were designed on the basis of sequence information , selected with the Primer 3 program ( http://bioinfo . ut . ee/primer3-0 . 4 . 0/primer3/ ) and purchased from Metabion Corp . ( Munich , Germany ) . Reverse transcription and RT-PCR were essentially performed as described [33] . Cdc42 ( GTP ) and Cdc42 ( GDP ) were identified in epithelial cells by pulldown experiments , followed by Western blot . To this end , A549 cells treated or not with 10 μM LAI-1 were suspended in ice cold RIPA buffer and incubated at 4°C for 10 min . Cellular debris was pelleted by centrifugation ( 10 min , 10 , 000 × g , 4°C ) . 1 mL of the supernatant ( 100–500 μg total cell protein ) was incubated for 1 h at 4°C with glutathione resin and GST-PBDPak1 ( the p21-binding domain ( PBD ) of p21-activated protein kinase ( PAK1 ) specifically binds to active Cdc42 ) according to the manufacturer’s recommendation ( Thermo Scientific ) . As controls , cell lysates were treated with GTPγS or GDP to yield the active or inactive form of Cdc42 . Subsequent Western blot was performed with an antibody recognizing Cdc42 ( GTP/GDP ) ( Abcam , 1:1 , 000 ) . Alternatively , the supernatant was transferred to a fresh centrifuge tube on ice , together with 20 μl of resuspended protein A/G PLUS agarose slurry ( Santa Cruz ) , incubated at 4°C for 30 min and pelleted by centrifugation ( 2 , 000 × g , 5 min , 4°C ) . The lysate was then incubated with primary anti-Cdc42 ( GTP/GDP ) antibody ( Abcam , 1:1 , 000 ) , 20 μl of resuspended AG PLUS agarose was added and incubated on a rotating device for 1 h at 4°C . Immuno-precipitates were collected by centrifugation ( 2 , 000 × g ) for 5 min at 4°C . The pellet was washed 4 times with 1 ml of RIPA buffer and after the final wash step resuspended in 40 μl of loading buffer . After boiling for 2–3 min , samples were subjected to SDS-PAGE and analyzed by Western blot using anti-Cdc42 ( GTP/GDP ) or anti-Cdc42 ( GTP ) antibodies ( Abcam , 1:1 , 000 ) . The amount of GAPDH , actin or α-tubulin in cell lysates was determined by Western blot using polyclonal antibodies ( Abcam , 1:1 , 000–1:2000 ) . For uptake experiments , D . discoideum ( 5 × 105 ) or RAW 264 . 7 macrophages ( 2 . 5 × 105 ) were infected ( MOI 10 ) for 1 h with GFP producing L . pneumophila wild-type or ∆icmT mutant bacteria and treated with different concentrations of LAI-1 ( 1 , 5 or 10 μM ) . Fluorescence of GFP-positive phagocytes was determined by flow cytometry [61] . To determine cytotoxicity of L . pneumophila strains or LAI-1 , D . discoideum or macrophages were seeded into 24 well plates , infected with L . pneumophila ( MOI 10 , 4 h ) and collected by scraping into 15 ml tubes . After centrifugation ( 240 × g , 10 min ) , the cells were resuspended in 500 μl SorC ( D . discoideum ) or PBS ( macrophages ) . Propidium iodide ( PI ) solution ( 2 . 5 μg μl-1 ) was added to the tubes , incubated for 10 min in the dark , and the PI-positive cells were analyzed by flow cytometry [61] . For immuno-fluorescence analysis , A549 cells were seeded on coverslips in a 24 well plate , treated or not with LAI-1 ( 10 μM ) and infected or not with L . pneumophila wild-type or ΔicmT strains . Cells were fixed with 3% paraformaldehyde for 15 min , washed three times with PBS , permeabilized with 0 . 1% Triton X-100 and blocked with 1% BSA . Cells were then incubated with antibodies against IQGAP1 , Cdc42 ( GTP/GDP ) or Cdc42-phospho-Ser71 ( Abcam; each 1:200 ) . The protein amount per cell was quantified using Image J software ( function “invert LUT” and “analyse/cell counter” ) . The microtubule network was analyzed with RAW 264 . 7 macrophages , infected or not with L . pneumophila ( MOI 10 , 1 h ) . After a washing step with BRB80 ( 80 mM PIPES , pH 6 . 8 , 1 mM MgCl2 , 1 mM EGTA ) the cells were fixed ( 50% BRB80 , 0 . 1% Triton X-100 , 0 . 5% glutaraldehyde ) for 5 min . Subsequently , the cells were washed with SorC ( Soerensen phosphate buffer containing 50 mM CaCl2 ) and blocked with 1 mg/ml sodium borohydrate in SorC for 10 min . The samples were stained with the anti-α-tubulin antibody WA3 ( gift from M . Schleicher ) . Appropriate secondary antibodies ( 1:200 ) were used . The number of microtubule fibers was counted along cross sections in the cell . Typically , four sections per cell were considered , and each peak represents one fiber ( S2 Fig ) . Actin was visualized in RAW 264 . 7 macrophages , seeded on coverslips in a 24 well-plate , using Texas red-phalloidin staining , followed by wash steps with PBS , permeabilization with cold 1% Triton X-100/PBS for 3–5 min and blocking with 1% BSA . Nuclei were stained with DAPI ( 0 . 1 μg/ml ) . The degree of actin re-localization was evaluation by visual inspection of single cells . If present , a layer of cortical actin was obviously visible , allowing scoring cells with or without cortical actin . The samples were analyzed with a Leica SP5 confocal microscope . Images were evaluated with ImageJ software . Further analysis was performed by using normalized background-subtracted band intensity values , defined as RIU ( Relative Intensity Units ) . All experiments were carried out in triplicates and significance was determined using an unpaired , two-tailed Student’s t test .
|
Legionella pneumophila is a ubiquitous environmental bacterium , which upon inhalation causes a severe pneumonia termed Legionnaires’ disease . The opportunistic pathogen employs the small molecule LAI-1 ( Legionella autoinducer-1 ) for bacterial cell-cell communication . LAI-1 is produced and detected by the Lqs ( Legionella quorum sensing ) system , which regulates a variety of processes including pathogen-host cell interactions . In this study , we analyzed whether LAI-1 not only plays a role for bacterial signaling but also modulates gene regulation and cellular responses of eukaryotic cells ( amoebae or macrophages ) . We discovered that the gene encoding the LAI-1 autoinducer synthase , lqsA , indeed promotes the inhibition of cell migration by L . pneumophila , and synthetic LAI-1 dose-dependently inhibits cell migration . LAI-1-dependent inhibition of cell migration required the scaffold protein IQGAP1 and the small GTPase Cdc42 , as well as the Cdc42 activator ARHGEF9 , but not other modulators of Cdc42 or small GTPases . Treatment with LAI-1 led to inactivation of Cdc42 and redistribution of IQGAP1 . In summary , our results reveal that the L . pneumophila signaling compound LAI-1 inhibits the migration of eukaryotic cells through a host signaling pathway comprising IQGAP1 , Cdc42 and ARHGEF9 .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[] |
2015
|
Inter-kingdom Signaling by the Legionella Quorum Sensing Molecule LAI-1 Modulates Cell Migration through an IQGAP1-Cdc42-ARHGEF9-Dependent Pathway
|
Glucagon , secreted from pancreatic islet α cells , stimulates gluconeogenesis and liver glycogen breakdown . The mechanism regulating glucagon release is debated , and variously attributed to neuronal control , paracrine control by neighbouring β cells , or to an intrinsic glucose sensing by the α cells themselves . We examined hormone secretion and Ca2+ responses of α and β cells within intact rodent and human islets . Glucose-dependent suppression of glucagon release persisted when paracrine GABA or Zn2+ signalling was blocked , but was reversed by low concentrations ( 1–20 μM ) of the ATP-sensitive K+ ( KATP ) channel opener diazoxide , which had no effect on insulin release or β cell responses . This effect was prevented by the KATP channel blocker tolbutamide ( 100 μM ) . Higher diazoxide concentrations ( ≥30 μM ) decreased glucagon and insulin secretion , and α- and β-cell Ca2+ responses , in parallel . In the absence of glucose , tolbutamide at low concentrations ( <1 μM ) stimulated glucagon secretion , whereas high concentrations ( >10 μM ) were inhibitory . In the presence of a maximally inhibitory concentration of tolbutamide ( 0 . 5 mM ) , glucose had no additional suppressive effect . Downstream of the KATP channel , inhibition of voltage-gated Na+ ( TTX ) and N-type Ca2+ channels ( ω-conotoxin ) , but not L-type Ca2+ channels ( nifedipine ) , prevented glucagon secretion . Both the N-type Ca2+ channels and α-cell exocytosis were inactivated at depolarised membrane potentials . Rodent and human glucagon secretion is regulated by an α-cell KATP channel-dependent mechanism . We propose that elevated glucose reduces electrical activity and exocytosis via depolarisation-induced inactivation of ion channels involved in action potential firing and secretion .
Blood glucose levels are under the control of two hormones released from the pancreatic islets of Langerhans . Islet β cells secrete insulin when glucose is high , decreasing glucose production by the liver and increasing glucose storage in multiple tissues . Regulated insulin secretion is relatively well understood , involving the metabolic stimulation of electrical activity , Ca2+ entry , and exocytosis [1] . Islet α cells secrete glucagon in response to decreased blood glucose , whereas elevated glucose levels suppress glucagon release . Glucagon is the principal factor stimulating glucose production by the liver . In diabetes , baseline glucagon release is elevated , and glucagon secretion in the low-glucose condition is blunted [2–4] . These effects contribute to chronic hyperglycaemia and to an increased risk for acute hypoglycaemic events . The mechanism regulating glucagon secretion is poorly understood and remains hotly debated [5] . Glucagon release in rodents may be regulated by paracrine signals , including γ-aminobutyric acid ( GABA ) [6 , 7] , Zn2+ [8] , and insulin [9 , 10] . Conversely , glucose may suppress glucagon secretion through a direct effect on α-cell activity [11–14] . There are also studies suggesting that glucagon secretion is under hypothalamic control [15 , 16] . Human in vivo studies provide conflicting evidence regarding the control of glucagon secretion by paracrine or intrinsic regulation of α cells [17–30] , and very little work has been done to examine this question in isolated human islets . Islet α cells express ATP-dependent K+ ( KATP ) channels [9 , 13 , 14 , 31 , 32] that can be closed by ATP [9 , 31] . Glucose increases intracellular free ATP in α cells [8 , 10] , although reports vary as to the ability of glucose to inhibit α-cell KATP channels [13 , 31 , 33] . Evidence from SUR1−/− mice implicate KATP channels as regulators of glucagon release [13 , 34 , 35] . However , because both α and β cells possess molecularly identical KATP channels [36 , 37] , it is not clear how KATP-mediated depolarisation would stimulate insulin secretion but suppress glucagon secretion . The answer may lie in the downstream machinery regulating electrical activity and Ca2+ entry . Unlike β cells , α cells possess a large voltage-dependent Na+ current that is essential for glucagon release [14 , 33] . We have previously proposed that the depolarisation-induced inactivation of this channel contributes to the cessation of action potential firing [14] . Additionally , activation of voltage-dependent Ca2+ channels ( VDCCs ) is essential for Ca2+ entry and α-cell function [38 , 39] . Accordingly , the α-cell intracellular Ca2+ ( [Ca2+]i ) oscillations ( reflecting α-cell electrical activity ) are suppressed in parallel with glucagon release by glucose [40] . Multiple VDCCs regulate glucagon release [14 , 41–43] , although the N-type channels appear to be particularly important for glucagon release evoked by hypoglycaemia alone [33 , 44 , 45] , at least in mouse islets . This is in contrast to the β cell in which the L-type VDCC functionally predominates [46] . We have now compared insulin and glucagon release and α- and β-cell Ca2+ responses in intact mouse , rat , and human pancreatic islets . We show that glucose retained the ability to suppress glucagon release from isolated islets during blockade of the Zn2+ and GABA paracrine pathways , and in the absence of stimulated insulin secretion or β-cell Ca2+ responses . Thus we now provide evidence in both rodent and human islets supporting the direct ( intrinsic ) glucose regulation of glucagon release from pancreatic α cells .
To examine a role for GABA and Zn2+ as paracrine mediators of glucagon secretion , we examined the ability of the GABAA receptor antagonist SR-95531 and Zn2+ chelation with Ca2+-EDTA to prevent the glucose-dependent suppression of glucagon release . Glucose , at concentrations ( 7 and 8 . 3 mM ) just above the threshold for insulin release ( see below ) , suppressed glucagon secretion from isolated mouse ( Figure 1A ) and rat islets ( Figure 1B ) by 60% ( n = 15 , p < 0 . 001 ) and 57% ( n = 10 , p < 0 . 001 ) , respectively . In both mouse and rat islets , the ability of glucose to inhibit glucagon secretion persisted in the presence of Ca2+-EDTA ( 43% and 48% , n = 10 , p < 0 . 001 , respectively ) and SR-95531 ( 31% and 46% , n = 8 and 10 , p < 0 . 01 and p < 0 . 001 , respectively ) ( Figure 1 ) . It is worth noting that in the presence of the GABAA antagonist , glucagon secretion was increased under both low- and high-glucose conditions , and furthermore , glucose was approximately 50% less effective in suppressing glucagon release from mouse islets ( 31% versus 60% , respectively ) ( Figure 1A ) . Additionally , somatostatin released from pancreatic δ cells is suggested to be a potential paracrine regulator of glucagon secretion . However , the somatostatin receptor 2 ( SSTR-2 ) antagonist PRL-2903 does not interfere with the ability of glucose ( at 3 and 7 mM ) to inhibit glucagon secretion from mouse islets [47] . Like the GABAA receptor antagonist , however , PRL-2903 increased glucagon secretion in low-glucose conditions [47] . Thus , although the present data do not entirely rule out these pathways as modulators of glucagon secretion , glucose is clearly able to suppress glucagon secretion independently of these . We next examined glucagon and insulin secretion during pharmacological manipulation of KATP channel activity . Here we have examined the role of KATP channels in intact islets by applying an indirect , but minimally invasive , technique . It would have been difficult ( if not impossible ) to study the effects of glucose on α-cell KATP channel activity using the patch-clamp technique because of the smallness of the α-cell resting conductance ( 0 . 15 nS/pF in the absence of glucose , of which two thirds is attributable to KATP channels ) [13] . We have instead used increasing concentrations of diazoxide and tolbutamide to “titrate” the influence of KATP channel activity on α-cell [Ca2+]i and glucagon secretion . Increasing concentrations of the KATP channel activator diazoxide demonstrated that moderate activation of KATP channels ( 0 . 3–10 μM diazoxide ) relieved the suppression of glucagon secretion from both mouse ( Figure 2A ) and rat ( Figure 2B ) islets . Stimulation of glucagon release was half-maximal at approximately 1 μM diazoxide , which is well below that required to inhibit insulin release , suggesting that “re-activation” of glucagon release was not secondary to reduced β-cell secretion . Increasing the concentration of diazoxide beyond 10 μM inhibited glucagon secretion in parallel with an inhibition of insulin release from both mouse and rat islets ( Figure 2A and 2B ) . When instead applied in the presence of 1 mM glucose ( at which concentration glucagon secretion is stimulated ) , increasing diazoxide produced a monotonic inhibition of glucagon secretion ( Figure 2C ) with a half-maximal inhibitory concentration ( IC50 ) that is much lower than what is seen under high-glucose conditions ( ∼2 μM versus ∼50 μM ) . In the complete absence of glucose , when KATP channels are expected to be open , the KATP channel antagonist tolbutamide also produced a biphasic effect on glucagon release . Augmentation of glucagon secretion was seen at tolbutamide concentrations of up to 1 μM ( stimulation being half-maximal at 0 . 1 μM ) , whereas inhibition was observed at higher concentrations ( Figure 2D ) . Importantly , in the presence of a maximally inhibitory tolbutamide concentration ( 0 . 5 mM ) , glucose was unable to produce any further inhibition of glucagon release ( Figure 3A ) . These data are inconsistent with the idea that β-cell secretion is the primary inhibitor of α-cell glucagon release . They also suggest that glucagon secretion is maximal within a “window” of intermediate KATP channel activity . To further investigate the role of α-cell KATP channels as regulators of glucagon secretion , we studied islets from Kir6 . 2Y12X mice , which posses a Tyr12STOP mutation in the Kcnj11 gene , leading to premature termination of the KATP channel pore-forming subunit [48] . At low-glucose concentrations , glucagon secretion from the Kir6 . 2Y12X islets was already suppressed compared with that from wild-type islets ( Figure 3B ) , similar to the effect of glucose stimulation or pharmacological KATP channel inhibition observed in Figures 2D and 3A . Consistent with a recent report [49] , higher glucose levels stimulated glucagon secretion , perhaps due to a direct effect of metabolism on secretion [33] . In wild-type islets , glucagon secretion exhibited a nadir at 5 mM glucose , and glucagon secretion at 20 mM glucose was 40% higher than at 5 mM glucose . Insulin secretion in the Kir6 . 2Y12X islets was elevated at low-glucose concentrations ( Figure 3C ) , whereas glucose-stimulated insulin secretion was blunted . It is notable that the increase in glucagon release from the Kir6 . 2Y12X islets was coincident with increased insulin secretion , reinforcing the view that inhibition of glucagon is not mediated by a factor released by the β cells . Furthermore , in control islets , the inhibition of glucagon secretion is maximal at a concentration of glucose ( 5 mM ) that is without stimulatory action on insulin secretion . We next examined the function of α and β cells in situ by monitoring the [Ca2+]i responses of single cells within intact mouse islets . These were functionally identified by their [Ca2+]i response to 0 . 5 , 2 , and 11 mM glucose [50] . Glucose stimulation suppressed the [Ca2+]i response of α cells by 51 ± 3% ( n = 63 cells in 15 islets , p < 0 . 001 ) ( Figure 4A and 4B ) . The suppressive effect of glucose was alleviated ( to 101 ± 11% of initial values ) by application of 2 μM diazoxide ( Figure 4A and 4B ) . This concentration of diazoxide was similar to that which produced maximal stimulation of glucagon secretion and much ( ∼15-fold ) lower than necessary to inhibit insulin secretion ( Figure 2B ) or β-cell [Ca2+]i responses to glucose ( Figure 4A ) . Little work has been done to examine the mechanism of glucose-regulated glucagon secretion from isolated human islets [51] . Glucagon and insulin secretion from islets isolated from healthy donors was examined . Similar to above , glucagon secretion from human islets was suppressed by 58 ± 3% ( n = 9 , p < 0 . 05 ) upon raising the glucose concentration from 0 to 10 mM . Tolbutamide ( 200 μM ) inhibited glucagon release by 43 ± 14% ( n = 7 , p < 0 . 05 ) . At the same time , insulin secretion was increased 6 . 4- and 4 . 4-fold by glucose and tolbutamide stimulation , respectively ( Figure 5A ) . Moderate re-activation of KATP channels with 2 μM diazoxide antagonized the glucose-induced suppression of glucagon release ( n = 7 , p < 0 . 01 ) ( Figure 5B ) . A 10-fold higher concentration of diazoxide had significantly less stimulatory effect ( p < 0 . 001 versus 2 μM diazoxide ) , and in the presence of 200 μM diazoxide , glucagon secretion was not different from that observed in the absence of the KATP channel activator . Importantly , the lowest concentration of diazoxide ( 2 μM ) , which produced the strongest stimulation of glucagon secretion , had no inhibitory effect on glucose-stimulated insulin secretion ( n = 5 ) , whereas 20 and 200 μM produced partial or complete inhibition of insulin secretion . The GABAA receptor antagonist SR95531 had no effect on glucose-induced inhibition of glucagon secretion from human islets , and whole-cell voltage-clamp measurements indicate that human α cells , unlike human β and δ cells , express few if any GABAA receptor Cl− channels ( M . Braun , R . Ramracheya , and P . Rorsman , unpublished data ) . Examination of the Ca2+ responses of single α and β cells within intact human islets demonstrated that glucose decreased [Ca2+]i by 62 ± 3% ( n = 42 cells in 7 islets , p < 0 . 001 ) in α cells , and that this effect could be completely reversed by 2 μM diazoxide ( Figure 6A and 6B ) . Furthermore , the re-activation of α-cell [Ca2+]i responses by diazoxide could be prevented by application of 100 μM tolbutamide , confirming the role of KATP channels ( Figure 6C ) . An analysis of the dose–response relationship to diazoxide demonstrated a maximally effective concentration of 1 . 7 μM ( Figure 6D and 6E; n = 38 cells in 9 islets ) , far below the concentration necessary to block β-cell responses ( Figure 6D ) . Increases in diazoxide above 20 μM blocked the [Ca2+]i responses of α cells and β cells in parallel , consistent with the effect on glucagon and insulin release . The above data suggest that glucose-dependent inhibition of α-cell KATP channels is involved in the suppression of glucagon secretion . It is not clear however , how KATP channel inhibition and membrane depolarisation result in suppression of secretion . Unlike in β cells , α-cell Na+ channels are active in the physiological range of membrane potentials . Previous work from our group suggested that the voltage-dependent inactivation of Na+ channels , which in mouse α cells is half-maximal at −42 mV [43] , contributes to cessation of electrical activity upon α-cell depolarisation [14] . We thus used the voltage-dependent Na+ channel antagonist tetrodotoxin ( TTX ) to test a role for these channels in α-cell function and glucagon secretion . TTX ( 0 . 1 μg/ml ) suppressed glucagon release from mouse islets by 56 ± 5% ( n = 10 , p < 0 . 001 ) under low-glucose conditions ( Figure 7A ) . This effect of TTX in the low-glucose condition was similar to what we observed with high-glucose stimulation alone . With TTX present , glucose was without further inhibitory action , suggesting that glucose inhibits glucagon secretion through a Na+ channel-dependent mechanism . TTX has no effect on glucose-stimulated insulin secretion in mouse islets ( Figure 7A ) ; again suggestive of a direct rather than indirect effect on α cells . In accordance with this , application of 0 . 1 μg/ml TTX to mouse islets reversibly abolished the α-cell [Ca2+]i response evoked by low-glucose concentrations ( Figure 7B and 7C; n = 18 cells in 4 islets ) , but had no effect on β-cell [Ca2+]i ( unpublished data ) . Downstream of Na+ channel activation , the opening of VDCCs allows Ca2+ into α cells , triggering the exocytosis of glucagon-containing vesicles . We applied whole-cell patch-clamp recordings to establish the α-cell Ca2+ channel complement . The integrated whole-cell Ca2+ current measured during 50-ms depolarisations from −70 mV to 0 mV amounted to 6 . 2 ± 0 . 8 pC ( n = 15 ) under control conditions , 2 . 6 ± 0 . 8 pC ( n = 10; p < 0 . 05 ) in the presence of 50 μM nifedipine , and 4 . 1 ± 0 . 5 pC ( n = 12; p < 0 . 01 ) in the presence of 1 μM ω-conotoxin . Thus , L- and N-type Ca2+ channels account for 58% and 34% of the α-cell Ca2+ current , respectively . Figure 8A compares glucagon secretion at 1 and 20 mM glucose under control conditions and in the presence of 100 nM ω-conotoxin and 20 μM nifedipine , respectively . We found that whereas the L-type channel blocker nifedipine had no effect on glucagon release at 1 or 20 mM glucose , the N-type Ca2+ channel blocker ω-conotoxin inhibited the release of the hormone to a level similar to that of high glucose , tolbutamide , and TTX . Furthermore , glucose exerted no additional inhibitory action in the presence of ω-conotoxin . Thus , glucagon secretion stimulated by low-glucose levels depends principally on Ca2+ influx through N-type channels , although these only account for one third of the Ca2+ current . We confirmed this Ca2+ channel dependence by conducting high-resolution single-cell capacitance measurements of exocytosis ( Figure 8B ) . Exocytosis elicited by 500-ms step depolarisations from −70 to 0 mV averaged approximately 150 fF ( corresponding to the fusion of ∼75 secretory granules with the plasma membrane [43] ) under control conditions ( Ctrl ) . This response was not significantly affected by 50 μM nifedipine ( nif ) , but was nearly abolished by 1 μM ω-conotoxin ( ω-con; Figure 8C ) . We next examined the mechanism by which N-type channels regulate glucagon release in response to membrane depolarisation/hyperpolarisation in mouse α cells . Non–L-type ( isradipine-resistant ) Ca2+ currents were elicited by step depolarization to 0 mV in the absence ( black ) or presence ( red ) of 1 μM ω-conotoxin following a 200-ms conditioning pulse to −70 mV ( Figure 9A ) or +10 mV ( Figure 9B ) . It is evident that the ω-conotoxin–sensitive component was abolished by the +10 mV conditioning pulse . Figure 9C summarizes the relationship between the conditioning voltage and the Ca2+ current amplitude in the absence ( open squares ) and presence ( circles ) of ω-conotoxin ( upper panel ) . The ω-conotoxin–sensitive N-type Ca2+ current component is shown in the lower panel of Figure 9C , and underwent voltage-dependent inactivation that was half-maximal ( V0 . 5 ) at −31 ± 6 mV ( n = 6 ) . A residual , ω-conotoxin–insensitive Ca2+ current also appears to undergo voltage-dependent inactivation ( Figure 9C ) . This accounts for less than 15% of the inactivating current , and may be attributable to T-type Ca2+ channels that have been reported in α cells [14] . In Figure 9D , we examined the voltage-dependent activation of α-cell exocytosis using step-wise depolarisations ( 500 ms ) from −70 mV to between −50 and 20 mV ( Figure 9D ) . The capacitance response versus voltage relationship ( Figure 9E ) demonstrates a marked increase in the capacitance response between −10 and 10 mV . Thus any reduction in action potential amplitude within this range would severely attenuate α-cell exocytosis . The voltage-dependent inactivation of α-cell exocytosis was examined next ( Figure 9F ) . Cells were held at conditioning potentials between −70 and −30 mV , after which exocytosis was elicited by depolarisation to 0 mV . Exocytosis stimulated from the −30 mV conditioning potential was only 25% of that elicited from −70 mV ( Figure 9G ) . This voltage-dependent decline in exocytosis is attributable to the voltage-dependent inactivation of the N-type Ca2+ channels . The fact that inhibition of exocytosis appears to occur at more-negative voltages than that documented for the inactivation of the N-type Ca2+ channels can be attributed to the brevity of the conditioning pulses in Figure 9A–9C ( 200 ms ) , whereas in Figure 9G the holding potential was varied .
Regulated glucagon secretion from pancreatic α cells is a major component of the counter-regulatory response to hypoglycaemia . Diabetes mellitus is associated with defects of glucagon secretion that exacerbate the consequences of impaired insulin secretion [2–4] . The mechanism regulating the release of this important hormone is incompletely understood and currently the source of much debate . In the present study we examined glucagon secretion and the [Ca2+]i response of in situ α cells of isolated rodent and human islets . We further characterised the Ca2+ currents and capacitance changes of single isolated α cells , to dissect the mechanism regulating downstream control of glucagon release . Previous work with genetic mouse models , including SUR1−/− [13 , 34 , 35] and Kir6 . 2−/− [16] mice , suggests an important role for KATP channels in glucagon secretion . Three pieces of evidence argue for the importance of islet KATP channels in glucagon secretion . First , introduction of the Tyr12STOP mutation into the KATP channel subunits in the Kir6 . 2Y12X islets results in suppression of glucagon secretion at low-glucose levels and causes loss of glucose-induced inhibition of secretion . Second , the KATP channel inhibitor tolbutamide when applied at maximally effective concentrations inhibits glucagon secretion from isolated islets . Third , a low dose of the KATP channel activator diazoxide restores α-cell Ca2+ responses and glucagon secretion in high-glucose conditions . What is not immediately clear from these arguments , and from the previous SUR1−/− genetic studies , is whether glucagon is being controlled by the α-cell or the β-cell KATP channels , as the latter may regulate α-cell function indirectly via paracrine pathways . We therefore examined insulin and glucagon secretion , and the Ca2+ responses of α and β cells in situ in parallel to determine the functional relationship between these cells . Furthermore , we measured the glucagon response to glucose during blockade of putative paracrine signalling pathways . Studies on rat islets support an important role for paracrine signals as regulators of glucagon release [33 , 52 , 53] , whereas the case for paracrine regulation of glucagon secretion from mouse and human α cells is less clear [9 , 10 , 14 , 17–21 , 27] . Although it is clear that Zn2+ , GABA , and somatostatin can exert a paracrine control of glucagon secretion under certain conditions , the data shown here firmly establish that glucose can suppress glucagon secretion independently of these pathways as demonstrated in Figure 1 ( and in [47] ) . Furthermore , and in agreement with recently published results [47] , maximal inhibition of glucagon release occurs at levels equal to or lower than 5 mM glucose , whereas the stimulation of insulin release requires glucose levels greater than 5 mM ( Figure 3 ) , suggesting that products of β-cell secretion are not required for suppression of glucagon release . This conclusion is further underpinned by the significant discordance between insulin and glucagon release , and the β- and α-cell Ca2+ responses , under several conditions: ( 1 ) low doses of the KATP channel opener diazoxide that stimulate glucagon secretion while not affecting insulin release; ( 2 ) high doses of diazoxide at which suppression of β-cell secretion would be expected to elevate glucagon release; ( 3 ) in the Kir6 . 2Y12X islets; and ( 4 ) in response to the Na+ channel blocker TTX . Therefore , our data argue that glucagon-producing α cells possess an intrinsic mechanism for regulation by glucose and that involves KATP channels . This is at variance with the data of Liu et al . [12] who report that tolbutamide has no effect on [Ca2+]i in single , isolated α cells , but in agreement with the conclusion of Ostenson et al . [54] , that sulfonylureas can inhibit glucagon secretion by a direct , non-paracrine mechanism . On the basis of our findings , we propose that α-cell glucagon secretion occurs within a narrow window of intermediate KATP channel activity ( and thus membrane potential ) ( Figure 10 ) . That is , if the α cell is either too hyperpolarised ( maximal KATP activity ) or too depolarised ( maximal KATP inhibition ) , then glucagon secretion is suppressed . This is supported by the biphasic effects of both diazoxide and tolbutamide . Whereas the former ( in high glucose ) brings the α cell in a dose-dependent manner through the membrane potential window supporting glucagon secretion in the depolarised ( low KATP ) to hyperpolarised ( high KATP ) direction ( Figure 10C ) , the latter ( in zero glucose ) brings the α cell through the window in the opposite direction ( from hyperpolarised to depolarised ) ( Figure 10B ) . Thus , glucose likely leads to the suppression of glucagon secretion by depolarising the α-cell membrane potential above the range that supports glucagon secretion ( Figure 10A ) . Indeed , in some ( but not all , see [55] ) studies , glucose was found to depolarize mouse and rat α cells and reduce action potential amplitudes [13 , 52] . It is interesting that whereas low concentrations of glucose were without stimulatory effect , tolbutamide ( 0 . 1–1 μM ) stimulated glucagon release beyond that observed at zero glucose . Thus , small depolarisations ( as previously documented for arginine [13] ) exert a positive “chronotropic” effect in α cells and thus stimulate glucagon release . The fact that glucose did not share this ability indicates that the depolarisation produced by 1 mM glucose results in sufficient inactivation of the currents to balance any increase in action potential frequency . Previous human studies have employed pharmacological KATP channel antagonists [23 , 24] or agonists [25 , 26] to examine the regulation of glucagon release in vivo . These were interpreted with the assumption that pharmacological KATP modulation only affects the β cells , and that any change in glucagon release was therefore secondary to altered β-cell function . Our data establish that KATP channel modulation has dramatic and direct effects on glucagon secretion and Ca2+ signalling in human α cells under conditions in which insulin secretion is unaffected . Thus , the in vivo manipulation of α-cell KATP channel activity in the above studies may well have involved direct effects on α-cell function that contributed to the observed changes in glucagon release . The importance of the human KATP channel pathway for glucagon release is nicely highlighted by a study investigating the effects of the common Glu23Lys ( E23K ) polymorphism of the Kir6 . 2 subunit of the KATP channel on insulin and glucagon secretion in non-diabetic human patients [29] . This variant of the channel leads to a slight decrease in the ATP sensitivity of the channel . The functional significance of this was examined by comparing hormone release during hyperglycaemic clamps in individuals carrying the polymorphism or not . Although insulin secretion in homozygous Glu23Lys individuals was not different from controls , glucose-induced suppression of glucagon release was blunted [29] . This becomes understandable in light of the effect of diazoxide on isolated human islets ( Figures 4 and 5 ) . Half-maximal activation of KATP channels occurs at diazoxide concentrations of 20–100 μM ( depending on the intracellular ATP level ) [56] . If the Glu23Lys polymorphism increases KATP channel activity to the same extent as 0 . 3–1 . 5 μM diazoxide , the concentration at which an effect on glucagon release is first seen , then the effect will be very difficult to detect with electrophysiology , perhaps explaining why some studies have failed to detect a functional effect of the polymorphism ( reviewed in [57] ) . Nevertheless , such small changes can have significant biological effects , as illustrated by the glucagon release data , and may contribute to pathological states such as impaired glucose tolerance and diabetes . Thus , the reduced ability of glucose to inhibit glucagon secretion in individuals carrying the Glu23Lys variant of the KATP channel likely results from the failure of these channels to undergo complete inhibition in response to glucose . We have proposed that glucagon secretion is stimulated within a window of intermediate α-cell KATP channel activity , and that glucagon release is suppressed by either increases or decreases in KATP channel activity . How is this accomplished ? Briefly , we suggest that this window is the result of ( 1 ) the ability of intermediate KATP channel activity to support regenerative electrical responses through the activation of voltage-dependent Na+ and N-type Ca2+ channels ( grey in Figure 10 ) ; ( 2 ) the failure of the Na+ and Ca2+ channels to activate when α cells are hyperpolarised by the opening of a major fraction of KATP channels ( above the grey in Figure 10 ) ; and ( 3 ) the voltage-dependent inactivation of the Na+ [14] and N-type Ca2+ channels when KATP channels are closed and the α-cell membrane potential is depolarised ( below the grey in Figure 10 ) . Thus , the differential responsiveness of α and β cells to diazoxide does not result from differential sensitivities of the KATP channels in these cells , but to the downstream responses to titrated KATP channel activity . One important difference between mouse α and β cells is that whereas the latter rely exclusively on voltage-gated Ca2+ channels for the upstroke of the action potentials , glucagon-producing α cells are equipped with voltage-gated Na+ channels . These channels undergo voltage-dependent inactivation at voltages more positive than −50 mV [14] . This will reduce the action potential amplitude , and indeed it is reported that glucose reduces the peak voltage in α cells from +11 mV to −1 mV [13] . This will in itself result in an approximately 35% reduction of exocytosis , which is steeply dependent on voltage between −10 and +10 mV ( Figure 9E ) . The functional significance of this is illustrated by the observations that glucagon secretion and α-cell Ca2+ responses at low-glucose concentrations are suppressed by the Na+ channel blocker TTX . The exact ion channel complement of human α cells remains to be established . However , the fact that glucagon secretion from human islets shows the same bell-shaped diazoxide concentration dependence as in mouse islets suggests that the depolarization-induced inactivation of ion channels involved in α-cell regenerative electrical activity underlies glucose-induced suppression of glucagon secretion in man as well . Although L-type VDCCs mediate the majority of Ca2+ current in α cells , this work and previous studies [33 , 44 , 45] demonstrate that it is the N-type ( ω-conotoxin–sensitive ) VDCCs that mediate the Ca2+ influx necessary for α-cell exocytosis and glucagon secretion under hypoglycaemic conditions . We now show that the α-cell N-type VDCCs are also subject to voltage-dependent inactivation at voltages more positive than −50 mV and furthermore that this is associated with reduced exocytotic capacity . It is pertinent that N-type VDCC-deficient mice exhibit reduced serum glucagon levels and improved glucose tolerance despite a parallel reduction in plasma insulin [44] . Although Ca2+ influx through L-type Ca2+ channels does not appear to contribute much to glucagon secretion under the experimental conditions used in this study , these channels become the predominant conduit of Ca2+ entry in the presence of agents increasing cAMP [41] ( unpublished data ) . The mechanism underlying this switch in Ca2+ channel dependence remains obscure , but may depend on the strength of depolarisation because glucagon secretion stimulated by strong depolarisation with increased K+ in combination with KATP channel block can be prevented by inhibition of L-type Ca2+ channels [47] . We finally point out that the model we propose here for the control of glucagon secretion shares many features with what is known about the metabolic control of insulin secretion . This in turn means that processes that interfere with the ability of , for example , glucose to stimulate insulin secretion will have the opposite effect on glucagon secretion . This would provide a simple explanation for the fact that both insulin and glucagon secretion become perturbed in diabetes and why oversecretion of glucagon exacerbates the hyperglycaemic effects of insufficient insulin secretion .
Islets from female NMRI mice were isolated by collagenase digestion and cultured in RPMI-1640 media ( 5 mM glucose ) at 37 °C and 5% CO2 for 2−24 h prior to secretion or intracellular Ca2+ assays . For single-cell studies , islets were dispersed in a Ca2+-free buffer and plated in 35-mm plastic dishes . Generation of the Kir6 . 2Y12X mice , which possess a Tyr12STOP mutation in the Kcnj11 gene on a BALB/c background , has been described previously [48] . This results in nonfunctional KATP channels in both α and β cells . These animals are not overtly diabetic , exhibiting normal fasting glucose , but are glucose intolerant ( R . Cox , unpublished data ) . For these experiments , age- and weight-matched wild-type C3HB mice were used as controls because the Kir6 . 2Y12X mice were back-crossed into this background over several generations . Human islets from four healthy donors were obtained from the Oxford DRWF Human Islet Isolation Facility . Experiments were performed in duplicate or triplicate using islets from each donor . Human islets were cultured in RPMI-1640 ( 10 mM glucose ) at 37 °C and 5% CO2 , and experiments were performed within 48 h of isolation . Intact islets were loaded with fluo-4-AM ( 1 μM ) and fura red ( 5 μM ) in 0 . 5 mM glucose buffer ( see below ) with 0 . 01% pluronic acid for 30 min at 37 °C . Islets were fixed with a wide-bore holding pipette within a continuously superfused and temperature-controlled ( 37 °C ) bath on an Axioskop 2 FS-mot microscope ( Carl Zeiss , http://www . zeiss . com ) . The perfusion buffer contained ( in mM ) : 140 NaCl; 3 . 6 KCl; 2 NaHCO3; 0 . 5 NaH2PO4; 0 . 5 MgSO4; 5 HEPES; 2 . 5 CaCl2; 0 . 5 , 3 , or 11 glucose; ( pH 7 . 4 with NaOH ) . Laser scanning confocal microscopy was performed using an LSM 510meta system ( Zeiss ) . Excitation was with a 488-nm argon laser , and emitted fluorescence was collected through 500–550-nm and 650–710-nm band-pass filters for the fluo-4 and fura red ( FuraR ) signals , respectively . Increases in intracellular Ca2+ are displayed as upward deflections . Images were acquired at 1 . 5-s intervals . Individual cells were selected as regions of interest , and the average ratio intensity ( RI = [Fluo+1]/[FuraR+20] × 128+1 ) of these were analysed over time with Origin v7 . 0220 ( OriginLab Corporation , http://www . originlab . com ) . Prior to experimental recordings , islets were perfused for 10 min with the appropriate control solution , and Ca2+ responses were monitored periodically during this time to ensure that responses ( or lack thereof ) were stable prior to beginning the experimental recording . Ca2+ responses were determined by baseline subtraction and calculation of the integrated response ( i . e . , the area under the curve ) . The slope of this response , reported here as arbitrary units ( AU ) , was calculated for the final 60%–90% of a given treatment period to allow for equilibration of the responses . Insulin and glucagon secretion were measured as described elsewhere [45] . Briefly , batches of ten freshly isolated islets were pre-incubated in 1 ml of Krebs–Ringer buffer ( KRB ) supplemented with 1 mM glucose for 30 min ( rodent ) or 1 h ( human ) followed by 1-h incubation in 1 ml of test KRB medium supplemented as indicated . For experiments in which Zn2+ was chelated or GABAA receptors were antagonised ( Figure 1 ) , the Ca2+-EDTA and SR-95531 , respectively , were present in both the pre-incubation and test solutions . When extracellular KCl was increased , NaCl was correspondingly reduced to maintain iso-osmolarity . Whole-cell currents and exocytosis were recorded using an EPC-9 patch-clamp amplifier ( HEKA Electronics , http://www . heka . com ) and Pulse software ( version 8 . 50 ) . Single α cells were identified by their small size and Na+ current inactivation properties [58] . This method was validated by combining the electrophysiological recordings with subsequent immunostaining for glucagon and insulin . The extracellular medium contained ( in mM ) : 118 NaCl; 20 TEA-Cl ( tetraethylammonium chloride ) ; 5 . 6 KCl; 2 . 6 CaCl2; 1 . 2 MgCl2; 5 HEPES ( pH 7 . 4 with NaOH ) ; and 5 glucose . Except for Figure 8B and 8C , in which the standard whole-cell configuration was used , the electrophysiological measurements were conducted using the perforated patch technique , and the pipette solution contained ( in mM ) : 76 Cs2SO4; 10 NaCl; 10 KCl; 1 MgCl2; and 5 HEPES ( pH 7 . 35 with CsOH ) . Exocytosis was monitored as changes in α-cell capacitance using the software-based lock-in function of the Pulse software . The standard whole-cell measurements ( Figure 8B and 8C ) were conducted using a pipette solution ( dialyzing the cell interior ) consisted of ( in mM ) 125 CsOH; 125 glutamate; 10 CsCl; 10 NaCl; 1 MgCl2; 5 HEPES ( pH 7 . 15 with KOH ) ; 3 Mg-ATP; and 25 μmol/l EGTA ( measured resting free Ca2+ , ∼0 . 2 μmol/l ) . Pulses were applied at low frequency ( <0 . 05 Hz ) to allow the exocytotic capacity to recover fully between the pulses . Data are presented as means and standard errors . Significance was examined by either the unpaired t-test or by multiple-comparisons analysis of variance ( ANOVA ) and post-test , as appropriate .
|
Glucagon is a critical regulator of glucose homeostasis . Its major action is to mobilize glucose from the liver . Glucagon secretion from α cells of the pancreatic islets of Langerhans is suppressed by elevated blood sugar , a response that is often perturbed in diabetes . Much work has focused on the regulation of α-cell glucagon secretion by neuronal factors and by paracrine factors from neighbouring cells , including the important islet hormone insulin . In contrast , we provide evidence in support of a direct effect of glucose on α cells within intact rodent and human islets . Notably , our work implicates an α-cell glucose-sensing pathway similar to that found in insulin-secreting β cells , involving closure of ATP-dependent K+ channels in the presence of glucose . Furthermore , we find that membrane depolarisation results in inhibition of Na+ and Ca2+ channel activity and α-cell exocytosis . Thus , we propose that elevated blood glucose reduces α-cell electrical activity and glucagon secretion by inactivating the ion channels involved in action potential firing and secretion .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"physiology",
"in",
"vitro",
"diabetes",
"and",
"endocrinology",
"homo",
"(human)",
"rattus",
"(rat)"
] |
2007
|
A KATP Channel-Dependent Pathway within α Cells Regulates Glucagon Release from Both Rodent and Human Islets of Langerhans
|
The architecture of dendritic arbors determines circuit connectivity , receptive fields , and computational properties of neurons , and dendritic structure is impaired in several psychiatric disorders . While apical and basal dendritic compartments of pyramidal neurons are functionally specialized and differentially regulated , little is known about mechanisms that selectively maintain basal dendrites . Here we identified a role for the Ras/Epac2 pathway in maintaining basal dendrite complexity of cortical neurons . Epac2 is a guanine nucleotide exchange factor ( GEF ) for the Ras-like small GTPase Rap , and it is highly enriched in the adult mouse brain . We found that in vivo Epac2 knockdown in layer 2/3 cortical neurons via in utero electroporation reduced basal dendritic architecture , and that Epac2 knockdown in mature cortical neurons in vitro mimicked this effect . Overexpression of an Epac2 rare coding variant , found in human subjects diagnosed with autism , also impaired basal dendritic morphology . This mutation disrupted Epac2's interaction with Ras , and inhibition of Ras selectively interfered with basal dendrite maintenance . Finally , we observed that components of the Ras/Epac2/Rap pathway exhibited differential abundance in the basal versus apical dendritic compartments . These findings define a role for Epac2 in enabling crosstalk between Ras and Rap signaling in maintaining basal dendrite complexity , and exemplify how rare coding variants , in addition to their disease relevance , can provide insight into cellular mechanisms relevant for brain connectivity .
Dendritic structure is critical for neuronal function , as the size and shape of the dendritic arbor defines the neuron's receptive field [1] . Generating and maintaining proper arborization is therefore crucial for neural circuit function [2] . The importance of maintaining dendritic arborization is illustrated by observations of loss of dendritic complexity in patients with neuropsychiatric disorders . Reduced dendritic arborization occurs in patients with psychiatric disorders with delayed onset , including schizophrenia [3] , as well as in autism spectrum disorders [4] , [5] , and disorders comorbid with autism , such as Rett [6] and Down syndromes [7]–[9] . Thus , the maintenance of dendritic arbor complexity for extended periods of time during development and into adulthood is likely to be crucial for the preservation of functional circuitry and connectivity relevant for learning and complex behaviors . Patterns of dendritic branching are integral to the computational ability of the neuron [10] , [11] . An increasing body of evidence suggests that the apical versus basal regions of the dendritic arbor are functionally specialized . These distinct dendritic compartments receive different inputs , integrate distinct signals , and are selectively regulated in physiological and pathological conditions [12] , [13] . Recent work probing the subcellular location of thalamocortical and intracortical connections has revealed tight spatial restriction of synapses to various somatic and dendritic compartments . Laminar positioning of target cells in the cortex and afferent cell type are critical determinants of synaptic positioning along the dendritic arbor . For example , ascending inputs target basal dendrites in layer 2/3 [13] . Basal dendrites are also the target of substantial inhibitory innervation by interneurons , allowing for the tight regulation of excitability [14] . From a computational perspective , even very small basal dendrites are capable of large effects on cell output [12] . Thus , subtle morphological alterations to the basal dendritic arbor may have large consequences for cellular and circuit function . Consistent with the selective function of dendritic compartments , there is evidence for the selective regulation and maintenance of apical versus basal dendritic compartments . Environmental enrichment appears to have region- and cell-specific effects on dendrites , but preferentially enhances basal arborization [15] , [16] , and sensory deprivation during a critical developmental period can prevent normal basal dendritic elaboration in the barrel cortex of rats [17] . While a few molecular alterations selectively affect distinct dendritic compartments , including PTEN or dopamine receptor D1 loss [18] , [19] , the molecular mechanisms that specifically govern basal dendrite maintenance in cortical neurons remain unclear . Regulators of Ras-like small GTPases have been extensively implicated in neuronal morphogenesis [20] . The EPAC2 gene encodes Epac2 ( exchange protein directly activated by cyclic AMP 2 ) , a guanine nucleotide exchange factor ( GEF ) for the Ras-like small GTPase Rap , which is highly enriched in the adult brain [21] and dendrites [22] . Previous studies utilizing an Epac-specific agonist have found that Epac activation can modulate synaptic plasticity [23] as well as memory retrieval in mice [24] , and EPAC null mice exhibit deficits in spatial reference memory and social interactions [25] , implicating Epac in brain function . Epac2 has been implicated in the outgrowth of neuronal processes in vitro [26] , [27] , but its role in dendritic morphogenesis within the cortex is not known . In the present study we observed that Epac2 knockdown robustly and selectively impaired basal dendrite maintenance in cortical pyramidal neurons in vivo and in culture . Recent genetic studies have detected numerous rare coding mutations in subjects with neurodevelopmental disorders [28] . While their significance for disease etiology remains to be elucidated , such mutations might provide insight into a protein's functional role in important cellular processes . Previously , four rare amino acid coding variants had been identified in EPAC2 in subjects with autism [29] . In the present study we observed that expressing one of these rare coding variants robustly and selectively reduced basal dendrite complexity in cortical pyramidal neurons and impaired Epac2's interaction with Ras . The use of a disease-associated point mutation as a method of probing molecular function revealed that Epac2 mediates crosstalk between Ras and Rap to specifically regulate basal dendritic complexity in cortical neurons . This approach exemplifies a more general “reverse translational” strategy for discovery of basic cellular mechanisms .
Epac2 , an upstream regulator of Rap activity , is involved in regulating synapse morphology [22] , but its role in regulating the architecture of the dendritic tree in cortical neurons is unknown . We first tested whether altered expression of Epac2 affected the maintenance of dendritic arbors in vivo , using in utero electroporation ( IUEP ) [30] to knock down protein expression ( Figure S1A ) . Using a previously characterized RNA interference ( RNAi ) construct selective for Epac2 [22] , we coinjected either Epac2-RNAi , or control ( pGSuper ) , with pCAG-eGFP into the subventricular zone of E16 . 5 mouse embryos to specifically target layer 2/3 neurons . Electroporated constructs were allowed to express until brain harvesting and sectioning on postnatal day 28 ( P28 ) . In vivo knockdown of the Epac2 protein via Epac2-RNAi was confirmed at P28 by immunohistochemistry of cortical sections . Quantification of GFP-positive cells revealed that Epac2 was knocked down by ∼75% compared to GFP-negative cells ( Figure S1B–C ) . Electroporated neurons were specifically found in layer 2/3 ( Figure 1A ) . When we analyzed dendritic morphology by measuring apical and basal dendritic number and length in GFP-positive cells , we observed that long-term knockdown of Epac2 expression specifically reduced basal dendrites . Examination of basal arbors of layer 2/3 neurons from P28 mice expressing GFP and Epac2-RNAi ( Figure 1B–F ) revealed a ∼42% reduction in basal dendritic branches and a ∼54% reduction in basal dendritic length ( p<0 . 001; Figure 1E–F ) compared to cells from P28 mice expressing GFP and control construct . This effect on basal dendritic morphology was mediated by a ∼30% reduction in secondary and a ∼65% reduction in tertiary order basal branch number ( p<0 . 05 , 0 . 001; Figure 1G , Table S1A ) . Additionally , basal dendritic length was reduced by ∼24% in secondary branches and ∼55% in tertiary branches ( p<0 . 05 , 0 . 001; Figure 1H , Table S1B ) . The effect of Epac2 knockdown from E16 . 5 to P28 was specific for basal dendritic arbors , as Epac2-RNAi had no effect on apical dendrite branch number or length in these cells ( Figure 1E–H , Table S1A–B ) . To eliminate the possibility of inter-individual and reporter expression variability , we took advantage of the sparse nature of IUEP gene transfer to directly compare dendritic morphology between neighboring non-Epac2-RNAi-expressing cells versus Epac2-RNAi-expressing cells within the same layer and cortical region from the same mice . Epac2-RNAi-expressing cells could be easily identified by the coexpression of pCAG-eGFP , while neighboring control cells not expressing Epac2-RNAi were identified by the lack of GFP expression . Pairs of GFP- and non-GFP-expressing neurons were filled with biocytin and stained with streptavidin-568 conjugated fluorescent probe to visualize dendritic morphology ( Figure 2A ) . Using 2-photon laser scanning microscopy , we imaged paired layer 2/3 cells in the anterior frontal cortex , an area previously described to display abnormal circuitry in disease-related animal models [31] . Cortical slices were cut at a thickness of 300 µm to allow reconstruction of the majority of the apical and basal dendritic fields of these cells . This approach further confirmed that loss of Epac2 resulted in a specific reduction of basal dendrite complexity: Epac2-RNAi-expressing cells displayed an overall decrease in basal dendritic branch number ( ∼47% ) and basal dendritic length ( ∼53% ) ( p<0 . 05 , 0 . 01; Figures 2B–C , S2A–B ) . Furthermore , Epac2-RNAi-expressing cells had ∼30% fewer basal secondary dendrites and ∼55% fewer basal tertiary dendrites ( p<0 . 05 , 0 . 001; Figure 2D–E , Table S2 ) . In addition , secondary basal length was reduced by ∼25% and tertiary basal length was reduced by ∼32% in Epac2-RNAi-expressing cells ( p<0 . 05 , 0 . 001; Figure 2D , F , Table S3 ) . This effect was driven by the absence of high order branches ( tertiary branches and beyond; Figure 2D–F ) . No effect on apical dendrite complexity ( Figure S2C , Table S2 ) or length ( Figure S2D , Table S3 ) was observed . Together , these data demonstrate that Epac2 signaling is required for maintaining higher order branching of basal dendrites in vivo . Interestingly , extended Epac2 knockdown in vivo also reduced dendritic spine density on both apical and basal dendrites , as compared to paired control electroporated cells ( Figure S2E–G ) . This effect contrasted with that of acute ( 5-d ) Epac2 knockdown in vitro [22] , which did not alter dendritic spine numbers , but suggests that prolonged reductions in Epac2 signaling can have pronounced effects on basal dendrites and more subtle effects on apical dendrites . Epac2 knockdown by in utero electroporation reduced Epac2 expression throughout development; however , it is not clear when Epac2 expression is required for normal dendritic morphology . To directly test the role of Epac2 in dendritic maintenance , we used an RNAi approach in mature cultured neurons , allowing perturbations in Epac2 expression levels after the dendritic arbor has already been established . This system has also been extensively used for mechanistic studies of structural plasticity [22] , [32]–[34] and allows examination of the potential molecular underpinnings of dendritic architecture . A number of studies have demonstrated that mature cultured pyramidal neurons develop pyramidal morphologies with primary ( classified as “apical” ) and non-primary ( classified as “basal” ) dendrites that resemble morphologies observed in vivo ( see Materials and Methods for description of criteria used for identifying apical and basal dendrites of cultured neurons ) [34]–[36] . We knocked down Epac2 expression in mature ( DIV 23–28 ) cultured cortical neurons ( Figure 3A ) , and used Sholl analysis [37] as well as dendritic length measurements to assess the complexity and morphology of basal or apical dendritic compartments ( Figure S3A–D ) . Consistent with our in vivo data , reduced Epac2 expression selectively decreased dendritic complexity in an asymmetric manner . Epac2 knockdown reduced basal dendritic intersections 25–175 µm from the soma , as well as basal dendrite length , without affecting apical dendrite length or complexity ( basal dendritic length: length ( µm ) , control: 956±164; Epac2-RNAi: 273±100; rescue: 647±124 , p<0 . 005; Figure 3B–C ) . Importantly , this deficit was rescued by overexpressing an RNAi-resistant mutant of Epac2 ( “Epac2-rescue” ) [22] . Epac2-rescue overexpression did not significantly alter basal or apical complexity compared to control , but significantly increased basal complexity 25–100 µm from the cell body and basal dendrite length compared to Epac2-RNAi ( Figure 3B–C ) . Similar to the effect of Epac2 knockdown in in utero electroporated neurons , Epac2 expression levels were reduced down by similar degrees ( ∼75% ) in both apical and basal dendritic compartments in cultured cells overexpressing Epac2-RNAi ( Figure S3E ) . Given that Epac2 knockdown occurred in mature neurons after their dendritic arbors had been established , these data suggest that Epac2 plays a role in maintenance of the basal dendritic arbor . Rare protein-coding variants of the EPAC2 gene have previously been identified in several subjects with autism [29] . One of these missense mutations ( Epac2-G706R ) , detected in four human subjects with autism from two families , is located within the Ras association ( RA ) domain of Epac2 ( Figure 4A ) , suggesting that it may affect one of Epac2's functional domains . To investigate the effect of this mutation on neuronal morphology , we expressed either Epac2-G706R or its wildtype counterpart in cultured cortical neurons . Expression of Epac2-G706R ( Figure 4B ) , followed by Sholl analysis , revealed a robust selective decrease in basal dendrite complexity 50–100 µm from the soma , compared to Epac2-WT , with no effect on apical dendrites ( Figure 4C ) . Furthermore , overexpression of Epac2-G706R reduced basal dendrite length relative to Epac2-WT ( p<0 . 05; Figure 4D ) , but did not affect apical dendritic length ( Figure 4D ) . We have previously shown that overexpression of Epac2-G706R in neurons does not affect basal Rap-GEF activity or dendritic spine morphology [22] . Comparison of the effects of Epac2-G706R overexpression to GFP alone revealed a decrease in basal complexity 25–50 µm from the soma , but no change in apical complexity , basal length , or apical length ( Figure S4A–C ) , suggesting that Epac2-G706R is a loss-of-function mutation . Taken together , these data suggest that Epac2-G706R , a variant that occurs in human patients , specifically alters basal dendrite maintenance , without affecting apical dendritic structure , synaptic morphology , or baseline Rap activation levels . We next reasoned that the location of the single amino-acid mutation in the Epac2 protein might offer insight into the mechanisms of asymmetric maintenance of dendritic compartments . The small GTPase Ras and its signaling partners have been implicated in neuronal morphogenesis [20] , [38]–[42] . Given that the G706R mutation is within Epac2's Ras-association ( RA ) domain ( Figure 4A ) , we hypothesized that this mutation might alter Epac2's interaction with Ras and that abnormal association with Ras could underlie the dendritic effects induced by Epac2-G706R . Epac2 has been shown to interact with Ras in non-neuronal cells [43] , but this interaction has not yet been established in cortical neurons . We therefore tested whether Epac2 interacted with Ras in rat cortical neurons by coimmunoprecipitation . We found that Ras coimmunoprecipitated with Epac2 in mature cortical neurons ( DIV 25 ) ( Figure 4E ) . This interaction was dependent on the activation state of Ras: Ras activation by incubation with GTPγS enhanced the interaction between Ras and Epac2 , whereas Ras inhibition by treatment with GDP reduced the interaction ( p<0 . 05 , 0 . 001; Figure 4F ) . This interaction was further confirmed by ectopic expression of HA-tagged Epac2-WT alone or with YFP-Ras in hEK293 cells ( Figure S4D ) . We then tested the ability of Epac2-G706R to interact with Ras by coexpressing YFP-Ras with HA-Epac2-WT or HA-Epac2-G706R in hEK293 cells , and immunoprecipitating with YFP-Ras ( Figure 4G ) . Indeed , quantitative analysis of coimmunoprecipitation revealed that Epac2-G706R displayed significantly impaired Ras interaction ( p<0 . 001; Figure 4H ) . These results demonstrate that a naturally occurring Epac2 variant specifically alters basal dendritic architecture , and that interaction with Ras may be a key feature of Epac2's role in regulating basal dendritic maintenance . The findings that Epac2 is required for the maintenance of basal ( non-primary ) dendrites , and that a rare coding variant that specifically disrupts Epac2's interaction with Ras mimics this selective morphological phenotype but has no affect on Epac2's basal Rap-GEF function [22] , suggest a role for Epac2 and Ras signaling in the maintenance of basal dendrite complexity . Ras is a small GTPase that has been strongly linked to structural plasticity in neurons [20] , [39]–[42] , [44] , [45] , but its specific role in the maintenance of basal dendrite complexity in mature neurons has not been directly tested . Thus , we tested whether disruption of endogenous Ras activity by the farnesyl transferase inhibitor II ( FTaseII ) could affect either apical or basal dendrite maintenance . We used time-lapse imaging of live mature ( DIV 25 ) cultured cortical pyramidal neurons , expressing GFP and treated with either vehicle or FTaseII ( 200 nM ) and measured dendritic complexity and length before and after treatment ( Figure 5A and S5A ) . Imaging of neurons for 2 h prior to treatment revealed a remarkable stability of the dendritic arbor ( Figure 5A–E ) with an almost equal gain and loss of apical and basal dendrites . Imaging of apical and basal dendrites for 6 h following vehicle treatment did not reveal any changes in basal or apical dendrite complexity or length ( Figure 5A–E ) . In contrast , incubation with FTaseII resulted in a robust retraction of basal dendrites over 6 h , as demonstrated by a reduction of basal complexity ( normalized basal dendrite intersections: 0 h: 1 . 04±0 . 03 versus 0 . 95±0 . 05; 6 h 1 . 02±0 . 5 versus 0 . 56±0 . 08; control versus FTaseII , p<0 . 001; Figure 5A–B ) , and a loss of dendritic length ( normalized basal dendrite length: 0 h: 1 . 06±0 . 02 versus 1 . 01±0 . 04; 6 h: 1 . 02±0 . 4 versus 0 . 54±0 . 06; control versus FTaseII , p<0 . 001; Figure 5A , C ) . This loss of length , driven by a progressive retraction of basal dendrites , was not seen in vehicle-treated neurons ( Figure 5D–E ) . There was no change in apical dendrites ( Figure 5B–C ) . We further confirmed these results in neurons fixed following treatment with FTaseII or vehicle for 6 h ( Figure S5B–D ) . Sholl analysis revealed that FTaseII treatment specifically reduced basal complexity 25–100 µm from the soma ( Figure S5C ) and basal dendritic length ( basal dendrite length ( µm ) ; control , 1 , 405±221; FTaseII , 602±155 , p<0 . 05; Figure S3D ) with no effect on apical dendrite complexity or length ( Figure S5C–D ) . The very specific effect of short-term FTaseII treatment on the dendritic tree is quantitatively similar to that of Epac2 knockdown or Epac2-G706R overexpression ( which displays impaired Ras binding ) , suggesting that interference with Ras signaling , but not other potential targets of FTaseII , in cortical neurons results in a selective reduction of basal complexity . To further investigate the impact of interference with Ras signaling on dendritic architecture , we mimicked Ras inhibition by expression of a dominant-negative Ras mutant ( Ras S17N; RasDN ) . Expression of RasDN alone ( Figure S5E–F ) , followed by Sholl analysis , revealed a robust selective decrease in basal dendrite complexity 50–100 µm from the soma , compared to GFP ( control ) , with no effect on apical arbors ( Figure S5F ) , paralleling the effects seen following short-term FTaseII treatment . Furthermore , overexpression of RasDN reduced basal dendrite length relative to control ( basal dendrite length ( µm ) ; GFP , 1 , 704±122; RasDN , 987±198 , RasDN+Epac2-WT 1 , 519±160 , p<0 . 05; Figure S5E–F ) , but did not affect apical dendritic length ( Figure S5G ) . Importantly , co-expression of Epac2-WT was sufficient to rescue RasDN-induced loss of basal dendrites and basal dendritic length ( Figure S5E–G ) . These data provide further support for the role of the Ras/Epac2 pathway in the maintenance of basal , but not apical , dendrites . The distribution of the Ras , Epac2 , and Rap proteins across the dendritic tree of cortical neurons has not yet been examined . We thus compared the relative amounts of Epac2 , Ras , and Rap immunostaining intensity in basal versus apical dendrites of cortical neurons in culture . All intensity measurements were limited to secondary apical or basal dendrites and were normalized to unit area ( µm2 ) to ensure that measurements of protein content between apical and basal dendrites of different thicknesses were comparable ( Figure 6 ) . We observed more intense labeling for each protein in apical dendrites than in basal dendrites ( p<0 . 05 , 0 . 01; Figure 6A–F ) . We also examined the distribution of phosphorylated ( active ) BRaf ( p-BRaf ) , a direct target of both Ras and Rap small GTPases [20] , [22] . As with Ras , Epac2 , and Rap , we found that p-BRaf was more abundant in apical dendrites than in basal dendrites ( p<0 . 001; Figure 6G–H ) . Interestingly the same subcellular distribution was observed for overexpressed Epac2-WT and Epac2-G706R ( p<0 . 01; Figure S6A–B ) . Epac2-G706R signal was reduced compared to that of Epac2-WT in basal dendrites , but not apical dendrites , in these cells ( p<0 . 05; Figure S6C ) . In order to determine whether this asymmetric pattern was specific for Ras/Epac2/Rap pathway , we also examined the distribution of kalirin-7 , a GEF for the small GTPase Rac , and the phosphorylated ( active ) form of p21-activated kinase ( p-PAK ) , a direct downstream effector of Rac [46] . In contrast with the above findings , kalirin-7 and p-PAK immunofluorescence was equally distributed across apical and basal arbors ( Figure S6D–E ) , indicating that the asymmetric distribution of Ras/Epac2/Rap was specific to this pathway . Because we observed asymmetry in the levels of Epac2/Ras/Rap across dendritic compartments of cortical neurons , we wondered if perturbations of this pathway might lead to differential signaling output in basal versus apical dendrites . To address this question , we specifically inhibited Ras signaling or reduced Epac2 expression by RNAi throughout the neurons , and examined p-BRaf immunofluorescence in individual dendritic compartments . Inhibition of Ras using FTaseII ( Figures 7A , S7A ) resulted in a ∼22% reduction in p-BRaf levels in apical dendrites compared to control ( Figures 7A–C , S7A ) , but produced a more profound reduction in p-BRaf levels in basal dendrites ( ∼45% compared to control basal levels ) , which was significantly different to both control and FTaseII-induced apical p-BRaf levels ( mean p-BRaf intensities relative to control levels: FTaseII; apical 0 . 76±0 . 04; basal 0 . 55±0 . 01; p<0 . 001; Figures 7B–C , S7B–C ) . Examination of endogenous Epac2 levels in control ( pGSuper ) and Epac2-RNAi cells revealed that Epac2 expression was less abundant in basal dendrites than apical dendrites after Epac2 knockdown ( reduced by ∼37% compared to apical levels; Figure S7D ) , suggesting that asymmetric localization of Epac2 is maintained under knockdown conditions . Moreover , comparison of Epac2 immunofluorescence in apical or basal compartments in control versus Epac2-RNAi-expressing neurons , relative to Epac2 levels in control apical dendrites , revealed that Epac2 expression in apical dendrites of Epac2-RNAi cells was not significantly different from Epac2 levels in control basal dendrites ( Figure S7E ) , suggesting that apical dendrites may contain enough Epac2 even in the knockdown condition to preserve apical dendrite morphology . Epac2 knockdown ( Figures 7D , S7F ) resulted in a greater reduction of p-BRaf levels in basal dendrites ( ∼47% ) versus apical dendrites ( ∼20% ) , when compared to control levels ( mean p-BRaf intensities relative to control levels: Epac2-RNAi; apical 0 . 80±0 . 07; basal 0 . 52±0 . 04; p<0 . 001; Figures 7D–F , S7F–H ) . Collectively , these data suggest that the signaling output of the Ras/Epac2/Rap pathway is asymmetric across dendritic compartments of cortical neurons .
Here we showed that Epac2 is important for the selective maintenance of basal dendrite complexity in cortical neurons . Utilizing a rare coding variant of Epac2 , found in human patients , to probe the molecular and cellular functions of Epac2 in the context of dendritic complexity of cortical pyramidal neurons , we identified Ras as a signaling partner of Epac2 in this pathway . Our findings support a model in which Epac2 , as a Rap-GEF , enables crosstalk between two morphogenic GTPase signaling pathways to maintain basal dendrites . The importance of maintaining dendritic architecture is illustrated by observations of dendritic complexity in pathological analysis of patients with neuropsychiatric and neurodevelopmental disorders . Reduced dendritic arborization occurs in patients with psychiatric disorders with delayed onset , including schizophrenia [3] , autism spectrum disorders [4] , [5] , and disorders comorbid with autism , such as Rett [6] and Down syndromes [7]–[9] . In Down syndrome , loss of dendritic arbors occurs in a progressive manner: prior to 2 y of age , dendritic hypertrophy was observed in the cortex of subjects with Down syndrome; thereafter , dendritic arbors were reduced in complexity relative to controls [7]–[9] , suggesting an initial period of dendritic overgrowth followed by a later stage in which maintenance mechanisms are potentially disrupted or lost . An early study of visual cortex from subjects with Down syndrome revealed a selective loss of basal dendrite complexity [47] , though subsequent studies have also detected apical deficits [8] . Rett syndrome , a monogenic disorder frequently accompanied by an autistic phenotype , features dendritic deficits , usually in basal dendrites and occasionally in apical dendrites [6] . Individuals with autism exhibited reduced dendrite numbers and dendritic cytochemical markers in the cortex [4] and hippocampus [5] . Thus , a more complete understanding of molecular mechanisms that contribute to the maintenance of specific aspects of the dendritic arbor may hasten the development of therapeutic strategies that aim to prevent the apparent progressive loss of dendritic complexity and to preserve functional cortical circuits in patients with neurodevelopmental and neuropsychiatric disorders in which dendritic structure is affected . Several lines of evidence support the selective regulation of apical versus basal dendritic compartments . Environmental enrichment has been shown to selectively increase basal dendrite length , while stress can reduce basal dendritic length , in layer 2/3 cells of the auditory cortex [15] . Sensory deprivation by whisker trimming during a critical developmental window , between P9 and P15 in rats , delayed normal basal dendritic elaboration of layer 2/3 pyramidal cells in the barrel cortex [17] . PTEN knockout results in selective outgrowth of apical dendrites , demonstrating that inhibition of mTOR blocks continued apical but not basal dendrite growth under stable conditions in mature animals [18] . Particularly relevant to Epac2 signaling is the recent report that knockout of dopamine receptor D1 in mouse cortex results in selective basal dendrite loss [19] , as Epac2 is also regulated by D1/cAMP signaling [22] . Given the ability of Epac2 to selectively regulate the basal arbor , signaling through Epac2 may be a key mechanism for control of select dendritic compartments . A current hypothesis is that the establishment and elaboration of apical and basal arbors occur at distinct time points , with basal dendrites developing after apical dendrites , and thus may involve distinct regulatory mechanisms [48] . Our knockdown data indicate that Epac2 is required for the maintenance of higher order basal dendrite branches . Elaboration of higher order basal dendrite branching , which increases basal dendrite complexity , occurs subsequent to the formation of the primary proximal basal arbor and the apical arbor [48] . Interestingly , this time point ( 3 wk postnatal ) coincides with a dramatic increase in Epac2 expression in cortical neurons [22] . Thus , our findings support a role for Epac2 in regulating the maintenance of basal dendritic complexity once these complex arbors have been initially established . Here we were guided by a rare coding mutation , naturally occurring in human subjects with autism , to identify the Ras/Epac2 interaction as important for the control of basal dendrite complexity in cortical neurons . Interestingly , a different point mutation in the RA domain of Epac2 , identified through sequence analysis rather than occurrence as a rare variant in human patients , has also been shown to disrupt the interaction with Ras in COS cells [49] , suggesting that single-residue sites in this domain of Epac2 may be crucial for its function in response to Ras , and that even single amino acid mutations occurring as rare variants may have functional or pathological consequences . Our strategy exemplifies how mutations identified in humans with neurodevelopmental or psychiatric disorders , beyond their relevance for disease , could provide functional insight into novel mechanisms underlying brain development and connectivity . A growing number of rare single amino acid mutations have been identified in neuropsychiatric disorders by recent genetic studies , and with the advent of whole exome or genome sequencing , their numbers are expected to increase dramatically [50] . While their significance for disease etiology remains to be elucidated , our approach taken in this study shows that such mutations might help identify cellular mechanisms that control crucial cellular processes , including dendrite arborization . About 20% of single amino acid mutations are thought to be damaging , with another 53% being mildly deleterious [51] . Given that rare mutations are thought to make up a significant fraction of the genetic architecture of complex diseases , functional characterization of such mutations may provide novel insight into both physiology and pathophysiology . In this study , we show that a pathway involving Ras/Epac2/Rap contributes to the maintenance of basal dendrite complexity . The G706R point mutation disrupted the Ras-Epac2 interaction and reduced basal complexity , and Ras inhibition experiments using FTaseII or RasDN overexpression specifically affected basal dendrite maintenance , suggesting that this pathway exerts specific control over basal dendritic complexity in pyramidal neurons . Our results establishing a role for Epac2 in linking Ras and Rap signaling to dendrite maintenance in mature cortical neurons are consistent with a number of previous findings . Dominant-negative Rap1-expressing layer 5 pyramidal neurons exhibit deficits in basal dendrite arborization during development [38] . In non-neuronal cells , activated Ras has been shown to recruit Epac2 to the plasma membrane , thereby activating membrane-associated pools of Rap [43] . In adrenally derived PC12 cells , Ras activation can recruit Epac2 to the plasma membrane , activate membrane-associated pools of Rap1 , and induce the outgrowth of neurite-like structures [49] . Recent characterizations of EPAC null mice reveal cognitive and behavioral phenotypes , illustrating the importance of Epac in complex behavior and brain function [25]; however , dendritic architecture of cortical neurons was not measured in this mouse model . Our data implicating Epac2 in the maintenance of basal arbors of cortical neurons provide a potential mechanism for the disruption of neuronal circuitry upon perturbations of this pathway . Our observation of asymmetric distribution of Epac2 , Ras , and Rap proteins is consistent with the selective effect of reduced Ras/Epac2 signaling on the maintenance of basal dendrites . Indeed , it is reasonable to expect that other redundant mechanisms are employed for the active maintenance of apical dendritic architecture , which may require more stability during the life of the neuron . Our data showing that disruption of a single pathway can alter basal maintenance are consistent with the intrinsic dynamism of basal dendrites , due to the higher demands for plasticity driven by sensory and inhibitory inputs to this compartment [11] , [13] , [14] . Taken together , our data support a model in which Epac2 couples with Ras signaling and actively maintains basal dendrites in cortical pyramidal neurons .
Farnesyl transferase inhibitor II was purchased from EMD Biosciences . We purchased the following antibodies: mouse anti-GFP monoclonal ( Millipore ) , chicken anti-GFP polyclonal ( Abcam ) , rabbit anti-Epac2 polyclonal and mouse anti-Epac2 monoclonal ( Santa Cruz ) , rabbit anti-HA polyclonal ( Enzo ) and mouse anti-HA monoclonal ( Santa Cruz ) , mouse anti-Ras monoclonal ( Upstate ) , mouse anti-Myc monoclonal ( Developmental Studies Hybridoma Bank , Iowa ) , mouse anti-MAP2 monoclonal ( Millipore ) , and rabbit anti-Rap polyclonal antibody ( Millipore ) . A rabbit anti-GFP polyclonal antibody was a gift from Dr . Richard Huganir ( Johns Hopkins University ) . The pCAG-EGFP construct was a kind gift from Atsushi Kamiya , Johns Hopkins University . The pEGFP-N2 construct was obtained from Clontech . Constructs encoding shRNA specific for Epac2 and a rescue construct ( an HA-Epac2 construct containing three silent point mutations in the RNAi target sequence ) were previously generated and validated [22] . Dissociated cultures of primary cortical neurons were prepared from E18 Sprague-Dawley rat embryos as previously described [22] . On DIV 21–23 , neurons were transiently transfected for 4 h with plasmids ( 1–3 µg DNA ) using Lipofectamine 2000 ( Invitrogen ) . For experiments utilizing soluble GFP , cultures were allowed to express the transfected constructs for 2 d . For RNAi and Epac2 mutant , constructs were expressed for 5 d . Rats were used in accordance with ACUC institutional and national guidelines under approved protocols . Treatment of live cultured neurons was performed in ACSF essentially as previously described [22] . Briefly , cultured neurons were transfected with GFP , and allowed to express for 2 d . Neurons were then pre-incubated in ACSF for 1 h , imaged 2 h and 1 h before beginning of treatment , and then imaged every hour for 6 h after beginning of treatment with either FTaseII ( 200 nM ) or vehicle . Coverslips were kept in culture plates throughout the experiment , and were returned to a 37°C incubator between imaging timepoints . Micrographs of healthy GFP-expressing neurons with pyramidal morphologies were acquired using a 10× objective ( NA = 0 . 17 ) and a Zeiss AxioCam MRm CCD camera . Dendrites were traced and binarized in ImageJ as described below . It is of note that whereas farnesyl transferase inhibitors were initially developed for their ability to inhibit Ras activity , a number of other proteins are also farnesylated , and therefore the potential contribution of other proteins to this specific loss of basal dendrites cannot be excluded . For dendritic spine morphologies in vivo ( Figure S2 ) , images of dendritic spines on biocytin-filled neurons were acquired with a Zeiss LD Lci Plan Apochromat 25×/0 . 8NA multi-immersion lens ( 440842-9870-000000 ) with a digital zoom of 4 . Volume imaging was acquired with 15–35 optical sections taken in 0 . 75 µm focal steps ( 2 . 13 µm axial resolution ) . For each condition , 5 neurons were imaged . Two dendrites between 50 and 100 µm in length per cell were measured: only spines on tertiary apical or secondary basal dendrites were imaged to reduce variability . Dendritic spine density ( number of spines per 10 µm ) was calculated using ImageJ . Cultured pyramidal neurons were fixed , immunostained , and imaged as previously described [22] . Protein clustering was imaged as z-series taken at 0 . 37 µm intervals using a Zeiss LSM5 Pascal confocal microscope and a 63× objective ( NA = 1 . 4 ) . Two-dimensional maximum projection images were reconstructed and analyzed using MetaMorph software ( Molecular Devices , Sunnyvale , CA , USA ) . Images were background-subtracted and thresholded equally to include clusters with intensity at least 2-fold above the adjacent dendrite . Regions along dendrites were manually outlined , and the linear density ( number per 100 µm of dendrite length ) and total gray value ( total immunofluorescence intensity ) of each cluster was measured automatically . Cultures that were directly compared were stained simultaneously and imaged with the same acquisition parameters . Experiments were carried out blind to condition and on sister cultures . Coimmunoprecipitations ( coIPs ) from hEK293 cells or rat cortical tissue were performed as previously described [22] , using RIPA buffer ( in mM: 150 NaCl , 10 Tris-HCl , pH 7 . 2 , 5 EDTA , 0 . 1% SDS , 1% Triton X-100 , 1% Deoxycholate , plus inhibitors ) . Precleared lysates were incubated with 2 . 5–5 µL of antibody for 3 h; 60 µL of protein-A Sepharose was added for 2 h at 4°C , after which samples were washed 3 times with 0 . 5 ml RIPA buffer , boiled for 5 min at 95°C in Laemmli buffer , and analyzed by SDS-PAGE and Western blotting . For treatment with GTPγS or GDP , cortical neurons were lysed in Mg2+ lysis buffer containing protease inhibitors . Cell lysates were then incubated with 100 µM GTPγS or 1 mM GDP at 30°C for 30 min . Reaction was stopped by the addition of 60 mM MgCl2 . Cell lysates were harvested as described above . C57BL6 female mice were checked for vaginal plugs ( E0 ) , and electroporation was performed at E16 . 5 . After proper sedation , both uterine horns were removed and placed on sterile , warm , and PBS-wetted pads . DNA solution was loaded into beveled glass micropipettes ( 100 µm oblique opening ) , and 0 . 26 µl was injected into the lateral ventricle through the uterus wall ( 4 injections; 65 nL/injection ) using a nanojector ( Drummond Nanoject II ) . DNA was electroporated into the neural precursor populations that reside on the ventricular zone by directed electroporation , by placing the ( + ) end of the electrode toward the developing neocortex . Unipolar electric pulse of 40 V was generated ( BTX ECM830 ) , and a total of five 50 ms pulses at an interval of 100 ms were applied to the cerebral wall . After electroporation , embryos were placed back into the abdominal cavity , and the rectus abdominis and abdominal oblique muscles were sutured with 5-0 coated vicryl suture for quick absorption and fast recovery . The skin was closed with LiquiVet tissue adhesive . Mice were allowed to recover and give natural birth . Injected animals were collected at P28 for further investigations . A preparation of DNA 3∶1 molar ratio was used to mix pCAG-EGFP construct [52] , which expresses EGFP under the chicken beta actin promoter , and Epac2-RNAi cloned into pGSuper expression vector [22] . 0 . 05% FastBlue was added to visualize DNA . In control experiments , pGSuper was mixed with pCAG-eGFP construct . Electroporated mice were anesthetized with sodium pentobarbital at 50 mg per gram of body weight , and fixed by transcardial perfusion of 4% paraformaldehyde in 0 . 1 M sodium phosphate buffer ( pH 7 . 4 ) at P28 . Brains were dissected out and sectioned into 50 µm coronal sections and immunostained with a chicken anti-GFP polyclonal antibody ( Abcam ) as floating sections , before being mounted onto glass slides and covered with glass coverslips . Cells exhibiting intact and healthy secondary and tertiary apical and basal dendritic arbors were imaged by taking 1 µm serial optical sections , 35–45 optical sections per cell , using a Zeiss LSM5 Pascal confocal microscope and a 40× objective ( NA = 1 . 3 ) . Following acquisition , images were projected as 2-D Z-projections using Fiji6/ImageJ ( http://imagej . nih . gov/ij/; NIH , Bethesda , MD , USA ) . Dendrites were analyzed using NeuronJ plugin [53] for Fiji6/ImageJ . Between 12 and 15 cells per condition were analyzed . At P28 , electroporated mice were deeply anesthetized with isoflurane and their brains were quickly removed . Brain sections were cut at a thickness of 300 µm using an off-sagittal slice angle to preserve apical and basal tufts of layer 2/3 cortical neurons of the anterior frontal cortex [31]; sections were cut in ice-cold carbogenated choline solution ( in mM: 110 choline chloride , 25 NaHCO3 , 2 . 5 KCl , 1 . 25 NaH2PO4 , and 0 . 5 CaCl2 , 7 MgSO4 , 25 D-glucose , 11 . 6 sodium ascorbate , 3 . 1 sodium pyruvate ) . Slices were transferred to carbogenated artificial cerebrospinal fluid ( ACSF , in mM: 126 NaCl , 2 . 5 KCl , 26 NaHCO3 , 2 CaCl2 , 1 MgCl2 , 1 . 25 NaH2PO4 , and 10 D-glucose ) and incubated for 30 min at 35°C . They were then maintained at room temperature for the remainder of the intracellular labeling procedure . To maximize the amount of dendritic arbor , we selected neighboring pairs of GFP-positive , Epac2-RNAi-expressing , and non-fluorescent control neurons with somata deeper than 60 µm from the surface of the brain slice ( average depth = 89±3 . 9 µm ) . Using a micropipette filled with biocytin intracellular solution ( in mM: 10 biocytin , 126 K-methylsulfate , 4 KCl , 10 HEPES , 4 ATP , 0 . 3 GTP , and 10 phosphocreatine ) , we dialyzed the neurons for at least 15 min and then allowed the cells to recover for at least 30 min before fixing the slices in 4% paraformaldehyde in 0 . 1 M sodium phosphate buffer ( pH 7 . 4 ) . Slices were then immunostained with a fluorescent streptavidin-568 conjugate ( Invitrogen ) and chicken anti-GFP polyclonal antibody ( Abcam ) as floating sections . Sections were mounted under a #1 . 5 coverslip with 2 #1 coverslips ( ∼150 µm thickness ) placed either side of the section to avoid damage to the tissue . Images were taken with a Prairie Ultima 2-photon in vivo microscope , using a Mira 900F laser at a wavelength of 795 nm ( 6 nm bandwidth ) to locally excite both Alexa-488 and -568 nm fluorescence , with a Zeiss LD Lci Plan Apochromat 25×/0 . 8NA multi-immersion lens ( 440842-9870-000000 ) . Volume imaging was acquired with 300–375 optical sections taken in 0 . 75 µm focal steps ( 2 . 13 µm axial resolution ) . The objective lens lateral resolution was defined to be 0 . 43 µm with 795 nm and NA = 0 . 8 and captured with pixels of 0 . 22 µm ( 2 , 048×2 , 048 , 440 µm field of view ) , 4 µs pixel dwell time . Best performance was achieved with Cargill Type FF immersion oil , an index of 1 . 479 , and using the glycerol with cover slip objective lens correction collar setting . Only pairs of cells exhibiting intact healthy secondary and tertiary apical and basal dendrites were imaged and used for quantification . Following acquisition , images were projected as 2-D Z-projections using Fiji6 . Dendrites were analyzed using NeuronJ plugin for Fiji6/ImageJ . 5 animals were analyzed . To quantify dendritic morphology in vitro , cultured neurons expressing GFP were imaged using a 10× objective ( NA = 0 . 17 ) , and micrographs were acquired using a Zeiss AxioCam MRm CCD camera . Dendrites were traced and binarized in ImageJ . The axon was identified by its distinct morphology and was eliminated from quantification . The following criteria for identifying apical and basal dendrites in cultured neurons were used . “Apical” dendrites were defined as the longest single protrusion , also referred to as the primary dendrite , which has the largest diameter proximal to the cell body [35] , [36] , whereas “basal” dendrites were identified as smaller and shorter protrusions , with a smaller diameter proximal to the cell body , compared to the primary dendrite ( Figure S3A–B ) . Examination of Golgi outposts in vitro and in vivo has demonstrated that the longest dendritic protrusions ( primary dendrite ) contain Golgi complexes in cultured neurons , and that in vivo , Golgi complexes are found in apical dendrites [34] , [35] . Indeed , we found that in the majority ( >90% ) of neurons in our cultures , only one dendrite , typically the longest one , was positive for giantin , a marker for the Golgi complex ( Figure S3B ) , which we have classified as the apical dendrite [34] . Dendritic length was measured in MetaMorph . For Sholl analysis , we used the Sholl analysis plugin for ImageJ ( http://biology . ucsd . edu/labs/ghosh/software ) to measure the number of dendritic processes that intersected with concentric circles spaced 25 µm apart starting at the center of the soma . For each parameter , 7–17 cells from 3–5 experiments were measured , and images were acquired and quantified by an experimenter blind to condition . For quantitative immunofluorescence experiments , coIPs , and dendrite length or number measurements , differences among condition means were identified by Student's unpaired t tests or ANOVAs performed in GraphPad Prism ( La Jolla , CA , USA ) or SPSS ( Armonk , NY , USA ) . Tukey-b or Bonferroni post hoc analyses were used for multiple comparisons . Error bars represent standard errors of the mean . For Sholl analysis , mixed model ANOVAs ( condition×distance from soma ) were conducted , with distance from soma as a repeated measure . Student's paired t tests were used to analyze paired cell morphology .
|
A fundamental feature of a neuron is the morphology of its dendrites , which are the processes that receive and integrate synaptic signals from other neurons . Neurons in the mammalian cortex exhibit two distinct dendritic arbors: apical dendrites , which extend far from the cell body , and basal dendrites , which elaborate locally around the cell body . After development , neurons must actively maintain each of these dendritic arbors to sustain their specific connectivity . Because several neurological and neurodevelopmental disorders are associated with disruptions in dendritic morphology , it is crucial to understand the molecular mechanisms that regulate the process of active maintenance of dendritic arbors . We find that disruption of a particular molecular pathway , the Ras-Epac2 pathway , can result in dramatic simplification of basal , but not apical , dendritic arbors in both cultured neurons and in the intact mouse brain . We show that a mutant form of Epac2 , identified in patients with autism , also impairs basal dendrite maintenance and disrupts its interaction with Ras . Our findings suggest that specific molecular pathways can regulate distinct dendritic regions , and that disease-related mutations can inform our understanding of the molecules that regulate important biological processes .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"molecular",
"neuroscience",
"cellular",
"neuroscience",
"neuronal",
"morphology",
"signaling",
"pathways",
"biology",
"neuroscience"
] |
2012
|
An Autism-Associated Variant of Epac2 Reveals a Role for Ras/Epac2 Signaling in Controlling Basal Dendrite Maintenance in Mice
|
Meiotic recombination is required for the orderly segregation of chromosomes during meiosis and for providing genetic diversity among offspring . Among mammals , as well as yeast and higher plants , recombination preferentially occurs at highly delimited chromosomal sites 1–2 kb long known as hotspots . Although considerable progress has been made in understanding the roles various proteins play in carrying out the molecular events of the recombination process , relatively little is understood about the factors controlling the location and relative activity of mammalian recombination hotspots . To search for trans-acting factors controlling the positioning of recombination events , we compared the locations of crossovers arising in an 8-Mb segment of a 100-Mb region of mouse Chromosome 1 ( Chr 1 ) when the longer region was heterozygous C57BL/6J ( B6 ) × CAST/EiJ ( CAST ) and the remainder of the genome was either similarly heterozygous or entirely homozygous B6 . The lack of CAST alleles in the remainder of the genome resulted in profound changes in hotspot activity in both females and males . Recombination activity was lost at several hotspots; new , previously undetected hotspots appeared; and still other hotspots remained unaffected , indicating the presence of distant trans-acting gene ( s ) whose CAST allele ( s ) activate or suppress the activity of specific hotspots . Testing the activity of three activated hotspots in sperm samples from individual male progeny of two genetic crosses , we identified a single trans-acting regulator of hotspot activity , designated Rcr1 , that is located in a 5 . 30-Mb interval ( 11 . 74–17 . 04 Mb ) on Chr 17 . Using an Escherichia coli cloning assay to characterize the molecular products of recombination at two of these hotspots , we found that Rcr1 controls the appearance of both crossover and noncrossover gene conversion events , indicating that it likely controls the sites of the double-strand DNA breaks that initiate the recombination process .
Meiotic homologous recombination is responsible for generating genetic variety among offspring as well as ensuring accurate chromosome segregation during meiotic cell divisions . The process of recombination is initiated by the formation of DNA double-strand breaks ( DSBs ) created by the highly conserved topoisomerase IV–like protein SPO11 [1] . These DSBs provide the sites at which chiasmata and crossovers form , events that are necessary for proper chromosome alignment and segregation in Meiosis I . When the repair of a DSB involves a homologous chromatid , the outcome can be recognized genetically as either a reciprocal exchange of genetic information between the homologous chromatids ( a crossover [CO] ) , or alternatively as the unidirectional acquisition of genetic information by the initiating chromatid from its non-initiating partner ( a noncrossover [NCO] , sometimes referred to as a gene conversion ) [2 , 3] . Studies in Saccharomyces cerevisiae show that COs and NCOs are the preferred outcomes of two alternative pathways in meiotic recombination , COs being predominantly produced by the Double-Strand Break Repair ( DSBR ) pathway , and NCOs predominantly produced by the Synthesis-Dependent Strand-Annealing ( SDSA ) pathway [4 , 5] . Importantly , in all organisms , meiotic recombination does not occur at uniform rates along chromosomes . In both yeast and mammals—the most extensively studied cases—recombination rates vary considerably along the length of a chromosome [6–10] . When examined at high resolution , the great majority of recombination , possibly all , occurs in restricted regions , termed hotspots , that are typically 1–2 kb long in humans and mice [11 , 12] . In contrast to the considerable body of information describing the participation of a variety of proteins in the overall processes of recombination , relatively little is presently understood about the factors determining the chromosomal locations and relative activity of recombination hotspots . In yeast , hotspots have been classified into three groups based on their presumed activation requirements . Activation of “α” hotspots requires transcription factors; the “β” hotspots require the presence of nuclease-sensitive chromatin ( a necessary , but not sufficient , condition ) ; and the “γ” hotspots are dependent on the G+C content of DNA ( reviewed in [13] ) . Many yeast hotspots can be assigned to more than one class due to multiple mechanisms involved in the initiation of recombination . There is no obvious consensus sequence defining hotspots in either S . cerevisiae or Schizosaccharomyces pombe with the exception of the class of hotspots in S . pombe , which have an 18-bp consensus sequence containing the CRE-heptameric cyclic AMP response element ATGACGT [14–16] . However , this sequence accounts for only a minority of hotspots in S . pombe; the defining elements of the remainder are unknown . A 13-bp consensus sequence has been identified that is present in 41% of human hotspots [17]; although this sequence is highly enriched in hotspots , its presence alone is not sufficient to initiate hotspot activity , suggesting that other presently unknown factors are also required . This 13-bp sequence also has the interesting property of serving as a site of spontaneous DNA breakage in mitochondrial DNA . That the location of a hotspot is not determined simply by its internal DNA sequence was shown by DSB mapping experiments in S . cerevisiae . The DSBs initiating recombination preferentially occurred within a window of 100–500 bp near the center of the hotspot [18–21] . However , replacement of these preferred sites for DSB formation did not eliminate DSB formation there , and DSBs now occurred at the replacement sequence [22 , 23] . Trans-acting factors controlling hotspot activation have been identified in several cases in yeast . Binding of the ATF1/PCR1 transcription factor is required for activity of the aforementioned ADE6-M26 hotspot in S . pombe [24] , and activity of the HIS4 hotspot in S . cerevisiae requires binding by the transcription factors BAS1 , BAS2 , and RAP1 [25] and GCN4 [26] . A more extensive analysis of the effect of BAS1 has shown that loss of this protein can either reduce or increase the recombination activity of a number of S . cerevisiae hotspots [27] . Recent data indicate that such trans-acting factors may act through posttranslational modifications of histones with attendant nucleosome rearrangements . For example , in the MAT2-MAT3 cold region in S . pombe , the cooperative action of histone deacetylases and histone methyltransferases contribute to recruitment of heterochromatin proteins , keeping the region both transcriptionally and recombinationally silent [28] and directing recombination to the adjacent mating-type locus [29] . Regulation of recombination by histone methyltransferases has also been shown in S . cerevisiae [30 , 31] and Caenorhabditis elegans [32] , and histone H2B ubiquitination has been shown to play a role in DSB formation , by recruitment and/or stabilization of DSB-initiating factors through RAD6-BRE1 [33] . The most detailed analysis of the influence of chromatin modifications on meiotic recombination has been achieved for the ADE6-M26 hotspot in S . pombe , at which a set of histone acetyltransferases and ATP-dependent chromatin remodeling factors alter chromatin structure , regulating both transcription and recombination [34] . In contrast to what is known in yeast , we know considerably less about possible trans-acting factors influencing the location and relative activity of mammalian hotspots , although there is now evidence that such factors exist . The MSTM 1a and 1b hotspots in humans vary considerably in activity among individual males even when they share the same haplotype within and around the hotspots themselves [35] , suggesting control by either trans-acting factors or very distant cis-acting factors; and in mice , Baudat and de Massy [36] have shown that initiation of recombination at the Psmb9 hotspot on chromosome 17 ( Chr 17 ) requires the presence of a trans-acting gene located some distance proximal to the site of the hotspot . To systematically explore the possible existence and identity of trans-acting factors controlling hotspot specificity in mammals , we have taken advantage of the possibilities inbred mouse strains provide for genetic analysis and compared the recombination maps generated along a region of mouse Chr 1 when this region was always heterozygous C57/BL6 ( B6 ) × CAST/EiJ ( CAST ) , and the rest of the genome either did or did not carry CAST alleles . In doing so , we found hotspots whose activity was either dependent upon , or suppressed by , the presence of CAST alleles at distant loci , whereas other hotspots were unaffected . Assaying the activity of specific hotspots in the sperm of males segregating in genetic crosses , we found that the activity of several hotspots depended on a single Mendelian factor we have designated Recombination regulator 1 ( Rcr1 ) that maps to a 5 . 30-Mb region on proximal Chr 17 . Molecular assays of individual products of recombination at these hotspots indicated that Rcr1 acts to control the initiation of recombination rather than the choice between the CO and NCO pathways of the recombination process . In their accompanying paper , Grey et al . [37] describe a similar trans-acting locus controlling the appearance of both CO and NCO gene conversions at the Psmb9 hotspot on Chr 17 of the mouse as well as elsewhere in the genome .
The existence of trans-acting genes became apparent when comparing the recombination maps obtained by two genetic crosses involving the B6 and CAST mouse strains , which were chosen for their genetic diversity . In the first cross , hereafter referred to as the interstrain cross , B6 mice were mated to CAST , and the F1 hybrids were backcrossed to B6 . In the second cross , hereafter referred to as the congenic cross , B6 were mated to B6 . CAST-1T , a congenic strain carrying 100 Mb of CAST DNA sequences from distal Chr 1 introgressed into C57BL/6J ( see Materials and Methods for details of this strain ) ; the resulting F1 hybrids were then backcrossed to B6 . In both crosses , the F1 hybrids shared the same heterozygous 100-Mb segment on Chr 1 . The fundamental difference was the presence of CAST alleles in the remainder of the genome in the interstrain B6xCAST F1 animals and their absence in congenic B6xB6 . CAST-1T F1 mice , which are homozygous B6 outside the 100-Mb Chr 1 region ( Figure 1 ) . In both crosses , recombination was tested in an 8-Mb region ( 183 . 5–191 . 5 Mb , National Center for Biotechnology Information [NCBI] build 36 ) located within the 100-Mb heterozygous B6/CAST region . The B6xCAST recombination map utilized the products of 6 , 028 meioses occurring in F1 animals ( 3 , 002 offspring of female F1 and 3 , 026 offspring of male F1 ) and was part of the entire map of Chr 1 described previously [10] . Among these offspring , 735 contained a single CO event in the 8-Mb region ( 264 arising in female F1s and 471 in male F1s ) , producing a total sex-averaged map length of 12 . 2 cM ( 8 . 8 cM in females and 15 . 6 cM in males ) , or 1 . 52 cM/Mb ( 1 . 1 cM/Mb and 1 . 95 cM/Mb in females and males , respectively ) . The B6xB6 . CAST-1T map utilized 2 , 083 meioses ( 1 , 173 offspring of female F1 and 910 offspring of male F1 ) , of which 175 meioses provided a single CO event in the 8-Mb region ( 83 from females and 92 from males ) . The sex-averaged map length of the region in this cross was 8 . 4 cM ( 7 . 1 cM in females and 10 . 1 cM in males ) , or 1 . 05 cM/Mb ( 0 . 89 cM/Mb and 1 . 26 cM/Mb in females and males , respectively ) . To test for any possible effect of genomic imprinting on recombination rates , half of the offspring of each cross were derived from F1 animals in which the dam was B6 and the sire was CAST or B6 . CAST-1T , and the other half were derived from a reciprocal parental combination; there were no significant differences in hotspot locations between these reciprocal crosses . The COs occurring in the 8-Mb region of Chr 1 were mapped to hotspot-level resolution using the same markers for both crosses . Figure 2 presents the female and male recombination maps obtained . Although most regions showed similar activity in both crosses , there were nine regions , indicated by arrows , where the results differed dramatically; six hotspots disappeared in the congenic cross ( Fbxo28 , Dusp10 , Hlx1 , D1Pas1 , Esrrg-1 , and Kcnk2 , named after their closest genes ) , and three regions that were devoid of recombination in the interstrain cross contained new , active hotspots in the congenic cross ( Capn2 , Kctd3 , and Ptpn14 ) . Three hotspots in the interstrain cross ( Hlx1 , Esrrg-1 , and Kcnk2 ) and one in the congenic cross ( Kctd3 ) showed statistically significant sex differences . In no case did we find a hotspot whose activity depended on a CAST allele in one sex , but not the other , in a statistically significant manner . Particularly notable were the twin hotspots Esrrg-1 and Esrrg-2 , which in the interstrain cross are separated by less than 5 kb and are significantly more active in males than in females . Only the hotspot proximal to the centromere , Esrrg-1 , disappeared in the congenic cross; the other , Esrrg-2 , remained active , although with reduced activity ( inserts on Figure 2A and 2B ) . The probabilities of these results being observed by chance are very low ( Table 1 ) . One of the regions ( 188 . 7–189 . 1 Mb ) in which recombination appeared in the congenic cross stretched across several markers , indicating that this region contains several distinct , activated hotspots . The rates at regions with centromere-proximal ends at 185 . 220 , 185 . 900 , and 187 . 485 Mb were similar , and others like those at 187 . 827 , 189 . 584 , 189 . 785 , and 190 . 001 Mb were active in both crosses but with different activities . It is apparent from these results that the products of CAST alleles of distant loci can activate or suppress the activity of individual hotspots without affecting other hotspots in the same chromosomal region . Two mapping crosses were used in searching for trans-acting genes regulating recombination at specific hotspots . In the first cross , F1 females derived from a cross between the B6 . CAST-1T congenic strain and CAST were backcrossed to B6 ( Figure 3A ) . The progeny of this cross were all heterozygous B6/CAST at the 100-Mb congenic region and segregated CAST alleles in the remainder of the genome . The advantages of this cross were that it allowed the detection of any relevant X-linked genes; that all male progeny were informative , and that any CAST alleles were always present in the heterozygous condition , as they were in the F1 animals where the differences in hotspot activity were originally detected . A total of 211 male animals from this cross were individually phenotyped for activity of three hotspots Hlx1 ( 186 . 316 Mb , located 110 kb away from the gene's 3′-end ) , Esrrg-1 ( 189 . 778 Mb , together with its neighbor Esrrg-2 located in intron 3–4 of Esrrg ) and Kcnk2 ( 191 . 027 Mb , located in intron 2–3 of its namesake gene ) . Phenotyping was carried out using allele-specific sperm DNA assays ( see Materials and Methods ) ; these assays gave a plus/minus phenotype for each of the three hotspots; a hotspot was either active or missing from genetically segregating mice ( Figure 4 ) . Unfortunately , because suitable single nucleotide polymorphism ( SNP ) combinations were not available , it was not possible to develop nested PCR assays for any of the three hotspots suppressed by the presence of trans-acting CAST alleles . To address possible dosage effects of CAST alleles on recombination activity , a second cross was carried out in which B6xCAST F1 animals were mated together; the half of the resulting F2 male progeny that were heterozygous at the distal part of Chr 1 ( including the entire congenic region between the microsatellite markers D1Mit145 located at 169 . 132 Mb and D1Mit510 located at 194 . 118 Mb ( Figure 3B ) were tested for hotspot activity . As with the first mapping cross , these animals were heterozygous for the distal part of Chr 1 , but segregated both B6 and CAST alleles in either the homozygous or heterozygous state in the remainder of the genome . In all , 98 animals from this F2 cross representing 196 meioses were phenotyped . Phenotyped animals from both crosses were genotyped with 165 SNP markers spaced across the genome , ensuring 20-Mb resolution ( Table S1 ) . The genotyping and phenotyping data were analyzed by the R/QTL-based software package J/qtl ( http://research . jax . org/faculty/churchill/software/Jqtl/index . html ) . The results showed strong linkage between hotspot activities and a single interval on proximal Chr 17 located between 5 and 25 Mb , with LOD scores above 30 for hotspots Hlx1 and Esrrg-1 and around 8 for hotspot Kcnk2 in the congenic backcross ( Figure 5A ) , and above 10 for hotspots Hlx1 and Esrrg-1 in the interstrain cross ( Figure 5B ) . The two crosses produced identical map locations , providing evidence that CAST alleles in either homozygous or heterozygous condition activate recombination at the analyzed hotspots . No other chromosome location showed significant linkage in either cross . The reason for the lower LOD scores with the Kcnk2 hotspot is the lower efficiency of the nested PCR assay for this hotspot , which although it never gave a false-positive result with control DNA samples , did not always give a positive result with samples known to contain COs . We have designated the Chr17 locus Recombination regulator 1 ( Rcr1 ) . To further refine the location of Rcr1 , all of the crossovers occurring between 5 and 25 Mb on Chr 17 were typed for a combination of microsatellite and SNP markers ( Figure 5C ) . The left border of the critical interval was located between 11 . 74 Mb ( D17Mit113 ) and 13 . 01 Mb ( NES15751522 ) , and the right border was located between 16 . 14 Mb ( NES12260613 ) and 17 . 04 Mb ( NES12247255 ) , showing that Rcr1 must lie in the 5 . 30-Mb interval between 11 . 74 and 17 . 04 Mb on Chr17 . We tested whether Rcr1 controls the earlier stages of recombination process , between the initiation of DSB and the formation of recombination intermediates , or the later decision to process the recombination intermediates into either COs or NCOs . If Rcr1 acts early in recombination , it should control the appearance of both COs and NCO gene conversions at susceptible hotspots . If , however , it acts on the choice between CO and NCO pathways as alternative outcomes of repairing the DSBs that initiate recombination , we would expect to see persistence of NCOs at susceptible hotspots in the absence of the Rcr1 CAST allele . A cloning assay counting the number of COs and NCOs at individual hotspots in F1 sperm DNA [38] was used to make this distinction . In essence , the region containing the hotspot was amplified from sperm DNA using primers common to both B6 and CAST . The amplified product was then cloned into E . coli such that the mammalian DNA in each clone is derived from a single strand of an individual sperm , and the individual clones were genotyped ( see Material and Methods ) . This assay was feasible for the Hlx1 and Esrrg-1 hotspots , in which the availability of suitable internal markers facilitated the detection of both COs and NCOs . In B6/CAST F1 sperm , in which these hotspots are expected to be active , both COs and NCOs were present at the two hotspots , but in B6/B6 . CAST-1T F1 sperm , in which the hotspots are expected to be inactive , NCOs as well as COs were entirely absent from these hotspots , suggesting that Rcr1 controls the initiation of recombination ( Table 2 ) .
Our major findings are , first , that the recombination activity of several hotspots on mouse Chr 1 is either activated or suppressed by the presence of CAST allele ( s ) of distant trans-acting loci; second , that activation of several of the hotspots is controlled by a locus , Rcr1 , located within a 5 . 30-Mb window on proximal Chr 17; and third , that Rcr1 exerts its effect at the initiation of recombination , prior to the CO–NCO choice . Rcr1 does not have a generalized effect on recombination rates as many of the hotspots within the 8-Mb segment of Chr 1 mapped here were not affected by the presence versus absence of a CAST allele at this locus; moreover , the number of regions exhibiting greater than 0 . 1% recombination in the congenic cross ( 21 ) is similar to the number of such regions ( 20 ) in the interstrain cross across ( Figure 2 ) . Although the total recombination rate across this 8-Mb region is somewhat lower in the congenic cross compared to the interstrain cross , presumably due to the loss in the congenic cross of several hotspots that were highly active in the interstrain cross , this appears to be a localized effect as the overall recombination rate across the larger interval 169–193 Mb was the same in both crosses ( unpublished data ) . Some of the hotspots in the 8-Mb tested region remained active in both crosses , leaving open the question of what controls their activity . Presumably , these hotspots are activated either by the B6 allele of Rcr1 , which was always present , or by other trans-acting genes . Previous observations suggest that there are approximately 13 , 500–14 , 000 hotspots across the mouse genome active in the same B6xCAST cross used in this study [10] , and the known variation in hotspot usage among different mouse crosses suggests a potentially even higher number for the species as a whole . This is true for the human genome as well , which is estimated to contain an even larger number of hotspots [9] . Given these high numbers , it is unlikely that each hotspot is controlled by its specific trans-acting gene , and it is more likely that , as in the present case , a trans-acting gene controls a family of hotspots . However , the only prior indications that such families might exist were the finding that the 13-bp consensus sequence CCNCCNTNNCCNC occurs in 41% of human hotspots ( [17] and the existence of the ATF1 . PCR1 transcription factor of S . pombe which binds to an 18-bp consensus sequence , effecting a chromatin reorganization in the region and activating recombination [16] . It is interesting that although activation of the Hlx1 hotspot requires the presence of a CAST allele at Rcr1 , former investigations showed that recombination at this hotspot initiates almost three times more frequently on the B6 chromatid then on the CAST one [10] . The origin of this seemingly contradictory effect may lie in the so-called “hotspot paradox” [39] , which postulates that whenever there are cis-acting sequences influencing the activity of hotspots , because in the process of recombination , the more active initiating chromatid acquires the DNA sequence of its less active partner more frequently than the reverse , hotspots will slowly degrade their activity over time as any mutations diminishing activity accumulate in the population . In the case of Hlx1 , which is activated by the CAST allele present in Mus musculus castaneus , but not by the B6 allele present in M . m . domesticus , it may be that over evolutionary time , the Hlx1 haplotype has been under selection pressure for diminished activity in M . m . castaneus but not M . m . domesticus , leaving the B6 Hlx1 haplotype closer to the more active primordial sequence . Although indirect , this may be the first experimental evidence that the hotspot paradox does operate over evolutionary time . These results with hotspot Hlx1 demonstrate that activity of a hotspot can be determined by the interaction of a trans-acting factor with cis-acting DNA sequences on each chromatid . A similar interaction between cis- and trans-acting elements has been reported at the Psmb9 hotspot on mouse Chr 17 [36] . In their accompanying paper , Grey et al [37] now report that this trans-acting element ( Dsbc1 ) is located on Chr 17 within a 6 . 7-Mb region between 10 . 1 and 16 . 8 Mb , and affects the distribution of recombination in other regions of Chr 17 , Chr 15 , and Chr 18 , as well as the Hlx1 hotspot described here . The Dsbc1 interval overlaps the location of Rcr1 , suggesting that Rcr1 and Dsbc1 are likely to be the same gene or at least members of the same gene family . The evidence that Rcr1 acts prior to the choice between processing DSBs into COs or NCOs derives from the fact that both CO and NCO products of recombination at hotspots Hlx1 and Esrrg-1 depend upon the presence of a CAST allele at Rcr1 . If mammalian recombination processes parallel those in the yeast , this choice is made very early in the recombination process [4 , 5 , 40] . Given that result , and the difficulty of sustaining any unrepaired DSBs during meiosis , it seems highly likely that Rcr1 influences the choice of sites for the SPO11 catalyzed DSBs that initiate the recombination process . At this point , the molecular identity of Rcr1 is unknown . The closest known phenotypic parallel is the ADE6-M26 hotspot in S . pombe that is activated via chromatin remodeling mediated by the ATF1 . PCR1 transcription factor . Somewhat less similar is the effect of histone deacetylase SIR2−/− mutants on recombination at multiple sites in S . cerevisiae [41] . In this latter case , the magnitude of the effects and their frequency differ appreciably from what is seen in mice; only 12% of recombination sites were affected , and less than 1% of those affected ( <0 . 1% of total sites ) showed greater than 5-fold difference between mutant and wild type . Moreover , the 12% of recombination sites affected at any level tended to be regionally concentrated depending on whether they showed increased or decreased recombination . Mieczkowski et al . [41] drew the reasonable conclusion that SIR2 likely affects regional chromatin configurations . If Rcr1 acts by modifying chromatin , it must do so over a very small range as the Esrrg-1 and Esrrg-2 hotspots differed markedly in their responses to the presence of an Rcr1 CAST allele despite being less than 5 kb apart . Whatever the molecular nature of the Rcr1 gene product , the existence of such trans-acting factors offers a potential means of explaining several presently enigmatic features of mammalian recombination , including varying hotspot activities ( the relative affinity of adaptors for their cognate sequences ) , sex differences in activities of individual hotspots ( differential rates of transcription ) , and the failure to find a consensus DNA sequence that accounts for the specificity of SPO11 cleavage ( if SPO11 acts by recognizing the existence of complex between a hotspot and its cognate adaptor , each family of hotspots may have its own , unique consensus sequence ) .
C57BL/6J and CAST/EiJ were obtained from the Jackson Laboratory . B6 . CAST-1T ( rs3022828–rs13476307 ) was kindly provided by Dr . Wesley Beamer [42] . The Jackson Laboratory is American Association for Laboratory Animal Science ( AALAS ) accredited , and the Jackson Laboratory Animal Care and Use Committee approved all animal procedures . The sample preparation was done as described earlier [10] . Sperm phenotyping used DNA from 12-wk-old animals . Sperm DNA was isolated by a modified protocol using the DNeasy tissue kit ( Qiagen ) as described before [38] . The E . coli cloning assay was applied to hotspots Hlx1 and Esrrg-1 as described previously [38] . The hotspot sequence was amplified with primers common to both parents , and the DNA fragments were cloned in E . coli such that each colony represents a single DNA strand from the initial meiotic event . Fluorescent SNP genotyping was carried out directly on an aliquot of E . coli cultures grown from each colony in 96-well plates . Recombination activity was detected by selectively amplifying recombinant DNA fragments at the hotspot sequence of interest . Sperm DNA samples were subjected to two rounds of nested PCR using allele-specific primers in each of the two rounds . The two pairs of primers were oriented in the B–C combination: the proximal pair was specific to B6 alleles flanking the hotspot , and the distal pair to CAST alleles . The primers were PTO-modified at the last three nucleotides of their 3′-end ( MWG-Biotech ) . All primer sequences and positions are summarized in Table S2 . The PCR conditions for each tested hotspot were empirically established to ensure that the amplified product is only from the recombinant class ( B–C ) and not the parental types ( B–B and C-–C ) ( Figure 4 ) . The PCR reaction was performed on an Eppendorf PCR system ( Eppendorf AG ) , using 50-ng initial amount of DNA template and 0 . 23 mM each dNTP , 0 . 23 μM of each primer , 1× TITANIUM Taq PCR Buffer , 0 . 4 U TITANIUM Taq DNA Polymerase ( Clontech ) for the first round of PCR . The amplified product was diluted 500 times and used for the second allele-specific PCR . The PCR cycling conditions for hotspot Hlx1 were first round—initial denaturizing step at 94 °C for 4 min , and 45 rounds of 94 °C for 1 min , 63 °C for 40 s , 72 °C for 3 min 20 s , and a final extension at 72 °C for 10 min; second round—initial denaturizing step at 94 °C for 1 min , and 35 rounds of 94 °C for 50 s , 64 °C for 35 s , 72 °C for 1 min , and a final extension at 72 °C for 7 min . The cycling conditions for the Esrrg-1 hotspot were first round—initial denaturizing step at 94 °C for 2 min , and 30 rounds of 94 °C for 1 min , 63 . 2 °C for 1 min , 68 °C for 2 min 30 s , and a final extension on 72 °C for 7 min; second round was the same as for hotspot Hlx1 . For hotspot Kcnk2 , the cycling conditions were first round—initial denaturizing step at 94 °C for 4 min , and 45 rounds of 94 °C for 1 min , 63 °C for 40 s , 72 °C for 4 min 30 s , and a final extension at 72 °C for 10 min; second round—initial denaturizing step at 94 °C for 4 min , and 35 rounds of 94 °C for 1 min , 64 °C for 30 s , 72 °C for 4 min , and a final extension at 72 °C for 10 min . The amplified product was run through standard 2% agarose gel ( Invitrogen ) with ethidium bromide and visualized under UV light . Fine mapping of recombination activities in the region of 183 . 5–191 . 5 Mb on Chr 1 was carried out using SNP markers and Amplifluor SNPs and the HT FAM-JOE System ( Millipore ) . All markers used in this study and their positions according to NCBI build 36 are summarized in Table S3 . For genome-wide association mapping , all progeny were genotyped at 20-Mb resolution using the KASPar genotyping system ( KBiosciences ) . The markers were selected from the Jackson Laboratory genotyping panel [43] . The data were analyzed by SNPviewer2 ( KBiosciences ) . Fine mapping of CAST alleles on Chr 17 was done by a combination of the microsatellite markers D17Mit48 , D17Mit57 , D17Mit113 , and D17Mit46 ( Invitrogen ) and SNP markers as shown on Figure 5C . Data analysis was performed using the J/qtl software package ( http://research . jax . org/faculty/churchill/software/Jqtl/index . html ) . The linkage between phenotyping and genotyping data was estimated by One QTL Genome Scan using the Imputation method and 10 , 000 permutations per scan . LOD scores above 3 were considered statistically significant ( p < 0 . 05 ) .
|
Recombination is an essential aspect of meiosis , ensuring proper contact and exchange of genetic material between homologous parental chromosomes , as well as their subsequent segregation to produce haploid gametes . In humans and mice , recombination events are located at preferential sites termed hotspots , whose placement and activity are tightly regulated . We have now identified a hotspot-regulating locus in mammals , Rcr1 , that simultaneously controls the locations of multiple hotspots . The discovery of Rcr1 indicates the existence of a newly emerging class of genes important in the recombination processes . Gaining further insights into their function may contribute to a better understanding of genetic factors underlying human fertility and evolution .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"evolutionary",
"biology",
"molecular",
"biology",
"genetics",
"and",
"genomics"
] |
2009
|
Trans-Regulation of Mouse Meiotic Recombination Hotspots by Rcr1
|
It is now established that the central nervous system plays an important role in regulating whole body metabolism and energy balance . However , the extent to which sensory systems relay environmental information to modulate metabolic events in peripheral tissues has remained poorly understood . In addition , it has been challenging to map the molecular mechanisms underlying discrete sensory modalities with respect to their role in lipid metabolism . In previous work our lab has identified instructive roles for serotonin signaling as a surrogate for food availability , as well as oxygen sensing , in the control of whole body metabolism . In this study , we now identify a role for a pair of pheromone-sensing neurons in regulating fat metabolism in C . elegans , which has emerged as a tractable and highly informative model to study the neurobiology of metabolism . A genetic screen revealed that GPA-3 , a member of the Gα family of G proteins , regulates body fat content in the intestine , the major metabolic organ for C . elegans . Genetic and reconstitution studies revealed that the potent body fat phenotype of gpa-3 null mutants is controlled from a pair of neurons called ADL ( L/R ) . We show that cAMP functions as the second messenger in the ADL neurons , and regulates body fat stores via the neurotransmitter acetylcholine , from downstream neurons . We find that the pheromone ascr#3 , which is detected by the ADL neurons , regulates body fat stores in a GPA-3-dependent manner . We define here a third sensory modality , pheromone sensing , as a major regulator of body fat metabolism . The pheromone ascr#3 is an indicator of population density , thus we hypothesize that pheromone sensing provides a salient 'denominator' to evaluate the amount of food available within a population and to accordingly adjust metabolic rate and body fat levels .
In relation to fat metabolism and energy balance , the central nervous system plays a more intricate role than historically thought . Initially believed to exert its effects on adiposity predominantly through promoting food intake , several studies have now demonstrated that the underlying neuronal circuits , genetic , molecular and endocrine pathways that regulate body fat reserves in the peripheral metabolic organs are distinct from those that regulate feeding behavior [1–6] . In addition to the pre-eminent role of the mammalian hypothalamus , the sensory nervous system has also been shown to play an important role in regulating whole body metabolism and physiology [7 , 8] . Broad sensory dysfunction in humans can be exemplified by ciliopathies such as Bardet-Biedl Syndrome , that leads to profound obesity [9] . In contrast , enhanced sensory environments improve metabolic homeostasis [10] . However , identifying discrete sensory neurons with instructive roles in lipid metabolism has been a challenging undertaking in any system . In the metazoan Caenorhabditis elegans , the nervous system is well-defined at an anatomic and functional level [11 , 12] . The sensory nervous system plays a profoundly important role in regulating whole body physiology and lifespan [13–15] . We and others have shown that the sensory nervous system is an important regulator of systemic lipid metabolism [3 , 16 , 17] . For example , the presence of food , relayed by serotonergic sensory neurons and amplified by the octopaminergic neurons ( octopamine is the invertebrate analog of noradrenaline ) is one salient input that regulates the magnitude of fat loss in the intestine [3] . The intestine is the predominant metabolic organ for C . elegans , and expresses all of the genes involved in lipid metabolic processes including fat synthesis , breakdown and its regulation [7 , 18 , 19] . Furthermore , conserved intestinal fatty acid beta-oxidation has been shown to play a central role in the biosynthesis of the ascarosides , a family of small-molecule pheromones that regulate many aspects of C . elegans physiology and behavior [20 , 21] . Thus , metabolic changes in the intestine effectively encapsulate whole body metabolism . The mechanisms governing lipid metabolism are ancient and well-conserved across metazoans [22–28] , therefore C . elegans offers an excellent platform to identify new genes and molecular mechanisms underlying neuronal control of fat metabolism , using unbiased approaches . To systematically examine the role of the sensory nervous system in regulating whole body lipid metabolism , we undertook a screen of the 19 ( of 21 ) viable Gα protein mutants for changes in body fat content [29] . This family of heterotrimeric G proteins is well-known to regulate intracellular signaling cascades in response to changes in the environment , which in turn control many aspects of physiology and behavior [30–32] . An added advantage of this family is that the majority of null mutants are viable , and the anatomical locations of these genes have been well-defined . One gene identified from this screen is the Gα protein , GPA-8 , the ortholog of the mammalian gustducin proteins that regulates intracellular cGMP . Previous work from our lab has shown that GPA-8 is expressed in the C . elegans body cavity neurons , and integrates oxygen-sensing with the sensing of internal metabolic state , to drive the rate and extent of fat loss [29 , 33] . Thus , we found that environmental oxygen serves as a second physiologically relevant sensory input for the regulation of lipid metabolism . The most potent 'hit' from our Gα protein screen is called GPA-3 , and is a member of the Go/Gi protein family . In this study , we identify the neurons in which GPA-3 functions , define its cellular mechanism of action and the critical downstream neurotransmitter required for its functions in promoting fat loss . In so doing , we define pheromone sensing as a new sensory modality for the regulation of lipid metabolism .
gpa-3 ( pk35 ) null mutants ( henceforth gpa-3 ) exhibit a significant decrease in body fat content , as judged by Oil Red O staining of fixed adult animals followed by quantification of lipid droplets in the intestinal cells ( Fig 1A and S1A and S1B Fig ) , and by biochemical extraction of triglycerides from whole adult animals ( Fig 1B ) . The reduced body fat content of gpa-3 mutants could not be explained by differences in locomotor behavior , which is indistinguishable between wild-type and gpa-3 mutants ( Fig 1C ) [34] . Our previous work had identified a highly conserved lipase called adipocyte triglyceride lipase ( ATGL-1 ) that is rate-limiting for fat loss via the conversion of triglycerides to energy by β-oxidation [3] . Work from other groups has shown that the ATGL-1 protein is stabilized by phosphorylation during fasting thus promoting fat loss [35] . atgl-1 is expressed in the intestine , and is transcriptionally induced in response to neuronal signals that stimulate fat loss . Changes in atgl-1 transcription are tightly correlated with rates of lipolysis [7 , 36] , thus changes in atgl-1 mRNA reflect physiological shifts in energy utilization . Relative to wild-type animals , gpa-3 mutants have a robust increase in ATGL-1 expression in the intestine ( Fig 1D and 1E ) . Our results indicate that increased fat utilization via induction of triglyceride hydrolysis underlies the reduced body fat of gpa-3 mutants . To corroborate our experiments using the atgl-1 reporter line , we conducted qPCR studies and found an approximately 2 . 5 fold increase in atgl-1 mRNA in gpa-3 mutants ( Fig 1F ) . Furthermore , RNA-mediated inactivation of ATGL-1 resulted in a nearly 2-fold suppression of fat loss in the gpa-3 mutants ( Fig 1G and 1H and S1C Fig ) . Together , these results show that increased triglyceride hydrolysis is one major mechanism underlying the decreased body fat stores of gpa-3 mutants . GPA-3 is orthologous to the mammalian cAMP-regulating Gαo/i class [31] , sharing 73% similarity ( 7e-139 ) . Gαo/i family members are known to regulate intracellular cAMP via inhibition of adenylyl cyclases . The C . elegans cAMP adenylyl cyclase ACY-1 is expressed in neurons , and viable loss-of-function ( nu239 ) mutants are available [37] . Although the acy-1 ( nu329 ) mutants did not show an appreciable difference in body fat ( Fig 2A and S2C Fig ) , removal of acy-1 in the gpa-3 mutants resulted in a near-complete suppression of the gpa-3 body fat phenotype . To provide a second , molecular readout of the fat loss , we measured atgl-1 mRNA by qPCR , and found that the gpa-3-mediated induction of atgl-1 in the gpa-3 mutants was also suppressed in the gpa-3;acy-1 double mutants ( Fig 2B ) . Thus , ACY-1 function is required downstream of GPA-3 in the regulation of body fat via the induction of ATGL-1-mediated lipolysis . acy-1 mutants have a mild locomotor defect and gpa-3;acy-1 double mutants have a statistically significant additive effect ( Fig 2C ) . However because gpa-3 mutants themselves do not have a locomotor phenotype , these effects are non-specific to the gpa-3 fat regulatory pathway . To determine the direction of the effect of increased cAMP on body fat , we exogenously administered a non-hydrolyzable analog of cAMP called 8-Bromo-cAMP ( 8-Br-cAMP ) , which led to a dose-dependent decrease in body fat stores ( S2A Fig ) . Together , our results indicate that GPA-3 controls fat utilization through inhibition of the adenylyl cyclase ACY-1 , and the resultant regulation of cAMP concentrations ( S2B Fig ) . Examination of our gpa-3-expressing transgenic lines showed that GPA-3 is solely expressed in the nervous system and not in the intestine , where its metabolic phenotype manifests ( Fig 2D ) . GPA-3 is expressed in 9 bilaterally symmetric pairs of amphid sensory neurons with ciliated endings that are directly exposed to the environment: ADF , ADL , ASE , ASG , ASH , ASI , ASJ , ASK , and sporadically in AWA ( Fig 2E and 2F ) , confirming previous observations [38 , 39] . The localization of GPA-3 to the amphid sensory neurons suggests that the subset of neurons from which GPA-3 regulates body fat either directly or indirectly regulate a long-range neuroendocrine factor that acts in the intestine to elicit fat loss . To identify the neurons in which GPA-3 acts to regulate fat stores in the intestine , we generated expression constructs to drive gpa-3 cDNA in subsets of amphid neurons in which gpa-3 is normally expressed . The transgenic rescue strategy is given in Fig 3A and 3B . In gpa-3 null mutants , restoration of gpa-3 cDNA using either 5kb or 7kb of endogenous gpa-3 upstream regulatory regions , gave significant restoration of intestinal fat content ( Fig 3C and S3A Fig ) , confirming that GPA-3 functions in sensory neurons to regulate intestinal fat stores . However , we noted that in the gpa-3 transgenic animals , neither the 5kb nor the 7kb promoter were sufficient to confer a complete restoration of body fat stores ( S3A Fig ) , which prompted us to examine the feeding behavior of gpa-3 mutants . Relative to wild-type animals , gpa-3 mutants displayed ~15–20% decrease in food intake ( S3B Fig ) . However , we found that re-expression of gpa-3 under either 5kb or 7kb promoters did not restore food intake to wild-type levels ( S3B Fig ) . We next wanted to test whether acy-1 mutants suppressed the decreased food intake of gpa-3 mutants , and accordingly measured food intake in the relevant mutants . We found that acy-1 mutants also displayed decreased food intake to a similar extent as the gpa-3 mutants . However , the gpa-3;acy-1 double mutants resembled either single mutant alone ( S3C Fig ) . Thus , unlike the suppression of GPA-3-mediated fat loss ( Fig 2A ) or its induction of atgl-1 ( Fig 2B ) , acy-1 mutants do not suppress GPA-3-mediated food intake . We wanted to further examine the role of GPA-3 in specific subsets of neurons . Accordingly , we devised a transgenic rescue strategy that allowed us to include or exclude a role for GPA-3 in subsets of neurons in which it is expressed ( Fig 3A ) . In gpa-3 null mutants , restoration of gpa-3 cDNA expression using the srb-6 ( ADL , ADF , ASH ) and nlp-7 ( ADL , ASI , ASE ) promoters , but not the gpa-14 ( ASH , ASI , ASK , ASJ ) promoter significantly restored intestinal fat content ( Fig 3C ) . This combinatorial strategy first eliminated a role for GPA-3 in ASI , ASH , ASK and ASJ neurons and second , identified a potential role for the ADL neurons because it is the only neuron pair that overlaps between the two rescuing promoters , srb-6 and nlp-7 . We next drove gpa-3 cDNA expression in the individual neurons ADL , ASG and ASE using neuron-specific promoters ( Fig 3B; the AWA neurons were not tested ) . Restoration of gpa-3 in the ADL neurons alone significantly restored body fat stores in the gpa-3 null mutants ( Fig 3C ) . Thus , GPA-3 function in the ADL neurons regulates body fat stores . Although the transgenic rescue strategy revealed a clear role for GPA-3 in controlling body fat in subsets of neurons , in no case were we able to restore the feeding phenotype of gpa-3 mutants , including the endogenous promoter that was sufficient to restore fat stores ( Fig 3C and S3A and S3B and S3D Fig ) . These results suggest the possibility that background effects unrelated to the gpa-3 gene contribute to the reduced feeding phenotype in these mutants , despite the gpa-3 mutant having been outcrossed 7 times . Another possibility is that although unusual in C . elegans [40] , additional regulatory elements further upstream from the chosen 7kb gpa-3 promoter region may play a role in controlling gpa-3 expression . However , our data also suggest that the reduced food intake only accounts for a small percentage of the net change in body fat stores , because expression of gpa-3 in the ADL neurons significantly restores body fat stores without altering food intake ( Fig 3C ) . Together our data suggest that gpa-3 functions in the ADL neurons to regulate fat content , independent of changes in food intake or locomotion . To determine the necessity of GPA-3 in the ADL neurons for the regulation of body fat , we conducted antisense mediated inhibition experiments [41] using an ADL-specific promoter . Inactivation of gpa-3 in the ADL neurons lowered fat content to 65% of that seen in wild-type animals ( Fig 4A and 4B and S4C Fig ) . Notably , eliminating gpa-3 in the ADL neurons in an otherwise wild-type background did not alter food intake ( Fig 4C ) , reinforcing our observations that gpa-3-mediated regulation of body fat via the ADL neurons occurs independently of feeding . Together with the transgenic rescue experiments , we find that GPA-3 expression in ADL neurons is necessary and sufficient to maintain body fat stores . Our genetic epistasis experiments ( Fig 2A and 2B and S2A Fig ) suggested that gpa-3 negatively regulates acy-1 to control intracellular cAMP . We next wanted to determine the extent to which this signaling pathway functions in the ADL neurons . Accordingly , we inactivated acy-1 solely in the ADL neurons in the gpa-3 mutant background using antisense inhibition . Relative to non-transgenic controls , inactivation of acy-1 selectively in the ADL neurons led to a significant suppression of the decreased body fat of the gpa-3 mutants , resulting in body fat content similar to wild-type levels ( Fig 4D and S4B Fig ) . As seen with global acy-1 loss ( Fig 2A ) , inactivation of acy-1 specifically in the ADL neurons also did not appreciably alter fat stores ( Fig 4D and S4B Fig ) . Together , these experiments reveal a role for GPA-3 as a negative regulator of ACY-1 and intracellular cAMP in the ADL neurons , for the control of body fat stores . In C . elegans , the stimulatory Gαs that activates adenylyl cyclase to increase intracellular cAMP [42] is called GSA-1 . To test the prediction that enhanced cAMP production in ADL neurons decreases body fat , we selectively expressed gsa-1 ( R182C ) , a dominant , gain-of-function mutation of C . elegans Gαs [43] in the ADL neurons , which resulted in a near-complete loss of intestinal fat ( Fig 4E and S4C Fig ) . Thus , enhanced cAMP signaling in the ADL neurons stimulates fat loss in the intestine , and the cAMP second messenger in ADL neurons is instructive for the control of fat stores in the intestine . Information from the ADL neurons to the intestine could be relayed either directly via the release of a neuroendocrine factor , or indirectly via modifying the properties of other neurons . These possibilities can be distinguished in the following way: long-range neuropeptides and neuromodulators are localized to dense core vesicles , which require the conserved Calcium-dependent Activator Protein for Secretion ( CAPS , UNC-31 in C . elegans ) for fusion with the plasma membrane [44–46] . On the other hand , the canonical neurotransmitters ( acetylcholine , ACh; γ-amino butyric acid , GABA; and glutamate ) are localized to small clear synaptic vesicles , which require a protein called UNC-13 ( MUNC-13 in mammals ) for fusion at the synapse [47 , 48] . Thus , loss of UNC-31 function blocks the release of neuropeptides and biogenic amines from neurons [46] , and loss of UNC-13 function blocks release of the canonical neurotransmitters [48] . We generated gpa-3;unc-31 ( e928 ) and gpa-3;unc-13 ( n2813 ) mutants and measured the body fat of the respective single and double mutants . Interestingly , we found that loss of unc-13 resulted in complete suppression of the fat loss seen in the gpa-3 mutants ( Fig 5A and S5A Fig ) . This result suggested that rather than neuropeptides and biogenic amines , the canonical neurotransmitters acetylcholine , GABA or glutamate are required for the effects of GPA-3 signaling . To determine which of the three canonical neurotransmitter pathways are required downstream of GPA-3 , we first examined mutants of the presynaptic re-uptake transporters for GABA ( snf-11 ) , glutamate ( glt-4 ) and ACh ( cho-1 ) . Loss of the re-uptake transporters of the conventional neurotransmitters would disrupt their steady-state levels in the synaptic cleft , and thus indicate a potential role in regulating body fat stores . snf-1 ( ok156 ) and glt-4 ( bz69 ) mutants did not appreciably alter body fat stores , whereas cho-1 ( ok1069 ) mutants had approximately 40% of the body fat of wild-type animals ( Fig 5B and S5B Fig ) , resembling gpa-3 null mutants . ACh synthesis and breakdown occur via mechanisms distinct from the other neurotransmitters: after release into the synapse , unbound ACh is cleaved to form acetyl-CoA and choline by the enzyme acetylcholinesterase within the synaptic cleft itself . Choline is then taken up into the pre-synaptic neuron by the CHO-1 re-uptake transporter , and this step is a rate-limiting source of choline for presynaptic ACh synthesis . Thus , cho-1 mutants are defective in the re-uptake of synaptic choline and are deficient in ACh [49 , 50] . gpa-3;cho-1 mutants display similar fat content as either single mutant alone ( Fig 5C and S5C Fig ) . To determine the extent to which changes in ACh signaling regulate body fat stores , we examined the available mutants in the acetylcholinesterase genes , ace-1 , ace-2 and ace-3 , which have increased synaptic ACh [51–53] . Relative to wild-type animals , ace-1;ace-2 double mutants had a significant increase in body fat stores , whereas ace-3 mutants did not show an appreciable difference ( Fig 5D and S5D Fig ) . These results suggested that alterations in synaptic ACh result in changes in body fat stores . To determine the relationship between gpa-3 signaling and ACh , and to identify the key acetylcholinesterase responsible for the effects of ACh on body fat , we crossed gpa-3 mutants with the ace-1;ace-2 mutants to generate each mutant combination , as well as the ace-1 and ace-2 single mutants . We found that the gpa-3;ace-1 mutants fully suppressed the reduced body fat of gpa-3 single mutants ( Fig 5E and S5E Fig ) , whereas the gpa-3;ace-2 double mutants did not , and resembled the gpa-3 single mutants alone ( Fig 5E and S5E Fig ) . ace-1 mutants also suppressed the transcriptional induction of atgl-1 seen in gpa-3 mutants ( Fig 5F ) . Thus , ACE-1 is required downstream of GPA-3 in the regulation of body fat , suggesting that the GPA-3-mediated fat regulatory signal is transmitted from the ADL neurons via the cholinergic pathway . We measured food intake and locomotion of the mutants in the cholinergic pathway , with and without gpa-3 ( Fig 6 ) . As described in S3B Fig , gpa-3 mutants had an ~15–20% decrease in food intake ( Fig 6A and 6B ) . Cholinergic signaling has been known to alter rhythmic behaviors [54 , 55] , and as expected , cho-1 mutants also have a significant reduction in food intake , albeit to a lesser extent than the gpa-3 mutants themselves . gpa-3;cho-1 double mutants resemble the gpa-3 single mutants with respect to feeding deficits ( Fig 6A ) . We next examined the ace genes with and without gpa-3 with respect to food intake . ace-1 mutants have decreased food intake similar to the gpa-3 mutants , and gpa-3;ace-1 double mutants do not suppress the gpa-3 phenotype . Rather , they resemble either single mutant alone ( Fig 6B ) . This is in contrast to the suppression of GPA-3-mediated fat loss , as judged by fat levels as well as by measuring the induction of atgl-1 by GPA-3 ( Fig 5E and 5F ) . ace-2 mutants have a negligible effect on food intake , and gpa-3;ace-2 mutants resemble gpa-3 mutants alone . Taken together , the ace-1-mediated suppression of fat loss of the gpa-3 mutants is specific , and is not accompanied by suppression of food intake ( Fig 6B ) . Similar results were observed with locomotion ( Fig 6C and 6D ) ; additionally , gpa-3 mutants and ace-1 mutants do not show appreciable differences in locomotion . Thus , the fat phenotype of gpa-3 mutants occurs as a selective consequence of a shift towards fat mobilization . The ADL neurons mediate avoidance behavior from aversive stimuli [56–58] and are also shown to modulate social feeding behavior in response to high O2 levels [59] . These effects are mediated predominantly through the TRPV channel , OSM-9 [59 , 60] . We found that osm-9 mutants had wild-type body fat levels , and fully suppressed the reduced body fat of gpa-3 mutants ( Fig 7A and S6 Fig ) . These results suggested that an aversive function encoded by ADL neurons was related to the GPA-3-mediated fat phenotype . ADL neurons detect an ascaroside pheromone called ascr#3 ( also called C9 ) and initiate an aversive response in N2 wild-type animals that is abrogated in osm-9 mutants [61] . Pheromone signaling in C . elegans was originally shown to control developmental fate decisions [62–64] . However , a recent body of evidence has shown that a chemically-diverse family of ascaroside-based pheromones function individually and in combination , to elicit behaviors that collectively transmit population structure and population density information [20 , 27 , 65] . We wondered whether ascr#3 , the ascaroside detected by the ADL neurons would alter body fat stores . Administration of ascr#3 at a dose known to elicit Ca2+ transients in ADL neurons [66] led to a robust decrease in body fat stores ( Fig 7A and S6 Fig ) . gpa-3 and osm-9 single mutants , and gpa-3;osm-9 double mutants , did not display a further reduction in body fat upon ascr#3 administration , suggesting that activation of ADL neurons by ascr#3 decreases body fat stores via GPA-3-dependent signaling ( Fig 7A and S6 Fig ) . As expected , administration of ascr#3 also robustly induced atgl-1 expression in the intestine ( Fig 7B and 7C ) . We propose a model in which pheromone-mediated regulation of cAMP signaling in the ADL neurons controls acetylcholine release in to-be-defined cholinergic neurons , which in turn regulates a fat-stimulatory signal to control body fat stores via the rate-limiting ATGL-1 lipase in the intestine ( Fig 7D ) . Under normal conditions , in wild-type animals , population-density-dependent levels of the ascaroside ascr#3 regulates the extent to which GPA-3 in the ADL neurons inhibits the downstream adenylyl cyclase ACY-1 thus controlling cAMP levels in the ADL neurons . This , in turn , controls the level of acetylcholine released from small , clear vesicles in cholinergic neurons , and initiates a signal to the intestinal cells to regulate the activity of ATGL-1 ( Fig 7D , left panel ) . In gpa-3 mutants , irrespective of ascr#3 levels , the GPA-3 mediated inhibition of ACY-1 is lost , causing more cAMP to be produced in the ADL neurons . This , in turn , causes more acetylcholine to be released in downstream neurons , ultimately up-regulating ATGL-1 in the intestinal cells , and a constitutive loss of body fat due to increased fat utilization ( Fig 7D , right panel ) . Receptors for GPA-3 have been identified in the context of the dauer developmental decision [67] . Two G protein coupled receptors , srbc-64 and srbc-66 require GPA-3 signaling from the ASK neurons to mediate the dauer decision in response to the dauer pheromone and the ascaroside C6 . srbc-64 and -66 are not reported to be expressed in the ADL neurons , nor known to be responsive to ascr#3 , and therefore likely function via distinct mechanisms . Our previous work has suggested that food and oxygen are salient environmental cues that regulate body fat stores via modulation of neuronal circuit function [3 , 29] . Based on the studies presented here , we now propose pheromone sensing as a third sensory modality that regulates body fat stores . As an animal encounters a new patch of food , it must adjust its metabolism to reflect its environment . A patch of food that contains other worms must necessarily be shared , whereas a patch of food without worms reflects a relatively greater amount of food . We speculate that ascr#3/GPA-3 signaling from the ADL neurons provides C . elegans a mechanism to discriminate between these distinct environments , and accordingly modulate its metabolism . Our experiments provide the first insights into the molecular mechanisms by which pheromone sensing from the nervous system regulates peripheral lipid metabolism . In future studies , it will be interesting to determine the extent to which these discrete sensory inputs intersect to coordinate body fat metabolism .
C . elegans was cultured as described [68] . N2 Bristol , obtained from the Caenorhabditis Genetic Center ( CGC ) was used as the wild-type reference strain . The mutant and transgenic strains used are listed in S1 Table . Animals were synchronized for experiments by hypochlorite treatment , after which hatched L1 larvae were seeded on plates with the appropriate bacteria . All experiments were performed on day 1 adults . Promoters and genes were cloned using standard PCR techniques from N2 Bristol worm lysates or cDNA and cloned using Gateway Technology ( Life Technologies ) . Promoter lengths were determined based on functional rescue and are available upon request . All rescue plasmids were generated using polycistronic GFP . Transgenic rescue strains were constructed by microinjection into the C . elegans germline followed by visual selection of transgenic animals under fluorescence . For the microinjections , 5–10 ng/μl of the desired plasmid was injected with 25 ng/μl of an unc-122::GFP or myo-2::mCherry coinjection marker and 65–70 ng/μl of an empty vector to maintain a total injection mix concentration of 100 ng/μl . In each case , 10–20 stable transgenic lines were generated . Two lines were selected for experimentation based on consistency of expression and transmission rate . Triglycerides were extracted from wild-type and mutant C . elegans as described [3] . Extracted lipids were quantified by liquid chromatography/mass spectrometry on an HP 1100 MSDTM , using a neutral lipid Pheromex Luna C5 column , following the methodology from Nomura and colleagues [69] . Data were normalized to protein , quantified by the Pierce BCA Protein Assay kit . Oil Red O staining was performed as described [3] . For Oil Red O experiments in which animals were treated with a non-hydrolyzable cAMP analogue , animals were added to plates containing either M9 vehicle or 20 , 200 , or 500μM 8-Bromoadenosine 3′ , 5′-cyclic monophosphate ( Sigma Aldrich ) . For Oil Red O experiments in which animals were treated with ascaroside ascr#3 , animals were added to plates containing either ddH2O vehicle or 80nM ascr#3 . Within a single experiment , roughly 3500 animals were fixed and stained , 100 animals were visually inspected on slides , following which 15–20 animals were imaged for each genotype/condition . All experiments were repeated at least 3 times . Wild-type and gpa-3 mutants were included as controls for each experiment . Thrashing rate was measured as previously described [70] . For each animal , a movement where the head and/or tail swung to the other side was counted as one thrash . 15–20 animals were assessed for each phenotype . RNAi experiments were conducted as previously described [3] . Plates were seeded with HT115 bacteria containing vector or the relevant RNAi clone four days prior to seeding larvae . Black and white images of Oil Red O stained animals and fluorescent images were captured using a 10X objective on a Zeiss Axio Imager microscope . Lipid droplet staining in the first four pairs of intestinal cells was quantified as described [3] . We have found that quantification of the anterior intestine reliably captures fat content of the entire intestine . For all atgl-1::GFP images , an equal number of worms were chosen blindly and lined up side by side . Fluorescence intensity for all chosen worms was quantified for each condition . Images were quantified using ImageJ software ( NIH ) . Total RNA was extracted using TRIzol reagent ( Invitrogen ) . Genomic DNA was removed using an RNase-free DNase kit ( QIAGEN ) . cDNA was prepared using a iScript Reverse Transcription Supermix for RT-qPCR kit ( BioRad ) according to the manufacturer’s instructions . Quantitative PCR was performed using the SsoAdvanced Universal SYBR® Green Supermix according to the manufacturer’s instructions . Data were normalized to actin mRNA . Primer sequences are available upon request . Animals of mixed developmental stages were incubated in a 1:200 dilution of DiI stain ( Life Technologies ) for 3 hours on a rotating rack . After staining , the animals were seeded onto a plate containing an OP50 bacterial lawn and allowed to dry for approximately 30 minutes . Fluorescent images of animals in the L2-L3 larval stages were captured using a 100X objective on a Zeiss Axio Imager microscope . Food intake was measured by counting pharyngeal pumping , as previously described [71] . For each animal , the rhythmic contractions of the pharyngeal bulb were counted over a 10 s period under a Zeiss M2 Bio Discovery microscope . For each genotype , 10 animals were counted per condition and the experiment was repeated at least three times . Wild-type animals were included as controls for every experiment . Error bars represent SEM . Student’s t-test , one-way ANOVA , and two-way ANOVA were used as indicated in the figure legends .
|
The central nervous system plays a vital role in regulating whole body metabolism and energy balance . However , the precise cellular , genetic and molecular mechanisms underlying these effects remain a major unsolved mystery . C . elegans has emerged as a tractable and highly informative model to study the neurobiology of metabolism . Previously , we have identified instructive roles for serotonin signaling as a surrogate for food availability , as well as oxygen sensing , in the control of whole body metabolism . In our current study we have identified a role for a pair of pheromone-sensing neurons in regulating fat metabolism in C . elegans . cAMP acts as a second messenger in these neurons , and regulates body fat stores via acetylcholine signaling in the nervous system . We find that the population-density-sensing pheromone detected by these neurons regulates body fat stores . Together , we define a third sensory modality , population density sensing , as a major regulator of body fat metabolism .
|
[
"Abstract",
"Introduction",
"Results",
"and",
"discussion",
"Materials",
"and",
"methods"
] |
[
"biotechnology",
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"caenorhabditis",
"neuroscience",
"animals",
"animal",
"models",
"caenorhabditis",
"elegans",
"model",
"organisms",
"oils",
"experimental",
"organism",
"systems",
"genetic",
"engineering",
"digestive",
"system",
"research",
"and",
"analysis",
"methods",
"lipids",
"genetically",
"modified",
"organisms",
"animal",
"cells",
"fats",
"biological",
"tissue",
"genetically",
"modified",
"animals",
"gastrointestinal",
"tract",
"biochemistry",
"cellular",
"neuroscience",
"anatomy",
"adipose",
"tissue",
"cell",
"biology",
"neurons",
"nematoda",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"pheromones",
"organisms"
] |
2017
|
Pheromone-sensing neurons regulate peripheral lipid metabolism in Caenorhabditis elegans
|
Lymphedema is a debilitating and disfiguring sequela of an overwhelmed lymphatic system . The most common causes of secondary lymphedema are lymphatic filariasis ( LF ) , a vector-borne , parasitic disease endemic in 73 tropical countries , and treatment for cancer in developed countries . Lymphedema is incurable and requires life-long care so identification of effective lymphedema management is imperative to improve quality of life , reduce the burden on family resources and benefit the local community . This review was conducted to evaluate the evidence for effective lymphedema self-care strategies that might be applicable to management of all types of secondary lymphedema . Searches were conducted in Medline , CINAHL and Scopus databases in March 2015 . Included studies reported before and after measures of lymphedema status or frequency of acute infections . The methodological quality was assessed using the appropriate Critical Appraisal Skills Program checklist . Descriptive synthesis and meta-analysis were used to evaluate effectiveness of the outcomes reported . Twenty-eight papers were included; two RCTs were found to have strong methodology , and overall 57% of studies were rated as methodologically weak . Evidence from filariasis-related lymphedema ( FR-LE ) studies indicated that hygiene-centred self-care reduced the frequency and duration of acute episodes by 54% , and in cancer-related lymphedema ( CR-LE ) home-based exercise including deep breathing delivered significant volume reductions over standard self-care alone . Intensity of training in self-care practices and frequency of monitoring improved outcomes . Cultural and economic factors and access to health care services influenced the type of intervention delivered and how outcomes were measured . There is evidence to support the adoption of remedial exercises in the management of FR-LE and for a greater emphasis on self-treatment practices for people with CR-LE . Empowerment of people with lymphedema to care for themselves with access to supportive professional assistance has the capacity to optimise self-management practices and improve outcomes from limited health resources .
Lymphedema is a high protein edema which forms when the lymphatic system is chronically overwhelmed . Earlier fluid rich stages progress gradually toward enlargement and fibrosis of the subcutaneous compartment and hyperkeratosis of the skin ( elephantiasis ) [1] . This can occur as a result of congenital factors ( primary lymphedema ) but is more commonly caused by an alteration in normal lymphatic function leading to secondary lymphedema . The majority of secondary lymphedema occurs through infection with a vector-borne , parasitic disease known as lymphatic filariasis ( LF ) which is endemic in 73 tropical countries where is it closely associated with poverty [2] . In developed countries lymphedema is more commonly a consequence of some cancer treatments which involve lymph node removal or irradiation . Global estimates of filariasis-related lymphedema ( FR-LE ) are 16 . 7 million cases [3] and cancer-related lymphedema ( CR-LE ) is estimated to affect between 15% and 80% of all cancer survivors [4] . Annual mass drug administration ( MDA ) of anti-filarial chemotherapy can prevent future transmission of LF [5] and improvements in surgical management of cancer should reduce the incidence of new CR-LE cases [6] but in both aetiologies onset of chronic symptoms may be delayed for months , years or even decades after exposure to the risk [4 , 7] . People with any impairment to lymphatic function bear a lifelong risk of developing secondary lymphedema [7] . Lymphatic vessels remove circulating fluid and large molecules from the extracellular spaces of almost all body tissues and transport them to the lymph nodes . This is essential for continuous clearance of pathogenic elements crossing the skin barrier and entering the subcutaneous compartment and in other tissues it is vital in maintaining correct extracellular fluid balance . Cleaned and filtered lymph is returned to systemic circulation via the vascular system . In lymphedema , when normal lymph transport is impeded , protein rich fluid accumulates , mostly in the subcutaneous compartment . Risk of infection is increased; namely acute dermato-lymphangio-adenitis ( ADLA ) in FR-LE , and cellulitis or erysipelas in CR-LE . Infection then exacerbates disease progression and as lymphedema advances , symptoms become increasingly disabling and disfiguring [8] . In areas endemic for LF , lymphedema causes social stigma , superstition and loss of opportunity to marry [9] . People with CR-LE report depression , poor quality of life and an inability to engage in paid employment [10] . Although essentially the same chronic disease , treatments for FR-LE and CR-LE follow different guidelines . The World Health Organization recommends community based home care ( CBHC ) [5 , 11–13] to improve hygiene and reduce ADLA episodes ( the main cause of lost working days ) in FR-LE . The program promotes frequent washing and drying of affected areas with particular attention to entry lesions ( potential sites of fungal and bacterial infection ) , passive elevation and range of motion ( ROM ) exercises , and the use of oral antibiotic or anti-inflammatory medications during acute events . Self-massage and compression bandaging are recommended in advanced stages but usually not implemented in resource-poor settings [14] . In contrast the gold standard for CR-LE is a two phase program with an initial , intensive period of therapist based treatment applying specialized lymphatic massage and multilayer compression bandaging to reduce limb size; followed by an ongoing maintenance phase of self ( or partner ) lymphatic massage with regular use of compression garments [15] . Meticulous skin care and remedial exercises are a component of both phases . Established lymphedema is considered to be irreversible , necessitating lifelong care , and family and psychosocial support [1] . Effective management of either FR-LE or CR-LE can improve quality of life for the individual , reduce the burden on family resources and benefit the local community . In developing countries there may be few or no health care services available for people with FR-LE and resources allocated to CR-LE treatment in many developed countries are also insufficient [16 , 17] . Poor access to lymphedema health services in both settings and limited evidence of treatment efficacy has resulted in misinformation and unproven management practices . Approaches to management will ultimately be shaped by access to resources and cultural , financial and political influences [17] , but effective core strategies that can be applied across cultural and economic borders need to be identified . This review was undertaken to evaluate the outcome of current self-care interventions for FR-LE and CR-LE . Differences and similarities were assessed with respect to self-care components included in the intervention , outcome measures used and the extent of any support services and monitoring . The results have illuminated beneficial practices that may inform health systems in any setting to increase the effectiveness of self-care strategies for people with secondary lymphedema .
Searches were conducted in Medline ( OVID ) , CINAHL and Scopus databases in October 2013 and updated during March 2015 . Medical Subject Headings ( MeSH ) were used in the Medline search and keywords for the Scopus and CINAHL searches were derived from the Medline MeSH terms to ensure consistency of search terms ( Table 1 ) . The search strategy was limited to publications in English and grey literature was not searched . Reference lists of studies and reviews , World Health Organization ( WHO ) summary reports and editorials were searched to find other original peer reviewed studies . After removal of duplicates the title and abstract of returned studies were screened by two authors ( JD , SG ) . Full inclusion and exclusion criteria used are detailed in S1 Tables . Two authors ( JD , SG ) independently appraised the included studies for methodological quality using the Critical Appraisals Skill Program ( CASP ) RCT appraisal checklist or cohort appraisal checklist appropriate to the study design [21] . A rating of strong , moderate or weak was awarded to each study according to the number of ‘yes’ , ‘can’t tell’ and ‘no’ responses . Studies without ‘no’ or with two or less ‘can’t tell’ answers were considered methodologically strong . Studies with one ‘no’ or three ‘can’t tell’ answers were rated moderate and studies with two or more ‘no’ answers rated weak [22] . When the independent appraisal process was completed , studies which did not achieve the same rating by both reviewers were discussed until agreement was reached . A third reviewer ( PG ) was available where discrepancies could not be resolved but was not required . One author ( JD ) independently extracted study data including number of participants , intervention , outcomes , outcome measures , and results . Special characteristics of gender , age , country/setting , lymphedema cause , affected site and stage ( or grade ) of lymphedema were noted . Data were extracted for change in lymphedema status and effect on acute episodes . Where results were reported consistently with appropriate estimate of precision , a meta-analysis was performed using Review Manager version 5 . 3 ( The Nordic Cochrane Centre , 2014 ) . Data on effect size and outcomes of clinical importance were extracted for narrative synthesis and to highlight differences between the FR-LE and CR-LE studies .
Ten RCT’s including one quasi RCT and one randomized cross over trial were included . Of these , three FR-LE trials [23–26] and two CR-LE trials [27 , 28] compared two or more groups performing different self-care protocols . The remaining five RCTs included only one group using self-care alone and these results were considered as uncontrolled , prospective cohorts [29–33] . Most cohort studies were prospective: ten on FR-LE , including two reports on the same cohort [34 , 35] , and four on CR-LE . There was one retrospective follow up on a previous FR-LE RCT and two retrospective reports on CR-LE . There were eighteen reports on seventeen studies about FR-LE and all were conducted in tropical countries endemic for Brugia malayi or Wuchereria bancrofti . All participants had leg lymphedema , three included people with arm lymphedema [14 , 25 , 40] and two did not specify the affected limb [33 , 41] . Both genders were represented with a greater proportion of females ( 42 . 5% - 87% ) and although children were admitted in 14 studies most participants were adults with a mean or median age between 35 and 57 years ( range 10 to 98 years ) . Sample sizes ranged between 14 and 1578 with 14 studies of more than 90 participants . In contrast to FR-LE , most participants in the ten CR-LE studies were women with arm lymphedema after breast cancer and only one RCT included males and participants with leg lymphedema [27] . Other than one Indian cohort [42] all were conducted in developed countries; sample sizes were smaller and four studies had less than 30 participants ( range 18–138 ) . Compared to FR-LE studies , the mean or median age of participants in CR-LE studies was higher , between 47 and 66 years ( range 25–87 years ) and none included adolescents . Population characteristics for all studies are provided in S2 Tables . Staging ( or grade ) of lymphedema was based on clinical assessment of limb size and skin changes . FR-LE studies used either the WHO criteria of Grades 1–3 [43] , WHO criteria of Grades I–IV [44] or a seven stage criteria developed by Dreyer , Addiss [11] . CR-LE studies did not state the staging criteria used . Two CR-LE [29 , 42] and three FR-LE studies [32 , 36 , 39] excluded participants with later stages of disease and participants with lymphedema stage 0 were included in two FR-LE studies [30 , 36] . The water displacement method [45] , which is considered the international gold standard [1] , was used to calculate limb volume in both CR-LE and FR-LE studies . Only studies on CR-LE used electronic measuring devices , specifically; multi-frequency bio-impedance spectroscopy which measures extracellular fluid loads [46] and perometry which calculates limb volume using a truncated cone formula from circumferences at 3 millimetre intervals [47] . Limb circumference by tape measure was included in most studies and reported as limb volume , calculated using a truncated cone formula at four or five centimetre intervals , as raw circumference values at fixed points or as a combined average of all points . Studies on unilateral lymphedemas frequently reported on relative limb volume ( difference between the affected and unaffected limbs ) and percentage change in relative limb volume ( RLV ) over time . Methods used to assess limb status in each study are provided in S2 Tables . Seven studies ( n = 1073 ) reported on change in either objective assessment of lymphedema stage or limb volume , or participant perception of limb status . WHO staging criteria was used more frequently ( four studies ) than the seven stage Dreyer system ( two studies ) . Limb volume was quantified by water displacement ( three studies ) or limb circumference ( three studies ) . Two studies reported on this outcome using either the Dermatology Quality of Life Index ( DQLI ) [56] or the WHO Disability Assessment Schedule II ( WHO DAS II ) [57] . Both studies used the seven stage criteria for FR-LE and reported significant improvement in either; all stages after 12 months ( n = 14 ) [41] or in stages 3–7 after 24 months ( n = 370 ) [34] . Results for changes in perceived disability and quality of life are provided in S3 Tables . No study assessed the effect of self-care alone on CR-LE . Five studies ( n = 460 ) used medicated creams or soap in the daily self-care protocol . Three RCTs were used to compare either medicated soap [23] or medicated cream [24 , 25] to plain soap or plain cream and medicated ointment was used daily by one self-care group in two other trials [30] [33] . All interventions were of 12 months duration and three studies followed participants for a further year after the intervention ceased [24 , 25 , 33] . Neither of the RCTs which compared medicated cream to plain cream found any between group differences . After the 12-month intervention a significant reduction in annual ADLA episodes of between 63 . 83% [25] and 77 . 57% [24] was recorded by all groups and at 24 months the mean annual incidence was still significantly lower than at baseline by 59 . 52% and 65 . 02% respectively . During the follow up year , annual incidence in the groups who had used plain cream during the intervention continued to reduce , whereas both groups who had used antibiotic cream experienced an increase ( not significant ) . The trial which compared antibiotic soap to plain soap also found no difference between groups and reported results as for a single cohort [23] . This cohort and the medicated cream groups in two trials ( n = 340 ) all reported significant reductions in mean ADLA episodes of between 62 . 5% and 65 . 6% after 12 months [23 , 30 , 33] . One group was followed for 12 months after the intervention and an overall reduction of 73% from baseline was recorded [33] . Results of the effect of medicated cream or soap on ADLA are shown in S3 Tables . Two trials reported on this outcome ( n = 120 ) . In a 12-month intervention on unilateral leg lymphedema [30] , raw circumference values for the affected limb reduced between 27 . 6% - 92% with the greatest reduction at the calf of participants with Grade 2 FR-LE ( WHO grades 1–3 [43] ) and the least reduction at the ankle in participants with Grade 3 ( Fig 3 ) . In this trial the difference in circumference between affected and unaffected limbs also reduced significantly at all time points . In the trial which used water displacement to measure limb volume at baseline and then again during ADLA [24] , limb volume increased during 80% of episodes and remained elevated in 73% of cases after two weeks . No CR-LE studies investigated the addition of medicated creams or soaps . No studies assessed the effect of prescribed exercises on FR-LE . Seven studies assessed exercise interventions of between eight weeks and six months duration . All participants were women with unilateral arm lymphedema after breast cancer ( n = 197 ) and six studies reported significant benefits in at least one outcome . Seven studies ( n = 295 ) reported improvement in one or more of these outcomes . Only one FR-LE study emphasized the use of SLD and instructed participants in daily self-bandaging . After three weeks of self-bandaging they were fitted with compression garments which were then worn daily for the remainder of the study [37] . Although the self-care protocol in the study by Wijesinghe , Wickremasinghe [14] recommended compression it was not routinely used and results of that study were reported in the basic self-care section . One study on both arm and leg lymphedema [27] and two studies on arm lymphedema [29 , 54] included SLD in the self-care program ( n = 121 ) . The RCT with the mixed arm and leg population investigated the benefits of using Aromatherapy ( essential oils ) in the massage medium compared to plain cream but found no significant difference between groups and reported most results as a single cohort using SLD [27] . No CR-LE studies included self-bandaging but regular use of a compression garment was recommended in all studies . New garments were supplied at commencement in four studies [28 , 29] of which two [51 , 52] allowed a control period before the intervention to adjust for the effect of the garment . The Measure Yourself Outcome Profile 2 [59] was used to assess quality of life in participants performing SLD with or without Aromatherapy and both groups reported significant improvements at all time points up to 6 months [27] .
In reviewing the evidence for self-care in FR-LE and CR-LE , marked differences were apparent both between and within settings and some key opportunities for improvements were identified . Evidence from the FR-LE population showed that basic self-care alone is effective in preventing ADLA which is consistent with results of a recent review of the effect of hygiene based interventions on FR-LE [60] . Hygiene alone may halt disease progression but is less likely to reduce limb volume and evidence from the CR-LE population indicated that greater volume reductions are achieved when activities such as progressive resistance exercise are included in the self-care routine . Whilst basic self-care was the primary intervention in almost 60% of studies on FR-LE , no CR-LE studies assessed basic self-care alone; rather the self-care group when included were always as controls . Best practice guidelines in CR-LE management are still dependent on therapist performed interventions; however evidence from the FR-LE population suggests that more effort to involve CR-LE patients in their own self-treatment may relieve the financial burden of therapist based care in this population . The exclusion of any group in comparative studies that received drug or therapist based interventions meant that although ten RCTs were reviewed the bulk of evidence came from observations of a single cohort in studies rated of moderate or weak methodological quality where only one or two groups were performing self-care . There was also inconsistency between assessment techniques and reporting methods . Therefore a review which includes all interventions for FR-LE and CR-LE may provide data for more rigorous meta-analysis . None the less , this first review to systematically examine the similarities and differences in self-care for CR-LE and FR-LE has opened a pathway for further investigation of transferrable strategies for lymphedema management in disparate settings . The available resources in each study setting were reflected in the simplicity or complexity of devices used to measure change in limb status . All CR-LE studies used water displacement , bio impedance spectroscopy , perometry or a truncated cone formula at small intervals to quantify limb volumes . These methods can detect very small changes which , although statistically significant , might be of minimal clinical significance . In contrast most FR-LE studies relied on less precise measures and less than one quarter used either water displacement or a truncated cone formula . A further 18% of studies used three or four fixed circumference points to compare affected and unaffected limbs or summed or averaged these measures . These methods , especially summed or averaged circumferences lack the precision to detect small changes in limb volume and this could account for variations in reported outcomes between the CR-LE and FR-LE groups . More frequently , FR-LE studies relied on assessment of lymphedema by stage and the use of criteria with only three or four groups was common , since even studies which used the seven stage criteria often grouped them into early , moderate or late stage disease . These graduations may lack the precision to detect small changes and participants who changed from a higher to a lower stage or vice versa may not have always been detected , thereby under or over estimating the effectiveness of the intervention . Most studies tried to minimise inter- or intra-observer variation in staging but the subjective nature of these assessments may also have contributed to the disparate results . These limitations may explain why some studies reported reduction in limb volume without a corresponding change in lymphedema grade . Overall studies which used more precise measuring protocols more often reported significant evidence for volume reduction in both settings . Less than 12% of FR-LE studies investigated the effect of self-care on subjective symptoms or functional deficits whereas this was reported in 70% of CR-LE studies . The evidence for improvement was weak , mainly due to the disparate range of measuring tools employed , but the overall trend was that a reduction in limb volume was accompanied by improvements in symptoms , perceived disability , overall wellbeing and quality of life . This suggests that self-reported parameters could be used as a proxy for objective measures when these are not available and provide valuable information about the lived experience of FR-LE . Oral or topical medications for ADLA were used in almost all FR-LE studies but the influence of these could not be adequately separated from the effect of other components of self-care and it is unclear what bearing this had on the results . Although groups treated specifically with oral antibiotic or deworming medications were excluded from the data synthesis some reports showed that the placebo drug groups performing basic self-care had better long term results than groups that had initially received oral medications [24 , 25 , 33] . Studies on topical medications for ADLA showed no additional benefit over self-care using placebo creams or soaps and the necessity of prophylactic oral medication for ADLA remains controversial . Similarly , in CR-LE studies the use of a compression sleeve was considered an integral aspect of self-care but few studies controlled for the effect of a new garment or included frequency of garment use in the statistical analysis . Where this was done the effect of the compression sleeve was shown to be significantly correlated with improvement in all groups . Whilst these treatments might be considered to be core elements of self-care , access to medication in poorly resourced settings and non-adherence to compression therapies in CR-LE warrant the investigation of self-care protocols that do not rely on these components . Effective self-care implementation requires some degree of education , instruction or demonstration and the role of the educated health worker or trained volunteer cannot be ignored . FR-LE studies which provided frequent monitoring and support were associated with greater improvements than studies which offered minimal or no support services . The study by Suma , Shenoy [55] which retrospectively reviewed participants in a previous drug based RCT indicated that without monitoring , program effectiveness is lost over time , an effect also found in a later follow up of the study by Addiss , Louis-Charles [36] which could not be included in this review [61] . Ultimately , the long term success of any self-care intervention will depend on individual ownership of and adherence to the daily self-care practices and the level of family or local support available . This was demonstrated clearly in the study by Akogun and Badaki [50] where one group was able to alter the program design to suit their immediate cultural and social constraints and reported good outcomes , whereas the two groups who could not alter the program design to suit their personal circumstances experienced a large loss to follow up . Despite the wide variation in measurement techniques and support services , it was apparent that hygiene-centred , basic self-care can reduce the frequency and duration of ADLA episodes by approximately 50% , and this was particularly beneficial for people with later stages of disease . There was less evidence for a reduction in limb volume but the study by Joseph , Mony [24] showed that ADLA led to an increased limb volume which persisted after the infection had been treated and Mues , Deming [35] demonstrated that ADLA is related to days of work lost . Thus a reduction in ADLA episodes without change in limb size may still improve outcomes for many individuals . This was illustrated histologically in the study by Wilson , Guarner [39] which showed that basic self-care improved skin integrity and prevented new infections while limb stage remained the same . Where a reduction in limb volume was reported in FR-LE , greater benefits were experienced among participants with early stages , suggesting that implementation of a self-care routine as soon as lymphedema is detected has the potential to curtail the number of cases that progress to advanced stages . Current guidelines for FR-LE will not assist program managers to find and address these earliest stages of lymphedema yet this may be the optimal time to intervene in terms of volume reduction and long term prevention of ADLA . Basic hygiene is enough to control ADLA but this review has shown that limb volume is more difficult to reduce in later stages regardless of the setting [27 , 48] . Advanced lymphedema is characterised by fibrosis and fatty induration of the tissues which become much more difficult to reduce than in early stages where the swelling is more characteristically due to protein rich fluid . So to reduce limb volume more intervention is required at an earlier stage than is currently indicated in the WHO guidelines . Reduction in limb volume was reported in all CR-LE studies , all of which included at least one additional component of self-care . Since publication of the 2005 study by Moseley , Piller [53] , specific deep breathing exercises appear frequently in interventions for CR-LE as was evidenced by their inclusion in six studies in this review . Home based exercise , including deep breathing , is easy to perform , require no financial resources , can be continued alone after minimal initial instruction and may contribute to overall improvement in health and wellbeing , but exercise advice included in the current protocols for FR-LE is very limited . Simple resistance exercise and deep breathing could be easily incorporated into CBHC particularly in cultures where activities such as Yoga or Tai Chi may be readily available and acceptable and the addition of such components to current WHO recommendations warrants investigation . In FR-LE , issues pertaining to infection , wet environments and lack of foot wear make compression therapies problematic . However , Bernhard , Bernhard [37] and other studies that could not be included in this review [62] have showed that people with FR-LE are capable of performing complex compression bandaging and daily use of compression garments . This requires a greater initial investment in the training and education of people with FR-LE so that they have a better understanding of the purpose and effect of each treatment component . The long term benefit of this increased investment was demonstrated in the study by Bernhard , Bernhard [37] where limb volume reduction was significantly greater in the self-treating group compared to the therapist treated group ( results for the therapist treated group were not included in this review ) . This supports consideration of compression therapies in FR-LE when possible and although the evidence is limited , also suggests that self-bandaging could be further explored for people with CR-LE . Although people with arm lymphedema were included in several FR-LE studies , results were not reported by limb . Only the CR-LE study by Barclay , Vestey [27] reported results by affected limb and in this study the impact of SLD on volume reduction was much greater for arms than legs . This limited comparison of results by limb implies that interventions may deliver different results depending on the location of the lymphedema . Similarly , some studies allowed participation of children as young as five years old but no study gave an analysis by age to determine if children had better or worse outcomes than the adult subjects , nor was the effect of gender explored . It is possible that age and gender influence self-treatment outcomes and investigation of different components of self-care by limb , age and gender should be considered . Since disability from existing LF will continue to increase for several decades even after transmission has been successfully interrupted , and increasing cancer survivorship is a primary focus of cancer research , it is probable that the incidence of new lymphedema cases from both causes will continue to increase for the foreseeable future . Focussing efforts toward greater emphasis on early intervention and prevention has the potential to alleviate this future burden . Implementation of morbidity management in the global effort to eliminate LF requires evidence based strategies to attract and maintain funding , and reducing the burden of CR-LE for all cancer survivors requires more research about lymphedema of the leg . In both cases high quality studies that investigate reversal of early stage disease and analysis of individual components of self-care by age , gender , stage and location of lymphedema are essential to determining optimal , financially sustainable , management .
|
Secondary lymphedema is a major cause of disability worldwide . The most common causes are treatment for cancer or infection with lymphatic filariasis . In both cases , the lymphatic system is damaged and unable to perform its normal function of removing extracellular fluid and wastes . Protein rich fluid accumulates in the affected area , and if left untreated may progress to ‘elephantiasis’ , which is characterised by a grossly enlarged limb and thickening of the skin . Lymphedema is incurable , requires lifelong care , and psychosocial support , but early intervention and good self-management practices can halt progression , preserve quality of life and maintain the ability to participate in normal work and social activities . Current approaches to treatment vary depending on the setting . In developed countries , cancer related lymphoedema is therapist-based and aims to intervene early and prevent disease progression . Lymphatic filariasis related lymphedema is associated with poverty , affecting people living in developing countries where minimal intervention is recommended or available for early stages . By identifying useful practices that can be transferred across cultural and economic borders , people living with lymphedema can be empowered to care for themselves and improve their long term outcomes .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"medicine",
"and",
"health",
"sciences",
"legs",
"pathology",
"and",
"laboratory",
"medicine",
"chemical",
"compounds",
"sports",
"and",
"exercise",
"medicine",
"salts",
"limbs",
"(anatomy)",
"database",
"searching",
"physical",
"activity",
"health",
"care",
"signs",
"and",
"symptoms",
"edema",
"research",
"and",
"analysis",
"methods",
"public",
"and",
"occupational",
"health",
"exercise",
"soaps",
"musculoskeletal",
"system",
"chemistry",
"lymphedema",
"arms",
"physical",
"fitness",
"diagnostic",
"medicine",
"quality",
"of",
"life",
"anatomy",
"database",
"and",
"informatics",
"methods",
"biology",
"and",
"life",
"sciences",
"sports",
"science",
"physical",
"sciences"
] |
2016
|
Self-Care for Management of Secondary Lymphedema: A Systematic Review
|
Individual rapid tests for serodiagnosis ( RDT ) of human African trypanosomiasis ( HAT ) are particularly suited for passive screening and surveillance . However , so far , no large scale evaluation of RDTs has been performed for diagnosis of Trypanosoma brucei gambiense HAT in West Africa . The objective of this study was to assess the diagnostic accuracy of 2 commercial HAT-RDTs on stored plasma samples from West Africa . SD Bioline HAT and HAT Sero-K-Set were performed on 722 plasma samples originating from Guinea and Côte d’Ivoire , including 231 parasitologically confirmed HAT patients , 257 healthy controls , and 234 unconfirmed individuals whose blood tested antibody positive in the card agglutination test but negative by parasitological tests . Immune trypanolysis was performed as a reference test for trypanosome specific antibody presence . Sensitivities in HAT patients were respectively 99 . 6% for SD Bioline HAT , and 99 . 1% for HAT Sero-K-Set , specificities in healthy controls were respectively 87 . 9% and 88 . 3% . Considering combined positivity in both RDTs , increased the specificity significantly ( p≤0 . 0003 ) to 93 . 4% , while 98 . 7% sensitivity was maintained . Specificities in controls were 98 . 7–99 . 6% for the combination of one or two RDTs with trypanolysis , maintaining a sensitivity of at least 98 . 1% . The observed specificity of the single RDTs was relatively low . Serial application of SD Bioline HAT and HAT Sero-K-Set might offer superior specificity compared to a single RDT , maintaining high sensitivity . The combination of one or two RDTs with trypanolysis seems promising for HAT surveillance .
Human African trypanosomiasis ( HAT ) or sleeping sickness is a fatal parasitic infection affecting rural populations in sub-Saharan Africa . During the last decade , active case finding by specialized mobile teams has considerably contributed to the reduction of the prevalence of HAT caused by Trypanosoma brucei ( T . b . ) gambiense . Since 2009 , the number of cases reported annually has dropped below ten thousand . At low prevalence , cost-effectiveness of active screening decreases and passive case finding becomes increasingly important [1] . This shift from the mobile team to the fixed health system for HAT detection requires an adapted diagnostic approach . Detection of trypanosome specific antibodies in blood with the card agglutination test for trypanosomiasis ( CATT ) , [2] is routinely applied for large scale active population screening . CATT is however ill-adapted to the conditions encountered in health-care centers . The limited shelf-life of the reconstituted CATT reagent at ambient temperature leads to considerable reagent loss when only few tests are performed per day . Another limitation of the CATT is the need of an agitator and a cold chain and therefore electric power , which are not always available in rural health-care centers . The venue of individual rapid tests for serodiagnosis of HAT that are stable at ambient temperature and can be performed without additional material [3–5] , is a key event in the development of an effective passive screening and HAT surveillance system [6] . Two rapid diagnostic tests ( RDT ) have been evaluated in phase 2 diagnostic trials [5 , 7] , show sufficient diagnostic accuracy and have been commercialized . So far , all RDT diagnostic evaluations have been performed on samples originating from Central-Africa , and no large scale evaluation has been performed for diagnosis of gambiense HAT in West Africa , while geographic variation in the accuracy of HAT serodiagnostic tests may occur [8] . The objective of this study was therefore to assess the diagnostic accuracy of 2 RDTs on stored plasma samples collected from HAT cases , negative controls , and serological suspects originating from Guinea and Côte d’Ivoire , two countries where HAT transmission is still active [9 , 10] .
Samples were collected during medical surveys conducted by the national HAT control programs . All participants were informed about the study objectives in their own language and gave written informed consent . Children less than 12 years old were excluded . For participants between 12 and 18 years old , informed consent was obtained from the parents . Approval for this study was obtained from the consultative committee for deontology and ethics ( Comité Consultatif de Déontologie et d’Ethique ) of the Institut de Recherche pour le Développement . In Côte d’Ivoire , the protocol was approved by the national ethical committee ( N°0308/MSLS/CNER-P ) . Plasma samples originated from subjects identified during active screening campaigns in the Dubreka , Boffa and Forecariah coastal mangrove HAT foci , situated north of Conakry in the Republic of Guinea and in the HAT foci of Oumé , Bouaflé , Sinfra , and Bonon in western central Côte d’Ivoire . All subjects underwent CATT/T . b . gambiense performed on whole blood ( CATT-WB ) . Blood was collected in heparinised tubes and for CATT WB-positive persons , the plasma end titre was determined . All CATT-pl ≥1/4 positive persons underwent parasitological examination by direct microscopic examination of the lymph node aspirate if swollen lymph nodes were present and/or mini-anion exchange centrifugation technique on buffy coat ( mAECT-BC ) [11] . Based on the CATT and parasitological result , four categories of study participants ( n = 722 ) were defined: 1° HAT: Parasitologically confirmed HAT patients with positive CATT-WB and CATT-pl end titer ≥1/4 ( n = 229 from Guinea , n = 2 from Côte d’Ivoire ) ; 2°Control: CATT-WB negative individuals for whom there was no suspicion for sleeping sickness infection ( n = 101 from Guinea and n = 156 from Côte d’Ivoire ) ; 3° SERO: Individuals with positive CATT-WB and CATT-pl end titer ≥1/4 ( Seropositives ) but no parasites detected ( n = 123 from Guinea , n = 42 from Côte d’Ivoire ) ; 4° SUSP: Individuals with positive CATT-WB but CATT-pl <1/4 ( Suspects ) in whom parasitological examinations were not performed ( n = 69 from Côte d’Ivoire ) . Samples were retrospectively tested in two commercial RDTs for serodiagnosis of gambiense HAT: SD Bioline HAT ( SD Diagnostics , Korea ) and HAT Sero-K-Set ( Coris BioConcept , Belgium ) . Both tests use purified native variant surface glycoproteins of T . b . gambiense variable antigen types LiTat 1 . 3 and 1 . 5 as antigens: SD Bioline HAT in two separate test lines ( line 1 and 2 respectively ) , HAT Sero-K-Set in a single test line consisting of a mix of both glycoproteins . The methodology applied was previously described for evaluation of RDTs for malaria diagnosis [12] . Tests were performed according to the indications of the manufacturers . Briefly , for SD Bioline HAT , 10 μls of test plasma were applied in the sample well , followed by 4 drops of assay diluent . For HAT Sero-K-Set , 15 μls of plasma were applied in the sample well , followed by 2 drops of BL-A buffer , after which the test device was re-inserted into its pouch . Tests were performed in batches of 10 . Reading was done in day light , 15 minutes after application of the buffer . In case the control line did not appear , the test result was considered invalid . A scoring system was used for estimating the individual test line intensity: negative ( no visible test line ) , faint ( barely visible test line ) , weak ( test line weaker than the control line ) , medium ( test line equivalent to the control line ) or strong ( test line more intense than the control line ) [13] . Reading was performed by 3 independent readers that were blind to other results ( 2 experienced and 1 less experienced that had been trained ) . The consensus test line intensity was based on consensus between two readers . In absence of consensus ( 3 different scores ) , the median score was taken . The test line was interpreted positive if the consensus test line intensity was faint or stronger . The HAT Sero-K-Set was positive if the test line was positive , the SD Bioline HAT was considered as positive if at least 1 test line was read as positive . For immune trypanolysis [14 , 15] , 25 μl of plasma were mixed with 25 μl of guinea pig serum and incubated for 30 minutes at room temperature . Blood of mice infected with T . b . gambiense was diluted in guinea pig serum to a final concentration of 107 trypanosomes/ml . 50 μl of this trypanosome suspension were added . After 90 min of incubation at room temperature , the suspension was examined microscopically at 400x magnification . Trypanolysis was considered positive when 50–100% of the trypanosomes were lysed , otherwise it was considered negative . Two trypanolysis series were run , one with T . b . gambiense variable antigen type LiTat 1 . 3 and one with LiTat 1 . 5 . A sample was considered positive in trypanolysis if it was positive with at least 1 variable antigen type . Diagnostic sensitivity and specificity with binomial exact 95% confidence intervals ( CI ) were calculated for the results obtained in respectively the HAT and control group . Specificities and sensitivities were compared using the McNemar chi-square test . Differences between independent groups were assessed using a Chi squared test . Taking into account that the SERO and SUSP group are heterogeneous and might contain individuals that ( i ) are or have been in contact with T . b . gambiense but did not have detectable parasitemia , or ( ii ) are CATT false positives [16] , immune trypanolysis was used as a reference test for presence of T . b . gambiense specific antibodies [15] .
For both RDTs , not a single invalid RDT result was observed . The line intensities scored by the 3 readers as well as the consensus intensity are shown in Table 1 . In HAT Sero-K-Set , the consensus test line intensity was negative for 370 persons , and faint to strong for 352 persons whom were considered positive . Absence of a consensus intensity or differences between individual scores larger than one grade occurred in 1 . 4% of readings ( 10/722 ) . The kappa values for a positive or negative test result , when comparing each of the readers , were between 94 . 2 and 95 . 8% . The consensus test line intensity in SD Bioline HAT line 1 and 2 were respectively 362 and 363 times negative and 360 and 359 times positive . In respectively 3 . 8% ( 28/722 ) and 1 . 7% ( 12/722 ) of readings of line 1 and 2 , at least one reader scored more than 1 grade different than another reader . At least 1 of both test lines scored positive in SD Bioline HAT for 396 persons . Kappas between readers were 91 . 7–95 . 0% for line 1 , and 92 . 8–94 . 5% for line 2 . The number and proportion of positive test results by study participant category are summarized in Table 2 . Sensitivities observed in HAT patients were respectively 99 . 6% ( CI 97 . 6–100 ) for SD Bioline HAT , and 99 . 1% ( CI 96 . 9–99 . 9 ) for HAT Sero-K-Set . There was no difference in sensitivity ( p = 0 . 6 ) between the 2 RDTs . Specificities in healthy controls were respectively 87 . 9% ( CI 83 . 3–91 . 7 ) for SD Bioline HAT and 88 . 3% ( CI 83 . 8–92 . 0 ) for HAT Sero-K-Set . There was no difference in specificity between the lines 1 and 2 in SD Bioline HAT ( p = 0 . 2 ) , nor was there any difference in specificity between the 2 RDTs ( p = 0 . 8 ) . HAT Sero-K-Set was slightly more specific ( p = 0 . 04 ) on samples from Côte d’Ivoire ( 91 . 7% , CI 86 . 2–95 . 5 ) than those from Guinea ( 83 . 2 , CI 74 . 4–89 . 9 ) , while no difference was observed with SD Bioline HAT ( p = 0 . 7 ) . Considering combined positivity in both SD Bioline HAT and HAT Sero-K-Set , increased the specificity significantly to 93 . 4% ( CI 89 . 6–96 . 1 ) compared to the single RDTs ( p≤0 . 0003 ) , while high sensitivity was maintained ( p>0 . 16 ) . Sensitivity and specificity of immune trypanolysis were respectively 100% ( CI 98 . 4–100 ) and 95 . 7% ( CI 92 . 5–97 . 8 ) . Immune trypanolysis was significantly more specific than SD Bioline HAT and HAT Sero-K-Set ( p<0 . 0009 ) . However , no significant difference in specificity could be observed between immune trypanolysis and the combination of SD Bioline HAT with HAT Sero-K-Set ( p = 0 . 2 ) . In SD Bioline HAT , respectively 64 . 6% of SERO and 40 . 6% of SUSP tested positive . These percentages were respectively 50 . 9 and 13 . 0% for HAT Sero-K-Set , and respectively 47 . 9 and 10 . 1% for the combination of the 2 RDTs ( Table 2 ) . In immune trypanolysis , respectively 46 . 7 and 4 . 3% of SERO and SUSP were positive . Thus , significantly more SERO tested RDT or trypanolysis positive than SUSP ( p≤0 . 001 ) . With HAT Sero-K-Set or trypanolysis a similar proportion of SUSP and controls were positive ( p≥0 . 1 ) , while significantly more SUSP than controls tested positive in SD Bioline HAT ( p<0 . 001 ) . Immune trypanolysis is considered to be the reference test for presence of trypanosome specific antibodies and T . b . gambiense contact . SD Bioline HAT and HAT Sero-K-Set were positive in respectively 93 . 5% ( 301/322 , CI 90 . 2–95 . 9 ) and 94 . 4% ( 304/322 , CI 91 . 3–96 . 7 ) of immune trypanolysis positive persons . In immune trypanolysis positives , there was no significant difference between both RDTs in number of positives ( p = 0 . 4 ) . In immune trypanolysis negative persons , respectively 76 . 3% ( 305/400 , CI 71 . 8–80 . 3 ) and 88 . 0% ( 352/400 , CI 84 . 4–91 . 0 ) were negative in SD Bioline HAT and HAT Sero-K-Set . In this group , HAT Sero-K-Set was significantly more negative than SD Bioline HAT ( p<0 . 0001 ) . Table 3 shows test line 1 and test line 2 results for SD Bioline HAT compared to trypanolysis with the corresponding variable antigen type , respectively LiTat 1 . 3 and LiTat 1 . 5 . In SD Bioline HAT , line 1 and 2 were significantly more positive than the corresponding variable antigen type in immune trypanolysis ( p≤0 . 0001 ) . Among the samples that were trypanolysis negative for both LiTat 1 . 3 and LiTat 1 . 5 , there was no significant difference in test line 1 or 2 positivity , nor was there in trypanolysis positive samples ( p values of 0 . 2 ) . Table 4 shows the number of positives in one or both RDTs combined with trypanolysis , considering only those subjects positive that are positive in all individual tests . Sensitivities in HAT patients were respectively 99 . 6% ( CI 97 . 6–100 ) for the combination SD Bioline HAT and trypanolysis , 99 . 1% ( CI 96 . 9–99 . 9 ) for HAT Sero-K-Set combined with trypanolysis , and 98 . 7% ( CI 96 . 3–99 . 7 ) for the combination of the 2 RDTs with trypanolysis . There was no difference in sensitivity between the 3 different test combinations ( p>0 . 2 ) . Specificities in controls were respectively 98 . 8% ( CI 96 . 6–99 . 8 ) for the combination SD Bioline HAT and trypanolysis , 98 . 1% ( CI 95 . 5–99 . 4 ) for HAT Sero-K-Set combined with trypanolysis , and 99 . 2% ( CI 97 . 2–99 . 9 ) for the combination of the 2 RDTs with trypanolysis . No significant differences were observed between the specificities of the different test combinations ( p>0 . 08 ) . However , the combination of one or 2 RDTs with immune trypanolysis was more specific than one or 2 RDTs without immune trypanolysis ( p<0 . 005 ) . The combination of one or two RDTs with trypanolysis was positive in 38 . 2–41 . 2% of SERO , 2 . 9% of SUSP ( Table 4 ) . Again , significantly more SERO tested positive than SUSP ( p≤0 . 001 ) , while a similar proportion of SUSP and controls were positive ( p≥0 . 6 ) .
This is the first study to report on HAT diagnostic accuracy on a large number of samples originating from West Africa , and also the first to perform both commercially available RDTs for serodiagnosis of HAT on the same sample set . Although sensitivity of the two tested RDTs for serodiagnosis of HAT in West Africa was high , specificity remained limited to 88% . Specificity significantly increased to 93% considering combined seropositivity in both RDTs . Using a combination of one or two RDTs with trypanolysis further improved specificity to 99% while maintaining sensitivity at 99% . For interpretation of the results , a selection bias caused by routine screening of the population at risk using the CATT test should be taken into account . This could result in an overestimation of test sensitivity and specificity , as CATT consists of whole fixed and stained trypanosomes of the LiTat 1 . 3 variable antigen type and the corresponding purified native VSG is one of the two antigens used in both RDTs as well . Furthermore the evaluation was done on stored plasma samples and not on fresh whole blood . We cannot exclude that this could influence the test results , although antibodies are well conserved after freezing . Subjectivity of scoring of the RDT test result was largely eliminated by the use of 3 independent readers . Absence of a consensus intensity or the occurrence of large differences between scores , were not frequent but can be explained by a non-uniform coloration of the test line . The RDT specificities around 88% observed in this study are close to the 87% specificity mentioned in the SD Bioline HAT test instructions ( version 53FK10–04-En-0 ) but below the previously observed specificities of 98 . 6% for HAT-Sero-K-Set [7] and of 94 . 6% for a SD Bioline HAT prototype [5] . Specificity of both RDTs was also below the 98 . 7% specificity of CATT on whole blood previously reported in West Africa [15] . Possible explanations could be regional differences [8] , cross reaction with other infections or superior challenge by animal trypanosomes to cause false positive reactions [15] , or other . Although immune trypanolysis has been considered 100% specific for HAT [15] , 4 . 3% of controls tested positive . It is not clear if this is due to false positivity , if previously treated HAT cases who did not declare themselves were included as controls , or if they were trypanotolerant individuals who became negative in CATT but remained immune trypanolysis positive [17] . The phenomenon of immune trypanolysis positive , CATT negative healthy controls requires further examination . Taking into account the high number of false positive test results observed , we examined the possible performance of combined positivity in both RDTs for diagnosis of HAT , taking the example of the strategy of serial testing applied with RDTs for diagnosis of HIV [18] . Although both RDTs actually available for serodiagnosis of HAT are based on identical antigens , considering combined positivity significantly increased specificity and reduced the number of false positives by almost half . Serial application of SD Bioline HAT and HAT Sero-K-Set could therefore be considered as an option for passive case finding , as long as no second generation RDTs for serodiagnosis of HAT are available based on different antigens . However , as the combined specificity of 93 . 4% is still suboptimal , the local context , on-site availability of parasitological confirmation tests and the relative cost should be taken into account when deciding on test algorithms . For surveillance of HAT , RDTs are actually being implemented in fixed health centres . In case of clinical suspicion and a positive RDT , and depending on the experience of the health centres in HAT diagnosis , local prevalence , and availability of sensitive confirmation diagnostic tests , blood on filter paper is sampled and sent to a reference centre for immune trypanolysis , either directly , or after unsuccessful parasitological examination . Those persons with a trypanolysis positive result are considered at high suspicion for infection , should be ( re- ) examined parasitologically and followed-up closely . Although in this study stored plasma samples were used for immune trypanolysis instead of filter paper , our results show the potential high diagnostic accuracy of a combined RDT-trypanolysis approach . In the final result no difference in accuracy occurred when combining one or two RDTs followed by trypanolysis . However , the serial application of two RDTs may present considerable advantages . The number of unnecessary parasitological examinations may be significantly reduced as well as the number of filter papers to be dispatched and tested in trypanolysis . Use of filter paper instead of plasma for immune trypanolysis , may further decrease of the number of trypanolysis positive SERO and SUSP individuals [19] thus further decrease the number of people to be followed up . Our data suggest that the specificity of actual RDTs for serodiagnosis of HAT might be lower than expected . Care should therefore be taken in interpretation of the result , especially since the future use of RDTs alone , without parasitological confirmation , for patient management has already been suggested [20] . Serological screening using serial application of SD Bioline HAT and HAT Sero-K-Set might offer superior specificity compared to a single RDT , maintaining high sensitivity . The combination of one or two RDTs with trypanolysis seems promising for HAT surveillance . However , the diagnostic accuracy and especially the specificity of applying a combination of RDTs on fresh blood for HAT diagnosis , without prior CATT selection , remains to be determined as well as their combination with trypanolysis on filter paper , not only in West Africa but also in Central Africa .
|
Screening for gambiense human African trypanosomiasis ( HAT ) or sleeping sickness is traditionally based on detection of trypanosome specific antibodies in blood . Whereas the card agglutination test is particularly suited for mass screening , individual rapid serodiagnostic tests ( RDTs ) are rather adapted for use in peripheral health-care centres . Two RDTs have been commercialized recently , and we assessed their diagnostic accuracy on stored plasma samples from West Africa . Immune trypanolysis was performed as a laboratory reference test for antibody presence . Although sensitivity for serodiagnosis of HAT in West Africa was high for both RDTs , their specificity was only 88% . Taking into account the high number of false positive test results , combined seropositivity in both RDTs was considered , raising specificity to 93% . Serial application of two RDTs should therefore be considered as an option for passive case finding , especially in settings with low HAT prevalence . A combination of one or two RDTs with immune trypanolysis further improved specificity for HAT to 99% , while maintaining sensitivity at 99% and seems promising for HAT surveillance .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] |
[] |
2015
|
Accuracy of Individual Rapid Tests for Serodiagnosis of Gambiense Sleeping Sickness in West Africa
|
Understanding emotion is critical for a science of healthy and disordered brain function , but the neurophysiological basis of emotional experience is still poorly understood . We analyzed human brain activity patterns from 148 studies of emotion categories ( 2159 total participants ) using a novel hierarchical Bayesian model . The model allowed us to classify which of five categories—fear , anger , disgust , sadness , or happiness—is engaged by a study with 66% accuracy ( 43-86% across categories ) . Analyses of the activity patterns encoded in the model revealed that each emotion category is associated with unique , prototypical patterns of activity across multiple brain systems including the cortex , thalamus , amygdala , and other structures . The results indicate that emotion categories are not contained within any one region or system , but are represented as configurations across multiple brain networks . The model provides a precise summary of the prototypical patterns for each emotion category , and demonstrates that a sufficient characterization of emotion categories relies on ( a ) differential patterns of involvement in neocortical systems that differ between humans and other species , and ( b ) distinctive patterns of cortical-subcortical interactions . Thus , these findings are incompatible with several contemporary theories of emotion , including those that emphasize emotion-dedicated brain systems and those that propose emotion is localized primarily in subcortical activity . They are consistent with componential and constructionist views , which propose that emotions are differentiated by a combination of perceptual , mnemonic , prospective , and motivational elements . Such brain-based models of emotion provide a foundation for new translational and clinical approaches .
To develop a model for emotion categories and test its accuracy in diagnosing the emotions being cultivated in specific studies , we constructed a generative , Bayesian Spatial Point Process ( BSPP ) model of the joint posterior distribution of peak activation locations over the brain for each emotion category ( see Methods and [38] ) . The BSPP model is a hierarchical Bayesian representation of the joint density of the number and locations of peak activations within a study ( i . e . , x , y , z coordinates ) given its particular emotion category . The BSPP model differs from standard univariate [12 , 13 , 39] and co-activation based [40 , 41] approaches to meta-analysis in several fundamental ways . For instance , Activation Likelihood Estimation ( ALE ) , multi-level kernel density analysis ( MKDA ) , and co-activation approaches are 1 ) not generative models of the emotion , and 2 ) not multivariate in brain space . Because they are not generative models , standard analyses provide only descriptive , summary maps of activity or bivariate co-activation for different psychological states . The BSPP , by contrast , can be used to predict the number and locations of activation in a new study given its emotion category and the probability that a new study will contain peak activations within a particular region or regions . The generative ( or ‘forward’ ) model estimates a set of brain locations , or ‘population centers’ , that are consistently active during instances of a given emotion category . Stochastic sampling from these population centers with study-level variation ( in methods , pre-processing , statistical analysis , etc . ) and measurement-level spatial noise is assumed to generate the observed data . The result is a rich , probabilistic representation of the spatial patterns of brain activity associated with each emotion category . Once estimated , the model can be used to [1] investigate the brain representations for each emotion implicit in the model and [2] infer the most likely emotion category for a new study based on its pattern of activation ( ‘reverse’ inference ) . The generative model concerns the process by which emotional instances of a given category produce observed peak activation foci , and the likelihood with which they do so . Activation data from studies or individuals are modeled at three hierarchical levels ( see Methods and [38] for more details ) . At Level 1 is the individual study data , in the form of peak coordinate locations . Level 2 models the activation centers across study with a data-generating focus that can result in variable numbers of reported locations depending on the smoothness in the image and analysis/reporting choices . Level 3 models the location of ‘true’ population centers for each emotion category with a probability distribution over space specified by the model . The model parameters—including the number and locations of population centers and spatial variation at study and peak levels—were estimated by fitting the model to peak activation coordinates from our database using Markov Chain Monte Carlo ( MCMC ) sampling with a generative birth-and-death algorithm for population centers . The MCMC procedure draws samples from the joint posterior distribution of the number and locations of peak activations in the brain given an emotion category . The posterior distribution is summarized in part by the intensity function map representing the spatial posterior expected number of activation or population centers in each area across the brain given the emotion category; this can be used to interpret the activation pattern characteristic of an emotion category ( Fig . 1A ) . Since the BSPP models the joint distribution of the number and locations of a set of peak coordinates , the posterior distribution also includes information about the co-activation across voxels; thus , MCMC samples drawn from it can be used to infer on the co-activation patterns and network properties for each emotion category ( discussed below ) .
We applied the BSPP model to ‘decode’ each study’s emotion category from patterns of brain activity in our meta-analytic database . Once the Bayesian model is estimated , it can be inverted in a straightforward manner to estimate the posterior probability of each emotion category given a set of brain activation coordinates ( see Methods ) . We used Bayes rule to obtain these probabilities , assuming no prior knowledge of the base-rate of studies in each category ( i . e . , flat priors ) , and used leave-one-study-out cross-validation so that predictions were always made about studies not used to train the model . The model performed the five-way classification of emotion categories with accuracy ranging from 43% for anger to 86% for fear to ( mean balanced accuracy = 66%; Fig . 1B; S1 Table ) ; chance was 20% for all categories , and absence of bias was validated by a permutation test . The BSPP model outperformed both a Naïve Bayes classifier ( mean accuracy was 35% ) and a nonlinear support-vector-machine based classifier ( mean accuracy was 33%; see Supplementary Methods for details ) , confirming its utility in distinguishing different emotion categories . Next , we examined whether the covariance between emotion categories and the methods and stimuli used to induce emotion ( S1 Fig ) contaminated the classification accuracies . For instance , 23% of the studies used recall and 50% used visual images to induce sadness , whereas 2% of studies used recall and 90% used visual images to induce fear . Thus , patterns for sadness vs . fear might be differentiable because the different stimuli elicit different brain responses . We verified that classification results were essentially unaffected by controlling for induction method ( 40–83% accuracy across the five emotion categories , and 61% on average; S1 Table ) . We also attempted to predict the emotion category using several methodological variables , including the method of elicitation ( the most common were visual , auditory , imagery , and memory recall ) , stimulus type ( faces , pictures , films , words , and others ) , participant gender , control condition , and imaging technique ( PET or fMRI ) . Several of these variables accurately classified some emotion categories in the five-way classification ( S2 Table ) , but no methods variable performed as well as the original BSPP model in accuracy . Stimulus type , task type , and imaging technique predicted emotion significantly above chance , at 32% , 26% , and 26% accuracy , respectively . Elicitation method , participant sex , and control condition for the emotion contrasts were at 24% , 21% , and 18% , respectively , all non-significant ) . Thus , although there are some dependencies between the methods used and the emotion categories studied , the emotion category patterns that we identified with our BSPP approach appeared to generalize across the different methods ( at least as represented in our sample of studies ) . Fig . 1C shows the intensity maps associated with each emotion category . The distinctiveness for each emotion category was distributed across all major regions of the cortex , as well as in subcortical regions , as supported by additional analyses described below . Notably , limbic and paralimbic regions such as the amgydala , ventral striatum , orbitofrontal cortex ( OFC ) , anterior cingulate cortex ( ACC ) , brainstem , and insula were likely to be active in all emotion categories , though with different and meaningful distributions within each region ( as shown in analyses below ) . In addition , regions typically labeled as ‘cognitive’ or ‘perceptual’ were also engaged and potentially differentially engaged across categories , including ventromedial , dorsomedial , and ventrolateral prefrontal cortices ( vmPFC , dmPFC , and vlPFC ) , posterior cingulate ( PCC ) , hippocampus , and medial temporal lobes , and occipital regions . To characterize the BSPP intensity maps , we calculated the mean intensity for each emotion category in 49 a priori regions and networks , which together covered the entire brain ( Fig . 2A ) . For cortical , basal ganglia , and cerebellar networks , we used results from Buckner and colleagues [42–44] , who identified seven networks with coherent resting-state connectivity across 1 , 000 participants . Each of the seven included a cortical network and correlated areas within the basal ganglia ( BG ) and cerebellum . We supplemented these networks with anatomical sub-regions within the amygdala , hippocampus , thalamus , and the brainstem and hypothalamus . We tested whether a ) broad anatomical divisions ( e . g . , cortex , amygdala ) showed different overall intensity values across the five emotion categories; and b ) whether the ‘signature’ of activity across networks within each division differed significantly across emotions ( S3 Table ) . Our broad goal , however , was not to exhaustively test all emotion differences in all regions , but to provide a broad characterization of each emotion category and which brain divisions are important in diagnosing them .
One of the important differences between the BSPP model and previous meta-analytic approaches is that is it sensitive to co-activation patterns across regions . By saving the average intensity values for each region/network from each MCMC iteration , we were able to estimate the co-activation intensity as the correlation between average intensity values for each pair of regions . We used a permutation test to threshold the co-activation estimates ( using the most stringent of the q < . 05 FDR-corrected thresholds across categories ) . Fig . 3 shows that each emotion category was associated with a qualitatively different configuration of co-activation between cortical networks and subcortical brain regions . In Fig . 3A , force-directed graphs of the relationships among the 49 anatomical regions/networks demonstrate very different topological configurations for the five emotion categories . In these graphs , regions and networks are represented by circles ( nodes ) , with significant co-activations ( edges ) represented as lines . The size of each circle reflects the region/network’s betweeness centrality [48 , 49] , a graph-theoretic measure of the degree to which a region/network is a ‘connector’ of multiple other regions . Colors reflect membership in six cerebral zones: Cortex , basal ganglia , cerebellum/brainstem , thalamus , amygdala , and hippocampus . Fig . 3B shows the same graph’s relationships in anatomical brain space . Finally , Fig . 3C shows estimates of average co-activation within ( diagonals ) and between ( off-diagonals ) the six cerebral zones . The co-activation metric reflects global efficiency , based on the average shortest path length , a graph theoretic measure of the shortest path connecting the regions in the co-activation graphs , and was calculated as the average ( 1/path length ) between pairs of regions/networks . Emotion patterns were distinguished by both their patterns of co-activation and by the regions that are ‘connectors . ’ The anger category is characterized by relatively strong and uniform connections across cerebral zones compared to other emotion categories , with strong co-activation among cortical , basal ganglia , and cerebellar networks , and other regions , particularly the right amygdala . Connectors ( >90th percentile in betweenness-centrality ) include the cortical frontoparietal network , right amygdala , and brainstem . ( Visual cortex was a connector for all emotion categories except sadness ) . In disgust , by contrast , cortical networks connect to basal ganglia regions and serve as a bridge to an otherwise isolated cerebellum . Connectors include the somatomotor basal ganglia network and brainstem . The fear category is marked by reduced co-activation among cortical networks and between cortex and other structures , but the basal ganglia are tightly integrated with the amygdala and thalamus . In happiness , intra-cortical co-activation is higher , but cortical-subcortical co-activation is low , and connectors include the limbic cortical network , motor thalamus , and visual basal ganglia and cerebellum . Sadness is characterized by dramatically reduced co-activation within the cortex , between cortex and other regions , and between cerebellum and other regions . Intra-thalamic , intra-basal ganglia , and intra-cerebellar co-activation are relatively preserved , but large-scale connections among systems are largely absent . Connectors include the limbic cerebellum , brainstem , two hippocampal regions , and the left centromedial amygdala .
Because it is a generative model , the BSPP model of emotion categories is capable of making predictions about new instances . Other methods—such as our previous MKDA analyses , ALE analyses , and bivariate co-activation analyses that we and others have developed—are not generative models , and would not be expected to be appropriate to or perform well in classification . In addition , unlike 'brute force' pattern classification algorithms , we can classify emotions with a single , generative representation of each emotion category . For example , when using Support Vector Machines to discriminate the five categories , ten separate classifier maps ( 5-choose-2 ) are required to predict the category , rather than relying on a single representation of each category and the likelihood that a particular study belongs to it . In addition , the nonlinear SVM model we adapted for meta-analytic classification performs substantially more poorly in classification . In the BSPP model , each representation includes information about both activation and co-activation across systems . However , unlike data-driven pattern classification models , this model can be queried flexibly—i . e . , here , we present graphs of bivariate ( 2-region ) co-activation strengths , but other , more comprehensive summaries can be used , including those that were not explicitly used in model training . For example , we demonstrate this by using non-negative matrix factorization ( NNMF ) to derive canonical profiles of activation across cortical and subcortical systems . We then re-calculate the model likelihood according to the new metric of canonical profile activation , without re-fitting the model , and are able to make statistical inferences about the differences among emotion categories in that new feature space . This flexibility is a hallmark of generative Bayesian models that provides advantages in allowing researchers to test new metrics , features , and patterns , rather than being limited to a fixed set of features such as pair-wise correlations . Beyond these considerations of methodology and broad interpretation , the present findings bear on theories of emotion , and the ways in which studies look for the hallmarks of particular emotional experiences , in novel and specific ways . We elaborate on some of these below . In this study , the five way decoding accuracy for emotion categories varies substantially across categories . Fear was the most accurate overall , with 86% accuracy , whereas anger was the least accurate , at 43% . These findings could indicate heterogeneity in the categories themselves . However , it could also reflect the signal detection properties of the test itself , as we explain below . Thus , it is premature to make strong claims about the diversity/heterogeneity of the emotion categories based on these results . Heterogeneity in the representation across categories occurs both because some of the methods used to elicit emotion are more diverse ( S2 Table ) and because the categories are likely to be inherently psychologically and neurally diverse . We think of each emotion category as a population of diverse instances , rather than a homogeneous set of instances . Thus , there may be multiple types of ‘anger’ that activate different subsets of regions and networks . What we observe is the population average across these potentially disparate features . This is analogous to dwellings containing disparate architectural features ( e . g . , an igloo vs . a castle ) being grouped into a common category ( ‘dwelling’ ) because of their cultural functions rather than their architectures . On a more mundane level , the ways in which researchers choose to study emotion categories can also contribute to the observed diversity ( and reduced classification accuracy ) ; researchers studying fear , for example , tend to sample very similar instances by using a small range of fear-inducing methods , whereas anger is elicited in more diverse ways . A second potential reason for differences and accuracy relates to the signal detection properties of the model . Sensitivity and specificity can always be traded off by changing the decision threshold for labeling an instance as ‘anger , ’ ‘fear , ’ etc . , and accuracy in 5-way classification is more closely related to sensitivity than specificity . Here , anger has the lowest sensitivity ( 43% ) , but the highest specificity ( 99% , Table 1 ) : Studies that are not anger are almost never categorized as anger . Such differences in threshold preclude making strong claims about the diversity/heterogeneity of the emotion categories themselves based on these results . However , we should not be too quick to dismiss all differences in decoding accuracy to methodological artifacts; true differences in category heterogeneity may exist as well . A number of important issues and limitations remain to be addressed . First , our analyses reflect the composition of the studies available in the literature , and are subject to testing and reporting biases on the part of authors . This is particularly true for the amygdala ( e . g . , the activation intensity for negative emotions may be over-represented in the amygdala given the theoretical focus on fear and related negative states ) . However , the separation of emotion categories in the amygdala was largely redundant with information contained in cortical patterns , which may not be subject to the same biases . Likewise , other interesting distinctions were encoded in the thalamus and cerebellum , which have not received the theoretical attention that the amygdala has and are likely to be bias-free . Secondly , these results are limited by the inherent resolution and signal properties of the original studies . Some regions—particularly the brainstem—are likely to be much more important for understanding and diagnosing emotion than is apparent in our findings , because neuroimaging methods are only now beginning to focus on the brainstem with sufficient spatial resolution and artifact-suppression techniques ( Satpute et al . , 2013 ) . Other areas that are likely to be important , such as the ventromedial prefrontal cortex ( e . g . , BA 25 and posterior portions of medial OFC ) are subject to signal loss and distortion , and are likely to be under-represented . Thirdly , there is always the possibility that differences in study procedures or the involvement of processes not directly related to emotional experience could partially explain some findings . A meta-analytic result is only as good as the data from which it is derived , and a brief look at S1 Fig indicates that there are some systematic differences in the ways researchers have studied ( and evoked instances of ) different emotion categories . We have tried to systematically assess the influence of methodology differences in this paper , but our ability to do this is imperfect . However , though we cannot rule out all possible methodological differences , we should not be too quick to dismiss findings in ‘sensory processing’ areas , etc . , as methodological artifacts . Emotional responses may be inherently linked to changes in sensory and motor cortical processes that contribute to the emotional response ( e . g . , [101] ) . This is a central feature of both early and modern embodiment-based theories of emotion [92 , 102–104] . In addition , most major theories of emotion suggest that there are systematic differences in cognitive , perceptual , and motor processes across emotion categories; and in some theories , such as the appraisal theories , those differences are inherently linked to or part of the emotional response [105] . Finally , the results we present here provide a co-activation based view of emotion representation that can inform models of functional connectivity . However , co-activation is not the same as functional connectivity . The gold-standard measures of direct neural connectivity use multiple single-unit recording or optogenetics combined with single-unit electrophysiology to identify direct neural connections with appropriate latencies ( e . g . , < 20 msec ) . Much of the information processing in the brain that creates co-activation may not relate to direct neural connectivity at all , but rather to diffuse modulatory actions ( e . g . , dopamine and neuropeptide release , much of which is extrasynaptic and results in volume transmission ) . Thus , the present results do not imply direct neural connectivity , and may be related to diffuse neuromodulatory actions as well as direct neural communication . However , these forms of brain information processing may be important in their own right .
The dataset consists of activation foci from 397 fMRI and PET studies of emotion published between 1990 and 2011 , which included a total of 914 unique study activation maps and 6 , 827 participants . Activation foci are coordinates reported in Montreal Neurologic Institute standard anatomical space ( or transformed from Talairach space ) . Foci are nested within study activation maps , maps of group comparisons between an emotion- or affect-related condition and a less intense or affectively neutral comparison condition . We used the foci associated with study activation maps to predict each map’s associated emotion category . Studies were all peer-reviewed and were identified in journal databases ( PubMed , Google Scholar , and MEDLINE ) and in reference lists from other studies . A subset of studies that focused specifically on the most frequently studied emotion categories were selected ( 148 studies , 377 maps , 2519 participants ) . Categories included anger ( 69 maps ) , disgust ( 69 maps ) , fear ( 97 maps ) , happiness ( 77 maps ) , and sadness ( 65 maps ) . The BSPP is built on a hierarchical marked independent cluster process designed for functional neuroimaging meta-analysis [38] . We model the foci ( peak activation locations ) as the offspring of a latent study center process associated with a study activation map . The study centers are in turn offspring of a latent population center process . The posterior intensity function of the population center process provides inference on the location of population centers , as well as the inter-study variability of foci about the population centers . Specifically , the model has three levels of hierarchy . At level 1 , for each study , we assume the foci are a realization of an independent cluster process driven by a random intensity function . These processes are independent across studies . The study level foci are made up of two types of foci: singly reported foci and multiply reported foci . For a given activation area in the brain , some authors only report a single focus , while others report multiple foci , however this information is rarely provided in the literature . These differences are attributable to how different software packages report results , and simply author preference . We assume that multiply reported foci cluster about a latent study activation center , while the singly reported foci can either cluster about a latent population center or are uniformly distributed in the brain . At level 2 , we model the latent study activation center process as an independent cluster process . We assume that the latent study activation centers can either cluster about the latent population center or are uniformly distributed in the brain . At level 3 , we model the latent population center process driven by a homogeneous random intensity ( a homogeneous Poisson process ) . The points that may cluster about the population centers are singly reported foci from level 1 and study activation centers from level 2 . We make inference on the latent population centers in a Bayesian framework . In particular , we use spatial birth and death processes nested within a Markov chain Monte Carlo simulation algorithm . The details of the algorithm and pseudo code are provided in [38] . The BSPP model estimates the posterior distribution for reported foci across studies . Foci reported in each emotion category can be modeled as an emotion-specific spatial point process . This leads to a joint model for foci from studies with different categories of emotions , and it can be used to classify the emotion category of studies given their observed foci , by choosing the emotion category that maximizes the posterior predictive probability . To be more specific , suppose we have n studies , let Fi and Ei denote the foci and the emotion category for study i respectively , for i = 1 , … , n . The BSPP model specifies the probability π ( Fi|Ei , λ ) , where λ represents the collection of all the parameters in the BSPP model . The posterior predictive probability of emotion category for a new study En+1 is given by Pr ( En+1=e| ( Fi , Ei ) i=1n , Fn+1 ) ∝Pr ( En+1=e ) ∫Πi=1nπ ( Fi|Ei , λ ) π ( Fn+1|En+1=e , λ ) π ( λ ) dλ , for e = 1 , … , m , where m represents the total number of emotion categories . Pr ( En+1 = e ) represents the prior probability of emotion category for the study type and π ( λ ) is the prior of parameters . The performance of the proposed classifier is evaluated via leave-one-out cross validation ( LOOCV ) on all the observed data , i . e . , leaving one study out . We conduct Bayesian learning of the model parameters on the foci reported from a set of training studies consisting of all studies except a left-out study , k . We then make a prediction for study k based on its reported brain foci . We repeat the procedure for each study ( 1…K ) and compute the classification rate across all studies . The above procedure for a Bayesian model can be very computationally expensive since it involves multiple posterior simulations . We employ an importance sampling method to substantially reduce the computation . See [106] for details . To investigate the similarities and differences among emotion categories in defined resting-state fMRI and anatomical networks , we identified a priori networks and regions from published studies as described above [42–44 , 107 , 108] ( see also Fig . 2 legend ) . These networks covered the entire cerebrum , excluding white matter and ventricles . Within each of the 49 regions , we calculated the average BSPP intensity value across voxels for each emotion category . Analyses of the mean intensity across regions/networks are visualized in Figs . 2 and S3 and S1 Text . Non-negative matrix factorization ( NNMF ) is a way of decomposing a complex data set into simpler , additive components that are particularly interpretable [109] . Here , we used it to decompose the matrix of activation intensities for each of the five emotions across subgroups of 49 regions into simpler , additive ‘profiles’ of activation shown in polar plots in Figs . 2 and S3 . The activation matrix A is decomposed into two component matrices W ( n•k ) and H ( k•m ) whose elements are non-negative , such that A = WH , with the number of components ( k ) selected a priori ( here , k = 2 for interpretability and visualization ) . The squared error between A and WH was minimized via an alternating least squares algorithm with multiple starting points . The rows of H constitute the canonical profiles shown in figures , and emotion-specific activation intensity values from the BSPP model are plotted in the 2-dimensional space of the two recovered canonical activation profiles . NNMF is a particularly appropriate and useful decomposition technique here , because activation intensity is intrinsically non-negative [110 , 111] . In such cases , NNMF identifies components that are more compact and interpretable than principal components analysis ( PCA ) or independent components analysis ( ICA ) , and better reflect human intuitions about identifying parts that can be additively combined into wholes . Here , the parts reflect interpretable , canonical activation profiles , and the whole is the observed activation profile for each emotion category across multiple brain systems .
|
Neuroimaging provides a unique way of understanding the ‘emotional brain’ by identifying patterns across multiple systems that imbue each instance of emotion with its particular qualities . In this meta-analysis across 148 studies , we ask whether it is possible to identify patterns that differentiate five emotion categories—fear , anger , disgust , sadness , and happiness—in a way that is consistent across studies . Our analyses support this capability , paving the way for brain markers for emotion that can be applied prospectively in new studies and individuals . In addition , we investigate the anatomical nature of the patterns that are diagnostic of emotion categories , and find that they are distributed across many brain systems associated with diverse cognitive , perceptual , and motor functions . For example , among other systems , information diagnostic of emotion category was found in both large , multi-functional cortical networks and in the thalamus , a small region composed of functionally dedicated sub-nuclei . Thus , rather than relying on measures in single regions , capturing the distinctive qualities of different types of emotional responses will require integration of measures across multiple brain systems . Beyond this broad conclusion , our results provide a foundation for specifying the precise mix of activity across systems that differentiates one emotion category from another .
|
[
"Abstract",
"Introduction",
"Results",
"Network",
"Co-activation",
"Differences",
"among",
"Emotion",
"Categories",
"Discussion",
"Methods"
] |
[] |
2015
|
A Bayesian Model of Category-Specific Emotional Brain Responses
|
Vertebrate hearts depend on highly specialized cardiomyocytes that form the cardiac conduction system ( CCS ) to coordinate chamber contraction and drive blood efficiently and unidirectionally throughout the organism . Defects in this specialized wiring system can lead to syncope and sudden cardiac death . Thus , a greater understanding of cardiac conduction development may help to prevent these devastating clinical outcomes . Utilizing a cardiac-specific fluorescent calcium indicator zebrafish transgenic line , Tg ( cmlc2:gCaMP ) s878 , that allows for in vivo optical mapping analysis in intact animals , we identified and analyzed four distinct stages of cardiac conduction development that correspond to cellular and anatomical changes of the developing heart . Additionally , we observed that epigenetic factors , such as hemodynamic flow and contraction , regulate the fast conduction network of this specialized electrical system . To identify novel regulators of the CCS , we designed and performed a new , physiology-based , forward genetic screen and identified for the first time , to our knowledge , 17 conduction-specific mutations . Positional cloning of hobgoblins634 revealed that tcf2 , a homeobox transcription factor gene involved in mature onset diabetes of the young and familial glomerulocystic kidney disease , also regulates conduction between the atrium and the ventricle . The combination of the Tg ( cmlc2:gCaMP ) s878 line/in vivo optical mapping technique and characterization of cardiac conduction mutants provides a novel multidisciplinary approach to further understand the molecular determinants of the vertebrate CCS .
Vertebrate hearts have evolved into multichambered structures requiring coordinated beating of their chambers to achieve antegrade blood flow throughout the organism . Unidirectional blood flow is achieved through two specialized structures that are unique to vertebrates: cardiac valves and the specialized cardiac conduction system ( CCS ) . In the adult heart , the initial electrical impulses are generated in the slow pacemaker sino-atrial ( SA ) node and then propagated across the atrium . This electrical impulse is delayed at the atrioventricular ( AV ) boundary through specialized slow conducting AV node cardiomyocytes . After the delay at the AV node , electrical propagation travels rapidly through the fast conduction network comprised of the His-Purkinje system , which coordinates ventricular activation to occur from the apex to the base of the heart . This apex-to-base activation allows for efficient ejection of blood from the ventricles into the outflow tracts ( OFTs ) at the base of the heart [1] . Despite extensive knowledge of the anatomy and physiology of the adult vertebrate CCS , the cellular and molecular events that govern the development of this specialized tissue remain unclear . Lineage tracing studies have revealed that the CCS is derived from cardiomyocyte progenitors [2 , 3] . Myocardial factors that regulate the specification of the CCS include Nkx2 . 5 and Tbx5 [2 , 4] . Loss of either transcriptional regulator leads to defects in the maturation and maintenance of the AV conduction system and subsequent AV heart block and bundle branch block . Additional studies have revealed the requirement of the endocardium for cardiomyocyte specification to form the fast conduction network within the ventricle [5–7] . Secreted factors from endocardial as well as other cardiac endothelial cells , such as Endothelin 1 and Neuregulin , are able to induce cardiac conduction markers in cultured embryonic cardiomyocytes and cultured hearts [7–9] . Furthermore , hemodynamic changes regulate the secretion of Endothelin 1 from endocardial cells , thereby affecting the development of the fast conduction pathway [6] . More recently , the role of the endocardium for the development of AV conduction delay has been investigated further using the zebrafish cloche mutant [5] , which lacks endothelial cells among other defects [10] . That study concluded that Neuregulin but not Endothelin 1 is required for the induction of AV conduction delay . Optical mapping of cardiac excitation using voltage- and calcium-sensitive dyes has allowed the spatiotemporal analysis of electrical excitation wave dynamics , not only advancing our understanding of the electrical activity during cardiac arrhythmias but also allowing for further analysis of CCS development [11] . However , the use of voltage- and calcium-sensitive dyes is associated with serious shortcomings , including a lack of cellular targeting , limited live animal experimentation , the need for physical loading of these indicators into cells , and cellular toxicity . To circumvent these problems , fluorescent calcium indicator proteins have begun to replace voltage- and calcium-sensitive dyes for physiologic in vivo analysis of tissue/organ electrical activity in different animal model systems including fly and mouse [12–14] . Yet , optical mapping of mouse hearts is currently limited due to explantation for ex vivo analysis . Thus , we have taken advantage of the external fertilization and translucency of zebrafish embryos to create a cardiac-specific fluorescent calcium indicator transgenic line , Tg ( cmlc2:gCaMP ) s878 , to perform in vivo optical mapping analyses throughout the stages of heart development . Here we describe a multidisciplinary approach using the zebrafish toward understanding CCS development . Utilizing the Tg ( cmlc2:gCaMP ) s878 optical mapping system , we identified four distinct physiologic developmental stages of the CCS that correspond to cellular and anatomical changes of the developing zebrafish heart . ( 1 ) Initially , a linear conduction travels across the heart tube from the sinus venosus to the OFT ( 20–24 hours postfertilization ( hpf ) ) ; ( 2 ) next , a significant AV conduction delay develops during cardiac chamber formation ( 36–48 hpf ) ; ( 3 ) as the heart loops and develops ventricular trabeculations ( 72–96 hpf ) , an immature fast conduction network develops within the ventricle; ( 4 ) finally , this fast conduction network fully matures to an apex-to-base activation pattern when the ventricular apex has formed . Furthermore , to identify regulators of CCS development , we performed a diploid ethylnitrosourea ( ENU ) mutagenesis screen and recovered several novel as well as known cardiovascular conduction/rhythm mutants , which we have analyzed using in vivo optical imaging techniques and classified according to the affected physiologic developmental stage of the CCS . Positional cloning of hobgoblin ( hob ) , a novel mutant with AV heart block , reveals that tcf2 , a homeobox transcription factor gene involved in mature onset diabetes of the young , also regulates conduction between the atrium and the ventricle . Thus , these detailed electrophysiologic and genetic analyses of wild-type and mutant hearts provide further insights into the development of the vertebrate CCS and will lead to a better understanding of the pathogenesis of cardiac arrhythmias .
Previous studies have utilized calcium green , a calcium-sensitive fluorescent indicator , in zebrafish hearts to observe cardiac conduction up to 48 hpf [5 , 15] . However , because loading these hearts and/or embryos with calcium green is technically cumbersome and is only temporary , we created a zebrafish transgenic line Tg ( cmlc2:gCaMP ) s878 that specifically expresses gCaMP , a genetically encoded calcium reporter based on a circular permutation of green fluorescent protein ( GFP ) [16] , at all developmental stages in the heart , using the cardiac-specific promoter cmlc2 [17] ( Figure S1 ) . Because cardiac contraction and blood flow begins at the linear heart tube ( LHT ) stage , we initiated our optical mapping studies on the LHT of 24 hpf Tg ( cmlc2:gCaMP ) s878 embryos . These experiments revealed that conduction travels unidirectionally in a relatively slow and linear pattern without significant pauses from the sinus venosus to the OFT , suggesting the presence of a functional SA node pacemaker activity ( Figure 1A and 1B and Video S1 ) . Additionally , an acceleration of conduction was observed within the OFT half of the heart ( Figure 1B , arrow ) . Utilizing the Tg ( cmlc2:eGFP-ras ) s883 line ( Jungblut B , Munson C , Huisken J , Trinh L , Stainier D , unpublished data ) , which outlines individual cardiomyocytes with membrane-bound GFP , confocal microscopy was performed to analyze further the cellular characteristics of the LHT . Despite displaying atrial and ventricular molecular changes [18] ( Figure 1E and 1F ) , 24 hpf cardiomyocytes maintain a nearly uniform squamous morphology on cross-sectional analysis ( Figure 1C ) . However , we observed a small subpopulation of cuboidal cardiomyocytes near the OFT ( Figure 1C , box ) . These cuboidal cardiomyocytes correlate with the acceleration of conduction observed in the optical mapping of the LHT . By 36–48 hpf , the zebrafish embryonic heart has developed a distinct AV canal that separates the cardiac chambers ( Figure 2C ) . Calcium activation travels from the sinus venosus across the atrium to the AV canal ( Figure 2A and 2B ) . At the AV canal , a dramatic slowing of the calcium activation wave was observed as illustrated by an increased number of isochronal lines at the AV boundary between the atrial and the ventricular chambers ( Figure 2B , arrowhead ) . Subsequently , ventricular calcium activation proceeded from the AV canal , accelerating laterally across the ventricular myocardium to the OFT where significant deceleration of conduction was observed ( Figure 2B , arrow , and Video S2 ) . These AV and OFT conduction delays may help to prevent regurgitant blood flow between the atrial and the ventricular chambers as well as between the heart and the bulbus arteriosus , respectively , thus resulting in efficient antegrade blood flow throughout the embryo . To determine whether cell morphology or orientation may correlate with these conduction velocity differences , we analyzed and measured cardiomyocytes of 48 hpf Tg ( cmlc2:eGFP-ras ) s883 hearts . On cross-sectional analysis , it appeared that atrial cardiomyocytes had maintained their squamous cell morphology while ventricular cardiomyocytes had become cuboidal ( Figure 2D and 2F ) and cardiomyocytes at the AV boundary had initiated apical membrane constriction , resulting in cells with a distinct trapezoidal shape ( Figure 2D , yellow box ) . Furthermore , examination of the surface of ventricular cardiomyocytes revealed that outer curvature ( OC ) cardiomyocytes ( the greater , convex curvature of the cardiac chamber ) became elongated while inner curvature ( IC ) cardiomyocytes ( the lesser , concave curvature of the cardiac chamber ) remained rounded ( Figure 2C and 2E ) , as previously described [19] . Interestingly , cardiomyocytes at the AV boundary also became significantly elongated , forming a ring of cells around the AV canal ( Figure 2C , yellow box and inset ) . The orientation of these AV cardiomyocytes was orthogonal to that of the OC ventricular cardiomyocytes . To determine whether these different populations of myocardial cells exhibit distinct electrophysiologic properties , calcium transients of the atrium , ventricle , and AV canal were recorded from fluorescence of a single pixel from each region . Distinct calcium transients were recorded from each region of 48 hpf Tg ( cmlc2:gCaMP ) s878 hearts ( Figure 3B and 3C ) . To further test these findings , action potentials ( APs ) were recorded in explanted 48 hpf wild-type hearts using patch pipettes and the current clamp technique ( Figure 3D ) . The APs recorded from zebrafish atrium were similar in morphology to those reported in embryonic mammalian myocardium , including spontaneous diastolic depolarization . The APs recorded from zebrafish ventricle were typical of mammalian ventricles , with a flat diastolic phase and an overt systolic plateau phase . The APs recorded from the AV canal were distinguishable from atrial and ventricular APs by the presence of a slow diastolic depolarization phase in the setting of an overt systolic plateau phase . The morphology of the APs recorded from the wild-type AV canal was similar to that reported for the mammalian AV node [20] , suggesting that the distinct electrophysiologic properties of AV myocardial cells contribute to the conduction delay between the cardiac chambers . To achieve efficient ejection of blood from the ventricle to the arterial system , the adult vertebrate ventricle contracts from the apex to the base of the heart ( i . e . , where the AV canal and OFT reside ) . This contraction pattern is achieved through the fast cardiac conduction network ( the His-Purkinje system ) , which passes through the ventricular septum in amniotes and allows for apex-to-base conduction across the ventricular myocardium . Previous studies have shown that the fast CCS in chick and mammalian hearts may initially develop along ventricular trabeculae after cardiac looping but prior to ventricular septation [6 , 21] . To understand further how the fast CCS develops , we analyzed zebrafish hearts at stages after cardiac looping . Using the Tg ( flk1:eGFP ) s843 line , which marks endocardial cells with green fluorescence [22] , we observed that rhodamine-phalloidin–stained 100 hpf zebrafish hearts not only have completed cardiac looping but also have initiated ventricular trabeculation ( Figure 4C ) . By 2–3 weeks postfertilization , the ventricle has developed an apex and increased its trabeculation ( Figure 4G ) . Optical mapping of 100 hpf Tg ( cmlc2:gCaMP ) s878 hearts revealed that ventricular conduction has transformed from a primitive linear propagation traveling from the AV canal to the OFT as observed at 48 hpf ( Figure 2 ) to a more complex conduction pattern propagating from the OC of the ventricle to the base/IC . After the AV conduction delay , the earliest ventricular calcium activation was observed along the trabeculae ( Figure 4A and 4B , activation sequence 7 ) . Next , calcium activation rapidly proceeded from the trabeculae to the adjacent peripheral cardiomyocytes , leading to conduction from the OC to the base/IC ( Figure 4A and 4B , activation sequences 8 and 9 , and Video S3 ) . Finally , similar to 48 hpf hearts , ventricular conduction terminated at the OFT where conduction velocity significantly decelerated ( Figure 4A and 4B , activation sequences 9–12 ) . Interestingly , trabeculae were not observed within the OFT ( Figure 4C ) , further suggesting that they may be responsible for rapid ventricular conduction . At 2–3 weeks of life , when the ventricle clearly forms an apex , we observed that the initial conduction along the trabeculae ( Figures 4E and 4F , sequence activation 4 ) breaks through at the apex , resulting in apex-to-base/IC conduction ( Figure 4E and 4F , activation sequences 5 and 6 ) . Thus , these results indicate that the fast cardiac conduction pathway develops prior to the formation of the apex , resulting in ventricular conduction from the OC to the base/IC to allow efficient ejection of blood from the heart shortly after cardiac looping . To understand further the development of the fast CCS , we examined the expression of gap junction proteins responsible for cardiac conduction . Connexin40 ( Cx40 ) is expressed in the atrium and the fast conduction system of mammalian hearts , and loss of this gap junction protein results in reduced cardiac conduction velocity and a predisposition to cardiac arrhythmias [23] . In contrast , Connexin43 ( Cx43 ) is expressed abundantly throughout the atrial and ventricular myocardium but in lower amounts in the fast conduction system [24 , 25] . Because both cx40 and cx43 have been suggested to be expressed in zebrafish hearts [26 , 27] , we performed immunostaining on these hearts with antibodies against Cx40 and Cx43 . Cx43 immunostaining is present throughout the heart from 24 hpf onwards ( Figure 5A , 5D , and 5G ) . However , Cx40 immunoreactivity is not observed at 24 hpf , but is detected very weakly at 48 hpf , and strongly by 100 hpf throughout the myocardium ( Figure 5B , 5E , and 5H ) . Thus , Cx40 immunoreactivity is present during the development of the fast CCS . Epigenetic factors , such as hemodynamic flow and cardiac contraction , previously have been suggested to influence the development of the CCS [6 , 9] . To determine the role of hemodynamic factors in the development of both slow/AV and fast/ventricular cardiac conduction pathways in zebrafish hearts , we performed optical mapping on the silent heart ( sihb109 ) mutant heart , which fails to contract due to a null mutation in the cardiac troponin T ( tnnt2 ) gene [15] . Optical mapping of 48 hpf Tg ( cmlc2:gCaMP ) s878; sih mutant hearts revealed an AV conduction delay , similar to that of 48 hpf wild-type hearts [5] ( see Figure 7F ) . However , in contrast to the ventricular OC-to-base calcium activation pattern observed in wild-type 100 hpf hearts , 100 hpf sih mutant hearts displayed ventricular calcium activation laterally across the myocardium from the AV canal to the OFT as well as intermittent AV heart block ( Figure 6A and 6B ) . Interestingly , this intermittent AV heart block occurred within the ventricle just after the AV canal ( Figure 6B and 6B′ ) . Similar conduction patterns also were observed in cardiac contractility mutants exhibiting weak contractions throughout the heart ( unpublished data ) . To determine possible etiologies for the conduction defects in sih mutant hearts , we analyzed rhodamine-phalloidin-stained Tg ( flk1:eGFP ) s843; sih mutants to assess the effects of the lack of blood flow and contraction on cardiac development . At 100 hpf , sih mutant hearts appear to have undergone cardiac looping and possess endocardium but do not exhibit trabeculae ( Figure 6C ) . Additionally , 100 hpf sih mutant hearts showed significantly diminished Cx40 immunoreactivity but unaffected Cx43 immunoreactivity compared to wild type ( Figure 6E and 6F ) . Thus , loss of hemodynamic blood flow and contraction results in the failure of the ventricle to develop trabeculae and the down-regulation of Cx40 , thereby possibly affecting the development of the fast CCS . We performed a large-scale ENU mutagenesis screen using the Tg ( cmlc2:gCaMP ) s878 line as a secondary screen for physiologic analysis of cardiac conduction to identify genes that specifically regulate the development of the CCS . Intercrosses of F2 families were screened by visual inspection of live embryos for aberrant heart rates and/or coordination of atrial and ventricular contraction at each developmental stage of the CCS ( 24 , 48 , and 96 hpf ) . F2 carriers of putative mutations were outcrossed into the Tg ( cmlc2:gCaMP ) s878 background to facilitate the physiologic analysis of cardiac conduction . Recovered mutations were organized into phenotypic groups and tested by complementation analysis . We identified 17 mutations defining 14 cardiac conduction regulating loci , four of which previously had been identified ( Table 1 ) . Physiologic analyses by optical mapping revealed that the identified mutations disrupt distinct developmental stages of the CCS . Representative examples of phenotypes observed are described below . We recovered several noncontractile ventricle mutants including s209 , s264 , s249 , s271 , and silent ventricles213 , s290 [28 , 29] . Previous work has suggested that a noncontracting ventricle phenotype may be due to contractile defects [15 , 19 , 30] . Utilizing the Tg ( cmlc2:gCaMP ) s878 , we discovered that the silent ventricle ( siv ) mutant heart fails to generate cardiac conduction across the ventricular myocardium ( Figure 7A ) . Further electrophysiologic characterization of siv mutant hearts revealed that the ventricular membrane potential was depolarized markedly . Hyperpolarization of mutant ventricle in current clamp mode resulted in spontaneous AP generation , restoring contractility [29] . We identified several mutants that had conduction defects across the AV myocardium . The hob s634 mutants develop AV heart block despite displaying a wild-type cardiac morphology and contractility ( Figure 7C ) . However , dococs215 , 226 ( dcc ) mutants display an uncoordinated contraction of the ventricle at 36–48 hpf , which was initially characterized by brightfield microscopy as ventricular fibrillation . However , optical mapping of 48 hpf Tg ( cmlc2:gCaMP ) s878; dcc mutants revealed that the AV conduction was blocked at the superior portion of the AV myocardium , resulting in heterogeneous and disorganized conduction across the ventricular myocardium ( Figure 7B ) . In contrast , endocardial mutants including cloche [5 , 31] , santa , and valentine [32] fail to display an AV conduction delay ( unpublished data ) . Interestingly , the slipjigs644 ( sli ) mutant , which has an endocardium and vasculature ( unpublished data ) , also lacks an AV conduction delay ( Figure 7D ) . Finally , we identified a class of cardiac mutants that loses its ventricular conduction at 96 hpf . The mutant , daredevil ( ddl ) , was initially characterized as a cardiovascular contractile mutant in which the ventricle became noncontractile by 96 hpf . Optical mapping of Tg ( cmlc2:gCaMP ) s878; ddl mutants revealed that these hearts developed heart block between 80–96 hpf and eventually lost all organized ventricular conduction by 96 hpf , resulting in a noncontractile ventricle ( Figure 7E ) . Interestingly , these particular conduction mutants manifest their cardiac phenotype at the developmental stage when the fast CCS develops , suggesting that the affected genes may specifically regulate the development of the fast CCS . Because of its resemblance to human AV heart block ( Figure 8A and Video S4 ) , we isolated the gene disrupted by the hob mutation . Fine mapping of 1164 diploid embryos positioned the hob gene to a genomic interval of ∼150 kb on LG15 ( Figure 8B ) . Sequence analysis of tcf2 within the critical region revealed a G to A base pair change at position 552 , leading to a premature stop codon after the dimerization domain thereby removing the homeobox and transactivation domains ( Figure 8C ) . Data from morpholino ( MO ) knockdown , in situ analysis , and mRNA rescue experiments support the claim that tcf2 is the gene affected by the hob mutation . Injection of 0 . 5 ng of tcf2 MO recapitulated the hob mutant cardiac phenotype ( Figure 8A ) . Whole mount in situ analyses reveal that tcf2 is expressed at 48 hpf around the AV canal as well as the OFT portion of the ventricle ( Figure 8D ) . Finally , injection of wild-type tcf2 mRNA rescued the heart phenotype in ∼82% of hob mutants , while injection of mutant tcf2 mRNA failed to rescue ( Figure 8E ) . Altogether , these data indicate that tcf2 contributes to the regulation of the cardiac conduction between the atrium and the ventricle .
The vertebrate CCS can be divided into the slow conduction pathway , which regulates SA pacemaker activity and central AV conduction delay , and the fast conduction pathway , which allows apex-to-base conduction [33] . Through optical mapping of developing wild-type hearts , we observed distinct developmental stages for each of these CCS processes . As early as the LHT stage , we observed that conduction propagates unidirectionally across the myocardium , suggesting the presence of SA node pacemaker activity [34] . Despite chamber specification , no significant conduction delay was detected at this stage . However , increased conduction velocity was observed in the ventricular half of the heart , suggesting that these cardiomyocytes have initiated cellular and molecular changes including the acquisition of relatively faster conducting properties . Interestingly , expression of natriuretic peptide precursor a , a marker of the OC and faster conduction velocities , recently has been observed in a subdomain of the ventricular portion of the LHT near the OFT [19] . The AV conduction delay was observed at 36–48 hpf , a time corresponding to the formation of the AV canal . Action potential and calcium transients of atrial , ventricular , and AV canal regions revealed electrophysiologic differences , suggesting that AV myocardial cells are distinct from atrial and ventricular cells . Whether these differences are due to unique channel expression versus distinct combinatorial effects of several cardiac channels remains unclear . However , weak expression of Connexin40 , a marker for fast conduction , was observed at 48 hpf within the atrium and ventricle but lacking from the AV canal . In addition , the AV cardiomyocytes positioned themselves circumferentially around the AV canal . This ring-like orientation of cardiomyocytes has been suggested to contribute to the AV conduction delay [35] . Previous findings suggest that AV endocardial-specific signals , Neuregulin and Notch1b , cause the overlying AV myocardium to differentiate into slow conducting myocardium [5] . Additionally , Tbx3 , a transcriptional repressor expressed in the developing conduction system in mouse [36] , has been observed in the AV myocardium of 48 hpf zebrafish hearts [37] . Future studies exploring how these myocardial and endocardial factors may impact AV cardiomyocyte morphology , orientation , and action potential properties will be of great interest toward understanding the development of the AV conduction delay . Overall , our findings support studies suggesting that AV cardiomyocytes actively differentiate to become precursors of the AV node [5 , 20 , 38] . Despite lacking a distinct interventricular septum , adult zebrafish hearts possess a functionally fast CCS , resulting in an apex-to-base conduction across the ventricular myocardium [39] . Previous studies in mouse and chick have suggested that ventricular trabeculae and Cx40 may be responsible for an apex-to-base conduction prior to ventricular septation [6 , 21 , 40] . Consistent with these results , the rapid cardiac conduction network within the zebrafish ventricle develops as early as 96 hpf , a stage when cardiac looping is completed , trabeculae have started forming , and Cx40 expression is present . Loss of blood flow and cardiac contraction prevented the transition of ventricular conduction from a primitive linear pattern to a mature apex-to-base propagation , as previously observed in chick [6] . The absence of this transformation correlated with the lack of trabeculae and Cx40 immunoreactivity . Furthermore , we observed that sih/tnnt2 mutants as well as weak cardiac contractility mutants also developed intermittent AV heart block below the AV canal , suggesting that this defect is within the proximal ventricular portion of the fast CCS rather than the slow AV conduction pathway . How these epigenetic factors modulate the development of the fast CCS remains unknown . However , similar to findings in chick , endocardial signals may be involved , as cloche mutant hearts also fail to develop trabeculae and a fast cardiac conduction network ( unpublished data ) . Finally , a mature apex-to-base ventricular conduction was observed at 21 days postfertilization ( dpf ) when the hearts develop a distinct apex with increased ventricular trabeculation . Altogether , these data suggest that there is an intermediate step during the development of the mature ventricular apex-to-base conduction prior to the formation of the apex . Overall , these cellular and electrophysiologic data provide the framework for future studies in zebrafish regarding the origin of the vertebrate CCS . Through lineage tracing and optical mapping studies , one should be able to address recent observations related to the ontogeny of the CCS from progenitors that also give rise to chambered myocardium [41] . Previously , a candidate gene approach in zebrafish was performed to identify AV endocardial factors , neuregulin and notch1b , as regulators of AV conduction development [5] . In contrast , we have taken an unbiased approach and performed a new physiology-based forward genetic screen to identify additional mediators of the CCS . The recovered mutants display a wide spectrum of conduction phenotypes that first appear at distinct developmental stages of the CCS ( Table 1 and Figure 9 ) . Positional cloning reveals that hob , an AV conduction block mutation , affects the homeobox transcription factor , Tcf2 . Given the increased prevalence of AV block in humans with type II diabetes mellitus independent of congestive heart failure and coronary artery disease [42] , Tcf2 , a mature onset diabetes of the young/diabetes-associated gene , may also directly mediate the development and maintenance of mammalian AV conduction . Interestingly , Tcf2 is known to regulate expression of Na+/K+-ATPase [43] , an ion channel gene necessary for cardiomyocyte electrical polarization . Cardiac glycosides , such as digitalis , which inhibit the Na+/K+-ATPase pump , also can lead to human AV conduction heart block [44] . Additionally , mouse Tcf2 in situ hybridization and real-time PCR analyses reveal its expression in the mammalian heart ( Bussen M , Martin G , unpublished data ) . Thus , this mutant may help to address primary cardiac risk factors for a form of hereditary diabetes from indirect effects of diabetes on heart function . Future analysis and isolation of the genes affected by these mutations promise to uncover new molecular clues and mechanisms that underlie both genetic and acquired forms of heart disease in humans . Novel genes discovered from this screen may be potential candidates for sequencing in population genetic studies of human cardiac arrhythmias . Conversely , candidate genes identified from human genome-wide association studies for cardiac arrhythmias may be validated rapidly and characterized by employing this in vivo optical mapping technique with MO knockdown experiments . Together , these studies will help develop therapeutic options aimed at maintaining and/or improving overall cardiac conduction .
Zebrafish were raised under standard laboratory conditions at 28 °C . We used the following transgenic lines: Tg ( flk1:EGFP ) s843 [22] and Tg ( cmlc2:eGFP-ras ) s883 ( Jungblut B , Munson C , Huisken J , Trinh L , Stainier D , unpublished data ) . We generated the Tg ( cmlc2:gCaMP ) s878 construct by cloning a 900 bp fragment of the cmlc2 promoter [17] upstream of a promoter-less gCaMP construct [16] . We injected 200 pg of linearized DNA into one-cell-stage embryos and selected individual transgenic carrier adults by screening for fluorescent progeny . Six Tg ( cmlc2:gCaMP ) founders were recovered with identical expression patterns and levels . Homozygous mutant embryos were obtained by incrossing Tg ( flk1:EGFP ) s84/+; sih/+ , Tg ( cmlc2:gCaMP ) s87/+; sih/+ [15] , and Tg ( cmlc2:gCaMP ) s87/+; conduction mutants double heterozygotes . Immunohistochemistry and confocal microscopy were performed as previously described [28 , 45 , 46] . The following antibodies were used at the following dilutions: rabbit polyclonal anti-Cx43 ( Sigma ) at 1:100 and rabbit polyclonal anti-Cx40 ( Sigma ) at 1:100 . Cardiomyocyte surfaces and cross-sections were analyzed using ImageJ software ( National Institutes of Health ( NIH ) , http://rsb . info . hin . gov/ij/ ) as described previously [19] . A total of 288 cardiomyocytes were measured from 15 wild-type embryos: 165 OC , 83 IC , and 40 AV canal cardioymyocytes . Circularity measurements discriminate circular cells from elliptical cells . In situ analysis was performed as described [47] . We screened approximately 9 , 076 F3 clutches from 2 , 392 ENU-mutagenized F2 families that were generated in the context of two different screens [28 , 48–51] . On the basis of the number of crosses per F2 family , we estimate that the screens surveyed 2 , 723 genomes . The specific locus test for each screen indicated a mutation rate of approximately 0 . 3% per gene per genome . Utilizing a set of simple sequence length polymorphism markers , we mapped hobs634 to linkage group 5 . Fine mapping with 1 , 164 mutant embryos narrowed the critical region to two bacterial artificial chromosomes: CHORI-211 77I17 and DanioKey pilot 118E10 . This region contained two putative open reading frames , tcf2/vhnf-1 and gamma-synergin . The s634 cDNA was isolated , sequenced , and analyzed for both genes . A point mutation was discovered in tcf2 and confirmed by sequencing s634 genomic DNA . Approximately 0 . 5–1 ng of an ATG MO against tcf2 ( Gene-Tools ) , 5′-CTAGAGAGGGAAATGCGGTATTGTG-3′ , was injected into one-cell-stage embryos . For mRNA rescue experiments , one-cell-stage embryos were injected with 50–100 pg of tcf2 mRNA . Individual zebrafish between 24 hpf and 21 dpf were placed on a coverglass . Electromechanical isolation was achieved with 10 mM 2 , 3-butanedione monoxime ( Sigma ) applied 15 min before imaging . Single-plane widefield epifluorescence images of the heart were obtained with a Nikon TE-2000 inverted microscope , using 20× and 40× Plan Apo air objectives , an Xcite-120 ( Exfo ) widefield epifluorescent source , and standard fluorescein isothiocyanate filter set . Images were acquired with a Coolsnap HQ camera ( Photometrics ) using Metavue software ( Molecular Devices ) in stream acquisition mode at a frame rate of 30 ms/frame ( 512 × 512 pixels ) for 48 hpf hearts and 15 ms/frame ( 256 × 256 pixels ) for 96 hpf hearts . Image processing first consisted of manual adjustment of minor spatial shifts of the image over a temporal imaging series . Then , the fluorescence intensity of each pixel in a 2-D map was normalized to its percentage between the minimum and the maximum recorded values of the pixel over the full series . Isochronal lines at 60 ms intervals were obtained by identifying the maximal spatial gradient for a given time point . The color-coded scheme in each panel and video describes progressive activation of the heart with white/red cells and black/blue cells indicating depolarization and repolarization , respectively . Software processing was performed with Metavue software and procedures written in Matlab ( Mathworks ) . Videos of the cardiac conduction wave were recorded with selective plane illumination microscopy [52] . The attenuated 488-nm laser line from a diode-pumped solid-state laser ( Coherent Sapphire , 30 mW ) was focused to a light sheet with a thickness of 6 μm . The sample was oriented such that a thin slice of atrium , AV canal , and ventricle was illuminated . The fluorescence was collected at 90 frames per second with a 20×/0 . 5 objective lens ( Leica ) and an emission filter ( Chroma , HQ 525/50m ) and imaged on an EM-CCD camera ( Andor DV885 ) . The microscope and camera were controlled with a Labview ( National Instruments ) program and analyzed with Matlab . In each sequence , several 15 × 15 pixel areas ( i . e . 12 × 12 μm ) were selected , and the intensity in these areas was plotted over time . The 48 hpf embryos were dechorionated and anesthetized with 0 . 02% tricaine for 1–2 min . The heart was dissected from the thorax en bloc , and all experiments were performed at room temperature . The recording chamber was perfused with the following solution containing ( in mM ) 140 NaCl , 4 KCl , 1 . 8 CaCl2 , 1 MgCl2 , 10 glucose , and 10 HEPES , pH 7 . 4 . Suction pipettes were made from borosilicate capillary tubes ( 8250 glass , A-M Systems ) and fire-polished to obtain resistances of 6–9 MΩ when filled with ( in mM ) 120 KCl , 5 EGTA , 5 K2ATP , 5 MgCl2 , and 10 HEPES , pH 7 . 2 . Transmembrane potential was measured using an Axoclamp 2A amplifier ( Molecular Devices ) in the bridge mode using the disrupted patch technique . The pipette was positioned adjacent to the heart , and a seal was formed by application of minimal suction . Through the use of this technique , stable spontaneous APs were recordable for up to 2 h . Transmembrane voltage was filtered at 10 kHz and digitized at a sampling frequency of 20 kHz with a 12-bit analog-to-digital converter ( Digidata 1322A Interface , Molecular Devices ) .
|
Aberrant electrical activity of the heart , otherwise known as cardiac arrhythmia , may disrupt heart contractions , leading to loss of consciousness and sudden death . Every year , approximately 450 , 000 individuals in the United States die suddenly from this event . Currently , the only proven preventive therapy for sudden cardiac death is the automatic implantable cardioverter defibrillator , which carries a significant burden and cost to the patient . Greater understanding of the cardiac conduction system , which coordinates rhythmic beating of the heart , may lead to novel and safer therapeutic options for these patients . Working with zebrafish , a productive model system for understanding human disease , we have developed a cardiac-specific fluorescent calcium indicator zebrafish transgenic line to analyze the formation of the cardiac conduction system . Using this fluorescent transgenic line , we have observed four distinct physiologic cardiac conduction stages that correspond to cellular and anatomic changes of the developing heart . Furthermore , we have designed and performed a new , physiology-based , forward genetic screen to identify cardiac conduction mutants that would have escaped discovery in previous screens . Overall , these studies may prove rewarding toward developing therapeutic options aimed at maintaining and/or improving overall cardiac health .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"developmental",
"biology",
"cell",
"biology",
"genetics",
"and",
"genomics",
"cardiovascular",
"disorders"
] |
2008
|
Genetic and Physiologic Dissection of the Vertebrate Cardiac Conduction System
|
Estimation of liver function is important to monitor progression of chronic liver disease ( CLD ) . A promising method is magnetic resonance imaging ( MRI ) combined with gadoxetate , a liver-specific contrast agent . For this method , we have previously developed a model for an average healthy human . Herein , we extended this model , by combining it with a patient-specific non-linear mixed-effects modeling framework . We validated the model by recruiting 100 patients with CLD of varying severity and etiologies . The model explained all MRI data and adequately predicted both timepoints saved for validation and gadoxetate concentrations in both plasma and biopsies . The validated model provides a new and deeper look into how the mechanisms of liver function vary across a wide variety of liver diseases . The basic mechanisms remain the same , but increasing fibrosis reduces uptake and increases excretion of gadoxetate . These mechanisms are shared across many liver functions and can now be estimated from standard clinical images .
Measurements of liver function are important to determine the optimal therapeutic strategy in cases of severe chronic liver disease ( CLD ) , and for prevention of post-treatment hepatic failure [1] . Estimating liver function is also important when planning surgical treatment , because postoperative hepatic function insufficiency is associated with both morbidity and mortality [2] . Sensitive biomarkers for liver function would also be useful for the management and early identification of drug-induced liver injury ( DILI ) , which is a leading cause of acute liver failure [3] and also of drugs being withdrawn from the market [4] . Different options for estimation of liver function are used clinically today , but they all have some shortcomings . For instance , the primary clinical screening tool for liver injury in clinical trials , serum alanine aminotransferase ( ALT ) , neither indicates the severity of liver injury nor estimates liver function [5] . In addition , ALT ( and other transaminases ) only indicates injury at a late stage when substantial tissue damage has already occurred [6] . Alternative methods for measuring liver function include Indocyanine-Green 15 retention rate ( ICGR15 ) [7] and Tc-99m galactosyl human serum albumin ( GSA ) [8] , which both measure the liver´s capacity to clear substances from the blood , and the galactose breath test [9] , which measures the liver’s metabolic capacity . These are all examples of global indicators that provide indirect measurements of liver function . Furthermore , GSA involves the injection of a radioactive isotope , which from a practical point of view is cumbersome and suffers from limited spatial and temporal resolution , and importantly is not widely available . In summary , biomarkers that are sensitive and respond early to changes in liver function would be beneficial both in a clinical setting as well as in the pharmaceutical industry and regulatory agencies [10 , 11] . Because of the low quality of available measures of liver function , little is known about the more detailed mechanisms of liver function , and about how these mechanisms change at different stages of CLD . One of the most promising state-of-the-art methods for assessing clearance-based liver function is to use magnetic resonance imaging ( MRI ) in combination with the liver-specific contrast agent gadoxetate ( Bayer Schering Pharma , Berlin , Germany ) . This method has the potential to allow investigation of liver function at a regional level without the need for any ionizing radiation . As a liver-specific contrast agent , gadoxetate is actively accumulated within hepatocytes [12] and is commonly used for characterizing lesions . The gadoxetate uptake is mainly associated with the function of the organic anion-transporting polypeptide 1 ( OATP1 ) family of transporters [13] . The subsequent excretion into the bile occurs via the multidrug resistance-associated protein 2 ( MRP2 ) transporter [14] . These transporter proteins have important functions , such as mediating the clearance of bilirubin , toxins , drugs , and other organic solutes [15] . For these reasons , gadoxetate MRI has the potential to facilitate study of the aspect of liver function that has to do with these uptake and excretion processes . There are a number of previous studies which indicate the high potential of gadoxetate MRI as a biomarker for liver function . Early studies established a correlation between gadoxetate MRI and common clinical markers for liver function [16] . A more recent study demonstrates the ability of gadoxetate MRI to predict liver failure after surgery [17] . Furthermore , a recent prospective follow-up study on patients with primary sclerosing cholangitis showed that quantitative gadoxetate MRI could predict solid clinical endpoints , such as liver transplantation , cholangiocarcinoma , and liver related death [18] . The approach used in that study separated the population into two clear groups , one with >90% survival and one with <60% survival . Finally , in rats , similar analyses have shown promising results of using gadoxetate MRI as a biomarker for DILI [19 , 20] . All the above clinical studies have used a simple analysis , called relative enhancement , which simply compares signal intensity before and after gadoxetate injection; therefore , the studies could not elucidate the detailed mechanisms of liver function . More advanced approaches can make use of the information in an entire time series of images and use this to extract different gadoxetate transport rates . Such methods require the use of mathematical models . One common class of such models is the perfusion-based model [21 , 22] . These models require images with a very high temporal resolution , which limits the spatial resolution , meaning that the images cannot be used for conventional radiological reading . An alternative to perfusion-based models is models based on simulation and optimization of ordinary differential equations . These models do not need such high temporal resolution , but can utilize the high-spatial low-temporal resolution images used in clinical MRI protocols today . One such model , describing how gadoxetate is distributed in the whole body , as well as taken up and excreted by the liver , was described by Forsgren and colleagues [23] . While the Forsgren model can arguably be viewed as the most realistic gadoxetate uptake model , it has several shortcomings . The model is the most realistic in the sense that it is the only compartment model to describe the dynamics in liver , blood , spleen , and extracellular extravascular space . Furthermore , the model has been validated in healthy humans with gadoxetate doses up to 20 times higher than the clinical dose used for model training . On the other hand , the model has not been personalized . One effective approach used for such personalization is non-linear mixed-effects modeling ( NLME; Fig 1A and 1B ) . NLME is effective because it can deal with low-informative data [24] , which could allow for fewer images , shorter clinical examinations , and more reliable parameters . However , NLME has not been applied to any gadoxetate uptake model . Another shortcoming with the Forsgren model is that it has not been tested in patients with liver diseases . Therefore , it is not known how the mechanisms in the model vary across different stages and etiologies of CLD . Furthermore , the model has not been validated with respect to other important independent measures such as biopsies and post-procedural blood samples . These limitations have been due to the lack of relevant data . In this study , a new modeling framework is created ( Fig 1C ) by combining i ) state-of-the-art MRI processing of high resolution gadoxetate-enhanced time series [25 , 26]; ii ) the mechanistic gadoxetate uptake model , [23]; and iii ) NLME model parametrization methods . To validate the model , a large clinical study was conducted by recruiting 100 patients , who were subjected to a variety of different measurements . These new data validate the model in three different ways . First , the extended model can describe patient variation across all stages of CLD . Second , the model can predict quantified images from later time points , which were not included in the estimation data; this implies the possibility of a shorter clinical protocol . Third , the model can predict independent validation data from blood samples and biopsies . Finally , it is demonstrated how the estimated model properties , such as OAPT1 and MRP2 transport rates , change with varying severity of CLD . These results point to a new avenue for estimation of liver function .
A total of 100 patients , with suspected CLD were included in the study and each underwent an MRI examination followed by two liver biopsies . Of these , eight patients were excluded because they aborted the examination and one patient was excluded due to poor data quality , giving a final cohort of 91 patients . The demographic characteristics and clinical diagnoses of the final study population are presented in Table 1 . None of the included patients showed signs of hepatic decompensation . All patients were given an injection of gadoxetate and several MR-images were acquired over a period of 30 minutes for each patient . Time series for each patient were made by quantifying the change in R1 relaxation rate ( ΔR1 ) , which is proportional to gadoxetate concentration [27] , in the liver and spleen . These time series were used to parametrize the mechanistic model ( Fig 1D ) for each individual patient using the NLME method . That is , the NLME algorithm was used to identify optimal model parameter values for each patient ( e . g . , describing the function of OATP1 and MRP2 ) such that the model predictions of the MRI data in the liver and the spleen matched the measured MRI data . Fig 2A and 2B shows the observed values and the model predictions for two typical patients , one without fibrosis ( F0 ) and one with cirrhosis ( F4 ) . Goodness-of-fit for was assessed each patient . In all but one of the 91 patients , the model predicted the observed MRI data without being rejected by the goodness-of-fit test . This indicated that the same mechanisms were at play in all stages and etiologies of CLD , and that only the quantitative details were different . Fig 2C–2F shows a comparison between the gadolinium concentrations ( the paramagnetic nucleus responsible for the contrast enhancement in gadoxetate ) measured in the blood samples and biopsies and the gadolinium concentrations predicted by the mathematical model . At a group level , there were no differences between the gadolinium concentrations predicted by the mathematical model and the concentrations measured using inductively coupled plasma sector field mass spectrometry ( ICP-SFMS ) ( Fig 2C and 2E ) . At an individual level , there was a moderate Lin’s concordance correlation between the predicted and measured gadoxetate concentrations in blood samples ( rc = 0 . 62; Fig 2D ) . However , there was only a low correlation in liver biopsy samples ( rc = 0 . 31; Fig 2F ) . To assess whether the NLME-model parameterization method outperforms the standard two-stage approach ( STS ) , and whether it is possible to reduce the examination time to 10 min from gadoxetate injection , the dataset for each patient was divided into two parts: all data points from within 10 min after gadoxetate injection were used as estimation data , and the validation data included the remaining later time points . Both the NLME and STS parametrization methods were used to parametrize the model with the estimation data . The resulting model predictions were compared to the estimation data and validation data , and goodness-of-fit was assessed for each patient individually . The NLME parameterization was implemented using a 'leave-one-out' design , meaning that one data set at a time was truncated while all other data sets were complete . This design was to demonstrate how NLME could be used in a clinical situation where the distributions of the parameters have already been determined in clinical studies . Four patients had insufficient data for this analysis ( there was no available data after 10 min ) , so the test included data from 87 patients . Both the NLME and STS methods produced model predictions that passed the statistical test for goodness-of-fit for the estimation data in all 87 patients . The NLME method predicted the data in the validation dataset in 81 patients ( 93% ) without being rejected , compared with 37 patients ( 43% ) with the STS method . Fig 3A and 3B shows an example of a patient for whom the NLME method could predict the validation data , while the STS method failed , particularly when predicting the liver signal . Significant Lin’s concordance correlation was observed between predicted and measured blood plasma gadoxetate concentrations when using the NLME parametrization to data from the first 10 min ( rc = 0 . 60 ) . Notably , there was no significant difference between the predicted and measured blood plasma concentrations with prediction based on parametrization to data from the first 10 min . Correlation between predicted and measured blood plasma gadoxetate concentrations based on the STS parametrization was not as strong as with the NLME parametrization ( rc = 0 . 24 ) . The array of model parameters for each patient , estimated by the NLME parametrization method using data from the first 10 min , were compared with the same parameters estimated by the NLME parametrization method using the full dataset ( Fig 3C–3H ) . The OATP function , i . e . hepatocyte uptake rate ( kph; Fig 3C ) , was unaffected by the sparse ( ≤10 min ) estimation with a highly significant correlation between sparse and full estimations ( Fig 3D ) . At a group level , the MRP2 function , i . e . hepatocyte elimination rate ( khb; Fig 3E ) , was unaffected by the sparse estimation data . However , the individual values for each patient were affected and hence the correlation was poor ( Fig 3F ) . The hepatocyte to plasma flux ( khp; Fig 3G ) was significantly affected by the sparse estimation data , with a non-significant correlation ( Fig 3H ) . In the liver , ΔR1 is lower in patients with increased fibrosis stage ( Fig 4A ) . In the spleen , ΔR1 appears to be unaffected ( Fig 4B ) . Furthermore , the hepatocyte uptake rate of gadoxetate by OATP1 , differentiated between fibrosis stages , is shown in Fig 4C . The figure shows that the uptake is decreased in patients with advanced fibrosis and cirrhosis . Furthermore , the hepatocyte excretion rates were differentiated between cirrhosis and both advanced fibrosis and mild fibrosis ( Fig 4D ) . Finally , Table 2 shows a confusion matrix of the ability of kph to identify patients with advanced fibrosis , i . e . ≥F3 , when using a cut-off of 0 . 00198 s-1 .
A new next-generation framework to measure liver function using MRI was developed . This framework was successfully applied and validated with liver biopsy and blood samples in a clinical setting in a diverse cohort . More specifically , the model could describe data from all patients and it was able to adequately predict gadoxetate levels in both blood plasma and biopsies . Furthermore , the introduced NLME method for parameter estimation is more robust on shorter protocols , compared to the previously used STS method; this allows for shorter examinations . Finally , the validated model allowed for the examination of how the biomedical mechanisms for clearance-based liver function vary across different stages of CLD . The new framework is validated in several different ways . First , the model was validated by the fact that it could be used to extrapolate data points not used for parameter estimation ( example in Fig 3A and 3B ) . Second , the model could also be fitted to data from patients with a wide variety of different chronic liver diseases ( Table 1 ) . Third , the concentrations of gadoxetate in both liver biopsy and blood samples were measured by ICP-SFMS ( Fig 2C–2F ) . On a group level , there was no significant difference between the predicted and measured gadoxetate concentrations . On an individual level , there was a moderate correlation in the blood , while there was a low individual correlation in the biopsy . This lower correlation may be due to contamination of the biopsy samples from gadoxetate in the bile ducts . These correlations do not necessarily mean that our modeling framework should be used for individual predictions of gadoxetate concentrations . However , the results do show that our modeling framework produces realistic parameter values . Last , the framework was also validated by the fact that the model parameters corresponding to OATP1 and MRP2 functions varied as expected in the patient population ( Fig 4 ) . The variation of the parameters across the patient population requires some additional remarks . First , the population covered a wide spectrum of both etiologies ( Table 1 ) and severity and we used liver fibrosis to indicate severity . Second , hepatic uptake via OATP1 transporters ( kph ) decreased significantly with increasing levels of fibrosis . Similar results were previously obtained in studies with perfusion-style model frameworks [28 , 29] . Possible reasons for reduction of OATP1 function include restricted access of gadoxetate to the hepatocytes , reduction in the number of functional hepatocytes , and competitive inhibition . Third , with respect to hepatic excretion via MRP2 ( khb ) , a significantly higher excretion rate was estimated in patients with cirrhosis , compared to patients with lower levels of fibrosis . Other studies have reported mixed results . Previously , a small study using a perfusion model indicated the opposite , that gadoxetate excretion decrease in cirrhotic humans [30] . Therefore , it is interesting to look at studies of gene expression . Some studies on cirrhotic rats have shown an upregulation of MRP2 [31–33] , which is consistent with our findings . In contrast , one study found a lower expression of MRP2 in rats with fibrosis [34] . In humans , the picture is also mixed and CLD has been found to be associated with either no difference [35] , a slight increase [36] , or in some CLD etiologies , a decrease [37] in MRP2 expression . By using the NLME parameterization scheme , the time needed for MRI-examinations could be reduced , while still being able to estimate reliable parameters , as well as predicting both the liver and spleen signals ( Fig 3 ) . This reduction was accomplished because NLME allows for information to be shared among the parameter estimations of all patients , thus requiring fewer new datapoints per patient . This reduction in the examination time is beneficial , since it reduces cost and patient discomfort , and requiring only a few images also allows for our method to be included for liver function evaluation in short abbreviated MRI-protocols . Such protocols , ( sometimes called AMRI ) are gaining popularity , e . g . when screening cirrhotic patients for hepatocellular carcinoma [38 , 39] . It can be noted that while the STS scheme failed to predict the liver signal , STS was still able to predict the spleen signal . The reason for this is that almost all dynamic information of the spleen signal is contained within the first ten minutes . Removing all later time points should therefore not affect the ability to predict the spleen signal . Furthermore , it can also be noted that while the χ2-test was used to evaluate the goodness-of-fit of the model , the NLME framework offers other methods for assessing model performance , such as visual predictive check and normal predictive distribution errors . However , since NLME was only used to increase the robustness of the predictions of the individual patients , the χ2-test should be enough . Comparing different methods would be interesting , but was unfortunately beyond the scope of this work . Another strength of this work is that data are presented from a new clinical study , where 100 patients were recruited . The patients were selected to represent the actual flow of patients being referred to a hepatology department , with a normal variation in both disease etiology and severity . This gives a more realistic picture of the clinical situation , as most other studies have either been small or not prospective . Additionally , the study included dual biopsies , as well as blood samples , from the patients , in conjunction with the MR examination . These rare measurements allowed for extensively validate the model . Lastly , the study was conducted over a span of around six years . Such a long time could be seen as a limitation , as changes occur to an MR-system over time , such as software upgrades . On the other hand , this could also be seen as a strength of the method , since it was found that all data could easily be analysed in the same framework . Although this methodology is still in the research phase , the methodology is better suited for clinical implementation , compared to other similar methods , for a variety of reasons . First , the modeling framework uses the same type of clinical images , already collected in routine examinations . Therefore , the liver function estimation can easily be included in clinical workflows or studies that already use gadoxetate MRI , by simply adding a few more breath holds . Second , the model is based on simulations of ordinary differential equations , which has additional advantages . For instance , the model , unlike previous non-simulation-based models [19 , 21 , 22] , can easily be combined with other models describing detailed processes in the liver , and thus can possibly characterize other aspects of liver function , such as metabolic aspects [40] . Third , the simulation-driven model can also be combined with more zoomed out whole-body models . The result of such combinations is multi-level models which can simultaneously describe multiple organs and processes in the body [41–43] . For all these reasons , the framework could be further extended and reused in a variety of different contexts , both regarding clinical implementation and research . In conclusion , this study presented a new integrative MRI-based framework for estimating liver function . The extendable framework has been validated in a variety of ways and has allowed for a new and deeper look into the variation of mechanistic parameters across a clinically relevant cohort .
Between 2007 and 2014 , 100 patients were recruited on referral to the Linköping University Hospital , Linköping , Sweden for evaluation of chronic ( > 6 months ) elevation of levels of one or more of ALT ( >1 . 10 μkat/L for men and >0 . 75 μkat /L for women ) , aspartate aminotransferase ( AST; >0 . 75 μkat /L for men and >0 . 60 μkat /L for women ) , and serum alkaline phosphatase ( ALP; >1 . 80 μkat /L regardless of gender ) . All patients who , on clinical indication or as part of a clinical study , needed a liver biopsy for histopathological evaluation were asked to participate in the study . Exclusion criteria included contraindications for MRI ( presence of pacemaker devices , implants with ferromagnetic properties , pregnancy , and claustrophobia ) and liver biopsy ( presence of primary or secondary coagulative disorder , prothrombin time > 1 . 5 times the international normalized ratio , platelet count <50×109 /L , hepatic malignancy , and clinical signs of decompensated cirrhosis ) . A diagnostic work-up was performed prior to MRI , including a physical examination , laboratory investigations , and ultrasonography . The pathologist was blinded to the results of the diagnostic work-up , and the radiologists were blinded to the diagnosis and the pathology findings . This study was approved by the regional ethics committee ( Reference No . M72-07 T5-08 ) . All patients gave informed consent to participate before the inclusion . MRI was performed within two months of the diagnostic work-up with a Philips Achieva 1 . 5 T MR scanner ( Philips Healthcare , Best , Netherlands ) and a phased-array body coil . Single-breath-hold symmetrically sampled T1-weighted gradient-echo two-point Dixon 3D images [44] were acquired using sensitivity encoding [45] . All patients received a bolus injection of gadoxetate ( gadolinium ethoxybenzyl diethylenetriamine pentaacetic acid , or Gd-EOB-DTPA , marketed as Primovist in Europe and Eovist in the USA by Bayer Schering Pharma , Berlin , Germany ) , at a dose of 0 . 1 ml/kg , administered intravenously at a rate of 1 mL/s by a power injector ( Medrad Spectris Solaris , Pittsburgh , PA , USA ) , followed by a 30 mL saline flush . Image time series were acquired prior to ( non-enhanced ) and directly following gadoxetate injection ( Fig 5 ) . The post-injection time series corresponded to the arterial and portal-venous phases , as well as 3 min , 10 min , 20 min , and 30 min following injection . Additional acquisitions between 3 min and 30 min were added from 2012 and onwards . The field of view ( FOV ) and acquisition matrix were adjusted to accommodate patients of different sizes . Higher temporal resolution was used during the initial contrast agent wash-in phase , the arterial phase . The non-enhanced and post-injection images were acquired using the following sequence parameters: repetition time = 6 . 5 ms , echo time = 2 . 3 ms and 4 . 6 ms , flip angle = 13° , typical acquisition matrix = 168×168 , typical FOV = 261 mm by 200 mm by 342 mm , slice thickness = 4 mm . We used interpolation , with zero padding in the z-direction , and up-interpolated from 4 to 2 mm . The acquired in-phase and opposite-phase images were reconstructed into separate water and fat images by our previously developed inverse-gradient method [26] . The signal intensity of MR-images is not absolute . Hence , if images are acquired in a time series , the signal intensity in the images can vary , even though all images are acquired using the same parameters . We corrected for this variation by using voxels of pure adipose tissue as an internal reference throughout the time series . [25] This was an important step in the quantification process . To extract signal intensities ( SIs ) for the quantitative analysis , two clinical radiologists ( BN , ND; with more than ten years of experience in abdominal radiology ) placed ROIs in the reconstructed water-image time series , in the liver ( N = 7 ) , and spleen ( N = 3 ) . Liver ROIs were placed in both the left and right liver lobes to avoid any large vessels or focal lesions , but not strictly following the Couinaud segmental division . The sizes of ROIs were arbitrarily chosen by the radiologists . However , the ROIs were adjusted to be equal in size and approximate position throughout the time series . Landmarks in the images were used to correct for movement of the liver between the acquisitions . Fig 5 shows an example of ROI placement . Mean SIs in the ROIs were normalized and the relaxation rate ( R1 ) was calculated as described previously [46] . The induced change in the relaxation rate ( ΔR1 ) was directly proportional to gadoxetate concentrations [27] . For quality assurance , the image data were inspected visually for quality issues and potential data exclusion . As the two radiologists independently placed ROIs in the images , they took particular notice of cases of poor image quality ( such as artifacts resulting from breathing or post-processing failure ) . Then , both radiologists reviewed these cases of potential poor quality and reached a consensus about whether to exclude the images , to return the images for manual image reconstruction , or to accept the images . After the radiologists were satisfied , the data analyst continued to search for any significant outliers in the extracted time series . The radiologists were then instructed to review these latter outliers , but they were not told why each case was to be reviewed . If they were still satisfied with the placement of ROIs and with the image quality , nothing was corrected . The lower limit of uncertainty in the extracted gadoxetate time series was estimated by calculating the average standard deviation in the normalized signal intensities ( Eq 1 , where S = SI ( t ) /SI ( t = 0 ) ) . The uncertainty was then averaged over the entire study population; for this averaging , each entry was in turn the mean of each patient’s liver and spleen ROIs . Fig 6 shows a histogram of the estimate of the lower limit of uncertainty , and the fitted normal distribution had a mean of 0 . 18 . Unless the standard error of the mean across the spleen or liver ROIs exceeded 0 . 18 , this lower limit value was used as the standard deviation in the following statistical test for the mechanistic model goodness-of-fit . The whole-body model devised by Forsgren and co-workers [23] was used here to quantify the liver function . Fig 1C shows a schematic diagram of the model , which has two parts: a dynamic model and a signal model . Briefly , the dynamic model describes five separate fluxes of gadoxetate: between the blood plasma and the extracellular extravascular space ( EES; kdiff ) ; elimination via the kidneys to urine ( CLr ) ; uptake into the hepatocytes ( through the OATP1 family transport proteins; kph ) ; back-flux from the hepatocytes into the blood plasma ( through the transport protein MRP3; khp ) ; and excretion from the hepatocytes into the bile ( through the transport protein MRP2; khb ) : dChepdt=kphCPAlb−khpChep−khbChep , ( 2 ) dCpdt= ( khpChep−kphCPAlb ) Vlvh−CLrCPAlb+ ( kdiffCees−kdiffCPAlb ) Vees+uVp , ( 3 ) dCeesdt=kdiffCPAlb−kdiffCees , ( 4 ) where Chep , Cp , and Cees is the gadoxetate concentration in the hepatocytes , blood plasma , and EES respectively . Vl , Vees , and Vp are the volumes of the liver , EES , and blood plasma respectively ( assumed to be 1 . 43 , 14 . 77 and 2 . 57 L ) . Alb is the fraction of Gadoxetate not bound to serum albumin ( assumed to be 0 . 9 ) , vh is the volume fraction of hepatocytes in the liver ( assumed to be 0 . 68 ) , and u is the injection of gadoxetate . CLr is assumed to be 118 mL/min . The signal model was used to predict ΔR1 in the gadoxetate MRI time series as a function of the gadoxetate concentrations in the compartments . The model takes into account the parenchyma volume fractions as well as the in situ tissue-specific relaxivity properties of gadoxetate [23]: ΔR1 , l=ξ ( Chepvhr1 , hep+Cpvp , lr1 , p+Ceesvees , lr1 , ees ) , ( 5 ) ΔR1 , s=ξ ( Cpvp , sr1 , p+Ceesvees , sr1 , ees ) , ( 6 ) where ΔR1 , l and ΔR1 , s are the ΔR1 in the liver and spleen respectively , vp , l and vees , l are the volume fractions of plasma and EES in the liver ( assumed to be 0 . 12 and 0 . 20 ) , and vp , s and vees , s are the fractions of plasma and EES ( assumed to be 0 . 35 and 0 . 20 ) in the spleen . ξ is an arbitrary scaling parameter and r1 , hep , r1 , p , and r1 , ees are the tissue-specific relaxivities in the hepatocytes , blood plasma , and EES respectively ( assumed to be 10 . 7 , 7 . 3 , and 6 . 9 mmol-1s-1 ) . The model was parametrized separately using the STS method and the NLME method , described in Fig 1A and 1B . The STS parametrization was performed by minimizing the following costfunction , which follows a χ2-distribution: V ( p^ ) =∑ ( y^i ( p^ , t ) −yi ( t ) ) 2σ2i ( t ) ∈χ2 ( df ) , ( 7 ) where y and σ are the measurements and standard deviation of the measurements respectively , y^ is the predicted data as a function of time and the estimated model parameters ( p^ ) , and the index i indicates liver or spleen . When using NLME , all the optimized parameters have two parts , a fixed effect and a random effect . The fixed effect is the same across all patients and represents the typical parameter value . The random effect describes how each individual deviate from the typical value and is thus allowed to vary across the population , but is still constrained to a normal distribution: pj=θp+ηpj ( 8 ) pj=θpeηpj ( 9 ) where pj is a generic parameter for patient j , θp is the fixed effect for parameter p , and ηjp is the random effect for parameter p for patient j . If p is postulated to be normally distributed , Eq 8 is used , while Eq 9 is used if p is lognormally distributed . More details of the STS parametrization are described in [23] and the details of NLME in [24 , 47] . Population distributions in the NLME model parametrization were defined as a normal distribution for the scale parameter ( ξ ) and lognormal distributions for the four rate parameters ( kdiff , kph , khp , and khb ) . The distributions were a priori parametrized from the results of parametrizing the model to healthy human patients , which has been described previously ( Table 3 in [23] ) , where the expectation values were ξ = 1 . 6 , kdiff = 1 . 7 ms-1 , kph = 4 . 7 ms-1 , khp = 28 ms-1 , and khb = 38 ms-1 . The a priori standard deviations were chosen such that the optimization algorithm would not be unnecessarily limited ( ξ = 1 , kdiff = 0 . 1 ms-1 , kph = 0 . 1 ms-1 , khp = 0 . 01 ms-1 , and khb = 0 . 01 ms-1 ) . The mechanistic model framework assumes that the compartments are well mixed containers ( a fundamental property of ordinary differential equation models ) . In addition , there are interfering effects from the bolus injection during the arterial and portal-venous phases . Therefore , only data at the 3 min point and later after contrast injection were included in the model parameterization . Immediately following the MR examination , 3 mL venous blood ( collected in a 3 mL BD Vacutainer sterile hematology tube with K2-EDTA ) was drawn from each patient for elemental analysis of gadolinium content . Samples were transferred to 4 mL sterile low-temperature freezer vials ( VWR , Sweden ) for freezing and storage at -80°C . The frozen samples were sent to an external laboratory ( ALS Scandinavia AB , Luleå , Sweden ) for elemental analysis by ICP-SFMS: 0 . 20 mL from each thawed blood sample was mixed with 1 . 00 mL ‘super pure’ HNO3 ( pure with respect to traces of metal ) and digested in a 600 W microwave oven operating at 75% power for 30 min . Each of these mixtures was then diluted up to 10 . 00 mL with MilliQ ultrapure water for ICP-SFMS analysis , which had a detection limit for gadolinium of 0 . 05 μg/L . Immediately after completion of the MR examination and blood sampling , two ultrasonographically guided liver biopsy procedures were performed , on an outpatient basis . The biopsy samples were obtained percutaneously with a 1 . 6 mm BioPince needle ( BioPince Full Core Biopsy Instrument , Argon Medical Devices , Plano , TX , USA ) in either the left or right liver lobe depending on which location offered the best combination of a successful biopsy and maximum patient safety . A histopathologist graded and classified one of the biopsy samples according to the Batts and Ludwig system [48] , through which fibrosis was staged as no fibrosis ( F0 ) , portal and/or perisinusoidal fibrosis ( F1 ) , periportal and perisinusoidal fibrosis ( F2 ) , bridging fibrosis ( F3 ) , and probable or obvious cirrhosis ( F4 ) . The biopsies were also graded for inflammation . All biopsy samples were graded by the same histopathologist . The second biopsy sample of each pair was weighed and directly frozen at -80°C . The frozen samples were later freeze dried , and the dry weight was measured prior to submission to our external analysis partner ( ALS Scandinavia AB ) for elemental analysis . The dried samples were digested by adding 2 . 50 mL ‘super pure’ HNO3 and 0 . 25 mL H2O2 followed by a 30 min treatment at 170°C in a microwave oven . The samples were then diluted to 5 . 00 mL with MilliQ ultrapure water , for ICP-SFMS analysis . The goodness-of-fit of the model to the data was investigated on a subject basis using a χ2 test ( Eq 7; α = 0 . 05 ) with degrees of freedom equal to the number of observations in the gadoxetate-enhanced MRI time series . Group differences were investigated using an unpaired two-tailed Mann–Whitney U-test ( α = 0 . 05 ) . A paired two-tailed Mann–Whitney U-test was used when comparing two model parametrization method estimates of model parameters ( α = 0 . 05 ) . Linear regression and Lin’s concordance correlation were used to investigate correlation between variables that measure or describe similar entities . For correlation between non-similar variables , a Pearson correlation coefficient was calculated . An ANOVA with Tukey’s post-test was used to investigate sources of variation and biomarker performance between fibrosis stages .
|
Being able to accurately and reliably estimate liver function is important when monitoring the progression of patients with liver disease , as well as when identifying drug-induced liver injury during drug development . A promising method for quantifying liver function is to use magnetic resonance imaging combined with gadoxetate . Gadoxetate is a liver-specific contrast agent , which is taken up by the hepatocytes and excreted into the bile . We have previously developed a mechanistic model for gadoxetate dynamics using averaged data from healthy volunteers . In this work , we extended our model with a non-linear mixed-effects modeling framework to give patient-specific estimates of the gadoxetate transport-rates . We validated the model by recruiting 100 patients with liver disease , covering a range of severity and etiologies . All patients underwent an MRI-examination and provided both blood and liver biopsies . Our validated model provides a new and deeper look into how the mechanisms of liver function varies across a wide variety of liver diseases . The basic mechanisms remain the same , but increasing fibrosis reduces uptake and increases excretion of gadoxetate .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"medicine",
"and",
"health",
"sciences",
"liver",
"immune",
"physiology",
"diagnostic",
"radiology",
"body",
"fluids",
"biopsy",
"spleen",
"fibrosis",
"liver",
"diseases",
"surgical",
"and",
"invasive",
"medical",
"procedures",
"magnetic",
"resonance",
"imaging",
"physiological",
"processes",
"developmental",
"biology",
"gastroenterology",
"and",
"hepatology",
"research",
"and",
"analysis",
"methods",
"imaging",
"techniques",
"animal",
"cells",
"hepatocytes",
"blood",
"plasma",
"radiology",
"and",
"imaging",
"diagnostic",
"medicine",
"blood",
"cell",
"biology",
"anatomy",
"physiology",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"excretion"
] |
2019
|
Model-inferred mechanisms of liver function from magnetic resonance imaging data: Validation and variation across a clinically relevant cohort
|
Glioblastoma multiforme ( GBM ) is the most aggressive type of brain tumor . Molecular heterogeneity is a hallmark of GBM tumors that is a barrier in developing treatment strategies . In this study , we used the nonsynonymous mutations of GBM tumors deposited in The Cancer Genome Atlas ( TCGA ) and applied a systems level approach based on biophysical characteristics of mutations and their organization in patient-specific subnetworks to reduce inter-patient heterogeneity and to gain potential clinically relevant insights . Approximately 10% of the mutations are located in “patches” which are defined as the set of residues spatially in close proximity that are mutated across multiple patients . Grouping mutations as 3D patches reduces the heterogeneity across patients . There are multiple patches that are relatively small in oncogenes , whereas there are a small number of very large patches in tumor suppressors . Additionally , different patches in the same protein are often located at different domains that can mediate different functions . We stratified the patients into five groups based on their potentially affected pathways , revealed from the patient-specific subnetworks . These subnetworks were constructed by integrating mutation profiles of the patients with the interactome data . Network-guided clustering showed significant association between each group and patient survival ( P-value = 0 . 0408 ) . Also , each group carries a set of signature 3D mutation patches that affect predominant pathways . We integrated drug sensitivity data of GBM cell lines with the mutation patches and the patient groups to analyze the therapeutic outcome of these patches . We found that Pazopanib might be effective in Group 3 by targeting CSF1R . Additionally , inhibiting ATM that is a mediator of PTEN phosphorylation may be ineffective in Group 2 . We believe that from mutations to networks and eventually to clinical and therapeutic data , this study provides a novel perspective in the network-guided precision medicine .
Cancer mostly occurs when somatic mutations accumulate and eventually change the behavior , structure and properties of the cell . Understanding which mutations cause cancer is of crucial importance . The large-scale cancer genome sequencing projects including The Cancer Genome Atlas ( TCGA ) [1] , the International Cancer Genome Consortium ( ICGC ) [2] , and smaller-scale gene/protein focused and genome-wide screenings have enabled us to explore a large volume of somatic mutations in human cancers . Heterogeneity in mutation profiles between and within tumors as well as among individuals of the same type of cancer is enormous . However , not every somatic mutation affects pathways involved in cancer . Mutations are conventionally divided into driver and passenger mutations based on their function in providing positive growth advantages to cancer cells . The main challenge is to discriminate the drivers from passengers . Lately , another class of mutations were defined which is called “latent” or “mini-driver” [3] . Latent mutations have a potential to behave like a driver or are not yet discovered to be as drivers . Although latent mutations are not significant mutations , they can be triggered to become driver mutations by the environmental factors or conformational changes in proteins . Ultimately , proteins of the driver genes are the favored molecular targets in drug discovery and cancer therapy . Also , having insights about the accumulation of mutations and their impact at the pathway level is equally important to understand the causes and mechanisms of cancer development and progression . All these together with epigenetic and post-translational factors determine the risk of cancer progression and the therapeutic resistance . One therapy that works in some patients might be ineffective in other patients . It is challenging even in a single patient for the same tumor type . Protein-protein interactions ( PPIs ) have critical role in regulating and performing many cellular functions . Disease-associated mutations are more likely to affect protein interactions and eventually the cellular functions [4] . Several studies have focused on the impact of the disease-associated alterations in protein-protein interaction networks [5–8] . Recently , IMEX consortium [9] started an effort to curate and catalogue the oncogenic and neutral mutations in protein interactions [10] . The combination of three-dimensional structural information with large-scale mutation information may assist clarify the impacts of cancer mutations [7 , 11–14] as a protein's biological functions and physical interactions are strongly linked to its structure . Various mutations in the same protein may result in distinct profiles of interaction and eventually distinct phenotypes of disease [15–19] . Mutations that destabilize a protein’s global structure can result in severe alterations in its overall interactions . Additionally , a mutation may affect only one interface of a multi-face protein and the lost and gained interaction partners through the affected site may give insights about the functional changes . This type of edgetic perturbations [4] in proteins thus require structurally resolved PPI networks and the 3D spatial position of the mutations in proteins [20–22] . Mutations in cancer have been evaluated in many studies based on their organization in proteins structures [5 , 6 , 23 , 24] . Niu et al spatially clustered the mutations from 19 different cancer-types and came up with the set of druggable functional mutations [24] . The functional effects of mutations on protein interactions and signaling networks have been extensively reviewed in [14] which nicely puts forward that biophysical studies complement omics and clinical data . Additionally , some other studies focused on patient-specific analysis of the molecular signatures in tumors in a network context [25–27] . The phosphoproteomic data from eight GBM patients have been previously used to demonstrate that the network-guided comparison reveals commonalities and differences across patients [27] . In another network-based approach , mutations , transcriptional and phosphoproteomic data were used to model patient-specific pathways in prostate cancer [25] . The network based stratification ( NBS ) approach integrated somatic mutation profiles with molecular interactions to divide a heterogeneous set of tumors into clinically similar clusters [28] which was successfully applied to various TCGA mutation profiles [28 , 29] . NBS was earlier used in conjunction with structural locations of cancer missense mutations to disclose the impacts of a mutation in the core or interface regions when the rebuilt networks are perturbed [30] . Network-based analysis was further used to distinguish driver mutations from passenger mutations in GBM [31] . Computational approaches are crucial for analyzing the effects of mutations on proteins , protein interactions and functional pathways in a patient-specific way , considering the big quantity of diverse data including mutations , protein structures , and known PPIs . We applied a systems level approach to the somatic missense , nonsense and frameshift mutations in 290 Glioblastoma ( GBM ) patients which is the most aggressive type of brain tumor . The mutation profiles are rarely common across the patients and they do not track with the known transcriptional subtypes of GBMs or the known biomarkers such as the IDH1 mutation . Despite this heterogeneity , mutations in different proteins functioning in the same pathway may result in phenotypically similar tumors . In order to overcome the heterogeneity in tumors and develop personalized therapeutic strategies , reverse engineering from mutations to networks and pathways is a key approach . In this work , we proceeded in two directions: ( i ) finding the spatial arrangement of the mutations as patches and ( ii ) reconstructing the sub-networks primarily affected by the set of mutations across patients ( see Fig 1 ) . Then , each patient-specific network was reduced into a significantly enriched set of pathways and patients were grouped to better classify them into clinically similar groups according to their pathway similarity . Toward the precision medicine , patient groups were analyzed based on their predominant patches and pathways , and were associated with their survival profiles . Eventually , drug sensitivity data in GBM cell lines were integrated with the signatures of patient groups and hypothetical therapeutic strategies for each patient group were inferred .
The missense , nonsense and frameshift mutations from 290 GBM tumors were first analyzed at the sequence level . There are 15 , 399 unique mutations and 14 , 308 of them match at least to one canonical protein whereas the rest matches to alternative isoforms of the proteins . The average number of mutations per patient is 50 . 43 . The mutations are rarely common across different GBM tumors where only 44 mutations are present in at least three patients and 213 mutations are present in at least two patients . The most frequent mutations with 13 patients are EGFR mutation A289V and IDH1 mutation R132H . Next , we mapped the GBM mutations on to protein structures and found that 4702 mutations were aligned to at least one protein structure either from PDB [33] or from ModBase [34] . The local organization of mutations in 3D was determined by their spatial proximity to each other which we call “patches” . A patch is a set of mutated residues that are either in physical contact with another mutated residue ( that is , at least one pair of atoms within 5Å distance ) , or there is another intermediate residue in close proximity connecting the two mutated residues . The term “patch” was used in previous studies , however we have to indicate that our patch definition is different from those [11 , 24] . We looked for continuous residue contacts instead of using a mutated residue as the center of the patch . The 3D spatial grouping of 4702 mutations resulted in 220 patches composed of 580 mutations and 4122 singletons ( a mutation that is not involved in a patch ) . We then split patches as intra- and inter- which represents patches that do not include any interface mutations and patches that have at least one interface mutation , respectively . The interpatch can consist of residues of a single protein or two partner proteins . In total , there are 160 intra- , 60 inter-patches in our dataset . A patch is present in a patient if the patient has at least one patch mutation . While each individual mutation is shared between 1 . 13 patients on average , each patch is present in 3 . 5 patients on average , which partially reveals some common patterns across patients . In Fig 2A , patches are sorted based on their frequency across patients . The patches in TP53 and PTEN are common among 20% of the patients , and the patches in EGFR are common in 8% of the patients , that yields slightly better detection of commonalities across patients . Then , we divided the patients into mutually exclusive groups based on their most common patches . We linked patient survival data to assess the advantage of patches ( 3D spatial grouping ) in overcoming heterogeneity . The 3D spatial grouping of patients is significantly associated with survival data ( P-value = 0 . 0001 for 162 patients having at least one mutation in patches ) . we found no significant association of individual mutations between survival and patient groups . ( P-value = 0 . 5115- S1 Fig ) . Previous studies suggested that 3D clustering of the mutations led to a better classification of different cancer types , driver mutations or novel cancer genes [5 , 11 , 24 , 35] . Our initial results suggest that patient-grouping is also possible with 3D patches . The strong association between the patient groups and their survival indicates that patients with similar 3D spatial organization in their proteins may have similar disease phenotypes which may represent similar affected functions and pathways in the tumor cells . In order to understand the possible functional effects of spatial organization , we mapped the patches on protein domains . We found that different patches are located in different functional protein domains . For example , PIK3R1 gene encodes the P85 which have two patches: Patch1 is on the inter-SH2 domain that has the inhibitory function on PIK3CA by binding the catalytic domain p110 , on the other hand Patch2 is on the SH2 domain where the protein binds to phosphorylated residues . Additionally , PTEN has two patches and one patch is on the phosphatase domain , the other on membrane binding domain of the protein . Mutations on PTEN Patch1 disrupts the phosphatase function which results in accumulation of PIP3 [36 , 37] in cell and thereby activation of the AKT pathway that leads to tumor growth . PIK3CA protein consists of four different domains where Patch1 is only in the region in which the catalytic domain of PIK3CA , p110 binds the P85 subunit which is the p110 inhibitor . We indicated the mutated residues of frequently mutated hub protein patches mainly located in the corresponding domains in Fig 2C . So far , our analysis did not differentiate driver mutations from passenger mutations . Driver genes can , when mutated , play a causal role in tumorigenesis and should be enriched for driver mutations . In our analysis , we wanted to eliminate the noise from passenger mutations , therefore a list of putative driver genes was assembled from various literature sources and databases including The Network of Cancer Genes [38] , Cancer Genome Interpreter [39] , COSMIC [40] and the Firehose data using CHASM [41] , MutSig [42] and Mutations Assessor [43] for GBM and the analysis focused on those genes . In total , we obtained 6270 driver mutations from 3789 driver genes . We then found the intersection between the TCGA GBM mutation dataset and the collected driver genes and driver mutations . Of all the mapped mutations , 6278 are located in a driver gene of which 2072 mutations map to at least one protein structure . When we filtered out our 3D patch dataset based on the driver genes , we obtained 112 intra- , 32 inter-patches . There are numerous patches which only contain two or three residues ( Fig 3A ) . These small patches tend to be intra-patches without any interface mutations . On the other hand , the larger patches happen to be inter-patches ( with at least one interface residue ) and the largest ones are found in the central proteins ( TP53 with 41 , PTEN with 43 residues as shown in Fig 3A ) . An example of a large patch in PTEN is illustrated in Fig 3B . Some proteins have various patches of comparatively small size such as EGFR ( its three patches are shown in Fig 3C ) . An example of inter-patches is in the PIK3R1-PIK3CA complex where both partner proteins have at least one mutation ( Fig 3D ) . We further compared the differences in 3D spatial organization of their mutations between tumor suppressors and oncogenes . Interestingly , we found that driver mutations of the tumor suppressors have a tendency to be located in patches whereas driver mutations of the oncogenes mostly remain as singletons ( P-value = 8 . 33x10-6/ Fisher’s Exact Test ) . These results may explain the reason why PTEN ( 43 and 2 residues in two different patches ) and TP53 ( 41 residues in only one patch ) , which are tumor suppressors , have relatively larger patches . On the other hand , PIK3CA ( 10 , 3 , 2 , 2 residues in three different patches of PIK3CA ) and EGFR ( 11 , 14 , 5 residues in three different patches ) oncogenes have smaller patches and also many singletons . These results agree that it is more difficult to make a protein more active or efficient , therefore mutations in oncogenes tend to pile up at very specific sites , i . e . all cancer-related Ras mutations are around the GTP binding site . Whereas tumor suppressors can be functionally impaired in a variety of ways and thus mutations could be more broadly distributed in large patches . Additionally , we further compared oncogenes and tumor suppressors based on the frequency of non-synonymous mutations and found that frameshift and nonsense mutations are significantly more frequent in tumor suppressors ( P-value = 1 . 84x10-15 ) . These types of mutations may disrupt functionality of tumor suppressors making cells more vulnerable to cancer . As mentioned in the previous sections , mutations are rarely common in patients with GBM . Mutations may be on distinct proteins but they may alter the same pathway . Therefore , we first reconstructed patient-specific subnetworks from mutation profiles and then reduced each network into enriched pathways . We detail the outcomes of the network reconstruction and grouping patients based on network similarities in the coming section . The significant association between patient groups and survival led us to further analyze the possible therapeutic targets in each group . The therapeutic information is very sparse in TCGA; therefore , we collected drug sensitivity data of the GBM cell lines treated with different drugs from CancerRxGene [55] . We also retrieved the target proteins and the target pathways of each drug . As a result , we collected 37 GBM Cell Lines having the mutation profile information in the Cell Model Passports database [56] . In total , there were 13243 mutations . We got the intersection of these mutations with the set of GBM mutations we used in our study and found that 23 mutations are common of which 16 are located in patches . These enriched patches are on PTEN , TP53 , EGFR , BRAF and RB1 proteins . As a result , we obtained 17 cell lines treated with 73 drugs that target 18 pathways . To link the patient groups to the drug response data of each cell line , we used the signature 3D patches . We found that 44 patches tend to be significantly present in one or multiple patient groups including PTEN , TP53 , EGFR , BRAF and RB1 patches which are also present in cell lines ( S3 Fig ) . According to our results , all patches of PIK3R1 , PTEN , TP53 and one patch of PIK3CA and BRAF have a strong tendency to be present in Group 5 . While RB1 , TP53 , PTEN , patches are enriched in Group 2; TP53 , PTEN patches without EGFR and RB1 have a bias to be in Group 1 . EGFR has 3 patches , and PTEN has two patches which are present in distinct groups . While all patches of EGFR are found only in Group 5 , Groups 4 and 3 only have one of the patches of EGFR . Patch 1 , 3 and 4 of PIK3CA are only found in Group 3 , Patch 2 of PIK3CA is found in Groups 2 , 3 and 5 . One of the very well characterized biomarkers in GBM is BRAF . BRAF mutations V600E and G596D form a 3D patch in only Group 5 . Group1 is linked to the cell lines having at least one mutation in TP53 Patch and PTEN Patch1 , Group 2 is linked to the cell lines having TP53 Patch and PTEN Patch1 and also RB1 Patch . Group 4 has the PTEN Patch and Group 5 has EGFR Patch2 , TP53 Patch , PTEN Patch1 together with the BRAF Patch ( Fig 8A ) . Our first therapeutic hypothesis is based on Pazopanib which is a multi-targeted receptor tyrosine kinase inhibitor . Pazopanib is linked to our patient groups through its targets CSF1R and PDGFRB that are significantly enriched in Group 3 and Group 5 , respectively ( Fig 8B ) . GBM cell line GI-1 is sensitive to Pazopanib . GI-1 has at least one mutation in TP53 patch which is predominantly available in Group 3 and Group 5 . CSF1 ( colony stimulating factor 1 ) binds to CSF1R and activates several signaling pathways , including Ras/Raf/MAPK , phosphatidylinositol 3-kinase ( PI3-kinase ) and JAK/STAT pathways . When we refer to the pathway enrichment results in Group 3 , Phosphatidylinositol and JAK-STAT pathways were enriched . Additionally , CSF1R is also on tumor-associated macrophages and microglia ( TAMs ) which are highly available in glioma microenvironment . CSF1 , the ligand of CSF1R , is responsible for the differentiation of TAMs to pro-tumorigenic . The inhibition of CSF1R results in the differentiation of the macrophages and makes them more anti-tumorigenic . Another target of Pazopanib is PDGFRB protein is a receptor tyrosine kinase and functions as a cell surface receptor . It activates cell proliferation and survival . Moreover , it is proven that PDGFRB is overexpressed in GBM cells and very important for self-renewal [57] . Therefore , we suggest that Group 3 and Group 5 might be sensitive to a treatment based on Pazopanib . Another example of a significant target is ATM , which is a target of ATM-inhibitor ( CP466722 ) . ATM is present in the network of Group 2 where PTEN Patch 1 is enriched ( Fig 8C ) . Two GBM cell lines ( KALS-1 , GMS-10 ) are resistant to this drug molecule . ATM is a mediator of PTEN phosphorylation and ATM targeting drugs are used to make a patient more sensitive to radiotherapy . In our therapeutic hypotheses , we suggest that Group 2 might be resistant to ATM-dependent therapy . Our last example is SRC protein as a target in Group 5 which is a non-receptor protein tyrosine kinase and plays an important role in many cellular processes such as growth , adhesion , and differentiation . It is also a component of several cell signaling pathways including EGFR , ERBB , and Rap1 signaling pathways . Group 5 is associated with the GBM cell line D-452MG through TP53 and PTEN patches . The kinase-inhibitor WZ3105 , that targets SRC , is resistant in D-452MG and we suggest that Group 5 might be possibly resistant to WZ3105 according to our therapeutic hypothesis ( Fig 8D ) .
The mutation landscape of GBM tumors is very heterogeneous and not discriminative to classify disease progression and subtypes . Given the impact of mutations in protein interactions and eventually cellular signaling pathways , reverse engineering from mutation profiles to patient-specific subnetworks can shed light on network-level changes to observe hidden commonalities . Therefore , we applied a systems-level strategy to the patient-derived information of 290 GBM tumors to gather knowledge about their commonalities . We first started with an in-depth analysis of individual mutations such as their spatial organization , physicochemical characteristics , and their effects in binding . Our results show that out of 15399 mutations , 4702 mutations have structural information and 10% of these mutations are spatially grouped into patches , while most mutations are spatially distant to other mutations , namely , singletons . Interestingly , distinct patches of a protein are located in distinct domains that could have distinct functional consequences . Despite a small portion of all mutations , 3D patches reduce the heterogeneity across patients and more commonalities can be identified . To show the success of the patches in overcoming heterogeneity , we associated them with the survival data . Indeed , grouping patients based on the patch information significantly discriminates the survival curves . For example , tumors in patients with at least one mutation in PI3K patches are more aggressive compared to the tumors with at least one mutation in TP53 patch . These results are a proof of concept that 3D spatial grouping of mutations can be related to clinical outcome and is useful in overcoming heterogeneity . In our follow-up analyses , we found that GBM mutations are significantly more frequent in interface regions than the rest . Although a small portion of all mutations is located in the core region; they may affect the function more severely . Mutations in the core tend to preserve their chemical classes while interface and surface mutations are significantly more prone to changes . Mutated residues are mostly populated on the surface and their functional effects are less severe than the rest ( interface and core ) . When we limit our analysis to only the mutations on cancer-driver genes , we still have the same results . Furthermore , tumor suppressors have large patches while oncogenes have multiple and relatively smaller patches . Having the mutations on very specific sites of oncogenes agrees with the known observation that making a protein more active is harder . On the other hand , tumor suppressors can be functionally damaged in several ways thus their mutations could be distributed in large regions . Also , nonsense and frameshift mutations are more frequent in tumor suppressors . Mutations in hub proteins are organized into very large patches that connect mutations in multiple binding sites through the core of the protein ( observed in tumor suppressors ) . Additionally , the organization of multiple 3D patches in hub proteins implies the importance of 3D patches in cancer vulnerability ( observed in oncogenes ) . 3D spatial organization of mutations in hub proteins may provide a fitness advantage to tumor cells . Not all mutations are equally damaging . Some mutations are neutral and some mutations cause damage in protein stability or protein binding . Our results suggest that hub proteins’ patch mutations are more disease-causing , whereas other proteins’ singleton mutations are more disease-causing . Some proteins repeatedly use a single interface to interact with their partners while some proteins have multiple interfaces . The characteristics of the mutations are also different in these interfaces . Interface mutations in proteins with a single interface stay distant from other mutations and are usually present as singletons; however , interface mutations in proteins having multiple interfaces are mostly located in patches . When we analyzed the patient-specific networks and the consensus network of these patients , we observed that although mutation profiles are very heterogeneous across patients and their pathway-level representation is very limited , the network-based analysis groups the patients better and reveals predominant pathways in each group . Additionally , the network-based similarity analysis shows that each group of patients carries a set of signature 3D mutation patches . For example , EGFR , TP53 , PIK3CA , PIK3R1 patches are frequently found in Group 3 , TP53 , RB1 , PTEN patches in Group 2 . Beyond the list of mutations , the network-guided analysis also reveals similarities across patients and overcomes the heterogeneity in mutation profiles by completing the interaction components that mutated proteins potentially affect . We found that there are significant differences across the patient groups in their survival . Additionally , several pathways are common in each patient group , such as the Jak-Stat pathway being enriched in three groups , the TGF-beta signaling pathway being present in only one patient group . This pathway-level outcome led us to link the available drug treatment data to our patient groups . Because the drug treatment data is very sparse in TCGA , we used the GBM cell lines for this purpose . Each group of patients is linked to each GBM cell line through its predominant patches . For example , we found that PDGFRB , the target of Pazopanib , is significantly present in the subnetworks of Group 5 which has the TP53 patch as the marker . GBM cell lines having a mutation in the TP53 patch and treated with Pazopanib are sensitive to this drug . We therefore proposed that Pazopanib may be efficient in Group 5 . This sort of therapeutic hypotheses was found and suggested for each group of patients as provided in Figs 8 and S4 . Overall , these results show that network-guided interpretation of mutations and their 3D organizations give a deeper insight into their impact and useful in overcoming the inter-tumor heterogeneity that is the main barrier in finding optimal treatment strategies . Despite the apparent diversity in the mutation profiles of GBM tumors , the 3D spatial grouping of mutations and network-guided clustering of tumors reveal several commonalities and enable us to link the gathered information to clinical outcomes and therapeutic data . Our approach from mutations to protein interactions and eventually to signaling networks and pathways transforms the tumor information into clinically interpretable knowledge . We believe that this study represents a good example of how networks can be used efficiently in precision medicine .
The missense , nonsense and frameshift mutations in Glioblastoma were retrieved from TCGA which has been published in [58] for 290 patients . First , all proteins that have at least one mutation was searched in PDB [33] . If a structure was not available , we made use of ModBase [34] homology models . The structural information of the protein interactions was collected from PDB , Interactome3D [7] , PRISM [44] and Interactome Insider [12] . PDB deposits protein complexes that are crystallized together . Interactome3D predicts the protein complexes through structure and domain similarity with a template structure . PRISM uses known interfaces to predict new protein interactions . We used the pre-runned PRISM results for a subset of the proteome , rather than the whole proteome . The other source for structural protein interactions was the Interactome Insider . It produces the binding sites on each partner of the protein interaction . Different than PRISM and Interactome3D , it does not give the structure or the pose of the predicted protein complex . We also downloaded the human proteome from UniProt [59] for cross-referencing from one data source to another . Additionally , the residue positions in sequence are not consistent with the residue positions in protein structures . A PDB entry or a homology model of a UniProt sequence may represent only a fragment of the given protein and the residue numbering may not be the same with the sequence positions . Therefore , we performed UniProt sequence to the sequence in the protein structure alignment to find the exact position of each residue . Moreover , we retrieved the known cancer genes from The Network of Cancer Genes [38] , cancer-related genes from Cancer Gene Census of The Catalogue of Somatic Mutations in Cancer ( COSMIC ) [40] , validated oncogenic mutations from Cancer Genome Interpreter [39] . The Network of Cancer Genes is a repository for predicted or known cancer driver genes that have been manually curated . In this analysis , we only used the known cancer genes . Similarly , Cancer Gene Census of COSMIC includes the manually curated cancer genes that behave as driver effect for human cancer . We included these genes into our analysis . On the other hand , dataset of Cancer Genome Interpreter gives the oncogenic mutations by using the information from DoCM , ClinVar , OncoKB , and IARC . We also took these cancer driver mutations . Additionally we obtained the mutation information for GBM from Broad Institute FireBrowser which includes MutSig2CV v3 . 1 [42] , Mutation Assessor [43] , CHASM 1 . 0 . 5 [41] . Mutation Assessor only considers the missense mutations and gives the gene names which missense mutations on and their functional impacts . In our analysis , we only considered the high and medium functional impact genes as significant genes . Secondly , MutSig2CV gives the gene significance according to mutations on the gene . In this project , we took the genes whose P-value is smaller than 0 . 05 as significant genes . Lastly , CHASM uses the missense mutations and gives the probability for each mutation due to the selective survival advantage that is provided to the cancer cells by the mutation . In this analysis , we only considered the mutations that have P-value smaller than 0 . 05 . To reach the gene name for the mutation in CHASM , we also needed Ensembl BioMart [60] for the conversion between RefSeq mRNA ID to Ensembl Transcript ID to reach the official gene symbol . Finally , the confidence weighted interactome deposited in iRefWeb has been downloaded for the reconstruction of patient-specific sub-networks inferred from mutation profiles of each patient . Afterward , the interactions having structural information also retrieved for constructing structural interactome for Omics Integrator analysis . Each cancer-related driver protein structure and protein complexes were converted into a network of residue-residue interactions . If any atom in a residue is in close proximity to any atom in another residue , then these two residues were considered to interact . The proximity was defined as the distance of less than 5Å between any atoms . We constructed a residue contact graph R ( v , e ) for each structure where v is the set of residues and e is the set of edges between these residues . We searched for all shortest paths between each mutated residue pairs with a length of less than 3 to identify the spatial clusters , which means that if two mutated residues are either directly connected or only one residue is in between them . Then all the extracted shortest paths merged to create a subgraph P ( v’ , e’ ) representing one spatial cluster , namely “patch” where v’⊂v and e’⊂e . Mutations that are not assigned to a patch were labelled as singletons , meaning that these residues are distant to other mutations in the same protein . Patients are grouped based on the presence of each patch . This grouping is performed iteratively . The first group is formed by the patient having at least one mutation in the most frequent patch . The second group is the patients having at least one mutation in the second most frequent patch , and having not any mutation in the most frequent patch . This iteration continues until each group has at least ten patients . In this way , each group is mutually exclusive , where there are no common patients across the groups . Proteins can be divided into three regions , namely , the core , surface and interface regions . The conventional approach for identifying these regions is to calculate solvent-accessible surface areas of each residue in the protein . FreeSASA [61] is a software designed for calculation of solvent accessible surface area at both residue level and molecule level . In general , if the relative solvent accessible surface area of a residue in its monomer state is greater than or equal to 5% , then this residue is labelled as the surface residue . Interface residues that are collected from structural interactome are excluded from the surface residue set . The rest is identified as core residues . However , we only considered the structure files whose length greater than 50 residues for this analysis . We used EVmutation and PolyPhen-2 web servers to calculate the effect of mutations if they are damaging or neutral . The EVmutation data provided the information for a limited number of proteins in text format where each position in a UniProt entry is substituted by the remaining 19 amino acid and the damage score is calculated . The more negative values of the calculated score means the more damaging mutation . The details of the calculation steps of EVmutation scores are in reference [49] . On the other hand , Polyphen2 gives the results as being probably damaging , possibly damaging or benign . We used mutation effect data to compare the damage of the mutations based on their localization and their role in the cancer progression ( tumor suppressors or oncogenes ) . In our setup , we focused on driver genes to reduce the noise caused by passenger mutations . We added each protein having at least one nonsynonymous mutation ( missense , nonsense and frameshift ) on a driver gene/protein in a tumor sample to the list as the base for network reconstruction and weighted each protein with their number of mutations . We used the probability-weighted protein-protein interactions in iRefWeb [62] as the reference interactome . This reference interactome further is filtered by the interactions having structural information . Additionally , Omics Integrator has a unique feature to avoid biasing the dominance of well-studied proteins or hubs in the final network . We used hub-penalizing parameters set to reveal more specific pathways for a better comparison of the patient-specific networks . Two different values of the scaling factor of hub proteins ( μ parameter , described in Methods ) were used for this purpose and resulting optimal networks were merged . Omics Integrator software was used to reconstruct patient-specific sub-networks . Given a reference graph G ( V , E , w ) where V is the node set {v|v ∈V} , E is the edge set {e|e ∈ E} and w is the edge weights , the Forest module of Omics Integrator solves the prize-collecting Steiner forest problem for a given set of nodes with predefined prizes . In our case , the terminal nodes were the mutated cancer proteins for each patient and the prizes were given according to the significance of the mutations included in each protein . If the mutation on the terminal node was significant , we added 1 as a prize to the terminal node and if it was not significant , the added prize was 0 . 5 . Therefore , the prize list composed of the proteins from cancer genes having at least one mutation . We retrieved the iRefWeb v8 . 0 interactome and filtered the interactions if they have structural information or not . Therefore , we used structural iRefWeb interactome as the weighted reference interactome in our modeling . To have a stringent setup , we filtered out interactions having a score less than 0 . 4 and also the proteins such as UBC , APP , ELAVL1 , SUMO2 , CUL3 and the proteins huge in size ( TTN , MUC16 , SYNE1 , NEB , MUC19 , CCDC168 , FSIP2 , OBSCN , GPR98 ) to limit the noise coming from random mutations in these proteins as in [63] . The parameter set ω ( omega ) = 10 . 0 , depth ( D ) = 6 and β ( beta ) = 10 was used for the reconstruction . Omega ( ω ) parameter was used for tuning the number of trees in the final network , depth is the number of edges from the root to the leaf nodes and beta ( β ) is a scaling factor to force more prize nodes to enter the final network . Finally , mu ( μ ) is another scaling factor to tune the dominance of hub proteins in the final network . We used two mu ( μ ) values ( 0 . 005 . 0 . 01 ) to recover the canonical pathways and more specific ones and merged the node and edge set of the reconstructed networks to come up with a single network for each patient . WebgestaltR [64] package evaluated each patient’s subnetwork to obtain the overrepresented KEGG pathways in each network . Pathways were assumed to be enriched in the sub-network if the False Discovery Rate ( FDR ) is less than 0 . 1 . In the resulting list of pathways , we eliminated disease pathways including infections , cancer , addiction related pathways . Then we prepared a matrix where rows are union set of enriched pathways , columns are patient barcodes and entries are the enrichment score ( ES ) of a pathway in the corresponding barcode’s sub-network . If the pathway is not enriched , 0 is inserted into that entry . We used this matrix for implementing the non negative matrix factorization without a network regularizer and then consensus clustering from pyNBS package which is a Python implementation of NBS [65] . The identified groups were searched for if any identified spatial patch tends to represent a group using hypergeometric testing . We linked each patient group to the GBM cell lines retrieved from Cell Model Passports through the mutation information . If at least one mutation belonging to a predominant patch in a group is also present in the GBM cell line then the patient group is connected with that cell line . The drug treatment data is obtained from CancerRxGene [55] where the sensitivity of the drugs to the cell lines are deposited . For each drug , target proteins and target pathways are also obtained .
|
Precision medicine aims to find the best treatment strategy based on the information about the patient’s tumor . Molecular heterogeneity is the main obstacle in developing treatment strategies . Therefore , transforming patient specific molecular data into clinically interpretable knowledge is fundamental in precision medicine . In this work , we tackle the mutation profiles of patients with Glioblastoma Multiform ( GBM ) which is the most aggressive type of brain tumors with a poor survival . Our main motivation is that different mutations , that are spatially in close proximity in the same protein , or function in the same pathway , may result in phenotypically similar tumors . 3D spatial clustering of the mutations , that we call “mutation patch” , significantly decreases the heterogeneity . We additionally identify the affected patient-specific subnetworks and pathways that are inferred from mutations . Indeed , grouping the patients based on the presence of mutations in close proximity together with network-guided grouping is significantly associated with their survival . These results also enable us to suggest several therapeutic hypotheses for each group based on available drug treatment data . We believe that from mutations to networks and eventually to clinical and therapeutic data , this study provides a novel perspective to the analysis of mutation effects towards the network-guided precision medicine .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"interaction",
"networks",
"protein",
"interactions",
"protein",
"interaction",
"networks",
"mutation",
"network",
"analysis",
"nonsense",
"mutation",
"protein",
"structure",
"mutation",
"databases",
"frameshift",
"mutation",
"research",
"and",
"analysis",
"methods",
"computer",
"and",
"information",
"sciences",
"proteins",
"biological",
"databases",
"proteomics",
"molecular",
"biology",
"biochemistry",
"database",
"and",
"informatics",
"methods",
"genetics",
"protein",
"domains",
"biology",
"and",
"life",
"sciences",
"macromolecular",
"structure",
"analysis"
] |
2019
|
3D spatial organization and network-guided comparison of mutation profiles in Glioblastoma reveals similarities across patients
|
Recent efforts to cure human immunodeficiency virus type-1 ( HIV-1 ) infection have focused on developing latency reversing agents as a first step to eradicate the latent reservoir . The histone deacetylase inhibitor , vorinostat , has been shown to activate HIV RNA transcription in CD4+ T-cells and alter host cell gene transcription in HIV-infected individuals on antiretroviral therapy . In order to understand how latently infected cells respond dynamically to vorinostat treatment and determine the impact of vorinostat on reservoir size in vivo , we have constructed viral dynamic models of latency that incorporate vorinostat treatment . We fitted these models to data collected from a recent clinical trial in which vorinostat was administered daily for 14 days to HIV-infected individuals on suppressive ART . The results show that HIV transcription is increased transiently during the first few hours or days of treatment and that there is a delay before a sustained increase of HIV transcription , whose duration varies among study participants and may depend on the long term impact of vorinostat on host gene expression . Parameter estimation suggests that in latently infected cells , HIV transcription induced by vorinostat occurs at lower levels than in productively infected cells . Furthermore , the estimated loss rate of transcriptionally induced cells remains close to baseline in most study participants , suggesting vorinostat treatment does not induce latently infected cell killing and thus reduce the latent reservoir in vivo .
Treatment of HIV-infected individuals with combination antiretroviral therapy ( cART ) effectively suppresses HIV to levels below the limit of detection of conventional assays and substantially reduces morbidity and mortality of HIV infected patients [1] . However , it does not eradicate the virus and treatment is lifelong [2] . Therefore , developing novel therapeutics to cure HIV infection remains an important research priority [3 , 4] . A major barrier to cure is the presence of a population of long lived latently infected cells [4] that can persist indefinitely in patients treated with highly potent cART [5] . Recent efforts have focused on strategies that activate HIV production in latently infected cells . The idea , termed ‘shock and kill’ [6] , is to first shock latently infected cells thereby activating HIV gene expression , such that the cells are then killed by viral cytopathic effects or immune-mediated cell death . Histone acetylation is one of several factors that regulate HIV transcription and is therefore important for establishing and maintaining latency [7] . Drugs such as histone deacetylase inhibitors ( HDACi ) enhance acetylation of both histones and proteins and thereby induce changes in gene transcription , including transcription of HIV [8] . Vorinostat , a histone deacetylase inhibitor licensed for the treatment of cutaneous T-cell lymphoma [9] , has been shown to activate HIV transcription in resting memory CD4+ T-cells in vivo [10 , 11] . In a recent clinical trial , 20 HIV-1 infected individuals on suppressive cART were treated orally with 400 mg a day of vorinostat for 14 days and then followed for an additional 70 days . Overall , vorinostat induced a rapid and sustained increase of cell-associated unspliced ( CA-US ) HIV RNA [10] . However , the response pattern was highly variable among the participants . For example , in half of the participants , after an initial significant increase in CA-US HIV RNA , the level of CA-US HIV RNA decreased rapidly within 1–3 days before increasing again , and in 14 of the 20 participants , the level of CA-US HIV RNA continued to increase after vorinostat was stopped . These puzzling observations raise important questions about the temporal impact of vorinostat treatment on HIV transcription and the design of treatment strategies to eradicate the latent reservoir . We constructed mathematical models to better understand the temporal changes in CA-US HIV RNA in individuals treated with vorinostat . Mathematical models have been widely applied to study viral dynamics in vivo [12–14] . They played an instrumental role in quantifying important parameters , such as the half-lives of virions and infected cells in vivo [14] . Recently , several models have been developed to understand the maintenance of the latent reservoir under cART treatment [15–17] , the viral rebound time distribution after latency reversing agent ( LRA ) treatment [18] , and the optimal time to start a LRA [19] . However , the dynamic response of HIV transcription in latently infected cells following treatment with a LRA has not been investigated . This question has important implications for future clinical trial design and optimizing treatment strategies to eliminate latently infected cells . Previous models have generally assumed that the HIV provirus in latently infected cells becomes fully activated following treatment with a LRA and that subsequent events will be identical to latently infected cells activated by normal immunological signals or through the T-cell receptor [18 , 19] . However , evidence suggests that current LRA treatments primarily activate HIV transcription and its impact on translation may be mild or minimal [20 , 21] . Here , we construct models that treat cells activated by a LRA and naturally activated cells separately . By fitting models to the clinical data , we show the complex dynamic response of latently infected cells to vorinostat can be explained . Furthermore , we use the models to quantify the extent to which vorinostat activates HIV transcription and induces cell death in vivo .
We first construct a mathematical model based on a previously published latency model by Rong et al . [15] . As the study participants were treated with suppressive cART for a medium of 5 years , we have chosen parameters in the Rong et al . model such that before vorinostat treatment the viral load is approximately 5 HIV RNA copies/ml , target cells levels are 750 T cells/μL similar to mean CD4 count in the 20 clinical trial subjects [10] and latently infected cell levels are 2/mL , which is approximately 2 . 7/million CD4 cells ( roughly consistent with previous studies [22 , 23] ) . The major innovation in the direct activation model is that we assume latently infected cells become transcriptionally induced and express CA-US HIV RNA directly upon vorinostat treatment ( Fig 1A; see Methods for full description of the model and Table 1 for parameter values ) . Thus , two equations are added to the Rong et al . model [15] , one for the number of cells that have HIV transcription induced by vorinostat , LA , and one for the average amount of CA-US HIV RNA per transcriptionally activated cell , R . Also , cells that are transcriptionally induced by vorinostat are assumed to be in a different state than cells that are naturally activated . We fitted this model to the clinical data collected during the entire 84-day study period ( see Methods for the fitting procedure and S1 Table for best-fit parameter values ) . In general , the direct activation model does not explain the data well especially during the first 1–3 days’ treatment ( S1 Fig ) and the period after treatment stops ( S2 Fig ) . First , in 10 of the 20 participants , the level of CA-US HIV RNA first increased upon initiation of vorinostat , and then decreased rapidly after the first 1–3 days , whereas the best-fit model curves in these patients have CA-US HIV RNA increasing continuously during vorinostat treatment . Second , in 14 out of the 20 participants , the level of CA-US HIV RNA increased at variable time points after cessation of vorinostat at day 14 , whereas the model predicts that the level always decreases over time after cessation of vorinostat . To better understand the initial peaking pattern observed in half of the study participants soon after the initiation of vorinostat , we fitted the model to data obtained during the first 7 days of treatment only ( Fig 2 ) . The peaking pattern was well described by the model in five participants ( VOR001 , VOR004 , VOR010 , VOR016 , VOR018 ) . In these individuals , the estimated loss rate of transcriptionally activated cells , dLA , ranged between 1 . 8 and 10 day-1 ( see S2 Table ) , and is much greater than the death rate of productively infected cells , i . e . 1 . 0 day-1 [27] . Rates of decrease of CA-US HIV RNA levels higher than 1 . 0 day-1 were also apparent in 5 other participants ( VOR008 , VOR019 and VOR021-023 ) . This decrease can potentially result from either death of cells transcriptionally activated by vorinostat or from shutdown of HIV transcription and loss of CA-US HIV RNA . We reason that this decline is not due solely to cell death , as it is unlikely that cells activated by vorinostat die at a faster rate than productively infected cells . Then why would HIV transcription shut down during vorinostat treatment ? Previous in vitro studies have shown the activation of HIV transcription is a transient stochastic process , and that the duration of this transient process is dependent on the strength of Tat transcriptional feedback [34–36] , as well as the availability and regulation of many host factors that are necessary for transcriptional activation , such as the NAD-dependent deacetylase sirtuin-1 , NF-κB , Yin Yang 1 and the positive transcription elongation factor , P-TEFb [37–41] . In latently infected cells , mostly memory T cells , these transcription factors are likely to be at low levels [42–44] , whereas many host enzymes such as Murr1 ( a gene product that restricts HIV-1 replication ) , human schlafen 11 and the lipid raft associated protein tetherin , actively inhibit HIV transcription initiation [45] , mRNA translation [46] , and viral release [47] . Therefore , before vorinostat treatment , the host factors/enzymes required for full HIV gene activation are most likely limiting in latently infected cells . After vorinostat treatment initiation , host genes undergo rapid differential regulation at 2 , 8 and 24 hours [10]; however , the immediate impact of vorinostat treatment may not be sufficient to induce HIV gene transcription sustainably . This unfavorable cellular environment and rapid changes in gene expression may lead to very short transcriptional pulses of Tat activity and CA-US HIV RNA production . Without further production of CA-US HIV RNA , the rapid decrease observed in the data may be a result of the loss of US HIV RNA by degradation and by splicing . Vorinostat treatment not only induces rapid changes in host gene expression but also induces changes after treatment cessation [10] . It is therefore plausible that the late increase in HIV transcription after vorinostat treatment is due to a longer-term impact on host gene transcription . To test this hypothesis , we extended the direct activation model to include a ‘transiently activated’ state ( LT ) and a ‘waiting’ state ( LW ) , and denote this as the ‘delayed activation’ model . We assume that upon vorinostat treatment latently infected cells first get activated transiently , i . e . , enter the transient activation state , LT , where CA-US HIV RNAs are produced for a short period of time . We assume the cells then enter a waiting state , LW , in which there is no CA-US HIV RNA production before becoming sustainably activated cells , LA ( Fig 1B ) . The waiting state reflects the time needed for the transcriptional programs to produce sufficiently high levels of host factors necessary for transcriptional activation such that the cellular environment becomes favorable for sustained HIV transcription . We assume that the cells in the transiently activated state and in the waiting state die at the same rate as in the latent state ( L ) as transient activation is not likely to be strong enough to produce the shock needed for kill . The ordinary differential equations ( ODEs ) describing this model are given in the Methods . Next , we tested whether the above hypotheses explain both the short-term and the long-term dynamics of CA-US HIV RNAs by fitting the delayed activation model to the full data set ( see S3 Table for best-fit parameter values ) , and found that the delayed activation model describes the data much better than the direct activation model in a majority of participants ( compare S3 and S4 Figs with S1 and S2 Figs , respectively ) . It successfully describes the initial pattern of CA-US HIV RNA change following initiation of vorinostat in most individuals as well as the dynamics of CA-US HIV RNA in 6 of the 14 individuals where the level of CA-US HIV RNA increased after cessation of vorinostat . For the other 8 patients , the delayed activation model does not predict the magnitude of the late increase in CA-US HIV RNA level at some time points ( S4 Fig ) . We speculate that the discrepancy may arise from the assumption of an exponentially distributed residence time for latently infected cells in the waiting state before becoming sustainably activated ( an assumption implicitly assumed in the ODE system ) . This assumption is valid when a single event is needed for the transition to sustained activation . However , it is likely that multiple events must occur before the transition to sustained activation , such as upregulation of several host factors , HIV RNA splicing and expression of tat and other regulatory proteins including rev . We , thus , further modified our model to assume that cells in the waiting state have to go through several stages before becoming sustainably activated as in previous work describing the multiple events needed to drive an initially infected cell into viral production [48] . We denote this model the ‘multistage delayed activation’ model and the equations describing this model are given in the Methods . See Table 2 for a summary of the assumptions made with regard to the impact of vorinostat on latently infected cells in the three different models . In this model , the LW state is divided into n identical sub-states , i . e . LW , 1 , LW , 2 , … LW , n . The transition rate from one sub-state to the next is set to nkw such that the average residence time in the overall waiting state is 1/kw . This model is equivalent to one in which we assume the transition out of the waiting state is stochastic with the delay described by a gamma probability distribution [49] . We let n change from 1 to 10 , and fitted these 10 model variants to the clinical data from all 20 participants ( S5 Fig ) . The fitting results show that this multistage delayed activation model describes the patterns of increases of CA-US HIV RNA after cessation of vorinostat as well as the initial peak following initiation of vorinostat ( Fig 3 and S6 Fig; see S4 Table for best-fit parameter values ) . We further performed model selection using the corrected Akaike information criterion ( AICc ) ( see Methods ) . The direct activation model significantly underperformed compared to the delayed activation model and the multistage delayed activation model in 19 out of the 20 participants . The multistage delayed activation model was significantly better than the delayed activation model in 12 participants ( Table 3 ) . These results support the hypothesis that the immediate impact of vorinostat treatment is to activate HIV transcription for a short period of time ( 1–3 days ) , possibly due to the limited availability of many host cellular factors and that sustained activation may take longer to attain and a number of events ( possibly in host cell transcriptional regulation ) must occur before sustained HIV transcription becomes possible . As the number of events required varied among the participants , cells in different individuals may be in different states of latency . Also , host gene expression patterns , which can differ among individuals , may play a role in determining the length of the delay . Analyzing the effect of changing the number of waiting stages on the model fit to the data using AICc shows that for 17 out of the 20 participants , using a model with more than 7 stages would be a good choice in general ( S5 Fig ) . We next examined the best-fit parameter values of the multistage delayed activation model to assess the impact of vorinostat on latently infected cells in vivo . First , we find that the estimated values of α , the rate of CA-US HIV RNA production induced by vorinostat in latently infected cells , varies over a wide range ( over 1 . 5 logs ) among the 20 patients , suggesting the response to vorinostat is very heterogeneous across participants . In a majority of patients , the estimated values of α are smaller than the production rate of CA-US HIV RNA in productively infected cells , αI ( Fig 4A; see Methods for calculation of αI ) . We then examined the estimated loss rate of transcriptionally activated latently infected cells , dLA , and found that in 12 participants the estimated loss rates are extremely low , close to the death rate of latently infected cells , dL ( Fig 4B ) . Although the estimated loss rates are higher than dL in other patients , we were not able to distinguish whether the loss is through shutdown of HIV transcription or through cell death . Nonetheless , the low estimates of the loss rate in most participants suggest that vorinostat treatment does not induce killing of transcriptionally activated latent cells in vivo in a majority of individuals , and thus , according to this model , the reductions in reservoir size were minimal or absent in most participants . We further tested the robustness of the parameter estimates to variations in our assumptions . First , we varied the values of two fixed parameter values that describe the intracellular dynamics of CA-US HIV RNAs , i . e . the rate of US HIV RNA export from the cell in the form of virions , ρ , which we initially assumed to be 0 , and the combined rate of RNA splicing and degradation , μ . We find the parameter estimates are robust to changes in the value of ρ ( S7 Fig ) , and that the estimated production rate of CA-US HIV RNAs , α , decreases approximately linearly with decreases in μ ( S8 Fig ) . Thus , if the rate of CA-US HIV RNA loss is lower in cells activated by vorinostat than we have estimated , the estimated production rate of CA-US HIV RNA would also be lower . Last , we tested the robustness of our results to the assumption in the Rong et al . ( 15 ) model about how the latently infected cell population is maintained by employing a different model based on the work of Kim and Perelson [50] in which the latent population is maintained by homeostatic proliferation rather than asymmetric division ( see Methods ) . We found the model fits to the CA-US HIV RNA data and the estimates of α and dL are largely unaffected ( S9 Fig ) .
We have constructed mathematical models to describe the dynamics of CA-US HIV RNA in HIV-infected individuals on ART who received multiple doses of the HDAC inhibitor vorinostat . By fitting these models to a clinical dataset , we have assessed the dynamic response of latently infected cells to vorinostat and estimated the quantitative impact of vorinostat on the latently infected cell population . Model analyses show that the multistage delayed activation model , can describe both the short-term and the long-term patterns of change in CA-US HIV RNA induced by vorinostat in most individuals . This model assumes that in response to vorinostat treatment , HIV transcription in latently infected cells is induced transiently . Afterwards , the cells rather than returning to their original latent state go through several waiting stages where CA-US HIV RNAs are not produced but host gene expression patterns may change before becoming sustainably induced . The sustained induction of HIV transcription may even occur after vorinostat treatment is stopped . The induction of HIV gene expression depends on the availability of the HIV Tat protein as well as many host factors [34–36 , 42–44] . In latently infected cells , the number of Tat proteins [34] and the host factors necessary for inducing HIV transcription , such as P-TEFb , are likely to be at low levels [42–44] , and at the same time , the presence of inhibitory molecules , such as Murr1 , human schlafen 11 and tetherin , prevent transcriptional activation [45–47] before and at the early stage of response to vorinostat treatment . A recent proteomics and transcriptomics study showed that after 24 h of vorinostat treatment of primary CD4+ T cells the expression of a large number of host genes and proteins as well as genes and proteins previously reported to be involved in HIV transcription was modulated , with some effects appearing to be stimulatory and others inhibitory for HIV reactivation [51] . Therefore , it is likely that the immediate impact of vorinostat on histone acetylation and host gene transcription lead only to a transient induction of HIV RNA transcription and sustained HIV transcription may depend on the longer-term impact of vorinostat on host gene transcription [10] . This delay in sustained transcriptional induction may explain the later increase in the level of CA-US HIV RNA after cessation of vorinostat seen in this study , and the observed refractory periods in response to multiple doses of vorinostat in another study [52] . Note that the effect of vorinostat on host genes may also include the generation of read through transcripts containing HIV RNA [53] , but a recent report suggests such transcripts are a minor fraction of total gag RNA [54] . Analyzing the model , we found that the number of stages latently infected cells goes through the waiting state and the total waiting period before sustained induced transcription varied among individuals . This suggests that latently infected cells in different individuals may be in different states , possibly due to variations in Tat protein copy number , host gene expression or alternatively different degrees of chromatin silencing or configuration potentially dependent on the sites of HIV integration . This , in turn , would cause responses to vorinostat to be heterogeneous . Interestingly , the maximal fold increase of CA-US HIV RNA was strongly correlated with the basal level of CA-US HIV RNA before vorinostat treatment [10] . Thus , it is plausible that the basal level of CA-US HIV RNA serves an indicator of the status of latency in a patient and the ease of induction of transcription using LRAs , suggesting that future treatment strategies may be able to be tailored to individual patients . We further assessed the impact of vorinostat on the rate of loss of cells in the sustained activated state . This estimated loss rate , which serves as an upper bound on the death rate of activated cells ( as cells could lose their activated state ) , is extremely low in most individuals , suggesting that vorinostat treatment does not induce killing of transcriptionally activated cells in most participants . This is in agreement with several previous in vitro and ex vivo studies showing vorinostat activates HIV transcription in only a subset of cells and that this level of HIV transcription and protein expression does not lead to cell death [21 , 52 , 53 , 55 , 56] . Interestingly , a recent in vitro study showed that vorinostat treatment only has significant impact on HIV transcriptional activation , with the impact on translation being minimal , suggesting that HIV proteins may not be produced sufficiently to lead to virion production or to induce viral cytopathic or cytotoxic T cell mediated cell death [20] . Thus , new treatment strategies aiming at both transcriptional and translational activation of HIV may be needed to induce efficient killing of latently infected cells . In the model , we have assumed that in each participant , latently infected cells are a homogeneous population and respond to vorinostat by going through two activation steps . However , because of limited data sampling there could have been additional transient activation steps that we were unable to detect . In addition , the latent state of individual infected cells in vivo may differ [57] and it is possible that within an individual , some cells go through a different number of activation steps or have different waiting periods before becoming sustainably activated . This may give rise to the minor discrepancies between the data and the model seen in some participants ( VOR010 , VOR018 , VOR019 , VOR021 , VOR023 in Fig 3 ) . Although a model that accounts for different responses of latently infected cell subpopulations or has additional activation steps might explain the data , such a model would have more unknown parameters than our current model with only a marginal improvement in model fit . Nonetheless , the possibility that there exist different cell populations in individual patients in terms of their response to vorinostat or additional activation steps cannot be excluded . Further experiments examining the dynamics of host factors/enzymes that are responsible for transcriptional activation and inhibition under LRA treatment could validate our model , improve our understanding of the impact of vorinostat , and ultimately aid the design of treatment strategies to eradicate the latent reservoir . To conclude , our results suggest that vorinostat induces both immediate transient induction and delayed sustained induction of HIV transcription . Similar dynamic patterns of CA-US HIV RNA were also observed in clinical trials of the LRAs panobinostat and romidepsin [58 , 59] , suggesting LRAs may induce both transient and delayed transcription activation in latently infected cells in general . Therefore , designing clinical trials with frequent longitudinal sampling during both treatment and the follow-up period would help quantify the impact of LRAs . To our knowledge , our work represents the first mathematical model to assess the impact of a LRA on the dynamics of CA-US HIV RNA in vivo . Our model can be easily adapted to study other LRAs as well as combinations of these agents once data are available . In addition , our model or variants of it could be used to assess the efficacy of different candidate treatments , such as those using anti-HIV monoclonal antibodies combined with LRAs , and ultimately suggest optimal drug combinations to eliminate latently infected cells in HIV-infected individuals .
As described by Elliot et al . [10] , 20 chronically HIV-infected adults receiving at least three antiretroviral agents , having plasma HIV RNA < 50 copies per mL for at least three years ( excluding single viral ‘blips’ ) , a CD4+ T-cell count > 500 cells/μL and documented subtype B HIV-1 infection were recruited into a vorinostat trial ( ClinicalTrials . gov , NCT01365065 ) . Participants received vorinostat 400 mg orally once daily for 14 days . Levels of CA-US HIV RNA were measured in peripheral blood mononuclear cells at 0 , 2 , 8 and 24 hours , and on days 7 , 14 , 21 , 28 and 84 ( as well as on day 3 for participants VOR019-023 ) . For each blood sample , four replicate q-PCR runs were performed to measure the levels of CA-US HIV RNA . The total number of data points in each patient ranges from 35 to 44 . The ordinary differential equations ( ODEs ) describing the model are: dTdt=s−dTT− ( 1−εRT ) βVITdIdt= ( 1−εRT ) ( 1−f ) βVIT−δI1 . 44+2 ( 1−pL ) aLdLdt= ( 1−εRT ) fβVIT−dLL−aL+2pLaL−υLdLAdt=υL−dLALAdRdt=α−μR−ρRdVIdt= ( 1−εPI ) [pVI+ρRLA]−cVIdVNIdt=εPI[pVI+ρRLA]−cVNIUS=RLA+US0 In this model , target cells , T , are produced at a constant rate , s , and die at per capita rate , dT . In the absence of cART , they are infected at per capita rate , βVI , where β is a rate constant and VI is the concentration of infectious virus . The effect of reverse transcriptase inhibitors ( RTI ) is to multiply the rate of infection by the factor 1-εRT , where εRT is the effectiveness of the RTI with 0 ≤ εRT ≤ 1 . Under protease inhibitor treatment , a fraction , εPI , of produced viruses are non-infectious ( VNI ) . Upon infection with viruses , a fraction , f , of infected cells becomes latently infected , and the remaining fraction , 1-f , becomes productively infected . Productively infected cells , I , die at per capita rate , δI1 . 44 , where the power 1 . 44 models the effects of both viral cytopathicity and cell-mediated immune responses [60 , 61] . We assume that latently infected cells , L , when activated by antigen at rate a , undergo an asymmetric division in which a daughter cell remains in latency with probability , pL , and becomes productively infected and produces virus with probability , 1-pL [15] . Latently infected cells die at per capita rate , dL . We assume that vorinostat causes latently infected cells to move into a transcriptionally activated state , LA , at rate ν . The transcriptionally activated cells are lost at rate , dLA . We also assume there is a pharmacological delay , t0 , in the effect of vorinostat upon treatment initiation such that the rate of activation remains 0 when t<t0 , and it becomes ν when t0≤t≤14 days . After treatment is terminated , we assume the effectiveness declines exponentially as νe-w ( t-14 ) for t>14 days , where w is the rate at which vorinostat is cleared from the system . We also model the population average of the amount of CA-US HIV RNA ( R ) within transcriptionally activated cells . We assume that CA-US HIV RNA is produced at a constant rate , α , once vorinostat becomes effective , i . e . when t≥ t0 . CA-US HIV RNA is encapsidated and exported as virions at per capita rate , ρ , and lost by degradation and splicing at per capita rate , μ . Here , we assume that the transcriptionally activated cells do not produce mature viral particles , i . e . ρ = 0 , because recent work shows that the production of virus from cells treated with vorinostat is minimal [20 , 53] . As other LRAs may induce viral production , we leave ρ in the model . Viruses are produced from productively infected cells at rate pv and from transcriptionally activated cells at rate ρR ( note that ρ is set to 0 for vorinostat ) . Viruses are cleared at per capita rate c . The amount of CA-US HIV RNAs , US , is calculated as the sum of the basal level of CA-US HIV RNA , US0 , and the average number of US HIV RNAs per activated cell , R , multiplied by the number of activated cells , LA . In the delayed activation model , the equations describing the transient activation state , LT , the waiting state , LW , the transcriptionally activated state , LA , and the total number of CA-US HIV RNAs , US , are dLTdt=υL−dLLT−kTLTdLWdt=kTLT−dLLW−kWLWdLAdt=kWLW−dLALAUS=R ( LT+LA ) +US0 where kT and kw are transition rate constants . The ODEs describing other variables are kept the same as in the direct activation model . The ODEs describing the multistage waiting states are: dLW , 1dt=kTLT−dLLW , 1−nkWLW , 1dLW , idt=nkWLW , i−1−dLLW , i−nkWLW , i i=2 , 3 , …… , ndLAdt=nkWLW , n−dLALA where the waiting state is divided into n identical sub-states , i . e . LW , 1 , LW , 2 , … LW , n . The transition rate from one sub-state to the next is set to nkw , so that the average residence time in the waiting state , 1/kw , is the same as in the delayed activation model . The ODEs describing other variables are the same as in the delayed activation model . To test the assumption about how the population of latently infected cells is maintained , we modified the multi-stage delayed activation such that the latently infected cells proliferate at a constant rate as in Kim and Perelson [50] instead of asymmetric division in the main text . The ordinary differential equations ( ODEs ) describing the productively infected population and latently infected population are shown below and other terms in the model are kept the same as in the main text . In this model , the latently infected cells becomes activated at per capita rate a and proliferate at per capita rate r . We set a = 0 . 0088 day-1 according to [62] and r = 0 . 0183 day-1 , such that the half-life of the latent reservoir is 44 months as estimated before [63] . The fixed parameter values are based on prior work to yield a baseline state with a plasma viral load of 5 copies/mL and a latently infected cell population of ~2 cells/million cells ( Table 1 ) . The production and death rates of uninfected CD4+ T cells were chosen to yield a baseline CD4 count similar to the average level in patients in Elliot et al . [10] . Changes in the values of the fixed parameters that govern the dynamics of target cells and productively infected cells do not impact the estimation of the fitted parameters ( S10 Fig ) . This is because cART is so effective that the contribution of new infections to the latent reservoir is negligible over the time period we study here . We also tested the robustness of the parameter estimates against changes in the fixed parameters that govern intracellular HIV transactivation in latently infected cells and to the structure of the model with regard to how latently infected cells proliferate ( S7–S9 Figs ) . The CA-US HIV RNA production rate in a productively infected cell , αI , is calculated based on the derivations in Ref . [28] . The basal transcription rate in eukaryotic cells is estimated to be approximately 40 nucleotides/s [64] . The HIV genome has around 9500 nucleotides . Thus , the basal production rate of CA-US HIV RNA production can be estimated as 40/9500≈0 . 0047 s-1≈406 day-1 . In the presence of Tat , the transcription is upregulated to 100 fold [65] . Then , the production rate of CA-US HIV RNA in a productively infected cell , α0 , can be calculated as approximately 40 , 000 transcripts day-1 . This maybe a minimal estimate if productively infected cells live approximately one day while producing virus , as Chen et al . [32] have estimated that a productively infected cell produces between 40 , 000 and 55 , 000 virions over its lifespan . To fit the direct activation model to the data , we varied 5 unknown model parameters: the production rate of CA-US HIV RNA in a transcriptionally activated cell ( α ) , the death rate of transcriptionally activated cells ( dLA ) , the rate of transcriptional activation ( ν ) , the initial delay of vorinostat effectiveness ( t0 ) and the baseline US HIV RNA level ( US0 ) . To fit the delayed activation model and the multistage delayed activation model , we allow 2 additional parameters to be estimated ( together with the 5 parameters in the direct-activation model ) : the transition rate , kT , from the transiently activated state to the waiting state , and the transition rate , kw , from one sub-state of the waiting state to the next . To estimate the parameter values in each model , we first calculate the residual sum of squares ( RSS ) between model predicted log CA-US RNA level and log transformed data , and then minimize the RSS using the Nelder-Mead algorithm [66] . 1 , 000 individual fits were performed for each model starting from parameter values randomly sampled within biologically plausible ranges . The parameter values with the smallest RSS among the 1 , 000 fits are taken as the best-fit parameter values for each model . We perform model selection using the corrected Akaike information criterion ( AICc ) to account for the low number of data points ( ranges from 35 to 44 ) for each patient [67] . The AICc score is calculated as AICc=nlog ( RSSn ) +2Knn−K−1 where n is the number of data points and K is the number of fitted parameters . When comparing models , the model with the lowest score is the best model , although small difference in AICc scores , e . g . ≤ 2 , is not significant [67] .
|
Combination antiretroviral therapy ( cART ) for HIV infection must be taken for life due to the existence of long lived latently infected cells . Recent efforts have focused on developing latency reversing agents to eliminate latently infected cells by activating HIV production . In this work , we assess the impact of a latency reversing agent , vorinostat , by fitting dynamic models to data from a clinical trial . Results show that vorinostat treatment induces HIV transcription transiently and that the sustained induction of HIV transcription may depend on the temporal impact of vorinostat on host gene expression . Our results also suggest that vorinostat treatment is not sufficient to induce killing of latently infected cells in a majority of HIV-infected individuals on cART .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[] |
2015
|
Modeling the Effects of Vorinostat In Vivo Reveals both Transient and Delayed HIV Transcriptional Activation and Minimal Killing of Latently Infected Cells
|
The treatment of HCV infection has seen significant progress , particularly since the approval of new direct-acting antiviral drugs . However these clinical achievements have been made despite an incomplete understanding of HCV replication and within-host evolution , especially compared with HIV-1 . Here , we undertake a comprehensive analysis of HCV within-host evolution during chronic infection by investigating over 4000 viral sequences sampled longitudinally from 15 HCV-infected patients . We compare our HCV results to those from a well-studied HIV-1 cohort , revealing key differences in the evolutionary behaviour of these two chronic-infecting pathogens . Notably , we find an exceptional level of heterogeneity in the molecular evolution of HCV , both within and among infected individuals . Furthermore , these patterns are associated with the long-term maintenance of viral lineages within patients , which fluctuate in relative frequency in peripheral blood . Together , our findings demonstrate that HCV replication behavior is complex and likely comprises multiple viral subpopulations with distinct evolutionary dynamics . The presence of a structured viral population can explain apparent paradoxes in chronic HCV infection , such as rapid fluctuations in viral diversity and the reappearance of viral strains years after their initial detection .
An estimated 3% of the global human population has been infected with the hepatitis C virus ( HCV ) , many of whom are unaware of their infection status . Unlike other members of the virus family Flaviviridae , HCV causes acute and chronic infection in humans . Symptoms of acute infection are typically mild and , despite the early response mounted by the immune system , viral clearance occurs in only 15–20% of untreated cases . In the remaining individuals who become chronically-infected , the virus can , over many years , cause liver cirrhosis , hepatocellular carcinoma , and other related diseases . Genetically , HCV is a very diverse virus and up to 50% of nucleotide sites may vary among HCV strains belonging to different genotypes . The high genetic diversity of HCV is the product of both a high rate of molecular evolution and a proposed long-term association of the virus with human populations [1] . Prior to the discovery in 2003 of an atypical HCV genotype 2 strain that can readily replicate in hepatoma cell lines [2] , the development of HCV-specific antiviral drugs was comparatively slow and , until recently , standard drug treatment for HCV infection was non-specific and involved long courses of interferon and ribavirin . However , newly approved direct-acting antiviral ( DAA ) drugs that target the HCV life cycle are highly effective , leading to viral clearance in >90% of patients within 12 to 24 weeks of treatment [3–7] . Interestingly , these clinical successes have been achieved despite relatively little being known about the in vivo dynamics of HCV replication , host cell infection , and evolution . Most of our understanding of HCV replication behaviour within infected individuals has come from mathematical models of virus kinetics [8] , which are typically fitted to measurements of viral load from longitudinal samples of peripheral blood . Simple models that employ a mass action mechanism of infection can explain the two-phase decline in HCV viral load following interferon-based drug therapy and have demonstrated ( i ) a high turnover of virions in peripheral blood [9] , ( ii ) a high variance among patients in the mean lifespan of infected cells , ranging from 2–70 days [9] , and ( iii ) that approximately 3% of virions in serum result from extra-hepatic replication [10] . More complex viral load dynamics , including triphasic decline and the failure of drugs to fully eradicate the virus , have been explained by adding proliferation of infected and non-infected hepatocytes to the model [11] . However , it is unclear whether division of infected hepatocytes requires active virus replication , or whether HCV is passively transferred between parent and daughter cells . An alternate hypothesis , that HCV persists in exceptionally long-lived cells during chronic infection , has been discounted [11 , 12]; yet it is known that uninfected hepatocytes have a significantly slower turnover than the main target cells of HIV ( CD4+ T lymphocytes ) and are thought to survive for years [13] . Further , despite the important insights revealed by mass-action models of HCV virus kinetics , they cannot fully reconcile all aspects of chronic HCV infection in vivo . This includes observations of cell-to-cell virus transmission [14–16] , foci of infection within the liver [17–21] , and viral re-emergence after drug therapy has temporarily reduced viremia in peripheral blood to undetectable levels [22–25] . Standard models of virus infection kinetics were initially developed in the context of HIV-1 infection [26 , 27] , in which virions sampled from peripheral blood appear to be representative of the contemporaneous population of actively-infected host cells . Whether this assumption is also true for HCV is difficult to ascertain because sampling of liver tissue , the primary site of replication , is invasive and rarely repeated longitudinally during infection . Molecular and mathematical analysis of individual liver biopsy samples indicates that HCV infection spreads locally within the liver and is likely to be seeded randomly by viruses from peripheral blood [28] . Further , analysis of virus gene sequences obtained from transplantation patients and from explant livers suggests that hepatic and extra-hepatic viruses can be genetically distinct and may form different sub-populations [29–32] . Recent experimental and clinical studies have suggested a more complex model of HCV replication , involving cell-to-cell transmission dampened by localized immune responses , as well as detectable virus replication in quiescent hepatocytes ( i . e . cells that are differentiated but in a resting state ) and non-hepatic reservoirs [10 , 33] . Importantly , these observations suggest that viral replication dynamics during HCV infection may be decoupled , at least in part , from host cell turnover . Viral gene sequences , sampled longitudinally through time from chronically infected patients , constitute a valuable and independent source of information about replication dynamics . The high mutation of HCV means that viral genomes accrue ~0 . 3 to 1 . 2 nucleotide substitutions per cell infection [34 , 35] . As a consequence , the genetic divergence between viruses sampled throughout infection will be influenced by both the mode and tempo of cell-to-cell infection . Investigation of longitudinally sampled virus sequences has proven useful for HIV-1 infection , leading to insights regarding the size of viral bottlenecks at transmission [36–38] , correlations between viral evolution and clinical outcomes [39 , 40] , and the relationship between within- and among-host virus evolution [41] . Studies of serially-sampled HCV sequences have also indicated a link between viral evolution and disease progression . First , the level of HVR1 diversity during acute infection has been associated with whether a patient successfully clears the virus [42] . Second , greater genetic diversity and synonymous divergence is observed in viral populations sampled from rapid progressors , which suggests that faster disease progression is associated with shorter viral generation times [43] , as has also been reported for HIV-1 [40] . Nonetheless , these observations are based on studies with limited number of patients and viral sequences , and which used only simple summary statistics ( e . g . pairwise diversity ) during analysis . To better understand the replication dynamics of HCV during infection , we undertake a comprehensive analysis of HCV evolutionary dynamics during chronic infection . We use statistically powerful Bayesian phylogenetic approaches to test hypotheses concerning the diversity and divergence through time of within-host HCV populations . In total , we analyse more than 4000 viral gene sequences obtained from 15 patients , sampled over 100 different time points . We compare our HCV results to those obtained from nine comparable HIV-1 infected subjects , and discover differences between the evolutionary dynamics of the two viruses during chronic infection . Most notably we observe significant heterogeneity in the molecular evolution of HCV , both among patients and over time , which contrasts with more consistent trends in HIV-1 infected patients . Our results support a complex model of HCV replication dynamics during chronic infection that reconciles apparent paradoxes observed in the natural history of this infection
The amount of diversity among viruses sampled at each time point is shown in Fig 1A , where the size of each circle is proportional to the mean pairwise sequence diversity ( MPD ) for that time-point . If we average the MPD scores across all subjects and time points then we obtain 0 . 009 changes/site for the HCV untreated group and 0 . 013 changes/site for the HCV treated group . The overall genetic diversity is higher for HIV-1 patients ( average MPD across all time points = 0 . 029 ) . We also found interesting differences between HIV-1 and HCV patients in the distribution of viral diversity among time points . Specifically , we find that the distribution of MPD scores for the HIV-1 group is much more symmetric ( skewness = 0 . 37 ) than for the two HCV cohorts ( skewness = 1 . 38 and 2 . 08 , for HCV treated and untreated subjects , respectively; Fig 1B ) . The strong positive skew observed for both HCV groups indicates that , during infection , HCV exhibits more extreme occasional shifts to high viral diversity , despite the fact that , on average , viral population diversity is low compared to HIV-1 infections . Treatment periods ( interferon and ribavirin ) in the HCV treated group do not appear to correlate with lower genetic diversity , although this cannot be formally tested because the relative timing of sampling times and treatment periods varied among subjects . To characterize change in the genetic structure of the within-host viral population we calculated Tajima’s D statistic for each time point in each patient ( Fig 1C–1E ) . This statistic varies significantly over the course of infection in HCV patients , with rapid fluctuations even between immediately adjacent time points ( Fig 1C and 1D ) . This demonstrates substantial changes in the frequency distribution of polymorphic sites . In other words , the viral population shifts back and forth between carrying many common polymorphisms ( D>0 ) and carrying many unique low-frequency variants ( D<0 ) . In contrast , the genetic structures of within-host HIV-1 populations are more stable through time and predominated by rare or low-frequency polymorphisms ( D<0; Fig 1E ) . For HIV , Tajima’s D statistic gradually rises through time but rarely exceeds zero ( Fig 1E ) . Consequently , when comparing the distributions of Tajima’s D values among the three cohorts , for both untreated and treated HCV patients we observed significantly greater variance and positive skew in Tajima’d D values compared to HIV-1 patients ( S1 Fig ) . In addition , for all groups of subjects , Tajima’s D values for each time point are positively correlated with viral genetic diversity ( S2 Fig; p<0 . 001 for all three groups; correlation test ) , such that when diversity is low , shared mutations are more likely to be rare . Theory predicts both MPD and Tajima’s D values will be low when a sampled population has recently experienced an expansion , either due to rapid population growth or a recent selective sweep . High values of both statistics are predicted when population structure or fluctuating selection maintains genetic diversity in a population . The mean rates of molecular evolution for each subject , as estimated using the lognormal relaxed molecular clock model , are shown in Fig 2A . The mean rate is notably lower in drug-treated HCV subjects than in the HCV untreated group ( Fig 2A; Mann-Whitney U test , p < 0 . 05 ) . The evolutionary rate is in general higher for HIV-1 than for HCV ( we place no emphasis on this comparison because the HIV-1 and HCV genome regions are not homologous ) . Fig 2B shows , for each patient , the degree to which the viral evolutionary rate varies during infection , which is quantified using the coefficient of variation ( COV ) of the relaxed molecular clock . Two patterns are evident . First , the COV statistic is more variable among HCV subjects than among HIV-1 subjects . Second , extremely high levels of viral rate variation are observed in some HCV subjects , but not in HIV-1 subjects ( estimated COV>1 for seven HCV patients , but only one HIV-1 patient ) . The values in some HCV subjects are unusually high ( COV>1 . 75 ) and represent exceptional rate variation among lineages ( Fig 2B ) . To test whether these estimates were robust to model misspecification , we implemented a new relaxed molecular clock that assumes that branch rate scalars follow a more flexible skew-normal distribution . Unlike the standard lognormal molecular clock , the skew-normal molecular clock allows the distribution of evolutionary rates among branches to be either positively or negatively skewed , or non-skewed . Both the skew-normal and lognormal molecular clocks give similar parameter estimates ( Fig 2; filled and open circles indicates estimates under log-normal and skew-normal rate distribution , respectively ) . Furthermore , the shape parameter of the skew normal model differed significantly from zero in only one patient ( S3 Fig ) , indicating that the distribution of among-branch rate variation was approximately symmetric . To explore why rates of molecular evolution are lower in the HCV treated group than in the untreated group ( Fig 2A ) we used a partition model to estimate rates of evolution for first and second codon positions ( 1+2cp ) and third codon positions ( 3cp; Fig 3 ) . These rates contain information about the action of positive and negative selection because the majority of mutations at 1+2cp and 3cp sites are , respectively , non-synonymous and synonymous . This approach is a good proxy for dN/dS values estimated with codon substitutional models , which for large temporally sampled datasets can be difficult to obtain due to slow MCMC convergence . However , we note that , unlike dN/dS ratios , the ratio of codon position rates cannot be used to formally test for positive selection ( Table 1 ) . Amongst HCV subjects , 3cp rates ( open squares ) are largely similar between the treated group and untreated group , whereas the 1+2cp rates ( filled squares ) are lower in HCV subjects that have received treatment ( Mann-Whitney U test , p <0 . 01; Fig 3 ) . Thus the reduced overall rate of virus evolution ( Fig 2A ) in the HCV treated group appears to be caused by reduced evolution at 1+2cp sites ( Fig 3 ) , suggesting that drug-therapy has reduced the ability of the viral population to undergo adaptive fixation ( Table 1 ) , but has not significantly reduced the fixation of 3cp changes that are likely to be selectively neutral ( see also S4 Fig ) . In contrast to HCV , seven of the HIV-1 subjects had a higher estimated evolutionary rate at 1+2cp sites than at 3cp sites ( Table 1 and Fig 3 ) , indicating greater positive selection and/or less negative selection on the HIV-1 sequences . Many previous studies have demonstrated adaptation of the HIV-1 env gene during infection due to positive selection ( e . g . [39 , 44 , 45] ) . For HIV-1 subjects , both the 1+2cp and 3cp rates are correlated with total evolutionary rate , whereas for HCV subjects , only the 1+2cp rate exhibits such a correlation ( S4 Fig ) . There are several notable differences between the estimated time-scaled phylogenies from HCV subjects compared to those from HIV-1 subjects . One representative phylogeny from each patient group is shown in Fig 4 , and all phylogenies are presented in S5–S7 Figs . The vertical dashed lines indicate yearly intervals in each patient phylogeny . Firstly , during HCV infection distinct lineages can persist for extended periods of times; in Fig 4A and 4B this can be between 7 and 9 years , respectively . To quantify this we calculate the ratio of external to internal branch lengths for the two HCV phylogenies in Fig 4 . The mean ratios are significantly less than one: 0 . 49 ( 95% HPD = 0 . 38 , 0 . 59 ) for the untreated HCV patient and 0 . 56 ( 0 . 48 , 0 . 66 ) for the treated HCV patient . In contrast the mean ratio for the HIV-1 phylogeny is 1 . 94 ( 1 . 65 , 2 . 25 ) , indicating that viral lineage turnover is faster ( Fig 4C ) . Further , the persistent lineages observed in HCV infection may go undetected for many years; hence the number of divergent lineages that are actually detected at any given sampling time may vary . Secondly , HCV sequences sampled from the same time-point on the same lineage tend to share a very recent common ancestor , giving rise to a distinctive phylogenetic pattern of long internal branches punctuated by ‘bursts’ of closely related or identical sequences . When only a single lineage is sampled at a given time point , this leads to a low observed MPD and a strongly negative value of Tajima’s D . This indicates that all the HCV sequences belonging to that lineage represent a viral subpopulation that has recently expanded or been subjected to a recent population bottleneck . However , when multiple HCV lineages are observed at a given time-point , then the sample MPD for that time-point is , by definition , higher and the corresponding Tajima’s D is typically closer to zero or positive . This association between phylogenetic structure and genetic diversity explains both the results for HCV in Fig 1 and the correlation between MPD and Tajima’s D ( S2 Fig ) . In comparison , sequences sampled from a given time-point during HIV-1 infection share a comparatively recent common ancestor on the persistent ‘backbone’ of the phylogeny ( Fig 4C ) . Furthermore , the high ratios of external to internal branch lengths in the HIV phylogenies are expected by theory if the viral population is unstructured and undergoing recurrent selective sweeps . This result explains the consistently negative Tajima’s D values and the steady changes in MPD observed for HIV-1 in Fig 1 . Changes in relative population genetic diversity during infection are illustrated by Bayesian skyline plots , which are superimposed over the phylogenies in Fig 4 ( the timescale of the skyline plots and phylogenies are shared ) . Note that the skyline plot represents the total diversity of the entire within-host viral population through time , including lineages that are inferred to be present but unsampled , whereas the MPD values in Fig 1 represent only the diversity that is actually sampled at each time point . There are no clear trends among patient groups in the dynamics of viral population diversity , although significant declines are perhaps more common in the HCV treated group than in either of the untreated groups ( S5–S7 Figs ) .
Patterns of viral genetic divergence and diversity during chronic infection depend on the structure and dynamics of the replicating viral population , and therefore they provide a source of information about infection kinetics that is independent from and complementary to mathematical models of longitudinal viral load measurements [8–11] . Further , molecular clock approaches like those used here may better resolve complex evolutionary dynamics than analyses of sequence summary statistics , which uses data less efficiently [46] . The results of our evolutionary analyses show that intra-patient HCV evolution is exceptionally heterogeneous , both within and among different subjects , compared with intra-patient HIV-1 evolution , and that this variation is present in both treated and untreated HCV-infected subjects . Specifically , for HCV we find ( i ) extreme heterogeneity in the rate of molecular evolution in some patients; ( ii ) a lower rate of non-synonymous change in patients that received interferon-treatment; ( iii ) significant fluctuations in viral genetic diversity through time; and ( iv ) unusual phylogenetic topologies containing multiple distinct lineages that coexist for long periods of time , combined with ‘bursts’ of closely-related sampled variants . These observations are not consistent with a well-mixed viral population with homogenous infection dynamics , but instead suggest that HCV infections are comprised of multiple sub-populations with distinct evolutionary and replication behaviours . While rates of HCV molecular evolution in vivo are comparable to those estimated for other RNA viruses ( e . g . HIV-1 and influenza ) [47] , we observe very high among-lineage rate heterogeneity only for some HCV infections . This suggests , at the very least , that circulating HCV lineages do not all accumulate substitutions in the same manner . Rates of viral sequence divergence are determined by mutation rates , population sizes , generation times , and mutational selection coefficients . Crucially , the latter three factors can only vary within an individual if the within-patient viral population is split into distinct subpopulations with separate dynamics . There is a growing body of independent evidence that indicates the presence of an HCV population structure in the liver . The existence of genetically distinct viral sub-populations ( compartmentalization ) has been demonstrated for viruses isolated from ( i ) plasma versus liver [29 , 30 , 48–53] , ( ii ) different locations within the same liver [32] , and ( iii ) between non-tumourous liver tissue versus tumour-associated liver tissue [32 , 54–57] . Experimental studies demonstrate that , within the liver , HCV tends to be localized to specific foci of infection [18] . If cell-to-cell transmission is more efficient than transmission via free virions , then models of HCV infection should incorporate local viral replication , adaptation and spread within the organ [58] . Viral population structure may also exist outside the liver , as HCV genetic compartmentalisation has been reported ( i ) among cirrhotic liver samples [50] , ( ii ) between plasma and PBMCs [29 , 30 , 49 , 53 , 59–61] , ( iii ) between PBMCs and the liver [29–31 , 49 , 59] , ( iv ) between liver and perihepatic lymph nodes [53] , and ( iv ) in the brain [62] . The heterogeneity in HCV evolutionary rate we report here is consistent with these experimental results , and we posit that it arises from distinct sub-populations of HCV ( hepatic or extra-hepatic ) whose replication is modulated by local fluctuations in host cell availability and turnover , and/or by anti-viral immune responses . The modulation of replication within separate viral sub-populations can also readily explain the unusual HCV phylogenetic topologies . Specifically , HCV lineages that are present in the body , but which are not directly observed in peripheral blood for prolonged periods of time , might represent sub-populations that are not shedding virions into circulation , either because they are replicating slowly , or because they are transmitting via cell-to-cell contact . Cell-to-cell transmission may allow HCV to replicate in the presence of neutralizing antibodies [14 , 58] and is again consistent with the detection of hepatic foci of infection [18 , 20] . One recent study has found that in vitro DAA drug-resistant viruses predominantly spread by this route of transmission [16] . Further , the ‘bursts’ of closely related sequences that we observe are consistent with the recent and rapid growth of previously restricted viral subpopulations . The causes of these bursts are unknown; possible explanations include viral adaptation or the local deterioration of immune control . Viral population structure and host immune responses could also account for the puzzling fact that most cells in the liver are uninfected [18 , 35 , 63] . This is surprising given that viral loads in serum are high ( 105–107 virions/mL ) , and that transplant livers are rapidly re-infected following transplantation . If the viral population is strongly structured then chronic infection requires only the establishment of a few long-lasting sub-populations that are not removed by host immune responses . It is not known whether the distinct lineages observed during HCV infection are antigenically distinct . If they are , this antigenic variation may contribute to the creation and maintenance of a persistent infection . The highly structured nature of HCV intra-host genetic diversity also has consequences for the evolutionary analysis of chronic infection . Specifically , it means that samples of HCV diversity from peripheral blood do not adequately characterise the genetic diversity of the infection as a whole [46] . We find that statistics of sample diversity ( MPD and Tajima’s D ) vary substantially through time within HCV subjects , but are more consistent ( Tajima’s D ) or less skewed ( MPD ) for HIV-1 subjects . Molecular clock phylogenetic analyses show that this is due to significant among lineage rate variation . Whilst PCR primers might fail to amplify some within-host HCV lineages , it is difficult to conceive how differential amplification might cause strong fluctuations in viral diversity through time within a single patient . Given that the number of sequences per time point in our data sets is comparatively small ( range n = 18 to n = 88 ) it could be argued that the intermittent detection of HCV lineages in peripheral blood is solely a consequence of sampling uncertainty . To explore this , let us suppose there are two lineages , in which case the probability of detection can be determined by the binomial distribution . If n = 40 and sampling is random , then a lineage whose population frequency is 5% will be detected at 87% of timepoints , but a lineage whose frequency is 0 . 5% will be seen at only 4% of timepoints . Lineages at frequencies between ~0 . 5% and ~5% are therefore likely to be intermittently detected in our study . However the key observation that HCV lineages are often dominant at one timepoint , but rare or absent at a later timepoint , is not an artefact of sampling uncertainty because the sample sizes used in our study will almost certainly detect all lineages whose frequencies exceed 15% . We also note that the evolutionary patterns in HCV subjects reported here have come from different cohorts generated using different sequencing approaches , and similarly structured within-host HCV phylogenies have been noted elsewhere [64–66] . Although we cannot pinpoint the anatomical locations of HCV genetic sub-populations , these are likely to be sites within the liver and/or extra-hepatic compartments such as PBMCs or the central nervous systems [30 , 62] . Cross-sectional genetic analysis of HCV diversity in explanted livers may help to address this question . Low recombination in HCV [67] could also potentially explain differences between the within-host molecular evolution of HIV-1 and HCV . Specifically , infrequent recombination can lead to stronger clonal interference , whereby beneficial mutations on different genetic background compete for fixation [68] , resulting in longer times to fixation of mutations and increased diversity at each sampling time . While this effect is likely to shape HCV molecular evolution , and may increase the length of some internal phylogenetic branches , it cannot account for the alternating appearance of divergent lineages in peripheral blood after long periods of absence . Further , low recombination would lead to complete selective sweeps and is therefore inconsistent with the long-term persistence of multiple lineages ( e . g . > 20 years in one treated HCV subject; see S6C Fig ) observed in our HCV cohorts . The hypothesis that strong HCV population structure and lineage rate variation contributes to viral persistence has consequences for the new DAAs that are highly successful in treating HCV infection . Although these treatments drastically reduce treatment times , a longer-follow up of patients may be prudent if there is a longer-term risk of viral relapse from unsampled reservoir populations within the body . A recent study has found that viral persistence is prevalent in patients who have spontaneously resolved the virus [69]; HCV RNA was detected in ~70% of patients ~6 years after clearing the virus . Furthermore , samples collected from PBMCs between 5 to 20 years after initial detection of HCV supports ongoing viral replication despite patients appearing non-viraemic [69] . Very late HCV breakthroughs have been reported from some clinical trials using DAA therapy [23–25] . Although these instances are infrequent , they do highlight that our understanding of the persistence of HCV at low levels is inadequate and requires further investigation . Lastly , the high evolutionary heterogeneity of HCV within hosts has important implications for molecular epidemiological analyses of HCV genetic diversity at the among-host level . In such studies each infected individual is typically represented by a single sequence that is interpreted as the ‘consensus’ of the within-host viral population at the time of sampling . For HCV , the intermittent detection in sera of diverse lineages means that the consensus sequence obtained may be highly dependent on when sampling occurs , and may not be representative of the virus that is transmitted . Crucially , this could explain in part why among-host HCV molecular clock phylogenies have proven difficult to calibrate from longitudinal samples of HCV sequences [70] .
We analysed a total of 15 HCV infected subjects . Subject and sampling information is provided in Tables S1-3 . Subjects from previously published studies were only included if HCV sequences were sampled longitudinally for at least 5 years . HCV sequences were obtained from seven untreated patients previously reported by [43 , 71] ( referred to as U1-U7 in this study ) . These subjects acquired HCV infection either perinatally ( U1-3 ) [43] or via transfusion ( U4-7 ) [71] . The date of infection was known and thus all time points represent time since infection . To enable direct comparison with other subjects , HCV sequences sampled during acute infection were removed ( U1 , U4 , U6: time points <3 months; U5: time points <9 months; U7: time points <8 months ) . The sequences represent partial E1/E2 gene sequences corresponding to positions 1308–1835 relative to the H77 HCV reference genome . Alignments from these patients included a total of 2246 sequences ( range 235–418 sequences per subject ) and an average of 8 . 7 time points per subject ( range 6–12 time points ) that cover an average duration of sampling of 13 . 6 years ( range 7 . 4–23 . 3 years ) . Sequences from an additional 8 subjects were obtained from sequential serum samples from a cohort of HCV patients from Bari , Italy . The hypothesised route of transmission was nosocomial infection following surgery: no other risk factors were observed and all patients were anti-HCV negative at the time of surgery , however none of them received a blood transfusion . These subjects ( denoted T1-8 ) were treated with interferon and ribavirin; all subjects received at least one period of therapy during the study , although duration and regimen varied among subjects . Sequences from these subjects were generated by amplifying segments of the E1/E2 gene region using multiple different primer pairs that spanned the hyper-variable region 1 ( HVR1 ) . Full sequencing details for this cohort can be found in Supporting Information ( S1 Text ) . Sequences were trimmed to match those obtained from patients U1-7 and corresponded to positions 1320–1799 relative to the H77 HCV reference genome . At least 18 clonal sequences were generated per time-point . Alignments from subjects T1-8 included a total of 1980 sequences ( range 132–395 per subject ) , with an average of 7 . 3 time points per subject ( range 4–10 ) covering an average of 7 . 9 years of infection ( range 5 . 2–8 . 8 years ) . The HVR1 region was targeted for sequencing in both untreated and treated HCV cohort as it is the most diverse region in the HCV genome , and consequently contains the strongest phylogenetic signal compared to other , more conserved genomic regions . A comparable set of previously published sequences from a cohort of untreated HIV-1 infected subjects ( HIV1-9 ) was analyzed concurrently [72] . All subjects were infected with subtype B and sequences represented the C2-V5 region of the gp120 gene ( corresponding to positions 7023–7286 in the HXB2 HIV reference genome ) . The total number of HIV-1 sequences was 1028 ( range 52–160 per subject ) , with an average of 11 . 7 time points per subject ( range 6–15 ) spanning an average of 8 . 2 years of infection ( range 6 . 1–11 . 2 ) . To verify and subtype the HCV sequences , an alignment was created containing the HCV sequences from all 15 subjects , plus reference sequences from each of the major HCV subtypes and genotypes . A neighbour-joining tree was reconstructed under the HKY nucleotide substitution model using MEGAv5 . 0 [73] . Two hundred bootstrap replicates were used to assess the robustness of the tree topology . Sequences from each subject clustered with each other , and not with sequences from other subjects , with high bootstrap support . In the untreated cohort , subjects were singly infected with subtypes 1a , 1b , and 4d , while in the treated cohort all patients were infected with subtype 1b . The genetic diversity of the intra-host viral population at each time point in each subject was estimated by calculating mean pairwise genetic distances among sequences using a Tamura-Nei substitution model with gamma distributed rates , as implemented in MEGA5 . 0 [73] . We also calculated Tajima’s D statistic for each sampling time in each subject , using DNAsp [74] . Tajima’s D statistic describes the relative frequency of common versus rare polymorphisms in the sample , and consequently describes whether the sample phylogeny is star-like ( long external branches ) or structured ( long internal branches ) . Tajima’s D is expected to be zero under a null model of constant size population with no natural selection or population structure . Negative D values indicate an excess of rare polymorphisms compared to this null model , which may result from a recent selective sweep or population growth . Positive D values indicate an excess of common polymorphisms , which may be caused by population contraction , or population structure , or by fluctuating selection . Rates of within-host molecular evolution ( divergence rates ) were investigated using the Bayesian Markov chain Monte Carlo framework implemented in BEAST v . 1 . 8 [75] . An initial set of model selection analyses were undertaken to explore different coalescent and molecular clock models ( in each case the codon-structured SDR06 nucleotide substitution model was used ) . Simple coalescent models ( constant size and exponential growth ) failed to converge for some HCV datasets , so final analyses were performed using the Bayesian Skyline coalescent model . Preliminary analyses indicated significant among-branch rate heterogeneity so a relaxed uncorrelated molecular clock was used . Analyses were first performed using the standard log-normal distribution model , for which the among-branch rate distribution is negatively skewed . However , we were concerned that this model may not adequately capture the rate variation in within-host HCV evolution . Therefore we also implemented a new molecular clock model in BEAST 1 . 8 with a skew-normal distribution of among-branch rate variation , which allows the among branch rate distribution to be either positively or negatively skewed , or unskewed ( see S2 Text for example XML code ) . Evolutionary rates were also estimated separately for ( i ) combined 1st and 2nd codon positions ( 1+2cp ) and ( ii ) 3rd codon positions ( 3cp ) , using a log-normal molecular clock model . MCMC convergence was generally slow and chain length varied between 100–200 million generations . Chains were sampled regularly to yield 10000 samples . Multiple independent runs were undertaken to ensure adequate mixing and stationarity had been achieved , as diagnosed using trace plots and effective sample sizes . Our historical data sets were generated using clonal Sanger sequencing and contain far fewer sequences per time point ( n = 18–88 ) than could be generated using modern next-generation sequencing ( NGS ) platforms ( 100s or 1000s of sequences per time point ) . To explore the potential effects of this on our estimates of statistics of viral genetic diversity , we simulated the process of undersampling upon previously published NGS datasets for both chronic HIV and HCV infections . We looked for NGS within-host data sets within which we could identify non-overlapping regions of varying genetic diversity that were 350-400nt length and which were represented at depth of 500 reads or greater . Suitable HIV data was found in Zanini et al [76] and Dialdestoro et al [77] , and comparable HCV data was obtained from Lu et al [78] . We randomly subsampled these NGS datasets to simulate the effects of undersampling . Specifically , in each case , we generated 100 randomly subsampled datasets containing 5 , 10 , 12 , 14 , 16 , 18 , 20 , 40 , 60 , 80 , and 100 sequences . For each replicate subsample we estimated mean MPD and Tajima’s D in exactly the same way as for the real data ( see above ) . These results are summarized in S8 and S9 Figs . In all cases , the variability and uncertainty in estimates of MPD and Tajima’s D drops quickly as sample size ( n ) increases above 10 . In our data sets , sample sizes per timepoint range from n = 18 to n = 88 ( shown in S8 and S9 Figs as red dashed lines ) . In this range of sample sizes , estimates of MPD and Tajima’s D are close to those obtained from the full ( non-subsampled ) dataset . In general , variance in estimates of these statistics stabilises between n = 5 and n = 18 sequences , and this is seen in both low and high diversity genome regions . This indicates that our estimates of MPD and Tajima’s D ( Fig 1 ) are very similar to those that would be obtained from NGS data sets comprising hundreds or thousands of reads , and that the observed variation in these statistics among time points is not due to sampling uncertainty ( or small sample sizes ) ; instead the variation is due to real changes in the viral population . While NGS datasets would undoubtedly reveal many more rare variants , such variants have very little effect , by definition , on statistics that summarise the genetic composition of the population as a whole . HCV isolates were obtained from adult patients with diagnosis of acute hepatitis C followed at the Clinic of Infectious Diseases , University of Bari . The study was approved by the local Ethical Committee ( EC University of Bari ) and a written informed consent was obtained from each patient .
|
Our knowledge of HCV within-host evolution is substantially limited , which is surprising given that highly successful therapies against the virus have been developed . Key aspects of HCV infection , such as rapid fluctuations in viral diversity and the reappearance of viral strains years after their initial detection , remain unexplained . To better understand this problem , we analyse viral sequences from HCV-infected patients sampled over several years . Our findings suggest that the replication dynamics during chronic HCV infection are distinct from those of HIV-1 , and dominated by the co-circulation of multiple viral strains . Although a major difference between the two chronic-infecting viruses is the level of recombination , our results indicate that HCV within-host evolution is most likely to be shaped by a structured viral population . Crucially , our study shows that HCV sampled from blood does not fully represent the within-host viral population at that time . This may have important implications for HCV treatment , especially in patients that have seemingly cleared the virus , as well as for molecular epidemiology studies investigating HCV transmission .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"taxonomy",
"organismal",
"evolution",
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"evolutionary",
"biology",
"hepacivirus",
"pathogens",
"population",
"genetics",
"microbiology",
"retroviruses",
"viruses",
"immunodeficiency",
"viruses",
"phylogenetics",
"data",
"management",
"rna",
"viruses",
"microbial",
"evolution",
"population",
"biology",
"computer",
"and",
"information",
"sciences",
"evolutionary",
"rate",
"medical",
"microbiology",
"hiv",
"microbial",
"pathogens",
"hepatitis",
"c",
"virus",
"hepatitis",
"viruses",
"hiv-1",
"viral",
"replication",
"evolutionary",
"systematics",
"viral",
"evolution",
"flaviviruses",
"virology",
"viral",
"pathogens",
"genetics",
"biology",
"and",
"life",
"sciences",
"lentivirus",
"evolutionary",
"processes",
"organisms"
] |
2016
|
Exceptional Heterogeneity in Viral Evolutionary Dynamics Characterises Chronic Hepatitis C Virus Infection
|
Buruli ulcer ( BU ) is a slowly progressing , necrotising disease of the skin caused by infection with Mycobacterium ulcerans . Non-ulcerative manifestations are nodules , plaques and oedema , which may progress to ulceration of large parts of the skin . Histopathologically , BU is characterized by coagulative necrosis , fat cell ghosts , epidermal hyperplasia , clusters of extracellular acid fast bacilli ( AFB ) in the subcutaneous tissue and lack of major inflammatory infiltration . The mode of transmission of BU is not clear and there is only limited information on the early pathogenesis of the disease available . For evaluating the potential of the pig as experimental infection model for BU , we infected pigs subcutaneously with different doses of M . ulcerans . The infected skin sites were excised 2 . 5 or 6 . 5 weeks after infection and processed for histopathological analysis . With doses of 2×107 and 2×106 colony forming units ( CFU ) we observed the development of nodular lesions that subsequently progressed to ulcerative or plaque-like lesions . At lower inoculation doses signs of infection found after 2 . 5 weeks had spontaneously resolved at 6 . 5 weeks . The observed macroscopic and histopathological changes closely resembled those found in M . ulcerans disease in humans . Our results demonstrate that the pig can be infected with M . ulcerans . Productive infection leads to the development of lesions that closely resemble human BU lesions . The pig infection model therefore has great potential for studying the early pathogenesis of BU and for the development of new therapeutic and prophylactic interventions .
Buruli ulcer ( BU ) , caused by infection with Mycobacterium ulcerans , is a human disease of the skin primarily affecting subcutaneous fat tissue and leading to ulceration of the overlying dermal and epidermal layers [1] , [2] . The disease is reported from countries worldwide but has its highest prevalence in West Africa [3] . Natural reservoirs of M . ulcerans as well as the mode ( s ) of transmission are not clearly identified [3] , [4] . While for a long time wide surgical excision was the only treatment option for BU , since 2004 the World Health Organization ( WHO ) recommends antibiotic therapy with rifampicin and streptomycin for 8 weeks [5] . This change in standard treatment has reduced recurrence rates to less than 2% [6]–[9] . M . ulcerans produces the polyketide exotoxin mycolactone that is responsible for the necrotizing nature of BU [10] . Three distinct non-ulcerative stages of the disease are described: subcutaneous , painless and movable nodules or papules , oedema and plaques . All three stages may progress to ulceration once the destruction of the subcutis results in collapse of the overlying epidermis and dermis [11] . Ulcerative BU lesions have been histopathologically well described through the analysis of excised tissue from surgically treated patients . Coagulative necrosis , fat cell ghosts and epidermal hyperplasia together with the presence of extracellular clusters of acid fast bacilli ( AFB ) in the absence of major inflammatory infiltrates in central parts of the lesions are considered hallmarks of the disease and can also be used for histopathological diagnosis [12] , [13] . However , early , pre-ulcerative stages have been described less frequently , because in particular in the African BU endemic regions patients are rarely reporting at treatment centres during early stages of the disease . Furthermore , with the replacement of surgical treatment by chemotherapy , tissue samples are not easily available any longer . Therefore , a suitable experimental animal infection model is required to contribute to the understanding of early host-pathogen interactions and pathogenesis in BU . A range of animal species have been reported of being naturally infected with M . ulcerans and of developing ulcerative lesions . These include koalas , possums , cats , dogs and horses [14]–[21] . Except for possums which appear to be unusually susceptible to the disease , these animal infections seem to occur only sporadically [22] . Experimental M . ulcerans infections have been performed with amphibians , armadillos , rats , mice , guinea pigs and monkeys , with a mouse foot pad model being most widely used for studying the efficacy of prophylactic and therapeutic interventions [23]–[29] . Here we propose the pig ( Sus scrofa ) as experimental M . ulcerans infection model , since pigs are closely related to humans in terms of many aspects of anatomy and physiology [30] , [31] . The pig is widely used as a model in dermatological studies because pig skin , in contrast to rodent skin , has striking similarities to human skin [32] . Not only the thickness of the epidermis and the dermis are comparable to human skin [33] , but also the presence of a subcutaneous fat cell layer is favouring the pig model over the mouse foot pad model commonly used for analysing BU pathogenesis . Furthermore , the porcine immune system reflects the human immune system in many aspects better than the murine immune system does [34] , [35] . For all these reasons we explored here the potential of the pig to serve as model for human M . ulcerans infection .
All animal experiments described here were approved by the Animal Welfare Committee of the Canton of Berne under licence number BE50/11 , and conducted in compliance with the Swiss animal protection law and with other national and international guidelines . The M . ulcerans strain used in this study was isolated in 2010 from a swab taken from the undermined edges of the ulcerative lesion of a Cameroonian BU patient [4] . Five passages of the strain after isolation were done in Bac/T medium ( Biomerieux ) at 30°C . For preparation of the inoculum , bacteria were cultivated in Bac/T medium for 6 weeks , recovered by centrifugation and diluted in sterile phosphate-buffered saline ( PBS ) to 375 mg/ml wet weight corresponding to 2×108 CFU/ml as determined by plating serial dilutions on 7H9 agar plates . From this stock solution suspension serial dilutions in PBS were prepared for infection . Specific pathogen-free 2-month-old pigs ( Large White ) from the in-house breeding unit of the Institute of Virology and Immunology ( IVI ) were kept under BSL3 conditions one week prior and during the time of experimental infection . Animals were checked once daily for macroscopic signs of infection , had ad libitum access to water and were fed daily with complete pelleted food . Pigs were infected on both flanks at four to six infection sites with 100 µl of M . ulcerans suspension , containing 2×107 , 2×106 , 2×105 , 2×104 or 2×103 CFU . Injection areas were wiped with 70% ethanol and bacterial suspensions injected subcutaneously with a 26G needle . Individual infection sites were encircled with a black marker and the labelling renewed at least once a week . Animals were euthanized at 2 . 5 weeks or 6 . 5 weeks post-infection and tissue samples taken as described below . In addition , the effect of mycolactone was studied directly by injecting 5 µg or 0 . 5 µg of synthetic mycolactone A/B [36] and analysing tissue specimens taken 2 . 5 weeks later . Pigs were euthanized by intravenous injection of pentobarbital ( 150 mg/kg bodyweight ) and subsequent exsanguination . Skin tissue at infection sites was extensively excised with a scalpel and scissors , including all layers of the skin down to the fascia , and samples were immediately transferred to 10% neutral-buffered Formalin solution ( approx . 4% formaldehyde ) . After fixation samples were transferred to 70% ethanol for storage and transport , dehydrated and embedded into paraffin . 5 µm thin sections were cut , deparaffinised , rehydrated and directly stained with Haematoxylin/Eosin ( HE ) or Ziehl-Neelsen/Methylene blue ( ZN ) according to WHO standard protocols [11] . Stained sections were mounted with Eukitt mounting medium ( Fluka ) . Pictures were taken with a Leica DM2500B microscope or with an Aperio scanner .
In order to assess early effects of the subcutaneous experimental infection of pigs with doses of 2×103 to 2×107 M . ulcerans CFU , injection sites were closely monitored for macroscopic changes of the skin . At 2 . 5 weeks after injection of the bacteria , first changes in colouration and thickness of the skin became apparent at the sites inoculated with the highest inoculation doses , 2×107 and 2×106 CFU ( Fig . 1 , B1 ) . Like nodular BU lesions in humans , these early lesions were elevated , movable , firm and palpable . When these skin areas were excised 2 . 5 weeks and 6 . 5 weeks after infection and vertically cut in half after fixation in formalin , roundish yellow structures reflecting coagulative necrosis in the dermis became macroscopically apparent ( Fig . 1 , B2 ) . A belt with reddish colour , reflecting infiltrating cells and bleeding into the skin , was observed around the necrotic core . While these structures were larger at sites inoculated with a dose of 2×107 CFU than at sites inoculated with 2×106 CFU , the general architecture observed with both inoculation doses was similar . At sites inoculated with <2×105 CFU , no macroscopically visible alterations of the skin were found 2 . 5 weeks after infection ( Fig . 1 , A1 and A2 ) . At 6 . 5 weeks after experimental infection , sites injected with the highest inoculation dose had either enlarged to an indurated plaque ( Fig . 1 E1 ) or ulcerated ( Fig . 1 , D1 ) . At sites injected with 2×106 CFU , nodular lesions were observed that were flatter and less palpable compared to those detected 2 . 5 weeks after infection ( Fig . 1 , C1 ) . These lesions were macroscopically clearly visible in cross sections through the tissue ( Fig . 1 , C2 ) . Nodular and ulcerative lesions exhibited greyish/reddish colour changes in the dermis and subcutis ( Fig . 1 , C2 and D2 ) . The plaque lesion developing after injection with 2×107 CFU appeared as long cord-like structure with a centre made of yellowish necrotic slough , surrounded by several layers differing in colouration ( Fig . 1 , E2 ) . Microcopically , infiltrating immune cells were found 2 . 5 weeks after infection at all sites inoculated with ≥2×104 CFU ( Fig . 2 , A1 , B1 , C1 and D1 ) . As expected from the macroscopically observed signs , the most pronounced histopathological alterations were associated with the two highest inoculation doses ( 2×107 and 2×106 CFU ) . The non-ulcerative lesions that developed between the dermis and the underlying muscle tissue displaced the fat layer ( Fig . 2 , A1 and B1 ) and caused the macroscopically visible elevation of the skin ( Fig . 1 , B1 ) . Microscopically , a necrotic core surrounded by large numbers of infiltrating cells and interspersed with fat cell ghosts was observed ( Fig . 2 , A2 and B2 ) . At sites infected with 2×105 CFU , no necrotic core structures but some fat cell ghosts and accumulations of infiltrating cells were found ( Fig . 2 , C1 and C2 ) . The infection with 2×104 CFU caused a small accumulation of infiltrating cells ( Fig . 2 , D1 and D2 ) and no signs of infection and/or inflammation were observed at sites inoculated with the lowest dose ( 2×103 CFU ) . At 6 . 5 weeks after experimental infection , histopathological changes were only found at sites that had been injected with 2×107 or 2×106 CFU . In contrast , the skin appeared macro- and microscopically healthy following infection with lower doses of M . ulcerans , exhibiting intact epidermis and fat cells , undistorted collagen fibre networks and no marked inflammatory infiltration ( Fig . 3 , D1 and D2 ) . Where the infection focus had started to ulcerate , strong infiltration towards the destroyed epidermis was observed ( Fig . 3 , A1 and A2 ) . No AFB were found in this region , indicative for loss of the necrotic core with the major burden of AFB through the ulceration ( Fig . 3 , A2 , Fig . 4 , B1 and B2 ) . Small clusters of AFB were found at deeper sites in the tissue , lateral to the ulceration site ( Fig . 4 , B3–B5 ) . Infiltration and destruction of collagen fibres extended into the lower part of the dermis and the upper part of the subcutis , reaching far beyond the area where the epidermis was destroyed ( Fig . 3 , A1 ) , indicating the formation of undermined edges ( Fig . 3 , A2 , dotted line ) . The overall architecture of the plaque lesion that had developed resembled the nodular stages seen 2 . 5 weeks after infection , i . e . a necrotic centre was surrounded by layers of infiltrating cells ( Fig . 3 , B1 ) . While large clumps of extracellular AFB were found in the necrotic core after injection of 2×107 CFU ( Fig . 3 , B2 ) , AFB were less abundant and bacterial clumps smaller when 2×106 CFU were used for infection ( Fig . 3 , C2 ) . Fig . 4A depicts the complex architecture of a plaque lesion ( 2×107 CFU dose ) with several distinct belts of infiltrating cells surrounding a central necrotic core which contained huge clusters of AFB but was completely devoid of infiltration ( Fig . 4A , Ring 1 , A1 and A2 ) . In the surrounding ring 2 , AFB were scarce and had mostly a beaded appearance . In addition to these single AFB , small globi-like clusters of AFB were found , along with Methylene blue stained remains of infiltrating cells ( Fig . 4A , Ring 2 , A3 and A4 ) . Ring 3 contained mostly small infiltrating cells that appeared intact , and some acid-fast bacterial debris ( Fig . 4A , Ring 3 , A5 and A6 ) . The outermost layer that could be distinguished did not contain AFB and was mainly built by macrophages and lymphocytes ( Fig . 4A , Ring 4 , A7 ) . Hence , the number and integrity of AFB decreased from the centre to the periphery of the lesion , whereas the integrity of the cellular infiltration showed an opposite trend , most likely reflecting levels of the cytotoxic macrolide mycolactone decreasing from centre to periphery . All key features of BU pathology in humans were also found in the experimentally infected pig skin . Already 2 . 5 weeks after infection , coagulative necrosis ( Fig . 5 , A1 ) , fat cell ghosts ( Fig . 5 , A2 ) and extracellular clusters of AFB ( Fig . 5 , A3 and A4 ) were detected . Slight epidermal hyperplasia was already observed at 2 . 5 weeks and became more pronounced 6 . 5 weeks after infection ( Fig . 5 , A5–A7 ) . At this time , typical histopathological hallmarks of more advanced human BU lesions also emerged in the infected pig skin , namely formation of granulomas ( Fig . 5 , A8 ) and presence of giant cells ( Fig . 5 , A9 ) . Not only experimental infection with M . ulcerans led to these typical alterations in the skin , but also the injection of synthetic mycolactone A/B ( Fig . 5B ) . Besides the general histopathological changes , another similarity to findings in human BU [37] was observed: the formation of satellite infection foci adjacent to the primary lesion . A striking example for this is depicted in Fig . 4B where two satellite foci with small clusters of AFB in a necrotic core were found peripheral to the ulcerated main infection focus ( Fig . 4B , Region 2 ) . Likewise in the plaque lesion depicted in Fig . 4A , clusters of AFB were found near the main infection focus ( Fig . 4A , Ring 5 , A8 ) .
Detailed studies on the early pathogenesis of BU in an animal model closely mimicking human BU would be very important for a better understanding of host-pathogen interactions and the relative importance of different effector functions of the innate and adaptive immune system against M . ulcerans . Here we explored the potential of the pig to serve as model for human M . ulcerans infection . After having infected pigs subcutaneously with high doses ( 2×106 or 2×107 CFU ) of M . ulcerans bacteria , we observed the development of different forms of BU lesions ( nodules , plaques and ulcers ) . Macroscopic and histopathological changes closely mirrored human BU . Challenge with lower doses ( 2×103 to 2×105 CFU ) resulted in limited tissue destruction and/or infiltration 2 . 5 weeks after infection , which resolved spontaneously until week 6 . 5 . Likewise , the dose of bacteria transmitted may be of critical importance for the outcome of a natural M . ulcerans infection in humans . Sero-epidemiological analyses in human populations living in BU endemic areas have indicated that exposure to M . ulcerans often leads to self-resolving , non-symptomatic infections , as indicated by development of M . ulcerans specific antibody responses [38] , [39] . While macrophages and other immune cells might be able to eliminate smaller numbers of scattered M . ulcerans cells , microcolonies of a critical size may develop a protective cloud of mycolactone around them . If the local concentration of the macrolide cytotoxin exceeds a certain level , infiltrating cells may be killed before they can reach the bacteria . This leads to the characteristic picture of clusters of extracellular AFB located primarily in the necrotic core of advanced lesions , which is devoid of living infiltrating immune cells , but contains debris of early inflammatory infiltrates [40] , [41] . In our study , we observed round elevations of the skin already 2 . 5 weeks after infection . These alterations were firm , movable and clearly palpable and hence displayed the characteristic features of human BU nodules [11] . Microscopic investigation of the infected skin sites revealed that most histopathological hallmarks of BU had already developed during the first 2 . 5 weeks of infection if 2×106 or 2×107 CFU of M . ulcerans was used . The experimentally induced nodules exhibited a necrotic core containing extracellular AFB surrounded by infiltrating cells and fat cell ghosts . Subcutaneous injection of the bacteria led to the formation of an infection focus in the lower dermis and subcutis , where it is also typically found in human BU [13] . In ulcerative human BU lesions AFB are typically focally distributed and not evenly dispersed in the affected tissue [37] , [42] . Ulceration leads to the shedding of necrotic tissue containing masses of AFB . Therefore the bacterial burden is usually higher in non-ulcerative lesions than in ulcers , where the majority of the remaining AFB reside in the undermined edges of the ulcers . Our histopathological analyses showed , that like in human BU disease [37] , satellite lesions may develop near the primary lesion . These may emerge from globi-like accumulations of AFB originating from bacteria that were internalized and transported to distant sites by phagocytic cells . Globi-like accumulations are also found in human BU [43] and in experimentally infected mice [44]–[46] . Again , these microcolonies may have to reach a critical size to be able to develop a protective cloud of mycolactone around them . The emergence of only small numbers of newly established microcolonies may explain why borders of advanced ulcers often appear to be very heterogeneous with respect to disease activity with some regions displaying progressive tissue destruction and others showing spontaneous healing tendencies . At 6 . 5 weeks after infection with 2×106 or 2×107 CFU of M . ulcerans , lesions that were still closed comprised a necrotic centre containing clumps of AFB surrounded by well stratified belts of infiltrating cells . Similarly , lesions consisting of a necrotic core surrounded by an inner belt of CD14 positive monocytes/macrophages and a more external belt of CD3 positive T-cells have been described in human BU [47] . The integrity and number of bacteria was decreasing to the outer rim of the lesion . In contrast , the density and integrity of the cellular infiltrates decreased towards the necrotic core . In the pig model first macroscopic signs of infection ( nodules ) developed relatively fast after injection of a high number of bacteria . For human BU disease in Uganda and Southern Australia incubation periods of 4–13 weeks and 5–38 weeks have been estimated , respectively [48] . However , incubation periods as short as 2–3 weeks have also been described [49] . Despite extensive analyses we did not find bacteria in the tissue with low inoculation doses at the 6 . 5 week time point . Therefore we assume that also at later time points lesions would not develop with these low infection doses . It is possible that pigs are more resistant to M . ulcerans infection than humans . Consequently , the size of the inoculum to achieve productive experimental infection may be higher for pigs than for the natural infection of humans . This high experimental inoculation dose may have led to fast progression of the disease . In conclusion , our findings indicate that the pig is a very good animal model to study many aspects of M . ulcerans infection . Pig skin represents a much closer model for human skin than murine foot pads , ears or tails with respect to physiology , structure and abundance of fat tissue [50] . In addition , the immune system of the pig resembles the human system more closely than that of the mouse [34] . In particular the development of new therapeutic and prophylactic interventions might benefit from the porcine M . ulcerans infection model .
|
Buruli ulcer caused by Mycobacterium ulcerans infection is a necrotizing disease of the skin and the underlying subcutaneous tissue . Since the skin of pigs ( Sus scrofa ) has striking structural and physiological similarities with human skin , we investigated whether it is possible to develop an experimental M . ulcerans infection model by subcutaneous injection of the mycobacteria into pig skin . Injection of 2×106 or 2×107 colony forming units of M . ulcerans led to the development of lesions that were both macroscopically and microscopically very similar to human Buruli ulcer lesions . In particular for the characterization of the pathogenesis of Buruli ulcer and of immune defence mechanisms against M . ulcerans , the pig model appears to be superior to the mouse foot pad model commonly used for the evaluation of the efficacy of chemotherapeutic regimens .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] |
[
"mycobacterium",
"ulcerans",
"bacterial",
"diseases",
"infectious",
"diseases",
"buruli",
"ulcer",
"medicine",
"and",
"health",
"sciences",
"anatomical",
"pathology",
"histopathology",
"animal",
"models",
"of",
"infection",
"pathology",
"and",
"laboratory",
"medicine",
"animal",
"studies",
"animal",
"models",
"of",
"disease",
"neglected",
"tropical",
"diseases",
"biology",
"and",
"life",
"sciences",
"tropical",
"diseases",
"actinobacteria",
"bacteria",
"organisms",
"research",
"and",
"analysis",
"methods"
] |
2014
|
Experimental Infection of the Pig with Mycobacterium ulcerans: A Novel Model for Studying the Pathogenesis of Buruli Ulcer Disease
|
Viruses use cellular machinery to enter and infect cells . In this study we address the cell entry mechanisms of nonenveloped adenoviruses ( Ads ) . We show that protein VI , an internal capsid protein , is rapidly exposed after cell surface attachment and internalization and remains partially associated with the capsid during intracellular transport . We found that a PPxY motif within protein VI recruits Nedd4 E3 ubiquitin ligases to bind and ubiquitylate protein VI . We further show that this PPxY motif is involved in rapid , microtubule-dependent intracellular movement of protein VI . Ads with a mutated PPxY motif can efficiently escape endosomes but are defective in microtubule-dependent trafficking toward the nucleus . Likewise , depletion of Nedd4 ligases attenuates nuclear accumulation of incoming Ad particles and infection . Our data provide the first evidence that virus-encoded PPxY motifs are required during virus entry , which may be of significance for several other pathogens .
Many viruses use the microtubule network of the host cell for transport to their site of replication ( i . e . the nucleus ) [1] . Access to the microtubule network is achieved through recruitment of cytoplasmic dynein motor proteins followed by efficient retrograde transport towards the nucleus [2] , [3] . Virus-induced cellular signaling cascades help stimulate the directionality and efficacy of the transport [4] . Viral interaction with dynein motor proteins occurs either directly through capsid proteins or indirectly via hijacking of adapters from existing transport pathways [5] . Most DNA viruses accumulate transiently at the microtubule organizing center ( MTOC ) prior to nuclear translocation [1] , [3] , [6] . How they release from the microtubules or the MTOC and transport to nuclear pores is poorly understood . MTOC release may involve a switch from dynein to kinesin mediated transport , the cellular ubiquitin/proteasome system and/or nuclear transport receptors [1] , [3] , [5]–[8] . Indirect evidence that the host's ubiquitylation machinery participates in parts of the viral entry process comes from studies using pharmacological inhibitors of the ubiquitin/proteasome system . For example , translocation of a murine coronavirus from the endosome to the cytoplasm is facilitated by the ubiquitin-proteasome system [9] . Similarly , influenza viruses appear to be trapped in an endosomal compartment upon pharmacological inhibition of the proteasome [10] . In contrast , blocking the proteasome increases the transduction efficiency of adeno-associated virus vectors and this correlates with ubiquitylation of capsid proteins [11] , [12] . The Semliki forest and the vesicular stomatitis virus , however , do not seem to be affected by proteasome inhibition during their entry suggesting different host factor requirements [10] . A role for the ubiquitylation machinery during egress of enveloped viruses is better understood . Egress involves the transport of assembled capsids , subviral structures or individual capsid proteins to assembly and budding sites at the cell surface or at intracellular membranes [1] . Budding , and potentially trafficking , to the egress site requires an intact class E vesicular sorting pathway ( VSP , [13] , [14] . The VSP is believed to involve the consecutive activity of three distinct heteromeric complexes termed endosomal sorting complexes required for transport ( ESCRT-I , -II and –III , [15] ) . The capsid proteins of several enveloped viruses encode ‘late domains’ that specifically interact with ESCRT components and redirect them towards the site of viral egress [13] . Some late domains of the PPxY motif type ( where x can be any amino acid ) require the binding of ubiquitin ligases of the Nedd4 family of HECT-E3 ubiquitin ligases ( Homologous to E6-AP Carboxyl Terminus ) for efficient ESCRT recruitment [13] . Nedd4 . 1 and its close relative Nedd4 . 2 are prototypic members of this family , which is conserved from yeast to mammals [16] . They encode a N-terminal C2 domain for Ca2+-dependent lipid interaction , a catalytical HECT domain and three to four WW-domains for protein-protein interactions with proline-rich domains such as the PPxY motif [16] , [17] . The exact role of PPxY-recruitment of Nedd4 ligases in VSP-mediated viral budding is still unclear . A possible link was recently shown by enhanced ESCRT ubiquitylation through Nedd4 . 2 overexpression [18] , [19] . Late domains , including the PPxY type , have also been found in some nonenveloped reoviruses but a general function in virus release remains to be shown [20] . PPxY type late domains where also described for the Ad capsid protein penton , which can interact with ubiquitin ligases of the Nedd4 family . However , its role in Ad infection is unclear [21] . At least in vitro , Ad infects cells by first attaching to primary receptors , including CAR , CD46 and sialic acid , via the fiber protein [22] . In some cells endocytosis of Ad may be triggered by penton base-mediated signaling through alpha ( v ) integrins [23]–[25] . In epithelial cells , Ad serotype 5 ( Ad5 ) particles undergo stepwise disassembly during entry , starting with detachment of the fiber at or near the cell surface and followed by a passage through endosomal compartments in which acidification serves as additional disassembly trigger for membrane penetration and cytosolic translocation [26] , [27] . Partial disassembly releases the internal capsid protein VI , which can lyse membranes in vitro via its predicted N-terminal amphipathic helix [28] . In the cytosol , the particle engages in microtubule-directed transport towards the MTOC and is translocated to the nuclear pore complex ( NPC ) for nuclear import of the genome [29]–[31] . In this study we address the mechanisms of Ad cell entry . We demonstrate that the internal capsid protein VI is rapidly exposed to antibodies during cell entry , possibly at the cell surface or immediately after endocytosis . We further determine that protein VI remains partially associated with Ad capsids as they traffic to MTOCs and the NPC . We identify a functional PPxY motif within protein VI that mediates the association of protein VI with Nedd4 E3 ubiquitin ligases and facilitates its ubiquitylation . Recombinant Ad5 in which the protein VI PPxY motif is mutated have normal capsid morphology , escape from endosomes with similar efficiency as wildtype viruses , but are defective in genome delivery to the nucleus . We show that the PPxY motif in protein VI is involved in its efficient microtubule-mediated transport and mutating it in the virus alters the intracellular targeting of Ads towards the MTOC region concomitant with a post-entry block in viral infectivity . Furthermore , Nedd4 . 1 and Nedd4 . 2 are involved in Ad infection and intracellular targeting of incoming virions to the MTOC . We propose that the PPxY motif , in other viral systems , may also function during entry and interact with novel cellular pathways for efficient viral entry .
The fate of Ad particles immediately after internalization is only partially characterized . In the context of endosome escape of Ad5 , very little is known about how this occurs and which , if any , cellular proteins are involved . From in vitro studies it was proposed that the internal capsid protein VI mediates Ad endosome escape [28] . Previous reports showed that protein VI dissociates from the Ad capsid very early after attachment [26] , [27] . To delineate the fate of protein VI during Ad entry we performed infection assays and followed the intracellular distribution of the viral capsid and protein VI as a function of time . To this end , fluorescently labeled Ad particles were adsorbed to either human retina epithelia pigment cells ( hTERT-RPE1 , Figure 1 ) or human osteosarcoma cells ( U2OS , Figure S1 ) at 4°C and then transferred to 37°C to synchronize internalization . Cells were fixed at various times and analyzed by confocal microscopy ( Figure 1 ) . Protein VI was detected using an affinity purified polyclonal-antiserum and the location of the MTOC was marked by detecting the primary cilia , which originates at the MTOC using antibodies against acetylated tubulin [32] . At 4°C viral particles accumulated at the cell periphery showing sporadic positive staining with the protein VI antiserum ( ∼1% ) possibly due to the recognition of protein VI from damaged particles ( Figure 1 first row ) . In contrast , 5 min after the temperature shift , Ad particles were still localized close to the cell periphery but approximately 40% of them gave a signal with the protein VI antibody indicating that more protein VI was accessible ( Figure 1 , second row ) . After 15 min , particles had entered the cell with some localized at the MTOC region ( Figure 1 , third row ) and some at the nuclear rim as described previously [6] . About 10% of the particles remained protein VI positive , including particles at the nuclear rim . After 45 min the majority of the particles were concentrated at the MTOC region ( Figure 1 , bottom row ) as previously reported [3] . Protein VI staining was also concentrated at the MTOC region but most of the signal was not particle associated . Similar results were obtained in U2OS cells ( Figure S1 ) . Together these data suggested that structural rearrangements leading to protein VI exposure take place at or close to the cell surface during Ad entry . In addition , the data showed that protein VI trafficked to the MTOC region and partially remained associated with the capsid . To identify possible trafficking determinants , we analyzed the sequences of protein VI from several Ad serotypes and identified a highly conserved ubiquitin-ligase interacting motif present in PPxY-type viral late domains ( PPxY , Figure S2 ) . To examine the role of this PPxY motif in Ad cell entry , we used an E1/E3-deleted Ad5 that had the protein VI PPSY motif mutated to PGAA ( Ad5-VI-M1 , detailed in Figure S3 ) [33] , [34] . This mutation , when introduced into Mason-Pfizer monkey virus , was previously shown to abolish late domain functions with no apparent structural changes , which would impair virus assembly [35] , [36] . To control for unintended mutations introduced during the cloning , we reverted the PGAA sequence back to PPSY ( Ad5-VI-wt ) . Because the ∼360 copies of protein VI appears to be in contact with several proteins in the mature capsid , modifications that disrupt the tertiary structure could also affect the capsid composition . In large-scale preparations , mutant and wt virus banded at identical densities and gave similar yields of particles as determined by genome and protein quantification . A biochemical analysis of the capsid composition of purified viral particles showed no apparent differences between wt and mutant viruses ( Figure 2A and data not shown ) . To confirm that viral capsid integrity between mutant and wt virus remained unchanged , we used negative stain electron microscopy . As shown in Figure 2B capsid integrity and morphology of the mutant virus was indistinguishable from the wt virus . In contrast , the infectious versus physical particle ratio of Ad5-VI-M1 was ∼20-fold lower than Ad5-VI-wt as assayed by plaque formation on monolayers of 911 cells ( Figure 2C ) . Because infectious vs . physical particles can vary between preparations , we assayed plaque size , which is more informative measurement of propagation rate . Plaques were significantly smaller for Ad5-VI-M1 versus Ad5-VI-wt ( see below ) , suggesting that the altered PPxY domain affects some stages of virus propagation . To determine whether the M1 mutation influences Ad cell entry , we performed a fluorescent focus forming assay and stained cells at 8 , 12 and 24 h post-infection for expression of the E2A protein , which marks the appearance of viral replication centers ( Figure 2D and data not shown ) . Compared to Ad5-VI-wt , the Ad5-VI-M1 virus produced approximately 20-fold fewer fluorescent foci when equivalent numbers of viral particles were used for infections . This suggested that steps prior to replication ( i . e . internalization ) require an intact PPxY motif in protein VI . To address trafficking using a different approach , we inserted a GFP expression cassette into the FRT site in the E1-deleted region of the wt and the M1 mutant virus ( see Figure S3 for details ) . The GFP expressing viruses showed no difference in the quantitative yield and biochemical composition after large-scale purification . We repeated plaque forming assays ( Figure 2E ) and single round infection assays ( Figure 2F ) , this time using GFP expression as the quantification method in non-complementing U2OS cells . We observed a reduction in infectivity in the same order of magnitude as previously ( compare Figures 2C with 2E and 2D with 2F ) . In addition the GFP expression allowed us to follow the plaque formation over time . As shown in Figure S4 the spread of the M1 mutant virus was significantly slower and led to fewer and much smaller plaques ( Figure 2E and Figure S4 ) . Next we asked whether Ad5 endosomal lysis efficiency following internalization is affected by the mutation in the PPxY motif of protein VI . We infected A549 cells with 30 , 300 or 3000 particles per cell of either Ad5-VI-wt or Ad5-VI-M1 in the presence of alpha-sarcin , a membrane impermeable toxin that inhibits translation when it enters the cytoplasm . Alpha-sarcin enters the cell by virus-mediated endosomal membrane lysis thus providing a quantifiable marker for endsome membrane lysis ( Figure 2G , [28] ) . We measured the incorporation of radiolabeled amino acids over time as a means to determine translational efficiency . As control we used the temperature sensitive Ad mutant Ad2ts1 that is closely related to Ad5 . Ad2ts1 mutants poorly lyse the endosomal membrane when the virus is grown at non-permissive temperatures due to an increased capsid stability that lacks intra-endosomal disassembly . Ad2ts1 mutants grown at permissive temperatures lyse the endosome with the same efficiency as the wt virus . Incorporation of radiolabeled amino acids was compared to alpha-sarcin treated cells . When we added alpha-Sarcin and Ad5-VI-wt , Ad5-VI-M1 or the Ad2ts1 mutant grown at a permissive temperature translation was diminished in a dose-dependent manner over at least two orders of magnitude , showing efficient cytoplasmic delivery of the toxin . Ad2ts1 grown at the non-permissive temperature did not inhibit translation in this assay ( Figure 2G ) . We observed no difference between the PPxY-mutant virus and the wt controls . This indicated that an attenuating effect of the PPxY-mutation occurs after endosomal lysis and prior to the onset of replication . We next characterized the release of protein VI from fluorescently labeled Ad-VI-M1 and Ad5-VI-wt following synchronous infections of U2OS cells using the infection assay as described for Figure 1 . Ad5-VI-wt released protein VI within 5 min ( Figure 3A , left panel ) . In contrast , we observed a delayed release of protein VI from Ad5-VI-M1 ( 15% of M1 after 5 min compared to 38% of wt , Figure 3A ) and an increase in colocalization of viral particles with protein VI at the nucleus ( Figure 3A ) . This observation suggests a delayed accessibility of protein VI within the virus or a defect in protein VI dissociation from the virion ( Figure 3A , images to the right ) . In addition to the increase of protein VI capsid association , Ad5-VI-M1 appeared to be more evenly distributed throughout the cell and did not efficiently accumulate at the MTOC region ( Figure 3B , images to the left ) . Quantification revealed that 45 min post-infection approximately 60% of the wt viral particles in proximity of the MTOC could be found within a 10 µm radius around the MTOC and 40% within 10–20 µm . In contrast , the localization for the M1 virus was 50% for each region showing a decreased targeting towards the MTOC ( Figure 3B , right panel ) . In summary these data suggest that the PPxY motif in protein VI is required for proper uncoating and normal nuclear targeting . To understand the accumulation defect of Ad5-VI-M1 at the MTOC , we expressed wt ( VI-wt ) , mutant ( VI-M1 ) and protein VI deleted in the amphipathic helix ( VI-ΔΦ ) fused to mRFP in cells and analyzed protein VI localization in relation to microtubules ( Figure 4 ) . We found VI-wt in a punctuated distribution throughout the cell suggesting association with a vesicular compartment or tubulo-vesicular structures associated with microtubules ( Figure 4 , top row ) . In contrast , the PPxY motif mutant VI-M1 localized to a more central compartment and was rarely associated with the microtubules ( Figure 4 , middle row ) . Deletion of the amphipathic helix , in contrast , abrogated membrane association , causing nuclear targeting of protein VI and redistribution of protein VI from membrane fractions into soluble fractions ( [37] , and Figure 4 bottom row , data not shown ) . Owing to its association with membranes , microtubules and the viral capsid , we next asked whether protein VI displayed intracellular dynamics that could explain virus trafficking . We therefore performed live-cell imaging ( LCI ) using cells expressing mRFP-VI-wt or mRFP-VI-M1 . We found that VI-wt was fast moving with short- and long-range movements whereas VI-M1 was essentially motionless ( Figure 5A and Video S1 ) . The length of the trajectories and the movement of >300 particles were plotted ( Figure 5B ) . We found that protein VI-M1 motility was greatly reduced compared to protein VI-wt . We next asked whether VI-wt motility depends on intact microtubules and/or actin filaments . Disrupting actin filaments with cytochalasin B had no apparent effect on VI-wt localization or motility ( Figure 5B , right panel ) suggesting that actin was not involved in the movement . In contrast , protein VI motility in nocodazole-treated cells was strongly reduced resembling the reduced motion observed for the M1 mutant ( Figure 5B , right panel , see also Video S2 in the supporting information ) . We asked whether VI-M1 showed only plus-end microtubule directed movement . We applied nocodazole to VI-M1 transfected cells , followed by washout of nocodazole . During treatment and after removal of nocodazole no movement or relocalization towards the cell center was observed for VI-M1 . In contrast VI-wt rapidly stopped and restarted bidirectional movement under these conditions ( data not shown ) . Together these data suggested that protein VI is a highly mobile protein that moves along microtubules , presumably in association with vesicular structures whose motion depends on the PPxY motif . PPxY domains are the physiological targets of ubiquitin ligases of the Nedd4 family [16] . Therefore we asked whether protein VI ubiquitylation depends on the PPxY motif . We adapted an in vitro Ad disassembly assay mimicking the partial capsid disassembly believed to occur during Ad entry by exposing virions to 48°C . This assay dissociates the vertices including fiber , penton , protein VI and peripentonal hexons , but leaves the remainder of the capsid intact ( Figure S5; [28] ) . Heat and mock-treated samples were subjected to in vitro ubiquitylation reactions , using free ubiquitin , recombinant E1 and E2 enzymes and purified cytosol as source for the E3 ubiquitin ligase ( s ) , and analyzed by western blot ( Figure 6 , [28] ) . Western blot analysis showed that partial capsid disassembly resulted in the appearance of protein VI reactive signals with discrete size increments suggesting predominant modification with two to three ubiquitin as well as some higher molecular weight bands ( lane 2 , Figure 6A ) . In contrast , the lack of capsid disassembly ( lane 1 ) or cytosol ( lane 3 ) showed no additional protein VI reactive bands . We also tested ubiquitylation of the capsid proteins fiber , protein IIIA and penton base as internal control . We only detected ubiquitylation of the penton base ( which also harbors two PPxY domains at its N-terminus ) , while the fiber and protein IIIA ( which lack PPxY motifs ) were not modified ( Figure S5 ) . The ubiquitylation of protein VI was also confirmed by using GST-ubiquitin in the above assay , followed by GST-pulldown to show covalent modification of protein VI with ubiquitin confirming the predominant modification with two to three ubiquitin-moieties ( data not shown ) . To address the role of the PPxY motif in protein VI ubiquitylation , we repeated the in vitro ubiquitylation assay using wt or M1 mutant protein VI purified from E . coli followed by western blot analysis . We detected protein VI-reactive bands , consistent with protein VI modified with two to three ubiquitin ( Figure 6B , lane 2 ) . In contrast , no modification was observed when the PPxY motif was mutated ( Figure 6B , lane 1 ) or in the absence of ATP ( Figure 6B , lane 3 and 4 ) . Using viral particles in in vitro ubiquitylation reactions ( following partial capsid disassembly ) protein VI of Ad5-VI-M1 was not ubiquitylated ( Figure 6C , lane 3 and 4 ) , while protein VI from Ad5-VI-wt was ( Figure 6C , lane 1 and 2 ) . Thus , the PPxY motif in protein VI is inaccessible in intact capsids but can recruit ubiquitin ligase activity from cytosol when protein VI is released from the capsid interior . Together our results show that protein VI ubiquitylation depends on i ) virus disassembly , ii ) an intact PPxY domain and iii ) the presence of a cytosolic ubiquitylation activity . To identify the ligase responsible for protein VI ubiquitylation , we focused on the Nedd4-family members Nedd4 . 1 , Nedd4 . 2 , AIP4/Itch , WWP1 and WWP2 because they can interact with viral late domains that harbor PPxY motifs [38] . We first co-expressed the VI-wt or VI-M1 mRFP fusion protein together with each of the E3 ligases fused to GFP in U2OS cells . When expressed alone , most ligases localized primarily to the cytoplasm ( data not shown , WWP1 localized to the plasma membrane and WWP2 accumulated in an uncharacterized intracellular membrane compartment ) . In contrast , when VI-wt is coexpressed with Nedd4 . 1 , Nedd4 . 2 or AIP4/Itch , the ligases are recruited to the same membrane compartment as protein VI ( Figure 7A , row 1–3 ) . WWP1 appears to sequester protein VI at the plasma membrane ( Figure 7A , row 4 ) . WWP2 does not colocalize with VI-wt ( Figure 7A , row 5 ) . We did not detect significant colocalization between VI-M1 and the E3 ligases , consistent with a PPxY-dependent interaction ( Figure S6 ) . To determine whether any of the ligases specifically interact with protein VI , we used purified cytosol from cells overexpressing GFP-tagged ligases and performed pull-downs with beads coated with recombinant protein VI-wt or VI-M1 . Two ligases , Nedd4 . 1 and Nedd4 . 2 , were highly enriched on VI-wt beads while none of the other ligases showed strong binding to VI-wt- or to VI-M1-beads ( Figure 7B ) . Taken together , these data suggest a preferential interaction between Nedd4 . 1 and Nedd4 . 2 and the PPxY motif in protein VI , which leads to relocalization of the ligases from the cytoplasm to a membrane compartment . To further characterize the interaction between protein VI and the ligases , we knocked down Nedd4 . 1 , Nedd4 . 2 , AIP4/Itch , WWP1 and WWP2 using siRNAs ( Figure 8A ) . The cells were then incubated with an Ad5 vector harboring a GFP expression cassette ( AdGFP ) at a low multiplicity of infection ( 30 physical particles per cell ) for 3 h to achieve approximately 20% transduced cells and limit the time of virus exposure . The following day the percentage of GFP-positive cells was quantified by flow cytometry . Most ligase knockdowns had no significant effects on transduction , but Nedd4 . 2 knockdown diminished transduction by 50% ( Figure 8A ) . Because Nedd4 . 2 showed the strongest effect on Ad transduction we determined whether bacterially expressed and purified Nedd4 . 2 could ubiquitylate purified protein VI in vitro . A minimal system where cytosol was replaced by recombinant Nedd4 . 2 was sufficient for protein VI ubiquitylation ( Figure 8B , lane 1 ) . The ubiquitylation pattern was similar to that obtained with cytosol ( Figure 6 ) and required an intact PPxY and a catalytically active Nedd4 . 2 ( Figure 8B , lane 2 and 3 ) . Nedd4 . 1 and Nedd4 . 2 both showed strong interaction with protein VI in pulldown assays . To further investigate the role of each ligase we transduced cells with lentiviral vector expressing shRNAs against either Nedd4 . 1 or Nedd4 . 2 or luciferase as a control . We transduced cells in a dose-dependent manner to achieve different levels of knockdown . Transduction efficiency was monitored using a GFP expressing lentiviral vector in control cells . Seven days post-transduction shRNA treated cells were infected with AdGFP virus as described above and the transduction rate was determined by flow cytometry . We observed a dose-dependent decrease in infectivity for two different shRNAs against Nedd4 . 2 , which was similar to what we observed when we used siRNAs ( Figure 8C ) . The results for shRNAs against Nedd4 . 1 were less clear . One shRNA also reduced viral infection at very high transduction rates but to a lesser extent than shRNAs against Nedd4 . 2 while a second shRNA showed no effect on Ad transduction ( Figure 8C ) . A combined treatment of cells with either siRNAs or shRNAs against Nedd4 . 1 and Nedd4 . 2 did not further decrease Ad-transduction indicating that the effects of Nedd4-ligase knockdowns on Ad transduction may be complex ( data not shown ) . A hallmark of Ad5 infection is its transient accumulation at the MTOC during entry [5] . The mutation of the PPxY in the Ad5-VI-M1 virus seemed to alter this localization ( Figure 3 ) . Similarly , the PPxY motif was required for Nedd4 . 2 dependent ubiquitylation of protein VI . Because knockdown of Nedd4 ligases also diminished transduction with AdGFP vectors we examined accumulation at the MTOC region in cells depleted with control shRNA and Nedd4 . 1 and Nedd4 . 2 specific shRNAs . We used the same strategy as in Figure 3 by quantifying viral particles in proximity to the MTOC , which was identified by stain for pericentrin . MTOC accumulation for Nedd4 . 1 and Nedd4 . 2 shRNA treated cells was reduced when compared to cells treated with control shRNA indicating that both ligases might be involved in proper targeting of viruses towards the MTOC ( Figure 8D ) . In summary , these data provide evidence that release of protein VI during entry and a possible interaction between the PPxY motif of protein VI and Nedd4-family ligases are determinants of Ad5 trafficking during infection .
In this study we show that the Ads internal capsid protein VI harbors a PPxY-motif that is involved in virus entry and infectivity . For Ads , reaching the nucleus requires a series of sequential steps: receptor-mediated endocytic uptake , partial capsid disassembly , endosomal rupture , microtubule based transport to the MTOC and nuclear trafficking . The link between these steps has been best exemplified in the case of the thermostable temperature-sensitive mutant Ad2ts1 . This mutant enters cells by receptor-mediated endocytosis , but remains in an endosomal compartment due to increased capsid stability . Therefore , Ad2ts1 particles are directed to lysosomes for destruction and/or recycled back to the surface thus precluding accumulation at the nuclear pore complex [26] , [28] , [39] . The role of facilitating endosomal escape during Ad entry was initially assigned to the penton base [40] . Later , Wiethoff and co-workers showed that most membrane lytic activity of Ad viral capsids comes from the predicted N-terminal amphipathic helix of the internal capsid protein VI , and that membrane lytic activity required partial capsid disassembly to release protein VI [28] . Here we present several lines of evidence that protein VI plays an additional and previously unidentified role in nuclear targeting of the Ad capsid . We show that protein VI exposure from Ad5 capsids occurs within minutes when pre-adsorbed Ad5 is shifted from 4°C to 37°C . This is consistent with the loss of the fiber prior to internalization and the rapid accumulation of Ad5 in the cytosol [41] , [42] . We found that significant amounts of protein VI remains partially associated with the viral capsid after the initial exposure and until the viral particle accumulates at the MTOC or the nuclear rim . We show that protein VI is engaged in rapid intracellular trafficking that depends on intact microtubules and requires the N-terminal amphipathic helix for microtubule association and the PPxY motif for motion . To our knowledge protein VI is the first Ad capsid protein described that possesses its own microtubule-dependent dynamics and future work has to address if other capsid proteins have similar properties . Inactivation of the PPxY in the viral context ( Ad5-VI-M1 ) resulted in a post-entry delay that reduced infectivity and prevented efficient accumulation of the entering virus particles at the MTOC . While we cannot exclude the possibility that the PPxY motif in protein VI also plays a role in adenoviral replication , assembly or egress , our single round infection assays showed that the majority of the titer reduction for the mutant was related to steps prior to the initiation of replication or the delivery and expression of a reporter gene . Moreover , our data show that the efficiency with which a toxin is delivered to the cytosol during Ad infection is similar for mutant and wt virus . This provides strong evidence that the PPSY motif in protein VI has a function during cell entry , but probably only after endosomal membrane lysis has occurred . Current structural data place protein VI inside the assembled capsid , therefore potentially precluding it from functions during egress at least when capsid associated [43]–[45] . Our observations show that protein VI is exposed after entry and that capsid disassembly is required for its ubiquitylation , which is consistent with our hypothesis that the PPxY motif is accessible only during Ad entry following partial capsid disassembly . Furthermore , it is currently not clear whether late domains containing PPxY motifs present in other viral systems , which are required for viral egress , have additional functions . Mutational inactivation of PPxY motifs in the VP40 structural protein of Ebola virus and matrix protein of rabies virus both showed an attenuation of the virus and a reduction of infectivity [20] , [46] . Interestingly , for Ebola virus , the PPxY mutants showed no budding defect , but virus production was reduced , which could indicate a disruption earlier on in the life cycle than previously thought [46] . For rabies virus , Wirblich et al . describe a budding defect of the late-domain mutant but also noted a reduction of early mRNA production [20] . Earlier studies have shown that PPxY type late domains bind to ubiquitin ligases of the Nedd4 family rather then directly to class E VSP proteins [13] . Our study showed that the PPxY motif in protein VI can interact with all Nedd4 ligases tested but preferentially binds to Nedd4 . 1 and Nedd4 . 2 and re-targets them to membranes . Owing to topological constrains , recognition of the PPxY domain should require membrane rupture because of the cytosolic localization of Nedd4 ligases , which according to our data does not seem to be affected by the PPxY mutation in the M1 virus . The PPxY motif in protein VI seems to favor interaction with Nedd4 . 1 and Nedd4 . 2 although we observed interactions with AIP4/Itch and WWP1 as well . It is possible that an interaction between protein VI and WWP1 , as we observe in transient transfections , is circumvented by exposure of the PPxY domain after the virus has entered the endosomal compartment . A role for Nedd4 . 1 or Nedd4 . 2 in Ad entry is underscored by our observation that Nedd4 . 2 can directly ubiquitylate protein VI via the PPxY motif and its depletion ( and to a lesser extent also depletion of Nedd4 . 1 ) reduces Ad transduction . In addition this depletion also reduced MTOC accumulation of viral particles following infection , which is similar to the effect of the PPxY mutation in the M1 virus . However the effects were modest , indicating that additional mechanisms contribute to Ad entry . Further studies will be needed to identify a specific role for each ligase in Ad entry and trafficking towards the MTOC and to determine whether other ligases like AIP4/Itch and WWP1 are involved . How ubiquitylation of protein VI or interaction with Nedd4 ligases directs accumulation of Ads at MTOCs remains unknown . Ubiquitylated protein VI could be specifically recognized by cellular factors . Alternatively , recruitment of Nedd4 . 1 and/or Nedd4 . 2 by protein VI could result in the ubiquitylation of other cellular factors that constitute an efficient transport means used by the virus . Recent work has shown that some members of ESCRT-I become ubiquitylated when Nedd4 . 2 is overexpressed [18] , [19] . Therefore , it is possible that Nedd4 . 2 ( or other Nedd4-ligases ) binding to protein VI could activate the ESCRT pathway via ubiquitylation . Whether membrane compartments or the ESCRT pathway plays a direct role in Ad virus transport during entry remains to be addressed . However , ESCRT components can be found at the endosomal compartments as well as associated with the centromeric region [47] . It is noteworthy that endosomal escape is also required for interferon induction by Ad via yet unknown mechanisms [48] . This pathway may also be related to protein interaction between protein VI and Nedd4-family ligases . Release and separation of protein VI from the capsid is clearly defective in the Ad5-VI-M1 mutant virus . A lack of functional PPxY may therefore block disassembly steps , preclude efficient endosomal escape following the membrane lysis event or fail in the recruitment of subsequent factors required to efficiently link the virus with retrograde transport pathways . Penton base also encodes PPxY motifs [49] . Because eliminating the PPxY motif from protein VI had only modest effects on MTOC accumulation , there may be functional overlap between the PPxY in protein VI and penton base . This is further supported by the observation that ubiquitylation of penton could be abolished by depleting the cytosol with protein VI indicating that maybe the same ligases are involved ( Figure S5 ) . In conclusion , our study is the first demonstration of a PPxY motif , previously described for viral late domains , present in a non enveloped DNA virus that exerts a function during entry . Conservation of the amphipathic helix and PPxY motif in all Mastadenovirus suggests that protein VI fulfils a key function . We suggest that after endocytosis cell or serotype specific capsid disassembly cues regulate exposure , membrane-interaction and ubiquitylation of protein VI ( and possibly other capsid proteins ) . This could facilitate the recruitment of a common cellular microtubule-dependent pathway for retrograde trafficking . This mechanism could be more important in vivo in highly polarized epithelia or neurons where long-range movement is crucial and might be less important in cell culture models [50] . Given the prevalence of viral late domains , ubiquitylation and trafficking towards the MTOC our results may have uncovered a more general mechanism by which viruses and other cargos achieve intracellular delivery and provide a rational to look for further “early” functions of PPxY motif containing late-domains in other viral systems .
Immunofluorescence experiments and infection experiments were performed using U2OS cells ( human bone osteosarcoma epithelial cells ) , hTERT-RPE1 cells ( human retina epithelia pigment cells , Clontech ) or A549 cells ( human alveolar basic cells ) . Cytoplasmic extracts for pulldowns were prepared from 293T cells ( human embryonic kidney cells ) . All cells except hTERT-RPE were grown in DMEM Glutamax™ ( Gibco ) supplemented with 10% of fetal calf serum ( FCS ) ( Biowest ) . hTERT-RPE1 cells were a kind gift from M . Bonhivers ( University of Bordeaux 2 ) and grown in DMEM/HamsF12 media supplemented with 10% FCS according to the suppliers instructions . Prior to infection experiments , cells were serum starved for 24h to induce primary cilia growth [32] . Recombinant Ad5-VI-wt and Ad5-VI-M1 viruses and their GFP expressing counterparts were constructed as described in the supplemental material . Amplification of viruses was done in 293 cells and purified using double CsCl2-banding . Virus particle to cell ratios were calculated based on the estimated copy numbers of viral genomes . Copy numbers were calculated according to Mittereder et al . [51] . Briefly , purified particles were diluted 1∶10 in virus lysis buffer ( 0 . 1% SDS , 10 mM Tris/HCl pH 7 . 4 , 1 mM EDTA ) and incubated for 10 min at 56°C to release the viral genomes and the OD260 was determined . Calculations were based on 1 OD260 = 1 . 1×1012 particles/ml [51] . Lentiviral vector production for shRNA encoding vectors was done by the service platform for lentiviral vector production of the Institute Federative de Recherche 66 of the Bordeaux 2 University . Prevalidated lentiviral vectors encoding shRNAs in the vector backbone pLKO . 1 against Nedd4 . 1 and Nedd4 . 2 were purchased from the Mission™ shRNA collection from Sigma . For downregulation of Nedd4 . 1 we used NM_006154 . 1-1753s1c1 ( sh4 . 1 ( 1 ) , CCGGCCGGAGAATTATGGGTGTCAACTCGAGTTGACACCCATAATTCTCCGGTTTTT ) against the coding sequence and NM_006154 . 1-3522s1c1 ( sh4 . 1 ( 2 ) , CCGGGCCTTTCTCTTGCCTGCATATCTCGAGATATGCAGGCAAGAGAAAGGCTTTTT ) against the 3′ UTR . For downregulation of Nedd4 . 2 we used NM_015277 . x-2772s1c1 ( sh4 . 2 ( 1 ) , CCGGGCGAGTACCTATGAATGGATTCTCGAGAATCCATTCATAGGTACTCGCTTTTT ) against the coding sequence and NM_015277 . x-3959s1c1 ( sh4 . 2 ( 2 ) , CCGGCCTGTTTGTATGCGTTTGCTACTCGAGTAGCAAACGCATACAAACAGGTTTTT ) against the 3′ UTR . Control vectors for shRNAs encoded for shRNA against luciferase ( Sigma ) and control vectors for transduction and estimation of the titer encoded for GFP . All sequences for protein VI were derived from Ad serotype 5 ( Ad5 ) and cloned into the Gateway™ compatible entry vector pDONR221 . Sequence verified DONR plasmids were used for recombination into Gateway™ compatible destination vectors for N-terminal fusion of mRFP ( L30-mRFP , kindly provided by E . Bertrand ) . Bacterial expression vectors for protein VI are based on pET15b . Site-directed mutagenesis was used to change amino acids 148-PPSY-151 to 148-PGAA-151 in protein VI . N-terminal tagged expression vectors for Nedd4 . 1 , and Nedd4 . 2 were provided by E . Bertrand [52] and tagged expression vectors for AIP4/Itch and WWP1 were a kind gift of Paul Bieniasz ( Rockefeller University , New York ) . Bacterial expression vectors for catalytically active murine GST-Nedd4 . 2 and the inactive GST-Nedd4 . 2-DN was kindly provided by S . Kumar [53] . siRNAs were purchased as duplexes from EuroGentec ( only the reverse strand is shown ) : Scramble ( 5′-CGCAAUUCGAUGUCCCGUGdTdT ) , Nedd4 . 1 ( 5′-AAACAACCCAGCCAGGCUCdTdA ) , Nedd4 . 2 ( 5′-CUGUGACUUUGUGUUGUGGdTdA ) , were previously described by Segura-Morales et al . ( 2005 ) , AIP4/Itch siRNAs ( 5′-UCAUCAUUCUGAGAAGCACdTdT , [54] , and WWP1 siRNA ( 5′-CUUCUACGAUCAUCAACUCdTdT ) was previously described by Chen et al . ( 2005 ) . The WWP2 siRNA was a Smartpool from Dharmacon . Depletions were performed in 12-well dishes using 2×105 U2OS cells . Cells were transfected after 24 and 48 h with 20 pmol of each siRNA duplex per well . Forty-eight hours after the first transfection cells were transduced using 30 physical particles per cell of Ad5-GFP virus for 3 h without prior pre-adsorption . Cells were harvested 24 h later and analyzed by flow cytometry for GFP expression and further processed for western blot analysis to verify knock-down efficiency . Acquisitions were done with FACSCalibur® or FACSCantoII® cytometer ( BD Biosciences ) and the data were processed and analyzed by the CellQuest® Pro and FACSDIVA® software ( BD Biosciences ) . Purified Ad particles were labeled using the Alexa-488 microscale protein labeling kit ( Invitrogen ) using the manufacturers protocol . Infectivity of labeled virus preparations was determined by quantification of GFP-transduction . Only preparations with >90% activity where used . For time course experiments , U2OS cells were seeded at semiconfluency on coverslips . Pre-binding was done with 5000 physical particles per cell in 100 µl at 4°C on a shaking platform for 1 h . At t0 coverslips where rinsed in cold DMEM and transferred to pre-equilibrated ( 37°C , 5% CO2 ) DMEM . At indicated time points the cells where fixed and processed for IF analysis . A549 lung epithelial cells were plated in 96-well plates at a density of 10 , 000 cells/well on the day before infection . The cells were washed once with DMEM without cysteine or methionine and supplemented , 2 mM glutamine , 10% dialyzed FCS , penicillin and streptomycin ( DMEM-SA ) and infected with the respective viruses in 50 µl DMEM-SA containing 0 . 1 mg/well α-sarcin ( Sigma ) . The infected cells were incubated 30 min at 4°C to facilitate virus attachment and 90 min at 37°C to facilitate virus internalization . After this 50 µl of DMEM-SA containing 0 . 1 µCi of [35S]L-methionine ( Hartmann Analytic ) was added to each well and the cells were incubated for an additional 60 min at 37°C for labeling . The cells were then washed with 100 µl PBS and extracted in 150 µl lyses buffer containing 1% Triton-X100 , 150 mM NaCl , 10 mM MgCl2 , 20 mM Tris-HCl ( pH 7 . 5 ) supplemented with 1× Complete™ protease inhibitor cocktail ( Roche ) . The lysates were clarified by centrifugation at 20 , 000 g for 12 min . To remove the residual [35S]L-methionine , 100 µl cleared lysates were further purified with Zeba Desalt Spin Columns ( Pierce ) . The incorporation [35S]L-methionine into the extracted fraction of newly synthesized proteins was measured by liquid scintillation using TRI-CARB 1900CA counter ( Packard ) . Cells grown on coverslips were rinsed in PBS and fixed with 4% PFA in PBS and blocked/permeabilized with IF-buffer ( 10% FCS in PBS and 0 . 1% Saponin ) . Primary and secondary antibodies where applied to the coverslip in IF-buffer for 1 h each . Cells were mounted in DAKO mounting media containing DAPI and analyzed by confocal- or epifluorescence microscopy . For IF involving microtubule staining cells were treated with crosslinkers prior to fixation . The following primary antibodies were used in this study: mouse anti-AcTubulin ( kind gift from C . Janke , Montpellier ) , Mouse anti E2A ( kind gift of T . Dobner , Hamburg ) , rabbit anti pericentrin ( Abcam ) and rabbit anti-protein VI antibodies raised against recombinant protein VI and affinity purified ( see supplemental material ) . Secondary antibodies Alexa546 anti mouse was from Affinity Research and Atto647 anti rabbit was from Sigma . Confocal pictures were taken on a Zeiss LSM 510 Meta confocal microscope or a Leica SP5 confocal microscope and epifluorescence pictures were taken on a Zeiss Axiolmager Z1 microscope with CoolSnap HQ Photometrics camera both equipped with Metamorph™ software . Confocal stacks where taken every 0 . 5 µm with a pinhole setting of 1 for all channels to achieve high local resolution . Images were processed using ImageJ and Adobe Photoshop™ . Counting of viral particles was performed using the semi-automated cell counting tool from ImageJ . Colocalization analysis: Stacks from confocal images where combined as Z-projection using maximum intensity , converted into 8-bit images for each channel . Colocalization between protein VI and Ad was then determined using the colocalization finder plugin from ImageJ . Live-Cell Imaging: U2OS cells were seeded in 3 . 5cm glass-bottom dishes and transfected with 1 . 5 µg of protein VI-expressing vectors . Twenty-four h later the medium was replaced by pre-warmed OPTI-MEM ( Gibco ) . Movies were acquired on a Nikon TE 2000 microscope with Cascade 512B 2 camera using Metamorph™ software for data acquisition ( 120 frames , 1 frame/sec ) . In some cases , 2 h before the acquisition , cells were incubated with either Nocodazole ( Sigma ) or Cytochalasin B ( Sigma ) diluted respectively at 0 . 4 µg/ml and 5 µg/ml in OPTI-MEM . Drugs were not removed during acquisition . Acquired movies were further processed using ImageJ to enhance protein VI particle detection by background subtraction and bleach correction . Particles were tracked using the spot-tracking tool of Imaris™ software to determine the length of their trajectories and the speed of their movement . Particle detection size was scaled to 0 . 75 µm and tracks were built with a maximum displacement of 1 . 5 µm between consecutive frames , a maximum gap size of 3 frames and a minimal track length of 20 s . At least 5 cells were analyzed for each condition that equals a minimum of 300 analyzed tracks per condition . Three microliter of purified sample virus was adsorbed to a carbon-coated film ( 200 mesh grids ) . The grids with adsorbed virus were floated onto a solution of the negative stain ( 1% solution of uranyl acetate ) . The film was picked up by a copper EM grid and then air-dried . Specimens were examined under a HITACHI H7650 electron microscope operating at 80 kV and images were further processed using ImageJ software . Affinity purified rabbit anti-protein VI antibodies were used at a dilution of 1∶2000 . Other antibodies used for the study were: rabbit polyclonal anti-Nedd4 . 1 and anti-Nedd4 . 2 antibodies that were a kind gift of O . Staub ( Lausanne , Switzerland ) ( dilution 1∶1000 ) , rabbit polyclonal anti-WWP2 antibody ( sc-30052 , Santa Cruz Biotechnology ) ( dilution 1∶200 ) , goat polyclonal anti-WWP1 antibody ( sc-11893 , Santa Cruz Biotechnology ) ( dilution 1∶200 ) , mouse monoclonal anti-AIP4/Itch antibody ( sc-28367 , Santa Cruz Biotechnology ) ( dilution 1∶100 ) and mouse monoclonal anti-GFP antibody ( Roche ) ( dilution 1∶500 ) . SDS-PAGE was done using 12% poly-acrylamide gels and transferred to nitrocellulose membranes . Membranes were blocked in TBS containing 10% of dry-milk and 0 . 01% of Tween 20 ( Sigma ) , followed by over-night detection of antigens using primary antibodies diluted in TBS containing 10% of dry-milk and 0 . 01% of Tween 20 ( Sigma ) . Primary antibodies were detected using HRP-conjugated secondary antibodies against rabbit , goat or mouse ( Sigma ) at a dilution of 1∶5000 . Specific signals were revealed using the enhanced chemiluminescence detection system ( ECL ) ( PerkinElmer ) . Data are presented as mean , error bars as STD . Statistical analysis if not indicated otherwise was done using unpaired students t-test ( *:P<0 . 05; **:P<0 . 01; ***:P<0 . 005 ) .
|
Viruses exploit cellular functions during entry and exit of cells . To redirect cellular functions for their own purpose , viruses encode high-affinity binding sites for key-cellular factors . One such domain is the PPxY motif , which is present in structural proteins of several , mainly enveloped viruses . This motif binds to ubiquitin ligases of the Nedd4 family and recruits their function to sites of virus budding from cells . Here we show that adenoviruses also encode a PPxY motif in the internal structural protein VI and that the PPxY motif has an unprecedented function in virus entry . Adenoviruses with mutations in the protein VI PPxY motif undergo normal endosomal uptake and membrane penetration but have reduced infectivity , altered intracellular targeting and lack efficient gene-delivery . We also find that protein VI is ubiquitylated by Nedd4 ligases in a PPxY dependent manner following partial capsid disassembly and displays rapid intracellular movement . Depletion of Nedd4 ligases also alters virus movement within cells during entry and reduces viral infectivity . Given that PPxY motifs are important for virus exit our findings might have uncovered an additional function for PPxY motifs in virus entry , potentially expanding the significance of PPxY motifs and functionally related domains for viral replication .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"virology/virion",
"structure,",
"assembly,",
"and",
"egress",
"virology/host",
"invasion",
"and",
"cell",
"entry"
] |
2010
|
A Capsid-Encoded PPxY-Motif Facilitates Adenovirus Entry
|
Endogenous retroviruses are cellular genes of retroviral origin captured by their host during the course of evolution and represent around 8% of the human genome . Although most are defective and transcriptionally silenced , some are still able to generate retroviral-like particles and proteins . Among these , the HERV-K ( HML2 ) family is remarkable since its members have amplified relatively recently and many of them still have full length coding genes . Furthermore , they are induced in cancers , especially in melanoma , breast cancer and germ cell tumours , where viral particles , as well as the envelope protein ( Env ) , can be detected . Here we show that HERV-K ( HML2 ) Env per se has oncogenic properties . Its expression in a non-tumourigenic human breast epithelial cell line induces epithelial to mesenchymal transition ( EMT ) , often associated with tumour aggressiveness and metastasis . In our model , this is typified by key modifications in a set of molecular markers , changes in cell morphology and enhanced cell motility . Remarkably , microarrays performed in 293T cells reveal that HERV-K ( HML2 ) Env is a strong inducer of several transcription factors , namely ETV4 , ETV5 and EGR1 , which are downstream effectors of the MAPK ERK1/2 and are associated with cellular transformation . We demonstrate that HERV-K ( HML2 ) Env effectively activates the ERK1/2 pathway in our experimental setting and that this activation depends on the Env cytoplasmic tail . In addition , this phenomenon is very specific , being absent with every other retroviral Env tested , except for Jaagsiekte Sheep Retrovirus ( JSRV ) Env , which is already known to have transforming properties in vivo . Though HERV-K Env is not directly transforming by itself , the newly discovered properties of this protein may contribute to oncogenesis .
Retroviruses are responsible for a broad range of diseases in animals and humans , the most common of which is the development of cancers . The mechanisms by which they contribute to oncogenesis are diverse and include: ( i ) insertional mutagenesis , due to activation of cellular proto-oncogenes by inserted proviruses , ( ii ) immunosuppression , by an immunosuppressive domain conserved in most retroviral envelope proteins and ( iii ) direct oncogenic activity , with some retroviruses encoding proteins with transforming activities leading to tumour formation . For example JSRV causes the development of contagious lung tumours in sheep [1] , and the Env protein alone has been shown to be responsible for the formation of the tumours in vivo [2] . It is also able to transform cell lines [3–8] and induce lung tumour formation in mice [9] . The transforming pathways involved are many , and depend on the direct action of Env itself , as well as the Env-receptor interaction [1] . Endogenous retroviruses ( ERVs ) are the remnants of past retroviral infections , which have been captured by the host during the course of evolution . They occupy around 8% of the human genome and are similar to the proviral forms of integrated retroviruses from which they derive . Whilst most ERVs are defective and have degenerated over time , others have retained some or all of their open reading frames ( ORFs ) and can encode potentially pathogenic viral proteins [10–13] . These elements are normally suppressed in healthy tissues but expression has been reported in animal and human cancers [14–17] . The HERV-K ( HML2 ) family ( hereafter shortened to HERV-K ) is remarkable in that it has recently amplified in humans and many of its ORFs are intact , making it the largest contributor of retroviral-derived proteins in the genome [18] . Expression of the associated proteins and viral particles has been detected in cell lines as well as in human cancers , including melanoma , breast and ovarian cancers [19–22] . In addition , reports indicate that HERV-K expression is important for the transformed phenotype of several cell lines . For example , in melanoma , downregulation of HERV-K Env by siRNA decreases the tumorigenic potential of the A-375 cell line [23] and HERV-K Env expression in the TVMA-12 cell line is necessary for the transition from a adherent to a non-adherent phenotype [24] . In several breast cancer-derived cell lines , HERV-K expression was also recently shown to be important for cell motility and growth , both in vitro and in vivo [25] . In this study , we investigated whether HERV-K Env could have oncogenic properties and be involved in the transformation process of the cells where it is expressed . We demonstrate that its expression induces epithelial to mesenchymal transition ( EMT ) , leading to an increase in cell motility . We also show that HERV-K Env activates the ERK1/2 pathway and promotes the expression of transcription factors involved in oncogenesis .
As HERV-K Env expression has been reported in several human cancers , including breast , we investigated whether it could have a causal role in the transformation process . For this , we used the non-transformed breast epithelial MCF10A cell line , widely used to study the process of oncogenesis [26 , 27] . Stable long-term expression of the HERV-K Env was obtained following lentiviral transduction . A control population was generated using an empty vector . After selection , transgene expression in the populations was measured by qRT-PCR using primers located in a region common to both vectors , and were found to express similar levels of lentiviral transcripts . The HERV-K Env population alone was also found to express HERV-K Env transcripts at high but still physiological level ( S1 Fig ) , and as expected , we also detected expression of Rec transcripts that are produced from the Env construct through internal splicing [28 , 29] . Interestingly , the HERV-K Env populations displayed an altered morphology ( Fig 1A ) , changing size ( Fig 1B ) and becoming more elongated . They also lost their typical acinus organisation , and are dispersed in the plate , a phenotype reminiscent of that observed on cells treated with TGFß , a known inducer of EMT . EMT is a process during which cells change their identity , and is important both in normal development and in epithelial cancer progression . It is characterized phenotypically by a loss of cell polarity and cell-cell adhesion , and at the molecular level by a decrease of E-cadherin , an increase of N-cadherin and fibronectin , as well as the induction of a few key transcription factors ( mainly Snai1 , Snai2 , Zeb1 , Zeb2 ) [30] . We quantified these markers in the different populations . As expected , TGFß-treated cells showed an increase of the expression of the mesenchymal markers fibronectin and N-cadherin , as well as EMT-associated transcription factors Snai1&2 and to a lesser extent Zeb1 ( Fig 1C ) . Interestingly , HERV-K Env , but not the control , also modified the expression of EMT-associated genes , with a significant increase in the levels of fibronectin and N-cadherin , and a decrease in E-cadherin . However , the most induced transcription factor was Zeb2 , unlike in TGFß treated cells . We also tested the different populations for changes in their motility properties using transwell assays . As shown in Fig 1D , the HERV-K Env expressing populations displayed an increase in cell migration and invasion , similar to what is observed with TGFß-treated cells ( Fig 1E ) . Altogether the data obtained with the MCF10A cells indicate that HERV-K Env possesses oncogenic properties , and modifies the cell physiology to induce a process related to , but not identical to TGFß-induced EMT . To further characterise HERV-K Env effects on cell physiology , we used the embryonic kidney 293T cell line . We searched for genes whose expression is modified by HERV-K Env using a non-biased transcriptomic approach . 293T cells were transfected in duplicates with a vector expressing the HERV-K Env or a control plasmid , under conditions adjusted to minimise cell toxicity due to the transfection ( Fig 2A ) . Of note , the two expression vectors are identical except for nt 6759 and 6762 that introduce premature stop codons in the control vector , leading to the production of a much shorter protein ( 102 aa instead of 699 ) , in order to rule out any effect due to the Env RNA per se . The transcriptomes were compared at 24 and 48 hours post transfection using whole-genome microarrays . Genes with statistically significant changes in expression ( ≥2-fold ) were further analysed . At 24 hours post-transfection , no significant difference between the two conditions was observed , probably due to very low protein expression levels at this time point . At 48 hours , we found 86 genes with modified expression levels ( 69 down-regulated , 17 up-regulated ) . After checking , the down-regulated genes proved to be genes induced by the transfection itself . The expression level of these genes increased between 24h and 48h after transfection , but to a lower extent in the HERV-K Env cells than in the control , leading to a seemingly downregulation of these genes by HERV_K Env when only the 48h expression data are taken into account . These genes were not investigated further . Amongst the identified upregulated genes ( Fig 2B ) , there was a strong excess of transcription factors ( 7 out of 17 , with five in the top six most upregulated genes ) . Interestingly , they include EGR1 , ETV4 , ETV5 and FosB that have been associated with EMT and tumour aggressiveness in several cancers [31–35] . We confirmed the induction of the top five genes from the list in the same samples by qRT-PCR ( Fig 2B ) . We noted some variations in the measurements between microarray and qRT-PCR data , but this is likely due to the specific sequences of the primers/probes used in each technique . We also verified that the induction of transcription factors was specific and not observed with another retroviral Env protein using an expression vector for the Amphotropic Murine Leukemia Virus Env as a control in an independent series of experiments . As shown in Fig 2C , qRT-PCR confirmed the increased expression of the four transcription factors ( EGR1 , ZCCHC12 , ETV4 and ETV5 ) by HERV-K Env . We noticed that the majority of genes activated by HERV-K Env are involved in the ERK1/2 MAPK pathway ( Fig 3A ) . Indeed , EGR1 , ETV4 and ETV5 are direct targets of ERK1/2 . DUSP6 , which we found induced at a lower level , is a secondary target involved in the negative retro-control of the pathway . This strongly suggested that HERV-K Env is an inducer of the ERK1/2 pathway . We tested this hypothesis by assessing the phosphorylation of ERK1/2 following transient transfection of 293T cells ( Fig 3B ) . As shown , cells transfected with HERV-K Env show a marked phosphorylation of ERK1/2 , not seen with the controls , while total amounts of ERK1/2 are similar with all plasmids . To investigate the specificity of this activation , we transfected a panel of retroviral Envs , encompassing several genera , in 293T cells and measured the expression of EGR1 , ETV4 and ETV5 , as well as the phosphorylation of ERK1/2 . As shown in Fig 4A & 4B , no other retroviral Env was able to induce both ERK1/2 phosphorylation and transcription factor expression , except JSRV Env . Interestingly , JSRV Env’s ability to activate ERK1/2 has been described in other cellular models and has been linked to its strong oncogenic effect [36–38] . This similarity suggests that HERV-K Env possesses oncogenic properties as well . Of note , the closely related MMTV Env was unable to activate the ERK1/2 pathway , indicating that the ability to activate the ERK1/2 pathway is not a general property of betaretroviral Envs . Finally , EGR1 , but not ETV4 and ETV5 , was also induced by the deltaretroviral HTLV-1 Env , consistent with previous data reporting the transactivation of EGR1 promoter by the accessory protein Tax [39] which is also produced from the HTLV-1 Env expression vector . Accordingly , HTLV-1 Env does not induce the phosphorylation of ERK1/2 ( Fig 4A ) . Due to the similarities with JSRV Env , we tested if expression of HERV-K Env was directly transforming in a classical transformation assay using rat 208F cells . [40 , 41] HERV-K Env was unable to induce the formation of transformed foci in this assay , but an otherwise fully infectious functional endogenous allele of JSRV ( enJSRV-18 ) was also negative in this assay ( Fig 4C ) . Furthermore , unlike HERV-K Env , the endogenous JSRV Env was also unable to activate ERK1/2 or induce expression of the transcription factors ( Fig 4D+4E ) . Functional expression of all Env constructs was confirmed by assessing by their ability to produce infectious pseudotyped viral particles ( Fig 4E ) . All the experiments described above have been performed with a consensus HERV-K Env , which theoretically corresponds to the protein in the progenitor virus responsible for the insertion of all modern human-specific HERV-K proviruses [42] . In order to assess the relevance of our observations in modern-day people , we tested the effect of six previously described natural “alleles” of HERV-K Env present in humans on ERK1/2 activation ( Fig 5A ) . Five out of these six , namely K108 , K109 , K113 , K115 , K17833 show some transcription factor induction activity ( Fig 5B ) . Among these , the three that correspond to the most functional HERV-K Env proteins when tested for other classical virological properties ( Fig 5C ) also induce significant ERK1/2 phosphorylation in our assay , suggesting that the ability of the endogenous Envs to activate the signalling pathway is linked to the canonical properties of the infectious retrovirus . Two proteins are produced from the HERV-K Env-Rec plasmid: the envelope glycoprotein and the accessory protein Rec . The latter consists of two exons and is completely contained within the Env ORF . We wanted to ascertain which of these two proteins mediates the effects seen earlier . We therefore used two previously described vectors [43]: one expressing only Rec , and the other HERV-K Env without Rec ( through silent point mutations targeting the splicing sites ) ( Fig 6A ) . Expression of the Rec protein did not induce expression of the transcription factors or phosphorylation of ERK1/2 , whereas the Env alone was as efficient as the consensus Env-Rec construct to activate signal transduction ( Fig 6B & 6C ) . We then set about mapping the domains in HERV-K Env required for ERK1/2 activation . Like other retroviral Envs , it is processed by cellular proteases into two subunits , SU , expressed at the cell surface , and TM , containing a single-pass transmembrane domain ( Fig 6D ) . Using previously described mutants [43] ( see Fig 6D ) , we found that the soluble SU ( mut 4 ) completely lost the ability to induce the expression of the transcription factors and the phosphorylation of ERK1/2 ( Fig 6E & 6F ) , indicating that the TM moiety of Env is required . HERV-K Env deleted of its cytoplasmic tail ( mut 1 ) showed some activity , but was markedly decreased . The cytoplasmic tail is therefore important for the activation of the ERK1/2 pathway , either directly or through modification of the intracellular localisation of HERV-K Env , but other domains are also involved . We also assayed HERV-K Env for potential effects on other cell signalling pathways . We first tested for activation of NFκB , which is often implicated in transformation , using the degradation of IκBα as a sign of activation of the pathway . None of the Envs that we tested had any effect on IκBα levels , unlike TNFα stimulation ( Fig 7 ) . HERV-K Env therefore does not modify cell physiology through the NFκB pathway . We then tested for p38 activation ( Fig 7 ) . As previously reported , JSRV Env induces the phosphorylation of p38 [36] . The role of p38 activation in JSRV pathogenesis is not clear , but it has been shown to have an inhibitory effect on ERK1/2 signalling . We found that HERV-K Env also activates p38 , but our microarray data indicates that ERK1/2 is the major signalling pathway activated by HERV-K Env . It is possible that the activation of p38 by JSRV and HERV-K Envs is a mechanism to regulate the level of ERK1/2 activation . Finally , we tested the PI3K/AKT pathway , previously shown to be important for JSRV Env mediated transformation in several cellular models [6 , 8 , 40 , 41 , 44 , 45] . Preliminary experiments showed that AKT is constitutively phosphorylated in 293T cells , whatever the culture conditions . We therefore tested for AKT activation in HeLa cells . As shown , HERV-K Env had no effect on the phosphorylation state of AKT , unlike JSRV Env ( Fig 7 ) . Finally , using a series of inhibitors ( Fig 8A ) , we characterised where HERV-K Env acts in the ERK1/2 pathway . 293T cells were transfected as before , treated with each inhibitor individually 18h later and ERK1/2 phosphorylation and expression of the transcription factors were measured at 48h ( Fig 8B ) . First , we used FTI-277 that targets H and K-Ras . As expected , this inhibitor efficiently suppressed the phosphorylation of ERK1/2 , with a corresponding decrease in transcription factor expression induced by the transfection of a constitutively active H-Ras ( Fig 8C & 8D ) . However , it did not affect the activation of ERK1/2 mediated by HERV-K Env or JSRV Env . This suggests that either these Envs act downstream of Ras or that they activate another form of Ras ( e . g . N-Ras ) . In contrast , TAK632 , a potent pan-Raf inhibitor , completely abolished ERK1/2 signal transduction and transcription factor induction by both JSRV and HERV-K Envs , indicating that the two glycoproteins act upstream of the kinase Raf . U0126 , a MEK1/2 inhibitor , also impaired the activation of ERK1/2 mediated by HERV-K and JSRV Envs , as expected . Of note , unlike JSRV Env , the inhibition was only partial for HERV-K Env .
In this paper , we report on the pro-oncogenic properties possessed by the Env protein of the HERV-K family and investigate the mechanism of action . Using the MCF10A cell line , we demonstrated that the stable expression of HERV-K Env , at a high but physiologically relevant level ( i . e . similar to the expression level naturally observed in some germ cell derived cell lines ) , induces clear changes in the expression of EMT-associated genes towards a more mesenchymal phenotype , with the cell morphology altered accordingly . Remarkably , these HERV-K Env-induced changes are accompanied by an increase in cell motility . The modification of the attachment proteins we observed is similar to that obtained after treatment with TGFß , but the transcription factors that are induced are different . A number of different transcription factors have been associated with EMT , and this generic term in fact covers several processes that can occur in different circumstances , either in normal development or in the course of disease [30] . Previous studies had already hinted at oncogenic properties for HERV-K Env , but it had only been shown that this Env protein can alter the phenotype of pre-transformed , cancer-derived cell lines [23–25] whereas we demonstrate here that it can also direct non-malignant cells in the path towards transformation . Given the change of the EMT markers and cell motility observed , it is likely that HERV-K Env expression by a tumour or a pre-tumour could also trigger further changes and favour metastasis . Using microarrays in a different cell model , we identified a very limited number of genes whose expression is induced following HERV-K Env transfection . We showed that the induction of these genes is due to activation of the ERK1/2 pathway in cells following HERV-K Env expression . Most of the identified genes are transcription factors that have already been associated with transformation and cancer . In fact , the list of the induced genes is remarkably similar to that observed in tumours with a mutated BRAF [46] . BRAF-activating mutations have been reported in several tumours , but are particularly common in melanomas [47] . Specific inhibitors of mutated BRAF have been developed and used to treat patients . They promote a dramatic improvement in patient health for up to 6 months until relapse , due to tumours developing resistance to the treatment [48] . HERV-K Env expression in melanomas has been reported by several independent groups [19 , 20 , 49–52] . It is possible that HERV-K Env expression is part of the mechanism used by the tumours to escape the treatment against mutated BRAF by re-activating the ERK1/2 pathway . Interestingly , a study recently showed that HERV-K expression in several breast cancer cell lines leads to an increase expression of ERK1/2 [25] , which indicates that HERV-K expression could increase ERK1/2 signalling by several independent mechanisms . The consensus HERV-K Env that we used to demonstrate ERK1/2 activation is more efficient than present-day alleles for most canonical virological properties . However , when tested under the same conditions , three out of the six previously characterised full-length “alleles” of HERV-K Env activated ERK1/2 nearly as efficiently as the consensus , giving credence to a possible role of HERV-K proviruses in tumour development . Of interest , these active alleles have been found to be spontaneously expressed in several human cell lines [52 , 53] . In addition , the effect of these alleles could be additive , especially when HERV-K proviruses are expressed due to general LTR activation ( eg the reported induction by MITF in melanoma [54] ) , instead of locus-specific expression . Remarkably , no other retroviral Env protein that we tested activated ERK1/2 , except for JSRV Env which is known to be a strong oncogene . Furthermore , the oncogenic activity of JSRV Env has been linked to ERK1/2 activation [36 , 38] , further supporting a role for HERV-K Env in tumour development . The effects observed with JSRV Env are admittedly stronger than those observed with HERV-K Env . However , it should be noted that the JSRV Env used in this study is that encoded by the infectious strain of the virus . Like HERV-K Env , the endogenous JSRV-18 Env was unable to transform 208F cells in our assays and no transforming effect has ever been reported for any of the endogenous copies [1] . It is not surprising that a gene showing such deleterious effects to the host should be lost very quickly following endogenisation . It is quite remarkable that the HERV-K family , which entered the primate lineage more than 40 million years ago , could have conserved some oncogenic properties for so long , even if these properties are slightly subdued . Indeed we showed that a recent , fully infectious endogenous JSRV Env protein has completely lost the ability to activate the ERK1/2 pathway , unlike some HERV-K Env alleles . It is possible that the last amplification of HERV-K elements in the human genomes is the product of a horizontal transmission of an infectious virus that would have remained active in other primate species , instead of the re-activation of previously dormant proviruses . It would be less unexpected for an exogenous , infectious virus to conserve such oncogenic properties that could play an important role in its propagation . Finally , we used mutant forms of HERV-K Env to map the domains responsible for ERK1/2 activation . We found that the TM subunit of Env is required , and that the cytoplasmic tail plays an important role , although there is still some ERK1/2 activation when it is deleted . Additionally , the 3 endogenous alleles of HERV-K Env-Rec that most strongly activate ERK1/2 all possess an intact cytoplasmic tail ( S2 Fig ) . However , the K115 allele possesses the same sequence and is much less potent for ERK1/2 activation . It is therefore likely that several domains of HERV-K Env cooperate for its oncogenic properties , as demonstrated for JSRV Env for which both SU and TM are involved in the transformation process [55] . Concerning JSRV Env , several intracellular interacting proteins expected to be involved in transformation have been reported recently [56 , 57] . In all cases , the reported interacting domain is the Env cytoplasmic tail . Since JSRV and HERV-K Envs show no homology in this region ( S3 Fig ) , these proteins are unlikely to interact with HERV-K Env . Using motif prediction software , we failed to identify any relevant motif in the sequence of the HERV-K Env cytoplasmic tail , and cannot therefore propose a likely mechanism of action . In the future , the elucidation of the cellular proteins interacting with the HERV-K Env cytoplasmic tail could lead to the development of specific inhibitors able to block its oncogenic properties , and could be of therapeutic interest in tumours where HERV-K Env is expressed , providing alternative treatments to those currently available .
HIV-1 derived particles were produced as described [43] , using a modified CSGW [58] expressing HERV-K Env or control proteins and the hygromycin resistance gene . pBabe-H-RasG12V [59] was kindly gifted by A . Puisieux . All Env transient expression vectors are CMV-driven . Plasmids for Ampho , IAPE , RD114 , GaLV , FeLV and HERV-K Envs have been described previously [60–64] . phCMV-JSRV Env was constructed by replacing the G protein ORF in phCMV-VSV-G ( GenBank accession no . AJ318514 ) with the JSRV Env ORF ( plus the 3’ LTR ) present in pCMV3JS21ΔGP [3] ( a gift from M . Palmarini ) . The Env gene from the enJSRV-18 provirus was similarly cloned by PCR on sheep genomic DNA using a forward primer on the ATG and the reverse primer indicated in [65] . phCMV-MMTV Env contains the MMTV Env ORF and the 3’ LTR from the pEnv vector [66] ( a gift from S . Ross ) . The HTLV-1 Env plasmid [67] was a gift from C . Pique . phCMV-HERV-K ( HML2 ) Env-LP was derived from phCMV-HERV-K ( HML2 ) Env by changing aa 103 and 104 into stop codons . pCMV-ß ( Clontech ) is an expression vector for beta-galactosidase . It was used as a control vector and designated “None” in the figures . It was also used to adjust total DNA content in transfection experiments . 293T ( ATCC CRL-3216 ) , HeLa ( ATCC CCL-2 ) and 208F ( ECACC 85103116 ) cells were maintained at 37°C , 8% CO2 , in DMEM with 10% heat-inactivated FCS , 100u/mL penicillin and 100μg/mL streptomycin ( PS ) . MCF10A cells ( ATCC CRL-10317 ) were cultured at 37°C , 10% CO2 , in DMEM:F12 medium supplemented with 5% horse serum , 5ng/mL EGF ( Peprotech ) , 10μg/mL insulin ( Sigma ) , 1ng/mL cholera toxin ( Sigma ) , 100μg/mL hydrocortisone ( Sigma ) and PS . Unless specified all reagents were from Life Technology . Lentiviral particles were produced as described [43] using JetPrime ( PolyPlus Transfection ) . Cells were infected with viral supernatants and selected with Hygromycin B ( 46u/mL , Calbiochem ) 3 days later . The populations of resistant cells were thereafter maintained in selection media . Cells in 12-well plates were transfected with 250ng total DNA ( 50 or 30ng of Env in 293T and HeLa cells respectively , supplemented with pCMV-ß ) using 1 . 25μL of Fugene6 ( Promega ) . Media were replaced 18 hours post transfection ( without FCS for Hela cells to minimize background ) . When used , inhibitors were added during the medium change ( FTI-277 ( 5μM , Sigma ) , TAK-632 ( 5μM , Selleckchem ) , U0126 ( 5μM , Cell Signaling ) ) . TNFα was used at 100ng/μL ( R&D Systems ) . Cells were used for protein or RNA extraction 48 hours post transfection . For Western blot analysis , cells were lysed in PBS , 1% NP40 or RIPA ( Life Technology ) complemented with Halt protease and phosphatase inhibitor cocktail ( ThermoScientific ) . Cell lysates were then subjected to SDS-PAGE as described [43] . Proteins of interest were detected using antibodies from Cell Signalling: p44/42 MAPK , phospho-p44/42 MAPK ( Thr202/Tyr204 ) , p38 MAPK , phospho-p38 MAPK ( Thr180/Tyr182 ) , pan-Akt , phospho-Akt , IKBα or Sigma ( Tubulin ) . HERV-K ( HML2 ) anti-Env antibody was previously described [61] . MLV ampho Env protein was detected using a goat antiserum directed against Rauscher leukemia virus gp70 ( from the National Cancer Institute , Frederick , MD ) . HERV-K Rec protein was detected using a polyclonal rabbit antiserum given by R . Löwer . HRP-conjugated secondary antibodies ( GE Healthcare Or Dako ) and ECL Plus Reagent ( GE Healthcare ) were used for Western blots . Membrane stripping was done using ReBlot Plus Strong ( Merck Millipore ) . Total RNA were extracted with the RNeasy extraction kit ( Qiagen ) and treated with DNase I ( Ambion ) . For microarray experiments , we compared duplicates of RNA from 293T transfected with either HERV-K ( HML2 ) Env or the control plasmid , HERV-K ( HML2 ) Env-LP , collected 24 and 48 hours post-transfection . Gene expression analysis was performed on Agilent SurePrint G3 Human GE 8x60K Microarrays ( Agilent Technologies , AMADID 39494 ) . Data were extracted using Feature Extraction software ( v10 . 5 . 1 . 1; Agilent Technologies ) and normalized using an empirical Bayes method . Top-ranked genes were selected for an absolute fold-change ≥2 using a False Detection Rate ( FDR ) <0 . 05 . DNase-treated RNAs were reverse-transcribed using the MLV reverse-transcriptase ( Applied Biosystems ) . qPCR was performed using the QuantiFast SYBR Green PCR kit ( Qiagen ) on the ABI PRISM 7000 system . Efficacy of the PCR reaction was checked for each primer pair . Transcript levels were normalized to RPLO employing the ΔΔCt method . MCF10A cells were resuspended in DMEM:F12 media without serum or additional additives . 5x104 cells in 500μL were seeded into the top of each transwell ( Corning , 24-well inserts , 8μm pore ) , and 750μL of complete culture medium was added to the bottom of each well as a chemoattractant . The cells were incubated for 22 hours before non-migratory cells were removed and the membrane fixed with methanol . Membranes were stained with DAPI and the migrated cells counted . Invasion assays were performed similarly except transwells were precoated with 16μg of Matrigel ( Corning ) . MCF10A cell populations were incubated in DMEM/F12 media supplemented with 5μM CellTracker Green ( Invitrogen ) for 30 minutes , and then cultured for an additional 60 minutes in complete media before fixation in 4% PFA . Cells were imaged to assess changes in morphology . Average cell size ( area ) was calculated by measuring the surface area covered by the cells ( green stain ) relative to the number of cells . 208F cells ( seeded at 2x105 cells per 3 . 5 cm dish the day before ) were transfected with 4μg DNA using Fugene 6 ( Promega ) . After 24h , they were reseeded in a 10 cm dish . When the cells reached confluence , medium was replaced by DMEM complemented with PS , 5% FCS and 1μM dexamethasone . Medium was replaced weekly . After 3–4 weeks , cells were stained with Leishmann to allow counting of transformed foci .
|
Nearly half the DNA of mammals consists of reitarated , selfish elements that can move and amplify within the genome . With time , some of these elements are recruited by the host and the proteins they encode are used to fulfill physiological functions , whereas other elements have conserved some of their pathological properties and contribute to the development of diseases . The human HERV-K ( HML2 ) elements originated from an ancestral infection of the primate germline by an infectious retrovirus that has been maintained and amplified in the human lineage . It is associated with several pathologies in modern humans , in particular cancer of the breast , germline and skin . We show that the HERV-K ( HML2 ) envelope protein is able to activate a major cellular signalling pathway often involved in human cancers , and that its expression promotes a series of cellular changes that are characteristic of cancer development . Altogether , this study indicates that the expression of HERV-K ( HML2 ) elements is not only a marker of cancer , but can also directly participate to tumourigenesis via the newly discovered oncogenic properties carried by the envelope protein .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] |
[
"phosphorylation",
"transfection",
"medicine",
"and",
"health",
"sciences",
"gene",
"regulation",
"293t",
"cells",
"regulatory",
"proteins",
"biological",
"cultures",
"rna",
"extraction",
"dna-binding",
"proteins",
"oncology",
"transcription",
"factors",
"bioassays",
"and",
"physiological",
"analysis",
"molecular",
"biology",
"techniques",
"extraction",
"techniques",
"research",
"and",
"analysis",
"methods",
"proteins",
"gene",
"expression",
"cell",
"lines",
"molecular",
"biology",
"microarrays",
"biochemistry",
"carcinogenesis",
"post-translational",
"modification",
"genetics",
"biology",
"and",
"life",
"sciences"
] |
2017
|
A human endogenous retrovirus-derived gene that can contribute to oncogenesis by activating the ERK pathway and inducing migration and invasion
|
Recent advances in epigenomics have made it possible to map genome-wide regulatory regions using empirical methods . Subsequent comparative epigenomic studies have revealed that regulatory regions diverge rapidly between genome of different species , and that the divergence is more pronounced in enhancers than in promoters . To understand genomic changes underlying these patterns , we investigated if we can identify specific sequence fragments that are over-enriched in regulatory regions , thus potentially contributing to regulatory functions of such regions . Here we report numerous sequence fragments that are statistically over-enriched in enhancers and promoters of different mammals ( which we refer to as ‘sequence determinants’ ) . Interestingly , the degree of statistical enrichment , which presumably is associated with the degree of regulatory impacts of the specific sequence determinant , was significantly higher for promoter sequence determinants than enhancer sequence determinants . We further used a machine learning method to construct prediction models using sequence determinants . Remarkably , prediction models constructed from one species could be used to predict regulatory regions of other species with high accuracy . This observation indicates that even though the precise locations of regulatory regions diverge rapidly during evolution , the functional potential of sequence determinants underlying regulatory sequences may be conserved between species .
Epigenomic modifications such as histone modifications and DNA methylation play critical roles in development , regulation , and diseases . The study of epigenetic modifications has made great strides in recent decades , and the specific combinations of different epigenome components in distinct biological conditions are rapidly being discovered [1] . In particular , epigenomic profiling is widely used to empirically identify regulatory regions including enhancers and promoters using chromatin immunoprecipitation with massively parallel DNA sequencing ( ChIP-seq ) . For example , genomic regions enriched for histone H3 lysine 27 acetylation ( H3K27ac ) are considered as active enhancers [2 , 3] . On the other hand , enrichment of histone H3 lysine 4 trimethylation ( H3K4me3 ) , in particular together with H3K27ac , indicates active promoters [4–6] . Beyond identifying regulatory regions , the next challenge is deciphering what factors determine and affect epigenomes . Among potential factors , the importance of cis-regulatory sequences on the epigenome is well appreciated . Several cis-regulatory sequences based predictive models have been constructed to classify regulatory regions [7–10] . For example , a recent study reported random forest classifier models from the human genome that could predict regulatory regions marked by H3K27ac and H3K4me3 modifications with relatively high accuracy [11] . Even though our understanding of the true nature of the relationship between specific histone modifications and regulatory regions is sure to undergo much more revisions , these technical advances in genome-wide epigenomic profiling brought new approaches to study evolution of regulatory regions . Instead of having to rely on experimentally characterized comparative transcription factor binding assays [12–14] and/or regions that retain sequence similarities [15–18] , enhancers and promoters can be identified based on the distribution of specific epigenomic modifications such as H3K4me3 and H3K27Ac across different species [6 , 19] . Interestingly , these studies show that at the genome-scale , chromosomal locations of enhancers are highly divergent between species [6 , 20 , 21] . Promoters are also found in divergent locations , although their positions are more constrained than enhancers , since promoters are typically adjacent to transcription units ( e . g . [6] ) . Thus , while regulatory regions can be reliably predicted from sequences within specific genomes [7–11] , the precise locations of regulatory regions , in particular of enhancers , diverge rapidly during evolution [6 , 18 , 20 , 21] . It is not necessarily straightforward to reconcile these two aspects of regulatory regions . In the simplest scenario , functional regions such as enhancers and promoters should be evolutionarily conserved since they are subject to purifying selection . Indeed , this idea has been successfully used to identify non-coding sequences with regulatory functions [16 , 17 , 22 , 23] . However , at the genome-scale , regulatory regions harbor little sequence similarities and their locations are highly divergent . Rapid turnover of transcription binding sites [12 , 24 , 25] and transcription rewiring [26–28] can explain some aspects of regulatory sequence evolution , but many questions still remain [29 , 30] . Here , utilizing the wealth of comparative data on epigenomically determined enhancers and promoters , we investigated whether we could identify specific sequence fragments that constitute enhancers and promoters , and if so , whether such sequence fragments were evolutionarily conserved between species . We first performed an exhaustive search to identify sequence fragments that are statistically over-represented in experimentally identified enhancers and promoters of several mammals [6] . A unique aspect of our study is that we focused on distinguishing regulatory regions from nearby regions . Genomic sequences of mammals such as humans are highly heterogeneous in many aspects such as GC contents , transposable element contents , genic contents , and other aspects [31 , 32] . By comparing regulatory regions to their nearby non-regulatory regions , we identified sequence fragments that distinguished regulatory regions from its local genomic backgrounds . Our comprehensive exhaustive search revealed numerous sequence fragments that were significantly enriched in regulatory regions compared to nearby regions . Due to the nature of the exhaustive search , some of the identified sequence fragments may be inter-related . To overcome this limitation and identify a subset of sequence fragments that are statistically independent , and to construct prediction models to test evolutionary hypotheses , we employed a machine learning method . Specifically , we used the least absolute shrinkage and selection operator ( LASSO ) method [33] , which can effectively select one variable among the set of highly correlated variables [34] . The LASSO method is also excellent at prediction accuracy [11 , 35] . From these procedures , we discovered numerous sequence fragments that are statistically enriched in experimentally verified regulatory regions ( referred to as ‘sequence determinants’ henceforth ) . Intriguingly , sequence determinants obtained from enhancers and promoters show remarkable differences with respect to their impact on functional regions . Moreover , even though sequence determinants themselves exhibit only moderate overlaps between species , prediction models constructed using sequence determinants from different species could be inter-changed to perform as well as prediction models from the focal species . We discuss potential implications of these findings .
We used experimental annotations of liver enhancers and promoters from a previous study [6] . Following the definition in this study [6] , we considered enhancers to be regions marked only with the H3K27ac mark and promoters to be regions marked with H3K4me3 ( with or without H3K27ac ) . We selected data from seven ‘high-quality’ mammalian genomes as indicated in [6] , including Home sapiens ( human ) , Macaca mulatta ( macaque ) , Bos taurus ( cow ) , Sus scrofa ( pig ) , Canis familiaris ( dog ) , Rattus norvegicus [32] , and Mus musculus ( mouse ) . Each enhancer or promoter was designated as foreground , and a segment of the same length 100 , 000 base-pairs ( 100kb ) apart from the foreground was selected as the background . We used these ‘regional’ backgrounds to control for potential chromosome effect and/or regional effects . The distance of 100kb between the foreground and background was selected since several genomic features such as linkage disequilibrium blocks and GC contents show correlations that extend to ~ 100kb [32 , 36] . We obtained the genome sequences using the R Bioconductor libraries “BSgenome” [37] . Backgrounds that had greater than 50% of nucleotides missing ( not sequenced ) were discarded ( Table 1 ) , and put information on overlapped proportions between foreground and background in S1 Table . Those enhancers and promoters found in orthologous locations across species were identified as conserved ( Table 1 ) . Specifically , for each human enhancer or promoter we retrieved the 17 eutherian EPO multiple alignment using Ensembl REST API [38] and determined if the region was conserved or not based on whether all other 6 species also showed the same histone mark ( s ) in the orthologous region . For species with different genome assemblies in the alignment , we converted the coordinates using Ensembl assembly converter [39] . We examined whether specific sequence fragments in the foreground were over-represented compared to the backgrounds by statistical testing . We used sliding windows with a specific length ( from 6-mers to 15-mers ) , moving from the 5’ end to the 3’ end in each foreground or background ( Fig 1 ) . As the window moved by a base-pair ( bp ) , a sequence fragment within that bin was captured and recorded . Following this sliding window analysis , counts of each sequence fragment in the foreground and background were obtained . For each sequence fragment , we constructed a 2×2 contingency table that contained counts of a sequence determinant in each of foreground and background region ( Table 2 ) , and we used the odds ratio ( OR ) as a measure of over-representation in foreground , compared to background . The magnitude of OR indicated how strongly over-enriched a specific element was in regulatory regions , which we also referred to as ‘effect size’ in this study . We used the χ2 test to test the following null and alternative hypotheses: H0:OR=1 , ( 1 ) H1:notH0 . ( 2 ) If the expected count of a sequence fragment in any of the cell in the 2×2 contingency table was lower than 5 , we used the Fisher’s exact test instead . The resulting P-values were corrected for multiple testing using the false discovery rate ( FDR ) approach [40] . Following these procedures , a ‘sequence determinant’ in the statistical sense was identified as a sequence fragment whose FDR Q-value was equal to or less than 0 . 05 and the OR was greater than 1 . In the process , we tested only sequence determinants that appeared over 100 times to avoid selecting rare sequence determinants of negligible biological relevance . For example , for 15-mers in the human enhancer data set , most sequence fragments ( 63 million out of 70 million ) occurred only once . We repeated this procedure for each of the seven species and identified ‘species sequence determinants’ . We identified ‘common sequence determinants’ as sequence fragments that are enriched in foreground regions compared to the background regions across the seven mammalian species . For the purpose , we used the Cochran-Mantel-Haenzel ( CMH ) test [41] to identify enrichment of sequence determinants from multiple data sets using a conditional variable , which is a nominal covariate such as the species index [41 , 42] . The CMH test is also equivalent to the score type test of logistic regression , which has advantages in the handling of sparse count data sets [42] . Consequently , we used the CMH to test the null hypothesis , H0:OR|species=1 , whereOR|speciesistheconditionalORinpresenceofthespeciesindex . ( 3 ) H1:notH0 . ( 4 ) Common sequence determinants were then defined as those whose OR|species>1 for all species and FDR Q-value from CMH ≤ 0 . 05 . We constructed prediction models that yield predictive scores for each region . We used the least absolute shrinkage and selection operator ( LASSO ) method [33] , which excels at prediction accuracy as well as covariate selection [11 , 35] . In the LASSO model , each foreground or background region was regarded as a binary observation ( foreground = 1 , background = 0 ) . The relative frequency of each sequence determinant was regarded as an explanatory variable . Because the space of all significant sequence determinants was extremely large ( S2 and S3 Tables ) , including all determinants in the LASSO model was not computationally feasible . Instead , we selected 10 , 000 sequence determinants , sampled according to their distribution of GC content and fragment length , to incorporate in the LASSO models using a stratified sampling approach [43] . Specifically , we stratified the whole sequence determinants by the combination of GC content ( ten uniform intervals: [0~0 . 1] , … , ( 0 . 9~1 . 0] ) and length ( ten lengths: 6 , … , 15bp ) . Then we selected samples from each of the stratified subsets so that its number out of the 10 , 000 was proportional to the number of determinants in the specific subset among the total determinants . To train LASSO models and estimate coefficient of each determinant , we used the R function “glmnet” from the package “glmnet” using R 3 . 4 . 0 . To construct prediction models , we used both the 10 , 000 species sequence determinants and the 10 , 000 common sequence determinants as input variables , so that we can compare the prediction performances of species determinants and common determinants . We performed two types of predictions . First , we performed same-species prediction , which evaluates prediction AUC through a 10-fold cross-validation process [11 , 35 , 44 , 45] . During the 10-fold cross-validation process , an optimal penalty parameter that provides the smallest test AUC is chosen . We regarded the smallest test AUC as same-species prediction AUC . For inter-species prediction , we used the optimal parameter to construct a prediction model from whole data set of a species and applied the model to the other species to calculate inter-species prediction AUCs . Workflow from the exhaustive search to LASSO is depicted in Fig 1 . In most prediction results , we provided two types of AUC , the first one is receive operating characteristic AUC ( ROC-AUC ) for general performance of prediction and the second one is precision-recall AUC ( PR-AUC ) for robustness of performance regardless of the ratio between numbers of foreground and background [46] . Among several machine-learning methods , we selected LASSO because of its ability to reduce the number of input variables so that those are not redundant and are statistically meaningful . However , other machine learning methods might be useful as well . For example , when many of sequence determinants have strong relationship in terms of correlation , elastic net that can capture more input variables would be useful to improve prediction performances [47] . We examined the presence of transcription factor binding sites ( TFBS ) in the sequence determinants using TOMTOM [48] . This tool assesses the similarity between individual sequence input and specific TFBS databases and provides P-values and Q-values adjusted by FDR . Known TFBS compiled in the JASPAR 2014 Core vertebrate database [49] , the HOCOMOCOv10_HUMAN and the HOCOMOCOv10_MOUSE [50] were used . We summarized the proportion of significant ( P <0 . 05 ) TFBS hits as ‘TFBS frequency’ . For example , each human sequence determinant was compared to the 641 known TFBS in the HOCOMOCOv10_HUMAN database . The number of significant comparisons out of the total 641 comparisons was referred to as ‘TFBS frequency’ . Due to the probabilistic nature of TF binding and the fact that sequence determinants might encode partial or full TFBS , TFBS frequency indicates versatility of a sequence determinant that can be a motif for TFBS binding . For instance , the CAGCCC determinant from the human genome yielded 18 of 641 significant hits , thus TFBS frequency of the determinant was 2 . 8% . We also used–log10min ( P ) instead of TFBS frequency to evaluate the best match between a k-mer and the motifs in the database . Sequence determinants from the exhaustive search as well as from the LASSO prediction models were further analyzed to explore relationships between their effect sizes and several biological factors such as GC content and TFBS binding properties . For this analysis , we used the following linear model; log2 ( OR ) i∼GCcontenti+TFBSfrequencyi+GCcontenti×TFBSfrequencyi+εi , ( 5 ) where i is the index of each sequence determinant and εi ~N ( 0 , σ2 ) . In this model , we log2 transformed the OR values to improve normality . We applied the model to enhancer and promoter sequence determinants from common , human , and mouse sets .
To identify sequence fragments that are significantly enriched in enhancers or promoters compared to nearby background regions ( sequence determinants ) , we first performed an exhaustive search . Briefly , we examined sequence fragments of lengths from 6 to 15 bp , using a sliding window approach ( Fig 1 ) . We tested statistical over-representation of the specific sequence fragment in the enhancers or promoters compared to their backgrounds using a contingency table test based on their ORs . The P-values were adjusted via the false discovery procedure [40] ( Materials and Methods ) . Following these procedures , we identified numerous sequence determinants associated with enhancers and promoters of each species ( referred to as ‘species sequence determinants’ , Materials and Methods ) . Fig 2 ( A ) and 2 ( B ) show the numbers of significant sequence determinants from human enhancers and promoters based on their OR and length . The majority of sequence determinants in enhancers and promoters were found in 7–11 bps . Human enhancer determinants were slightly yet significantly longer than promoter determinants ( mean lengths for human enhancers and promoters were 9 . 20 and 9 . 01 , P <1×10−5 by two sample t-test ) . However , there was no consistent pattern across the seven mammals when comparing the length of sequence determinants in enhancers and promoters . Sequence determinants were also generally GC-rich and TFBS-rich compared to non-significant sequence fragments ( see below ) . Remarkably , with respect to OR , sequence determinants from enhancers and promoters were highly distinct . Strongly enriched sequence determinants , such as those with OR ≥ 2 . 0 , were 140-fold more abundant in promoters than in enhancers ( Fig 2 ) . Accordingly , the ORs of sequence determinants were significantly higher in promoter sequence determinants than in enhancer sequence determinants ( P < 10−15 by Wilcoxon’s rank sum-test in all seven species , Fig 3 ) . We then examined sequence determinants that occurred more frequently than expected in all seven mammalian species , which we referred to as ‘common sequence determinants’ ( Materials and Methods ) . Similar to the results from the above analysis , common sequence determinants had higher ORs in promoters than in enhancers ( P < 10−15 by Wilcoxon’s rank sum-test , Fig 2 ( C ) and 2 ( D ) , S4 Table ) . When we compared the entries of common sequence determinants to those of species sequence determinants , we found that 39% and 57% of all human enhancer and promoter determinants overlapped with common enhancer and promoter sequence determinants , respectively ( S5 Table ) . Therefore , regardless of their species-wise distribution , sequence determinants that mark promoters tended to have significantly greater OR thus presumably stronger effects on regulatory potential of target regions in terms of marginal effect size , compared to those found in enhancers . The exhaustive search allowed us to identify all sequence determinants that were marginally enriched . However , some sequence determinants might be highly correlated with each other , because they were extracted from overlapping regions ( Fig 1 ) . The LASSO approach is capable of selecting one variable among the highly correlated variable sets , in addition to selecting variables of substantial effect [33] . Therefore , we next used the LASSO approach to select essential variables among the many correlated variables , and to construct prediction models that discriminate enhancers and promoters from their corresponding background regions ( Materials and Methods ) . The total numbers of sequence determinants from the human enhancers and promoters were 107 , 287 and 101 , 625 , respectively ( S2 and S3 Tables ) . AUCs increased as the number of input sequence determinants increased , to stabilize around 7 , 000 sequence determinants ( S1 Fig ) . We thus chose 10 , 000 sequence determinants for each set of sequence determinants using a stratified sampling approach [43] , to select a subset that is representative of the original distribution with respect to GC contents and lengths ( Materials and Methods ) . Following these steps , prediction models were constructed for both same-species prediction and inter-species prediction . We investigated the distribution of ORs and the lengths of selected sequence determinants from the LASSO approach ( ‘LASSO-selected sequence determinants’ ) , and from same-species prediction . The same-species prediction model of human enhancers and promoters had a total of 4321 and 1343 LASSO selected sequence determinants , respectively ( S6 and S7 Table ) . Consistent with the results from the exhaustive search , marginal ORs from the enhancer models were significantly lower than those from the promoter models in all species ( Wilcoxon test , P < 10−15 , S2 Fig ) . We investigated the relative frequencies of individual LASSO-selected sequence determinants in foreground and background regions , shown as density plots in S3 Fig . In promoters , marginal density of the relative frequencies of LASSO-selected sequence determinants is highly distinct from that of the background , which is consistent with the high effect size of LASSO-selected promoter sequence determinants . On the other hand , marginal densities of LASSO-selected enhancer sequence determinants are similar to those in the background . This observation indicates that in addition to having weaker marginal effects than promoter sequence determinants , the frequency distribution of enhancer sequence determinants is similar between foreground and background . Interestingly , LASSO selected sequence determinants were significantly longer for enhancers than for promoters ( mean lengths of 9 . 22 in enhancers and 8 . 32 in promoters in human , P <1×10−15 by two sample t-test , S6 and S7 Table ) . This pattern was consistent in other species ( P < 10−15 by two-sample t-test in all cases ) . When we applied LASSO approach to 10 , 000 common sequence determinants , we observed similarly significant differences of effect size and length between enhancer and promoter sequence determinants ( S2 and S4 Figs ) . We examined two aspects of sequence determinants to understand what features affect enhancer and promoter potentials of specific sequence fragments . Specifically , we used a linear model to analyze the effect of the frequency of G and C nucleotides ( GC content ) and the frequency of transcription factor binding sites ( TFBS frequency ) . The effect sizes of sequence determinants were response variables , and GC content , TFBS frequency , and their interaction term were explanatory variables . When we analyzed the results of the LASSO-selected sequence determinants , several patterns became clear . First , this model explained a large amount of variation observed in promoter sequence determinants , but only a modest portion of those in enhancer sequence determinants ( Table 3 ) . Nevertheless , we found that main factors of GC content and TFBS frequency were positively correlated with the log2-transformed OR of sequence determinants both in enhancers and promoters ( Table 3 , S8 and S9 Tables ) . However , interaction terms between the two main factors were significantly negative only in promoters . Thus , while GC content and TFBS frequency worked additively to determine the strength of regulatory potential for enhancer sequence determinants , these two factors were antagonistic with each other in promoter sequence determinants ( Fig 4 ( A ) and 4 ( B ) ) . This observation is consistent with previous studies that found a lack of transcription factor binding enrichment at GC-rich promoters compared to GC-poor promoters [51] . We also evaluated–log10min ( P ) instead of TFBS frequency to evaluate the best match between a k-mer and the motifs in the database , and obtained highly similar results for the same models ( S10 Table ) . In summary , TFBS frequency was positively correlated with effect size in both of enhancer and promoters when GC content was low . On the other hand , the estimated coefficients of GC content and TFBS frequency were higher in promoters than in enhancers , indicating that the effects of these factors were stronger in promoters compared to in enhancers . Accordingly , the R2 of the linear models were substantially higher for promoters than for enhancers ( Table 3 , S8 and S9 Table ) . Second , the relationships between GC contents and TFBS frequency were negative in both of enhancer and promoter analysis ( Fig 4 ( C ) and 4 ( D ) ) . Accordingly , sequence determinants that were GC-rich tended to lack TFBS , and low GC sequence determinants tended to harbor more TFBS than high GC sequences [51] . The whole set of sequence determinants obtained from exhaustive search yielded similar results ( S11 Table ) . The prediction accuracy of the human promoter same-species prediction model was very high , with an AUC of 0 . 97 ( Fig 5 ) . Same-species prediction models from other six species exhibited similarly high AUCs ( S12 and S13 Table ) , indicating that promoters can be accurately predicted from sequence determinants . We also evaluated prediction AUCs using 10 , 000 non-sequence determinants , while matching the distributions of GC content and length as those of sequence determinants . We then constructed prediction models using LASSO for enhancers and promoters in human and mouse , respectively . We iterated the process five times to measure variability of the AUCs . Results are shown in S6 Fig . The AUCs of models using non-sequence determinants were lower than AUCs with sequence determinants . For example , human and mouse enhancer prediction AUCs with non-sequence determinants showed 0 . 507 and 0 . 002 , and 0 . 500 and 0 . 007 for mean and standard deviation , respectively . These results indicate that non-sequence determinants had poor prediction performances . In case of promoters , the mean and standard deviation of AUCs were 0 . 636 and 0 . 006 for human , and 0 . 608 and 0 . 004 for mouse , respectively . These values were higher than those of enhancers , likely reflecting the effect of GC contents ( e . g . , [52] ) . Nevertheless , they were substantially lower than the AUCs with sequence determinants , indicating that sequence determinants have superior prediction performances than non-sequence determinants . Next , we tested if prediction models constructed from one species could be used in different species , to investigate if different genomes use similar sequence determinants to encode promoters . Indeed , when we calculated AUCs of inter-species prediction between seven species of promoters , the AUCs were all above 0 . 9 , indicating high accuracy ( Fig 5 ) . On the other hand , the LASSO prediction models of enhancers had the following differences from those of promoters . First , the enhancer models using 10 , 000 species determinants had 2 . 5- to 4 . 2-fold greater numbers of explanatory variables than the promoter models ( S12 Table ) . However , their AUCs were generally lower than those of the promoter models ( Fig 5 ) . We found that same-species prediction AUCs for enhancer models were greater than 0 . 7 , and the highest was when mouse model were used to predict mouse enhancers , 0 . 76 ( S12 Table ) . Nevertheless , inter-species prediction results using enhancer models showed similar AUCs to same-species enhancer predictions ( Fig 5 ) . We tested if the high inter-species prediction accuracies were driven by the presence of highly conserved regulatory elements across different mammalian species . The proportions of conserved enhancer regions among the seven species were much smaller than those of promoter regions , as previously described [6] ( Table 1 ) . Interestingly , we observed similar AUCs before and after removing highly conserved regulatory regions at both enhancers and promoters ( S14 Table ) , suggesting that conserved regulatory regions were not responsible for the high predictabilities across species . We then extracted 10 subsets of 10 , 000 sequence determinants from human enhancer and promoter sequence determinants ( all subsets were mutually exclusive with each other subset ) and constructed LASSO models to apply to the same-species ( human ) prediction and inter-species ( mouse enhancer ) prediction . We found that the AUCs of these 10 subsets were highly similar ( S6 Fig ) . Thus , even though the regulatory regions themselves were not conserved in terms of their precise location , mammalian enhancers and promoters have inter-changeability in terms of prediction between species . We also constructed LASSO models using 10 , 000 common sequence determinants from all seven species . AUC values for promoter prediction were highly similar to those obtained from models using species sequence determinants ( Fig 5 , S13 Table ) , indicating that sequence fragments that were commonly enriched in all 7 species harbor sufficient signals for promoter prediction . On the other hand , enhancer prediction results using 10 , 000 common sequence determinants showed slight decrease of AUC compared to same-species prediction ( mean AUC of prediction with species determinants: 0 . 723 , that with common determinants: 0 . 679 ) . Interestingly , mean numbers of LASSO selected common sequence determinants were significantly lower than the species ones in enhancers ( 1613 and 3970 for common sequence determinants and species sequence determinants , respectively; P <1×10−5 by paired t-test ) , while they were not significantly different in promoter models ( 1138 and 1342 for common sequence determinants and species sequence determinants , respectively; P = 0 . 2114 by paired t-test ) . This implies that each of the common enhancer sequence determinants may have higher predictive capabilities than species sequence determinants . While background and foreground of enhancers exhibit similar GC distribution , foreground regions of promoters are substantially skewed towards GC-rich regions ( the average difference was 10 . 0% , higher in promoters than in enhancers ) ( S7 Fig ) . Therefore , we investigated how GC content difference between foreground and background might affect prediction analyses . First , to measure the impact of GC content alone in prediction performances , we calculated AUCs using only GC content as a predictor ( Table 4 ) . Second , we constructed LASSO models using sequence determinants of low-GC content ( GC content ≤0 . 5 ) to measure prediction performances without effects of high GC content sequence determinants . For this analysis , we randomly selected 10 , 000 sequence determinants with stratification of GC content and sequence length . These results were then compared to those of the original AUCs . We found the AUCs using only GC content reflected the amount of GC content differences between foreground and background ( S7 Fig ) . For example , average AUCs using only GC content were 0 . 589 and 0 . 837 in enhancers and promoters , respectively . However , both of those AUCs were considerably lower than the original AUCs ( differences of 0 . 134 and 0 . 107 in enhancers and promoters , respectively ) , meaning that GC content could not explain all of the variation between foreground and background . This observation is consistent with a prior study utilizing a similar approach [52] . Moreover , models with low-GC sequence determinants had higher AUCs than those using only GC contents . In other words , models without high GC content sequence determinants outperformed the AUCs with only GC contents . Interestingly , mean AUCs with low-GC sequence determinants in enhancers were even higher than those of the original AUCs , which may imply that low-GC enhancers sequence determinants had better prediction performances than high-GC sequence determinants when they were jointly used for prediction . In conclusion , prediction performances of the sequence determinants detected by LASSO cannot be attributed to their GC contents .
Understanding specific histone modifications marking enhancers and promoters has opened the way to identify these regions using ChIP-seq , which complements and scales up traditional transcription factor binding assays [1 , 6 , 53] . Even though our understanding of the exact molecular nature of regulatory regions continues to improve , technical advances in epigenomic assays have opened a new opportunity to study evolution of regulatory regions using unbiased genome-wide epigenomic profiling . We were motivated by two observations: that regulatory regions identified from epigenomic assays can be predicted with high accuracy in case of same-species prediction [7–11] , yet that they are highly divergent between different species [6 , 18 , 20 , 21] . The fact that regulatory regions can be predicted with high accuracy implies that specific sequence fragments can encode regulatory function . Indeed , previous studies often referred to such fragments as cis-regulatory motifs . Since they encode function , they are likely to be subject to natural selection ( largely purifying selection ) and thus evolutionarily conserved . However , genome-wide studies indicate that regulatory regions , especially enhancers , are highly divergent between species . To investigate this potentially paradoxical pattern of evolution of regulatory regions , we used a powerful approach to examine every possible sequence fragments for their statistical enrichment in experimentally verified enhancers and promoters of seven mammalian species . This approach , which we named exhaustive search , revealed that numerous sequence fragments were statistically over-represented in enhancers and promoters ( which we named as sequence determinants ) . Sequence determinants underlying enhancers and promoters exhibited intriguing differences with respect to their degree of enrichment ( effect size ) , GC content , and the frequencies of known TFBS . Notably , the degree of statistical enrichment was significantly higher for promoter sequence determinants compared to enhancer sequence determinants . This observation suggests that sequence determinants may have greater impacts on the regulatory potential of promoters than of enhancers . This idea is also consistent with the fact that promoters are more evolutionarily conserved than enhancers [6] . We next applied a machine-learning method , LASSO , to reduce interdependence among sequence determinants and construct prediction models based on the non-redundant sequence determinant set . Same-species prediction models generated from these sequence determinants had high AUCs for enhancers and promoters ( Fig 5 and S12 and S13 Tables ) , affirming the predictor power of sequence determinants [11 , 52] . The AUCs from these models are on par with those from previous studies that utilized different approaches ( e . g . , [11] ) . We observed that enhancer models utilized greater numbers of predictors yet exhibited lower accuracy compared to promoter models , which can be explained by promoter sequence determinants associated with significantly higher effect sizes compared to enhancer sequence determinants ( Figs 2 and 3 , S2 and S4 Figs ) . Furthermore , we applied prediction models generated from one mammal to other mammals , to directly test whether sequence determinants from one species could be used to predict regulatory regions in other species . Remarkably , even though the sequence determinants themselves had only moderate overlaps between species ( S5 Table ) , models constructed from one species could predict promoters in other species with high accuracies ( S12 and S13 Tables ) . As for enhancer models , AUCs from inter-species prediction models were also comparable to same-species predictions ( Fig 5 ) . In other words , the extent to which prediction models could be inter-changed between species was similar between enhancers and promoters ( Fig 5 ) . We used a cutoff effect size for sequence determinants as 1 , for the following reasons . First , many sequence determinants have extremely low p-values despite low effect sizes due to their abundance , especially those with shorter lengths . For example , 25% of human enhancer sequence determinants among those of top 10 , 000 lowest p-values have effect sizes smaller than 1 . 2 . Second , when we constructed a human enhancer prediction model using randomly selected 10 , 000 sequence determinants with effect sizes smaller than 1 . 2 , the resulting AUC was 0 . 715 , which is equivalent to the original AUC . Moreover , when we applied this model to mouse , the inter-species AUC was 0 . 680 , even higher than the original AUC ( 0 . 647 ) . Therefore , setting an arbitrary cutoff value is likely to result in the loss of true sequence determinants that are important in terms of prediction performances . Integrating the main findings that 1 ) there are a large number of sequence determinants that potentially contribute to the regulatory roles of enhancers and promoters; 2 ) the strength of statistical enrichment of sequence determinant is greater for promoters , which are more evolutionarily conserved than enhancers; 3 ) prediction accuracies of models generated using sequence determinants from different species are comparable to each other , we hypothesize the following . Even though the specific motifs that encode regulatory regions are different between species [6 , 18 , 20 , 21] , the function of specific sequence determinants could be conserved between species . There may exist a large reservoir of potential sequence determinants that can contribute to regulatory regions of many species .
|
Regions of the genome that do not encode genes but affect expression of other genes , such as enhancers and promoters , are referred to as regulatory regions . Because of their regulatory functions , it was thought that enhancers and promoters should be evolutionarily conserved . Regulatory regions can be now epigenomically identified because they are marked by specific modifications of histone tails at the chromatin level . Interestingly , when we compare epigenomically identified regulatory regions from different mammals , the specific positions of regulatory regions are often divergent between species . Enhancers in particular are highly divergent between species . In this study , we show that we can find sequence fragments that are statistically enriched in enhancers and promoters of different species , and that the degree of statistical enrichment can explain different levels of evolutionary sequence conservation between enhancers and promoters . We further constructed predictive models of enhancers and promoters using the enriched sequence fragments , and show that these models can not only accurately predict enhancers and promoters of the same species , but works comparably well when applied to other species . These results indicate that even though the specific positions of regulatory regions have diverged between species , the functions of sequence fragments that comprise those regions may be conserved .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion"
] |
[
"comparative",
"sequence",
"analysis",
"chemical",
"characterization",
"statistics",
"vertebrates",
"animals",
"mammals",
"mathematics",
"forecasting",
"sequence",
"motif",
"analysis",
"epigenomics",
"research",
"and",
"analysis",
"methods",
"sequence",
"analysis",
"genomics",
"bioinformatics",
"mathematical",
"and",
"statistical",
"techniques",
"biological",
"databases",
"binding",
"analysis",
"eukaryota",
"sequence",
"databases",
"database",
"and",
"informatics",
"methods",
"genetics",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"computational",
"biology",
"amniotes",
"statistical",
"methods",
"organisms"
] |
2018
|
Functional conservation of sequence determinants at rapidly evolving regulatory regions across mammals
|
Independently evolving populations may adapt to similar selection pressures via different genetic changes . The interactions between such changes , such as in a hybrid individual , can inform us about what course adaptation may follow and allow us to determine whether gene flow would be facilitated or hampered following secondary contact . We used Saccharomyces cerevisiae to measure the genetic interactions between first-step mutations that independently evolved in the same biosynthetic pathway following exposure to the fungicide nystatin . We found that genetic interactions are prevalent and predominantly negative , with the majority of mutations causing lower growth when combined in a double mutant than when alone as a single mutant ( sign epistasis ) . The prevalence of sign epistasis is surprising given the small number of mutations tested and runs counter to expectations for mutations arising in a single biosynthetic pathway in the face of a simple selective pressure . Furthermore , in one third of pairwise interactions , the double mutant grew less well than either single mutant ( reciprocal sign epistasis ) . The observation of reciprocal sign epistasis among these first adaptive mutations arising in the same genetic background indicates that partial postzygotic reproductive isolation could evolve rapidly between populations under similar selective pressures , even with only a single genetic change in each . The nature of the epistatic relationships was sensitive , however , to the level of drug stress in the assay conditions , as many double mutants became fitter than the single mutants at higher concentrations of nystatin . We discuss the implications of these results both for our understanding of epistatic interactions among beneficial mutations in the same biochemical pathway and for speciation .
The nature of epistasis is critical to broad-scale evolutionary phenomena . If all possible alleles have the same effect in all genetic backgrounds , we might expect populations that diverge initially to converge to a similar genotype and/or phenotype over time at the fitness optimum . In contrast , if some alleles are beneficial only in certain backgrounds , early genetic changes will limit future genetic options , and populations may diverge genotypically and phenotypically . Thus , the shape and “ruggedness” of the fitness landscape is directly determined by the prevalence of sign epistasis [2–4] . The type of epistasis can also shape the rate of adaptation . In the case of positive epistasis , when early mutations increase the beneficial fitness effects of subsequent mutations , adaptive evolution can accelerate over time . In contrast , when epistasis is negative , i . e . , when first-step mutations reduce or oppose the advantage of subsequent mutations , evolution will decelerate . The deceleration of adaptation over time has been previously found in a number of experimental evolution studies [5–8] . Even the formation of new species rests upon epistasis between alleles present in different nascent species . A major driver of postzygotic reproductive isolation between species is the buildup of Bateson–Dobzhansky–Muller ( BDM ) genetic incompatibilities . These incompatibilities represent reciprocal sign epistasis , where alleles that work well together within a species perform poorly when combined with alleles from the other species in a hybrid individual , leading to hybrid inviability or sterility [9] . Reproductive isolation can also arise from non-reciprocal sign epistasis , where the double mutant is less fit than only one parent , reducing gene flow into the more fit parental population . With enough such asymmetric barriers acting in opposite directions , gene flow may cease entirely between populations . All models of speciation agree that sign epistasis , and particularly reciprocal sign epistasis , is important for speciation , but they differ on why species carry different alleles . Among the models of speciation by natural selection , the classic explanation , proposed by Darwin [10] , is that populations diverge into species because they experience different environments and so adapt in ways that often do not work well together . Because of the focus on environmental differences , this explanation has become known as “ecological speciation” [11] . A contrasting hypothesis , known as “mutation-order speciation” [11] , focuses on the chance order in which mutations arise and spread in different populations when facing the same selective environment . Even if the mutational steps that have occurred in each population are independently beneficial , combining mutations across populations need not be . The specifics of the selective environment ( s ) likely have a major influence on the nature of epistasis between beneficial mutations . In environments where adaptation can occur via the elimination of a single biosynthetic pathway , complete loss-of-function mutations at one step in the pathway may lead mutations in downstream genes to become irrelevant to fitness . Indeed , Bateson [12] originally coined the term “epistasis” in 1909 to describe this type of interaction , in which the action of one gene was blocked by that of another , and this is primarily how molecular geneticists continue to define the word [13] . Considering instead partial loss-of-function mutations , genotypes combining multiple mutations may be more fit than single mutants if flow through the biosynthetic pathway is reduced by each additional mutation . In either case , we would expect double mutants to have equal or greater fitness than single mutants if knocking out a pathway is beneficial ( as long as there are no pleiotropic effects beyond the pathway ) , and consequently sign epistasis and reproductive isolation should not arise . On the other hand , if an intermediate phenotype is optimal in a particular environment , mutations that are beneficial on their own may overshoot the optimum when combined , causing a reduction in fitness . In this type of environment , theoretical work predicts that sign epistasis should be particularly frequent between independently selected mutations that have relatively large effects on the phenotype [14] . There is also increasing evidence that epistasis is more often negative for mutations in functionally related genes . In a large-scale screen for genetic interactions in which mutations in most of the ~6 , 000 genes in the yeast S . cerevisiae were tested pairwise in 23 million double mutants ( including mutations in both nonessential and essential genes ) , Costanzo et al . [15] found that combinations of genes involved in the same biological process were enriched for negative interactions . This enrichment suggests , counter to intuition , that strongly negative fitness interactions , of the form that give rise to reproductive incompatibilities , may be more likely to accumulate between populations experiencing the same selective environment compared to those experiencing different environments . To date , few incompatibilities between or within species have been genetically characterized , although recent advances in genomic sequencing technology have greatly aided the discovery of the genetic basis of speciation . For natural populations , the majority of incompatible alleles ( “speciation genes” ) that have been characterized are found between species adapted to different local environments , presumably representing cases of ecological selection ( documented in [16]; see their Tables S1 and S2 ) . For example , the buildup of a suite of plant-specific traits has allowed one species of Drosophila to utilize a different , normally toxic , host plant [17] , and selection on soils of different salinity has caused the accumulation of quantitative trait loci associated with salt tolerance in a hybrid species of Helianthus sunflowers beyond what is found in its parental species [18] . In other cases , genetic incompatibilities between natural populations have been identified for which there is no clear connection to the external selective environment , including BDMs caused by the reciprocal silencing of alternative duplicate gene copies [19] or the differential accumulation of selfish genes and suppressors ( see examples in [20] ) . The exact history of selection is unknown in natural populations , thus it is difficult to know whether these cases represent mutation-order or ecological selection . Natural populations of yeast also show environment-specific genetic incompatibility ( including one characterized two-locus BDM [21] ) , although , as in other taxa , we have no knowledge of the evolutionary history that led to these interactions . Experimental evolution studies allow direct control over the form of environmental selection , and sign epistasis has been found in some studies that combined mutations from populations adapted to both different and similar selective environments . Dettman et al . [22] evolved different populations of Neurospora crassa to high salinity and low temperature . When the evolved strains were mated , lineages adapted to different environments exhibited reduced reproductive success relative to matings between lineages adapted to the same environment , and this reduction was consistent with the action of BDM incompatibilities . A parallel study that examined populations of S . cerevisiae evolved to high salinity and low glucose for 500 generations found very similar results [23] . Follow-up work identified a BDM incompatibility between an allele of PMA1 ( a proton efflux pump ) that arose under high salt adaptation and an allele of MKT1 ( a global regulator of mRNAs encoding mitochondrial proteins ) that evolved in low glucose [24] . This was the first reported BDM interaction among known genes isolated from experimentally evolved strains , to our knowledge . Sign epistasis has also been documented when combining mutations between experimentally evolved populations adapting to the same environment . Kvitek et al . [25] investigated populations of asexually propagated haploid S . cerevisiae evolved under glucose limitation in continuous culture for 448 generations [26] . Mutations in two genes , MTH1 and HXT6/HXT7 , appeared several times in independent lineages during the experiment but never together . These mutations were shown to be individually beneficial , but they had lower competitive fitness when combined in a double mutant than either single mutant or the ancestor , showing reciprocal sign epistasis [25] . Negative epistasis was also prevalent among five additional strains constructed to bear two adaptive mutations that arose in different lineages , with significant negative epistasis in four out of the five comparisons , including one example of sign epistasis [25] . Chou et al . [27] similarly investigated epistasis using an engineered strain of Methylobacterium extorquens with a modified central metabolism that was dependent on a foreign pathway artificially introduced on a plasmid . These bacteria were evolved for 900 generations under conditions that utilized this pathway . All adaptive mutations decreased expression of the introduced pathway . Combining mutations , the authors found that expression levels were well predicted by the independent effects of each mutation but that expression mapped nonlinearly onto fitness , leading to sign epistasis in many cases . Collectively , these experiments demonstrate that BDMs can arise rapidly in experimental evolution studies when populations experience either different or similar selective pressures , providing support for both ecological and mutation-order speciation . What remains unknown from long-term experiments of populations evolved under the same selective pressure is how frequently early adaptive mutations could contribute to reproductive isolation . This raises the question of whether mutation-order speciation occurs because of incompatibilities among mutations that would be beneficial in either population or because the fixation of different initial mutations alters the subsequent selective environment experienced in different populations ( i . e . , divergent selection due to differences in genetic background ) . We investigate , for the first time , fitness interactions among all pairwise combinations of genes bearing first-step adaptive mutations to a common selective environment . Specifically , we measured epistasis between beneficial mutations acquired in the yeast S . cerevisiae grown in the presence of the fungicide nystatin [28] . Briefly , Gerstein et al . [28] isolated 35 first-step mutations in 4 μM nystatin , performed genome-wide sequencing , and found that all strains carried a single mutation in one of four genes in the ergosterol biosynthesis pathway ( Fig 2; genomic analysis revealed either no or only one other mutation present in the strains used herein , details below ) . We focused on one mutation in each gene and investigated the fitnesses of all six pairwise double mutants between these four mutations . For two of these genes ( ERG6 [SGD ID: S000004467] and ERG3 [SGD ID: S000004046] ) , many of the mutations found by Gerstein et al . [28] were consistent with a complete loss of function ( e . g . , early stop codons , similar sterol phenotype to the whole gene knockout ) . The mutations occurring in the most upstream ( ERG7 [SGD ID: S000001114] ) and downstream ( ERG5 [SGD ID: S000004617] ) genes in the pathway , however , were not [28] . The erg7 mutation is a nonsynonymous change close to the end of the gene , and deletion of ERG7 is inviable . The erg5 mutation is an in-frame deletion and is unlikely to be a null mutation because the full gene deletion is respiratory deficient [30] , which is not observed for this mutant ( BMN35 in [28] ) . Thus , we also assessed whether upstream mutations in the biosynthetic pathway generally mask the effects of downstream mutations or if masking is limited to complete loss-of-function mutations . Overall , we found that strong negative epistasis , of the type that causes some degree of reproductive isolation between strains fixed for different mutations , was surprisingly common among these first-step mutations . Indeed , the interactions were so negative that they reversed the direction of effect in over half of the double mutants , causing beneficial mutations to become deleterious when in combination and double mutants to be less fit than at least one of the two single mutants ( sign epistasis; Fig 1 ) . Furthermore , in one third of the comparisons , the double mutants were less fit than both single mutants ( reciprocal sign epistasis ) . We assayed mutational effects in both haploid and diploid backgrounds , finding similar results regardless of ploidy , indicating that these epistatic relationships are likely to hold across stages of the yeast life cycle . Epistatic relationships for fitness were not well predicted by sterol profiles or pathway position of the mutants , however , suggesting that selection does not simply act via flux through the pathway to ergosterol . Finally , we investigated epistasis in different concentrations of nystatin to determine how epistatic relationships , and therefore reproductive isolation , might change under different levels of environmental stress . Previous work with antibiotic resistance in bacteria has shown that the shape of fitness landscapes can be strongly dependent on antibiotic concentrations [31] . Interestingly , we found that the negative interactions observed between beneficial mutations at lower concentrations of nystatin reversed sign and became increasingly positive at higher concentrations of nystatin . Indeed , only the double mutants exhibited substantial growth in the higher concentrations of nystatin tested . Thus , although combining single-step mutations generally reduced fitness in the historical nystatin environment , these same combinations were more likely than the individual mutations to allow colonization of even harsher environments .
We characterized the epistatic interactions between pairs of mutations that act in the ergosterol biosynthesis pathway and individually confer increased fitness when exposed to the antifungal drug nystatin . Maximum growth rate of ancestral , single mutant , and double mutant genotypes was characterized in haploid strains of both mating types in a rich medium composed of yeast extract , peptone , and dextrose ( YPD ) + 2 μM nystatin ( “nystatin2” ) . Outlier data points were detected statistically and removed from further analyses , although we note where inclusion of outliers would have affected the results ( for further details , see the section “Outlier Detection and Removal” ) . The effect of mating type ( and its associated auxotrophy ) was not significant ( p = 0 . 19 ) , and the data for the two haploid mating types will be considered together , except where noted ( see Dryad file for additional statistical methods and results [32] ) . Using a mixed-effects model , all main effects of individual mutations were positive , confirming that the mutations improved growth in nystatin ( Table 1 ) . To assess epistasis , least-squares means of maximum growth rates were inferred from the model and compared between double and single mutants and between single mutants and ancestral strains , correcting for multiple comparisons . Double mutants were never significantly more fit than the best of the single mutants ( top right panels in Fig 3 ) , and all pairwise interactions exhibited significant negative ( antagonistic ) epistasis ( Table 1 ) . The double mutant was significantly less fit than the fittest single mutant in four cases ( “sign epistasis”: erg3 erg5 , erg3 erg6 , erg3 erg7 , and erg6 erg7 ) and significantly less fit than both single mutants in two cases ( “reciprocal sign epistasis”: erg3 erg6 and erg6 erg7 , Table 1 , Fig 3 ) . The results are similar when fitness is measured by optical density ( OD ) after 24 hours of growth instead of maximum growth rate over 24 hours ( S1 Fig ) . The strong negative interactions indicate that these alleles , when combined , confer genetic incompatibilities between the strains . We characterized epistatic interactions of maximum growth rate for the homozygous diploid strains in nystatin2 and compared them to the haploid results to determine whether the interactions were ploidy dependent . As in haploids , single mutations generally improved the growth of diploid homozygotes in nystatin2 , although the erg5 mutation did not do so significantly in a pairwise comparison with the ancestral strain ( Fig 4 ) . Qualitatively , epistatic interactions were also similar to the haploids ( Table 2 , Fig 4 ) whether fitness was measured by maximum growth rate or OD after 24 hours of growth ( S2 Fig ) . When we categorized the type of epistasis statistically for maximum growth rate , most interactions were of the same type ( sign epistasis: erg3 erg5; reciprocal sign epistasis: erg3 erg6 and erg6 erg7; negative epistasis: erg5 erg7 ) . There were , however , several quantitative differences . The erg6 erg7 double mutant was so unfit in diploids that we were often not able to standardize it properly in the growth assays ( low growth , as measured by OD , was observed in all concentrations of nystatin tested , S3 Fig ) . Furthermore , in two cases , epistasis was qualitatively similar , but the differences were no longer statistically significant ( sign epistasis: erg3 erg7; negative epistasis: erg5 erg6 ) . To visualize the full diploid fitness landscape , we repeated the analysis including all heterozygous strains ( open symbols in Fig 4 , pairwise comparisons in S4 Fig ) . Low F1 hybrid fitness was typical; double heterozygous strains ( open diamonds ) were uniformly low in fitness when compared to the homozygous single mutants ( not significantly so when compared with the weak erg5/erg5 mutant ) . Mutations were generally partially to fully recessive and did not have a large effect on fitness when comparing heterozygotes to wild type at a gene , either when the other gene was wild type ( open triangles ) or homozygous mutant ( open circles ) . To determine the extent to which epistasis reflected gross fitness defects not specific to nystatin resistance , we repeated the analysis on maximum growth rate in YPD , a rich growth medium . As in nystatin2 , mating type ( and its associated auxotrophy ) had no significant effect ( p = 0 . 98 ) , and results were averaged over mating types . The single mutations were generally deleterious in YPD ( note the negative coefficients for the individual mutations , Tables 3 and 4 ) , consistent with previous characterization of these mutations [28] . The exception is the haploid erg5 mutant , which is not significantly less fit than the ancestor in a pairwise comparison of maximum growth rates ( bottom left panels in Figs 3 and 4 ) . As observed in nystatin2 , the double mutant often had lower fitness than the single mutants in YPD , although the strength of epistasis was generally weak ( most interactions resemble a parallelogram , Figs 3 and 4 ) . Significant sign epistasis was only observed in a single diploid case ( erg3 erg7 ) . Epistatic interactions in YPD were qualitatively different from those observed in nystatin2 and often differed between haploids and diploids ( Tables 3 and 4 ) . In contrast to the prevalence of negative epistasis in nystatin2 , significant positive epistasis was observed in some cases ( the double mutant was more fit than expected under the additive model ) . The low growth in YPD of most double mutant strains suggests that the negative relationships observed in nystatin2 may , in part , be due to intrinsic growth problems , perhaps due to the instability of the cell membrane without proper ergosterol synthesis . To assess whether the genetic interactions depended on the concentration of drug , growth was measured as OD after 24 hours over a range of nystatin concentrations ( 0 , 1 , 2 , 4 , 8 , 16 , 32 , 64 , 128 , 256 μM ) . We focused here on OD to assess the range of environments in which the yeast strain could grow , even if slowly , and because of the massive replication required . Although OD is thought to reflect the efficiency of cells’ ability to turn nutrients into cellular material rather than the rate of growth , OD and maximum growth rate were correlated for the single mutants analyzed here [28] , and the interactions observed were qualitatively similar for the concentrations of nystatin used in both the maximum growth rate and OD assays ( 0 μM and 2 μM ) . As before , mating type was not found to have a significant effect on OD in the haploid data ( linear model that included mating type , concentration of nystatin , and strain identity as fixed effects; mating type: F = 0 . 23 , df = 1 , p = 0 . 63; concentration of nystatin: F = 600 . 12 , df = 1 , p < 10−15; strain: F = 31 . 95 , df = 10 , p < 10−15 ) , and data were pooled across mating types . We found that the form of gene interactions changed when measured over a range of concentrations of nystatin ( haploid results: Fig 5 ) . As observed previously , the double mutant generally had equivalent or lower growth than the two parent mutants at low concentrations of nystatin ( 0–4 μM ) , but at high concentrations ( 32–64 μM ) , the double mutant strains became the only strains able to grow well . That is , a preponderance of negative epistasis shifted towards a preponderance of positive epistasis as nystatin concentrations rose . This dependence of the sign of epistasis on the concentration of the drug ( not only on the presence or absence of the drug ) indicates that the outcome of mutation or hybridization will depend heavily on the specifics of the environment in which the yeast is found . Homozygous diploid strains showed qualitatively similar patterns of growth to the haploid strains , with the exception of the erg6/erg6 erg7/erg7 double mutant ( S3 Fig ) . When we compared all diploid strains ( including heterozygous strains ) , some interesting patterns emerge ( S5 Fig ) . In many cases , the double heterozygous strain exhibited more growth than either single heterozygous strain ( as observed by a “bump” in the middle of the figure ) , particularly at higher concentrations of nystatin . This may indicate a net beneficial effect of carrying two heterozygous mutations or may reflect an increased potential for loss of heterozygosity ( LOH ) . LOH , in which a locus that is initially heterozygous for a mutant allele becomes homozygous , would be beneficial in our fitness assay because being homozygous for either mutant allele increases growth in nystatin ( compare middle point in S5 Fig to those second from either end ) . This may have occurred during the course of the fitness assay , affecting our final measures of fitness . LOH was previously observed for the single heterozygous mutants over a 72-hour timescale [33] , and being heterozygous for two mutations may increase the chance of LOH for at least one of the two . The unexpected increase in fitness in the double heterozygotes may also be indicative of an epistatic interaction providing some benefit to having two heterozygous mutations within the ergosterol pathway compared to full recessivity ( i . e . , no benefit ) with only a single heterozygous mutation [33] . To determine whether epistasis for fitness was consistent with the sterol phenotypes exhibited by the strains , we extracted and measured the sterol profile of all MATa strains . In ancestral samples , we see the characteristic four-peaked curve between 240 and 300 nm that is produced by ergosterol and the late sterol intermediate 24 ( 28 ) dehydroergosterol [34] . Only the latter sterol shows an absorption band at 230 nm , allowing quantification of ergosterol , but we found the peak between 200 and 230 nm to be very sensitive to the standard used ( e . g . , newly mixed heptane and ethanol versus heptane layer from extraction performed with no yeast cells and ethanol ) and thus limit ourselves to a qualitative description of the results . All of our single mutants show similar results to those presented by Gerstein et al . [28] for these same mutants ( Fig 6 ) . The two potential loss-of-function mutants ( erg3 and erg6 ) also have similar sterol profiles to knockout mutants of these genes [35 , 36] . Double mutants show a variety of profiles , as can be seen in Fig 6 . Notably , most double mutants resemble one of the two parent single mutants , with the exception of the erg6 erg7 double mutant , which is intermediate between the two single mutants in absorbance over much of the measured range ( suggesting a mixture of sterols present ) . All double mutants that include the mutation in ERG3 tend to show similar profiles to the erg3 single mutant . Thus , the sterol profiles were not predicted by gene position in the ergosterol biosynthesis pathway ( as ERG6 is upstream of ERG3 ) . Furthermore , the similarity in sterol profiles between double and single mutants did not generally predict the patterns observed for maximum growth rate ( with the exception of the erg5 erg7 haploid and diploid , which behaved like erg7 , and the erg3 erg7 diploid , which behaved like erg3 ) , indicative of a disconnect between sterol profile and fitness .
We find that the interactions were so negative that the double mutant grew less well than at least one of the parent single mutants ( sign epistasis ) in four of the six gene combinations assayed in haploids . In half of these cases , the double mutant grew significantly less well than both single mutants ( reciprocal sign epistasis ) . Similar interactions were observed in diploids ( three cases of sign epistasis , two of which were reciprocal ) . The observation of reciprocal sign epistasis is of particular interest , as this type of BDM incompatibility underlies postzygotic reproductive isolation among speciating lineages . The high frequency of reciprocal sign epistasis observed , even among first-step beneficial mutations acquired in the same environment , confirms the possibility that isolated populations experiencing similar selective pressures can diverge and eventually speciate simply through the order of mutations that happen to arise and fix ( mutation-order speciation ) . The prevalence of sign epistasis among our specific set of beneficial mutations is somewhat surprising given the linearity of the biosynthetic pathway in which all of the affected ergosterol genes act ( Fig 2 ) . Our results were not consistent with the expectation that the phenotype and fitness of double mutants would be determined by the upstream mutation . In terms of phenotype , the sterol profile of the double mutant was similar to that of the most upstream mutant in only two cases ( the erg5 erg7 and erg3 erg5 double mutants , Fig 6 ) . In terms of fitness , the growth rate of the double mutant differed significantly from that of the most upstream single mutant in three ( haploids ) and four ( diploids ) out of six pairwise comparisons ( Figs 3 and 4 ) . In the combination of two loss-of-function–type mutations ( erg3 erg6 ) , neither sterol phenotype nor fitness matches that of the upstream ( erg6 ) mutation . These results indicate that there remain substantial interactions between the mutations in the ergosterol pathway potentially due to partial activity of the upstream genes creating low levels of substrate for the remainder of the pathway , due to downstream genes acting on alternative sterol substrates , or due to interactions among the intermediate sterols themselves . From previous work in the yeasts S . cerevisiae and Candida albicans , it has been shown that ERG6 plays a role in offshoot sterol synthesis in mutants of ERG3 [38] , and it is known that intermediate sterols are found at different levels in different compartments of the cell [39] and may impact fitness in a variety of ways ( e . g . , altering temperature tolerance [40] and virulence [41] ) . There was also no clear relationship between sterol phenotype and fitness in these strains . Sterol phenotype for most double mutants resembles one of the two single mutants ( Fig 6 ) , but this similarity in sterol phenotype did not generally predict maximum growth rate in nystatin2 ( with the possible exception of erg5 erg7 in haploids and diploids and erg3 erg7 in diploids , Figs 3 and 4 ) . Future analyses that determine the processing of sterols in the single and double mutants , as well as their pleiotropic effects , would further elucidate these genetic interactions . Interestingly , the type of epistasis depended strongly on the concentration of nystatin . At lower concentrations of nystatin , similar to those used to acquire the mutations ( ≤ 4 μM nystatin ) , epistatic interactions were typically negative ( Fig 5 ) , with the double mutant showing similar or lower densities after 24 hours of growth than the single mutants . By contrast , at higher concentrations of nystatin , the interactions were often positive , with double mutants typically able to outgrow both single mutants . Emblematic of this phenomenon , the best growing haploid double mutant strains at 32 μM nystatin ( erg3 erg6 , erg3 erg7 , erg6 erg7 ) were also those that exhibited the most negative epistasis at lower concentrations . This implies a trade-off between growth at low versus high concentrations of the fungicide . Conceptually , this trade-off suggests that the double mutant initially overshoots the optimum when nystatin concentrations are low , because the costs associated with each ergosterol mutation are combined ( perhaps destabilizing the plasma membrane ) ; by contrast , when nystatin concentrations are high , the optimum is shifted even farther away , and extreme reductions in ergosterol and potentially other sterols are needed for the yeast to survive , at which point the double mutant is most fit ( see , e . g . , Blanquart et al . [42] for a theoretical exploration of this phenomenon ) . Because membrane damage can trigger cell cycle arrest in yeast [43] , another possible explanation for the results observed at high concentrations of nystatin is that single mutants experience cell cycle arrest , reducing growth rate , whereas the additional stress caused by the combination of two mutations and high concentrations of nystatin may cause a checkpoint failure in double mutants , allowing the cells to bypass arrest and continue dividing ( personal communication , C . Nislow to J . Ono ) . The shifting nature of epistasis as a function of the severity of the environment also has implications for speciation and has not been widely discussed ( but see [44] for discussion about environment-dependent epistasis and [45] for an example of an environment-dependent negative epistatic interaction on feeding and growth performance in F2 hybrid stickleback ) . Our results show that BDMs can be environment-specific , and thus gene flow between species might vary according to the environment in which secondary contact occurs [46] . Counterintuitively , our results further suggest that harsher environments may be more conducive to gene flow because of the possible benefit of combining adaptive mutations from different populations . Indeed , environments that are so harsh that only strains combining mutations survive ( as we observed at high concentrations of nystatin ) might promote hybridization and potentially lead to hybrid speciation ( reviewed in [47] ) . For example , extreme desert environments have selected for combinations of traits that improve drought tolerance , allowing hybrid Helianthus sunflowers to colonize and proliferate [48] . Because cell volume to surface area ratios are different for haploids and diploids [49] , we might expect differences in growth and epistasis between haploids and diploids , particularly in the face of a selective pressure like nystatin that impacts the cell membrane . By and large , however , our results were consistent across ploidy levels , with diploid homozygous mutants and haploid mutants showing similar patterns of epistasis . One exception was the erg6/erg6 erg7/erg7 double mutant , which was so unfit in diploids that yields were often too low to obtain initial cell densities similar to other strains in our growth rate assays , even when given multiple extra days of growth . The haploid version of this same double mutant , however , also showed very low fitness . During the initial isolation of the haploids from spores , the double mutant colonies were identifiable by their noticeably smaller size compared to those produced by single mutant and ancestral genotypes . A haploid double mutant strain also exhibited reversion in one instance during growth in 10 mL YPD ( Sanger sequencing revealed a secondary mutation in the same codon as the original mutation , reverting the amino acid ) . Considering the various diploid heterozygotes , we confirmed that the ergosterol mutations were largely recessive , as found previously for the single heterozygous mutant strains [33] . There were more signs of nystatin resistance in the double mutant strains than in the single mutants , however . One indication of this was the double heterozygous strains showing a slight increase in biomass produced ( as measured by OD ) compared to the single heterozygous strains across a range of concentrations of nystatin ( S5 Fig ) . Despite this , the double heterozygous strains were uniformly of low growth rate in nystatin2 ( open diamonds in Fig 4 ) , with similar sensitivity as found in the ancestor . The generally poor performance of the double heterozygous diploid is of particular interest because this genotype would be the first hybrid product of crosses between strains fixed for different beneficial mutations ( see also S4 Fig ) . Thus , F1 hybrid inviability in the double heterozygotes , as well as reciprocal sign epistasis , contributes to reproductive isolation between these strains . Overall , we find that the very earliest stages of divergence within a common selective environment can generate postzygotic reproductive isolation , observing sign epistasis , reciprocal sign epistasis , and F1 hybrid inviability in double heterozygotes among the first-step adaptive mutations isolated in the presence of nystatin . Although we did not assay incompatibilities at other stages ( e . g . , meiotic incompatibilities ) , we expect that further BDMs might be revealed by analyzing other stages in the life cycle ( indeed , it was very difficult to sporulate some double mutant strains , particularly erg5/ERG5 erg7/ERG7 ) . We speculate that genetic incompatibilities may be especially likely in scenarios such as the one investigated here , in which selection favors large effect mutations . In the initial experiment in which mutations were acquired , the concentration of nystatin was chosen to inhibit growth , so that only mutations capable of rescuing fitness were isolated [28] . Such large effect mutations might have more costly pleiotropic effects and/or be more likely to overshoot the fitness optimum when combined , showing negative epistasis for fitness even if their effects are multiplicative or additive on the underlying trait . If large effect mutations are more likely to interact negatively , which is consistent with our results and others [5–8] , short periods of severe selection might be more likely to lead to speciation than longer periods of mild selection . Future experiments comparing genetic incompatibilities among strains with similar levels of divergence but consisting of a few large effect or several small effect genetic differences would be extremely valuable . We also speculate that independent populations experiencing directional selection to the same environmental change might be more likely to speciate than those experiencing directional selection to different environments because the beneficial mutations that accumulate in the former case may be more likely to involve similar pathways and thus be more likely to interact negatively ( as has been shown in interaction studies , see [15 , 50] ) . Indeed , even though the beneficial mutations that we assayed were all in the same “linear” pathway and acquired in the same selective environment , we found that the type of epistasis that underlies speciation was common , providing experimental support for the mutation-order speciation hypothesis .
We assayed all pairwise interactions in both haploids and diploids between four beneficial mutations acquired in the fungicide nystatin , one in each of ERG3 , ERG5 , ERG6 , and ERG7 ( Table 5 ) . Each mutation was initially isolated in the BY4741 haploid background ( MATa his3Δ1 leu2Δ0 met15Δ0 ura3Δ0 , derived from S288C ) and struck down to a single colony to remove standing variation . Mutations were detected by whole genome sequencing on an Illumina HighSeq 2000 , followed by alignments to the S288C reference strain [28]; few other mutations were detected besides those in the ERG genes . For the strains used here , only the strain containing the mutation in ERG7 also carried a secondary mutation ( in DSC2 [SGD ID: S000005434] ) , the presence or absence of which did not substantially alter the presented results ( see details in S1 File ) . For a complete description of the isolation of these initial strains , see [28] . All possible haploid and diploid genotypes for each pair of ERG genes were created via mating and sporulation . A brief overview of strain construction will be given here , but for a detailed description , see S1 File . To create singly heterozygous strains , each original single mutant strain was mated to BY4739 ( MATα leu2Δ0 lys2Δ0 ura3Δ0; Open Biosystems ) , which is isogenic with BY4741 except for the auxotrophies . MATα single mutant strains were isolated by sporulation of the heterozygous diploids followed by dissection and testing of the resulting tetrads . Throughout strain construction , histidine and lysine auxotrophies were consistently kept with the same mating types so that all haploid strains were either MATa his3Δ1 or MATα lys2Δ0 . Plates lacking methionine did not efficiently select against the met15Δ0 mutation carried by the original single mutant strains , suggesting a weak effect of this mutation , and the methionine auxotrophy was not tracked . The MATα single mutant strains were then mated to the original MATa single mutant strains to create strains that were either homozygous for one mutation or heterozygous for two mutations . The haploid double mutant strains were created through sporulation and dissection of the doubly heterozygous strains . All haploid double mutant strains were confirmed by Sanger sequencing . We failed to obtain the MATa erg5 erg6 double mutant haploid strain through crossing and sporulation because the two genes are linked ( 48 kb away , flanking the centromere of chr XIII ) . For this strain , a transformation was performed by electroporation using a protocol based on [55] to insert the mutation within ERG6 into the MATa erg5 genetic background; this mutation was then checked by Sanger sequencing . Strains with one heterozygous and one homozygous mutant locus as well as double homozygous mutant strains were created by mating the MATa single mutant and double mutant strains to the MATα double mutant strains . A diploid ancestral strain was created by mating BY4741 and BY4739 . We conducted a set of growth rate ( fitness ) assays under nystatin stress and in rich medium ( YPD ) . The experimental design sought to ensure that data were gathered for each combination of wild-type and mutant strains across batches performed on different days . Specifically , within a batch , for a given pair of mutations in haploids and for each mating type , each ancestral strain and each single mutant was assayed twice , while each double mutant was assayed four times ( the double mutant was assayed more often because it was the only genotype unique to that pair of mutations ) . For each pair of mutations in diploids , all possible combinations of the two genes in both heterozygous and homozygous forms ( including the nonmutant ) were present twice within a batch . We measured growth in YPD and YPD + 2 μM nystatin ( “nystatin2” ) using the Bioscreen C Microbiological Workstation ( Thermo Labsystems ) , which measures OD in 100-well honeycomb plates . Nystatin2 was used to assay fitness , because previous studies with these mutants found that 2 μM nystatin inhibits the growth of the ancestral strains while also allowing the growth of all mutant strains [56] . OD was measured automatically using the wideband filter at 30-min intervals for 24 hours from cultures growing at 30°C with maximum continuous shaking . Longer assays were avoided because mutations and LOH events began to accumulate [33] . The maximum growth rate over 24 hours was determined by the spline with the highest slope from a loess fit through natural log transformed OD data , using a custom script written by Richard FitzJohn in R [57] ( see Dryad for code [32] ) . For complete details on how strains were initially grown from frozen and standardized ( “pre-assays” ) before measuring growth ( “assays” ) , see S1 File . Briefly , each yeast replicate was grown from frozen in YPD + 0 . 5 μM nystatin in 100-well honeycomb plates for 72 hours in the pre-assays unless very poor growth of the strain required otherwise , and OD was then determined . YPD + 0 . 5 μM nystatin was used to help prevent reversion of strains with severe growth defects in YPD and was not found to affect subsequent measures of growth compared to a pre-assay in YPD ( the first pre-assay was conducted in YPD , see details in S1 File ) . For the main assays , honeycomb plate wells were filled with 148 . 5 μL of YPD or nystatin2 . The yeast was then transferred from the pre-assay plates into one well each of YPD and of nystatin2 , with the volume transferred determined by the maximum pre-assay OD reading ( the minimum volume transferred was 1 . 5 μL , while the maximum was 7 . 5 μL ) . Note that these transfers decreased the concentration of nystatin in the individual wells but never by more than 0 . 1 μM . Strains were randomized within plates using the same map for the pre-assays and assays in a given batch . There were not equivalent numbers of replicates for all strains after omitting some data due to low growth ( if the volume to be transferred to the assay plate exceeded 7 . 5 μL ) , lack of growth , mechanical error , or because some strains had to be re-run ( for details , see S1 Table ) . Nevertheless , at least two replicates per day on at least 2 days were measured for all strains in each medium ( with the exception of erg5/erg5 erg6/erg6 , for which 14 replicates were all run on a single day , S1 Table; for exact numbers and days on which the replicates were run , see Dryad [32] ) . Although the different numbers of replicates led some crosses to have less power than others , the cross with the least amount of data ( erg6 by erg7 ) was also the one in which the double mutant was particularly unfit , which contributed to the difficulties in assaying fitness but also meant that epistasis was readily detected . In all cases , data for each double mutant was collected simultaneously with data on the ancestor and single mutants , allowing day effects to be factored out in the analysis . Growth at different concentrations of nystatin was assessed following similar procedures to the growth rate assays . To prepare the strains for tolerance assays , pre-assays were again conducted to standardize initial cell concentrations . Stocks were first grown from frozen in four 96-well plates filled with 198 μL of YPD + 0 . 5 μM nystatin and inoculated with 2 μL of frozen culture . Strains were distributed among the four plates so that there was one replicate of the entire balanced design per plate , randomized within plate . In order to fit all strains on a single plate , some strains were excluded ( MATa erg5 erg6 and MATa erg3 erg5 ) . These strains were chosen because initial assays indicated that these double mutants most closely resembled the stronger ( non-erg5 ) single mutant . The plates were covered with aluminum lids and incubated at 30°C with continuous shaking at 200 rpm in a container with wet paper towels to minimize evaporative water loss . Prior to removal of the aluminum lid , plates were always spun for 1 min at 3 , 700 rpm to ensure that all liquid was collected at the bottom . After 72 hours , all wells were manually mixed and OD was measured on a BioTek plate reader at 630 nm . The well with the minimum OD value among the four pre-assay plates was identified and used to calculate the amount of YPD to add to each pre-assay well to standardize cell density across cultures . Wells containing only medium , those containing erg6/erg6 erg7/erg7 ( see below ) , and one well that appeared not to have been inoculated were excluded from standardization . Two μL from each well was used to inoculate the assay plates . Assay plates were prepared with 198 μL of YPD + 0 , 1 , 2 , 4 , 8 , 16 , 32 , 64 , 128 , and 256 μM nystatin , with four plates per concentration . The assay plates were covered with aluminum lids and incubated at 30°C in containers with wet paper towels , shaking at 150 rpm . Exceptions to the pre-assay protocol had to be made for strains with slower growth . Ten ml of 0 . 5 μM nystatin was inoculated with 15 μL of erg6/erg6 erg7/erg7 from frozen 2 days before all other strains were inoculated , allowing additional growth time for this unfit strain . On the day that all other strains were inoculated from frozen , the erg6/erg6 erg7/erg7 culture was concentrated into ~900 μL ( although growth was not observable ) , and 200 μL of this culture was used to replace the medium from the appropriate wells in the pre-assay plates . In addition , erg6 erg7 ( both MATa and MATα ) and erg6/erg6 were inoculated with 2 . 67 μL of frozen culture ( as opposed to the 2 μL used for all other strains ) to compensate for their lower growth rate from frozen . Twenty-four hours after inoculation , the aluminum lids were removed , wells were manually mixed , and the OD of each assay plate was read on a BioTek plate reader at 630 nm . Some wells had lost volume due to cracks that had developed in the plates , and these wells were omitted from analysis . Prior to analysis , the OD of the medium itself was subtracted from the final OD measurements . To determine whether the sterol profiles of the single mutants , along with their position within the ergosterol pathway , predict the sterol profiles of the double mutants and whether differences in sterol profiles predict differences in fitness , a spectrophotometry-based assay was used to compare the sterol profiles of the ancestral , mutant , and double mutant MATa strains . Sterols were extracted using the alcoholic potassium hydroxide method [34] , as previously performed on the single mutant strains [28] . MATa strains were struck from frozen onto YPD plates and grown for 65 hours . Three colonies for each strain were inoculated into two separate tubes filled with 10 mL of YPD ( total of 20 mL per replicate ) and incubated at 30°C on a rotor for 48 hours . After growth , cells were harvested by centrifugation at 2 , 700 rpm for 5 min , combining culture from the two tubes by performing two successive spins . The pellets were washed twice with sterile distilled water and 1 . 2 mL of 25% alcoholic potassium hydroxide was added to each . The tubes were vortexed for 1 min then incubated in an 80°C water bath for 1 hour . After cooling the samples to room temperature , 0 . 4 mL of sterile distilled water and 1 . 2 mL of n-heptane were added to each sample , and the tubes were vortexed for 3 min . Samples were collected by taking 220 μL of the heptane layer and adding it to 880 μL of 95% ethanol in a 1 . 5-mL tube . These tubes were stored at −20°C for 2 days before reading the absorbance every 3 nm between 200 and 300 nm in a quartz microcuvette using a Thermo BioMate 3 spectrophotometer . Due to a posteriori observations that different heptane/ethanol mixtures led to different peak heights near 220 nm , we chose to use one replicate of the erg6 erg7 strain that showed no evidence of growth ( suggesting an inoculation failure ) , but was otherwise identically treated , as a control for standardization . As a result , only two replicates of erg6 erg7 are presented . Outliers in microbial fitness assays often represent either contamination by a different strain or evolution over the course of the fitness assay . In order to prevent these events from having undue influence on our analyses , we detected outliers for maximum growth rate after omitting some wells due to lack of growth and mechanical error ( see details in S1 File ) . For outlier detection , we first normalized for plate within each day . We did so by finding the global mean maximum growth rate for all ancestral strains over all days and calculating the difference between this and the mean of all ancestral strains on a given plate , yielding a plate correction value . This correction value was added to the maximum growth rate for each strain from the corresponding plate . Outliers were detected by performing a two-sided Grubbs test , allowing us to detect a maximum of one outlier per strain and medium , using the R package outliers and the method grubbs . test [57 , 58] . A total of eight replicates in nystatin2 and six replicates in YPD were marked as outliers and removed from all presented statistical and graphical analyses . All qualitative relationships between strains and the main statistical conclusions were insensitive to the exclusion or inclusion of the identified outliers , with two main exceptions for the haploids in nystatin2 ( see S6 and S7 Figs for versions of Figs 3 and 4 that include all outliers ) . These exceptions are noted in the Results and described in detail in S1 File . Epistasis for maximum growth rate was assessed with mixed-effects models run on either all haploid or all diploid strains together , including the genotype at each gene , their pairwise interactions , and mating type ( for the haploids ) as fixed effects and plate within day as a random effect , fit using restricted maximum likelihood with the lmer function from the lme4 package in R [57 , 59] . For diploids , the models were first run using only strains that were homozygous ( either mutant or ancestral ) for comparison to the haploid data . Significance of interaction terms ( and mating type ) was determined by performing an ANOVA between the full model and a model dropping that term using the anova function in R and fitting models using maximum likelihood . To determine the type of epistasis present for each pair of genes , the package lsmeans [60] was used to both determine the least-squares mean for each strain in the model and to make comparisons between strains using the contrast function . The type of epistasis was determined by comparing the double mutant to each single mutant and each single mutant to the ancestor , and only these planned comparisons were performed . The p-value was adjusted for the number of tests performed using the multivariate t distribution ( mvt method ) in lsmeans . To be conservative , we based our categorization of epistasis solely on statistically significant differences . For example , if the double mutant had a lower growth rate than both single mutants but this difference was only significant in one of the two cases , it was considered an example of sign epistasis ( significantly lower than one single mutant but not the other ) rather than reciprocal sign epistasis . A similar procedure was then undertaken including heterozygous diploid strains . A model was run using the lmer function including all diploid strains together , with plate within day as a random effect . Least-squares means were determined for all diploid genotypes from this model , and comparisons were performed between each diploid genotype and all other diploid genotypes that were one mutational step away . The double heterozygous strains were compared to all other strains for that pair of genes because the potential progeny of the double heterozygote includes all possible genotypes and these comparisons are therefore of biological interest . For the tolerance assay assessed across a range of concentrations of nystatin , we performed Welch’s t-tests of OD after 24 hours between each double mutant and its single mutant parents ( day effects were not estimated as all measurements were gathered on the same day ) . Because we were focused on the changing nature of epistasis , rather than any particular pairwise comparison , a correction for multiple comparisons was not performed . Data and analyses deposited in the Dryad repository: http://dx . doi . org/10 . 5061/dryad . vs370 [32] .
|
We crossed yeast bearing different genetic mutations to determine the fitness of their hybrid offspring . These strains had previously evolved in the presence of the fungicide nystatin . Even though the initial strains had nearly identical genomes , differing only in the mutation they carried within the biosynthetic pathway leading to ergosterol , the hybrid offspring were less fit than expected based on parental fitness . These negative interactions were so strong that beneficial mutations often became deleterious in the presence of one another ( sign epistasis ) . In one third of crosses , the hybrid double mutant grew less well than either single mutant ( reciprocal sign epistasis ) . This work indicates that the first step toward speciation , partial postzygotic reproductive isolation , can evolve rapidly between populations under similar selective pressures , even with only a single genetic change in each .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"antimicrobials",
"medicine",
"and",
"health",
"sciences",
"drugs",
"population",
"genetics",
"microbiology",
"antifungals",
"epistasis",
"mutation",
"sign",
"epistasis",
"pharmacology",
"population",
"biology",
"mycology",
"lipids",
"fitness",
"epistasis",
"mutant",
"strains",
"sterols",
"reciprocal",
"sign",
"epistasis",
"biochemistry",
"heredity",
"genetics",
"microbial",
"control",
"biology",
"and",
"life",
"sciences",
"nystatin",
"evolutionary",
"biology"
] |
2017
|
Widespread Genetic Incompatibilities between First-Step Mutations during Parallel Adaptation of Saccharomyces cerevisiae to a Common Environment
|
Light plays a critical role in the regulation of numerous aspects of physiology and behaviour , including the entrainment of circadian rhythms and the regulation of sleep . These responses involve melanopsin ( OPN4 ) -expressing photosensitive retinal ganglion cells ( pRGCs ) in addition to rods and cones . Nocturnal light exposure in rodents has been shown to result in rapid sleep induction , in which melanopsin plays a key role . However , studies have also shown that light exposure can result in elevated corticosterone , a response that is not compatible with sleep . To investigate these contradictory findings and to dissect the relative contribution of pRGCs and rods/cones , we assessed the effects of light of different wavelengths on behaviourally defined sleep . Here , we show that blue light ( 470 nm ) causes behavioural arousal , elevating corticosterone and delaying sleep onset . By contrast , green light ( 530 nm ) produces rapid sleep induction . Compared to wildtype mice , these responses are altered in melanopsin-deficient mice ( Opn4-/- ) , resulting in enhanced sleep in response to blue light but delayed sleep induction in response to green or white light . We go on to show that blue light evokes higher Fos induction in the SCN compared to the sleep-promoting ventrolateral preoptic area ( VLPO ) , whereas green light produced greater responses in the VLPO . Collectively , our data demonstrates that nocturnal light exposure can have either an arousal- or sleep-promoting effect , and that these responses are melanopsin-mediated via different neural pathways with different spectral sensitivities . These findings raise important questions relating to how artificial light may alter behaviour in both the work and domestic setting .
In addition to its familiar visual function the mammalian retina mediates a broad range of non-image forming responses to light , including the entrainment of circadian rhythms [1] , regulation of pineal melatonin synthesis [2] , pupillary light constriction [3] and the regulation of sleep [4–6] . Research on these responses led to the identification of a novel retinal photoreceptor , consisting of a subset of photosensitive retinal ganglion cells ( pRGCs ) expressing the photopigment melanopsin ( OPN4 ) [7 , 8] . These melanopsin pRGCs project to a wide range of brain targets , which are thought to mediate the effects of light on different aspects of physiology and behaviour . For example , entrainment of circadian rhythms is mediated by projections to the suprachiasmatic nuclei ( SCN ) [7–9] , whereas sleep induction is thought to involve the ventrolateral preoptic area ( VLPO ) [4–6] . Non–image-forming responses to light persist in the absence of rods and cones [1 , 2] , demonstrating a critical role of melanopsin pRGCs in these processes . However , loss of melanopsin does not abolish these responses , indicating that rods and cones can at least partially compensate for the absence of pRGC photosensitivity [7 , 10 , 11] . By contrast , loss of the pRGCs themselves produces a dramatic loss of non–image-forming responses , suggesting that these cells provide the principal conduit for the light input from rods and cones [12] . As such , whilst melanopsin pRGCs play a critical role in mediating non–image-forming responses to light , it is clear that these cells also receive inputs from rods and cones , which contribute to these processes [13–17] . Moreover , recent studies have shown that visual responses may also be modulated by irradiance information detected by melanopsin , for example , in light adaptation [18] . Surprisingly , given the role of melanopsin in the regulation of light-induced sleep , nocturnal light exposure in rodents has also been shown to result in a rise in plasma corticosterone [19 , 20] , an arousal response that is physiologically incompatible with sleep induction [21] . To date , no studies have addressed the relationship between the sleep- and arousal-promoting effects of acute light exposure in rodents , nor investigated the role of melanopsin in these responses . Here , we investigate the effects of different wavelengths of light on sleep induction to determine if there is a differential effect of wavelength on sleep versus arousal . In view of previous findings , we were surprised to find that blue light mediates behavioural light aversion and elevated corticosterone via melanopsin pRGCs , whilst green light results in rapid sleep induction . Significantly , sleep induction in response to blue light is actually enhanced in Opn4-/- animals . Collectively , our data demonstrate that different wavelengths of light exert different effects on sleep and arousal , and these opposing responses are mediated by different photoreceptor pathways . Our data suggest that the role of melanopsin in the regulation of sleep is much more complex than originally envisaged .
Previous studies have shown that sleep induction in melanopsin knockout mice ( Opn4-/- ) in response to nocturnal white light stimuli is attenuated [4–6] . These data suggest a critical role for melanopsin in the regulation of sleep . To investigate this further , we studied behaviourally-defined sleep induction in response to different wavelengths in C57BL/6 mice . To prevent confounds from animals already being asleep , studies were performed at ZT14 when sleep pressure is low and animals show highest activity levels . In mice , the maximum absorption of rod opsin peaks at 498 nm [22] with M-cone opsin at 508 nm [23] , S-cone opsin around 360 nm [24] , and melanopsin around 480 nm [3] . We studied sleep onset in response to a 1 h isoquantal light exposure at ZT14 in wildtype C57BL/6 mice . We used three different wavelength light stimuli—violet ( 405 nm ) , blue ( 470 nm ) , and green ( 530 nm ) ( S1A Fig ) to produce differential activation of UVS cones , melanopsin , and rods/MWS cones ( S1B Fig ) . Given existing data on the role of melanopsin in sleep regulation , we predicted that blue light would result in the most rapid sleep induction , as this wavelength most closely corresponds to the peak sensitivity of melanopsin ( ~480 nm ) . All three wavelengths of light resulted in sleep induction at ZT14 , but with different latencies ( Fig 1A and 1B ) . Green light produced a very rapid sleep onset , which was observed almost immediately after light onset ( 2 ± 1 . 33 min ) . By contrast , both violet and blue light showed delayed sleep onset ( Fig 1B ) . Contrary to our prediction , sleep onset in response to blue light was significantly delayed ( 17 . 5 ± 1 . 64 min ) compared to either violet ( 7 . 5 ± 2 . 5 min ) or green light exposure ( S1 , S2 , S3 and S4 Videos ) . This delayed sleep induction under blue light resulted in a reduction of total sleep duration during the 1 h light pulse compared to either violet or green stimuli ( Fig 1C ) . To confirm that this effect was consistent at different times , sleep induction in response to blue and green light was also studied at ZT22 . A comparable delay in sleep induction in response to blue light was also observed at this later time ( S2 Fig ) , although differences were not as dramatic . This is most likely due to preceding differences in sleep/wake behaviour during the nocturnal active phase . To investigate whether the different effects of blue light on sleep and light aversion are mediated via melanopsin , we then performed the same experiments in melanopsin knockout ( Opn4-/- ) mice , on a congenic C57BL/6 background , using violet , blue and green light ( Fig 2A–2C ) . When compared with wildtype responses , sleep onset in Opn4-/- mice was significantly advanced under blue light ( Fig 2D ) , whereas under violet and green light it was significantly delayed . Sleep duration was reduced in response to violet and green light in Opn4-/- mice , but unaffected in response to blue light ( Fig 2E ) . Consistent with previous findings , delayed sleep induction and total sleep duration in response to white light were observed—comparable to the green light condition ( S3A–S3C Fig ) . Overall , when responses to different wavelengths were compared in Opn4-/- mice , no significant differences were evident ( S3D–S3F Fig ) . These results show that melanopsin is necessary for the acute wavelength-dependent effects of light on sleep . To determine why sleep onset was delayed in response to blue light , we studied behavioural light aversion to investigate if different wavelengths of light were associated with increased anxiety . Behavioural light aversion was measured using the light/dark box at ZT14 , comparing the time spent in the hidden ( dark ) zone of the apparatus when the transparent zone of the box was illuminated with different wavelengths ( Fig 3A ) . Data were analysed based upon the initial response to the test arena ( first entry to dark zone and time in dark zone during the first minute of the trial ) and over the whole course of the trial ( time in the dark zone from 2–10 minutes ) . A control condition in which both sides of the apparatus were unlit ( dark ) was also used . The latency from placing the mouse into the box until the first entry to the dark zone was significantly shorter under blue light ( 10 . 98 s ± 1 . 68 ) compared to green light ( 46 . 52 s ± 4 . 99 ) , violet ( 46 . 47 s ± 4 . 42 ) , and control ( 22 . 95 s ± 6 . 45 ) conditions ( S4 Fig ) . During the first minute of testing , mice under blue light spent more time in the hidden zone than under other conditions ( Fig 3B ) . By contrast , during the first minute of testing , mice under green or violet light remained in the illuminated area significantly longer than either the blue light or control group , suggesting that their exploration of this novel environment was not inhibited by light . Over the whole trial , mice spent significantly more time ( 87 . 5 ± 2 . 8% ) in the hidden zone under blue light compared to either green or violet light , as well as compared to control conditions ( Fig 3C ) . Under violet or green light , mice spent slightly more than 50% of their time in the dark zone ( green 64 . 15 ± 1 . 76% , violet 51 . 91 ± 1 . 59% ) and did not differ from that of control condition where the apparatus was unlit ( Fig 3C; control 66 . 54 ± 2 . 68% ) . Overall , these data demonstrate that mice find blue light aversive , both immediately within the first minute and over the 10 min timecourse of the experiment . This aversive effect of light may relate to the delay in sleep onset observed under blue light . By contrast , violet or green light does not evoke an aversive response , and over the first minute of the trial may result in novelty-evoked exploratory behaviour . In the home cage setting , this may in turn relate to the rapid sleep induction in response to green light . We subsequently compared behavioural light aversion in Opn4-/- mice under blue and green light conditions to determine if melanopsin plays a role in these responses . As we previously found no difference in light aversion between violet and green light ( Fig 3B and 3C ) , violet light was not studied . We found that in comparison to wildtype controls , Opn4-/- mice showed reduced light aversion to blue light over the whole trial and no difference to green light except in the first minute of the test ( Fig 3D and 3E ) . In the first minute of light exposure ( Fig 3D ) , in comparison to wildtype animals , Opn4-/- mice spent less time in the hidden zone under blue light , but more time in the hidden zone under green light . Rather than an increase in anxiety , this may simply reflect no preference for either zone ( as this was not significantly different to chance ) , unlike the increased preference for the lit zone seen in wildtype animals . Over the remaining test period ( Fig 3E ) , the reduced aversion to blue light in Opn4-/- mice was found to persist . However , the difference between wildtype and Opn4-/- mice in response to green light was not sustained over this period . When light aversion responses to blue and green light were compared in Opn4-/- mice , no effect of wavelength was observed ( Fig 3D and 3E ) . These data are consistent with the effects of light on sleep in Opn4-/- mice . The reduced aversive response to blue light in melanopsin-deficient animals is consistent with an advanced sleep onset , whereas the enhanced aversive response to green light at the beginning of the stimulus is consistent with delayed sleep onset . To determine if the different wavelengths of light result in different effects at a molecular level , we measured the expression of the immediate early gene Fos , and the light-regulated clock genes Per1 and Per2 following a 1 h light pulse in the SCN and adrenal gland ( Fig 4A ) . Consistent with previous studies , all three wavelengths evoked increased expression of Fos , Per1 , and Per2 in the SCN . Different wavelengths resulted in differential levels of induction of all three genes in the SCN , with blue light evoking significantly greater increases compared to either violet or green light . Violet and green light resulted in comparable levels of gene induction in the SCN . Nocturnal light exposure has previously been shown to result in induction of immediate early genes in the adrenal gland [19 , 25] . We reasoned that if light input is relayed from the retina to the adrenal gland via the SCN , light-induced changes in the adrenal should reflect those observed in the SCN . In the adrenal gland , we found that responses to blue light were significantly greater than to either green or violet light . Both blue and green light induced Per1 expression , with significantly greater Per1 induction compared to either green or violet light . Fos and Per2 expression were also induced by blue light , but not violet or green light . Given the high-amplitude changes in gene expression in response to blue light , we next investigated whether these responses were melanopsin-dependent . We compared gene expression in the SCN and adrenal gland between wildtype and Opn4-/- mice exposed to blue light ( Fig 4B ) . As the previously observed molecular responses to violet and green light were of much lower amplitude , these were not studied , as it would be extremely difficult to attain sufficient statistical power to resolve any difference . In response to blue light , Opn4-/- mice showed light-induced gene expression for all three genes in the SCN . However , these responses were attenuated compared to wildtype animals . In the adrenal gland , only Per1 and Per2 were increased in comparison to dark controls in response to light . All three genes showed attenuated responses in comparison to wildtype mice . Together , these results show that molecular responses to blue light are impaired in Opn4-/- mice at the level of both SCN and adrenal gland . The effects of light on SCN and adrenal , and the high levels of gene induction in response to blue light in particular , suggested that the behavioural responses to blue light may be accompanied by changes in plasma corticosterone . Previous studies have shown that acute white light exposure at ZT16 produces an increase in plasma corticosterone levels related to light-induced adrenal Per1 expression [19 , 20 , 25] . To determine if the observed sleep , behavioural , and molecular responses to blue light are accompanied by a rise in plasma corticosterone , we exposed wildtype C57BL/6 mice to 1 h of violet , blue or green light . All three wavelengths produced significantly elevated plasma corticosterone levels compared with dark-exposed controls ( Fig 5A ) . However , blue light resulted in significantly higher corticosterone levels compared to violet , green or dark conditions . Responses to green light were significantly lower compared to either violet or blue light . To investigate whether the effects of different wavelengths of light on Opn4-/- mice are reflected at the level of plasma corticosterone , we measured plasma corticosterone levels in Opn4-/- mice exposed to either blue or green light ( Fig 5B ) . Opn4-/- mice still show an elevation of plasma corticosterone levels in response to light , but responded differently in comparison to wildtype mice . Responses to blue light were significantly attenuated in Opn4-/- animals , although responses to green light were enhanced . As a result of these changes , no significant differences in corticosterone levels between blue and green stimuli were apparent in Opn4-/- mice . This finding is consistent with the loss of wavelength-dependent effects of light on sleep induction and behavioural light aversion in melanopsin-deficient animals . Surprisingly , baseline levels of corticosterone at ZT14 were slightly elevated in Opn4-/- mice compared to wildtype controls . To determine if blue and green light exert different effects at the level of retinorecipient targets , we studied responses to these wavelengths at the level of the SCN and VLPO ( Fig 6 ) . Mice were exposed to blue and green light stimuli and after 30 min SCN and VLPO were collected for gene expression analysis using quantitative PCR ( qPCR ) . Consistent with our previous findings ( Fig 4A ) , we found that blue light produced a larger Fos induction in the SCN than green light . However , in the VLPO we found a greater response to green light than blue light . As the sleep active neurons of the VLPO are galaninergic [26] , we also studied expression of Gal in both SCN and VLPO . In the VLPO ( but not the SCN ) , we found that Fos induction was accompanied by increased Gal expression . These data suggest that the different behavioural effects of blue and green light are mirrored by molecular responses at the level of the SCN and VLPO , suggesting that different pathways may mediate the wake-promoting and sleep-promoting effects of light on behaviour . It is unclear whether the inhibitory effects of blue light on sleep are mediated via elevated plasma corticosterone or whether elevated corticosterone levels are a downstream consequence of the heightened state of arousal produced by this stimulus . To investigate this question , we blocked the effects of corticosterone via systemic injection of the glucocorticoid receptor antagonist mifepristone ( RU-486 ) in wildtype C57BL/6 mice at ZT12 . These mice were then exposed to a blue light pulse at ZT14 . In comparison to vehicle-treated animals , mifepristone-treated mice showed elevated basal sleep ( S5 Fig ) . However , they did show significantly enhanced sleep onset in comparison to vehicle treated animals ( two-way ANOVA for treatment and time , treatment F ( 1 . 98 ) = 16 . 473 , p ≤ 0 . 001 , posthoc Tukey drug versus control first 10 min pulse p = 0 . 003 , 10–20 min p ≤ 0 . 001 , 20–30 min pulse p = 0 . 01 ) . These findings show that the effects of blue light on sleep induction can be attenuated by blocking the effects of elevated plasma corticosterone .
Here , we show that sleep induction and light aversion responses in mice are differentially affected by wavelength . Green light is associated with rapid sleep induction , whereas blue light results in an arousal response , characterised by reduced sleep , behavioural light aversion , and elevated plasma corticosterone . Notably we show that in melanopsin-deficient animals , these opposing effects of wavelength are abolished . As such , melanopsin-deficient animals show attenuated sleep induction under green light or enhanced sleep onset under blue light . These data suggest that the effects of nocturnal light exposure are more complex than first envisaged , and that sleep induction in rodents depends upon a balance between these arousal-promoting and sleep-promoting effects of light . Given the established role of melanopsin in the regulation of sleep in response to light [4–6] , we predicted that blue light would be most effective at inducing sleep in mice . To our surprise , we found that blue light was less effective , with a delayed sleep onset compared to other wavelengths . These findings appear to be due to a specific arousal-promoting response to blue light , which inhibits the normal sleep-promoting effects of light . In support of this hypothesis , the delayed sleep induction in response to blue light exposure is accompanied by behavioural light aversion , increased immediate early gene expression in both the SCN and adrenal gland , and elevated levels of plasma corticosterone . By contrast , green light results in rapid sleep induction in the home cage or transient increases in exploratory behaviour in a novel environment . These responses are accompanied by markedly lower immediate early gene induction in the SCN and adrenal gland and low levels of plasma corticosterone . The arousing effects of blue light suggest a potential role for melanopsin and , as predicted , we found that these responses were attenuated in melanopsin-deficient mice . As this arousal response to blue light normally inhibits sleep induction , melanopsin-deficient mice show enhanced sleep induction . Whilst this may appear to contradict previously published findings [4–6] , this is not the case . We show that melanopsin-deficient mice do indeed show attenuated sleep induction under green or white light . This can be explained by the fact that most white light sources have an emission spectrum optimised for human photopic vision ( with a peak sensitivity ~555 nm ) , which will be more comparable to our 530 nm rather than 470 nm wavelength stimulus [15] . To test this assumption , we compared the light available to each photoreceptor class from our different wavelength stimuli with that from commonly used white light sources—incandescent , fluorescent , and LED . As predicted , these white light sources are more comparable with the green ( 530 nm ) stimulus used in this study than the blue ( 470 nm ) ( S6A and S6B Fig ) . As such , sleep induction in melanopsin-deficient mice appears dependent upon the spectral power distribution of the light source used and may depend upon the balance of sleep-promoting and arousal-promoting effects of different wavelengths . In addition , previous studies have shown that melanopsin-deficient mice show normal behavioural light aversion under white light conditions [27] . This is consistent with the data presented here , where over the whole timecourse of light exposure under green light , melanopsin-deficient mice show light aversion comparable to wildtype animals . There is considerable evidence that the SCN relays photic information via the autonomic nervous system ( ANS ) to the adrenal gland to regulate corticosterone secretion in response to light [19 , 25 , 28] . This pathway is comparable to the well-known pathway mediating melatonin suppression in response to light , involving signals relayed via the SCN and sympathetic nervous system [2] . In agreement with these findings , we show that gene expression in both the SCN and adrenal gland is significantly increased after light exposure , and that the effects of different wavelengths of light on behaviour are mirrored by molecular responses in SCN and adrenal as well as in plasma corticosterone . In addition , we show that elevated responses to blue light are attenuated in melanopsin-deficient mice . This is the first demonstration that melanopsin plays a role in the regulation of adrenal corticosterone in response to light , adding to the repertoire of physiological and behavioural responses involving these photoreceptors . Glucocorticoids are important mediators of homeostasis and stress . They exert their complex effects via glucocorticoid receptors , which are expressed in both neurons and glia in the brain [29] . To determine if the increase in corticosterone observed is a downstream marker of the arousing effects of blue light , or whether corticosterone exerts a direct effect on the regulation of sleep , we assessed responses to blue light following administration of a glucocorticoid receptor antagonist . This resulted in enhanced sleep induction in response to blue light , suggesting that glucocorticoids exert a direct effect on the regulation of sleep . These changes in sleep latency appear to fall within the timescale of transcriptional regulation in response to glucocorticoid signalling . Moreover , recent data on the effects of stress on sleep have shown that chronic corticosterone administration affects sleep via activation of the locus coeruleus ( LC ) , which in turn suppresses GABAergic neurons of the VLPO . In this study , RU486 was shown to inhibit the effects of corticosterone on sleep by blocking the action of glucocorticoid receptors at the level of the LC . This provides a putative mechanism by which corticosterone induction in response to light may inhibit sleep induction [30] . Our data suggest that melanopsin provides an important drive for arousal in response to light . In response to blue light , arousal is high , and in the absence of melanopsin , this arousal response is attenuated . By contrast , the attenuated sleep responses to green and white light seen in melanopsin-deficient mice are more difficult to account for and do not appear to be due to impaired melanopsin signalling to the VLPO as previously assumed . Due to the overlapping spectral sensitivity of rods , cones , and melanopsin , high irradiance stimuli of different wavelengths will be expected to exert comparable effects on different non–image-forming responses . The finding that different wavelengths of light exert opposing effects on these responses could be due to either spectral opponency at the level of the retina or SCN , or independent pathways involving different photoreceptor projections . Whilst spectral opponency is consistent with recent findings of colour-sensitive neurons within the SCN [31] , it seems unlikely that this can account for our findings , as increased responses to shorter wavelength violet light would be expected , and these were not observed in the present study . By contrast , different non–image-forming responses to light are thought to be mediated via different retinorecipient areas of the brain , with the SCN mediating circadian entrainment and corticosterone induction and the VLPO mediating sleep responses [7 , 28 , 32] . This provides a plausible mechanism by which melanopsin projections to the SCN may mediate the arousing effects of light ( including the elevation of plasma corticosterone ) , whereas rod–cone signals may mediate acute sleep induction via activation of the VLPO . This mechanism is supported by the differing Fos-induction responses observed between the SCN and VLPO . Given the known projections of melanopsin pRGCs to the VLPO [32] , such responses could arise from different pRGC subtypes to the SCN and VLPO , with SCN input from M1 pRGCs and VLPO input from non-M1 cells , which may be more dependent upon rod/cone input [33] . Moreover , whilst attention has focused on the sparse innervation of the VLPO by melanopsin pRGCs , information on the innervation of the VLPO by non-melanopsin RGCs is limited [26] . Finally , it is also possible that photic information from the retina may mediate VLPO responses indirectly via its wide range of inputs [34] . If melanopsin mediates the arousal-promoting effects of blue light , why do melanopsin-deficient mice show attenuated responses to green light ? Unlike the innervation of the SCN by M1 pRGCs mediating the arousing responses to blue light , the pRGC subtype innervating the VLPO may be more dependent on rod and cone input in addition to its intrinsic photosensitivity . Melanopsin has recently been shown to provide an independent measure of irradiance to provide optimal settings for visual circuits [18] . If rod and cone signals are primarily mediating the sleep-promoting effects of light , a loss of melanopsin-dependent light adaptation may account for the impaired sleep induction observed in this and previous studies on the role of melanopsin in sleep . Under these bright light conditions , rods and cones will be saturated as they will be unable to adapt . This explanation is attractive as it also accounts for the loss of chromatic responses observed in melanopsin-deficient mice . This finding suggests that under bright light conditions , specific deficits may occur in wavelength discrimination in the absence of melanopsin . Finally , it should not be overlooked that the elevated basal corticosterone levels in melanopsin-deficient mice may also inhibit the sleep-promoting effects of green light . Our findings can be summarised as a model incorporating both arousal-promoting and sleep-promoting pathways ( Fig 7 ) . This model accounts for the opposing effects of different wavelengths of light , whereby blue light enhances arousal via SCN projecting M1 pRGCs , resulting in elevated corticosterone via autonomic innervation of the adrenal . Clearly , we would also anticipate that additional arousal pathways may be involved . For example , melanopsin pRGCs also innervate the subparaventricular zone ( SPZ ) , which has a direct projection to the dorsomedial hypothalamus ( DMH ) that in turn regulates both VLPO and lateral hypothalamus ( LH ) [35] . By contrast , we propose that green light promotes sleep induction via non-M1 pRGCs that are more dependent upon rod/cone input , which in turn project to the VLPO . These data are supported by the differential effects of blue and green light on Fos induction in the SCN and VLPO , respectively . In melanopsin-deficient animals , the arousal-promoting response to blue light is attenuated , resulting in an enhanced sleep onset ( due to a reduced arousal response ) . Moreover , melanopsin is also critical for adaptation of rod/cone pathways , and under bright light conditions , the absence of melanopsin results in attenuated sleep induction to green light . This in turn also accounts for the loss of chromatic responses observed in melanopsin-deficient mice . This model provides a tentative explanation for our data . However , given that different pRGC subtypes show an overlap in their projections [36] , a simple model based upon mutually antagonistic responses is perhaps overly simplistic , particularly given that the subtypes of pRGC projecting to the VLPO remain largely uncharacterised . Behavioural light aversion ( mediated via projections to regions such as the amygdala ) and adrenal corticosterone responses ( mediated via the SCN and sympathetic nervous system ) may involve separate neuronal circuits operating over different timescales . However , there is considerable evidence of an intimate relationship between adrenal stress hormones and arousal pathways involving the amygdala [37 , 38] . As such , we propose that these pathways operate to provide a consistent behavioural response to promote arousal to environmental stimuli , but given the diversity of pathways known to modulate sleep and arousal [35] , the model we propose in Fig 7 is likely to be an oversimplification . It is unclear why separate neural pathways are necessary to regulate arousal- and sleep-promoting pathways . One possible explanation is that the detection of environmental irradiance required for non–image-forming responses such as circadian entrainment and hormonal responses are well-served by the response characteristics of the melanopsin pRGC system . However , sleep involves reciprocal interactions between a wide range of sleep- and wake-promoting brain regions , giving rise to a “flip-flop” switch [39] . This bistable feedback loop is thought to be important in preventing intermediate states , minimising the time during which transition states between sleep and waking states occur . As such , the acute switch between sleep states in response to light is very different from other non–image-forming responses , as the critical signal is the transition between dark and light rather than ambient irradiance levels . As rods and cones will provide more reliable indicators of light transitions , this may account for their greater role in sleep induction . Consistent with this explanation , melanopsin is essential for the maintenance of light-induced sleep , where long-term environmental irradiance is a more reliable stimulus [40] . It is also possible to place our results into a broader ecological context . Around twilight , there is a relative enrichment of blue light across the dome of the sky [41] . This would result in a change in the balance between blue light- and green light-sensitive pathways ( S6C Fig ) . The spectral sensitivity of the blue light-aversive pathways may act to delay mice emerging from their burrows during early dusk and encourage a retreat to their burrow at early dawn . Such responses may help reduce the risk of predation at dawn and dusk , when there is a conflicting requirement for exploratory activity and safety . By contrast , exposure to longer wavelengths ( green or white light ) during the day may serve to limit daytime activity during waking bouts and promote sleep . Human studies have demonstrated an important role of light in the regulation of arousal . This involves activation of multiple brain areas involved in alertness , cognitive function , and melatonin production [42–47] . These studies report an increased effect of short wavelength light ( 470 nm or lower ) associated with the increased suppression of melatonin , reduction in subjective sleepiness , reduced reaction times , and changes in EEG power in the delta-theta frequency range [43 , 44 , 48] . Moreover , blue light exposure prior to sleep has been shown to reduce slow wave activity in the 0 . 75–5 . 5 Hz range in the first sleep cycle with a subsequent increase in the third sleep cycle . REM sleep duration during these cycles was also observed to be shortened [49] . These findings are consistent with recent studies on the effects of light-emitting devices on both melatonin rhythms and sleep latency [50] . These data suggest a key role of blue light exposure in inducing arousal and wakefulness in humans and are supported by our findings in mice . As C57BL/6 mice are naturally melatonin deficient [51] , the mechanisms mediating this effect of blue light are unlikely to be melatonin dependent . However , our data suggest that despite the differences between nocturnal and diurnal species , light may play a similar alerting role in mice as has been shown in humans , providing a valuable animal model to study the effects of light on cognitive function [48] . Whilst the arousal-promoting effects of light are consistent with human data , the sleep-promoting effects of light may differ between nocturnal and diurnal species . Here , we show that despite the arousal-promoting effects of blue light , mice still eventually go to sleep in response to nocturnal light exposure . By contrast , in diurnal species , nocturnal light may be expected to promote wakefulness , which may therefore be potentiated rather than antagonised by the arousing effects of blue light . In summary , here we show that light may have either arousal-promoting or sleep-promoting effects on mice , with corresponding changes in behaviour and molecular responses , including the regulation of adrenal corticosterone . These opposing effects of light are dependent upon wavelength , with blue and green light resulting in arousal and sleep , respectively . These coordinated responses appear to involve different neural pathways , which are both impaired in melanopsin-deficient mice . This results in a reduction in the arousal-promoting effects of blue light , as well as attenuated sleep-promoting responses to green light . In addition to providing the first demonstration of a role for melanopsin in the regulation of adrenal corticosterone , these findings challenge our existing understanding of the role of melanopsin in the regulation of sleep . Finally , the identification of a light-dependent arousal-promoting mechanism in mice provides a valuable animal model to study the effects of light on human cognitive brain function . With the increasing demands of the 24/7 society resulting in rising levels of both artificial lighting and shift work , there is an urgent need to understand the biological effects of artificial light on sleep versus alertness .
All experiments were conducted in accordance with the Home Office ( UK ) regulations , under the Animals ( Scientific Procedures ) Act of 1986 . All procedures were reviewed by the clinical medicine animal care and ethical review body ( AWERB ) , and were conducted under PPL 30/2812 , under PILs IF656800A and I869292DB . Wild type male C57BL/6 mice ( 3/7 mo old ) and melanopsin knockout ( Opn4-/- ) on a C57BL/6 background ( ten generations of backcrossing ) were used throughout . Before all experiments , mice were singly housed under a 12:12 light dark cycle with white light 250 lux when measured at the bottom of the cage , with food and water available ad libitum . Animals were housed in specific pathogen free ( SPF ) conditions , and the only reported positives on health screening over the entire time course of these studies were for Helicobacter hepaticus and Entamoeba sp ( Envigo , Alconbury UK ) . To assess behavioural responses ( including activity , sleep induction/duration and light aversion ) to different wavelengths of light we monitored home cage behaviour with miniature near infrared ( NIR ) video cameras ( Sentient Mini night vision CCTV camera , Maplin , UK ) mounted above each cage . Behaviour was tracked for 3 h and analysed in real time using ANY-maze ( Version 4 . 98 , Stoelting , US ) . Sleep onset was determined from episodes of immobility of 40 s or longer , and data were grouped into 10 min bins . This method has previously been validated , showing that the extended immobility correlates very highly ( correlation coefficient = 0 . 95 ) with EEG/EMG under both baseline conditions as well as following administration of pharmacological agents . Moreover , we have also shown that this method is capable of accurately identifying the dose-dependent effects of sedatives , stimulants and light [52] . To estimate the effect of monochromatic light pulses on immobility induction we housed Opn4-/- and WT mice under a LD cycle ( white light 250 lux ) . During video tracking , all mice have been kept in polypropylene cages , and the top of the cage was open so that the entire cage could be used as open field arena . The water bottle as well as the feeder ( a drawer ) were installed outside of the cage . White light as well as monochromatic light ( 530 nm , 470 nm , and 405 nm ) were provided using cool white LEDs ( Luxeon Star , Quadica Developments Inc , Brantford , Ontario ) . These wavelengths were used to produce spectral separation between stimuli , due to the overlapping photopigment absorption curves . The photon flux of all three monochromatic LEDs was isoquantal at 14 . 9 log quanta when measured at the bottom in the middle of the cage . The wavelength as well as the photon flux were measured using a calibrated Ocean Optics spectrometer . For acclimatisation in the new cages mice were kept for 5 days under a white light LD cycle ( ~250 lux ) . A 1 h monochromatic light pulse was administered at ZT14 . Mice which exhibited behaviourally-defined sleep prior to the light pulse were excluded from the analysis , as light was found to result in waking ( rather than sleep ) under such conditions . As such , unlike circadian responses to light , the behavioural state of animals prior to light exposure was found to be critical when considering sleep induction . Therefore , only mice which showed activity 10 min before light pulse were considered for sleep analysis . Moreover , as baseline sleep is lowest at the start of the dark phase , sleep induction light pulses were administered at ZT14 ( rather than later in the dark phase ) to coincide with this period of greatest waking . Sleep latency for each animal was determined as the time during which the first bout of behaviourally-defined sleep occurred . Total sleep was defined as the total period of behaviourally-defined sleep during the 1 h light pulse . Light aversion was measured in naive mice using a dark chamber light chamber paradigm as previously reported [53] . Behavioural testing was performed for 10 min at ZT14 and analysed using ANYmaze software . To estimate the difference in aversion behaviour induced by monochromatic light , the testing was performed with 470 nm , 405 nm , and 530 nm . Control studies were performed with no light . Each mouse was gently placed into the illuminated chamber of the box facing the portal to the dark chamber . A video tracking system ( see experimental setup for behavioural test ) was used to monitor behaviour and to quantify the total time that mice spent in the illuminated versus the dark chamber during a 10 min trial . The chambers were thoroughly cleaned between animals . For ex vivo phase analysis we exposed wildtype and Opn4-/- to a 1 h light pulse and collected trunk blood after 30–40 min , the time of peak corticosterone induction [19 , 25] . Dark-treated mice were used as a control . A parallel set of mice was equally treated with light pulse , and one hour after light exposure ( the expected peak of Fos and Per1 ) SCN and adrenal gland were collected and used for Per1 , Per2 , and Fos gene expression analysis using qPCR . After 1 hr light pulse , animals were killed under dim light by cervical dislocation . To prevent any photic stimulation to the SCN , eyes were immediately removed . Removed brains were placed into a brain matrix ( Kent Scientific , Torrington CT , US ) . Two skin graft blades ( Kent Scientific , Torrington CT , US ) one positioned at Bregma −0 . 10 mm and the second at Bregma −1 . 10 caudal from the first , were used to cut a 1 mm thick brain slice . SCN punches were collected from the brain slices by using a sample corer ( 1 mm internal diameter , Fine Science Tools GmbH , Heidelberg , Germany ) . SCN punches were stored at −80°C prior to RNA extraction . Wildtype mice were exposed to a 1hr green light pulse before SCN and VLPO punches were collected 30–40 min after the end of the light pulse . This time point is characterised by a peak of Fos in the SCN [5 , 54 , 55] . To prevent any photic stimulation to the SCN eyes were immediately removed . Three skin graft blades were used ( Kent Scientific , Torrington CT , USA ) one positioned at Bregma −0 . 10mm and the second at Bregma −1 . 10 caudal from the first for the SCN ( as described above ) , and the third 1 mm before 1 . 10 mm for VLPO . Two 1 mm thick brain slices were then dissected . SCN as well VLPO punches were collected as described previously [5 , 55] using a sample corer ( 1mm internal diameter , Fine Science Tools GmbH , Heidelberg , Germany ) . Samples were stored at −80°C prior RNA extraction . Total RNA of SCN and VLPO punches was extracted using the microRNeasy column method ( QIAGEN , Hilden , Germany ) . Total RNA of adrenal gland was extracted using the miniRNeasy column method ( QIAGEN , Hilden , Germany ) . The quantity of RNA was estimated using Nanodrop1000 ( Thermo Fisher Scientific , Waltham , MA USA ) . cDNA was synthesised from RNA samples with a qScript cDNA synthesis kit ( Quanta Biosciences , Gaithersburg , MD ) . qPCR was performed with Sybr green and SDS7700 thermal cycler ( Applied Biosystems , Foster City , CA ) . Relative quantification of transcript levels was conducted as described previously [56] . The geometric mean of three housekeeping genes Actb , Gapdh , and Arbp was used for normalisation . Galanin was used as a positive control as described in previous studies [6] . Expression levels were normalised to control ( dark ) values . Per1 , Per2 , Fos and Gal expression were estimated using following primers: Per1 forward AGTTCCTGACCAAGCCTCGTTAG , Per1 reverse CCTGCCCTCTGCTTGTCATC , Per2 forward GGGGTGAGATTCGTCATTGAACTTG , Per2 reverse AGGACATTGGCACACTGGAAAGAG , Fos forward ATCGGCAGAAGGGGAAAGTAG , Fos reverse GCAACGCAGACTTCTCATCTTCAAG , Gal forward ATGCCTGCAAAGGAGAAGAGAGGT , Gal reverse TCTGTGGTTGTCAATGGCATGTGG . Trunk blood was collected in microfuge tube containing anticoagulant EDTA ( 10 ul for 500 ul blood ) , kept for 10 min on ice and centrifuged . Obtained plasma was stored at −20°C until corticosterone measurement . Corticosterone was measured from 1:10 diluted samples using an ELISA kit ( Assaypro LLC St . Charles , MO ) according to manufacturer’s instructions . To block glucocorticoid receptor mediated effects , we injected mice with the glucocorticoid antagonist RU-486 ( Sigma Aldrich ) at ZT12 and started measuring behaviour from ZT13 until ZT15 . RU-486 was administered as described previously [57] . Briefly , RU-486 was dissolved in 20% dimethyl sulfoxide ( DMSO ) /80% polythelyne glucol ( PEG ) -300 . Drug dosage ( 100 mg/kg body weight ) was based on the previous studies [57 , 58] . As control the same amount of ( 1 ml/kg body weight ) of vehicle ( saline diluted in 20%DMSO/80% ( PEG ) -300 ) was injected . To determine the light available to the different photoreceptors of the mouse retina for the different wavelength stimuli used in this study as well as white light sources , the Rodent Toolbox v1 . 0 was used [15] . For the 405 nm , 470 nm , and 530nm LEDs used in this study , spectral power distributions were measured using a calibrated Ocean Optics spectrometer . For white light sources , standard illuminant functions were used from the Toolbox . For daylight and twilight data , previously published data were used [59] . Comparable results for daylight were obtained using the D65 standard illuminant function . Data were expressed in alpha-opic lux . However , as it is not possible to directly compare different photoreceptor channels , these were expressed as relative to the total alpha-opic lux across all four channels . Results are presented as mean ± SEM . Statistical analysis was performed using SPSS Statistics 22 . 0 . All data for statistical analysis were first checked for normal distribution using Kolmogorov Smirnov test . Non-normally distributed data were log transformed prior to statistical analysis to ensure normality and homogeneity of variance . Statistical significance of group comparisons was tested with one-way ANOVA and posthoc Tukey . Multivariate experiments such as genotype and treatment were analysed using two-way ANOVA with Tukey’s posthoc analysis . Time course experiments such as across multiple hours or min were analysed using rep . ANOVA . Assessment of relative photoreceptor activity under the 405 nm , 470 nm , and 530 nm wavelengths has been performed using the previously published rodent toolbox [15] .
|
Light exerts profound effects on our physiology and behaviour , setting our biological clocks to the correct time and regulating when we are asleep and we are awake . The photoreceptors mediating these responses include the rods and cones involved in vision , as well as a subset of photosensitive retinal ganglion cells ( pRGCs ) expressing the blue light-sensitive photopigment melanopsin . Previous studies have shown that mice lacking melanopsin show impaired sleep in response to light . However , other studies have shown that light increases glucocorticoid release—a response typically associated with stress . To address these contradictory findings , we studied the responses of mice to light of different colours . We found that blue light was aversive , delaying sleep onset and increasing glucocorticoid levels . By contrast , green light led to rapid sleep onset . These different behavioural effects appear to be driven by different neural pathways . Surprisingly , both responses were impaired in mice lacking melanopsin . These data show that light can promote either sleep or arousal . Moreover , they provide the first evidence that melanopsin directly mediates the effects of light on glucocorticoids . This work shows the extent to which light affects our physiology and has important implications for the design and use of artificial light sources .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"white",
"light",
"medicine",
"and",
"health",
"sciences",
"sleep",
"vertebrates",
"light",
"mice",
"animals",
"mammals",
"electromagnetic",
"radiation",
"physiological",
"processes",
"animal",
"behavior",
"adrenal",
"glands",
"visible",
"light",
"zoology",
"behavior",
"gene",
"expression",
"light",
"pulses",
"physics",
"rodents",
"anatomy",
"physiology",
"endocrine",
"system",
"genetics",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"amniotes",
"organisms"
] |
2016
|
Melanopsin Regulates Both Sleep-Promoting and Arousal-Promoting Responses to Light
|
Alveolar macrophages ( AM ) are critical for defense against bacterial and fungal infections . However , a definitive role of AM in viral infections remains unclear . We here report that AM play a key role in survival to influenza and vaccinia virus infection by maintaining lung function and thereby protecting from asphyxiation . Absence of AM in GM-CSF-deficient ( Csf2−/− ) mice or selective AM depletion in wild-type mice resulted in impaired gas exchange and fatal hypoxia associated with severe morbidity to influenza virus infection , while viral clearance was affected moderately . Virus-induced morbidity was far more severe in Csf2−/− mice lacking AM , as compared to Batf3-deficient mice lacking CD8α+ and CD103+ DCs . Csf2−/− mice showed intact anti-viral CD8+ T cell responses despite slightly impaired CD103+ DC development . Importantly , selective reconstitution of AM development in Csf2rb−/− mice by neonatal transfer of wild-type AM progenitors prevented severe morbidity and mortality , demonstrating that absence of AM alone is responsible for disease severity in mice lacking GM-CSF or its receptor . In addition , CD11c-Cre/Ppargfl/fl mice with a defect in AM but normal adaptive immunity showed increased morbidity and lung failure to influenza virus . Taken together , our results suggest a superior role of AM compared to CD103+ DCs in protection from acute influenza and vaccinia virus infection-induced morbidity and mortality .
Alveolar macrophages ( AM ) are lung-resident macrophages important for the maintenance of surfactant homeostasis in the alveolar space [1] . Their importance for lung physiology becomes evident in a rare human syndrome termed “pulmonary alveolar proteinosis” ( PAP ) , which is characterized by the accumulation of surfactant material and a varying degree of respiratory insufficiency [2] . PAP patients have a higher risk for pulmonary infections with opportunistic pathogens [3] . PAP typically occurs in patients that spontaneously develop GM-CSF autoantibodies [4] or carrying mutations in the GM-CSF receptor α chain [5] associated with impaired function and/or reduced numbers of AM . Similarly , mice lacking GM-CSF ( Csf2−/− ) or the receptor β chain ( Csf2rb−/− ) develop PAP [6] , [7] , [8] , [9] and display increased susceptibility to a range of bacterial and fungal infections , which is associated with impaired innate functions of AM [10] , [11] , [12] , [13] , [14] . High phagocytic activity and expression of pattern recognition receptors provides them with the capacity to respond to bacterial and fungal pathogens . However , AM were also described to have anti-inflammatory properties based on direct inhibition of the antigen-presenting function of lung DCs [15] and production of immunosuppressive mediators such as IL-10 and nitric oxide [16] . In addition , AM were proposed to sequester pulmonary antigen and thereby interfere with efficient priming of immune responses by DCs [17] . In contrast to the well-described functions of AM in bacterial and fungal infections , their precise role during viral infections is poorly understood . AM have the capacity to endocytose adenovirus particles , which was impaired in cells isolated from Csf2-deficient mice [10] . Furthermore , AM are potent producers of type I IFNs upon pulmonary virus infection [18] , [19] . A beneficial role of AM in influenza infection has been proposed based on AM depletion experiments [20] , [21] , [22] and treatment of mice with GM-CSF that increased numbers of AM [23] , [24] . However , besides the described role of GM-CSF for AM function under steady-state conditions , GM-CSF also influences DCs as demonstrated by its positive effects on the homeostasis of CD103+ DCs [25] or the upregulation of CD103 on CD11b− DCs [26] . Indeed , a recent report proposed that GM-CSF protects from lethal influenza virus infection by enhancing CD103+ DC mediated anti-viral T cell responses [27] . In this study , we revisited these controversies by comparing DC subpopulations and anti-viral CD8+ T cell responses in the presence or absence of AM with the outcome of respiratory viral infection in Csf2−/− , Csfrb2−/− , Batf3−/− , CD11c-Cre/Ppargfl/fl ( with a defect in AM ) and wild-type mice . Csf2-deficient mice developed severe morbidity and hypersuceptibility to influenza virus and vaccinia virus infection due to the complete absence of AM , although CD103+ DC mediated anti-viral T cell responses were unaffected . In contrast , virus-induced morbidity was moderately affected in Batf3-deficient mice , which mounted impaired anti-viral T cell responses due to the absence of CD8α+ DCs and lung CD103+ DCs . Selective depletion of AM using clodronate liposomes prior influenza virus infection resulted in morbidity comparable to Csf2-deficient mice . Moreover , mice with functionally impaired AM due to conditional deletion of Pparg showed increased disease severity and respiratory failure despite normal T cell responses . In contrast , selective postnatal AM reconstitution in Csfrb2−/−mice prevented pulmonary pathology and morbidity following influenza virus infection . Overall , our study provides evidence for a vital function of AM for the maintenance of lung function upon pulmonary viral infections .
Mice lacking GM-CSF ( Csf2−/− ) or the β subunit of its receptor ( Csf2rb−/− ) have been reported to develop alveolar proteinosis due to a defect in terminal maturation and surfactant catabolism of AM [6] , [8] . However , AM characterized as CD45+CD11c+Siglec-F+ cells with high autofluorescence were completely absent in lung and BAL of adult Csf2−/− mice ( Figure 1A , B and Figure S1 ) . Staining with a viability dye showed a striking enrichment of dead non-hematopoietic ( CD45− ) cells , presumably consisting mainly of lung epithelial cells ( Figure 1C ) , indicating that the removal of dead cells ( efferocytosis ) in the respiratory tract is impaired in the absence of AM . Indeed , following tracheal instillation of labeled apoptotic thymocytes , we found that AM are by far the most potent cell type compared to DCs and neutrophils in removal of apoptotic cells from the bronchoalveolar space ( Figure 1D and Figure S1 ) . The few remaining cells that showed up in the CD11c+Siglec-F+ population of Csf2−/− mice were contaminating dead cells with high autofluorescence as demonstrated by comparing the analysis with and without prior exclusion of dead cells ( Figure S1 ) . Absence of AM resulted in the accumulation of lipoproteinaceous and eosinophilic material in the BAL and lung of Csf2−/− mice ( Figure 1E–G ) . Moreover , while BAL of WT mice consisted almost entirely ( >90% ) of AM , BAL of Csf2−/− mice contained a substantial proportion of neutrophils ( Figure 1H ) . This indicates low-grade inflammation possibly induced by the accumulation of dead cells ( Figure 1C ) . Nonetheless , arterial oxygen saturation in Csf2−/− and WT mice was comparable indicating unimpaired gas exchange in the lung in the absence of AM under steady state ( Figure 1I ) . To determine whether AM require cell intrinsic GM-CSFR signaling for their development , we generated mixed bone marrow ( BM ) chimeras by adoptive transfer of a 1∶1 mixture of BM cells from Csf2rb−/− ( CD45 . 2+ ) and WT ( CD45 . 1+ ) mice ( as well as CD45 . 2+WT and CD45 . 1+WT BM as controls ) . The reconstituted AM in the BAL and lung of Csf2rb−/−:WT chimeras were exclusively of WT donor origin and prevented development of alveolar proteinosis . ( Figure 1 J , K and data not shown ) . In agreement with a recent report , these data show a requirement of GM-CSF signaling for the reconstitution of AM post-irradiation [28] . GM-CSF can be expressed by a variety of different cell types including T cells , macrophages , mast cells and non-hematopoietic cells [29] . To assess the source of GM-CSF for the development of alveolar macrophages , we generated 4-way Csf2−/− and WT crisscross BM chimeras . Irradiated WT recipients that received BM from either Csf2−/− or WT mice did not develop any pulmonary alveolar proteinosis during 16 weeks post-reconstitution . In contrast , Csf2−/− recipients reconstituted with WT or Csf2−/− BM showed a massive and comparable proteinosis ( Figure S2 ) . These results identify radio-resistant non-hematopoietic cells such as epithelial cells as the primary source of GM-CSF in the lung , which is absolutely essential for development of alveolar macrophages . Development of lung CD103+ DCs has been reported to depend on GM-CSF [25] . However , comparing adult WT and Csf2−/− mice , we found similar numbers of pulmonary CD103+ DCs , although CD103 surface expression on CD11b−CD11c+ DCs was strongly impaired in Csf2−/− mice ( Figure 2A–C ) consistent with a previous study using Csf2rb−/− mice [26] . Interestingly , CD103+ DCs were strongly reduced in 6-day-old Csf2−/− pups ( Figure 2D , E ) . In addition , injection of BM from Csf2rb−/− donors to irradiated WT recipients failed to reconstitute pulmonary CD103+ DCs ( Figure 2F ) [25] , while other myeloid cell populations in the lung and peripheral blood were comparably well reconstituted from Csf2rb−/− and WT bone marrow cells ( Figure S2 ) . These data suggest that GM-CSF plays an important role in development of CD103+ DCs , although a population of CD103lowCD11b− DCs can substitute typical CD103+ DCs in the absence of GM-CSF possibly under inflammatory conditions . To determine whether absence of AM and/or impaired CD103+ DC compartment affect resistance of Csf2−/− mice to respiratory viral infection , we infected WT and Csf2−/− mice with a sublethal dose of influenza virus PR8 . Compared to WT mice , Csf2−/− animals showed an earlier onset of disease and a much more pronounced loss of body weight and body temperature ( Figure 3A , B ) . Strikingly , whereas all WT mice recovered from infection , moribund Csf2−/− mice had to be euthanized between day 10 and 12 post-infection ( Figure 3C ) . Moreover , Csf2−/− mice showed an increased lung virus titer at the peak of viral load ( i . e . day 6 ) and a slight delay in viral clearance ( Figure 3D ) , which were nevertheless to small to explain the moribund state of the Csf2−/− mice . Importantly , the fatal outcome of infection resulted from the absence of GM-CSF production by non-hematopoietic cells as demonstrated by a higher susceptibility of WT→Csf2−/− compared to Csf2−/−→WT BM chimeras ( Figure S3 ) . Lung CD103+ DCs have been reported to induce CD8+ T cell responses by cross-presentation of influenza virus [19] , [30] . Notably , Csf2−/− mice mounted intact anti-viral CD8+ T cell effector responses in BAL , lung , and draining LN ( Figure 3E , F ) despite a 2-fold reduction in lung CD103+ DC frequencies and decreased CD103 surface levels during influenza virus infection ( Figure 3G ) . Nonetheless , total numbers of CD103+ DCs were similar in Csf2−/− compared to wild-type mice ( Figure 3H ) . In contrast , influenza virus infected Batf3-deficient mice were almost entirely devoid of lung CD103+ DCs ( Figure 3G , H ) and showed a strong reduction in virus-specific CD8+ T cells and anti-viral effector cytokine ( i . e . TNFα and IFNγ ) production [19] . Regardless , morbidity and disease severity remained almost completely unaffected in Batf3-deficient compared to wild-type mice , while Csf2−/− mice succumbed to infection ( Figure 3I ) . Notably , Batf3−/− mice were able to efficiently clear the virus despite impaired anti-viral CD8+ T cell responses ( Figure S4 ) . To further characterize the adaptive response and rule out a possible defect in anti-viral antibody responses in the absence of GM-CSF , we assessed anti-viral B cell responses by measurement of influenza HA-specific antibodies . Csf2−/− mice showed increased levels of virus-specific IgA and IgG2a antibodies in BAL and serum compared to wild-type mice ( Figure S4 ) . Furthermore , we found similar NK cell recruitment and increased activation in Csf2−/− compared to wild-type mice ( Figure S4 ) . Together these results demonstrate that CD103+ DC mediated anti-viral CD8+ T cell responses or B cell responses are intact and therefore not the underlying reason for the disease severity in Csf2−/− mice . As shown above , absence of AM in mice lacking GM-CSF results in accumulation of surfactant material and dead cells under homeostatic conditions . Interestingly , influenza virus infection strikingly aggravated pulmonary alveolar proteinosis in Csf2−/− mice indicated by an obstruction of alveoli with aggregates of eosinophilic material ( Figure 4A ) , increased total protein concentration ( Figure 4B ) and accumulation of surfactant material in the BAL fluid at days 6 and 10 p . i . ( Figure 4C ) . Moreover , the BAL of Csf2−/− mice was highly enriched in dead cells and cellular debris indicating impaired clearance of apoptotic cells ( Figure 4D–F ) . Accordingly , respiratory function as measured by arterial oxygen levels was critically impaired at the time when Csf2−/− mice succumb to infection ( Figure 4 G , H ) . These results suggest that AM prevent influenza induced morbidity by maintenance of lung function through removal of dead cells and surfactant material . AM were described to maintain independently from blood monocytes through self-renewal [31] . We and others recently showed that AM originate from a fetal monocyte precursor in the lung around E17–19 , which subsequently differentiates into mature AM during the following 5–7 days ( [32]; Schneider et al . , unpublished data ) . Single intranasal transfer of CD45+ lung cells from E18 . 5 fetuses into Csf2rb−/− newborns completely restored development of mature AM ( Figure 5A , B ) but did not contribute to the pool of BM-derived short-lived DCs including CD103+ DCs 6 weeks after transfer ( Figure 5C ) . Thus , we generated mice lacking Csf2rb in every cell but AM . Presence of wild-type AM in Csf2rb−/− mice protected from pulmonary proteinosis , severe morbidity and mortality following influenza infection ( Figure 5 D–F ) . Moreover , viral load in Csf2rb−/− mice with restored AM compartment was reduced to levels observed in wild type controls ( Figure 5G ) . These results unequivocally demonstrate that the absence of AM is responsible for disease severity and rule out a putative contribution of DCs or other immune cells in mice lacking GM-CSF or its receptor . To further rule out that a defect in CD103+ DCs contributed to the susceptibility of Csf2−/− mice to influenza virus infection , we selectively depleted AM in wild-type mice prior to infection using clodronate liposomes . Although DC subsets remained unaffected in clodronate-treated mice ( Figure S5 ) , AM-depletion resulted in increased morbidity ( i . e . temperature and weight loss ) ( Figure 6A , B ) and mortality ( Figure 6C ) , which was associated with hypoxia ( Figure 6D , E ) basically resembling the phenotype observed in Csf2−/− mice . These results further support the conclusion that absence of AM and respiratory failure rather than impaired CD103+ DC-mediated CD8+ T cell responses are responsible for the high morbidity of Csf2−/− mice to influenza virus infection . To assess whether the AM play an important role in resistance to other pulmonary viral infections , we assessed the outcome of infection with vaccinia virus ( WR ) in Csf2−/− and control WT mice . Control of vaccinia virus mainly depends on innate and T helper cell mediated B cell antibody responses , whereas the cytotoxic T cell response plays only a marginal role [33] , [34] . When infected with a virus dose that did not induce any morbidity in WT mice , Csf2−/− mice showed a pronounced loss of body weight and temperature ( Figure 6F , G ) starting already during the innate immune response at day 4 p . i . and displayed increased lethality ( Figure 6H ) , reminiscent to the outcome of influenza virus infection . Altogether , these results establish a direct role of AM in prevention of fatal respiratory viral infections . Mice lacking Pparg specifically in macrophages ( LysM-Cre/Ppargfl/fl ) develop pulmonary alveolar proteinosis [35] . Moreover , using CD11c-Cre/Ppargfl/fl mice that lack Pparg in AM and DCs , we recently found that development of AM is arrested in an immature state with a defect in surfactant catabolism and lipid metabolism while development of DC subsets remained unaffected ( data not shown ) . To determine whether absence of functional AM affects resistance of CD11c-Cre/Ppargfl/fl mice to respiratory viral infection , we infected WT and CD11c-Cre/Ppargfl/fl mice with influenza virus PR8 . Compared to WT , CD11c-Cre/Ppargfl/fl mice showed an earlier onset and more pronounced loss of body weight and body temperature ( Figure 7A , B ) associated with increased lethality ( Figure 7C ) , although lung virus titers were insignificantly elevated ( Figure 7D ) . Besides a slightly increased number of neutrophils , recruitment of inflammatory cells to the lung was in general normal in CD11c-Cre/Ppargfl/fl mice ( Figure S6 ) . Moreover , we found similar frequencies and total numbers of anti-viral CD8+ T cells as well as IFNγ-producing CD4+ and CD8+ T cells ( Figure 7E and Figure S6 ) indicating that DC-mediated T cell priming and effector responses were unaffected in CD11c-Cre/Ppargfl/fl mice . Likewise , levels of anti-influenza HA-specific antibody levels ( i . e . IgG2c , IgA and IgG1 ) were comparable in BAL and serum of CD11c-Cre/Ppargfl/fl and WT mice ( Figure 7F and Figure S6 ) . Together these results indicate that innate and adaptive anti-viral immune responses were largely intact in CD11c-Cre/Ppargfl/fl mice . Lungs of infected CD11c-Cre/Ppargfl/fl mice contained increased amounts of total protein in BAL fluid with a high proportion of lipid-engorged foam cells and alveoli filled with cellular infiltrate and debris ( Figure 7G–I and Figure S6 ) . Moreover , CD11c-Cre/Ppargfl/fl mice displayed significantly lower lung function as measured by arterial oxygen levels ( Figure 7J , K ) . Taken together , CD11c-Cre/Ppargfl/fl mice with immature and dysfunctional AM show increased morbidity and respiratory failure due to an impaired removal of dead cells and debris following influenza virus infection similar to , but not as pronounced as Csf2−/− mice that are completely devoid of AM . GM-CSF-deficient mice lacking alveolar macrophages showed 5–10-fold increased viral titers and a delayed viral clearance . Therefore , we next asked whether alveolar macrophages might contribute directly to viral clearance besides their role in prevention of respiratory failure following viral infection . Infection of C57BL/6 mice with PR8 or NS1-GFP influenza virus showed a considerable proportion of AM containing viral material as determined by staining of viral NP or expression of NS1-GFP ( Figure 8A , B ) . In addition , comparison of gene expression profiles of sorted AM from naïve and influenza virus infected mice by microarray analysis showed that the latter displayed a typical interferon signature with a 4- to 65-fold upregulation of several interferon-induced genes including the interferon-induced transmembrane protein 3 ( Ifitm3 ) and Ifitm6 , which were upregulated 42- and 9-fold , respectively ( Figure 8C ) . Ifitm3 has recently been described as a key component in restricting viral spread and the morbidity and mortality following influenza virus infection [36] . Thus it is tempting to speculate that AM act like a virus sink and prevent morbidity at least partially through Ifitm3 .
In this study , we revisited the role of GM-CSF in AM homeostasis and function of this cell population in respiratory viral infection . According to the current understanding , Csf2- and Csf2rb-deficient mice develop pulmonary alveolar proteinosis ( PAP ) due to a defect in terminal maturation of AM involving impaired lipid catabolism . However , using 8-parameter flow cytometry in combination with dead cell exclusion , we found that Csf2−/− mice were completely devoid of AM and presented a massive accumulation of dead CD45-negative cells , which are presumably epithelial cells , in the BAL and lung consistent with an important role of AM in removal of dead cells ( efferocytosis ) . Indeed , usage of a viability dye in combination with CD45 staining was inevitable for the exclusion of highly auto-fluorescent dead cells and avoidance of misidentification as AM . We also observed increased numbers of neutrophils in the BAL of naive Csf2−/− mice supporting an association of impaired efferocytosis and chronic inflammatory lung disease [37] . Using mixed bone marrow chimeras we demonstrated that AM exclusively differentiated from WT and not from Csf2rb−/− BM . The absence of Csf2rb−/− AM in mixed BM chimeras also excludes the possibility that these cells have developed and died subsequently due to a functional defect in surfactant metabolism and accumulation of cellular debris , as the mixed chimeras did not develop PAP pathology . These results demonstrate a cell intrinsic requirement of GM-CSFR signaling for licensing development of AM from a precursor cell . Our results are in line with a recent report describing highly reduced numbers of AM in Csf2−/− mice and a requirement of GM-CSF signaling for the reconstitution of AM post-irradiation [28] . The license is provided by GM-CSF secretion of radio-resistant lung cells , as shown by our 4-way BM chimeras . Whether during homeostasis GM-CSF also influences the functionality of AM needs to be determined using inducible knock out strategies or antibody-mediated neutralization . The role of GM-CSF for the development of DCs has been extensively studied . Most lymphoid tissue DCs develop normally in the absence of GM-CSF in vivo [38] . In contrast , GM-CSF has been shown to play a pivotal role for the development of non-lymphoid tissue DCs in the lamina propria [39] and the skin [40] . A recent report suggested that GM-CSF is critical for the homeostasis of tissue-resident CD103+ DCs in particular in the lung and skin [25] , while others found that GM-CSF merely regulates CD103 expression levels on DCs and not development of this DC subset [26] . In keeping with the latter study by Edelson et al . , we found that total CD11b−CD103+ DC cell numbers were not affected in the lung of adult Csf2-deficient mice , while CD103 surface expression on CD11b− DCs was clearly reduced . Interestingly , in lungs of neonatal Csf2−/− mice the CD103+ DC subpopulation was strongly reduced in cell numbers . Similarly , the vast majority of CD103+ DCs present in lungs of mixed bone marrow chimeras originated from WT but not Csf2rb−/− bone marrow and the few Csf2rb−/− BM-derived CD11b− lung DCs were CD103low . Whether impaired CD103 expression affects DC function is not clear . CD103+ DCs possess a unique potential in phagocytosis of virus-infected epithelial cells and capacity in cross-presentation of viral antigens to CD8+ T cells [41] . Several reports suggested that CD103+ DCs are the main inducers of CD8+ T cell responses to influenza virus infection with supportive evidence mainly based on depletion experiment using langerin-DTR mice [19] , [30] , [42] . It should be noted , however , that langerin is also expressed on CD8α+ DCs from spleen and LNs [43] , [44] , which are accordingly also depleted in langerin-DTR mice [45] . Indeed , LN-resident CD8α+ DCs as well as CD11b+ lung tissue DCs have also been implicated in influenza antigen transport , presentation and CD8+ T cell priming [30] , [46] , [47] . Batf3-deficient mice that are devoid of both CD8α+ DCs and CD103+ DCs showed a substantial reduction in anti-viral CD8+ T cell responses indicating that both CD8α+ DCs in spleen and lymph node together with lung migrating CD103+ DCs contribute to the induction and maintenance of T cell responses to influenza virus . In contrast , despite an almost 2-fold reduction in frequencies , total lung CD103+ DC numbers in influenza virus-infected Csf2−/− mice were comparable to wild-type mice and accordingly , antiviral CD8+ T cell numbers and cytokine responses in the BAL , lung , and the draining LN remained unaffected . However , the influenza-induced pathology and morbidity was far worse in Csf2-deficient mice lacking AM but having intact B and T cell responses as compared to Batf3-deficient mice with constrained CD8+ T cell responses due to the absence of both CD103+ DCs and CD8α+ DCs but intact AM indicating that only the latter are critical for survival of respiratory viral infection . PAP was strikingly aggravated in influenza virus infected compared to naive Csf2−/− mice . The accumulation of cellular debris , dead epithelial cells and surfactant material in the respiratory tract in the absence of AM resulted in impaired gas exchange and eventually in a fatal hypoxia . Similar results were obtained by clodronate-mediated depletion of macrophages without affecting DCs in wild-type mice before viral infection . In keeping with our data , transgenic mice with lung-restricted GM-CSF overexpression ( SPC-GM ) and increased AM numbers [48] were better protected from influenza virus-induced morbidity [24] . However , using the same transgenic mice , another study linked the GM-CSF-induced protection to enhanced CD103+ DC-mediated antiviral T cell responses , although numbers of DCs and virus-specific CD4+ and CD8+ T cells in SPC-GM mice were comparable to non-transgenic WT mice and the viral load was significantly reduced as early as day 3 post-infection prior CD8+ T cell expansion and effector function [27] . Moreover , this study concluded that increased susceptibility of Csf2−/− mice is a consequence of defective CD103+ DC-mediated CD8+ T cell responses [27] . In contrast , protection from virus-induced pathology and mortality by reconstitution of AM development in Csf2rb−/− mice ( Figure 5 ) unequivocally demonstrates that the absence of AM is the underlying reason for hypersusceptibility to influenza virus in mice lacking GM-CSF or its receptor . Notably , Csf2−/− mice showed also increased morbidity and pronounced lethality in response to pulmonary vaccinia virus infection , which does not depend on CD8+ T cells [33] , [34] . These data suggest a vital role of AM in resistance to respiratory viral infection in general , although it should be noted that depletion of AM in mice infected with RSV did not alter subsequent disease development[49] . Induction of host cell death is a hallmark of viral infection including influenza virus infection . Indeed , increased cell death has been associated with the pathogenicity of highly virulent influenza virus [50] . Moreover , we found a correlation between viral burden , PAP , and severity of hypoxia in wild-type mice . Our findings suggest that the pivotal function of AM is the removal of dead cells and cellular debris to prevent clogging of the airways and maintain gas exchange during respiratory viral infection . In addition , AM may contribute to viral clearance or interfere with virus-induced morbidity by yet unknown mechanisms . Consistent with previous reports [51] [19] , we found that influenza virus can directly infect AM ( Figure 8A , B ) . Interestingly , transcriptome analysis of sorted AM from infected and uninfected mice revealed a striking upregulation of interferon response signature genes including high levels of Ifitm3 . Ifitm3 has recently been described as a key component in restricting viral spread and the morbidity and mortality following influenza virus infection [36] . Moreover , high Ifitm3 expression in influenza-specific lung-resident CD8+ memory T cells confers resistance to infection and enhances survival of these cells upon recall infection with the virus [52] . Thus , induction of Ifitm3 in AM could serve as a mechanism to promote AM survival and thereby limit the loss of this vital cell type during influenza infection . Furthermore and in addition to their crucial role in maintaining respiratory function , AM could have a direct antiviral role serving as a sink for influenza virus consistent with slightly elevated virus titers in mice lacking AM . Taken together , we identified a key function of alveolar macrophages in phagocytosis of dead cells and maintenance lung function in respiratory viral infections . Mice lacking Csf2 or Csf2rb are highly vulnerable to influenza virus infection due to the absence of AM but not potentially impaired DC/T cell immunity . These results have implications for therapies targeting Csf2 ( GM-CSF ) .
Csf2−/− mice ( originally kindly provided by A . Dunn , Ludwig Institute for Cancer Research , Royal Melbourne Hospital , Victoria , Australia ) [7] were backcrossed to BALB/c for 7 generations [53] . Csf2−/− ( C57BL/6 ) and Csf2rb−/− ( C57BL/6 ) mice were kindly provided by B . Becher ( University Zurich ) . C57BL/6 and BALB/c mice were purchased from Charles River ( Germany ) . It should be noted that most of the results shown were done with Csf2−/− ( C57BL/6 ) and validated with Csf2−/− ( BALB/c ) . B6 . 129S ( C ) -Batf3tm1Kmm/J ( Batf3−/− ) mice [54] were purchased from the Jackson Laboratory . Ppargfl/fl mice [55] originally kindly provided by P . Chambon ( Université Louis Pasteur , Illkirch Cedex , France ) and backcrossed for 6 generations to C57BL/6 before crossing to CD11c-Cre [56] mice in our facility to generate CD11c-Cre/Ppargfl/fl mice . All animals were housed in individually ventilated cages under specific pathogen free conditions at BioSupport AG ( Zurich , Switzerland ) and used for experiments at between 8 and 12 weeks of age . Mice were sacrificed by an overdose of pentobarbital sodium i . p . BAL was isolated by canalization of the trachea with a catheter . The lungs were flushed with 3×400 µl PBS and BAL cells were harvested by centrifugation . Lungs were digested with 2 mg/ml of type IV collagenase ( Worthington ) and 0 . 02 mg/ml DNaseI ( Sigma ) at 37°C for 45 minutes and subsequently passed through a 70 µm cell strainer . Multiparameter analysis was performed on a FACSCanto II or LSR Fortessa ( BD ) and analyzed with FlowJo software ( Tree Star ) . Monoclonal antibodies specific to mouse CD11c ( N418 ) , CD11b ( M1/70 ) , Ly-6C ( HK1 . 4 ) , Siglec-F ( E50-2440 , BD Biosciences ) , CD103 ( 2E7 ) , CD115 ( AFS98 , eBioscience ) , CD45 ( 30-F11 ) , CD45 . 1 ( A20 ) , CD45 . 2 ( 104 ) , CD4 ( GK1 . 5 ) , CD8α ( 53-6 . 7 ) , MHC class II ( M5/114 . 15 . 2 , eBioscience ) , Gr-1 ( RB6-8C5 , eBioscience ) , CD49b ( DX5 , eBioscience ) , CD69 ( H1 . 2F3 ) , TNF-α ( MP6-XT22 ) , IFN-γ ( XMG1 . 1 ) were purchased from Biolegend unless otherwise stated . Dead cells were stained using eFluor780 ( eBioscience ) . PE-conjugated peptide-MHC class I tetramers ( H-2Db/NP34 ) with the peptide NP34 ( NP366-374; ASNENMETM ) from the nucleoprotein of influenza virus A/PR/8/34 were generated as described [57] . For detection of intracellular NP expression , cells were incubated with a monoclonal mouse anti-influenza NP antibody ( HB-65 , homemade ) followed by staining with AF647-conjugated anti-mouse IgG ( Life Technologies Co . ) . Prior to all stainings , FcγIII/II receptors were blocked by incubation with homemade anti-CD16/32 ( 2 . 4G2 ) . Thymocytes were isolated from C57BL/6 mice and apoptosis was induced by exposure to 60 mJ UV radiation ( Spectrolinker XL-1500; Spectronics Corporation ) . After 2 h incubation at 37°C in IMDM+10% FCS , cells were labeled with 5 µM eFluor670 ( eBioscience ) according to the manufacturer's instructions , washed extensively with IMDM+10% FCS and delivered i . t . in PBS . 3 and 24 h after administration , efferocytosis by cells in the BAL and lung was assessed by flow cytometry . For mixed bone marrow chimeras , CD45 . 1+CD45 . 2+ mice were lethally irradiated ( 9 . 5 Gy , using a caesium source ) and reconstituted with 5–10×106 BM cells of a 1∶1 mixture of CD45 . 1+WT∶CD45 . 2+WT or CD45 . 1+WT∶CD45 . 2+Csf2rb−/− . Mice were analysed 10 weeks post-reconstitution . For 4-way Csf2−/−/WT bone marrow chimeras , WT and Csf2−/− mice were lethally irradiated and reconstituted with 5–10×106 Csf2−/− or WT BM cells . Mice were analysed 16 weeks post-reconstitution . Total BAL protein concentration was measured by BCA Protein Assay ( Thermo Scientific ) according to the manufacturer's instructions . For fatty acid analysis , BAL lipids were extracted as described by Moser et . al [58] using methanol-methylene chloride ( 3∶1 , v/v ) . Thereafter , two internal standards , deuterated C17:0 ( Cambridge Isotope Laboratories , Inc . , Andover , MA , USA ) and deuterated C22:0 ( Dr . Ehrenstorfer GmbH , Augsburg , Germany ) were added to the samples . Derivatization was done by adding acetyl chloride followed by an incubation at 100°C for 1 h . The reaction was stopped with 7% K2CO3 and the extraction of fatty acid methyl esters was performed with hexane . After centrifugation for 20 min at 2500 rpm , the hexane layer was dried under a continuous nitrogen stream and resuspended in heptane . For the fatty acid analysis , gas chromatography-mass spectrometry was applied using a Finnigan PolarisQ ion trap gas chromatography-mass spectrometry system ( Thermo Quest , Austin , TX , USA ) . For quantification of fatty acid methyl esters , the specific masses were extracted . Analytes were identified with authentic standards by comparison of their retention time and their mass spectrum . Influenza virus strain PR8 ( A/Puerto Rico/34 , H1N1 ) was originally provided by J . Pavlovic , University Zurich . The influenza strain PR8 NS1-GFP carrying a GFP reporter in the NS segment was kindly provided by A . García-Sastre [51] . Vaccinia virus WR was kindly provided by J . Mercer , ETH Zurich . For infections , the mice were anaesthetized and intratracheally inoculated with indicated doses of virus in 50 µl endotoxin-free PBS . To determine influenza virus titers in the lungs , samples were collected on various days after infection , homogenized and serially diluted with MDCK cells as described [59] . Infected cells were detected using a monoclonal mouse anti-influenza NP antibody ( HB-65 , homemade ) . For the depletion of AM , mice were treated with 100 µl clodronate liposomes i . t . 2 days prior to infection . Clodronate was a gift from Roche Diagnostics GmbH and liposomes were prepared as previously described [60] . Control mice were treated with PBS liposomes . For restimulation , 1 . 5×105 bone marrow-derived dendritic cells ( BMDC ) were incubated overnight with 1×106 pfu UV-inactivated PR8 virus in 96-well plates . BMDC were pulsed with 1 µg/mL NP147 ( Balb/c ) or NP34 ( C57BL/6 ) peptide for 2 hours before BAL , lung or LN cells from individual mice were added and restimulation was performed for 4–5 h in the presence of 2 µM monensin ( Sigma-Aldrich ) . After surface staining and formalin-fixation , intracellular cytokine staining was done in the presence of 0 . 5% saponin using anti-mouse TNF-α FITC and IFN-γ APC and analysed by flow cytometry . Serum or BAL fluid from indicated time points post-infection was measured for influenza HA-specific antibody levels . Ninety-six well plates ( Maxisorp; Nunc ) were coated with 5 µg/ml recombinant PR8 influenza virus HA ( a kind gift of M . Bachmann , Cytos ) in PBS overnight at 4°C . After blocking , serum and BAL fluid from individual mice were serially diluted and incubated at RT for 2 hours . Plates were washed and incubated with alkaline-phosphatase-labelled goat anti-mouse isotype-specific antibodies ( Southern Biotech Technologies , Inc . ) and developed using substrate p-nitrophenyl phosphate ( Sigma-Aldrich ) . Optical densities were measured on an enzyme-linked immunosorbent assay reader ( Bucher Biotec ) at 405 nm . The femoral artery was catheterized in anaesthetized ( 2% isoflurane in oxygen ) mice and the wound was locally anaesthetized by the application of 2% lidocaine before the cut was closed and the catheter was sewn to the thigh to be held in place . The application of isoflurane was stopped and mice regained consciousness and were kept restrained in a dark card tube while normally breathing room air for 10 min to equilibrate blood gas . Subsequently , 100 µL arterial blood was taken from the catheter and blood gas composition was measured on an ABL800Flex blood gas analyzer ( Radiometer , Denmark ) before mice were sacrificed . The lungs were removed , fixed in formalin and processed for Hematoxylin and Eosin ( H&E ) staining . Histological sections were evaluated according to general inflammation . Fetal CD45+ cells were sorted from the lungs of CD45 . 1+ wild-type E18 . 5 fetuses using a FACSAria IIIu ( BD ) . Neonatal Csf2rb−/− recipient mice were anesthetized using Isoflurane and 1×105 fetal cells were administered i . n . in 10 µl PBS . Reconstitution of AM in the BAL and lung was assessed by flow cytometry 6 weeks post-transfer and mice were used for infection experiments at 8 weeks of age . Lungs of naive or influenza-infected animals at d5 post-infection were processed as described and stained with eF780 , anti-mouse CD45 , CD11c , CD11b and Siglec-F . AM were sorted as eF780−CD45+CD11chighautofluorescencehighSiglec-F+ ( BD FACSAria IIIu ) and RNA was prepared using PureLink RNA Mini Kit ( ambion , Life Technologies Co . ) , amplified and hybridized on the Affymetrix Mouse Gene 1 . 1 ST . Mean values , SD , SEM , and Student's t test ( unpaired ) were calculated with Prism ( GraphPad Software , Inc ) . p<0 . 05 ( * ) , p<0 . 01 ( ** ) , p<0 . 001 ( *** ) , p<0 . 0001 ( **** ) .
|
Acute respiratory viral infections can cause severe morbidity and pneumonia in infected individuals . Alveolar macrophages and various subsets of dendritic cells have been implicated in innate immunity and induction of anti-viral T cell responses that contribute to host defense against influenza virus infection . However , their relative importance in protection from pathology and disease severity has never been compared side by side . In this report , we demonstrate that mice lacking alveolar macrophages succumb to infection with low dose influenza virus and vaccinia virus infection due to respiratory failure . In contrast , mice lacking lymphoid CD8α+ and lung CD103+ DCs survived and showed little if any differences in disease severity compared to infected wild-type mice . These results indicate that therapies supporting AM and lung function may be beneficial during severe respiratory viral infection .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"immunopathology",
"immune",
"cells",
"cell",
"biology",
"animal",
"cells",
"clinical",
"immunology",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"immunology"
] |
2014
|
Alveolar Macrophages Are Essential for Protection from Respiratory Failure and Associated Morbidity following Influenza Virus Infection
|
Animal venoms are theorized to evolve under the significant influence of positive Darwinian selection in a chemical arms race scenario , where the evolution of venom resistance in prey and the invention of potent venom in the secreting animal exert reciprocal selection pressures . Venom research to date has mainly focused on evolutionarily younger lineages , such as snakes and cone snails , while mostly neglecting ancient clades ( e . g . , cnidarians , coleoids , spiders and centipedes ) . By examining genome , venom-gland transcriptome and sequences from the public repositories , we report the molecular evolutionary regimes of several centipede and spider toxin families , which surprisingly accumulated low-levels of sequence variations , despite their long evolutionary histories . Molecular evolutionary assessment of over 3500 nucleotide sequences from 85 toxin families spanning the breadth of the animal kingdom has unraveled a contrasting evolutionary strategy employed by ancient and evolutionarily young clades . We show that the venoms of ancient lineages remarkably evolve under the heavy constraints of negative selection , while toxin families in lineages that originated relatively recently rapidly diversify under the influence of positive selection . We propose that animal venoms mostly employ a ‘two-speed’ mode of evolution , where the major influence of diversifying selection accompanies the earlier stages of ecological specialization ( e . g . , diet and range expansion ) in the evolutionary history of the species–the period of expansion , resulting in the rapid diversification of the venom arsenal , followed by longer periods of purifying selection that preserve the potent toxin pharmacopeia–the period of purification and fixation . However , species in the period of purification may re-enter the period of expansion upon experiencing a major shift in ecology or environment . Thus , we highlight for the first time the significant roles of purifying and episodic selections in shaping animal venoms .
Venom is an intriguing evolutionary innovation that is utilized by various animals for predation and/or defense . This complex biochemical cocktail is characterized by a myriad of organic and inorganic molecules , such as proteins , peptides , polyamines and salts that disrupt the normal physiology of the envenomed animal . Evolution of venom has been intensively investigated in more recently diverged lineages ( for simplicity , we refer to them as ‘evolutionarily younger’ lineages ) , such as advanced snakes and cone snails , which originated ~54 [1] and ~33–50 [2 , 3] million years ago ( MA ) , respectively . Several venom-encoding genes in these animals have undergone extensive duplications [4 , 5] and evolve rapidly under the influence of positive selection [6–10] . In contrast , the evolution of venom in most of the ancient lineages , such as cnidarians ( corals , sea anemones , hydroids and jellyfish ) , coleoids ( octopus , squids and cuttlefish ) , spiders and centipedes , remains understudied , if not completely overlooked . Perhaps the only exhaustively investigated ancient venomous clade are the scorpions , which originated in the Silurian about 430 MA [11 , 12] . Moreover , certain potent toxins in species separated by considerable geographic and genetic distance can exhibit remarkable sequence conservation ( Fig 1 ) . Yet , research to date has solely focused on how positive selection has expanded the venom arsenal , while completely ignoring the role of negative ( purifying ) selection . Phylum Cnidaria consists of animals such as sea anemones , jellyfish , corals and hyrdroids that originated in the Ediacaran Period , approximately 600 MA [13–15] . They are characterized by unique stinging organelles called nematocysts with which they inject venom . Cnidaria represents the oldest venomous lineage known and includes some of the most notorious animals , such as the sea wasp ( Chironex fleckeri ) , a species of box jellyfish . Coleoids , which first appeared in the Early Devonian 380–390 MA [16] , represent yet another neglected lineage of ancient venomous animals . Although the venomous nature of coleoids was established as early as 1888 [17] , their venoms have received scant attention from toxinological research [17–20] . Centipedes are amongst the oldest living terrestrial venomous animals , with the fossil record extending back to ~420 MA [21] . All ~3 , 300 species of centipedes ( class Chilopoda ) described to date belong to five extant orders: Craterostigmomorpha , Geophilomorpha , Lithobiomorpha , Scolopendromorpha and Scutigeromorpha . They inject venom into victims via modified first pair of trunk limbs ( forcipules ) and use venom for predation and defense . Venoms of certain centipedes can cause excruciating pain , paresthesia , edema , necrosis [22 , 23] and can be fatal to mammals as large as dogs [24] . Yet , only a handful of centipede toxins have been pharmacologically characterized to date . Similarly , despite their remarkable ability to target a diversity of ion channels , only toxins from certain medically significant species of spiders have been investigated to date [25] . Thus , the evolutionary history and phyletic distribution of venom from these aforementioned ancient lineages , which represent the first venomous animal groups , remain understudied [18 , 26–28] . It should be noted that the divergence times of these lineages can be safely assumed to be equivalent to the time of origin of venom in those respective lineages , as all of the examined lineages are ( i ) venomous , ( ii ) do not share between them a common venomous ancestor , and/or ( iii ) for most of them the fossil data clearly indicates the presence of a venom delivery apparatus [29–36] . By examining a large number of nucleotide sequences from a diversity of species , we report for the first time the molecular evolutionary histories of a number of venom protein families in centipedes , spiders and Toxicofera ( clade of venomous snakes and lizards ) lizards . In contrast to the rapid evolution of venom in evolutionarily younger lineages , we report an unusually high conservation of venom in centipedes and spiders , despite their long evolutionary histories . Moreover , molecular evolutionary assessments of toxin-encoding genes distributed across the tree of life , has unraveled a surprisingly strong influence of negative selection on the venoms of ancient animals . Our findings reveal contrasting trajectories of venom evolution in ancient and evolutionarily young clades , and emphasize the significant roles of purifying and episodic selections in shaping animal venoms . Further , these results enabled the postulation of a new model of venom evolution that captures their evolutionary dynamics , and the rise and fall in evolutionary rates of animal venoms .
Despite the fact that several centipede and spider toxins are capable of exhibiting a diverse array of pharmacological effects , their venoms remain poorly studied . To date , very few studies have examined the evolutionary mechanisms responsible for the diversification of toxins in centipedes [27 , 28] and spiders [37–40] . Hence , we assessed the molecular evolutionary regimes of 17 and 10 gene families encoding toxins in the major lineages of centipedes and spiders , respectively . We computed the ratio of non-synonymous ( dN ) to synonymous ( dS ) substitutions , called omega ( ω ) , where ω greater than , less than or equal to one is characteristic of positive , negative and neutral selection , respectively . A large proportion of centipede venoms are characterized by β-pore-forming toxins ( β-PFT ) that are similar to aerolysins and epsilon toxins from bacteria [28] . They are theorized to be responsible for myotoxic and edematogenic activities of centipede venoms [23 , 28 , 41] . β-PFT has undergone substantial gene duplication and diversification in centipedes [28] . In contrast to venom-encoding genes in evolutionarily younger lineages that continue experiencing positive selection when they diversify via recurrent duplication events , we find that β-PFTs are evolutionarily extremely constrained under negative selection , as indicated by ω smaller one ( Table 1 ) . Centipede venoms are also chiefly constituted by cysteine-rich secretory proteins , antigen 5 , and pathogenesis-related 1 ( CAP ) family members and toxins with low-density lipoprotein receptor Class A ( LDLA ) repeats [28] . While certain CAP proteins in the venoms of centipedes are characterized by trypsin inhibitory activities [42] , the precise role of LDLAs remain unknown . We found that both these toxin classes have experienced a strong influence of purifying selection ( Table 1 ) . Scoloptoxin ( SLPTX ) is a family of cysteine-rich peptides found in the venoms of several centipedes , where different members exhibit a diversity of pharmacological activities [28] . SLPTX1 appears to be similar to insect peritrophic matrix proteins and has been theorized to be one of the most basally recruited toxins in centipedes [28] . Despite its long evolutionary history , this toxin exhibited lower-levels of sequence variations due to the influence of negative selection ( Table 2 ) . SLPTX10 and SLPTX15 families were reported to have undergone a functional radiation , where members exhibit neurotoxicity by targeting various voltage-gated ion channels: calcium ( Cav ) , potassium ( Kv ) and sodium ( Nav ) ion channels [28] . Similarly , certain SLPTX11 family members are known for their anticoagulant and Kv channel inhibitory activities [28 , 43] . We found that even these putatively potent toxins in centipedes were extremely well conserved under the influence of purifying selection ( Table 2 ) . While SLPTX family 13 appears to have convergently adopted an inhibitory cysteine knot ( ICK ) scaffold , which is characteristic of various potent toxins from scorpions and spiders , SLPTX16 has adopted a Von Willebrand factor type C ( VWC ) -like domain . These peptides were highly conserved despite their putative role in prey envenoming and long evolutionary histories . Certain taxonomically restricted toxin families , called ‘novel families’ were recently reported in centipede venoms [28] . Only one ( ‘novel family 6’ ) amongst the four of these novel families examined was found to have evolved rapidly , while the rest were negatively selected ( Table 3 ) . Overall , centipede venom-encoding genes were found to have evolved under the heavy constraints of purifying selection ( Fig 2A ) . Spiders are known to have originated 416–359 MYA in the Devonian [44] . All spiders , with the exception of a few species , employ venom for predation . However , toxinological research to date has solely focused on characterizing venom from the medically significant species of spiders . Yet , venom from only 0 . 4% of the currently cataloged spider species have been characterized to date [25] . We determined the rate of evolution of several venom protein superfamilies in a diversity of spider lineages , such as the lethal latrotoxins secreted by widow spiders [Theridiidae: 223–180 MYA [45]]; Kunitz-type serine protease inhibitors and huwentoxins from tarantulas [Theraphosidae: 250–200 MYA [46]]; the magitoxin family from tarantulas and certain funnel-web spiders [Hexathelidae: 250–200 MYA [46]]; sphingomyelinase-D ( SMase D ) in the medically significant venoms of recluse spiders [Sicariidae: 145+ MYA [46]]; lycotoxin family [47] from wolf spiders [Lycosidae: ~120+ MYA [46]]; and super family E ICKs [48] secreted by tarantulas and brushed trapdoor spiders [Barychelidae: 250–200 MYA [46]] . These venom proteins are secreted in large amounts by the respective spider lineage and are known for a diversity of biochemical activities , such as insecticidal presynaptic neurotoxicity and the ability to stimulate neurotransmitter secretions [latrotoxins: [49 , 50]] , dermonecrotic properties [SMase D: [51]] , Nav channel targeting capability–with some members additionally capable of targeting Cav channels [huwentoxin-1 family: [52 , 53]] , serine protease inhibition and the ability to block Kv channels [Kunitz toxins: [54]] , insect Nav channel targeting [magi-1 family [55]] , insect Cav channel targeting [ω-hexatoxins: [56]] , insect Calcium activated potassium channel ( KCa ) targeting [κ-hexatoxins: [57]] , and the Nav modulation and Cav blocking capabilities [Super Family E ICKs: [48]] . The computed ω values suggested a greater influence of purifying selection on nine out of ten toxin families examined , highlighting the slower evolution of spider venoms ( S1 Table; Fig 2B ) . Computed ω values for the vast majority of venom-encoding genes in all ancient lineages examined in this study and in previous studies [18 , 26 , 38 , 58] , highlighted the significant role of negative selection , which was in stark contrast to those of evolutionarily younger lineages , such as the advanced snakes and cone snails [S1 Table; [6 , 7 , 59–62]] . We also evaluated the molecular evolution of venom families from Toxicofera lizards that originated ~166 MYA [63] , and thus represent an intermediate state between ancient and recently originated lineages . Although , relative to advanced snakes , these lizards do not rely on venom for predation or defense to the same degree [6] , the evolutionary rates of some of their largely secreted venom proteins exhibited rapid evolution as demonstrated by their high number of positively selected sites ( Fig 3; S1 Table ) . Three ( Kallikreins , CRiSPs and crotamines ) among the six gene families examined exhibited an evidence for rapid evolution ( ω>1 and/or more than 10 positively selected sites ) , while the remaining were found to be extremely well conserved ( S1 Table ) . Further , we plotted site-wise ω against their respective amino acid position for each of the genes examined . Results indicated that a majority of sites in most venom proteins of ancient lineages evolved under the strong influence of negative selection ( Figs 2–5 ) . In contrast , a large proportion of sites in toxins of evolutionarily young lineages rapidly mutated under the significant influence of positive Darwinian selection ( Fig 6 ) . Thus , a stark difference was found in the evolutionary regimes of ancient and evolutionarily young lineages ( Fig 7 ) . Using the mixed effects model of evolution ( MEME ) several sites that experienced short periods of diversifying selection were also identified in all the examined venomous clades , which indicated that certain sites in these toxin proteins undergo episodic adaptation ( Tables 1–3; S1 Table ) . Considering the long evolutionary histories of these toxin types , we tested for nucleotide substitution saturation ( see methods ) . These tests did not detect saturation in any of the examined datasets ( S2 Table ) . We performed regression analyses to evaluate the possibility of the length of the toxin determining its rate of evolution by plotting ω values for various toxin types against their respective lengths ( S1 Fig ) . The coefficient of determination ( r2 ) for toxin types in each of the examined lineages suggested an absence of correlation between the length of the toxin and its ω value , indicating that venom proteins have undergone rapid evolution or extreme sequence conservation irrespective of their size . While most conotoxins are of relatively shorter lengths , snake venom components such as three-finger toxins ( 3FTxs ) and Snake Venom Metalloproteinases ( SVMPs ) are characterized by lengths of 80 and 600 amino acids , respectively . Despite such stark size differences , these toxins evolved rapidly . Similarly , several venom components in ancient lineages were characterized by a range of lengths . For example , most sea anemone and scorpion neurotoxins were of relatively shorter lengths ( 40–60 amino acids ) , while several pore-forming toxins were 450–550 amino acids long . Yet , these toxins were found to have evolved extremely slowly under the influence of purifying selection . Our results thus indicated that the stark differences in ω values for venom proteins of ancient and evolutionarily younger lineages did not result from the differences in size .
Animal venoms are assumed to rapidly diversify under the unabated influence of positive Darwinian selection . They have been theorized to undergo a chemical arms race with prey animals , where the evolution of venom resistance in prey and the invention of efficient toxins in the predatory venomous animal exert reciprocal selection pressure [64] , as postulated in the Red Queen hypothesis of Van Valen [65] . While the influence of positive selection is widely recognized , the role of purifying selection in shaping animal venoms has rarely been considered . Investigation of a large number of toxin-encoding gene families in this study has revealed a significant influence of negative selection on venom . Whilst positive selection increases the diversity of venom proteins , purifying selection probably aids in preserving the potency of the venom by filtering out mutations that negatively affect toxin efficiency . However , rare mutations that increase the potency of the venom arsenal ( e . g . , evolution of novel biochemical activity or increased binding efficiency ) are likely to be propagated and preserved in the population . In the absence of a conservatory evolutionary force , neutral or positive selection could modify key residues and result in the reduction of potency or , for worse , the complete loss of bioactivity , which could severely decrease the fitness of the animal . Thus , purifying selection pressure appears to be vital for sustaining the potency and , consequently , shaping the animal venom arsenal . It has been recently demonstrated that PFTs in Cnidaria , which bind to cell membranes and punch holes , evolve under the heavy constraints of negative selection [26 , 66] . The lack of variation in this group of toxins , which includes several unrelated toxin types ( e . g . , aerolysin-related toxins in sea anemones , independently recruited hydralysins in hydroids , actinoporins and jellyfish toxins ) , was theorized to be a result of their complex multi-subunit packaging [67] and their ability to attack highly conserved molecular targets , such as cell membranes [26] . Toxins that undergo oligomerization in other classes of animals have also been noted to evolve relatively slowly as a result of structural constraints like the need to conserve sites involved in subunit interaction . While most 3FTxs in snake venoms diversify rapidly , κ-3FTxs , which undergo dimerization , were found to accumulate relatively fewer variations [59] . Similarly , toxins that may function in a ‘non-specific’ manner may also experience negative selection . Here , non-specificity of action is defined as the ability to target regions in a structural/biochemical property dependent ( e . g . , surface electrostatic charge ) and target motif independent manner . For example , cytotoxic 3FTxs and β-defensin toxins—two very potent snake venom proteins , induce cytotoxicity by non-specifically binding to negatively charged cell membranes using hydrophobicity [68] and positively charged molecular surface [69] , respectively . As a result , unlike most snake venom components , these proteins remain evolutionarily constrained [59 , 62] . Similarly , scorpion lipolytic toxins were also theorized to be evolutionarily constrained because of their non-specific mechanism of action [58] . We found that β-PFTs in centipede venoms , which are similar to the aerolysin-like toxins , evolve under the significant influence of negative selection ( Table 1 ) . The lack of variation in this group of toxins may suggest that they either undergo oligomerization like their aerolysin homologues in other lineages or the possibility that they may employ a non-specific mechanism of action . A plot of site-specific ω against their respective amino acid positions reveals the extreme conservation of such toxin types that employ unique strategies for causing toxicity in envenomed animals ( S2 Fig ) . As it allows the targeting of a wide variety of animals , the strategy of exerting toxic action non-specifically or by targeting highly conserved molecular sites , appears to be advantageous and follows a contrastingly different evolutionary regime in comparison to toxins that specialize in attacking highly plastic molecular receptors . A comparison of evolutionary regimes of ancient and evolutionarily younger lineages suggests a fascinating strategy of venom evolution . When venomous animals venture into novel ecological niches , they encounter new types of prey and predatory animals . Consequently , in order to adapt and conquer niches , they would need to fine-tune venom proteins to efficiently target these new animals . Several sites detected as episodically adaptive—i . e . , sites that experience short bursts of adaptive selection , in these ancient clades may be reflective of such shifts in ecology . We propose that these earlier periods in the evolutionary history of a venomous species are accompanied by the significant influence of diversifying selection on the venom arsenal , which would expand the range of target sites and/or result in the origination of novel biochemical activities . This is particularly advantageous , since novel toxins generated may facilitate the efficient and rapid incapacitation of newly encountered prey and predatory animals . The period of expansion is followed by longer periods of purification , where the significant influence of negative selection preserves the potency of the toxin . Whenever there is a major shift in ecology or environment , the aforementioned stages of evolution repeat . Thus , we propose that venom-encoding genes mostly employ a ‘two-speed’ mode of evolution , where episodic diversifying selection accompanies the earlier stages of ecological specialization ( e . g . , diet and range expansion ) , resulting in the rapid diversification of the venom arsenal , followed by a longer period of purification and fixation that ensure the sustainability of venom potency . The low sequence variation in venom-encoding genes of ancient clades could be reflective of such long periods of purification and fine-tuning . In contrast , advanced snakes and cone snails , being evolutionarily very young , could still be undergoing the period of expansion and , consequently , exhibit a pronounced signature of positive Darwinian selection . However , it should be noted that the ‘two-speed’ model of evolution is likely applicable to venoms that serve predominantly predatory roles . Due to limited toxin sequence information from venoms that are employed for non-predatory functions ( e . g . , intraspecific competition in platypus , exclusively defensive roles in fishes , etc . ) , it remains to be seen whether they too follow our proposed evolutionary model . To conclude , in addition to unraveling the evolutionary regimes of toxin families in centipedes and spiders , which are amongst the first terrestrial venomous lineages , our findings highlight the pivotal roles of purifying and episodic selections in shaping animal venoms . Our findings enabled the postulation of a new theory of venom evolution in the animal kingdom that emphasizes the dynamic nature of these complex biochemical cocktails .
Toxin homologues were identified in the recently published genome of the coastal European centipede Strigamia maritima [70] by querying amino acid sequences of each toxin type against all six reading frames using the tblastn tool [71] . Translated nucleotide sequences were aligned using MUSCLE 3 . 8 [72] . The best-fit model of nucleotide substitution for individual toxin datasets was determined according to the Akaike’s information criterion using jModeltest 2 . 1 [73] and model-averaged parameter estimates were used for the reconstruction of trees . Phylogenetic trees were built using PhyML 3 . 0 [74] , where node support was evaluated with 1 , 000 bootstrapping replicates . Maximum-likelihood ( ML ) models [75] implemented in Codeml of the PAML package [76] were utilized to identify the influence of natural selection on toxin families[6] . As no a priori expectation exists , we compared likelihood values for a pair of models with different assumed ω distributions: M7 ( β ) versus M8 ( β and ω ) [77] . Only when the alternate model ( M8 ) shows a better fit than the null model ( M7 ) in the likelihood ratio test ( LRT ) , are its results considered significant . LRT is estimated as twice the difference in ML values between the nested models , and is compared with the χ2 distribution with the appropriate degree of freedom—the difference in the number of parameters of the two models . Further , we used the Bayes empirical Bayes ( BEB ) approach [78] in M8 to detect amino acids under positive selection by calculating the posterior probability ( PP ) that a particular site belongs to a given selection class ( neutral , conserved , or highly variable ) . Sites with PP ≥ 95% of belonging to the ‘‘ω > 1 class” are inferred to be positively selected . HyPhy’s [79] FUBAR approach [80] was used to detect sites evolving under pervasive diversifying and purifying selection pressures . MEME [81] was also employed to identify episodically diversifying sites . Sequence alignments used for selection assessments have been made available as a zipped file ( S1 File; see S3 Table for accession list ) . Nucleotide substitution saturation was tested using DAMBE 5 . 5 . 9 [82] using the recommended protocol [83] .
|
While the influence of positive selection in diversifying animal venoms is widely recognized , the role of purifying selection that conserves the amino acid sequence of venom components such as peptide toxins has never been considered . In addition to unraveling the unique strategies of evolution of toxin gene families in centipedes and spiders , which are amongst the first terrestrial venomous lineages , we highlight the significant role of purifying selection in shaping the composition of animal venoms . Analysis of numerous toxin families , spanning the breadth of the animal kingdom , has revealed a striking contrast between the evolution of venom in ancient and evolutionarily young animal groups . Our findings enable the postulation of a new theory of venom evolution . The proposed ‘two-speed’ mode of evolution of venom captures the fascinating evolutionary history and the dynamics of this complex biochemical cocktail .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[] |
2015
|
The Rise and Fall of an Evolutionary Innovation: Contrasting Strategies of Venom Evolution in Ancient and Young Animals
|
Spike-timing-dependent plasticity ( STDP ) has been observed in many brain areas such as sensory cortices , where it is hypothesized to structure synaptic connections between neurons . Previous studies have demonstrated how STDP can capture spiking information at short timescales using specific input configurations , such as coincident spiking , spike patterns and oscillatory spike trains . However , the corresponding computation in the case of arbitrary input signals is still unclear . This paper provides an overarching picture of the algorithm inherent to STDP , tying together many previous results for commonly used models of pairwise STDP . For a single neuron with plastic excitatory synapses , we show how STDP performs a spectral analysis on the temporal cross-correlograms between its afferent spike trains . The postsynaptic responses and STDP learning window determine kernel functions that specify how the neuron “sees” the input correlations . We thus denote this unsupervised learning scheme as ‘kernel spectral component analysis’ ( kSCA ) . In particular , the whole input correlation structure must be considered since all plastic synapses compete with each other . We find that kSCA is enhanced when weight-dependent STDP induces gradual synaptic competition . For a spiking neuron with a “linear” response and pairwise STDP alone , we find that kSCA resembles principal component analysis ( PCA ) . However , plain STDP does not isolate correlation sources in general , e . g . , when they are mixed among the input spike trains . In other words , it does not perform independent component analysis ( ICA ) . Tuning the neuron to a single correlation source can be achieved when STDP is paired with a homeostatic mechanism that reinforces the competition between synaptic inputs . Our results suggest that neuronal networks equipped with STDP can process signals encoded in the transient spiking activity at the timescales of tens of milliseconds for usual STDP .
Organization in neuronal networks is hypothesized to rely to a large extent on synaptic plasticity based on their spiking activity . The importance of spike timing for synaptic plasticity has been observed in many brain areas for many types of neurons [1] , [2] , which was termed spike-timing-dependent plasticity ( STDP ) . On the modeling side , STDP was initially proposed to capture information within spike trains at short timescales , as can be found in the auditory pathway of barn owls [3] . For more than a decade , STDP has been the subject of many theoretical studies to understand how it can select synapses based on the properties of pre- and postsynaptic spike trains . A number of studies have focused on how STDP can perform input selectivity by favoring input pools with higher firing rates [4] , [5] , with synchronously firing inputs [6] , or both [7] , detect spike patterns [8] and rate-modulated patterns [9] , and interact with oscillatory signals [10] , [11] . The STDP dynamics can simultaneously generate stability of the output firing rate and competition between individual synaptic weights [6] , [12]–[15] . In order to strongly drive the postsynaptic neurons , which we refer to as robust neuronal specialization . [16] . When considering recurrently connected neurons , the weight dynamics can lead to emerging functional pathways [17]–[19] and specific spiking activity [20] , [21] . Recent reviews provide an overview of the richness of STDP-based learning dynamics [22] , [23] . The present paper aims to provide a general interpretation of the synaptic dynamics at a functional level . In this way , we want to characterize how spiking information is relevant to plasticity . Previous publications [24] , [25] mentioned the possible relation between STDP and Oja's rate-based plasticity rule [26] , which performs principal component analysis ( PCA ) . Previous work [27] showed how STDP can capture slow time-varying information within spike trains in a PCA-like manner , but this approach does not actually make use of the temporal ( approximate ) antisymmetry of the typical STDP learning window for excitatory synapses; see also earlier work about storing correlations of neuronal firing rates [28] . Along similar lines , STDP was used to perform independent component analysis ( ICA ) for specific input signals typically used to discriminate between PCA and ICA [7] , [29] . In particular , STDP alone did not seem capable of performing ICA in those numerical studies: additional mechanisms such as synaptic scaling were necessary . On the other hand , additive-like STDP has been shown to be capable of selecting only one among two identical input pools with independent correlations from each other , also referred to as ‘symmetry breaking’ [13] , [17] . In addition to studies of the synaptic dynamics , considerations on memory and synaptic management ( e . g . , how potentiated weights are maintained ) have been used to relate STDP and optimality in unsupervised learning [30] , [31] . To complement these efforts , the present paper proposes an in-depth study of the learning dynamics and examines under which conditions pairwise STDP can perform ICA . For this purpose , we consider input spiking activity that mixes correlation sources . We draw on our previously developed framework that describes the weight dynamics [15] , [23] and extend the analysis to the case of an arbitrary input correlation structure . This theory is based on the Poisson neuron model [6] and focuses on pairwise weight-dependent STDP for excitatory synapses . Mutual information is used to evaluate how STDP modifies the neuronal response to correlated inputs [32] . This allows us to relate the outcome of STDP to either PCA and ICA [33] . Finally , we examine the influence of the STDP and neuronal parameters on the learning process . Our model captures fundamental properties shared by more elaborate neuronal and STDP models . In this way , it provides a minimal and tractable configuration to study the computational power of STDP , bridging the gap between physiological modeling and machine learning .
In order to examine the computational capabilities of STDP , we consider a single neuron whose excitatory synapses are modified by STDP , as shown in Fig . 2A . Our theory relies on the Poisson neuron model , which fires spikes depending on a stochastic rate intensity that relates to the soma potential . Each presynaptic spike induces variation of the soma potential , or postsynaptic potential ( PSP ) , described by the normalized kernel function , shifted by the axonal and dendritic delays , and , respectively ( Fig . 2B ) . The size of the PSP is scaled by the synaptic weight . We stress that the novel contribution of the present work lies in considering general input structures , i . e . , when the matrix of cross-correlograms is arbitrary . This extends our previous study [15] of the case of homogeneous within-pool correlations and no between-pool correlations , the matrix in ( 3 ) is diagonal ( by block ) . We focus on the situation where the average firing rates across inputs do not vary significantly . This means that rate-based plasticity rules cannot extract the spiking information conveyed by these spike trains . In this case , pairwise spike-time correlations mainly determine the weight specialization induced by STDP via , dominating rate effects lumped by in ( 2 ) . The key is the spectral properties of , which will be analyzed as follows: In order to illustrate the above analysis , we consider input configurations that give to “rich” matrices of pairwise correlations . Model input spike trains commonly combine stereotypical activity and random “background” spikes . Namely , to predict the evolution of plastic synaptic weights , it is convenient that the statistical properties of the inputs are invariant throughout the learning epoch ( e . g . , the presentation of a single stimulus ) . Mathematically , we require the input spike trains to be second-order stationary . In this way , the input firing rates and the spike-time correlograms in ( 2 ) are well-defined and practically independent of time , even though the spike trains themselves may depend on time . The formal definitions of and in Methods combine a stochastic ensemble average and a temporal average . This allows to deal with a broad class of inputs that have been used to investigate the effect of STDP , such as spike coordination [6] , [13] , [14] , [36] and time-varying input signals that exhibit rate covariation [12] , [27] , [45] , as well as elaborate configurations proposed recently [46]–[48] . Most numerical results in the present paper use the spike coordination that mixes input correlation sources . In the last section of Results , rate covariation will also be examined for the sake of generality . This first application shows how STDP can perform PCA , which is the classical spectral analysis for symmetric matrices . To do so , we consider input pools that have multiple sources of correlated activity , which gives within-pool and between-pool correlations . In the example in Fig . 5A , inputs are partitioned into pools of 50 inputs each that have the same firing rate . Some pools share common references that trigger coincident firing as described in ( 11 ) : pools and ( from left to right ) share a correlation reference with respective correlation strengths and for the concerned inputs; pools and share with ; and pools and share with . The overline indicates pool variables . All references correspond to coincident firing ( ) and the rate of correlated events is . The matrix is composed of blocks and given by ( 12 ) Each row of corresponds to a single correlation source here . We further assume that all synapses have identical kernels . Combining ( 11 ) and ( 3 ) , their covariance matrix reads ( 13 ) The covariance matrix in ( 13 ) is symmetric and thus diagonalizable , so it has real eigenvalues and admits a basis of real orthogonal eigenvectors . Here the largest real eigenvalue is isolated in Fig . 5B . The theory thus predicts that the corresponding spectral component ( solid line in in Fig . 5C ) dominates the dynamics and is potentiated , whereas the remaining ones are depressed . Numerical simulation using log-STDP agrees with this prediction , as illustrated in Fig . 5E . By gradually potentiating the correlated inputs , weight-dependent STDP results in a multimodal weight distribution that can better separate the mean weights of the pools . The final weight structure in Fig . 5F reflects the dominant eigenvector . Despite the variability of individual weight traces in Fig . 5D due to the noise in the weight update and rather fast learning rate used here , the emerging weight structure remains stable in the long run . Following the specialization induced by STDP , the modified weight distribution tunes the transient response to the input spikes . To illustrate this , we examine how STDP modifies the neuronal response to the three correlation sources in the previous configuration in Fig . 5 . Practically , we evaluate the firing probability during a given time interval of consecutive to a spike from input , similar to a peristimulus time histogram ( PSTH ) . Before learning , the PSTHs for ( red ) , ( green ) and ( blue ) are comparable in Fig . 6A , which follows because , and . After learning , pools and that relate to and are much more potentiated than pool by STDP in Fig . 5F . Consequently , even though pool is potentiated and transmits correlated activity from , the spike transmission after learning is clearly stronger for and than in Fig . 6B . The respective increases of the areas under the PSTHs are summarized in Fig . 6C . The overall increase in firing rate ( from about 10 to 30 sp/s ) is not supported equally by all . To further quantify the change in spiking transmission , we evaluate the mutual information based on the neuronal firing probability , considering correlated events as the basis of information . In this way , the increases in PSTHs are compared to the background firing of the neuron , considered to be noise . In contrast to previous studies that examined optimality with respect to limited synaptic resources [30] , [31] , we only examine how STDP tunes the transmission of synchronous spike volleys . We define with respect to the event ‘the neuron fires two spikes or more within the period ’ , denoted by ; is its complementary . Hereafter , we denote by and the occurrence of a correlated event and its complementary , respectively The mutual information is defined as ( 14 ) with and . The probabilities are defined for the events occurring during a time interval . We have and the realization of can be evaluated using a Poisson random variable with intensity ( 15 ) In the above expression , can be evaluated via the baseline firing rate for and the PSTHs in Fig . 6B for . Namely , we adapt ( 48 ) and ( 49 ) in Methods to obtain ( 16 ) Using the simulation results for the mean and we obtain the predicted values ( crosses ) for in ( 14 ) in Fig . 6D . They are in reasonable agreement with evaluated from the simulated spike trains after dividing the 100 s duration in bins of lengths . In a clearer manner than with the ratio in Fig . 6C , shows that the strong potentiation induced by STDP leads to the reliable transmission ( considering that Poisson neurons are noisy ) of the correlated events involved in the strong spectral component of , namely and , while that for remains poor . For the input firing rate used here , STDP potentiates the weights such that the postsynaptic neuron fires at after learning . Because the frequency of correlated events for each source is also 10 times per second , is not so high in our model . Perfect detection for corresponds to firing three spikes for each corresponding correlated event and none ? ? otherwise . In this case , , and , yielding the maximum . In comparison , for and the baseline log-STDP with ( results not shown ) , the firing rate after training is roughly eightfold that before learning . Then , for instead of about in Fig . 6 . For the Poisson neuron especially , high firing rates lead to poor because of the noisy output firing rate . Performance can be much enhanced by using inhibition [9] , but we will not pursue optimal detection in the present paper . The high neuronal response to both correlation sources in Fig . 6C arises because pools , and in Fig . 5F exhibit strong weights , in a similar manner to the example with Oja's rule in Fig . 1D . However , it is possible to obtain a much better neuronal selectivity to either or , as illustrated in Fig . 6E–F for two distributions set by hand . The corresponding mean weights were chosen such that favors the desired correlation source compared to others under the constraint of positive weights; cf . in ( 12 ) and indicates the matrix transposition . We use mutual information as a criterion to evaluate whether kSCA resembles PCA or ICA [33] . Here the analysis of independent spectral component for the postsynaptic neuron means a strong response to only one correlation source in terms of . To separate correlation sources as in Fig . 6E–F , stronger competition between the synaptic inputs is necessary . When tuning the weight dependence of log-STDP toward an additive-like regime , input weights corresponding to the dominant spectral component are more strongly potentiated . This increase results in higher with in Fig . 7A2 for both and , as compared to in Fig . 7A1 . However , the neuron still responds strongly to in addition to , as indicated by the ratio between the respective . So long as STDP causes the weights to specialize in the direction of the dominant spectral component , pool is the most potentiated and the neuron does not isolate . Even for log-STDP with or add-STDP ( not shown ) , we obtain . This follows because of the positive input correlations used here . We need a mechanism that causes inputs excited by distinct correlation sources to compete more strongly to drive the neuron . The synaptic competition induced by a negative postsynaptic single-spike contribution satisfactorily increases the ratio in Fig . 7B–C compared to A ( except for B1 ) . One drawback is that the larger negative is , the smaller the mean equilibrium weight become , cf . ( 6 ) . Consequently , even though the ratio increases , decreases and the transmission of correlations is weakened . To compensate and obtain sufficiently large weights after learning , one can use a positive presynaptic single-spike contribution . This gives both for and large ratios in Fig . 7D2–E2 , but not in Fig . 7D1–E1 . We conclude that , in order that the neuron performs ICA and robustly selects , STDP itself should also be sufficiently competitive to obtain robust selectivity , see Fig . 7B–E with compared to . By homogeneously weakening all weights after each output spike in addition to strong STDP-based LTP , only the inputs that most strongly drive the output firing remain significantly potentiated . In other words , introduces a threshold-like effect on the correlation to determine which inputs experience LTP and LTD . In agreement with our prediction , this “dynamic” threshold becomes more effective for large output firing rates , which only occurs when STDP leads to strong LTP ( ) . This is reminiscent of BCM-like plasticity for firing rates [49] . Note that we found in simulation ( not shown ) that using alone did not lead to ICA; this only increases the mean input weights . To further examine the effect of STDP parametrization and assess the generality of our analysis , we examine common trends and discrepancies in the weight specialization for different schemes for weight dependence for plain STDP: log-STDP [16] , nlta-STDP [13] , mlt-STDP [38] and add-STDP [12]; as well as the influence of single-spike contributions with log-STDP+SCC , nlta-STDP+SCC and add-STDP+SCC [6] . We consider the configuration represented in Fig . 8A where two sources of correlation excite three pools among four . The third pool from the left is stimulated by the same source as the second pool after a time lag of 20 ms . The corresponding spectrum of is given in Fig . 8C , leading to two dominant spectral components with equal real part , one for each correlation source . Due to the large imaginary parts of the complex conjugate eigenvalues related to , the final distribution in Fig . 8D does not reflect the green component in the sense that pool is not potentiated , but depressed . This follows because its correlated stimulation comes late compared to pool . Therefore , the weights from pool become depressed when the weights from pool become large . The final weight evolution differ from the initial splitting whereas both weight sets grew ( not shown ) , as expected by the theory . For log-STDP , the weight dependence regulates the number of selected components . Both red and green components are represented in in Fig . 8D , whereas the green component dominates in Fig . 8E . Nlta-STDP can also generate graded distribution as log-STDP does . The synaptic competition in Fig . 8G is comparable to that in Fig . 8E . In comparison , mlt-STDP induces weaker competition , although the asymptotic weights reflect the spectral components in Fig . 8I . On the other hand , add-STDP in Fig . 8J generates a bimodal distribution of weights , which is a thresholded version of Fig . 8D , E or G . In the case of add-STDP+SCC , the neuronal selectivity is controlled via the equilibrium mean weight that is constrained by the single-spike contributions in ( 6 ) . The situation is more complex for weight-dependent STDP+SCC , as the kernels is modified by as the weights evolve . Nevertheless , similar effects were observed in simulations . For log-STDP+SCC ( Fig . 8F ) and nlta-STDP+SCC ( Fig . 8H ) , the qualitative profile of the final weights is similar to that for plain STDP , with the additional competition induced by that depresses pool and favors , as was described in Fig . 7 . In the case of add-STDP+SCC , the instability of the dynamics leads to more sensitivity to the single-spike contributions . With and in Fig . 8K , only the weights from pool are potentiated at the end of the learning epoch . However , with and in Fig . 8L , the competition is weakened and all weights from pools and are potentiated , in agreement with the theoretical prediction . Interestingly , some weights from the uncorrelated pool are mildly potentiated , whereas those from the positively correlated pool are more strongly depressed toward zero because of the time lag associated to . Now we examine how the postsynaptic response affects the weight competition . This turns out to be particularly important when the correlograms have a temporal extension , that is , richer than just narrowly correlated inputs with a peak at . We consider the configuration in Fig . 9A where inputs from the pool tend to fire a time lag before those of pool . Namely , correlation is generated following ( 11 ) using a reference with , , and . Pool has no correlation . The matrix in ( 3 ) averaged over pools is not symmetric: ( 17 ) Following ( 4 ) , the PSPs and delays affect the kernel ( here identical for all synapses ) , hence and the resulting weight selection . In Fig . 3C , the same STDP learning window is combined with different PSP kernels and synaptic delays . We first use the baseline parameters in Fig . 9B1: a rise constant and a decay constant for the PSP kernel and purely axonal delays . They correspond to the blue curve in Fig . 3C . In this case , the matrix may be rather antisymmetric ( outside its diagonal ) : ( 18 ) cf . the values of the blue curve indicated by the arrows in Fig . 3C . The eigenvalues are represented in Fig . 9B2 . This indicates that the ( correlated ) pool fires “late” with respect to pool , from the point of view of STDP . It follows that the second pool is depressed while the first pool is potentiated , as illustrated in Fig . 9B3 . In contrast , a different weight selection occurs for the same axonal delays , but longer PSP time constants in Fig . 9C: , ( the purple curve in Fig . 3C ) ; as well as dendritic delays with the same short PSP time constants in Fig . 9D ( green curve in Fig . 3C ) . In both cases , this follows because has the following form: ( 19 ) which is “more” symmetric compared to Fig . 9B , and thus does not depress the late pool . The change in affects the spectrum , which results in the potentiation of both correlated pools and , as illustrated in Fig . 9C2 and D2 . For the case of a delay of the dendritic delay in Fig . 9D3 , the late pool is more strongly potentiated than the early pool as corresponds to the peak of the kernel , cf . the right arrow and the green curve in Fig . 3C . This illustrates that the effect of pool on the output firing felt at the synapse ( i . e . , after twice the dendritic delay ) coincides with the firing of pool , namely after pool .
For pairwise STDP , the weight dynamics can be predicted provided the firing rates and pairwise cross-correlations are well defined . The corresponding expressions ( 35 ) and ( 36 ) in Methods highlight the separation of timescales between rate-based and spike-based effects , which is determined by the learning window function . Spike-time correlations arise when coordinated firing between neurons is consistently repeated over time , such as a repeating spatiotemporal pattern embedded in random spiking activity and peaked PSTHs in experimental data . In the correlation structure induced by such pattern presentations , strong spectral components correspond to dense and peaked clusters of pattern spikes , in a similar fashion for both spike coordination and rate covariation [9] . Our framework can account for a rich variety of input configurations , in particular , stimuli that were used with STDP for categorization and/or representation in previous studies [8]–[10] , [50]–[52] , as well as recently proposed elaborate input configurations [46]–[48] . Time-varying signals can also generate significant spike-time correlations and thus weight specialization ( Text S2 and Fig S2 ) . The present framework aims to provide a unified description of the STDP dynamics for the many configurations that have been used in previous studies . Following the observations by Gerstner and Kistler [25 , Ch∼11] , STDP potentiates and depresses weights depending on the spectral components of . This matrix embodies the STDP-specific effects and is determined by the input correlation structure and kernels . The kernels are determined by the STDP learning window and PSP responses , cf . ( 4 ) . In a sense , the cross-correlograms in are “seen” by the neuron through the kernels . This is especially important when input correlograms have a temporal extension ( Fig . 9 ) or when the shape of the STDP learning window function varies across synapses . When using long timescales for PSPs with usual time constants for the learning window , the matrix tends to be symmetric and the PCA performed by STDP can result in slow-feature extraction [27] . Another point is that the input correlation structure as a whole determines for the weight specialization . In Fig . 8L for example , uncorrelated inputs are not as depressed by STDP as some positively correlated inputs . The present study has focused on Hebbian STDP for excitatory synapses ( Fig . 2C ) , but the same framework can be used for any arbitrary learning window , as well as the case of plastic inhibitory synapses [53] . A neuron can thus generate elaborate representations of the stimulating inputs in its weight structure , which illustrates the versatility of STDP . The present study has focused on STDP contributions up to the second order ( pairs of pre- and postsynaptic spikes ) and the learning dynamics that arise from the effect of pairwise spike-time correlations . This means that higher-order correlations only play a role via their collective second-order effects . In contrast , triplets or bursts of spikes can significantly modulate the weight updates in other models [40] , [60] . The model proposed by Appleby and Elliott requires multispike interactions ( i . e . , higher-order correlations ) for synaptic competition to emerge [41] . More elaborate STDP models also present advantages for spike computation and/or reproducing experimental data [7] , [31] , [34] , [40] , [61] , [62] . In addition to the effect of spike-time correlations considered here , some of these models are sensitive to firing rates . Likewise , when spike pairs contributing to STDP are restricted ( whereas all pairs are included in our model ) , the equilibrium mean weight depends on the input firing rates [4] , [5] and the balance between spike and rate effects is affected . Our results are expected to hold at least partially when pairwise effects dominate the STDP dynamics . Extending our study is left for subsequent work , but making use of higher-order correlations appears promising to perform ICA [58] . Although our STDP update incorporates noise , our analysis neglects it and assumes that the weight drift ( i . e . , mean change or first stochastic moment ) dominates the dynamics . In extreme cases , a fast learning rate can compromise the stability of the emerged weight structure [14] . The present analytical study is based on the “linear” Poisson neuron , which allows a tractable analysis . Its stochastic firing mechanism generates rather noisy and unreliable spike trains compared to deterministic neuron model where the stochasticity arises from the inputs , e . g . , integrate-and-fire neurons . Similar weight dynamics for both models have been demonstrated previously for slow STDP learning [6] , [13] . As mentioned above , a nonlinear firing response may be useful to perform ICA . In order to go beyond the linear input-output regime for integrate-and-fire neurons [63] , it is necessary to study how the neuron model shapes the input-output covariance; see ( 32 ) in Methods . In most neuron models , larger excitatory weights induce stronger input-output correlations for correlated inputs . This results in a positive-feedback loop for learning , which is captured by the Poisson neuron model . Dendritic integration of synaptic inputs are expected to bring interesting nonlinearities to the kernel defined in ( 4 ) . Moreover , depending on regional competition between and within dendritic branches [64] , [65] , different components can be represented in distinct areas of a single neuron . Including such refinements opens promising ways to understand spike-based computations . Finally , our results support the idea that neurons equipped with STDP can operate as self-adapting filters that process information based on the transient firing response of neurons . The input-output spike-time covariance ( in our model ) is simply the average of the transient response over all input statistics . STDP tunes these input-output correlations based on the input cross-correlation structure ( ) . Extending previous results focusing on a single correlated pathway [12] , Fig . 6 illustrates the modification of the transmission of coincidentally spiking activity using mutual information as a measure of signal-to-noise . This view is consistent with the hypothesis that the coordinated activity of cell assemblies can serve as a basis for the neuronal code [66] . In a more general scheme , spiking information should also consider the detailed shapes of the correlograms , not just their integral value as here . Because of the temporal dimension , coding using correlations appears richer than rate-based coding , as was observed in experiments [67] . Propagation of coordinated transient spiking activity , which can be seen as a generalization of PSTHs or spike patterns [9] , appears suitable for coding/decoding and naturally interacts with STDP . Depending on the more or less peaked shapes of the corresponding correlograms , the neurons may operate in either closer to a spike-based or a rate-based regime; these two forms of neuronal coding in feedforward networks are actually the two sides of the same coin [68] . Here correlations involve multiple input spike trains and all neurons belonging to the same assembly exhibit pairwise correlograms that have “coordinated” shapes , in a similar manner to cliques in graphs . Although a formal quantification has yet to be defined , the information in can intuitively be understood in terms of the diversity and arrangement of cross-correlograms . The kernels then define a “similarity measure” on matrices : the respective shapes of the correlograms and kernels determine the effective strength of spectral components . In a network , heterogeneity in the synaptic properties ( PSP response and delays ) and STDP learning windows leads to distinct kernels among the synapses , so neurons can extract different components from a common input correlation structure . This can be used by ( inhibitory ) STDP to extract the frequency of rhythmic neuronal activity [53] , which has been observed in many brain areas . Large inhomogeneities are expected to affect the weight specialization for oscillatory signals [10] , [11] , [45] , [69] . They may also play a role in encoding of slow signals at the shorter timescale of STDP [70] , [71] . Likewise , partial input connectivity allows neurons to see only part of the same global input structure , leading to differentiated specialization that may represent many spectral components . However , further developments are necessary to extend this analysis to the case of recurrent connections , which constrain the correlation structure [19] , and incorporate possibly plastic inhibitory connections . This theory aims to better understand how neurons can process spiking information in a distributed fashion [72] . Interesting applications have been proposed recently [73] , [74]: STDP can preprocess temporal signals within a recurrently connected network that act as a ( huge ) reservoir of functions of the inputs , which enhances the performance of the so-called liquid state machine [75] . Cessac et al . also showed that STDP can change the network activity such that observables ( e . g . , firing rates , spiking synchrony ) obey Gibbs distributions [76] . Together , these efforts will hopefully lead to novel interpretations on how neurons can process spike trains .
Pairs of pre- and postsynaptic spikes , as well as single spikes , completely determine the contributions to STDP . This choice has limitations compared to more elaborate models that include , for example , triplets or bursts of spikes in their analysis [40] , [69] or models for which pairwise correlations do not generate competition [41] . This choice allows us to focus on the next stochastic order after firing rates ( first order ) while keeping the analysis tractable . For a pair of pre- and post-spikes whose effects reach the th synaptic site at times and , respectively , the weight is modified by the following additive terms ( 20 ) In general , we assume that the STDP-specific update depends on the current value of the weight [13] , [38] , [39] , [42] , in agreement with experimental evidence [37] . This weight dependence alone can stabilize the weight distribution for ‘plain STDP’ , i . e . , without single spike contributions . However , in the absence of weight dependence , single-spike contributions are necessary to enforce partial stability on the weights , namely homeostasis on their mean [6] , [35] , [36] . Note that both mechanisms can also be successfully used together [15] . We will refer to the case where as ‘STDP+SSC’ , in contrast to ‘plain STDP’ ( or ‘STDP’ alone when there is no possible confusion ) for . The second case is often regarded as more biologically plausible for excitatory STDP and will be the focus on this work . Although their effect is not considered in detail , the weight update in ( 20 ) involves a learning rate , which determines the speed of learning , and variability in the pair-specific contribution , which is modeled by the white-noise random variable that has zero mean and variance . As mentioned above , the contribution specific to spike pairs depends on the relative timing of pre- and postsynaptic spiking activity felt at the synaptic site . For the synapse described in Fig . 2B , a pulse fired by the presynaptic neuron at time and a pulse fired by the postsynaptic neuron at time correspond to ( 21 ) Typically for excitatory STDP , leads to potentiation ( LTP ) and , conversely , to depression ( LTD ) . Thus , can be expressed as ( 22 ) Here decaying exponentials are used for illustration purpose . We compare several schemes for the weight dependence that is defined by the scaling function : Baseline parameters used in numerical simulations are recapitulated in Table 1 . The analysis is constrained to a single neuron excited by external inputs indexed by . The spike trains of the neuron and external input are denoted by and , respectively . We use a previously developed framework [6] , [36] to analyze the effect of weight-dependent STDP on the input plastic weights . The tractability of the present analysis relies on the condition that both the firing rates and covariances are quasi-invariant with respect to time ( but not for the time lag ) . We assume that learning occurs sufficiently slowly compared to the other neuronal mechanisms ( i . e . , PSP time constants and delays ) and that the noise is not too strong , such that the drift ( or first stochastic moment ) of the weight dynamics essentially determines the emerging structure [14] , [77] . Under this “adiabatic” assumption , the weight evolution can be described by ( 26 ) The weight update in ( 26 ) is the summation of two additive contributions . First , the rate-based contributions embodied by involve the time-averaged firing rates and for input and the neuron , respectively , cf . ( 35 ) . For weight-dependent STDP , it involves the integral value of the learning window ( as a function of the current weight ) ( 27 ) Second , the covariance coefficient incorporates the effect of the STDP on the time-averaged spike-time covariance between the neuron and input : ( 28 ) Note that the noise does not play a role in the weight drift evaluated here . In the Results section , we will show that the predicted weight specialization is valid even for a medium level of noise in STDP . In order to analyze the learning equation ( 26 ) , we need to evaluate the neuronal firing rate and covariance coefficients in terms of the input parameters . For this purpose we need to specify the neuronal firing mechanism . In the Poisson neuron model [6] , [13] , [14] , [19] , [36] , the neuronal spiking mechanism is approximated by an inhomogeneous Poisson process driven by an intensity function in order to generate an output spike-time series . A presynaptic spike from input induces a variation of referred to as the postsynaptic potential ( PSP ) , which is determined by the synaptic weight , the kernel function , and the sum of the axonal and dendritic delays . We require and , in order to preserve causality , for . For illustration purposes , we choose a double exponential function for all : ( 29 ) with rise and decay time constants , respectively . The “soma potential” sums the PSPs for all input spike times ( 30 ) Following ( 30 ) , we obtain the consistency matrix equations for the firing rates and spike-time correlations: ( 31a ) ( 31b ) Here and are -row vectors and a -column vector , whose elements are , and , respectively; bold capitals will be used for row vectors and bold lower-case characters for column vectors . The matrices have elements that correspond to pairs of inputs : ( 32 ) is reproduced in ( 3 ) . Note the respective roles of indices and . The input covariance is assumed to be quasi-independent of time , so in ( 32 ) only depends on through the weights , which slowly evolve due to STDP . The kernel functions in ( 33 ) describe the interplay between STDP and the postsynaptic response kernels that affects the weight dynamics: ( 33 ) The convolution indicated by concerns the variable . This equation is reproduced in Results , cf . ( 4 ) . This means that the postsynaptic response crucially determines the effect of synaptic plasticity [19] , [27] . In particular , the dendritic delay plays a distinct role compared to the axonal delay in that it shifts the kernel as a function of to the right , namely implying more potentiation for . Because of the weight dependence , the kernel is modified via the scaling of both potentiation and depression for when the strength evolves , as illustrated in Fig . 2C . The combination of ( 26 ) and ( 31 ) leads to ( 2 ) , where the dependence over time is omitted . The following expressions allow us to deal with general inputs while at the same time satisfying the requirement for mathematical tractability . We denote by the spike train ( Dirac comb ) of input . The corresponding time-averaged firing rate is defined as ( 34 ) and , for a pair of inputs and , the spike-time cross-covariance is given by ( 35 ) A double averaging is used in the above definitions: The separation of timescales implies that only correlations convey fast spiking information , whereas firing rates imply low-pass filtering . The covariance in ( 36 ) slightly differs from our previous framework [15] , [36] . It is actually the sum of two contributions: the stochastic covariance between the spike trains averaged over , which relates to ‘spike coordination’: ( 36 ) and the temporal covariance of the underlying rate functions , which we refer to as ‘rate covariation’: ( 37 ) For inputs generated using doubly stochastic processes [25] , a double ensemble average has been used in a similar fashion to the combination of ensemble average and temporal integration here . With our convention , the graphical interpretation of correlogram is that peaks for positive values of ( right side ) indicate that input tends to fire earlier than . For oscillatory inputs , if the closest peak to is on the right side , is phase-advanced compared to . Here we examine the conditions under which there exists at least one fixed point such that ( 2 ) for plain STDP vanishes for all coordinates , namely ( 38 ) where denotes the convolution of the correlation with the PSP kernels , reorganizing ( 33 ) . We make a couple of assumptions here: We define for sake of simpler notation the following functions such that ( 39 ) reads with ( 39 ) where denotes the whole set of weights . For each given , the sign of the first term in is given by alone and does not depend on : ( 40 ) where the circled signs indicate positive and negative values . The second term is zero for and for a sufficiently large , it becomes negative ( or barely positive ) with the assumptions that LTP vanishes and . For all , the sign of is given by and scales its modulus linearly: ( 41 ) Taken together , we have for an arbitrary small ( 42 ) and there exists a set of constants such that ( 43 ) These are sufficient conditions to prove that the expressions in ( 39 ) taken for all have at least a global fixed point . We first examine the illustrative case of weights . For any fixed , the expression of in ( 39 ) satisfies the two properties of being positive on the axis and becomes negative for large , following ( 43 ) and ( 44 ) . Consequently , for all , there is at least one zero of as a function of , and this zero is strictly positive and smaller than the upper bounds . Moreover , the expression in ( 39 ) for is continuous with respect to both and , so the zeros form a continuous curve . Reciprocally , by inverting the indices , there is a similar zero for . Because of continuity , there is at least one intersection point for the sets of the zeros as in Fig . 4A , which nullifies for and . In the general case of weights , the same properties in ( 43 ) and ( 44 ) ensure that , for each given , ( 39 ) is positive on the hyperplane and negative on . It follows that there is at least one zero for each , . Thus , the continuous surface that contains the zeros of ( 39 ) for a given contains a manifold of dimension . In the -dimensional hypercube , all such manifolds for have at least one intersection point , since the “constraint” for being on the -th manifold only concerns . On the non-empty intersection set , all derivatives vanish , meaning it consists of the fixed point ( s ) for the weight dynamics . The structure of these manifolds is actually simple and allows us to determined to the global stability of the fixed point ( s ) . For each , the corresponding manifold separates the hypercube into two subspaces . On the side containing , we have , whereas on the other side , . Each manifold is thus a global attractor for the coordinate , which guarantees global stability of their intersection set . The arrows in Fig . 4 illustrate the derivatives of and , which drive to its fixed point there . Now , for negative correlations , ( 43 ) or ( 44 ) may not hold anymore and the zero of ( 39 ) may become negative or even not exist for some values of . There is then no guarantee of a realizable global fixed-point , as illustrated in Fig . 4B . The analysis in this case will not be pursued here . A similar demonstration applies for STDP+SCC when is positive for and decreases with . This is the case when , for which LTP and LTD vanish at the upper and lower bounds enforced on the weights , respectively , in addition to . With the further condition that ensures a fixed-point for the mean weight , the equivalent to decreases when the output firing rate increases . Putting it all together , the existence of a fixed point is ensured for output firing rate that are not too high ( and positive correlations ) . In the early period of the weight evolution , we can approximate the weight vector as proportional to the dominant eigenvector ( s ) . Firstly , we consider the case of a single dominant eigenvalue , namely . The spike-based term of ( 2 ) can be rewritten ( 44 ) Decomposing for some initial factor and , the first term of the rhs is dominated by its component . This follows because is the largest eigenvalue . Now we further assume that the weight dependence is “weak” with respect to . By this , we require the second term of the rhs above to be dominated by the first term . Together , this means that the vector elements of are ordered as those of . The fixed point of the dynamics in ( 2 ) can be approximated by ( 45 ) We assumed earlier that the weight dependence is such that the -th component of is a decreasing of . The implicit relationship in ( 46 ) indicates that the fixed point of is given by the reciprocal function to applied on , which has its vector elements sorted in the same order as as we just explained . In other words , is expected to reflect the final weight distribution under the mentioned assumptions for a single dominant eigenvalue . In the case of two complex conjugate dominant eigenvectors , a large imaginary part for implies a strong rotation-like evolution even at the early stage: . In this case , the equilibrium weight distribution may significantly differ from the initial splitting in the direction of . As a illustrative example , we consider two weights with ( 46 ) This expression corresponds to the cases of time-lagged correlated inputs in Figs . 8 and 9 . When , and , has complex conjugate eigenvalues . Larger absolute values for and imply large imaginary parts . The spike-based effects on give . Starting from the homogeneous condition , it follows from that increases faster . If becomes so large that , STDP results in LTP for and LTD for at the end of the learning epoch . This means that , despite an initial growth in the case ( which is predicted by the eigenvectors ) , is eventually depressed . In the general case , we also expect that some weights may become depressed because others experience stronger LTP due to STDP . In any case , the most strongly potentiated weights at the initial splitting should eventually be the winners of the synaptic competition . Here we examine the spike transmission after learning , which is used to quantify mutual information in Results . To fix ideas , we present simple calculations for a neuron excited by a correlated pool of inputs with homogeneous weight and correlation strength . The firing probability during a given time interval of consecutive to a spike from input , similar to a peristimulus time histogram ( PSTH ) , can be seen as a measure of spike-based information transmission . It amounts to , which relates to the correlation term of , namely , evaluated for and rescaled by the spike rate of . For an isolated spike at time , i . e . , outside a correlated event such as that related to a reference in ( 11 ) , the above integral can be approximated by ( 47 ) Likewise , for a spike involved in a correlated event , the average increase of firing probability is scaled up by the mean number of coincidentally firing inputs: ( 48 ) When the neuron has many inputs and a non-zero background firing activity , the group effect dominates with , so we can neglect the term in in ( 48 ) . The ratio between ( 49 ) and ( 48 ) then becomes ( 49 ) To maximize this ratio , the optimal lies beneath the values for which most of the integral of is covered . Larger values for beyond the timescale of the PSP kernel ( e . g . , several hundreds of ms as used for rate-based coding ) lead to a smaller gain . With our parameters , we choose such that . The lower the equilibrium mean firing rate , the stronger this signal-to-noise ratio is . For the Poisson neuron , is also the variance of the firing rate , which can also be thought as a source of noise for rate coding . Note that from ( 26 ) with plain STDP , the equilibrium weight for a pool of instantaneously correlated inputs with strength satisfies , which gives a theoretical prediction of the expressions above . Text S1 . This section focuses on the situation where the spectrum of contains imaginary eigenvalues . For add-STDP , this can lead to an oscillatory-like behavior of the weights . In contrast , weight-dependent STDP stabilizes the weight distribution . Figure S1 . Example of quasi-periodic evolution for plastic weights modified by add-STDP . Text S2 . We show that spike coordination and rate covariation can induce correlations of similar strength . We consider a neuron stimulated by inputs that have a common periodic firing rate . We show how the weight evolution is determined by the frequency of the input rate , the postsynaptic response and the STDP learning window . Figure S2 . Example of weight evolution that depends on the frequency of oscillatory input firing rates . The postsynpatic neuron can be trained to represent only a certain frequency range , similar to a band-pass filter .
|
Tuning feature extraction of sensory stimuli is an important function for synaptic plasticity models . A widely studied example is the development of orientation preference in the primary visual cortex , which can emerge using moving bars in the visual field . A crucial point is the decomposition of stimuli into basic information tokens , e . g . , selecting individual bars even though they are presented in overlapping pairs ( vertical and horizontal ) . Among classical unsupervised learning models , independent component analysis ( ICA ) is capable of isolating basic tokens , whereas principal component analysis ( PCA ) cannot . This paper focuses on spike-timing-dependent plasticity ( STDP ) , whose functional implications for neural information processing have been intensively studied both theoretically and experimentally in the last decade . Following recent studies demonstrating that STDP can perform ICA for specific cases , we show how STDP relates to PCA or ICA , and in particular explains the conditions under which it switches between them . Here information at the neuronal level is assumed to be encoded in temporal cross-correlograms of spike trains . We find that a linear spiking neuron equipped with pairwise STDP requires additional mechanisms , such as a homeostatic regulation of its output firing , in order to separate mixed correlation sources and thus perform ICA .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"computational",
"neuroscience",
"biology",
"neuroscience"
] |
2012
|
Spectral Analysis of Input Spike Trains by Spike-Timing-Dependent Plasticity
|
Inflammatory monocytes can be manipulated by environmental cues to perform multiple functions . To define the role of monocytes during primary or secondary infection with an intra-phagosomal pathogen we employed Leishmania major-red fluorescent protein ( RFP ) parasites and multi-color flow cytometry to define and enumerate infected and uninfected inflammatory cells in the skin . During primary infection , infected monocytes had altered maturation and were the initial mononuclear host cell for parasite replication . In contrast , at a distal site of secondary infection in mice with a healed but persistent primary infection , this same population rapidly produced inducible nitric oxide synthase ( iNOS ) in an IFN-γ dependent manner and was critical for parasite killing . Maturation to a dendritic cell-like phenotype was not required for monocyte iNOS-production , and enhanced monocyte recruitment correlated with IFN-γ dependent cxcl10 expression . In contrast , neutrophils appeared to be a safe haven for parasites in both primary and secondary sites . Thus , inflammatory monocytes play divergent roles during primary versus secondary infection with an intra-phagosomal pathogen .
Immature bone marrow-derived monocytes are cells of the innate immune system that undergo maturation to populate numerous peripheral cell subsets [1–4] . Under different inflammatory and steady state conditions monocytes have been shown to acquire effector [5] , regulatory [6] , suppressor [7] , homeostatic [1] or repair functions [8 , 9] and can also prime T helper 1 ( Th1 ) adaptive immunity [10 , 11] . The recruitment of monocytes to sites of inflammation , their immature phenotype , and their plasticity suggests that these cells may be targets for infection and modulation by intra-phagosomal pathogens , such as Mycobacteria tuberculosis , Cryptococcus neoformans , Salmonella enterica , and Leishmania spp . [12] . Infection of mice with Leishmania major ( L . major ) is an established model to study inflammation and infection in the skin [13 , 14] . In nature , disease occurs when infected sand flies deposit parasites into the skin of a mammalian host during blood feeding , a process associated with significant tissue damage and inflammatory cell recruitment that is independent of the presence of the parasite . Once in the skin , Leishmania are predominantly engulfed by neutrophils , but are not killed . After 24–48 hours parasites transition into poorly defined CD11b+ mononuclear phagocytes , where they proliferate [15–17] . IFN-γ-producing T helper 1 ( Th1 ) CD4+ T cells mediate protective immunity against Leishmania infection by activating infected cells to produce nitric oxide ( NO ) and kill parasites [18 , 19] . Healed but persistent primary Leishmania infection , in which viable parasites are maintained at low levels for the life of an infected individual , mediates rapid immunity at a distal site of secondary challenge and is the gold standard of protective immunity in both mice and people [20–22] . Understanding the nature of this immunity is critical to developing an effective Leishmania vaccine . Remarkably , neither the phagocytic cell that mediates parasite killing during secondary challenge , nor the phenotype of the secondary host cell during acute primary infection , have been carefully defined in the skin . Rather , previous work has focused on the role of monocyte-derived cells late in primary infection , and/or employed an inadequate set of phenotypic markers at acute time points [5 , 10 , 11 , 15 , 17 , 22] . It is not known if the phenotype or effector function of infected inflammatory cells during primary or secondary infection differ , and whether or not this is related to infection outcome . In this study , we employed intra-dermal inoculation of L . major-RFP parasites and multicolor flow cytometry to track the fate of L . major and identify divergent roles for inflammatory monocytes during primary or secondary infection . These studies identify a critical early window during which protective immunity must act to prevent monocyte modulation and parasite expansion .
In order to define the phenotype of inflammatory cells during the transition of the L . major parasite from neutrophils to secondary phagocytic cells between approximately 1 and 4 days we initially employed CX3CR1+/gfp reporter mice ( Fig 1A–1C , and S1A Fig-gating strategy ) ; [15 , 23] . Prior to challenge , CD11b+Ly6ChiCX3CR1+ inflammatory monocytes were abundant in the blood but rare in the skin ( Fig 1A versus 1B , 0hr . ) . Following needle inoculation of L . major-RFP , CD11b+Ly6ChiCX3CR1+ inflammatory monocytes and Ly6CintCX3CR1-Ly6G+ neutrophils infiltrated the dermal site of infection ( Fig 1B ) . Ly6C clearly defined inflammatory monocytes at 10 hours post infection ( p . i . ) and these cells co-expressed CCR2 but were MHCII- and CD11clo ( S1B Fig ) . Analysis of the cellular infiltrate following exposure to sand fly bites , the natural mode of Leishmania inoculation , also revealed the robust infiltration of Ly6Chi inflammatory monocytes and neutrophils ( S2 Fig ) . Of note , this robust recruitment following exposure to sand fly bites is not dependent upon the parasite or pre-existing immunity . At 2 and 4 days p . i . , CX3CR1 was required to definitively track the CD11b+Ly6C+CX3CR1+ inflammatory monocyte population due to decreased expression levels of Ly6C ( Fig 1B; Ly6C MFI , 10hr 12634 +/-SD 1891 versus 2 days 4071 +/-SD 600; p<0 . 0001 , n = 6 ) . At 10h p . i . the vast majority of RFP+ infected cells ( S1C Fig ) were neutrophils , as previously shown ( Fig 1C; [15] ) . However , at 2 days p . i . , RFP+ cells were largely Ly6Cint/hiCX3CR1+ , strongly suggesting a transition of the majority of parasites into inflammatory monocytes . A similar pattern was observed at 4 days p . i . . We next extended the set of phenotypic markers to definitively track inflammatory monocytes according to a gating strategy outlined in Fig 1D and based upon previous observations [3] . Kinetic analysis employing counting beads to enumerate the inflammatory cell infiltrate into the ear revealed the total number of CD11b+ cells per ear increased significantly by 2 and 10 hours p . i . ( Fig 1E ) . This increase was attributable to CD11b+CD24+Ly6G+ neutrophil recruitment , which represented an increasing proportion ( Fig 1F , Population 1 ) and number ( Fig 1G , Neut . ) of total CD11b+ cells in the skin , and corresponded with a decrease in the proportion of CD64low-hiLy6C-CCR2-CX3CR1- dermal macrophages ( Population 6 ) , and pre-existing CD64-Ly6C- cells , that contain dermal DCs ( Population 7 ) ( Fig 1F ) . Following the initial recruitment of neutrophils , when neutrophils significantly outnumbered monocytes ( Fig 1G , 2h and 10h , p≤0 . 0001 , n = 6 ) , inflammatory monocytes , defined as CD11b+CD24-CD64low-hiLy6C+CCR2+ cells , which are virtually all CX3CR1+ ( Fig 1D , Population 2 ) , were recruited into the skin ( Fig 1F and 1G ) , likely from the blood ( S3A Fig ) where they pre-exist as MHCII-CD11c- cells ( S3B Fig ) . These cells increased dramatically in proportion ( Fig 1F ) and number ( Fig 1G ) between 10h and 2d post-infection and significantly outnumbered neutrophils on day 2 p . i . ( Fig 1G; p≤0 . 0001 , n = 6 ) . Kinetic analysis of the total number of RFP+ infected cells ( S1C Fig and S3C Fig ) revealed an increase in infected cell numbers due to neutrophil uptake of inoculated parasites at 2 and 10h p . i . ( Fig 1H and 1I; [15 , 17] ) . Between 10h and 2 days the phenotype of infected cells transitioned dramatically from 90 . 2% ( +/-SD 2 . 4 , n = 6 ) neutrophils to 83 . 5% ( +/-SD 1 . 1 , n = 6 ) inflammatory monocytes or putative inflammatory monocyte derived cells ( Defined as Ly6C+CCR2low/-CX3CR1+ , Population 3 , or Ly6C-CCR2+CX3CR1+ , Population 4 , cells ) ( Fig 1I ) , without a significant change in the total number of infected cells ( Fig 1H ) . This transition was also reflected in the absolute number of infected inflammatory cells ( Fig 1J ) , where RFP+ neutrophils significantly outnumbered RFP+ inflammatory monocytes at 10h ( p<0 . 0001 , n = 6 ) , but then dropped dramatically on day 2 p . i . , when the number of infected monocytes underwent a corresponding increase and significantly outnumbered infected neutrophils ( p = 0 . 0003 , n = 6 ) . Importantly , there was no significant difference ( p = 0 . 80 , n = 6 ) in the number of infected neutrophils at 10 hours p . i . ( Population 1: Mean 2890 +/-SEM 532 . 2 , n = 6 ) versus the number of infected inflammatory monocytes or recent inflammatory monocyte derived cells at 2 days p . i . ( Populations 2 , 3 , and 4: Mean 2718 +/-SEM 411 , n = 6 ) . This data suggests that L . major transitions from neutrophils to inflammatory monocytes without significant parasite expansion or elimination as determined by the number of infected cells . Infected neutrophils do not kill L . major at acute time points p . i . and inoculation of these infected neutrophils efficiently initiates disease [15] . In contrast , monocytes have been implicated in parasite killing following primary L . major infection of the intra-peritoneal ( i . p . ) cavity [24] . To determine the fate of parasites during the transition phase we first employed 2-photon intra-vital microscopy ( 2P-IVM ) to sequentially image the same site of L . major-RFP deposition in the skin following exposure to infected sand fly bites ( Fig 2A ) . Enumeration of RFP+ parasites employing a slice-by-slice visual analysis of the image z-stack revealed the total number of parasites per bite site and the number of infected cells did not significantly change until day 8 post-transmission , when it increased ( Fig 2B ) . In addition , the number of parasites per cell did not significantly change until day 6 of infection ( Fig 2A and 2B , right panel ) , which is consistent with a lack of parasite killing as well as a delay in parasite proliferation during neutrophil to monocyte transition . We also sorted infected CD11b+Ly6Cint/hiCX3CR1+Ly6G-RFP+ inflammatory monocytes from the skin 60h post in-vivo-infection ( Fig 2C ) and assayed parasite growth in-vitro . Cell sorting did not employ CCR2 , as optimal CCR2 staining required incubation at 37°C that reduced the yield of infected cells . Analysis of the total number of intra-cellular parasites per 25 RFP+ monocytes and the number of parasites per monocyte showed significant proliferation at 24 and 48h of in-vitro culture , demonstrating these cells are permissive host cells for L . major ( Fig 2D and 2E ) . To ensure our observations were not specific to murine monocytes we also infected elutriated human monocytes ( S4 Fig ) in-vitro . We observed robust expansion of L . major between 6 and 72h as assessed by the number of infected cells , parasites per single infected cell , and parasites per 100 total cells ( Fig 2F and 2G ) . Parasites grew despite triggering respiratory burst as assessed by Dihydrorhodamine 123 ( DHR ) ( Fig 2H and 2I ) . Remarkably , virtually every infected monocyte was positive for DHR ( Fig 2J ) suggesting these cells are activated , but fail to kill L . major . The ability of L . major to survive respiratory burst may be due to the fact that ROS have been shown to be excluded from parasite containing phagolysosomes in infected cells [25] . Therefore , inflammatory monocytes serve as a permissive host cell during acute primary infection in the skin . We next turned our attention to the role of inflammatory monocytes at secondary sites of infection . We hypothesized that rapid activation of infected immature inflammatory monocytes during neutrophil to monocyte transition is the effector mechanism by which healed but persistent primary infection mediates protective Th1 immunity [20 , 22] . Mice with a healed but persistent primary Leishmania infection where generated by inoculation of 104 L . major metacyclic promastigotes into the footpad [26] . As expected , mice with a persistent primary infection mounted a robust Th1 immune response at a secondary site of challenge in the ear ( LmLm group ) ( Fig 3A and 3B ) , coincident with significant control of parasite numbers versus mice with a primary infection ( NaLm group ) starting on day 4 ( Fig 3C ) . Evaluation of the absolute number of cells per ear revealed no difference in the total number of CD11b+ cells in LmLm versus NaLm mice at 10h post-challenge ( post-ch ) ( Fig 3D ) , and no difference in the number of neutrophils ( Fig 3E , left side ) , suggesting the acute neutrophil response at primary or secondary sites of infection is similar . In stark contrast , LmLm mice had a dramatic almost 10-fold increase in the number of CD11b+ cells on day 4 post-ch , correlating with the recruitment of IFN-γ+CD4+ T-cells into the ear ( Compare Fig 3D and 3A ) . This increase was attributable to huge increases in the number of inflammatory monocytes and CCR2low/- recent inflammatory monocyte derived cells ( Fig 3E ) . These differences were not due to an alteration in monocyte frequencies in the blood of naïve animals ( NaNa ) versus those with a healed lesion but persistent parasites ( LmNa ) prior to challenge ( S5A Fig ) . Neutrophil numbers were also maintained at secondary versus primary sites on day 4 and 8 post-ch . ( Fig 3E ) . Despite large numbers of IFN-γ producing CD4+ T cells in the skin , monocytes and monocyte derived cells had lower levels of MHCII expression at the population level during secondary infection compared to primary infection ( Fig 3F ) . This is likely due to the large and sustained numbers of infiltrating immature monocytes during secondary infection ( Fig 3E ) . Analysis of RFP+ infected cells during primary ( NaLm ) or secondary ( LmLm ) infection revealed NaLm mice had a faster transition of parasites from neutrophils to Ly6C+CCR2+ inflammatory monocytes between 10h and 2d ( Fig 3G ) as indicated by higher frequencies of infected monocytes on day 2 . This occurred despite a smaller proportion of inflammatory monocytes within the total CD11b+ population versus LmLm mice ( Fig 3H ) . However , by day 4 the frequency of total infected cells that were inflammatory monocytes was similar in both settings ( Fig 3G ) . While RFP expression does not definitively assess parasite viability , increased frequencies of RFP+ neutrophils at secondary versus primary sites at 2 and 4 days post-challenge suggests these cells are acting as a safe haven for parasites during secondary infection ( Fig 3G ) [15] . We also analyzed gene expression by RT-qPCR over time in NaLm and LmLm mice . As expected we saw an early ifnγ response at 10h post-ch in LmLm but not NaLm mice ( Fig 3I ) and this was associated with increased levels of the chemokines cxcl10 and cxcl9 , both of which are induced by exposure to IFN-γ , and in the case of CXCL10 , is involved in the recruitment of monocytes [27 , 28] . Peak cxcl10 expression occurred at 10h p . i . , correlating with subsequent monocyte recruitment . We also assessed inducible-nitric oxide synthase ( inos ) expression , the enzyme responsible for NO production . We found inos expression was not significantly upregulated until day 2 post-challenge in LmLm versus NaLm mice , suggesting the up regulation of inos expression is delayed until after the recruitment of IFN-γ producing CD4 T cells ( Fig 3I and 3J ) . In contrast , expression of tnfa , il17 , ccl2 , il10 , il4 , and ptgs2 , were not significantly different in NaLm versus LmLm mice . In addition , the low levels of ccl2 expression strongly suggest that the IFN-γ inducible chemokines CXCL9 and specifically CXCL10 , is the primary driver of monocyte recruitment at the secondary site of challenge . Therefore , at a site of secondary challenge inflammatory monocyte recruitment and infection correlated with parasite killing at day 4 p . i . . To determine the fate of parasites that transitioned into inflammatory monocytes during primary versus secondary infection we performed experiments as outline in Fig 4A . Infected and uninfected monocytes were sorted from ear skin 60h post in-vivo-infection as described in Fig 2E ( Fig 4B ) . Infected CD11b+Ly6Cint/hiCX3CR1+Ly6G-RFP+ inflammatory monocytes were virtually all CD64+ and CCR2int-high ( Fig 4C ) . Sorted RFP- and RFP+ monocytes from NaLm or LmLm mice were then subjected to limiting dilution analysis , in which a known number of cells underwent 2-fold dilution to determine the number of cells containing at least one viable parasite that was able to grow after 7–10 days of in-vitro culture in the titration plate . Our analysis revealed there was no difference between the number of RFP+ monocytes from NaLm mice with at least one viable parasite versus the 256 RFP+ cells plated in the assay ( p = 0 . 055 , n = 4 ) . In contrast , only approximately 4 of the 256 RFP+ monocytes from LmLm mice contained parasites capable of expanding in-vitro ( Fig 4D ) . Cytospin analysis of sorted RFP+ monocytes from LmLm mice directly ex-vivo confirmed that these cells do contain intact parasites ( Fig 4E , 0h time-point ) . However , unlike parasites from NaLm mice , these parasites fail to proliferate over a 48-hour period when plated in-vitro ( Fig 4E ) . Therefore , inflammatory monocytes from secondary sites of infection are not a permissive host cell for L . major . Infected monocytes from LmLm mice also contained fewer parasites per infected cell ( Fig 4E , 0h ) . This may be due to a lower in-vivo multiplicity of monocyte infection due to the slower transition of parasites into inflammatory monocytes at the 60h time point as suggested in Fig 3G . Alternatively , it may reflect increased parasite killing and/or suppression of parasite growth in LmLm mice prior to single cell sorting and analysis . Previous studies suggest CCR2+ monocytes undergo maturation to become dendritic cells during late primary infection [5] . We measured Ly6C and CCR2 expression on CD11b+CD24-CD64+CX3CR1+ putative monocytes on day 3 post-ch to determine the influence of infection and pre-existing immunity on monocyte maturation ( Fig 4F ) . Ly6C expression was much higher on cells from secondary versus primary sites of infection , likely due to the sustained recruitment of these cells in LmLm mice ( Fig 3E ) , while CCR2 expression was lower . Comparison of RFP+ versus RFP- cells revealed RFP+ infected cells had lower expression of Ly6C in LmLm mice , while in NaLm mice there was no change . CCR2 was reduced on RFP+ versus RFP- cells from both NaLm and LmLm groups but this was significantly enhanced in LmLm mice ( Fig 4F , right panel ) . Prior to challenge , these differences were not present on putative CX3CR1+ monocytes in the blood ( S5B Fig ) . Therefore , pre-existing immunity and infection status alter the phenotype of the inflammatory CD64low-highCX3CR1+ monocyte population in the skin , with secondary sites of infection containing monocytes with higher Ly6C expression and lower CCR2 expression . The lower levels of Ly6C and CCR2 on RFP+ infected versus uninfected monocytes during secondary infection suggest these cells are undergoing maturation . Subsequent analysis of Ly6C+CCR2+CX3CR1+/- monocytes on day 3 post-challenge , when monocytes represent the majority of infected mononuclear cells , revealed MHCII expression was significantly reduced on RFP+ versus RFP- monocytes from NaLm mice but this difference was lost by day 6 ( Fig 4G and 4H , far left panel ) , suggesting early modulation of monocyte maturation by Leishmania during acute primary infection . In contrast , RFP+ cells from LmLm mice had higher expression of MHCII versus RFP- cells or RFP+ cells from NaLm mice ( Fig 4H ) . CD11c and CD86 were increased on RFP+ cells in both NaLm and LmLm mice , however , CD86 was higher on RFP+ cells from LmLm mice , suggesting that infected monocytes at secondary sites of infection undergo enhanced maturation versus those at primary sites . While the frequency of monocytes that were CD11c+MHCIIhi monocyte derived dendritic cell was significantly higher among infected versus uninfected cells in LmLm mice ( Fig 4H , far right panel and I ) , the vast majority ( Mean 84 . 5% +/- SD 3 . 5% n = 6 ) had not matured to become CD11c+MHCIIhi DCs at this time-point as defined employing a DC gate based on CD11c and MHCII expression levels on dermal DCs in the CD64-Ly6C- population ( Fig 4J ) . Because Ly6C+CCR2+CX3CR1+ monocytes originate as MHCII-CD11c-CD86- immature cells in the blood ( Fig 4K; see also S5C Fig and S3C Fig ) , our data suggest that L . major actively prevents the maturation of inflammatory monocytes towards an MHCII+ phenotype at the site of infection , but that this does not occur at sites of secondary challenge . We next wished to determine the phenotype of iNOS+ cells at sites of secondary challenge ( Fig 5A ) . On day 3 post-ch , LmLm mice contained large numbers and frequencies of CD11b+iNOS+ cells ( Fig 5A , left panel , and Fig 5B and 5C ) . Phenotypic analysis revealed that CD11b+iNOS+ cells were almost exclusively inflammatory monocytes followed by a small proportion of CCR2low-neg recent inflammatory monocyte derived cells ( Fig 5D ) . Loss of CCR2 on CD64low-highLy6Chi monocytes was not associated with changes in the proportion of these cells that can make iNOS , suggesting maturation towards a CCR2- phenotype was not a prerequisite for iNOS production ( Fig 5E ) . iNOS+ inflammatory monocytes expressed higher levels of MHCII , CD11c , and CD86 versus iNOS- cells ( Fig 5F ) . However , iNOS production did not correlate with an increase in the proportion of cells that were CD11c+MHCIIhi DCs ( Fig 5F , lower right ) . As expected , NaLm mice contained large frequencies of RFP+iNOS- monocytes , while LmLm mice contained large numbers iNOS+ cells ( Fig 5G ) . Cells that were iNOS+ were 2 . 4 times as likely to be RFP+ versus iNOS- cells ( Fig 5H ) and , while RFP- monocytes did contain high frequencies of iNOS+ cells , the RFP+ infected population contained significantly more ( Fig 5I ) . Lastly , among iNOS+ monocytes , RFP+ cells had higher expression levels of iNOS versus RFP- cells ( Fig 5J ) , demonstrating that infected cells are both more likely to be iNOS+ and are activated to produce higher levels of iNOS . Therefore , iNOS production and infection are correlated . Of interest , analysis of the infected iNOS- phagocyte population revealed that neutrophils were once again the predominant RFP+ cell type , suggesting neutrophils are acting as a safe haven during secondary infection ( Fig 5K ) . Interferon-γ is the predominant effector cytokine associated with the activation of phagocytic cells to kill intra-phagosomal pathogens [12] . Treatment of LmLm mice with anti-IFN-γ Ab negated the control of parasite numbers on day 4 post-ch . ( Fig 6A ) and significantly reduced the frequency of CD11b+ per ear ( Fig 6B ) , although this did not reach significance when the absolute number of cells was compared ( Fig 6C ) . Importantly , treatment significantly reduced both the frequency ( Fig 6D ) and absolute number ( Fig 6E ) of Ly6C+CCR2+CX3CR1+/- inflammatory or Ly6C+CCR2-CX3CR1+ inflammatory monocyte derived cells in LmLm mice , demonstrating that IFN-γ drives a proportion of monocyte recruitment during secondary infection . Treatment of LmLm mice dramatically decreased the frequency of CD11b+iNOS+ cells to levels observed in naïve mice ( Fig 6F ) and reduced the absolute number of CD11b+iNOS+ cells 14-fold ( Fig 6G ) and the number of iNOS+ inflammatory monocytes 20-fold ( Fig 6H ) . Treatment also dramatically reduced the proportion of monocytes in the skin that were iNOS+ ( Fig 6I ) . Anti-IFN-γ Ab did not alter ifn-γ gene expression but did reduce the production of inos , cxcl10 , and cxcl9 ( Fig 6J ) , suggesting that CXCL10 production induced by IFN-γ is driving monocyte recruitment . Blockade of IFN-γ in LmLm mice resulted in MHCII down regulation on RFP+ monocytes to levels that were indistinguishable from NaLm mice and significantly lower than LmLm mice treated with control Ab ( Fig 6K ) . Similar to NaLm mice , RFP+ monocytes from anti-IFN-γ Ab treated LmLm mice also had lower percentages of CD11c+MHCIIhi expressing cells versus RFP- monocytes from the same site of infection , while control LmLm mice had increased frequencies ( Fig 6L ) . Continuous blockade of IFN-γ in LmLm mice resulted in a prolonged reduction in the frequency of iNOS+CD11b+ cells and enhanced parasite loads on day 7 p . i . ( S6A and S6B Fig ) . Under these conditions , immune mediated swelling significantly increased ( S6C Fig ) and neutrophils became both more frequent ( S6D Fig , left side ) and the predominant infected cell type ( S6E Fig ) . Increased neutrophil-mediated immuno-pathology under conditions of reduced monocyte recruitment has been reported previously following Toxoplasma gondii infection in the gut [6] . These observations demonstrate that IFN-γ drives a proportion of the recruitment of Ly6C+CCR2+ monocytes to the skin and the expression of MHC II and iNOS by these cells . We next employed CCR2-diptheria toxin receptor ( DTR ) mice in which CCR2 expressing cells also express the DT receptor to formally demonstrate the role of CCR2+ cells at secondary sites of infection ( Fig 7A ) . Analysis of the skin and blood revealed short-term DT treatment efficiently eliminated CCR2 expressing cells ( Fig 7A ) . Following DT treatment we observed highly significant decreases in the total number of CD11b+ cells per ear in both NaLm and LmLm mice 7 days post-challenge ( Fig 7B ) . These decreases were largely attributed to the absence of Ly6C+CCR2+ inflammatory monocytes , and Ly6C+CCR2- inflammatory monocyte-derived cells , but also extended to other cell types ( Fig 7C ) , confirming that during the inflammatory response at a dermal site of infection , CCR2+ monocytes undergo maturation and populate these cellular niches . As expected , the depletion of CCR2+ cells in DT treated mice resulted in a corresponding increase in the frequency of other CD11b+ cell types in the skin versus non-depleted mice , most notably neutrophils ( S7A Fig ) . Depletion of CCR2+ cells also led to a dramatic decrease in the frequency and total number of iNOS+ cells in the ear of LmLm mice ( Fig 7D and 7E ) and a loss of parasite killing ( Fig 7F ) . The decrease in the number of iNOS+ cells was most dramatic in the Ly6C+CCR2+ and Ly6C+CCR2- populations ( S7B Fig , note log axis ) , confirming that CCR2+ inflammatory monocytes or inflammatory monocyte derived cells are largely responsible for iNOS production and parasite killing at a site of secondary infection . Remarkably , DT treatment also led to a dramatic decrease in the frequency of skin-derived CD4+ T cells with the capacity to produce IFN-γ upon ex-vivo antigen-restimulation ( Fig 7G ) , an observation that was reflected by the absence of immune mediated swelling at the challenge site ( Fig 7H ) . To investigate this further we employed direct intracellular staining ( dICS ) [20] and counting beads to enumerate the total number and number of in-situ IFN-γ producing CD4+ T cells per ear ( Fig 7I–7K ) . We found a small but significant reduction in the frequency of in-situ IFN-γ+ cells at the challenge site of DT- versus PBS-treated LmLm animals; suggesting CCR2-derived cells play a role in antigen presentation in the skin ( Fig 7I ) . In addition , we observed a large and significant reduction in both the total number of CD4+ T cells per ear ( Fig 7J ) and the total number of CD4+IFN-γ+ cells per ear ( Fig 7K ) in LmLm DT-treated mice . This reduction was not due to off-target depletion of T cells in CCR2 . CFP-DTR mice following DT treatment as the number of CD4+ T cells in the dLN was the same in PBS versus DT treated groups ( S8A Fig ) . In addition , DT treatment of wild-type litter-mate controls did not alter the total number , or number of IFN-γ+ T cells in the ear versus PBS treated CCR2 . CFP-DTR mice ( S8B and S8C Fig ) , suggesting DT treatment did not alter the CD4 T cells response . Therefore , in addition to parasite phagocytosis and killing at a secondary site of challenge , monocytes are also important for down stream elicitation of adaptive immunity , as previously shown [10] . While we were initially surprised that parasites were able to maintain infection in the absence of monocytes acting as host cells following DT treatment , this is likely due to the compensatory infection of neutrophils in these animals , which appear to provide a safe haven for Leishmania ( S7C Fig ) , similar to our observation following anti-IFN-γ treatment ( S6E Fig ) . Therefore Ly6C+CCR2+ inflammatory monocytes are the primary cell involved in parasite killing at secondary sites of challenge .
Following infection with intra-phagosomal pathogens , inflammatory cell recruitment may favour the establishment of disease due to the provision of permissive host cells [12 , 16] . In the context of Leishmania infection , where the majority of parasites initially infect neutrophils , immature inflammatory monocytes would appear to be an ideal secondary target cell versus pre-existing mature tissue macrophages and dendritic cells . We were able to define CD11b+CD24lowCD64+Ly6C+CCR2+CX3CR1+ inflammatory monocytes as the preferred mononuclear host cell for L . major following the initial neutrophil phase of infection . This contrasts our previous data that relied largely on Ly6C expression and reported only 10–20% of infected cells were inflammatory monocytes at early time points [17] , emphasizing the need to use additional markers to track these cells . Following infection , parasite proliferation was delayed until the monocyte phase of infection , and sorted monocytes supported replication ex-vivo . Similar results were observed employing in-vitro infection of human monocytes . CD11b+CD64low-highLy6C-CCR2-Ly6G- macrophages or CD11b+CD64-Ly6C- DCs that , based upon their phenotype and presence in naïve ears prior to challenge are pre-existing resident cells , accounted for a remarkably small proportion of infected cells at early time points p . i . CX3CR1+Ly6C- ‘patrolling’ monocytes were also rarely infected . Inflammatory monocyte infection by L . major at day 3 post primary infection was associated with lower expression levels of MHCII versus uninfected cells . Because inflammatory monocytes originate as MHCII negative cells in the blood , this strongly suggests that infection prevents their maturation to MHCII+ cells , contrasting the more conventional macrophage-centric model in which infection down-regulates MHCII+ expression in already mature cells . A potential implication of these observations is that in-vitro infection of monocytes , rather than matured macrophages , may be a more relevant model with which to study phagocyte modulation by the Leishmania parasite . A common property of intra-phagosomal pathogens is their relatively slow rate of replication and dependence on the priming of CD4+ ‘Th1’ cell-mediated adaptive immunity for host resistance [12] . Despite considerable knowledge around the requirements for efficient Th1 CD4 T cell priming and memory cell generation , effective vaccination against any intra-phagosomal pathogen has been elusive [29–32] . The results presented here suggest a rapid response is required to counteract the modulation of immature inflammatory monocytes by Leishmania . At sites of secondary infection , iNOS+ cells were exclusively inflammatory monocytes or very recently inflammatory monocyte derived cells , not tissue resident cells . While iNOS production , increased MHCII expression , and maturation to a DC phenotype was IFN-γ dependent , the vast majority of iNOS+ cells maintained a largely immature phenotype , contrasting with observations in the dLNs of chronic sites of primary infection where iNOS production is associated with a CD11c+MHCIIhi mature DC phenotype [5] . Therefore , monocytes can acquire functionality without differentiation to a macrophage or DC phenotype in-vivo , similar to observations describing their role in steady state antigen transport [1] . Infected monocytes purified from dermal sites of secondary infection contained intact parasites that failed to expand upon in-vitro culture . These observations are in agreement with those of Muller et . al . [33] , in which iNOS production was shown to suppress parasite metabolism without necessarily resulting in direct parasite killing . In addition , while we found significant correlation between infection and iNOS production within monocytes during secondary infection , we also found high frequencies of iNOS+ monocytes within the RFP- population; supporting the idea that IFN-γ can also induce iNOS production by uninfected bystander cells [34] . Our results mirror those reported by Reiner et al . [35] in which pre-exposure of human monocytes to IFN-γ prevented Leishmania-induced inhibition of monocyte activation and extend the idea that CD4 T cell-mediated immunity is a major factor in shaping innate inflammation [36 , 37] . Our observations also bear similarity to studies examining the role of vaccine-induced immunity in shaping the innate inflammatory response in the spleen and liver or stomach , specifically the activation of inflammatory monocytes as a strong correlate of protective immunity [38 , 39] . CCR2-depleter mice allowed for the timed removal of monocytes following primary infection but prior to secondary challenge . Our results demonstrate that blood-derived CCR2+ monocytes are the definitive source of iNOS+ cells and are essential for rapid parasite elimination at a secondary site in infection . They also revealed the extensive degree to which these cells undergo maturation and populate numerous phenotypic niches [3] . Interestingly , we saw no evidence for parasite killing by neutrophils in the absence of monocytes and repeatedly found that neutrophils were acting as a safe haven for parasites , similar to their role in primary L . major infection . Our observations extend a model of disease whereby Leishmania parasitizes the innate inflammatory response in the skin following infection [15 , 40] . The need for a rapid immune response to condition monocytes immediately upon infection may explain the failure of conventional vaccination strategies that generate memory cells against intra-phagosomal pathogens . In these settings , the time required for the expansion and acquisition of effector function by memory cells residing in secondary lymphoid organs may be too long to counteract pathogen-mediated monocyte modulation .
Generation , maintenance and metacyclic promastigote purification of L . major ( MHOM/IL/80/Friedlin ) or L . major-RFP was performed as described previously [15] . Parasite loads were determined by limiting dilution analysis ( LDA ) . Briefly , two fold serial dilutions in 96-well flat bottom microtiter plates were performed by overlaying 100μl of the diluted tissue suspension onto 50 μl NNN medium containing 20% defibrinated rabbit blood . The dilutions were made in duplicate . The plates were scored microscopically for growth and the number of parasites in each tissue was determined from the highest dilution at which parasites could be grown out after 7–10 days incubation at 26°C . For two-fold limiting dilution of known numbers of sorted RFP+ infected cells , cells were plated in a multiple of two ( typically 256 cells in the first well ) in quadruplicate based on the number of available cells post-sorting . Female C57BL/6 WT and B6 . 129P ( Cg ) -Ptprca Cx3cr1tm1Litt/LittJ ( Jax strain 005582 , CX3CR1-GFP ) were originally obtained from Jackson Laboratories . CX3CR1-GFP mice were kindly provided by Drs . John Grainger and Yasmine Belkaid ( Laboratory of Parasitic Diseases , NIAID , Bethesda , MD ) or Dr . Paul Kubes ( Snyder Institute for Chronic Diseases , University of Calgary , Calgary , AB ) . cx3cr1gfp/+ mice ( B6 . SJL-Cd45a ( Ly5a ) /Nai x B6 . 129P ( Cg ) -Ptprca Cx3cr1tm1Litt/LittJ F1 ) were bred in house . C57BL/6 LYS-eGFP knock-in mice ( TAC 0342 ) were obtained through a supply contract between the National Institute of Allergy and Infectious Diseases ( NIAID ) and Taconic Farms . CCR2-depleter mice ( CCR2 . CFP . DTR mice ) were kindly provided by Dr . Eric G . Pamer ( Memorial Sloan Kettering Institute , New York , NY ) . All mice were bred and maintained at the University of Calgary Animal Resource Centre or at the NIAID animal care facility under specific pathogen-free conditions . Chronically infected mice were generated by injecting 104 L . major metacyclic promastigotes subcutaneously in the hind footpad in a volume of 40μl and used at 12 to 20 weeks post-primary infection when footpad lesions had completely resolved . Naïve mice and mice with a chronic primary infection , were challenged with 2x105 L . major-RFP metacyclic promastigotes intra-dermally ( i . d ) in the ear in a volume of 10μl . Two-to-four day old Phlebotomus duboscqi females were obtained from a colony initiated from field specimens collected in Mali . Flies were infected by artificial feeding through a chick skin membrane on heparinized mouse blood ( drawn intracardially from BALB/c mice ) containing 5 . 106 L . major promastigotes/ml of blood . Blood engorged flies were separated and maintained at 26°C and 75% humidity and were provided 30% sucrose ad libitum . After 13–14 days , 9–10 flies per experimental group were anesthetized with CO2 , killed in 5% soap solution , and whole midguts , including the crop , were dissected and transferred into tubes containing 25 μl 1× PBS . The guts were macerated briefly using a plastic pestle . A 10-μl sample of the supernatant was counted under a hemocytometer and the numbers of metacyclic promastigotes , non-metacyclic forms , and total parasite number , as determined by morphology and movement , were counted . Leishmania infections were allowed to mature for 14–16 days within the sand fly midgut . One day before transmission the sucrose diet was removed . On the day of transmission , 4–5 flies were transferred to small plastic vials ( volume 12 . 2 cm2 , height 4 . 8 cm , diameter 1 . 8 cm ) covered at one end with a 0 . 25-mm nylon mesh . Mice were anesthetized by intraperitoneal injection of 30μl of ketamine/xylazine ( 100 mg/ml ) . Specially designed clamps were used to bring the mesh end of each vial flat against the ear , allowing flies to feed on exposed skin for a period of 2–3 hours in the dark at 23°C and 50% humidity . Ear tissue was prepared as previously described [13 , 20] . In experiments employing direct intracellular staining ( dICS ) , ears were removed and placed in 70% ethanol for 2–5 minutes and then allowed to dry . Separated dorsal and ventral sheets of ears were then incubated at 37°C for 90 minutes in 1ml of DMEM containing 20μg/ml of Brefeldin A and 16μg/ml of Liberase that was pre-warmed to 37°C . Following Liberase treatment tissue was homogenized for 3½ minutes in a Medicon using a Medimachine ( Becton Dickinson ) . The tissue homogenate was then flushed from the medicon with 10 ml RPMI media containing 0 . 05% DNase I and filtered using a 50 μm-pore-size cell strainer . In experiments employing dICS , 2μg/ml of Brefeldin A was added to DNase media pre-warmed to 37°C and the ear homogenate returned to 37°C and 5% CO2 for an additional 1 . 5–2 hours in media containing 2μg/ml of Brefeldine A . Ear draining LN ( dLN ) were removed and homogenized with a syringe plunger , and the cell suspension was filtered through a 70-μm strainer . Blood was obtained by intra-cardiac bleed and red blood cells were eliminated by using ACK lysing buffer ( Lonza ) . For monocyte purification , ear tissue was prepared as described above 60h post-infection , and CD11b+Ly6Cint/hiCX3CR1+LY6G- infected ( RFP+ ) and uninfected ( RFP− ) monocytes were sorted from dermal tissue using a FACSAria ( BD Biosciences ) cell sorter . Cells were either stained with Giemsa following cytospin directly post-sort to determine intracellular infection or resuspended in complete RPMI and incubated at 37°C for 24h and 48h for subsequent analysis . Tissue derived cells were re-stimulated as described previously [22] . Briefly , single-cell suspensions were stimulated with 106 T cell-depleted ( Miltenyi Biotech ) naïve spleen cells ( APCs ) , with 50 μg/ml freeze-thaw Leishmania major antigen . During the final 4–6 hours of culture , 1μg/ml of Brefeldin A ( Sigma ) was added . For immunolabelling , cells were washed and labeled with Live/Dead fixable BLUE or AQUA at a 1:500 dilution of the manufacturer suggested stock solution ( Invitrogen ) to exclude dead cells . Cells were incubated with anti-Fc III/II ( CD16/32 ) receptor Ab ( 2 . 4G2 ) , followed by surface staining with various combination of the following antibodies for 30min at 4°C in the dark: PE-Cy7 anti-Ly6G ( 1A8 ) ; -CD4 ( RM4-5 ) , APC or BV650 anti-MHCII ( M5/114 . 15 . 2 ) , APC-eFluor 780 anti-Ly6C ( HK1 . 4 ) , Brilliant Violet 421 , APC or PE-Cy7 anti-CD64 ( X54-5/7 . 1 ) , -TCR-β ( 145-2C11 ) , PerCp-Cy5 . 5 or Brilliant Violet 605 CD11b ( M1/70 ) , Brilliant Violet 785 anti-CD11c ( N418 ) , FITC or BV421 anti-CD24 ( M1/69 ) , and/or PE-Cy7 anti-CD86 ( GL-1 ) . Staining for CCR2 employed AlexaFluor 700 anti-CCR2 ( 475301 ) and was done prior to surface staining at 37°C for 20-30min . In some experiments , cells were then fixed with BD Cytofix/Cytoperm ( BD Biosciences ) according to the manufacturer instructions and stained with FITC or APC anti-IFN-γ ( XMG1 . 2 ) , or Alexa Fluor 488 or APC anti-nos2 ( CXNFT ) . Isotype controls employed were rat IgG1 ( R3-34 ) and rat IgG2b ( A95-1 or eBR2a ) . All Abs were from eBiosciences , BD Biosciences , Biolegends or R&D systems . Data were collected using FACSDiva software on a FACSLSRII or FACSCANTO II flow cytometer ( BD Biosciences ) , and analyzed using FlowJo software ( TreeStar ) . Forward-scatter and side-scatter width was employed to exclude cell doublets from analysis . To determine the absolute number of cells , a portion of each sample was removed for counting with AccuCheck Counting Beads ( Invitrogen ) as described previously [41] . Diphtheria toxin ( DT ) was obtained from Milipore , reconstituted at 1 mg/ml in PBS , and frozen at −80°C . Mice received 25 ng/g DT via the intra-peritoneal route in 0 . 2–0 . 3 ml PBS . The toxin was injected every other day starting one day prior to the challenge of the mice with L . major parasites . The efficiency of the depletion was verified 8 days post-depletion , in ears cells , bone marrow and blood . Anti-IFN-γ monoclonal antibody anti-mouse clone XMG1 . 2 ( BioXCell ) was employed to block IFN-γ in vivo . A single dose of 0 . 5 mg/mL of XMG1 . 2 was administered by intraperitoneal injection 2h before the infection . The mice euthanized 7 days post-infection were also treated with the same dose of the mAb 4 days after infection . As an isotype control we treated the mice with 0 . 5 mg/mL of a mAb rat IgG1 anti Horseradish Peroxidase ( BioXCell ) . Human peripheral blood monocytes were obtained from healthy volunteers by counterflow centrifugal elutriation at the NIH Blood Bank under Institutional Review Board–approved protocols of NIAID and the Department of Transfusion Medicine . Human monocytes in vitro culture were performed in RPMI 1640 medium ( Life Technologies ) supplemented with 10% heat-inactivated FCS , 4 mM L-glutamine , 10 mM HEPES , 100 U/ml penicillin and 100 μg/ml streptomycin . Experiments were carried out on the day of collection . Monocytes were plated at 105/ml cells in 6-well tissue culture plates and infected with L . major metacyclic promastigotes at a multiplicity of infection ( MOI ) of 1:4 . At 6 hours , excess parasites were removed by two washes at cell speed . At 6 and 72h cells were recovered and cytospin and stained with Diff quick . ROS production was measured as previously described [42] . Sequential two-photon intravital imaging of single bite sites and image analysis was performed as reported and described previously [22 , 43] . Briefly , anesthetized mice were imaged in the lateral recumbent position that allowed the ventral side of the ear pinna to rest on a coverslip . A strip of Durapore tape ( 3 M ) was stuck to a bench top several times ( to ensure that subsequent removal would not cause undue damage ) and placed lightly over the ear pinna and affixed to the imaging platform in order to immobilize the tissue . Care was taken to minimize pressure on the ear . Images were acquired using an inverted LSM 510 NLO multiphoton microscope ( Carl Zeiss Microimaging ) enclosed in an environmental chamber that was maintained at 30°C . This system had been custom fitted with 3 external non-descanned PMT detectors in the reflected light path . Images were acquired using either a 20×/0 . 8 air objective or a 25×/0 . 8 NA water immersion objective . Fluorescence excitation was provided by a Chamelon XR Ti:Sapphire laser ( Coherent ) tuned to 920 nm for eGFP and RFP excitation . Voxel dimensions were 0 . 64×0 . 64×2 μm using the 20× objective and 0 . 36–0 . 51×0 . 36–0 . 51×2 μm using the 25× objective . Raw imaging data were processed with Imaris ( Biplane ) using a Gaussian filter for noise reduction . All images are displayed as 2D maximum intensity projections . For analysis of cytokine and chemokine gene expression ears cells were stored in RLT buffer at -80°C until the day of the RNA extraction . Homogenates were then passed through QIAshredder columns , and RNA was purified using an RNeasy midikit according to the manufacturer’s protocol ( Qiagen ) . Reverse transcription was performed using the SuperScript III first-strand synthesis system for reverse transcription-PCR ( RT-PCR ) ( Invitrogen Life Technologies ) . Real-time PCR was performed on an ABI Prism 7900 sequence detection system using the primer probe sets designed by Applied Biosystems . The quantity of the product was determined by the comparative threshold cycle method using 2−ΔΔCT ( where CT represents the cycle threshold ) to determine the fold increase . Each gene was normalized to the 18S rRNA endogenous control and to the average ΔCT of naive mice as sample control . Data were compared using the student’s t-test . Comparisons between multiple groups were done using ANOVA with Holm-Sidek’s post-test . Parasite load determined by LDA and absolute number data were log transformed before statistical analysis . All p-values are two-sided . Statistical calculations were done in Graphpad PRISM 6 . 0 ( www . graphpad . com ) . **** p<0 . 0001; *** 0 . 0001< p<0 . 001; ** 0 . 001<p<0 . 01; * 0 . 01<p<0 . 05 . Human peripheral blood monocytes were obtained from healthy volunteers at the NIH Blood Bank under Institutional Review Board–approved protocols of NIAID and the Department of Transfusion Medicine . All experiments were approved by the University of Calgary Animal Care Committee ( Protocol number AC14-0142 ) in compliance with the Canadian Council for Animal Care or by the NIAID animal care and use committee ( Protocol no . LPD68E ) in compliance with the Animal Welfare Act and the PHS Policy on Humane Care and Use of Laboratory Animals .
|
Many infectious diseases are initiated in the context of inflammation . This inflammatory response may be initiated by the pathogen itself or by damage to barrier sites associated with the infectious process . In the case of the vector-transmitted intra-phagosomal pathogen Leishmania , the parasite must contend with the robust inflammatory response initiated by the bite of an infected sand fly . Traditionally , rapid infection of macrophages in the skin and manipulation of these cells was seen as the mechanism by which the parasite avoided elimination by inflammatory cells . In the present study , we find that this is not the case following primary infection . After transient residence in neutrophils , Leishmania parasites transitioned into immature inflammatory monocytes , where they underwent proliferation and suppressed the maturation of these cells . In stark contrast , in a setting of pre-existing immunity , inoculation of parasites at a secondary site of infection resulted in parasite killing by monocytes in an IFN-γ dependent manner . Therefore , the role of monocytes is dependent upon the primary or secondary nature of the infection site into which they are recruited , emphasizing both the plasticity of this cell population and the central role these cells play during Leishmaniasis .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] |
[
"blood",
"cells",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"pathology",
"and",
"laboratory",
"medicine",
"respiratory",
"infections",
"ears",
"immunology",
"parasitic",
"diseases",
"parasitic",
"protozoans",
"pulmonology",
"otology",
"ear",
"infections",
"protozoans",
"signs",
"and",
"symptoms",
"leishmania",
"neutrophils",
"white",
"blood",
"cells",
"inflammation",
"animal",
"cells",
"head",
"otorhinolaryngology",
"immune",
"response",
"diagnostic",
"medicine",
"cell",
"biology",
"monocytes",
"anatomy",
"leishmania",
"major",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"organisms"
] |
2017
|
Divergent roles for Ly6C+CCR2+CX3CR1+ inflammatory monocytes during primary or secondary infection of the skin with the intra-phagosomal pathogen Leishmania major
|
Cells are naturally surrounded by organized electrical signals in the form of local ion fluxes , membrane potential , and electric fields ( EFs ) at their surface . Although the contribution of electrochemical elements to cell polarity and migration is beginning to be appreciated , underlying mechanisms are not known . Here we show that an exogenous EF can orient cell polarization in budding yeast ( Saccharomyces cerevisiae ) cells , directing the growth of mating projections towards sites of hyperpolarized membrane potential , while directing bud emergence in the opposite direction , towards sites of depolarized potential . Using an optogenetic approach , we demonstrate that a local change in membrane potential triggered by light is sufficient to direct cell polarization . Screens for mutants with altered EF responses identify genes involved in transducing electrochemical signals to the polarity machinery . Membrane potential , which is regulated by the potassium transporter Trk1p , is required for polarity orientation during mating and EF response . Membrane potential may regulate membrane charges through negatively charged phosphatidylserines ( PSs ) , which act to position the Cdc42p-based polarity machinery . These studies thus define an electrochemical pathway that directs the orientation of cell polarization .
Cell polarization arises from the asymmetric accumulation of cellular components near a region of the plasma membrane . Although the roles of polarity proteins such as small GTPases and cytoskeletal elements have been studied extensively [1] , much less is known about the possible contribution of electrochemical elements . Recent studies identifying certain ion transporters in regulating processes such as cell migration and polarized cell growth indicate potential roles of local pH , ion fluxes , and membrane potentials at the plasma membrane [2]–[8] . How these elements interface with established modules of polarity networks remains to be defined . The importance of electricity in cell polarization is illustrated by the ability of electric fields ( EFs ) to direct cell polarization . It has been appreciated for decades that most cells—ranging from bacteria , fungi , and amoebas to animal cells—are electrotactic , and robustly orient polarity , migration , or division to applied exogenous EFs [9]–[14] . EFs of similar intensities as those used in these experiments naturally surround cells in tissues , and even individual cells such as fungal cells [10] , [15] , [16] . The physiological relevance of endogenous EFs has been demonstrated in fungal infection [17] , immune cell response [18] , wound healing , regeneration , and development [6] , [10] , [19] , [20] . These findings have led to the proposal that in addition to responding to chemical and mechanical signals , cells may also be responding to endogenous electrotactic signals to guide cell polarization [20] . The response of cells to exogenous EFs provides a powerful tool to study electrochemical elements in cell polarization . The molecular mechanisms of cell polarity are currently best understood in the budding yeast , Saccharomyces cerevisiae . Polarized cell growth in these cells is tightly controlled by intrinsic and extrinsic spatial cues . Haploid budding yeast cells display an axial budding pattern , in which new buds form adjacent to previous bud sites , while diploid cells exhibit a bipolar pattern , in which buds emerge at sites of previous division or growth [21] , [22] . During mating , cells of opposite mating type polarize towards each other in response to gradients of secreted pheromones; exogenous application of the pheromone α-factor causes cells to grow a mating projection , forming a pear-shaped “shmoo . ” The core polarity machinery required for both bud and shmoo formation is organized around the small GTPase Cdc42p , which coordinates actin assembly and exocytosis [23]–[25] . Bud site selection is specified by a Ras-like protein Rsr1p and its regulators [23] . During mating , these spatial cues used to direct budding are turned off , so that cells can polarize towards the mating partner . This reorientation of polarity involves Far1p and its interactions with the receptor-coupled Gβ protein and Cdc42 GEF [25]–[27] . As demonstrated by mutants affected in the regulation of only shmoos or only budding [23] , [28] , [29] , there are specific molecular differences in the mechanisms governing budding and shmoo polarity . In general , still little is appreciated about electrochemical aspects of cell polarization in this cell type . Here , we show that cell polarity can be directed by exogenous EFs in budding yeast . Although EFs have been shown to direct polarized growth in Schizosaccharomyces pombe [13] and Candida albicans [30] , [31] , there have been no reports to date in S . cerevisiae . We find that although EFs do not appear to affect wild-type ( WT ) budding cells , they do have robust effects on cells in the presence of pheromone and on mutants defective in bud site selection . We find a potassium channel and membrane lipid charges as components mediating EF responses . We further show , using a light-activated rhodopsin , that local membrane potential itself is capable of directing polarization . Our results demonstrate the importance of electrochemical signaling in cell polarity and begin to define mechanistically how they contribute to polarized cell growth .
We tested whether exogenous EFs can influence cell polarization in budding yeast . Yeast cells were grown in the presence of EFs in microfluidic channels , which allow for defined EF lines and heat control [13] . Haploid WT cells were mostly resistant to EF effects and budded at their normal axial position ( Figure 1A and 1B ) . The bud site selection mutant rsr1Δ forms buds in random directions , in the absence of EF . In the EF , however , almost all new buds emerged at the cathode-facing side of the rsr1Δ cells after 1 h of exposure to an EF of 50 V/cm ( Figure 1A and 1B; Movie S1 ) . Cells did not exhibit any major signs of stress , cell death , or stress pathway activation [32] , but grew with slightly reduced growth rates and prolonged cell cycle length as controls ( Figure S1 ) . Cathodal bud orientation displayed dose dependence on EF intensity and duration of application ( Figure S2A and S2C ) . Diploid WT cells also polarized towards the cathode significantly more than WT haploids; this may reflect a less stringent regulation of budding pattern in diploids ( Figure 1B ) . Thus , the EF was not able to efficiently override the normal spatial cues involved in axial budding , but could direct bud site polarization if these cues were absent or weak . The application of EFs also directed the site of shmoo tip formation but , surprisingly , in the opposite direction . In the presence of uniform concentrations of α-factor and an EF , budding yeast cells showed a strong polarization towards the anode ( Figures 1C , 1D , Figure S1S2B , S2D , and S2E; Movie S2 ) . Changing the EF direction induced the formation of a second shmoo tip towards the new anode ( Figure 1E; Movie S3 ) . To rule out possible effects of adding mating factor exogenously , we also noted similar effects in mating pairs of cells . The EF disrupted mating and caused cells to polarize towards the anode of the EF instead of towards each other ( Figures 1G and S2F ) . Cells that were induced to shmoo without external pheromones , by overexpressing Ste4p , the β subunit of the G protein involved in pheromone response , also polarized toward the anode [33] ( Figure 1F and 1G; Movies S4 and S5 ) . Thus , although bud and shmoo formation use many of the same components of the polarity machinery [21] , [22] , there is a striking difference in directionality ( cathodal versus anodal ) for how budding and shmooing yeasts respond to EFs . We next tested whether cell polarization in response to EF requires the same polarity machinery normally used in budding or shmooing . The highly conserved small GTPase Cdc42p was required to polarize buds and shmoos in the absence or presence of the EF , as assessed with the loss-of-function mutant allele cdc42-118 ( Figures 2A , 2B , and S3A ) [34] . In addition , mutants specifically defective in establishing polarity during mating but not budding , such as bem1-s1 ( a point mutant in the scaffold protein Bem1p [29] ) and the formin null mutant bni1Δ [28] , showed similar polarization defects in the absence or presence of the EF ( Figures 2B , S3B , and S3E ) . Imaging GFP-Cdc42 [35] and the associated components Cdc24-GFP [36] ( a GEF for Cdc42p ) and Bem1-GFP [37] revealed that polarity caps assembled and oriented to the EF prior to bud or shmoo emergence ( Figure 2C–2E ) . Bem1-GFP cap assembly was dependent on Cdc42p in the presence or absence of EF ( Figure S3C and S3D ) . Actin also appeared to similarly mediate cell polarization in both instances . In budding cells , actin was dispensable for EF-induced Bem1-GFP cap cathodal orientation , although actin depolymerization appeared to accelerate polar cap accumulation at the cathodal side . In shmooing cells , actin inhibition caused rapid disappearance of the cap in the presence or absence of the EF [27] ( Figures 2D , 2F , and S3F ) . Together , these data show that the EF acts in reorienting polarized cell growth through the normal polarity machinery , including Cdc42p and its regulators . To investigate how EF directs mating projections , we tested the role of Far1p and Cdc24p . Mutant far1-s and cdc24-m cells have a specific orientation defect in response to α-factor , as they are not able to orient appropriately towards gradients of α-factor , and polarize instead using bud site selection cues [25]–[27] . In saturating concentrations of α-factor , we found that both of these mutants polarized towards the cathode of the EF ( the opposite direction as WT cells ) ( Figure 2G ) . This reversal was also observed at non-saturating concentrations of pheromones ( Figure S3G ) . As rsr1Δ mutants in the absence of α-factor bud towards the cathode , this suggests that far1-s and cdc24-m mutants may use machinery that orients buds to direct shmoo projections to the cathode . EFs are thought to affect cellular processes at or outside the plasma membrane , but not in the cell interior . They have been postulated to generate subcellular asymmetries in transmembrane potentials ( TMPs ) [13] , [38] , [39] , and/or displace charged membrane proteins at the cell surface [40] , [41] . To test whether membrane transporters mediate EF responses , we screened a set of well-characterized mutants and inhibitors affecting transport at the membrane . We found that calcium , sodium , and proton transport systems are not critical for EF sensing for bud or shmoo reorientation ( Figure S4 ) . We found , however , that a potassium transporter mutant trk1Δ was defective in the anodal orientation of mating projection , but not in budding orientation; these cells oriented shmoos to the cathode , in a similar manner as far1-s and cdc24-m mutants ( Figure 3A; Movie S6 ) . Trk1p is a high-affinity inward potassium transporter that displays conserved features in bacteria , plants , and fungi . In yeast , Trk1p is a major TMP regulator [42] , [43] , and trk1Δ cells exhibit hyperpolarized resting potential ( Figure 3B and 3C ) [42] . A trk2Δ mutant , in the secondary K+-import system ( Trk2p ) , did not display any orientation defect in the EF , however [44] . Similarly to far1-s and cdc24-m mutants , trk1Δ mutants formed shmoos with normal morphology and timing , but were defective in mating ( efficiencies of ∼10% of WT; Figure 3D ) , and displayed significant defects in polarizing in the correct direction in mating pairs ( Figure S5B ) . In contrast , trk1Δ had no defects in bud emergence and haploid axial patterns ( Figure S5A ) . We found that Trk1-GFP was located throughout the plasma membrane , but was reduced in emergent growing buds and shmoo tips , in a pattern similar to that of other membrane transporters [45] , [46] . In shmooing cells , measurements of fluorescence intensity showed a stable back-to-front gradient , with a concentration ratio of about 3-fold ( Figures 3E and S5C ) . In the presence of the EF , we observed a similar depletion of Trk1-GFP at the shmoo tip growing towards the anode , without noticeable change in protein distribution prior to tip growth ( Figure S5D and S5E ) . Together , these data suggest that a natural gradient of Trk1p leading to local differences in potassium import may contribute to polarity regulation for shmoo tip orientation and EF response . To shed more light on why cells may polarize in these different directions , we performed computational simulations and analytical calculations of the local EF strengths and electric potentials along the membrane of S . cerevisiae cells ( Figure S6 ) . This showed that sites of bud and shmoo emergence correspond to the minimum and maximum local EF potentials , and to sites of depolarized and hyperpolarized TMPs , respectively . This analysis thus led to the prediction that if EF-induced polarity is sensitive to TMPs , shmoos should emerge at sites of hyperpolarized TMP , while buds should emerge at sites of depolarized TMP . To directly test the nature of the electrochemical signaling orienting polarity , we developed an optogenetic approach to locally modulate TMPs and/or ion fluxes [47] . Microbial opsins are light-gated transmembrane channels or pumps that have been used to modulate TMPs for neuron activation or silencing [48] , as well as in other cell types such as yeast [49] , [50] . We expressed different opsins tagged with GFP , and found that Halorhodopsin-GFP ( NpHR ) displayed the most robust expression and plasma membrane targeting , although there was some low level accumulation of Halorhodopsin-GFP in internal membranes , as often seen in other cell types [51] ( Figure S7A ) . Halorhodopsin is a reversible inward chloride pump that causes rapid hyperpolarization of the TMP upon activation with green/yellow light [48] . We confirmed that Halorhodopsin could drive membrane hyperpolarization upon light activation in budding yeast , by measuring changes in global membrane potential in single cells following laser exposure , using the sensitive dye DiBAC4 ( 3 ) ( Figure S7B and S7C ) . We implemented a photoactivation assay to locally hyperpolarize mating and budding yeast cells at specific sites on the plasma membrane [52] . Cells were illuminated on a small square-shaped region at the cell surface with a yellow laser for 20 min , and subsequently filmed for 2 h to compute polarized growth orientation ( Figure 4A and 4B ) . Laser exposure did not cause the cells to die or halt growth , but we did note a reduction in growth rate of ∼10%–15% in cells exposed to the laser compared to non-exposed controls in the same field . Accordingly , measurement of stress pathway activation revealed a minor stress response that remained negligible compared to typical osmotic stress responses ( Figure S8A–S8C ) . Strikingly , many cells expressing Halorhodopsin subsequently grew mating projections towards the site of the laser illumination ( Figure 4C and 4D ) . This effect on orientation caused by light was similar to the one caused by 20 min of EF exposure ( Figure S2D ) . Control cells that either did not express Halorhodopsin or expressed an unrelated GFP-tagged transmembrane protein , Hxt3-GFP , with similar localization [53] polarized in directions independent of the laser , showing that this effect was opsin-dependent and not due to cellular damage from the laser itself [54] ( Figures 4D and S8D ) . Similar treatments in budding cells did not orient bud site emergence however ( Figure 4D ) . These data suggest that the direction of mating projections can be controlled by local hyperpolarization of membrane potentials . We next asked how local changes in membrane potential influence the Cdc42-based polarity machinery . Although membrane potential could impact proton transport and local pH [13] or the transport of other ions , our candidate screen did not reveal any obvious role for proton or other ion transport systems other than Trk1p ( Figure S4C and S4D ) . Another way by which membrane potential may affect polarity is through membrane electrostatics by affecting charged lipid flipping [55]–[57] . PS is a negatively charged lipid that acts as an electrostatic platform at the inner leaflet to regulate membrane binding of proteins including Cdc42p [58] . In budding yeast , PS concentrates at sites of shmoo and bud emergence [58] . A PS synthesis mutant cho1Δ has defects in Cdc42p recruitment , shmoo polarity , and mating [58] . We found that this mutant also exhibited an abnormal EF response in that it oriented mating projections to the cathode of the EF , much like trk1Δ , far1-s , and cdc24-m mutants ( Figure 5A ) . Conversely , mutants in a lipid flippase complex , dnf1-2Δ or lem3Δ , which may have increased PS and negative surface charges [59] , showed significant increased anodal shmoo orientation in the EF . PS and membrane charges affected EF response only in shmoos , not in buds ( Figure 5B ) . Next , we imaged PS localization using a GFP-Lact-C2 probe [58] , [60] . In shmooing cells , PS rapidly accumulated and persisted at the anodal side , long before shmoo appearance . In budding cells , PS also initially accumulated at the anodal side , but then reverted to the cathodal side immediately prior to bud emergence , often leaving a secondary patch at the anodal side ( Figure 5C and 5D ) . Thus , asymmetries in membrane potential may bias the localization of Cdc42p and other polarity factors through effects on PS and membrane charge .
These are the first studies , to our knowledge , showing the input of membrane electrochemistry in the regulation of cell polarity in S . cerevisiae . We find that EFs direct the site of bud formation and mating projections in different directions . Our optogenetic experiments further show that in shmooing cells , local hyperpolarization of membrane potential is actually sufficient for polarity reorientation ( Figure 6 ) . The mating defects of trk1Δ and cho1Δ mutants [58] , for instance , demonstrate that this pathway contributes to cell polarization even in the absence of EFs . Our results suggest a model in which the asymmetric segregation of Trk1p and possibly other transporters produces positive charges at the back of the cell and negative charges on PS lipids at the front of the cell , which promotes the polarized distribution of Cdc42 . In the absence of EFs , the asymmetry of Trk1p localization may arise from initial polarization of membrane insertion . These electrochemical pathways may thus represent a positive feedback loop that stabilizes the axis of Cdc42-based polarity for chemotropism . A surprising finding of this study is the different behavior of budding versus shmooing cells . Although these polarization systems share downstream polarity regulators , we found clear differences in the requirement for upstream electrochemical elements . rsr1Δ cells bud towards the cathode , while the same strain shmoos towards the anode . Mutations in Far1 , Cdc24 , Trk1 , and Cho1 all cause cells to shmoo in an abnormal direction in response to EF and have mating defects in the absence of EF , but have little or no effect on bud site selection [27] , [58] . The cathodal orientation of budding cells in EFs suggests that buds originate at sites of depolarized TMP . However , genetic and optogenetic analysis demonstrate that they may not be dependent on gradients of membrane potential or PS levels , suggesting that regulation of bud site selection is determined by a distinct mechanism . Although it is not yet known what elements act upstream of Cdc42p to drive cathodal growth , a plausible hypothesis is that EFs may localize some charged membrane proteins by direct electrophoresis , as suggested in other systems [13] , [40] . The EF thus causes a tug-of-war between two competitive pathways that steer polarity in different directions , with the anodal one being dominant in response to mating factors . The observed responses of cells to exogenously applied EFs lead to a question of whether EFs normally contribute to polarity regulation . Tissues and even individual polarized cells are surrounded by EFs , which may arise from asymmetries in ion transport [61]–[65] . We speculate that fungi may respond to their own EFs , possibly during mating , in the context of fungal communities such as biofilms , and in their natural environment to guide them during invasion of host tissues , for instance . It would be interesting to examine the role of genes such as Trk1 on various fungal behaviors . Mechanisms of electrochemical regulation of cell polarity are likely conserved . Our data are consistent with recent findings implicating a similar set of actors in fission yeast , neutrophils , keratocytes , and slime molds [9] , [10] , [12] , [13] , [40] . Fission yeasts respond to EFs by orienting their growth axis perpendicular to the EF , producing bent morphologies . Cdc42 , formins , and the Pma1 proton pump at the plasma membrane are identified as critical elements . Pma1 affects cell polarity and actin assembly even in the absence of exogenous EFs , indicating a role for membrane potential and intracellular pH in regulating normal tip growth [13] . Migrating neutrophils , keratocytes , and slime molds orient migration to exogenous EFs , possibly through effects on membrane potential [10] , [66] . These responses involve the phosphorylation and charge additions on phosphatidylinositol lipids—mediated by PI3-kinase—that recruit and activate Rho GTPases for polarized migration . Recent studies on plasma membrane pumps and channels are beginning to reveal the pivotal role of membrane potential , pH , and/or local ion transport in cell migration [67]–[69] , mitotic rounding [70] , asymmetric aging [71] , and tissue patterning [4] , [7] , [72] . A Na+-H+ exchanger , Nhe1 , is needed for directionality in fibroblast migration; this transporter has been shown to control local pH , which affects the ability of a Cdc42 GEF to bind to the plasma membrane [69] , [73] . Similarly , the membrane targeting of Dishevelled needed for planar cell polarity activation in fly epithelia may rely on charge interaction and pH [4] , [69] . An inward-rectifier potassium channel influences patterning of zebrafish skin stripes , leading to a model in which membrane potential controls a directional switch in cell migration and consequent cell–cell arrangement in the tissue [7] . The establishment of a highly tractable system in yeast to study the mechanisms of electrochemical regulation will serve as a foundation to understand the diverse roles of membrane electrochemistry in processes related to cell polarity .
Standard methods for S . cerevisiae media and genetic manipulations were used . Strains and plasmids used in this study are listed in Tables S1 and S2 , respectively . Microscopy was performed at room temperature ( 23–25°C ) with either an inverted wide-field fluorescence microscope or a spinning-disk confocal microscope . Images were acquired , processed , and analyzed with Micro-Manager or Metamorph . Chambers to apply the EF to the cells were adapted from previously described methods [13] . Microchannels were approximately 200 µm high , 500 µm wide , and 4 cm long and were fabricated in PDMS . S . cerevisiae cells were immobilized by adding 1% of low-melting agarose to the medium . For shmooing experiments with saturating pheromones , 50 µM of α-factor was added to the medium , and cells were placed in the channel 30 min prior to EF application . Because of their slow growth , cho1Δ cells were placed in the channel 1 h prior to EF application . Reservoirs connecting electrodes to the channels contained 4% agarose blocks made of medium , which protect cells in the channel from potentially toxic products emanating from the electrodes . Electrodes connected to a generator were immersed in liquid medium added on top of the reservoirs . In these conditions , growth rate and cell cycle periods were almost unaffected , and no significant stress was induced ( Figure S1 ) . The optogenetic assay used a 535-nm laser , with a power of ∼5 mW , interfaced with an iLas system ( Roper Scientific ) mounted on a confocal spinning disk and a 63× objective . This allowed irradiation of multiple regions of interest ( of 20×20 px2 ) in a given field of view . Cells were placed on a 2% agar pad containing 20 µM of all-trans retinal and 50 µM of α-factor for shmooing experiments . Cells were put on the pad 30 min prior to laser excitation . The laser was turned on for a continuous period of 20 min , and the cells were subsequently filmed for 2 h to monitor polarized growth . Laser exposure did not induce major changes in growth rate or stress levels ( Figure S8A–S8C ) . All inhibitors were prepared at the indicated concentration and applied before EF application . Latrunculin A ( LatA ) ( Sigma ) was used at a final concentration of 100 µM from a 100× stock in DMSO . The calcium ionophore A23187 was used at a final concentration of 10 µM . The calcium chelating agent EGTA was used at a final concentration of 2 mM . <1 ? tpb +2pt ? > Efficiency of mating in trk1Δ mutants was assayed by quantitative counting of mating diploids . WT Mat α ( AC 131 ) , WT Mat a ( AC 129 ) , and trk1Δ Mat a ( AC 31 ) cells were grown to mid-log phase in YPD medium and concentrated to 10 OD/ml; WT Mat α cells were incubated at a 10∶1 ratio with target WT or trk1Δ Mat a cells , and collected into a soft pellet by centrifugation . After 4 h of mating at 30 °C , cells were suspended in liquid YPD , and serial dilutions were plated on medium selective for diploids . Mating efficiency was compared between WT and trk1Δ by counting the number of diploid colonies obtained at different dilutions . About 500 colonies were counted for each condition , and the assay was repeated twice . In addition to this assay , we also counted the number of genuine zygotes by microscopy after 4 h of mating . To this aim , WT Mat α , WT Mat a , and trk1Δ Mat a cells were grown in YPD liquid medium to mid-log phase and concentrated to 10 OD/ml . WT Mat α cells were stained with calcofluor for 5 min , subsequently rinsed with YPD , and incubated at a 10∶1 ratio with target WT or trk1Δ Mat a cells . A 10-µl drop of each mixture was then spotted onto a YPD plate and incubated for 4 h at 30 °C . The mixtures were then imaged on a microscope , and mating efficiency was computed as the ratio of genuine zygotes to the total number of Mat α cells in the field of view . About 500 cells were counted for each condition , and the assay was repeated twice . This assay yielded a mating efficiency in the trk1Δ of about 30% of the WT . To test the role of Trk1p in axial budding , we generated a trk1Δ strain in the W303 background with a WT copy of the BUD4 gene ( AC 134 ) . Cells were then grown to mid-log phase in YPD medium and stained with calcofluor for 5 min to mark bud scars . Axially budding cells were counted when more than three scars were clustered at one site on the surface . To compare the efficiency of chemotropism in WT versus trk1Δ cells , we used a previously described assay that takes advantage of the fact that WT cells grow towards their mating partners irrespective of previous bud site selection , while mutants with defective mating polarity ( like far1-s or cdc24-m ) use bud site selection cues to grow shmoos [74] . WT Mat α , WT Mat a , and trk1Δ Mat a cells were grown in YPD liquid medium to mid-log phase and concentrated to 10 OD/ml . WT Mat a and trk1Δ Mat a cells were stained with calcofluor for 5 min , subsequently rinsed with YPD , and incubated at a 1∶1 ratio with target WT Mat α cells . A 10-µl drop of each mixture was then spotted onto a YPD plate and incubated for 4 h at 30°C . The mixtures were then imaged to assess the position of the bud scar relative to the fusion site in each newly formed zygote . Only zygotes with a single fluorescent bud scar were counted . Zygotes were scored as proximal if the bud scar was in the one-third of the cell adjacent to the fusion site , medial for the middle one-third , and distal for the one-third away from the fusion site ( Figure S5B ) . To measure global membrane potential in single budding yeast cells , we used the membrane potential dye DiBAC4 ( 3 ) ( Invitrogen ) , which absorbs in blue light and depicts increased membrane fluorescence upon membrane depolarization , with a sensitivity of nearly 1% per millivolt [7] . Cells were incubated with a concentration of 50 µM dye for 30 min , and images were taken on a confocal spinning disk . Relative membrane potential values were then quantified as membrane signal subtracted from background signal . To assess membrane hyperpolarization by Halorhodopsin , cells were immobilized at the bottom of a microfluidic chamber , between a dialysis membrane and the coverslip [75] . Cells expressing Halorhodopsin-GFP were bleached by long-time exposure with a blue laser . Medium was subsequently exchanged with YPD containing 50 µM DiBAC4 ( 3 ) dye , and the dye was left to stain the cells for 30 min . Dye staining intensity at the membrane was then measured in the same pre-bleached cells at two consecutive time points spaced by 3 min ( I0 and I1 ) to compute dye photo-bleaching . These cells were then exposed to the yellow laser for 3 min , and the final dye staining was computed again ( I2 ) . The specific loss of dye staining associated with Halorhodopsin effects on membrane potential was then computed as , which is expected to be negative for membrane hyperpolarization and positive for membrane depolarization . Computer simulations were performed using the Matlab Partial Differential Equation Toolbox ( MathWorks ) . The cell surfaces as well as the channel sides were considered as perfect insulators , while the cell interior and the surrounding medium as conductors . The electric potential , Φ , created by the applied EF , , was analytically computed by solving the Laplace equation: ΔΦ = 0 , with the boundary condition at the insulating membrane , with the vector normal to the membrane , and the limit condition at infinity . S . cerevisiae cells were represented by a sphere , leading to the classical results [76] for the potential at the membrane Φm and the field at the membrane : and , with R the radius of the sphere and θ the angle with the field .
|
The ability of cells to orient towards spatial cues is critical for processes such as migration , wound healing , and development . Although the role of electrochemical signals is well characterized in processes such as neuronal signaling , their function in cell polarity is much less understood or appreciated . Application of exogenous electric fields can direct cell polarization in many cell types , and electric fields of similar magnitude surround cells and tissues naturally . However , the significance and mechanism of these responses remain poorly understood . Here , we introduce budding yeast ( Saccharomyces cerevisiae ) as a powerful model system to study electrochemical regulation of cell polarity . We show that application of electric fields causes budding yeast to polarize in particular directions . We begin to identify key proteins involved in this response , which implicate an electrochemical pathway involving membrane potential , membrane charge , and an ion channel , which ultimately regulate the central polarity factor Cdc42p . These key proteins are not only needed for response to electric fields , but also contribute to cell polarity more generally . To test whether a change in membrane potential is sufficient to control cell polarization , we introduce a light-sensitive ion channel into yeast and show that we can now control the site of polarization simply by using a focused laser beam . Thus , our study shows that electrochemical regulation is an integral component of cell polarity pathways .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"biophysics",
"cell",
"biology",
"biology",
"and",
"life",
"sciences",
"ion",
"transport"
] |
2014
|
Electrochemical Regulation of Budding Yeast Polarity
|
The impact of global climate change on the transmission dynamics of infectious diseases is the subject of extensive debate . The transmission of mosquito-borne viral diseases is particularly complex , with climatic variables directly affecting many parameters associated with the prevalence of disease vectors . While evidence shows that warmer temperatures often decrease the extrinsic incubation period of an arthropod-borne virus ( arbovirus ) , exposure to cooler temperatures often predisposes disease vector mosquitoes to higher infection rates . RNA interference ( RNAi ) pathways are essential to antiviral immunity in the mosquito; however , few experiments have explored the effects of temperature on the RNAi machinery . We utilized transgenic “sensor” strains of Aedes aegypti to examine the role of temperature on RNA silencing . These “sensor” strains express EGFP only when RNAi is inhibited; for example , after knockdown of the effector proteins Dicer-2 ( DCR-2 ) or Argonaute-2 ( AGO-2 ) . We observed an increase in EGFP expression in transgenic sensor mosquitoes reared at 18°C as compared with 28°C . Changes in expression were dependent on the presence of an inverted repeat with homology to a portion of the EGFP sequence , as transgenic strains lacking this sequence , the double stranded RNA ( dsRNA ) trigger for RNAi , showed no change in EGFP expression when reared at 18°C . Sequencing small RNAs in sensor mosquitoes reared at low temperature revealed normal processing of dsRNA substrates , suggesting the observed deficiency in RNAi occurs downstream of DCR-2 . Rearing at cooler temperatures also predisposed mosquitoes to higher levels of infection with both chikungunya and yellow fever viruses . This data suggest that microclimates , such as those present in mosquito breeding sites , as well as more general climactic variables may influence the dynamics of mosquito-borne viral diseases by affecting the antiviral immunity of disease vectors .
There is a great deal of uncertainty concerning how climate change will affect the distribution and incidence of infectious diseases , particularly vector-borne diseases . Although there have been predictions of increased vector-borne disease with increasing global temperatures [1] , modeling such changes is difficult because of the complex relationships between vector-borne diseases and temperature [2] , [3] , [4] , [5] , [6] . For example , higher temperatures can shorten the time it takes mosquitoes to become competent vectors of human disease agents [7] , [8] , but also shorten the mosquito lifespan; thus , decreasing their ability to serve as successful vectors [5] . Temperature-dependent effects on the susceptibility of mosquito vectors to viral infection also contribute to the complexity of disease transmission . An inverse relationship appears to exist between the temperature at which mosquito larvae undergo development and their future susceptibility to virus infection [9] , [10] , [11] . Lower temperatures in particular have been shown to adversely affect a vector's ability to modulate viral infection [12] , [13] , [14] and increase rates of transovarial transmission [15] . Kramer et al [12] demonstrated that infection levels of western equine encephalomyelitis virus ( WEEV; genus Alphavirus ) in Culex ( Cx . ) tarsalis decreased as a function of increasing temperature . These researchers hypothesized that at higher temperatures , mosquitoes were better able to modulate virus infection . This ability to modulate virus infection was heritable , suggesting a genetic basis for the phenotype . Similar findings were made by Kay and Jennings [14] using Ross River virus ( RRV; genus Alphavirus ) in Ochelerotatus vigilax reared at 18°C . Turell [10] also found significantly higher rates of disseminated infections for both Rift Valley fever virus ( RVFV; genus Phlebovirus ) and Venezuelan equine encephalomyelitis virus ( VEEV , genus Alphavirus ) following the ingestion of an infectious blood meal when mosquitoes were reared at 19°C as compared to 26C . Most recently , Westbrook et al . [11] found that the infectivity of CHIKV for Ae . albopictus increased with decreasing rearing temperature ( 18°C>24°C>32°C ) . Despite these observations , no mechanism has been implicated that might explain how lower temperatures predispose mosquito vectors to higher infection rates . It is clear that small RNA pathways are critical in controlling virus infections of the mosquito host [16] , [17] , [18] , [19] . Infection of the mosquito with an arbovirus results in the production of two different classes of small RNAs; 21 nt small interfering RNAs ( vsiRNAs ) [16] and 24–30 nt piwi-interacting RNAs ( vpiRNAs ) , with the latter possessing a clear ping-pong signature [19] , [20] . Suppressing production of viral small RNAs in the mosquito leads to increased viral replication and mortality [16] . The generation of vsiRNAs depends on the activity of dicer-2 ( DCR-2 ) [21] . However , processing by DCR-2 alone is insufficient to control virus replication . Rather , this depends on the slicer activity of Argonaute 2 ( AGO-2 ) , a component of the RNA induced silencing complex ( RISC ) [22] , [23] , [24] . The RISC undergoes a maturation process in which one strand of a siRNA duplex is selected as a guide , directing the slicer activity of AGO-2 to RNA targets present within the cell . The specific molecules involved in the biogenesis and activity of vpiRNAs have not yet been identified . Here , we correlate loss of RNA silencing in mosquitoes reared at 18°C and increased susceptibility of Aedes spp . mosquitoes for infection with chikungunya virus ( CHIKV; genus Alphavirus ) and yellow fever virus ( YFV; genus Flavivirus ) , suggesting a molecular basis for previous observations of increased transmission of viral pathogens by disease vector mosquitoes exposed to cooler temperatures .
All mosquito strains [Ae . aegypti Liverpool , khw ( white eye ) and all transgenic strains , Ae . albopictus Wise County strain] were reared in environmental chambers with a light cycle of 14∶10 ( day∶night ) and humidity of 80% . For all experiments , mosquitoes were hatched under vacuum for ∼30 minutes , counted , and dispensed ( n = ∼400 ) into 4 L of water pre-chilled at 18°C or pre-warmed at 28°C . Unless otherwise stated in the text , all stages of development occurred at the indicated temperature . CHIKV strain 37797 was produced from an infectious clone as described previously [19] . YFV ( Asibi strain ) was obtained from the Centers for Disease Control and Prevention . Recombinant Sindbis ( SIN ) viruses containing Ae . aegypti dcr-2 or ago-2 fragments were as described in [25] . Assays were performed as described previously [26] , with the following exception . As vertebrate cells inoculated with small quantities of virus require additional infectious cycles before observation of cytopathic effects ( CPE ) , rapidly proliferating vertebrate cell monolayers may begin to deteriorate after too many doublings and the exhaustion of medium . Deterioration of cell monolayers inoculated with relatively small quantities of virus may obscure scoring of CPE . To militate against this , clarified mosquito homogenates were inoculated onto C6/36 mosquito cell monolayers and placed at 28°C for 48 hours; the supernatant from inoculated C6/36 cells was added to BHK-21 cell monolayers , which were then scored for cytopathic effects . Transgenic Ae . aegypti were generated as described previously [25] , [27] . Briefly , freshly deposited Ae . aegypti ( khw strain ) embryos were microinjected with 500 µg/µl Donor plasmid and 300 µg/µl pKhsp82 Mos1 Helper plasmid . The 3×P3-RG Donor plasmid was identical to that described in [25] , with the exception that the final 3×P3-EGFP inverted repeat cassette was deleted . Surviving individuals were crossed with the parental strain and progeny screened for the presence of DsRED fluorescence in the eyes . SYBR Green-based qPCRs for EGFP , dcr-2 and ago-2 mRNA levels were performed as described in [25] . Detection and quantitation of CHIKV ( + ) strand RNA was performed as described in [19] , [28] . YFV RNA was quantified with a standard TaqMan® assay as recommended by the manufacturer ( Life Technologies , Grand Island , NY ) . Small RNA libraries derived from head tissue of transgenic sensor strain mosquitoes were constructed using Illumina's small RNA sample prep kit and sequenced on an Illumina GAII ( single replicate per temperature ) . Small RNA libraries derived from whole female Ae . aegypti infected with CHIKV ( three biological replicates per temperature ) were barcoded and constructed with Illumina's TruSeq™ small RNA prep kit and sequenced on an Illumina HiSeq . All small RNA reads were mapped to the sensor transgene or the CHIKV genome using Bowtie ( v12 . 7 ) [29] after removal of the 3′ adapter sequence . Differential expression of viRNAs were calculated as described in Morazzani et al [19] . All small RNA libraries described in this study are available for download from the Gene Expression Omnibus ( GEO accession # GSE46204 ) .
We previously described two different Ae . aegypti transgenic “sensor” strains that express both EGFP and an inverted repeat derived from EGFP in an eye-specific manner; with knockdown of DCR-2 or AGO-2 resulting in increased EGFP expression ( Fig . 1A and [25] ) . In order to determine if rearing mosquitoes at a cooler temperature adversely affects the ability of the RNAi machinery to silence the EGFP transgene , we reared the two independent sensor strains at 18°C or 28°C . We observed a loss of silencing of the EGFP transgene in both strains when mosquitoes were reared at 18°C ( Fig . 1B ) . To verify that the observed increase in EGFP fluorescence was due to a corresponding increase in the steady-state levels of EGFP mRNA , we performed real-time qPCR on cDNA obtained from the heads of mosquitoes reared at 18°C or 28°C . In both sensor strains reared at the lower temperature , we observed a 4–8 fold increase in EGFP mRNA levels ( Fig . 1C ) . To verify that this was due to a temperature-dependent effect on RNAi , and not non-specific transcriptional or post-transcriptional changes affecting the expression of the transgenes , we generated additional transgenic strains ( termed 3×P3-RG ) carrying a similar construct as our sensor but without the inverted repeat sequence ( Fig . 1A ) . Two different lines were established , P10 and P11A . As expected , neither line P10 nor P11A exhibited any changes in EGFP fluorescence ( Fig . 1B ) or EGFP mRNA levels ( Fig . 1C ) when reared at 18°C . Unlike 3×P3-sensor mosquitoes , the 3×P3-RG transgenic lines did not show any change in EGFP protein or mRNA levels with the loss of DCR-2 or AGO-2 ( Fig . 2A , 2B ) , suggesting that the inverted repeat sequence targeting EGFP is essential for both RNAi-based silencing of EGFP and temperature-dependent loss of silencing . In order to determine if temperature-dependent loss of silencing could be rescued , mosquitoes reared at 18°C were transferred to 28°C one-day post emergence . EGFP mRNA levels were silenced by seven days ( Fig . 3A , 3C ) , whereas mosquitoes that remained at 18°C failed to silence EGFP ( Fig . 3E ) . This indicated that the RNAi pathway is not irrevocably damaged during development at cooler temperatures . Likewise , rearing mosquitoes at 28°C , with a subsequent shift of the newly emerged adults to 18°C resulted in a partial loss of silencing ( Fig . 3B , 3D ) , suggesting that this phenomenon is not necessarily fixed to a specific developmental stage . In order to determine if rearing at cooler temperatures perturbed the biogenesis of small RNAs , we sequenced small RNAs isolated from the heads of sensor strain mosquitoes reared at 28°C or 18°C ( Fig . 4 ) . Small RNA biogenesis was not affected by the lower temperature , suggesting that the observed effects on RNA silencing occur downstream of the initial dicing step . Interestingly , we observed an increase in 28 nt small RNAs derived from the EGFP inverted repeat transgene in the 18°C cohort , suggesting the possibility of increased targeting by the piRNA pathway ( Fig . 4 ) . To determine if the observed loss of transgene silencing correlated with increased susceptibility to viral infection , we fed sensor strain mosquitoes reared at 28°C or 18°C blood meals containing equivalent amounts of CHIKV . When infectivity was assayed eight days after the infectious blood meal , sensor mosquitoes reared at 18°C had significantly ( Fisher's exact test , P<0 . 03 ) higher infection rates with CHIKV in comparison with the cohorts reared at 28°C ( Fig . S1 ) . To determine if cooler temperatures also increase susceptibility of non-transgenic mosquitoes to CHIKV , Liverpool strain Ae . aegypti were offered blood meals containing serial dilutions of the virus . Eight days after challenge , the number of CHIKV-infected mosquitoes was determined . Even though both groups were held at 28°C for the duration of the viral extrinsic incubation period , prior exposure to a temperature of 18°C substantially increased the number of CHIKV-infected mosquitoes following ingestion of the blood meal , as compared with those reared at 28°C ( Fig . 5A ) . Next , we injected mosquitoes reared at 18°C or 28°C with equivalent amounts of virus , transferred both groups to 28°C , and processed 8 hours later; this time point was chosen specifically to limit the number of infectious cycles . This route of infection also bypassed the midgut , to verify that the temperature dependent effect was systemic , and not localized to the midgut tissue . Sequencing small RNA populations from these mosquitoes revealed significantly increased production of vsiRNAs ( exact Poisson test , P<0 . 0001 ) and vpiRNAs ( exact Poisson test , P<0 . 0001 ) in cohorts reared at the colder temperature ( Fig . 5B ) . Real-time quantitative PCR analysis performed on mosquitoes reared at 18°C indicated significantly higher levels of viral mRNA than in those reared at 28°C ( Fig . 5C ) . Thus , despite their increased presence in mosquitoes reared at 18°C , the dicer products in these mosquitoes were not effective at controlling virus replication , likely due to impairment of RNAi at a downstream step [21] , [22] , [30] . To confirm that these findings were not virus or host-specific , we offered infectious bloodmeals containing YFV to a recently colonized Ae . albopictus strain ( 2008 ) reared at 18°C or 28°C , and compared these results to those obtained in Ae . aegypti reared at 28°C . Similar to our previous results , prior exposure to a temperature of 18°C significantly increased YFV infections in Ae . albopictus ingesting blood meals containing the virus ( Fig . 6 ) . Our results also indicated that the Liverpool strain of Ae . aegypti was more susceptible to infection with YFV when reared at 28°C than was the Ae . albopictus strain reared at 28°C or 18°C . However , the difference was not as great between Ae . aegypti reared at 28°C and Ae . albopictus reared at 18°C ( Fig . 6 ) .
Our data suggest post-dicer inhibition of RNA silencing in disease vector mosquitoes subjected to low temperature ( 18°C ) . Mosquitoes reared at the same low temperature and subsequently infected with CHIKV or YFV , proved significantly more susceptible to these viruses than mosquitoes reared at 28°C . These results also corresponded with significantly increased virus replication in mosquitoes exposed to the lower temperature . In mosquitoes infected with CHIKV , increased virus replication occurred despite a significant rise in production of vsiRNAs , suggesting a diminished potency for these dicer products after mosquitoes were reared at the colder temperature . These results are consistent with those of others that have shown the production of vsiRNAs is insufficient to control virus replication in the absence of AGO-2 mediated slicing [21] , [22] , [30] . Thus , we propose that the temperature-dependent effects observed in both transgenic and non-transgenic mosquitoes are due to either direct or indirect inhibition of AGO-2-dependent slicing . Interestingly , temperature-dependent defects in RNAi have also been described in several plant species [31] , [32] . However , in contrast with our results these studies found that dicing of viral dsRNAs was inhibited in plants exposed to cooler temperatures . Nevertheless , the similarity of the plant phenotypes to those described here suggests a convergent evolutionary response to lower than optimal temperatures in both plants and insects . Previous work by our group has demonstrated that suppression of the RNAi response by the protein B2 increases the midgut infection rate for arboviruses [16] . Similar results were observed by others following silencing of DCR-2 or AGO-2 [17] , [18] . These data emphasize the idea that arbovirus entry and uncoating into midgut cells is followed by an effective RNAi response that eliminates many or all of these initial events so that a productive midgut infection is avoided . In our experiments , mosquitoes fed an infectious bloodmeal were treated in an identical manner after virus exposure . Thus , we expect the effects of temperature were limited to the time of virus exposure only . A reduction in RNA silencing at this time would be expected to reduce the ability of initially infected cells to mount a response early in the infectious process; the end result being an increase in the number of productive midgut infections . Once the arbovirus has established this initial foothold , the RNAi response , even if functioning normally might not be able to completely clear the infection , thus , these mosquitoes remain infected for life . Our data suggest that exposure to cooler temperatures after virus infection would also reduce RNAi activity; however , such temperatures would also reduce the replicative capacity of the virus . A prediction of this model is that if lower temperatures negatively affect virus replication more than RNAi , the extrinsic incubation period will increase with decreasing temperatures , a trend observed for most arbovirus-vector interactions [7] , [8] , [33] . In contrast , for virus vector combinations where RNAi is adversely affected more than viral replication , virus production will decrease with increasing temperatures ( ie , increased modulation by RNAi ) . This scenario is consistent with observations previously reported for WEEV and RRV , in which mosquitoes were unable to modulate viral infections at lower temperatures [12] , [13] , [14] . Current models of arboviral disease transmission consider temperature as it relates to parameters such as mosquito longevity; time to complete development , and the extrinsic incubation period following exposure to a given arbovirus [5] , [34] , [35] . However , our data re-emphasize the importance of temperature on mosquito physiology in the time prior to virus exposure . Thus , current models may be improved by also taking into consideration the microclimates present in shaded or secluded breeding sites , such as those preferred by Aedes mosquitoes [4] , [36] . For example , during the epidemic dengue transmission season in Buenos Aires ( Jan-Mar ) , temperatures in shaded microenvironments were estimated to be ∼10°C cooler ( 22–25°C vs 30–37°C ) than those in sunlit areas [37] . Similarly , during an outbreak of CHIKV in La Réunion , Ae . albopictus were found to prefer shaded breeding sites with average temperatures as low as 12 . 6°C [38] . During the same epidemic , a single amino acid substitution is known to have altered the infectivity of a CHIKV strain for Ae . albopictus , a non-traditional vector for this virus [39] , [40] . Transient , temperature-induced increases in infectivity similar to those demonstrated here may increase opportunities for arboviruses to acquire adaptive mutations that permanently modify infectivity for a specific vector species . Urban epidemics resulting from the introduction of arboviral pathogens into densely populated areas are highly dependent on transmission by infected peridomestic vector species . Disconcertingly , we have shown that the infectivity of YFV and CHIKV for two important peridomestic vector species is improved when the aquatic stages of the insects' lifecycles occur at an environmentally relevant low temperature . Thus , our results , as well those published by others [9] , [10] , [11] , suggest that temperature-dependent effects on vector competence may be applicable to many Aedes species and for alpha- , flavi- , and bunyaviruses . Finally , we note that strategies relying on RNAi to generate transgenic mosquito strains with pathogen-resistant phenotypes have been in development for a considerable amount of time [41] , [42] . Our data suggest that RNAi-based effecter genes may need to be re-evaluated for temperature-dependent effects on pathogen-resistance .
|
Although a link between the increased susceptibility of mosquitoes for arthropod-borne viruses and exposure to lower rearing temperatures has been known for many years , the molecular basis of this has remained unknown . We investigated this phenomenon using an engineered strain of mosquito where the expression of a reporter was dependant on the status of the RNA interference pathway ( RNAi ) . Our studies indicate a correlation between the virus-susceptibility phenotype and temperature-dependent deficiencies in antiviral immunity . Specifically , we demonstrate that RNAi , a critical antiviral immune pathway in mosquito vectors of human disease , is impaired in insects reared at cooler temperatures . This suggests for the first time a molecular explanation for previously described observations , findings that may lead to a better understanding of how global climate change will affect the transmission of mosquito-borne viruses , and new criteria for evaluating genetic control strategies based on RNAi . Our studies also suggest a novel mechanism for arbovirus adaptation to otherwise incompetent vector species .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] |
[
"mosquitoes",
"rna",
"interference",
"emerging",
"infectious",
"diseases",
"viral",
"transmission",
"and",
"infection",
"vector",
"biology",
"virology",
"gene",
"expression",
"genetics",
"gene",
"regulation",
"molecular",
"genetics",
"emerging",
"viral",
"diseases",
"biology",
"microbiology",
"viral",
"vectors"
] |
2013
|
Cooler Temperatures Destabilize RNA Interference and Increase Susceptibility of Disease Vector Mosquitoes to Viral Infection
|
Increasing seed oil content is one of the most important breeding goals for soybean due to a high global demand for edible vegetable oil . However , genetic improvement of seed oil content has been difficult in soybean because of the complexity of oil metabolism . Determining the major variants and molecular mechanisms conferring oil accumulation is critical for substantial oil enhancement in soybean and other oilseed crops . In this study , we evaluated the seed oil contents of 219 diverse soybean accessions across six different environments and dissected the underlying mechanism using a high-resolution genome-wide association study ( GWAS ) . An environmentally stable quantitative trait locus ( QTL ) , GqOil20 , significantly associated with oil content was identified , accounting for 23 . 70% of the total phenotypic variance of seed oil across multiple environments . Haplotype and expression analyses indicate that an oleosin protein-encoding gene ( GmOLEO1 ) , colocated with a leading single nucleotide polymorphism ( SNP ) from the GWAS , was significantly correlated with seed oil content . GmOLEO1 is predominantly expressed during seed maturation , and GmOLEO1 is localized to accumulated oil bodies ( OBs ) in maturing seeds . Overexpression of GmOLEO1 significantly enriched smaller OBs and increased seed oil content by 10 . 6% compared with those of control seeds . A time-course transcriptomics analysis between transgenic and control soybeans indicated that GmOLEO1 positively enhanced oil accumulation by affecting triacylglycerol metabolism . Our results also showed that strong artificial selection had occurred in the promoter region of GmOLEO1 , which resulted in its high expression in cultivated soybean relative to wild soybean , leading to increased seed oil accumulation . The GmOLEO1 locus may serve as a direct target for both genetic engineering and selection for soybean oil improvement .
Soybean ( Glycine max ( L . ) Merr . ) is an important food and oil crop . Soybean seeds accumulate large amounts of oil and protein and have been intensively targeted for human consumption during long-term domestication and cultivation . Given the high percentage of oil in soybean seeds , the demand for soybean oil production has increased dramatically due to the increasing demand for vegetable oils and expanded use of biodiesel , and the seed composition improvement is of particular interest in terms of increasing awareness of health issues around dietary fats [1] . However , oil accumulation in the seed is a complex metabolic process that is environmentally sensitive; thus , stably expressed oil-enhancing key genes that can be applied to soybean molecular breeding have rarely been reported , and the mechanism of the variance of oil content in soybean remains largely unknown . In plants , accumulated oil in seeds is generally stored as triacylglycerols ( TAGs ) . TAG synthesis is initiated from glucose in the cytosol , and the resulting products from glycolysis are transported into the plastid for fatty acid ( FA ) synthesis . The FAs are processed by a series of key enzymes to produce C16:0 and C18:0 acyl chains and desaturated products , such as C18:1 . FA products are then exported to the endoplasmic reticulum ( ER ) to form TAGs via the acyl-CoA-dependent and acyl-CoA-independent pathways [2] . The resulting TAGs are present in subcellular spherical lipid droplets in various plant tissues; the lipid droplets stored in seeds are usually called oil bodies ( OBs ) and have been extensively investigated previously in studies of , for instance , the structure and composition of an OB and the essential role of OB-related proteins , such as oleosins , in OB formation , mobilization , and oil accumulation [3–10] . Previous studies have indicated that oleosins play conserved roles in OB formation in seeds in several oilseed plants [7–8] . Suppression of a soybean oleosin produces micro-OBs [10] , while the effects of OBs on seed oil accumulation have rarely been reported in soybean . The soybean genome contains 13 putative oleosin-encoding genes [11] , and if any of them are involved in oil accumulation remain unexploited . By linkage and linkage disequilibrium mapping , over 300 quantitative trait loci ( QTLs ) associated with seed oil content have been identified across all 20 chromosomes in the soybean genome over the past decades ( SoyBase , https://soybase . org ) . These studies have revealed the polygenic nature of oil regulation , and the majority of loci were found to have varying additive , epistatic or QTL×environment effects [12–15] , implying that traditional breeding based on genetic crossing and phenotypic selection may be inadequate for oil improvement . Recent studies have shown that increased oil in soybean could be achieved by genetic engineering of transcription factors involved in oil accumulation [16–18] or a QTL gene controlling seed coat bloom [19] . However , QTLs directly related to seed oil accumulation in soybean have not been cloned; thus , the underlying mechanism has not been thoroughly elucidated to date . Therefore , identifying an environmentally stable major QTL regulating seed oil content is urgently needed to substantially enhance seed oil content and understand the underlying regulatory mechanism in soybean . To reveal the genetic basis of seed oil content and elucidate how oil accumulation is regulated , we investigated the oil content variation in 219 diverse soybean genotypes across six different environments and conducted a high-density genome-wide association study ( GWAS ) using 201 , 994 genome-wide single nucleotide polymorphisms ( SNP ) . In total , three QTLs were identified to be significantly associated with soybean oil content across at least two environments , with GqOil20 on chromosome 20 stably expressing across all six environments . We also found that an oleosin-encoding gene , GmOLEO1 , in the GqOil20 linkage disequilibrium ( LD ) block was exclusively expressed in developing seeds and that its expression level was significantly correlated with oil content within selected genotypes . We subsequently verified that GmOLEO1 contributed to oil accumulation in soybean seeds by conducting a series of molecular assays . Our results reveal an environmentally stable QTL/gene controlling oil accumulation in soybean seeds , provide new insight into oil accumulation in soybean and offer new directions for breeding soybean varieties with enhanced seed oil content .
To identify the genetic variation in seed oil content , we measured the oil contents of 219 soybean genotypes with diverse genetic backgrounds across six different environments . Seed oil content exhibited large amounts of natural variation within the association panel in each environment and showed relative consistency across the six environments ( S1A and S1B Fig , S1 Table ) . The mean oil content for the 219 accessions ranged from 18 . 10% to 18 . 97% across the six environments , and the observed maximum oil content reached 27 . 69% in Environment 1 ( E1 ) , which was approximately three times higher than the minimum value ( 9 . 64% ) observed in E6 ( S1A Fig , S1 Table ) . The distribution of oil content for the association panel in each environment was approximately normal ( S1C Fig ) . Analysis of variance ( ANOVA ) indicated a significant difference ( p < 0 . 001 ) in oil content among the genotypes , the oil content was significantly affected by environments ( p < 0 . 001 ) ( S1 Table ) , and the heritability was 0 . 64 . Because of the wide variation in seed oil content in the panel across the environments , we performed GWAS for the oil content in six environments ( E1 to E6 ) and the best linear unbiased prediction ( BLUP ) using 201 , 994 genome-wide SNPs with a minor allele frequency ( MAF ) ≥ 0 . 05 in an effort to identify the genetic loci associated with soybean oil content . In total , 110 SNPs on three chromosomes ( 8 , 12 , and 20 ) were identified as significantly associated with oil content across at least two environments ( S2 Fig , S2 Table ) . For the sake of simplicity , we empirically classified closely adjacent SNPs located within 5 Mb into one locus , as previously described [20] . The 110 SNPs were classified into three genomic loci , which were subsequently designated GqOil8 , GqOil12 , and GqOil20 ( S2 Fig , S2 Table ) . Of these QTLs , the most significantly associated SNPs were identified in GqOil20 , which was in physical proximity to oil-related QTLs identified in previous studies ( S2 Table ) [21–24] . Importantly , GqOil20 was consistently identified across all the environments and BLUP except E4 ( Fig 1A ) , and it explained 13 . 4–24 . 4% of oil variation , representing the most stably expressed QTL for oil content in soybean . It is known that oil content is a key domestication trait undergoing artificial selection [25] , and the regulatory genes involved were likely selected during domestication . Thus , a comparison of the genetic diversity at the three loci between cultivated soybean ( G . max ) and wild soybean ( G . soja ) , the progenitor of G . max , could be helpful in determining the most likely regions containing the oil-controlling gene ( s ) . To this end , we calculated genetic differentiation ( Fst ) within the 140 kb regions upstream and downstream of each leading SNP per locus within a group containing this association panel ( 272 G . max accessions ) and a panel of 122 G . soja accessions genotyped with the same microarray , as previously described [26] . After the comparison , we found that Fst showed variation , and the Fst across the entire group ( G . max and G . soja ) was lower than the average Fst in the association panel ( G . max only ) in two QTLs ( GqOil8 and GqOil12 ) . In contrast , most of the Fst values for GqOil20 were significantly higher in the G . max-G . soja group than in the association panel , suggesting that artificial selection might have occurred in this genomic region in relation to oil accumulation ( S3 Fig ) , consistent with the fact that soybean oil content is a domestication trait [25] . In this regard , GqOil20 likely harbors a gene or genes that have important functions in the regulation of soybean oil accumulation . Thus , we next focused on GqOil20 to identify oil-related genes . To identify the candidate gene , we analyzed the LD region harboring the leading SNPs using BLUP as a phenotype . GqOil20 contained a total of 33 significant SNPs located within a strong LD with an average r2 = 0 . 66 ( Fig 1C ) . Of these genes within the LD according to the G . max Wm82 . a2v1 reference genome ( https://phytozome . jgi . doe . gov ) ( S2 Table ) , we found that a gene , Glyma . 20G196600 , encoding a putative oleosin protein , colocated with the significant SNP AX-93661332 ( P = 4 . 98 × 10−10 ) ( Fig 1A and 1B , S3 Table ) . Glyma . 20G196600 is an ortholog of Arabidopsis AtOLE1 ( AT4G25140 ) , an oleosin-encoding gene with demonstrated roles in oil body formation [9] , while other gene models in this block are annotated to be involved in defense responses ( S3 Table ) . Thus , Glyma . 20G196600 might be the candidate gene underlying GqOil20 , and we designated it GmOLEO1 for further study . To investigate whether GmOLEO1 underlies the domestication region GqOil20 , we examined the expression patterns and sequence variations of GmOLEO1 alleles in 38 soybean accessions comprising 27 cultivated and 11 wild genotypes with significant differences in oil content between two subgroups ( Fig 1D ) . Consistent with the observed high oil content in G . max relative to G . soja , GmOLEO1 showed significantly higher expression in cultivated soybeans than in wild soybeans ( Fig 1D and 1E ) , indicating a correlation between the transcript abundance of GmOLEO1 and oil content . Next , a 2 . 3-kb genomic region extending from -1 , 500 bp upstream of the start codon ( ATG ) to the 3’-untranslated region ( UTR ) of GmOLEO1 was sequenced and analyzed . Sequence analyses identified 12 nucleotide variants that divided the 38 germplasm into six haplotypes ( Hap ) , which were clearly classified into two subgroups ( cultivated and wild ) by a phylogenetic tree ( Fig 1H ) . Moreover , the six haplotypes represented six levels of seed oil content ( Fig 1F and 1G ) , with Hap1 seeds containing the highest oil content . Of the 12 nucleotide variants , seven variants were found to be significantly associated with soybean oil content ( Fig 1F ) , with four located at -442 ( A/C/- , P = 5 . 22×10−4 ) , -281 ( G/C , P = 2 . 83×10−4 ) , -237 ( AAA/— , P = 1 . 26×10−3 ) , and -167 ( 13-bp insertion/deletion , P = 3 . 13×10−3 ) being detected in the promoter region and three ( PS26 = 5 . 36×10−3 , PS265 = 0 . 02 , PS393 = 0 . 04 ) occurring in the exon . Of these variants , those in the promoter region represent the most significant variation associated with the variation in seed oil , suggesting that resulting differences in the expression of GmOLEO1 among the haplotypes might account for the oil content variation . To further determine whether variation in the promoter affected gene expression , we compared the transcriptional activity of the promoters of Hap1 and Hap5 ( Hap1_pro and Hap5_pro ) using a dual luciferase reporter gene assay . As shown in Fig 1I , Hap1_pro exhibited 3 . 58-fold higher activity than Hap5_pro , consistent with the observed higher expression of Hap1 than Hap5 ( Fig 1I ) . These results suggest that GmOLEO1 is a strong candidate for GqOil20 and that expression level instead of exon variation is an important factor affecting seed oil content . Given that GmOLEO1 was a strong candidate associated with seed oil content , we characterized its protein structure , phylogeny , and expression pattern . BLASTp showed that GmOLEO1 is an ortholog of Arabidopsis AtOLE1 ( AT4G25140 ) , an oleosin-like protein with demonstrated roles in OB formation [9] ( Fig 2A and 2B ) . Similar to previously described OLE orthologs in other species , GmOLEO1 contains three conserved structural domains ( Fig 2A ) [5] . Two amphipathic domains are located at the N- and C-termini , respectively , and a hydrophobic domain is located at the center . In the central hydrophobic domain , GmOLEO1 also contains the conserved “proline knot” sequence ( PX5SPX3P ) , which can form a loop including a hydrophobic hairpin that penetrates into the TAG matrix and two arms located on both sides of the knot ( Fig 2A ) [5] . This domain organization allows oleosins to be anchored on the surface of an OB , as illustrated in a previous study [3] . Phylogenetic analysis revealed that OLEO-like proteins from the Faboideae , Brassicaceae , and grass clades clustered separately , suggesting functional conservation within the clade and possible functional diversity between clades ( Fig 2B ) . In addition , sodium dodecyl sulfate-polyacrylamide gel electrophoresis ( SDS-PAGE ) analyses showed that GmOLEO1 has a low molecular mass ( ~16 KD ) ( Fig 2C ) , consistent with previous findings [9 , 27] . These results indicated that the uncharacterized gene GmOLEO1 encodes a putative OB protein that might play roles associated with OB formation or oil accumulation in soybean . To determine the temporal and spatial expression pattern of GmOLEO1 , the expression levels of GmOLEO1 were examined in ten different tissues and two soybean varieties with different seed oil contents ( H101 , a high-oil variety; H112 , a low-oil variety ) ( Fig 2D ) . Quantitative real-time PCR ( qPCR ) results showed that GmOLEO1 transcripts were undetectable in nonseed tissues , including the roots , stems , leaves , and flowers of both varieties , but its transcripts could be detected in developing seeds beginning at the seed-filling stage ( Fig 2D ) . The abundance of GmOLEO1 transcripts in seeds increased with the number of days after flowering ( DAF ) , with the highest expression level observed in developing seeds at 40 DAF , which was immediately before that seeds had completely matured ( Fig 2D ) . Overall , the expression level of GmOLEO1 in the developing seeds of H101 was significantly greater than that in H112 seeds at all tested stages . These results indicated that GmOLEO1 functions specifically during seed maturation and that transcript abundance positively correlated with oil content ( Fig 1G ) . We next investigated whether GmOLEO1 was spatially related to OBs . We expressed a 35S::GmOLEO1-GFP ( green fluorescent protein ) construct in tobacco ( Nicotiana benthamiana ) leaf epidermal cells by agro-infiltration followed by staining with Nile Red , a lipophilic dye used to visualize OBs [4] . Confocal microscopy analysis revealed that GmOLEO1-linked GFP fluorescence and Nile Red fluorescence signal from OBs were colocalized in seed cells ( Fig 2E ) , indicating that GmOLEO1 was localized to accumulated OBs . Taken together , the results of haplotype analysis , diversity analysis , phylogenetic analysis , expression analysis , and subcellular localization supported the GWAS results and collectively indicated that GmOLEO1 was a strong candidate gene underlying GqOil20 associated with oil accumulation in soybean seeds . To further demonstrate whether GmOLEO1 is functionally involved in oil accumulation in soybean seeds , we overexpressed GmOLEO1 in soybean using an improved cot-node transformation protocol [28] . Successful transformation was determined by detecting both the expression of the selective bar gene using the strip test and the presence of 35S::GmOLEO1 ( Fig 3A ) using polymerase chain reaction ( PCR ) analysis in T0 plant leaves ( S4 Fig ) . Transgenic soybean lines were self-pollinated through three generations to obtain homozygous lines harboring 35S::GmOLEO1 . Three independent homozygous transgenic lines ( OE-9 , OE-16 , and OE-18 ) were selected and used for further analysis . We first quantified the expression of GmOLEO1 in developing seeds ( 10 , 20 , 25 , 30 , and 40 DAF ) . As shown in Fig 3B , GmOLEO1 expression increased during seed development in both the OE lines and wild type ( WT ) , while its expression in the OE seeds was significantly higher in OE than in WT at each stage of seed development . GmOLEO1 exhibited a sharp increase in its expression in the OE lines at 25 DAF , which was five days earlier than the increase in expression observed ( 30 DAF ) in WT seeds . In agreement with the observed expression difference between the two soybean varieties described above ( Fig 2D ) , expression of GmOLEO1 in both the OE lines and WT increased continuously as the seeds developed and reached its highest levels at 40 DAF . This gene expression result was further verified by comparative Western blot analysis of GmOLEO1 protein between WT and the OE lines using an antibody against GmOLEO1 . A higher expression of GmOLEO1 was observed in the three OE lines than in the WT at 25 and 40 DAF ( Fig 3C ) . Compared with WT , mature seeds from OE lines had shinier surfaces with more yellowish colors and smaller sizes ( Fig 4A ) . The oil contents of the seeds of the three OE lines were 22 . 35% , 21 . 91% , and 22 . 14% , respectively , which were all significantly higher ( an absolute average increase of 2 . 12% , a relative increase of 10 . 6% , P = 4 . 6 × 10−6 ) than that in WT seeds ( 20 . 01% ) ( Fig 4F ) . Not surprisingly , the increase in oil content in the OE lines resulted in a significant decrease ( P = 0 . 006 ) in protein content ( Fig 4G ) . To further verify the oil increase in the OE seeds , we conducted a series of microscopy analyses of developing OE seeds ( OE-9 and OE-18 ) at 25 DAF , where sharp increases in the expression of GmOLEO1 and oleosin were observed ( Fig 4C–4F ) . Microscopy analyses of cross-sections from developing seeds stained with Oil Red O showed that OE seeds have markedly stronger Oil Red O staining than WT seeds , indicating that OE seeds contain a higher level of neutral lipid accumulation than WT seeds ( Fig 4B ) . A further examination of the seed cells using an optical microscope ( Nikon , Eclipse Ci , Japan ) showed that more OBs were deposited in the two OE lines ( OE-9 and OE-18 ) than in WT ( Fig 4C ) , and a consistent result was found via staining with toluidine blue O ( Fig 4D ) . These observations were further verified by a comparative analysis of Nile Red staining of accumulated oil between OE and WT seeds using a confocal microscope ( Nikon , C2 , Japan ) ( Fig 4E ) . These results visibly illustrated that OE plants overexpressing GmOLEO1 contained higher levels of oil accumulation in seed cells than WT . In addition , we observed that overexpression of GmOLEO1 has pleiotropic effects on other agronomic traits . The phenotypic evaluation indicated that the overexpression resulted in a significant decrease in 100-seed weight in the OE lines compared with that in WT ( Fig 4I ) . However , a significant increase ( P < 0 . 01 ) in pod number per plant and a slight increase in plant height in the OE lines compared with WT lines ( Fig 4H and 4J ) were observed , which led to an increase ( P = 0 . 017 ) in seed yield per OE plant compared with WT plants ( Fig 4K , S4 Table ) . We also compared the seed germination between two lines . We found that seed germination and root growth were faster in the OE lines than in WT ( Fig 4L ) . These results indicate that GmOLEO1 is involved in oil accumulation in soybean seeds with pleiotropic effects on yield-related traits , and no yield penalty was found in the current preliminary study . In light of the role of GmOLEO1 in oil accumulation , we further measured and compared the fatty acids between WT and OE seeds to test whether GmOLEO1 affected FA composition ( Fig 5 ) . Compared with the WT , OE seeds contained a higher average total FA content of 12 . 7% ( P = 3 . 2× 10−4 ) . Further analysis of five important oil components ( TAGs ) indicated that two polyunsaturated oil components , linoleic acid ( 18:2 ) and linolenic acid ( 18:3 ) , were significantly increased by 14 . 4% and 14 . 9% ( P = 1 . 2× 10−4 and 9 . 7× 10−5 , n = 3 ) , respectively , in the OE seeds compared with WT , while no significant changes in the contents of palmitic acid ( 16:0 ) , stearic acid ( 18:0 ) and oleic acid ( 18:1 ) were observed between the OE seeds and WT ( Fig 5 , S5 Table ) . This result indicates that the overexpression of GmOLEO1 also led to increased accumulation of polyunsaturated FAs . Last , we compared the OBs of OE and WT seeds using transmission electron microscopy . At 25 and 40 DAF , the OBs of WT seeds showed typically spherical and ovoid structures and were distributed mostly between protein bodies at the periphery of the cells ( Fig 6 ) . In contrast , OE seed cells contained apparently smaller OBs than those of WT ( Fig 6 ) . To better understand the molecular mechanism by which GmOLEO1 increased oil accumulation in soybean seeds , we compared the transcriptomes of OE and WT seeds at three seed development stages ( 20 , 25 and 40 DAF ) using RNA-seq analysis . In total , 796 , 1238 , and 1417 differentially expressed genes ( DEGs ) were identified by comparing OE with WT seeds at 20 , 25 , and 40 DAF , respectively ( Fig 7A , S6–S8 Tables ) . The RNA-seq result was validated by qPCR analyses of 16 randomly selected genes ( S9 Table , R2 = 0 . 84 ) . We observed a trend toward increasing numbers of DEGs as DAF increased ( Fig 7B ) . This increasing trend in the number of DEGs is consistent with the pattern of oil content increase in seeds as DAF increases ( Fig 7B ) . This result indicated that overexpression of GmOLEO1 resulted in significant changes in the transcriptomes in the developing OE seeds , and the changes became more dramatic as the seeds developed . To understand the biological processes in which GmOLEO1 participates , we performed Gene Ontology ( GO ) enrichment analysis for these DEGs . In addition to the enrichment of GO terms associated with the regulatory pathways essential for plant growth and development , such as seed development and germination , amino acid and sucrose metabolism , and response to growth hormone , we found that GO terms associated with linoleic acid metabolism , fatty acid transport , lipid metabolism and storage were also significantly enriched for these DEGs ( Fig 7D ) . The RNA-seq results were further verified by the increased expression of several known genes participating in TAG biosynthesis in OE seeds as shown by qPCR , such as diacylglycerol acyltransferase ( DGAT1 ) [18] , wrinkled 1 ( WRI1 ) [29] , zinc-finger protein ( GmZF351 ) [17] , two Arabidopsis OLEO orthologs ( AtOLE2 and AtOLE3 ) [30] , and oil body associated protein 1 ( OBAP1A ) [31] ( Fig 7C ) , indicating that the expression of these genes may be affected by GmOLEO1 overexpression . These results indicated that overexpression of GmOLEO1 promoted the expression of TAG biosynthesis-related genes and led to the enhancement of TAG biosynthesis . Because higher expression of GmOLEO1 in cultivated soybean than in wild soybean was observed , we hypothesized that the variations in its promoter region were under selection . Statistical analyses were performed using a large population of 302 soybean accessions [24] . We first evaluated Fst for different comparisons , including wild soybean vs . cultivar , wild soybean vs . landrace , and cultivar vs . landrace . The results showed that the Fst between wild vs . cultivated soybean is considerably higher than that between cultivar vs . landrace , especially in the promoter region ( Fig 8A and 8B ) . The nucleotide diversity ( π ) analysis showed that π was higher in wild soybean than cultivated soybean in the promoter region and the coding region ( Fig 8A ) . Tajima's D in the promoter region was 2 . 00 , 0 . 06 and -0 . 84 for wild , landrace and cultivar , respectively , while Tajima's D in the coding region was 1 . 22 , -0 . 819 , and 0 . 04 for wild , landrace and cultivar , respectively ( Fig 8A ) , implying that positive selection had occurred in the promoter region . A phylogenetic analysis using variants in the promoter and coding regions identified three clusters , corresponding to wild soybean , landrace , and cultivar ( Fig 8C ) . Taken together , these results indicated that the promoter region was subjected to artificial selection during domestication .
It is known that seed oil content has been subjected to artificial selection targeting higher oil content [25] . This finding was further validated by our study , in which a significant difference in seed oil content between cultivated and wild soybeans was observed ( Fig 1D ) . Unlike other domestication traits in soybean , such as stem growth habit [32] and pod shattering [33] , soybean oil content is highly complex; it is regulated by many genes of small effect and is easily influenced by various environmental factors [1] . Our GWAS study across multiple environments allowed us to identify a new environmentally stable QTL , GqOil20 , and an underlying candidate gene , GmOLEO1 that is capable of increasing seed oil content in soybean . Notably , the GmOLEO1 locus was previously identified as a possible candidate for an eQTL associated with seed oil accumulation [34] , and it is physically close to other oil-related QTLs previously identified by linkage mapping [21–23] . GmOLEO1 may have been identified in this study because the corresponding alleles were fixed with respect to oil variation during domestication . Given the complexity of oil metabolism , the observed phenotypic variation ( 23 . 7% ) could be due to the combined effects of GmOLEO1 and other genes at this locus . Other oleosin genes , regardless of sequence variation , with increased expression at the gene or protein level [35–36] during seed filling/maturation may also substantially affect oil accumulation . Nevertheless , our study further functionally verified that human-selected GmOLEO1 might be involved in seed oil accumulation , possibly by indirectly affecting oil biosynthesis via efficient feedback . Our study and previous studies have indicated that the improvement of seed oil content in soybean during domestication was achieved by artificial selection of multiple major genes , and some of those genes may not be directly involved in oil biosynthesis , such as B1 [37] and GmZF351 [17] . In contrast to previous studies that identified oil-related genes using a reverse genetic approach , GmOLEO1 was pinpointed in an artificially selected locus , GqOil20 , using an integrated strategy of high-density genetic mapping and genomics . Haplotype and expression analyses of GmOLEO1 between cultivated and wild soybean in our study provided additional evidence of selection at the GmOLEO1 locus . This artificially imposed selection pressure on the expression of GmOLEO1 could be an important factor affecting the observed difference in oil accumulation in soybean , because the overexpression of Hap2 of GmOLEO1 resulted in enhanced oil accumulation in transgenic soybean ( Figs 1G and 4F ) . Whether other genes in this block have functions associated with oil accumulation requires further determination . Seed yield and quality represent two of the most important traits in soybean improvement . Breeding soybeans with high oil stability across environments while maintaining protein content and yield has been difficult due to the complex genetic architecture of oil regulation . In our study , GmOLEO1 was functionally identified as a candidate for the environmentally stable QTL GqOil20 . Overexpression of GmOLEO1 significantly elevated oil content and the percentage of polyunsaturated FAs without detriment to the overall plant performance , especially yield , in our preliminary study , making GmOLEO1 a promising candidate gene for use in breeding high-oil soybeans with improved levels of healthy polyunsaturated FAs . Although Hap2 ( Williams 82-type ) is not the most favored haplotype for increasing oil accumulation , enhanced seed oil accumulation was observed in this study , indicating that the GmOLEO1 allele in Hap2 enhanced oil accumulation independent of the amino acid substitution ( Ala265Pro ) . One possible reason for this finding is that the substitution , which does not change hydropathy , may not affect the secondary structure of oleosins in OBs . The strong correlation between Hap1 and high expression levels of GmOLEO1 alleles ( Fig 1 ) suggests the importance of unique variation ( Hap1 ) in the promoter region in enhancing the expression of GmOLEO1 . The discovery of the molecular function DNA marker , Indel P237167 , from the unique variation in the promoter of GmOLEO1 will facilitate marker-assisted selection ( MAS ) in soybean high-oil breeding programs . The importance of oleosins in lipid accumulation and oil body formation in seed plants has been gradually recognized over the past three decades , and it has demonstrated an important role in the maintenance of OBs and preventing them from coalescence [9] . The Arabidopsis genome contains 17 oleosin genes [38] , of which AtOLE1 has been reported to be involved in lipid biosynthesis [39] . In soybean , 13 putative oleosin genes were found in the G . max reference genome , while only GmOLEO1 colocalized with the associated SNP ( AX-93661332 ) in GqOil20 in our study ( S10 Table ) . The overexpression of GmOLEO1 showed consistent results , as observed in AtOLE1 , revealing conserved functions between GmOLEO1 and AtOLE1 in increasing oil accumulation . Despite being rarely studied in other species , the high similarity in amino acid sequence and structural domains ( Fig 2A and 2B ) suggests that GmOLEO1-like proteins from the Faboideae , Brassicaceae , and grass clades might have a conserved function in determining OB size but lineage-specific roles [7] . For example , expression of GmOLEO1 correlated with oil content in our study , and OE seeds with increased oil content contained smaller OBs; conversely , the expression of oleosin genes was independent of oil content in maize , and a high-oil maize strain contained larger , more spherical OBs than did low-oil maize [40] . The conserved role of OLEOs from various plant species in enhancing oil accumulation suggests that GmOLEO1 orthologs have considerable potential for oil improvement in other oil-producing crops . It has been demonstrated that oleosins have important functions in OB formation , stabilization , and transgenic addition of oleosin increased oil content in Arabidopsis , Brassica and yeast [10 , 31 , 41–42] , but oleosins’ role in increasing oil accumulation in soybean seeds has rarely been reported previously . In addition to these potentially cross-species functions in determining the size of OBs and affecting oil accumulation , our study showed that increased oleosins resulted in apparent reductions in OB size and increases in OB number in seed cells ( Fig 6 ) and increased seed oil contents ( Fig 2 ) , in agreement with a study in which suppression of OLEO1 resulted in larger OBs and reduced total lipid levels in seeds [9–10] . Oil accumulation was gradually regulated as the seeds matured , possibly due to a gradual increase in the expression of GmOLEO1 , which was concomitant with the enhanced TAG metabolism during seed maturation identified by RNA-Seq ( Fig 7 ) . Thus , it is logical that a positive correlation between oil content and GmOLEO1 expression was observed in our study ( Fig 1G ) . A similar correlation has also been observed in Brassica napus , where the expression levels of OLEO/oleosin in high-oil genotypes were considerably higher than those in low-oil seeds [41] . The higher expression levels of GmOLEO1 in high-oil soybean varieties might be attributed to the unique variation present in the promoter of Hap1 . The presence of putative seed-maturation-related cis-elements ( abscisic acid ( ABA ) response and seed regulation , Fig 1F ) in the promoter region of GmOLEO1 may be responsible for its exclusive expression during seed maturation . In addition to stabilizing the structures of lipid droplets ( LDs ) , oleosins also serve other functions , including enzymatic and signaling roles . Some of these proteins are ubiquitous in cells with and without LDs , thus exerting broader functions in seeds and other organs [43] . In peanut , oleosin3 ( OLE3 ) was shown to exhibit bifunctional activities and was phosphorylated by STYK ( AhSTYK ) to regulate MGAT and PLA2 activity; it could be involved in the biosynthesis and mobilization of TAGs during seed maturation and germination [44] . However , a recent report showed that the bifunctional enzymic motifs are present in only peanut oleosins and not in those of other plants [7]; thus , another possibility is that oil accumulation increases as a result of GmOLEO1 overexpression , which might lead to efficient feedback by producing smaller OBs [45] . The detailed mechanisms underlying the regulation of gene expression by GmOLE1 must be deciphered in future work . The level of oleosin itself is regulated during seed development and germination . When seeds germinate , oleosin degradation occurs prior to OB degradation . A recent study revealed that the ubiquitin binding protein PUX10 and division cycle 48 homolog A ( CDC48A ) are core components of an LD-associated ERAD-like degradation machinery , which facilitates the dislocation of oleosins from LDs [46–47] . In our study , faster seed germination of OE lines might be associated with higher levels of some oleosin-degradation proteins ( e . g . , PUX10 and CDC48A ) [46–47] , but this hypothesis remains to be experimentally determined . Based on the results of our preliminary study and a previous finding [9] , we proposed that the biosynthesis of TAGs was enhanced in the OE lines , possibly because of the affected TAG metabolic pathway , as a result of increased expression of GmOLEO1/oleosins . Smaller OBs gradually accumulated as the newly produced TAGs reached the minimum size that could be completely covered by the increased number of oleosin proteins , given that oleosins serve as surfactant to prevent OBs from coalescence [10 , 48] . Thus , increased oleosin production in OE seeds , which resulted in reduced size but increased turnover of OBs during seed maturation , could be a more efficient way to use the limited intracellular space than larger OBs , leading to increased total oil content in OE seeds .
The association panel for GWAS consisted of a diverse collection of 219 soybean accessions ( including 195 landraces and 24 elite varieties ) originating from 26 provinces across six different agroecological regions in China , ranging from latitudes 53 to 24°N and longitudes 134 to 97°E [49] . Field experiments were performed in the 2009 , 2011 , 2012 , 2013 and 2014 growing seasons at four different geographic locations as previously described [50] . Briefly , soybean plants were examined under field conditions at the following experimental stations: Jiangpu Experimental Station of Nanjing Agricultural University ( 32 . 1°N 118 . 4°E ) , Nanjing , in 2009 ( designated as Environment 1 , E1 ) ; Maozhuang Experimental Station ( 34 . 8°N 113 . 6°E ) of Henan Agricultural University Zhengzhou , in 2009 ( E2 ) and 2011 ( E3 ) ; the Fangcheng Experimental Farm ( 33 . 2°N 112 . 9°E ) of Henan Agricultural University in 2012 ( E4 ) , and Yuanyang Experimental Station of Henan Academy of Agricultural Sciences , Zhengzhou , in 2013 ( E5 ) and in 2014 ( E6 ) . A randomized block design was used for all field trials . In all environments , each accession was planted in a three-row plot , with each row 200 cm long and 50-cm row spacing . Mature soybean seeds were harvested and air-dried , and fully filled seeds were used for oil content measurement . Measurement of soybean oil , protein , and FA components was conducted using a near infrared spectrophotometer ( NIR ) seed analyzer ( DA7200 , Perten Instruments , Huddinge , Sweden ) as previously described [51] . This association panel was genotyped using the NJAU 355K SoySNP array as previously described [26] , and a total of 292 , 035 high-quality SNPs were used for association mapping . Phenotypic data for soybean seed oil across different environments were subjected to an ANOVA using the PROC GLM ( general linear model ) mixed model of SAS version 9 . 2 ( SAS Institute , 2002 ) . The linear statistical model includes the effects of genotype , environment and the environment × genotype interaction . The BLUP for each line was calculated with PROC MIXED in SAS ( SAS Institute , 2002 ) and used as the phenotypic input for the subsequent GWAS . The violin plot was drawn using the R package vioplot [52] . The heritability of oil content was calculated using h2 = Vg/ ( Vg+Ve ) , where Vg and Ve represent genetic and environmental variation , and each term was extracted from the ANOVA results . GWAS was conducted using the compressed mixed linear model with TASSEL 5 . 0 [53 , 54] using SNP with minor allele frequency greater than 0 . 05 , and the threshold was determined with Bonferroni threshold of ≤ 4 . 95 × 10−6 ( P = 1/n ) [55] , where n is the SNP number used in GWAS . The population structure and the relatedness were described previously [26] . The Manhattan plot was drown using the R package qqman [56] . The LD heat map was plotted using the LDheatmap R package [57] . Expression of the candidate gene was examined in different soybean tissues , including roots , shoots , leaves , flowers , pods , and developing seeds at different developmental stages ( 10 , 20 , 25 , 30 and 40 days after flowering ) . Total RNA was isolated from the tissues using the RNAsimple Total RNA Kit ( TaKaRa , Japan ) , and 1 μg of RNA was treated with 10 units of RNase-free DNase I ( TaKaRa ) prior to cDNA synthesis . The first strand of cDNA was synthesized using the SuperScript III First-Strand Synthesis System ( Invitrogen , USA ) following the manufacturer's instructions . Gene expression was determined using the Bio-Rad CFX96 Touch Real-Time PCR System ( Bio-Rad , California , USA ) . The PCRs contained 5 μL of the first-strand cDNA , 0 . 5 μL of 10 μmol L−1 gene-specific primers ( S11 Table ) , and 10 μL of Real-Time PCR SYBR Mix ( PC3302; Aidlab ) . The PCR conditions were as follows: 94°C for 3 min and 40 cycles at 94°C for 15 s and 60°C for 15 s . The soybean tubulin gene ( GenBank: AY907703 . 1 ) was amplified as an internal reference , and a negative control reaction was performed using water instead of cDNA . Three biological replicates per sample were used , and each reaction was performed in triplicate . In the protoplast transient expression experiments , the dual luciferase assay vector pGreenII 0800-LUC was used to analyze the activity of the different promoters . This vector contains a firefly luciferase ( LUC ) reporter gene that can be driven by the target promoter and a Renilla luciferase ( REN ) reporter gene driven by 35S . The purified DNA fragment of the target promoter was fused with the LUC reporter gene in the vector digested with HindIII and SalI enzymes to construct the recombinant vector . The vector pGreenII 0800-LUC without promoter insertion before the LUC reporter gene was used as a control . The recombinant vector and the control were individually transformed into Arabidopsis protoplasts via PEG-calcium transfection . The isolation of Arabidopsis protoplasts and protoplast culture were performed according to standard protocols [58] . The ratio of LUC and REN activity ( LUC/REN ) was used to reflect the activity of the target promoter . The LUC/REN value was determined using the dual luciferase reporter assay system ( Promega , USA ) . The complete coding sequence of GmOLEO1 was amplified from the cDNA of Williams 82 by regular PCR using gene-specific primers ( S11 Table ) . The PCR product was subcloned into the pMD-19 T vector ( TaKaRa , Japan ) for sequence verification . The verified GmOLEO1 sequence was then cloned into the dicotyledon expression vector pCAMBIA3300 , which contains a selection marker gene , phosphinothricin acetyltransferase ( bar ) , using the ClonExpress Entry One Step Cloning Kit . The resulting recombinant pCAMBIA3300-GmOLEO1 construct was transformed into Williams 82 via the Agrobacterium tumefaciens-mediated soybean cotyledon node transformation system as previously described [59] . Extraction of genomic DNA from the leaves of PPT-resistant plants and nontransformed plants was performed using the cetyltrimethylammonium bromide ( CTAB ) method [60] . Transformants were verified by leaf-painting assay with herbicide phosphinothricin ( PPT ) , PCR analysis for the presence of introduced GmOLEO1 and bar ( 482 bp ) , and LibertyLink strip detection for the expression of the bar gene using the QuickStix Kit ( EnviroLogix Inc . , ME , USA ) were considered positive transgenics for further analysis . For LibertyLink strip detection , a total of 100 mg leaf tissue was collected and ground completely in the bottom of a conically tapered 1 . 5 ml tube by pestle rotation , followed by adding 0 . 5 mL of extraction buffer and a strip into the tube . After ten minutes , strips containing only the control line were negative for PAT protein expression , while those with two lines ( control line and test line ) were positive for PAT protein expression [61] . The full-length GmOLEO1 cDNA was amplified and cloned into the pBWA ( V ) HS-osgfp vector to obtain the pBWA ( V ) HS-osgfp-35S::GmOLEO1-GFP construct under control of the cauliflower mosaic virus ( CaMV ) 35S promoter ( Biorun Co . , Ltd ) . The binary vector 35S::GmOLEO1-GFP was transiently coexpressed in the leaves of Nicotiana benthamiana via agro-infiltration . Then , the tobacco leaf epidermal cells agro-infiltrated with the GmOLEO1-GFP construct were stained with Nile Red , a lipophilic dye used to visualize OBs [4] . Fresh leaves were placed in a solution containing Nile Red stock ( 100 mg/mL dimethyl sulfoxide ) diluted 100× with 1×PBS for 10 min and washed with PBS twice for 30 s each time . Fluorescence signals were detected using a confocal laser scanning microscope ( Nikon C2-ER , Japan ) 2–3 days after infiltration . GFP , mKate and Nile Red were excited at 488 , 561 and 559 nm , and their emission was detected at 510 to 540 , 580 to 620 and 570 to 670 nm , respectively . All of these fluorescence experiments were independently repeated at least three times . Immunogenicity peptides of GmOLEO1 protein were predicted by bioinformatics analysis . The sequences of the peptides were as follows: MAELHYQQQHQYPHR and KDYGQQQISGVQAS . The peptides were commercially synthesized and purified ( Wuhan GeneCreate Biological Engineering Co . , Ltd , China ) . Two male Japanese White rabbits were used for the immune procedure . Next , a polyclonal antibody of GmOLEO1 protein was separated and purified for immunoblot analysis . Proteins of fresh soybean seed were extracted by Triton X-100 lysate ( 0 . 5% ) . Then , 30 μg of protein extracts mingling with 2× SDS-PAGE sample loading buffer ( Solarbio , Beijing , China ) were loaded and subjected to SDS-PAGE . Afterward , protein bands were transferred onto polyvinylidene fluoride ( PVF ) membranes ( Solarbio , Beijing , China ) . The membranes were blocked with 5% skimmed milk powder solution for 2 h at room temperature , followed by incubation with a polyclonal antibody against GmOLEO1 diluted to 1:10000 in phosphate-buffered saline overnight at 4°C . Finally , the blot was detected with horseradish peroxidase ( HRP ) -conjugated goat-anti-rabbit secondary antibody ( Santa Cruz Biotechnology , USA ) for another 1 h . The protein bands were visualized using a chemiluminescence system ( Pierce , Rockford , Illinois , USA ) . Transcriptomes were compared between pooled OE seeds at 20 , 25 , and 40 DAF , respectively , with the WT seeds at the corresponding developmental stage . For each time point , two developing seeds from each of the three OE lines ( OE-9 , -16 and -18 ) were collected and pooled as one biological replicate , and three biological replicates were used per sample . Library construction was performed as previously described [62] . The library was sequenced with the Illumina HiSeq 2500 analyzer at Biomarker Technologies ( Beijing , China ) , producing 200-bp paired-end reads . An average of 6 . 47 gigabases of clean data per sample was generated . Differential gene expression was determined using the DESeq R package [52] . A gene with an adjusted P < 0 . 05 and a fold change ( FC ) >1 . 5 were defined as DEGs . Enrichment analysis of Gene Ontology of biological pathways ( GOBPs ) was performed using the GOseq R packages [63] to compute P values that indicate the significance of each GOBP being represented by the genes . GOBPs with P < 0 . 01 were identified as enriched biological processes . Fresh immature soybean seeds harvested at 25 and 40 DAF were fixed in FAA fixation solution for at least 24 h . The main experimental steps for Oil Red O staining are as follows: cutting the whole sample into small blocks , removing excess water with tissue paper , immersing the small tissue blocks in Oil Red O ( Servicebio , G1016 , Wuhan , China ) solution and incubating at 37°C for 60 min . Excess staining solution was removed by rinsing with tap water . The stained tissue blocks were immersed in 75% ethanol for 30 min or until no fading occurred; then , they were preserved in 4% paraformaldehyde and kept in the dark . Photos were taken using a digital camera ( Canon 7D ) . The fixed tissue samples were embedded with OCT compound ( Sakura , Japan ) . Frozen sections ( 8–10 μm ) were obtained with Cryostat Microtome ( Thermo , CRYOSTAR NX50 , USA ) and mounted on a prechilled glass slide . The frozen sections were stained with 0 . 1% Nile Red ( Servicebio , G1073 , Wuhan , China ) and Oil Red O ( Servicebio , G1016 , Wuhan , China ) . Image observation for Nile red staining was performed using a Nikon confocal scanning microscope ( Nikon , C2 , Japan ) . The excitation wavelength was 488 nm , the emission wavelengths were 593–654 nm , and the OBs were imaged at 800× magnification . Oil Red O staining was imaged using an optical microscope ( Nikon , Eclipse Ci , Japan ) at 800× magnification . Tissues ( 1×3 mm3 in size ) of developing soybean seeds were fixed in 2 . 5% glutaraldehyde buffered with 0 . 1 M phosphate buffer ( pH 7 . 2 ) for 12 h . Postfixation was subsequently conducted in 1% osmic acid in 0 . 1 M phosphate buffer ( pH 7 . 2 ) for 5 h . The blocks were then washed , dehydrated through an ethanol series of 30–100% , and embedded in EMbed 812 media . The samples were cut into 1 μm slices using an ultramicrotome ( Leica UC7 , Germany ) , stained with alkaline toluidine blue O solution ( Servicebio , G1032 , Wuhan , China ) , and then imaged ( 800× ) using an optical microscope ( Nikon , Eclipse Ci , Japan ) . For transmission electron microscopy ( TEM ) , the samples were cut into 60 nm slices using an ultramicrotome ( Leica UC7 , Germany ) and then separately stained with uranyl acetate and lead citrate for 15 min . The slice samples were photographed under a TEM ( HT7700 , Hitachi , Japan ) . The 2 . 3-kb genomic region spanning from 1 , 500 bp upstream from the translation start codon ( ATG ) to the 3’-untranslated region ( UTR ) of GmOLEO1 was sequenced and analyzed . Haplotype analysis was performed by resequencing this region in 20 high-oil , 20 low-oil and 10 moderate-oil accessions . All primers ( S11 Table ) used in this study were designed using the Primer 3 online tool ( http://frodo . wi . mit . edu/primer3/ ) . All sequences were verified manually , and all observed polymorphisms were reverified by resequencing of another amplicon . All the verified sequences were aligned using ClustalX version 1 . 83 [64] . The polymorphism data were analyzed using DnaSP version 4 . 10 [65] to identify sequence variation . Prediction of cis-elements in the promoter region was carried out using the online web tool PlantCARE [66] . FA components in soybean seeds were analyzed as previously described [17] . Briefly , 10 mg fine powder of soybean seeds was used for FA isolation . FAs were extracted with 1 mL of extraction buffer ( 2 . 5% [v/v] H2SO4 in CH3OH ) at 85°C for 1 h . The supernatant ( 500 μL ) was mixed with 300 μL of hexane and 600 μL of 0 . 9% ( w/v ) NaCl . FA methyl esters were redissolved in 200 μL of ethyl acetate and analyzed immediately with a gas chromatography system ( GC-2014; Shimadzu , Beijing , China ) . Peaks corresponding to each FA species were identified by comparison to a FA methyl ester analytical standard ( Supelco , Poole , UK ) . Concentrations of FA species were normalized against the internal control heptadecanoic acid ( Sigma-Aldrich , USA ) . Five biological replicates per line were analyzed in this experiment . The seeds were surface-sterilized with chlorine gas for 4 h prior to germination in darkness in Petri dishes ( 90 mm in diameter ) on two sheets of filter paper moistened with deionized water ( 15 seeds per Petri dish ) . Germination tests were carried out in an incubator ( MGC-400B , YIHENG , Shanghai , China ) equipped at 25°C with 75% humidity . The filter paper was replaced once a day , and germinated seeds with healthy roots were counted . Root length was measured using a ruler at 2 , 3 and 7 days postgermination . Three replicates per treatment were performed . The published whole genome sequencing data were used for gene diversity analysis [25] . VCFtools was used to estimate gene diversity ( Fst , nucleotide diversity ( π ) and Tajima’s D ) [67] . SNPRelate combined with APE was used to construct the phylogenetic tree [68 , 69] .
|
Soybean seed oil is an important quality trait targeted during domestication and breeding . However , the molecular mechanism of soybean oil regulation is largely unknown due to its complex genetic architecture and environmental sensitivity . In this paper , we integrated GWAS across multiple environments , haplotype analysis , genetic transformation , and diversity analysis to study the genetic architecture of oil content and the underlying mechanism in soybean . This combined analysis enabled us to identify an environmentally stable QTL ( GqOil20 ) and functionally verified that GmOLEO1 positively regulates total seed oil accumulation in soybean seeds . In addition , we found that GmOLEO1 showed a higher level of expression in cultivated soybean seeds than in wild soybean seeds , possibly as the result of the positive selection of the promoter , resulting in seed oil accumulation . Moreover , we identified an elite GmOLEO1 haplotype that correlated strongly with high oil content in soybean , holding great potential for assisting oil improvement in soybean breeding . Our study provided a new genetic resource for oil content improvement in soybean and other oilseed crops .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Material",
"and",
"methods"
] |
[
"genome-wide",
"association",
"studies",
"plant",
"anatomy",
"quantitative",
"trait",
"loci",
"plant",
"products",
"genetic",
"mapping",
"plant",
"science",
"oils",
"genome",
"analysis",
"crops",
"lipids",
"genomics",
"crop",
"science",
"gene",
"expression",
"seeds",
"genetic",
"loci",
"agriculture",
"vegetable",
"oils",
"biochemistry",
"haplotypes",
"agronomy",
"soybean",
"heredity",
"genetics",
"biology",
"and",
"life",
"sciences",
"computational",
"biology",
"human",
"genetics"
] |
2019
|
Artificial selection on GmOLEO1 contributes to the increase in seed oil during soybean domestication
|
An estimated 2 million inhabitants are infected with Chagas disease in Mexico , with highest prevalence coinciding with highest demographic density in the southern half of the country . After vector-borne transmission , Trypanosoma cruzi is principally transmitted to humans via blood transfusion . Despite initiation of serological screening of blood donations or donors for T . cruzi since 1990 in most Latin American countries , Mexico only finally included mandatory serological screening nationwide in official Norms in 2012 . Most recent regulatory changes and segmented blood services in Mexico may affect compliance of mandatory screening guidelines . The objective of this study was to calculate the incremental cost-effectiveness ratio for total compliance of current guidelines from both Mexican primary healthcare and regular salaried worker health service institutions: the Secretary of Health and the Mexican Institute for Social Security . We developed a bi-modular model to analyze compliance using a decision tree for the most common screening algorithms for each health institution , and a Markov transition model for the natural history of illness and care . The incremental cost effectiveness ratio based on life-years gained is US$ 383 for the Secretary of Health , while the cost for an additional life-year gained is US$ 463 for the Social Security Institute . The results of the present study suggest that due to incomplete compliance of Mexico’s national legislation during 2013 and 2014 , the MoH has failed to confirm 15 , 162 T . cruzi infections , has not prevented 2 , 347 avoidable infections , and has lost 333 , 483 life-years . Although there is a vast difference in T . cruzi prevalence between Bolivia and Mexico , Bolivia established mandatory blood screening for T . cruzi in 1996 and until 2002 detected and discarded 11 , 489 T . cruzi -infected blood units and prevented 2 , 879 potential infections with their transfusion blood screening program . In the first two years of Mexico’s mandated program , the two primary institutions failed to prevent due to incomplete compliance more potential infections than those gained from the first five years of Bolivia’s program . Full regulatory compliance should be clearly understood as mandatory for the sake of blood security , and its monitoring and analysis in Mexico should be part of the health authority’s responsibility .
Chagas disease is caused by the unicellular parasite Trypanosoma cruzi , capable of movement directly from one person to another via blood transfusion , organ transplant , or maternal-fetal transfer [1 , 2] . Although the most prevalent mode of transmission is via the excreta of infected reduviid bugs , where vectors are not present , iatrogenic trypanosomiasis is considered the most important [3–5] . An estimated minimum 10 million individuals are infected worldwide with corresponding incidence of 41 , 200 cases per year [6] . Approximately 99% of inhabitants infected with Chagas disease ( CD ) reside in Latin America , where between 25 and 90 million persons are at infection risk via one of the multiple infection modes . The disease burden for CD in the Latin American and Caribbean region , based on disability-adjusted life-years ( DALYs ) is five times greater than malaria , and is approximately one-fifth that of HIV/AIDS [6 , 7] . Despite overall prevalence estimates for the Latin American region , there are an estimated 1 . 1 to 2 million Mexicans infected with T . cruzi [8–11] , with highest estimated prevalence in the southern half of the country [12] . Rural to urban population migrations in the last decades , have provoked largely unplanned urban development and landscape modifications surrounding cities , which are important amplifiers of zoonotic hosts and pathogens , and improved opportunities for 32 triatomine species to persist [12] . More than half of the T . cruzi infected vector-exposed Mexican population now lives in urban areas . Infected inhabitants are rarely diagnosed for T . cruzi infection since there is an overall lack of epidemiological surveillance for its transmission or for disease , and if an infection is detected due to blood donation screening , patients are rarely treated with anti-parasitic drugs [13] . Clinical and public health personnel have little knowledge regarding Chagas disease ( CD ) , its transmission , clinical diagnosis , or treatment , due to neglect by healthcare system policies . Most individuals with T . cruzi infection or Chagas disease ( CD ) are asymptomatic or symptomatic without clinical recognition of etiology ( cardiac insufficiency or megaviscera ) , and unaware , as are healthcare personnel , of potential blood transfusion risk [14] . Third level hospitals in Mexico City report from 0 . 37% ( National Institute of Cardiology ) [15 , 16] to 0 . 17% ( National Institute of Pediatrics ) [17] of blood donations with antibody to T . cruzi . In contrast , 7 . 7% of blood donations from the Puebla Mexican Institute for Social Security ( IMSS ) have antibodies to T . cruzi [18] . In some Mexican blood banks , T . cruzi seroprevalence is higher than that of HIV , Hepatitis B , and Hepatitis C , corresponding more closely to the high seroprevalence detected in Mexican populations in the US [19–21] . There are twice as many blood donations from urban ( > 10 , 000 inhabitants ) as compared to rural populations in Mexico , which implies the need to adjust overall seroprevalence accordingly when these estimates are extrapolated to open population . The vast majority ( > 90% ) of T . cruzi infections in Mexico are in fact detected by blood donation screening , with the exception of those cases detected by research groups . Interrupting blood transfusion of T . cruzi depends upon effective donor or blood donation screening . Guidelines formulated in 1994 by Mexico’s national legislation , the “Official Mexican Standard for disposition with therapeutic aims of human blood and its components ( NOM-003-SSA2-1993 ) ” , mandated blood screening for T . cruzi “if” donors resided in CD endemic areas [22] . However , endemic areas were not defined by this legislation , and at that time little if any cases were reported due to a lack of epidemiological surveillance . Most recent guidelines ( NOM-253-SSA2-2012 ) replace those from 1994 , and now mandate nationwide T . cruzi blood donation screening , using tests with at least 95% sensitivity and specificity , as established by the National Institute for Diagnostics and Epidemiological Reference ( Instituto Nacional de Diagnóstico y Referencia epidemiologica , InDRE ) [23] . Positive blood units detected by screening tests are discarded for therapeutic use , although they must be tested with two tests by approved reference laboratories . There has been no evaluation of the impact of the new guidelines on screening efficacy , costs , life-years gained , or CD case detection ( epidemiological or clinical follow-up ) . The objective of the present study has been to fill that gap and analyze the impact of complete vs . incomplete compliance of the new guidelines for the Secretary of Health ( MoH ) and for the Mexican Institute for Social Security ( IMSS ) . Combined , these two institutions attend approximately 70% of the Mexican population [24] , while the former is also normative and heads the primary prevention and health care programs for vector-borne diseases in the country .
Two scenarios were developed for each health institution , based on current documented estimates , and for 100% compliance ( NOM-253-SSA2-2012 ) . The first scenario reflects the known status of non-compliance , assigned based on donation center response to a screening questionnaire conducted in 2007 and categorized as “not all are screened and not all positives are confirmed” . The second scenario considers complete compliance of current guidelines from the category “all are screened and all positives are confirmed” . An analytical model for compliance and costs was constructed using two modules: 1 ) a decision tree for the most common donation screening algorithm based on most common practices at donation centers from each institution , and 2 ) a Markov transition model simulating the natural history of the illness and a standard care protocol for both institutions . The model of natural and/or clinical evolution of the illness is an application of the model developed previously by the group [11] . Professional software was used to construct the models ( TreeAge Software , Williamstown , Massachusetts ) , the first of which is divided into two parts: decision trees from each blood donation center where a screening assay is conducted , and the follow-up procedure for confirmation of positive samples . The model structure of decision trees for both health institutions are illustrated in Fig 1 . Confirmation of MoH positive donations is conducted at state public health diagnostic laboratories , and all positive samples , in addition to 10% of negatives , are retested at the InDRE for quality control ( National Institute for Diagnostics and Reference , a part of MoH , located in Mexico City ) . Confirmatory tests for IMSS samples are run in-house at one of four centralized reference laboratories ( two in Mexico City , one in Guadalajara , and one in Monterrey ) . A decision tree was developed separately for each institution , since confirmation procedures were not the same . Parameters for each scenario and health institution are summarized in Table 1 . Infected recipients of undetected blood units enter the natural history of the disease , modeled by the Markov transition module . The model assumes that infected donors are in the indeterminate asymptomatic phase of CD ( or their health status would have excluded them upon initial screening interview ) and that the prevalence of infected donors is the same as that of the general population . Independent of whether an infected individual has or not been diagnosed for T . cruzi , the person enters an additional Markov model module for disease evolution [11] . This Markov module has five health phases: acute , chronic asymptomatic , symptomatic chronic phase , no progression phase , and death . Each time-step length is one month in the acute phase and one year for later phases . Changes in time steps are managed as follows: in the acute stage , each time step represents one month by introducing monthly transition probabilities , whereas the time of life accumulated , runs as 1/12 per cycle . Similarly , the discount rate runs with the time divided by 12 . The chronic asymptomatic , symptomatic chronic , and no progression phases are driven by annual transition probabilities and with annual accumulated time of life , as well as the discount rates . A middle-step correction was introduced in the model . Infected donors in the asymptomatic phase , are randomly distributed across the average duration of the phase . Infected recipients enter the model in the acute phase . The simulation runs until the entire cohort dies ( Fig 2 ) . Donors in the model are characterized by age and their infection status ( apriori assigned ) ; a person may be either uninfected ( truly not infected with T . cruzi ) or infected ( truly infected with T . cruzi ) . The infection status of donors is assigned randomly based on national population prevalence . The model identifies blood units as true positives or as false negatives , depending on results from the screening tests . Donors to be screened are selected randomly based on the screening rates for each scenario . If a donor is detected positive , the model assumes that the person will begin specific anti-parasitic drug treatment . When a person is infected by an undetected infected blood unit , both the donor and the recipient remain undiagnosed and continue with the natural history of the disease . Screening tests modeled for MoH were those most frequently used by donor centers and reported to the National Blood Transfusion Center . These were a recombinant antigen indirect immunofluorescence assay ( Architect Abbott ) , a recombinant antigen ELISA ( ChagasScreen Plus ) , and a crude antigen ELISA ( Chagatest Wiener Lab ) [42] . Screening tests modeled for the IMSS were chemiluminescence ( PRISM Abbott ) and a recombinant antigen ELISA ( ChagasScreen Plus ) , which were the most frequently used by in-house donation centers and registered with the IMSS Medical Infrastructure Planning Coordination [43] . Donors with positive results in the screening tests were randomly selected for confirmatory tests based on the confirmation rate for each scenario . The confirmation procedure for MoH ( InDRE ) consisted of two simultaneous tests , a crude antigen ELISA and an indirect hemagglutination test ( Interbiol ) . If the tests were discordant , both were run a second time , and if the tests persisted discordant , a Western Blot test ( bioMérieux ) was run . The criterium for a positive sample was that two out of three tests be positive . The confirmation test for IMSS was a single lysate ELISA ( BioChile Chagas ELISA II ) . The comparative performance between current and complete compliance was analyzed by comparing Chagas-specific mortality , new infections produced , and the incremental cost-effectiveness ratio of life-years gained . Percentage of blood screening by the Mexican Institute of Social Security was 87% and its confirmation rate was 99% , whereas for the Ministry of Health 40% of donations screened and 39% confirmed ( based on a 2007 survey ) . Total cost is the sum of direct costs for medical care and indirect costs . Only the monetary value of work days lost was considered based on a modified social perspective . All costs are expressed as the 2014 value of the US dollar . Effectiveness variables generated were life-years gained and cases detected . Both costs and effectiveness variables were discounted at 5% per year . A second order Monte Carlo simulation was used to simulate a cohort of 100 , 000 donors with 500 different sets of parameters for recalculations; 100 , 000 donor screening outcomes were obtained from 500 replicates using random sampling of the distributions assigned to each parameter . All parameters used to feed the model were introduced as statistical distributions: cost inputs are set as gamma distributions and the effectiveness and probabilities of transition are beta distributed . The Monte Carlo method , an alternative to analyze sensitivity , first selects a random set of input data values drawn from their individual probability distributions . These values are then used in the simulation model to obtain certain model output variable values . The result is a probability distribution of model output variables and system performance indices which result from variations and possible values of all input values [44–45] . Since all distributions are sampled in a Second Order Monte Carlo calculation , no independent sensitivity analysis was necessary .
The sum of costs for screening and confirmation tests , healthcare , and labor costs due to work days lost for detected and undetected cases , and blood costs per 100 , 000 donors , is US$ 23 . 2 million dollars for the MoH . Healthcare and labor costs of undetected cases are 62 . 9% of the total cost , 18 . 3% correspond to healthcare and labor costs of detected cases , 18% to blood cost , and the remaining to screening and confirmation tests . If there is complete compliance , the total cost is US$ 31 . 6 million , 36% greater than incomplete compliance . Healthcare and labor costs of detected cases represent 83 . 8% of the total cost for 100% compliance ( Table 2 ) . The total cost of the current compliance for IMSS is US$ 32 . 7 million , 71 . 6% of which is due to healthcare and labor costs of detected cases , 12 . 8% due to blood donation costs , 7 . 9% to healthcare and labor costs of undetected cases , and 7 . 7% due to screening and confirmation tests . The cost of complete compliance for IMSS is US$ 34 . 3 million , 5% greater than current incomplete compliance ( Table 2 ) . Effectiveness for all compliance scenarios and for both institutions are summarized in Table 3 . In the current scenario for MoH , 190 cases are confirmed , there are 157 new T . cruzi infections detected , and 4 , 195 life-years are gained . If the MoH attains 100% compliance , 1 , 185 cases are confirmed ( 1 , 105% increase ) , 3 new T . cruzi infections are identified ( 154 new T . cruzi infections avoided ) , and 26 , 079 life-years are gained , which is 5 . 2 times greater the life-years gained . A 15% increase in the number of confirmed cases identifies 28 additional T . cruzi infections avoided ( 93 . 3% ) , and 15% of life-years gained were identified from complete compliance in IMSS . The incremental cost effectiveness ratio ( ICER ) for case detection by MoH is US$ 54 , 438 and US$ 383 for each life-year gained . The ICER for an additional case detected by IMSS is US$ 56 , 744 and US$ 463 for each additional life-year gained . The cost-effectiveness acceptability curves ( CEAC ) for the simulations suggest that willingness to invest is attractive above US$ 500 per year of life gained and US$ 8 , 000 per new case detected , based on 80% of cases falling below these thresholds ( Fig 3 ) . The Mexican government is willing to pay if the effectiveness unit is equivalent to the per capita value of the National Gross Domestic Product ( GDP ) . The current Mexican GDP is approximately US$ 9 , 300 , although a lower willingness to pay per unit of effectiveness is desirable for low and middle-income countries [46] .
Serological screening of blood donations or donors for T . cruzi was mandated historically after 1990 in certain Latin America countries . Coverage of transfusion blood screening expanded to all Southern Cone Initiative countries after 1991 , to some Central American countries after 1997 , to most Andean Initiative countries after 1999 , and to the Amazonian basin countries after 2004 [47–48] . Blood donation screening for T . cruzi in the United States became mandatory in 2011 , before that in Mexico . Despite the fact that Mexico signed international agreements along with other countries and the World Health Organization to strengthen national blood banks and health policies to ensure safe blood supply , T . cruzi infected blood units were transfused in Mexico prior to 2007 with minimal blood screening ( < 30% ) . Despite the fact that legislation for donation screening in Mexico was only approved finally in 2012 , there is no information , monitoring or independent validation of screening compliance , or regarding infected-population follow-up . Incomplete compliance of Mexico’s national transfusion blood screening legislation affects costs and health outcomes , and hence should be analyzed using modified social and economic perspectives . In 1991 , a World Health Organization ( WHO ) expert committee recommended the use of either a single indirect hemagglutination test ( IHA cutoff at 1:8 ) or a single latex agglutination test for donor or donation screening [49] , while the Pan American Health Organization ( PAHO ) advocated in 1994 for the parallel use of at least two different serological tests for all donations [50] . However , in 2002 , another WHO expert committee recommended a single enzyme-linked immunosorbent assay ( ELISA ) to screen blood donors or donations [51] , while PAHO recommendations and other guidelines from Brazil [52] , Chile [53] , and Spain [54] suggested once again the use of two simultaneous different serological techniques run in parallel for T . cruzi screening ( one of which should be an ELISA ) . The basis for this latter recommendation was that although ELISAs may occasionally give false positive results , they are the most sensitive , and confirmation could be run using a second confirmatory test [55] . Alternatives to existing serology have been developed and immunochromatographic test strips ( ICS ) , also known as rapid tests , have recently been compared for primary healthcare level and blood bank use , given their lower cost and simplicity of use [56] . In most cases , rapid tests cost less than US$2 to the end user and a product cost of approximately US$0 . 25 . However there have been few studies across indigenous and mestizo populations of Latin America to measure sensitivity , specificity , and agreement with existing serological assays , and none with joint analysis of cost and effectiveness [57] . Quantitative parasitological diagnosis of infection in patients is currently advancing rapidly with real time PCR [58–60] , although validation needs to include all ethnic populations , infection and disease phases , and economic scenarios , according to targeted use ( blood donation , early population-based diagnosis , chronic patients , congenital transmission , treatment efficacy ) . Current Brazilian guidelines recommend molecular screening only when serological tests are inconclusive [61] . Although control of T . cruzi transfusion transmission is an integral component of all CD prevention and control programs , few studies analyze costs or effectiveness of blood donor or donation screening , and none have analyzed both under different compliance scenarios . A Markov model has been used to estimate annual cost per person ( US$ 4 , 660 ) and that for lifetime care ( US$ 27 , 684 ) across countries with vector and non-vector transmission [62] . Bolivia established mandatory blood screening for HIV , hepatitis B , hepatitis C , and T . cruzi , and between 1996 and 2002 , 11 , 489 T . cruzi -infected blood units were detected and discarded , and 2 , 879 potential infections prevented [63] . The cost of discarding one infected unit was US$ 96 and for preventing one potential infection was US$ 385 . Blood donation screening to detect a positive CD case in Mexico is more expensive than in Bolivia , principally due to lower T . cruzi seroprevalence . The cost for preventing one additional potential infection in Mexico was estimated to be US$ 55 , 000 for the MoH , calculated along with social costs , the most important case cost component . It is important to note that blood product recipients are generally high risk , and may be even immunosuppressed , thereby having potentially early CD symptoms . In principle , these patients could be monitored , diagnosed and treated , which would lower cost estimates . However , in practice , T . cruzi infection induced by blood transfusion is not suspected or monitored due to lack of training or education regarding this neglected disease [11] . Mexican populations not included in this study were federal and state civil servants , public sector institutions ( PEMEX ) , the armed services , and private health service providers . All but the latter two would have compliance equivalent to that of MoH , since MoH institutions are their primary provider of transfusion blood screening and confirmatory testing . The Mexican armed services and private health providers are reportedly screening blood donations at a rate similar to or greater than IMSS . Considering the number of blood units donated in 2012 , and assuming equivalent compliance and a single blood unit per donor [64] , present data indicate that during 2013 and 2014 incomplete compliance of national legislation by the MoH failed to confirm 15 , 162 T . cruzi infections , did not prevent 2 , 347 avoidable infections , and lost 333 , 483 life-years . The IMSS failed to confirm 2 , 184 T . cruzi infections , prevent 392 avoidable infections , and lost 47 , 986 life-years over the same two year period . Incomplete compliance and lack of oversight by the National Health Council for national blood transfusion legislation passively allows an avoidable economic burden for the population , principally due to work days lost . The current cost in Mexico due to healthcare per CD patient is around US$ 2 , 540 , and the cost to the patient , due to work days lost , is approximately US$ 7 , 620 [11] . One of Mexico´s two principal health care institutions falls significantly short of blood donation screening compliance for T . cruzi , thereby affecting healthcare costs , case detection , and preventable life years . This study demonstrates that there is very little uncertainty that the decision to enforce complete compliance of blood donation screening is correct from a cost-effectiveness point of view . However , complete compliance will require unprecedented transparency of blood services´ information and rigorous monitoring programs for all healthcare institutions , particularly for reference networks and from government institutions . Until Mexico´s health , economy , and governance sectors recognize their responsibility for the continued burden of partial compliance of legislation , the Mexican population will continue to bear the weight of CD , and transmission risk will rise into the future .
|
Chagas disease continues to be a neglected disease in Mexico and Latin-American . Although an estimated 96% of Trypanosoma cruzi transmission to humans occurs via 32 triatomine vector species , the only transmission prevention in Mexico has been sparse and based on heterogeneous blood donation screening . Despite mandating serological screening of blood donations or donors for T . cruzi since 1990 in most Latin American countries , Mexico only finally included mandatory serological screening nationwide in official Norms in 2012 . In 2005 , a survey of blood donor centers in Mexico was conducted to compare T . cruzi prevalence in donations with that of Mexican migrants in the US . Since there was little coincidence between data from that survey and official screening or confirmed case rates , and screening for the social security system only initiated in 2010 , the objective of this study was to calculate the incremental cost-effectiveness ratio for total compliance of current guidelines from both Mexican primary healthcare ( the Secretary of Health ) and regular salaried worker health services ( the Mexican Institute for Social Security ) . A bi-modular model to analyze compliance was developed using a decision tree for the most common documented screening algorithms for the two principal health institution , and a Markov transition model for the natural history of illness and care . The incremental cost effectiveness ratio based on life-years gained is US$ 383 for the Secretary of Health ( MoH ) , while the cost for an additional life-year gained is US$ 463 for the Social Security Institute ( IMSS ) . Using survey compliance data for MoH , and that published by IMSS , failure to detect current infections , to avoid new infections , and life-years lost were calculated for 2013 and 2014 for both institutions . The MoH has failed to confirm 15 , 162 T . cruzi infections , did not prevent 2 , 347 avoidable infections , and lost 333 , 483 life-years over the two year period . Full regulatory compliance should be mandatory and timely monitoring should be part of the health authority’s responsibilities for the sake of blood security in Mexico .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"medicine",
"and",
"health",
"sciences",
"clinical",
"laboratory",
"sciences",
"body",
"fluids",
"enzyme-linked",
"immunoassays",
"cost-effectiveness",
"analysis",
"economic",
"analysis",
"geographical",
"locations",
"social",
"sciences",
"parasitic",
"protozoans",
"health",
"care",
"north",
"america",
"organisms",
"protozoans",
"screening",
"guidelines",
"transfusion",
"medicine",
"immunologic",
"techniques",
"research",
"and",
"analysis",
"methods",
"public",
"and",
"occupational",
"health",
"immunoassays",
"economics",
"hematology",
"people",
"and",
"places",
"trypanosoma",
"cruzi",
"health",
"screening",
"trypanosoma",
"diagnostic",
"medicine",
"blood",
"anatomy",
"physiology",
"biology",
"and",
"life",
"sciences",
"blood",
"transfusion",
"health",
"care",
"policy",
"mexico"
] |
2016
|
Cost-Effectiveness of Blood Donation Screening for Trypanosoma cruzi in Mexico
|
Triatomines are vectors of Trypanosoma cruzi , the etiological agent of Chagas disease in Latin America . The most effective vector , Triatoma infestans , has been controlled successfully in much of Latin America using insecticide spraying . Though rarely undertaken , surveillance programs are necessary in order to identify new infestations and estimate the intensity of triatomine bug infestations in domestic and peridomestic habitats . Since hosts exposed to triatomines develop immune responses to salivary antigens , these responses can be evaluated for their usefulness as epidemiological markers to detect infestations of T . infestans . T . infestans salivary proteins were separated by 2D-gel electrophoresis and tested for their immunogenicity by Western blotting using sera from chickens and guinea pigs experimentally exposed to T . infestans . From five highly immunogenic protein spots , eight salivary proteins were identified by nano liquid chromatography-electrospray ionization-tandem mass spectrometry ( nanoLC-ESI-MS/MS ) and comparison to the protein sequences of the National Center for Biotechnology Information ( NCBI ) database and expressed sequence tags of a unidirectionally cloned salivary gland cDNA library from T . infestans combined with the NCBI yeast protein sub-database . The 14 . 6 kDa salivary protein [gi|149689094] was produced as recombinant protein ( rTiSP14 . 6 ) in a mammalian cell expression system and recognized by all animal sera . The specificity of rTiSP14 . 6 was confirmed by the lack of reactivity to anti-mosquito and anti-sand fly saliva antibodies . However , rTiSP14 . 6 was recognized by sera from chickens exposed to four other triatomine species , Triatoma brasiliensis , T . sordida , Rhodnius prolixus , and Panstrongylus megistus and by sera of chickens from an endemic area of T . infestans and Chagas disease in Bolivia . The recombinant rTiSP14 . 6 is a suitable and promising epidemiological marker for detecting the presence of small numbers of different species of triatomines and could be developed for use as a new tool in surveillance programs , especially to corroborate vector elimination in Chagas disease vector control campaigns .
Control programs for Chagas disease in South America , such as the ‘Southern Cone Initiative’ have relied mainly upon vector control using insecticide spraying [1] . These campaigns have reduced the distribution of T . infestans to an area of 14 . 6% of the initial endemic area . However , especially the Gran Chaco region ( Bolivia , Argentina , Paraguay ) , Andean Bolivia , western Argentina and a small area in south Peru are now still harbouring significant vector populations , some of these regions with sylvatic foci of T . infestans [1]–[3] . In controlled and Chagas disease free areas , i . e . free of vector-borne transmission , such as Argentina and Uruguay , surveillance activities have been greatly reduced allowing the re-establishment of T . infestans [4] . Moreover , after elimination of domestic vectors , peridomestic or sylvatic bug populations and species persist and may replace the former domestic populations due to changes in the ecological balance [5] . Current methods to assess the prevalence and intensity of triatomine bug infestations in domestic and peridomestic sites involve timed manual collections using an irritant spray or artificial shelter units . These methods are costly , require skilled staff , and usually lack the sensitivity and precision necessary for detection of low-density populations . Additionally , current methods are too expensive for large-scale surveillance campaigns and are not easily adaptable to many peridomestic sites [6]–[8] . Thus , new methodologies are required to detect re-emerging T . infestans populations and for long-term monitoring of previously endemic regions for Chagas disease [9] . Hematophagous arthropods have evolved a wide range of salivary anti-hemostatic compounds such as anti-coagulants , anti-histamines , vasodilators and inhibitors of platelet aggregation , sodium channel blockers , immunosuppressors , pore forming molecules and complement inhibitors that are injected into the host when feeding on blood to overcome host defence mechanisms ( hemostasis , inflammation , immunity ) [10]–[15] . Salivary proteins can elicit humoral immune-responses in their hosts [16]–[20] . The detection of antibodies to salivary antigens has been used as an epidemiological tool and biological marker of exposure to disease vectors including mosquitoes , ticks , sand flies and tsetse flies [19] , [21]–[27] . The humoral immune response to salivary proteins of triatomines were studied in chickens , guinea pigs , mice , rabbits and humans; the latter studies using saliva and focusing on epidemiology [16] , [28]–[31] . Recently , we described the anti-saliva immune responses of chickens and guinea pigs which had been experimentally exposed to T . infestans [32] . Antibody responses were detected as soon as two days after the first bug bites . Salivary antigens of 14 and 21 kDa were recognized by all chicken sera and a 79 kDa protein by all guinea pig sera . Sera from animals naturally exposed to triatomines in Bolivia also reacted with these antigens . In the present paper we describe the development of a highly sensitive exposure assay resulting in a specific recombinant salivary antigen to be used as an epidemiological tool for Chagas disease surveillance .
Triatoma infestans , T . brasiliensis , T . sordida , Rhodnius prolixus and Panstrongylus megistus were reared at 27±1°C , 60–70% relative humidity , with a 16/8-h light/dark cycle and were fed on chickens [33] . T . infestans originated from a domestic population from Northern Chile , the Cachiyuyo village ( 29°1′48 . 90″S , 70°53′55 . 53″W , 808 m ) , at the border of the provinces Atacama and Coquimbo [34] , [35] . T . brasiliensis and T . sordida were originally collected from a chicken house from Sítio do Cleniro and Bairro Sosó , state of Piauí , Brazil ( GAS ) . R . prolixus originated from San Juan de Arama , Meta , Colombia ( obtained from A . D'Alessandro-Bacigalupo ) and P . megistus from Minas Gerais , Brazil ( obtained from J . Jurberg , Departamento de Entomologia , Instituto Oswaldo Cruz , Rio de Janeiro , Brazil ) . All experiments were performed with pooled saliva obtained from about 300 fifth instars and adults using capillary pipettes [36] . Typically 0 . 5–1 µl saliva were obtained from each bug . The saliva was desalted with a 4 kDa cut-off centrifugal concentrator Fugisep-Mini ( Intersep ) , and the protein concentration was determined using a BCA Protein Assay Kit ( Perbio Science ) according to the manufacturer's instructions . Aliquots of saliva , containing 30 µg protein/µl , were stored at −80°C . Adult Anopheles freeborni from Marysville ( California , USA ) , Aedes aegypti ( Liverpool black eye strain , UK ) and Culex quinquefasciatus from Vero Beach ( Florida , USA ) were maintained at 28°C , 75% relative humidity and a 12/12 h light/dark cycle . They were provided with 10% sucrose solution for maintenance and blood-fed on anesthetized BALB/c mice [37] , [38] . Lutzomyia longipalpis ( Jacobina strain , Brazil ) were reared according to Modi and Tesh [39] with modifications . Briefly , adult sand flies were maintained at 26°C and 70% relative humidity with a 14/10 h light/dark cycle and fed either with 30% fructose solution or on anesthetized C57Bl/6 mice [32] . The early IgG-response and serial challenges of chickens and guinea pigs by T . infestans have been described previously [32] . Briefly , for the early response five chickens were exposed to starved adult T . infestans ( 5 per chicken ) for 1 h and blood samples were taken daily for five days . For serial challenge , 12 chickens and 10 guinea pigs were exposed every two weeks over a period of 1 h or 30 min and for 19 or 23 weeks , respectively , to a low ( 5 adults ) or a high ( 5 adults and 20 fourth and fifth instars ) number of T . infestans . Sera from animals taken prior to the first feeding served as negative control . For the positive control , sera were pooled from chickens which had been used for routine maintenance of triatomines for at least six months . Groups of three chickens were exposed weekly either to five adults ( T . brasiliensis , T . sordida , R . prolixus or P . megistus ) or An . freeborni , Ae . aegypti or Cx . quinquefasciatus ( approx . 500 insects/animal/expsoure ) for one month . Triatomines were allowed to feed 1 h and mosquitoes about 30 min ( until about 90–100% had fed ) . Blood samples were collected before and always five days after exposure ( triatomines ) or at the end ( mosquitoes ) , and the exposure sera were pooled [32] . Pooled serum was also prepared after one month from five mice exposed weekly for 30–45 min to L . longipalpis ( approx . 100 insects/animal/exposure ) . From September to November 2007 , T . infestans were collected at peridomestic sampling sites and blood samples from animals were only taken if either 1–12 bugs ( low exposure group ) or ≥100 bugs ( high exposure group ) were collected by 3–5 persons within 30–60 min . Blood samples from 28 chickens ( taken from the brachial vein ) and 26 guinea pigs ( taken from the ear vein ) were collected from 16 out of 17 households in the following rural villages in the Department of Cochabamba: Sipe Sipe ( 17°27′2 . 78″S , 66°21′38 . 91″W , 2555 m; 2 out of 3 households colonized ) , Lipez ( 17°33′47 . 12″S , 66°15′27 . 643″W , 2542 m; 5 households ) , Chajra Corral ( 18°1′18 . 30″S , 64°55′25 . 157″W , 1796 m; 3 households ) , Pampas ( 18°3′26 . 812″S , 64°54′35 . 01″W , 1708 m; 3 households ) , Peña Colorada ( 18°10′5 . 288″S , 64°52′0 . 309″W , 1583 m; 1 household ) and Arpita ( 17°33′51 . 62″S , 66°4′15 . 049″W , 718 m; 2 households ) . The blood was centrifuged at 10 , 000×g for 10 min at room temperature , and the sera were stored at −20°C . All animal procedures described in this study and carried out at the Ruhr-Universität Bochum were approved by the Landesamt für Natur , Umwelt und Verbraucherschutz Nordrhein-Westfalen , Recklinghausen , Germany . All animal studies carried out at the The National Institute of Allergy and Infectious Diseases ( NIAID ) were approved by the Animal Care and Use Committee at NIAID , Bethesda , MD , USA . In Bolivia the blood was taken by veterinarians of the Faculty of Veterinary , Universidad Mayor de San Simón , Cochabamba , Bolivia . Salivary gland proteins of T . infestans were separated by isoelectric focusing ( IEF ) using a Multiphor II Electrophoresis Unit ( GE Healthcare ) , pH 4–10 [40] . Saliva ( 40 µg protein ) was dissolved in reducing buffer containing 1% ß-mercaptoethanol and 0 . 5% Ampholine carrier ampholytes , pH 3 . 5–10 ( GE Healthcare ) . Immobilized pH gradient ( IPG ) strips ( 7 cm , pH 4–10 ) were rehydrated with the protein solution and the proteins focused at 10°C using the following gradient: 0–200 V , 2 mA , 5 W for 5 min , then 200–3500 V , 1 mA , 3 W for 90 min and 3500 V constantly , 1 mA , 3 W for 60 min . For the second dimension electrophoresis , the IPG strips were equilibrated in Laemmli loading buffer [41] . The strips were placed onto 15% SDS-PAGE gels and the proteins separated according to their molecular weight using a Hoefer SE 600 apparatus ( GE Healthcare ) . Ten independent samples of saliva were separated to recognize differences in the protein profile . Gels were stained either with colloidal Coomassie blue or silver nitrate with modifications to ensure mass spectrometry-compatibility [42] , [43] . Molecular weights were calculated with reference to the mobility of standard proteins ( Prestained Protein Marker , New England Biolabs ) included in one lane of the gel using the software ImageMaster 2D Elite , version 4 . 3 ( GE Healthcare ) . Western blot analyses were carried out with replica gels of the 2D-gel electrophoresis using individual sera from chickens and guinea pigs exposed to low and high numbers of triatomines in the long-term exposure study [32] . The Western blot images were compared to silver stained 2D-gels in order to determine the recognized salivary gland proteins . Salivary proteins of T . infestans which were recognized by immune animal sera in the 2D-Western blot experiments were identified by mass spectrometry . Colloidal Coomassie stained protein spots were decolorised using alternating solution A ( 20 mM ammonium bicarbonate ) and solution B ( 10 mM ammonium bicarbonate , 50% acetonitrile ) . Silver stained spots were decolorised with 50 mM sodium thiosulfate and 15 mM potassium hexacyanoferrate ( III ) followed by washing with solution A and B [44] . Proteins from all spots were trypsin-digested and extracted [44] . The peptides were subject to liquid chromatography-electrospray ionization-tandem mass spectrometry ( nanoLC-ESI-MS/MS ) analysis with collision induced dissociation experiments using the Ultimate 3000 HPLC system ( Dionex LC Packings ) coupled online to the HCTultra PTM Discovery System ion trap mass spectrometer ( Bruker Daltonics ) [45] . For protein identification , the MS output files ( mgf format ) were stored in the Proteinscape database ( Bruker Daltonics ) and the MS/MS data were compared to the protein sequences of the NCBI database ( Update: 07/05/2007 ) using both the Mascot and Sequest algorithms . To improve the protein identification the MS/MS data were also compared to expressed sequence tags of a unidirectionally cloned salivary gland cDNA library from T . infestans combined for statistical reasons to the yeast protein sub-database of the NCBI database ( TaxID 4932 , 07/05/2007 ) [46]–[48] . The following search parameters were selected: peptide mass accuracy of 0 . 6 Da ( mono-isotopic ) , fragment mass accuracy of 0 . 2 Da ( mono-isotopic ) , variable modification due to oxidation of methionine , acrylamide modification of cysteine , two maximal missed cleavage sites in case of incomplete protease digests . As database searches did not consider all possible types of fragment ions , the quality of the fragment ion spectra explaining identified peptides was analysed using the ESI Compass 1 . 3 , DataAnalysis 4 . 0 software ( Bruker Daltonics ) and by theoretical fragmentation with the MS-Product software tool ( http://prospector . ucsf . edu/cgi-bin/msform . cgi ? form=msproduct ) . For positive protein identification , a minimum of two unique peptides , adequately explained by additional manual interpretation of the respective fragment ion spectra , and at least 5% sequence coverage of the respective protein were required . Complementary DNAs from the T . infestans library encoding the salivary antigens were amplified by PCR , cloned into the VR2001-TOPO plasmid and purified as described previously ( for primer sequences see Table S1 ) [47] , [49] . To aid purification , a six histidine-tag coding sequence was added to the C-terminus of the cDNAs by PCR in a single step using reverse primers ending with the histidine-tag and the stop codon . Plasmids were analyzed for their correct insert and the insert orientation by sequencing using a CEQ 2000 DNA sequencer ( Beckman Coulter ) as previously described [49] . Plasmid purification was carried out using the GenElute HP Endotoxin-Free Plasmid Megaprep Kit ( Sigma-Aldrich ) according to manufacturer's instructions . DNA was concentrated with a Centricon plus-20 centrifugal filter device ( Millipore ) of a 100 kDa cut-off . After measurement of the concentration using the NanoDrop spectrophotometer , the DNA solution was sterile filtered using a 0 . 22 µm Millex-GS Filter Unit ( Millipore ) and stored at −80°C . FreeStyle 293-F Cells ( Invitrogen ) ( 1×106 cells/ml ) were transfected with plasmids coding for salivary gland proteins following the manufacturer's instructions and incubated for 72 h at 37°C , 8% CO2 on a stirrer plate with the propeller of the flask rotating at 135 rpm . Forty-eight hours after transfection , cells were centrifuged and the supernatant filtered through a 0 . 8 µm filter unit ( Nalgene Labware ) . The supernatant was concentrated using an Amicon ultrafiltration device with a 10 kDa cut-off membrane ( Millipore ) in presence of Buffer A ( 20 mM NaH2PO4 , 20 mM Na2HPO4 , 500 mM NaCl , pH 7 . 4 ) . After concentration , the expressed recombinant proteins were dialysed using Slide-A-Lyzer Dialysis Cassettes ( 3 kDA cut-off , Pierce ) against PBS at 4°C overnight . The proteins were purified by HPLC using a HiTrap Chelating HP column ( GE Healthcare ) . The concentrated supernatant was loaded onto a prepared HiTrap Chelating HP column ( GE Healthcare ) and the column connected to a Summit HPLC System with a P680 HPLC pump and a PDA-100 photodiode array detector ( Dionex ) . The column was washed for at least 30 min with buffer A and the following gradient was used to elute the proteins at a flow rate of 1 ml/min: 0–10 min buffer A , from 10–25 min buffer B ( buffer A plus 50 mM imidazole , pH 7 . 4 , ) , 25–45 min 80% buffer B and 20% buffer C ( buffer A plus 500 mM imidazole , pH 7 . 4 ) , 45–105 min buffer C and 105–106 min buffer A . The elution of the proteins was detected at 280 nm , and protein fractions were collected every minute in a 96 deep well microtiter plate using a Foxy 200 fraction collector ( ISCO ) . The purity of the fractions containing the recombinant proteins was tested by SDS-PAGE using a 4–12% NuPAGE Novex 4–12% Bis-Tris gel and the XCell SureLock Mini-Cell electrophoresis system ( Invitrogen ) . The gel was silver stained using the SilverQuest silver staining kit ( Invitrogen ) . Only fractions containing the recombinant proteins were pooled and dialysed against PBS , and the purity of the proteins was tested again by SDS-PAGE and silver staining . The protein concentration was measured using the NanoDrop spectrophotometer . Only the 14 . 6 kDa recombinant salivary protein was obtained in sufficient amount and stored at −80°C . The 14 . 6 kDa recombinant salivary protein of T . infestans was deglycosylated using the enzymatic deglycosylation kit Glyko according to the manufacturer's instructions ( ProZyme Inc . ) . Three aliquots each containing 5 µg recombinant protein were incubated with N-Glycanase PNGase F , a mixture of PNGase F and Sialidase A or a mixture of PNGase F , Sialidase A and O-Glycanase , respectively . After incubation for 3 h at 37°C , the effect of deglycosylation was assessed by SDS-PAGE , and the peptides visualised by SimplyBlue SafeStain solution ( Invitrogen ) . Concentrations of anti-saliva IgG in pooled chicken and guinea pig sera from the long-term T . infestans exposure study were measured by ELISA using 0 . 5 µg recombinant salivary protein per well of a 96-well microtitre plate ( Immunolon , Nunc , Wiesbaden , Germany ) as previously described [32] . Individual serum samples from Bolivia were tested either using 0 . 5 µg recombinant salivary protein or crude saliva of T . infestans per well as the antigen . Responses to challenge with other hematophagous insects to measure cross-reactivity used capture ELISA with recombinant salivary gland protein of T . infestans as described above . Analysis of data was performed using SigmaStat 3 . 1 ( Systat Software Inc . ) . Comparison of responses to bug exposure in chickens and guinea pigs in the long-term study was carried out using a Friedman Repeated Measures Analysis of Variance on Ranks ( One Way RM ANOVA ) with a Pairwise Multiple Comparison Procedure ( Tukey test ) as these data were normally distributed . In the other experiments , depending on the distribution of the ELISA test data , either an Unpaired t-test or a Mann-Whitney Rank Sum test was carried out to compare ELISA assay results using the recombinant salivary protein or the crude saliva of T . infestans analysed with sera from former laboratory studies or field collections sites in Bolivia [32] . The level of significance for all tests was p≤0 . 05 . Initially , the amino acid sequence of the protein was compared to the NCBI protein database to identify species with putative similar proteins . Found sequences were aligned using Clustal X 2 . 0 . 3 , and graphically displayed using BioEdit 7 . 0 . 9 . 0 [50] , [51] . The presence of a signal secretion peptide was predicted by SignalP 3 . 0 [52] .
2D-gel electrophoreses of T . infestans salivary proteins revealed consistently ( n = 10 ) that most proteins were present between pH 4 . 6 and 7 . 3 and between 25 and 31 kDa ( Fig . 1A ) . In 2D-Western blots using serum from T . infestans challenged chickens and guinea pigs , four different proteins ( protein spots 1 , 3 , 4 and 5 ) of 30 , 26 , 14 and 12 kDa reacted with sera of all T . infestans challenged chickens ( Fig . 1B ) , while a 79 kDa protein ( spot 2 ) was recognized by 8 out of 10 challenged animals . Sera from all challenged guinea pigs reacted with a 79 kDa protein ( spot 2 ) and weakly recognized a 30 kDa protein ( spot 1 ) ( Fig . 1C ) . Partial peptide sequences from these five protein spots were identified by mass spectrometry and compared to the NCBI database and to a salivary gland cDNA library from T . infestans . Eight different protein matches were identified within these spots ( Table 1 ) . As an example , Fig . 2A and 2B present MS/MS spectra of two unique peptides of the 14 . 6 kDa salivary secreted protein [gi|149689094] from spot 4 . In all cases there was a difference between theoretical and measured pI , and also the theoretical molecular weights derived from the cDNA sequence differed from the molecular weight determined by the 2D-gel; protein spot 4 contained not only a protein of 14 . 6 kDa but also a 21 . 4 kDa salivary protein . This also applies to all other protein spots except for protein spot 1 ( Table 1 ) . Four out of eight identified salivary proteins of T . infestans were selected for recombinant expression . These were the truncated 79 kDa salivary apyrase precursor ( spot 2 ) , the truncated unknown salivary protein ( spot 3 ) , the salivary lipocalin and the salivary secreted protein ( both in spot 4 ) . Of these four proteins , the 14 . 6 kDa salivary secreted protein of T . infestans ( spot 4 ) was successfully produced as recombinant protein in the mammalian expression system in sufficient amount and purified by HPLC ( Fig . 3 ) . All other proteins were expressed at very low levels and were not used for further analyses . SDS-PAGE analyses of the recombinant 14 . 6 kDa T . infestans salivary protein ( rTiSP14 . 6 ) ( Fig 3 , lane 1 ) gave an apparent molecular weight of 28 kDa . Enzymatic deglycosylation resulted in the expected molecular weight of 14 kDa on SDS-PAGE ( Fig . 3 , lanes 2–4 ) . The earliest antibody reaction with rTiSP14 . 6 was detectable from the second day onwards after a single challenge of five bugs per chicken ( n = 5 , mean O . D . 492 nm = 0 . 032 ) . In the long-term exposure study , chickens had detectable responses in the low ( O . D . 492 nm = 0 . 249 ) and high ( O . D . 492 nm = 0 . 279 ) exposure groups , at the first time point of sera collection ( five days after the first exposure ) ( Fig . 4A ) . Subsequently , responses increased up to O . D . 492 nm = 0 . 734 and O . D . 492 nm = 2 . 598 for the low and high exposure groups , respectively , at the end of the 19 or 23 week exposure period . Overall , the antibody reaction measured with rTiSP14 . 6 was significantly stronger in high exposure versus low exposure chicken sera ( One Way RM ANOVA , p≤0 . 001 ) . The serum reactivity declined rapidly during the first month post-exposure to O . D . 492 nm = 0 . 125 ( low exposure ) and O . D . 492 nm = 0 . 285 ( high exposure ) . Afterwards , the serum reactivity remained stable for three months and then declined to zero . For all guinea pig sera , only a very weak reaction occurred in the low ( maximum mean O . D . 492 nm = 0 . 042 ) and high ( maximum mean O . D . 492 nm = 0 . 100 ) exposure groups , possessing no statistically significantly differences ( p>0 . 05 ) ( Fig . 4B ) . The intensity of the reaction also decreased rapidly after 20 weeks of exposure during four months post-exposure time . Antibody responses to crude saliva and rTiSP14 . 6 of T . infestans were measured using sera from peridomestic hosts from villages with low and high T . infestans infestation densities ( Fig . 5 , for detailed antibody reactivities see Table S2 ) . In the case of chickens , a significantly higher serum antibody response to both crude saliva ( Mann-Whitney Rank Sum test , p = 0 . 001 ) and rTiSP14 . 6 ( Mann-Whitney Rank Sum test , p = 0 . 006 ) was detected in the high infestation households as compared to the low-infestation households ( Fig . 5A ) . In the case of guinea pigs , this difference was only manifested in the serum antibody response to crude saliva ( Mann-Whitney Rank Sum test , p = 0 . 001 ) ( Fig . 5B ) . The value of rTiSP14 . 6 as a marker of triatomine challenge would be compromised if salivary antigens eliciting cross-reactive antibodies are expressed in other blood feeding insects feeding on the same host . A search of the NCBI database was therefore undertaken using the rTiSP14 . 6 sequence and similar peptide sequences were found in Ae . aegypti ( 34% identity ) , An . gambiae ( 33% identity ) and Cx . pipiens quinquefasciatus ( 33% identity ) ( Fig . 6 ) . Challenge experiments with these hematophagous insects and other triatomine species were then undertaken to determine if cross-reactive antibodies were elicited . No antibody responses to rTiSP14 . 6 were detected in sera from chickens exposed to An . freeborni ( chickens , n = 3 ) , Ae . aegypti ( chickens , n = 3 ) and Cx . quinquefasciatus ( chickens , n = 3 ) and in sera from mice ( n = 5 ) exposed to L . longipalpis . Serum samples from chickens ( n = 3 ) exposed only once to T . brasiliensis ( mean O . D . 492 nm = 0 . 075 ) , T . sordida ( mean O . D . 492 nm = 0 . 078 ) , R . prolixus ( mean O . D . 492 nm = 0 . 063 ) and P . megistus ( mean O . D . 492 nm = 0 . 093 ) all reacted with rTiSP14 . 6 . Comparing reactions of rTiSP14 . 6 with sera from chickens exposed to other triatomines and the pooled serum of chickens to T . infestans the latter was 2 . 4 fold higher .
In this study , we used immune sera to identify and characterize immunogenic salivary proteins of T . infestans and made the first steps towards developing a novel immune-epidemiological tool to assess domestic and peridomestic infestations by T . infestans . In Western blots , sera of challenged chickens recognized salivary antigens of 30 , 26 , 14 and 12 kDa , and a 79 kDa antigen was recognized by sera from challenged guinea pigs . Differences between the molecular weights and pIs of these proteins estimated from the 2D-gels and the theoretical characteristics derived from the MS and cDNA sequences were most likely due to post-translational modifications , e . g . proteolytic cleavage , partial glycosylation , protein-protein complex formation [53] . In MS analyses , four proteins were identified as lipocalins . These comprise 55% of the putative secreted salivary proteins of T . infestans [47] and are also the most abundant secretory proteins in the saliva of T . brasiliensis and R . prolixus [54] , [55] . Lipocalins are typically small extracellular proteins that possess a highly conserved eight-stranded antiparallel ß-barrel which forms a central hydrophobic ligand binding cavity . These secreted proteins are carriers of ligands involved in the retinol and pheromone transport , cryptic coloration , olfaction , prostaglandin synthesis and regulation of immune responses as well as the mediation of cell homoeostasis [56]–[58] . The salivary protein pallidipin , previously characterized in T . pallidipennis , is a lipocalin [59] and was highly immunoreactive at 30 kDa in the Western blots . This protein is an inhibitor of collagen-induced platelet aggregation and inhibits the release of ATP from platelets . Its effect on blood platelets is reversible , and it does not influence other types of platelet aggregation [59] . The 79 kDa salivary apyrase precursor from spot 2 is a member of the 5′-nucleotidase family and also a platelet aggregation inhibitor by hydrolysis of ATP or ADP to AMP [15] , [60]–[63] . This glycosylated protein is frequently found in the saliva of different blood-feeding insects and ticks and was recently described in T . infestans [60] , [64] . The apyrase was also detected in T . brasiliensis , but at 10-fold lower expression levels than in T . infestans [55] , [65] . R . prolixus also possesses a similar apyrase [15] , [66] . Four of eight identified proteins were selected for expression studies . The truncated 79 kDa salivary apyrase precursor [gi|148468017] of protein spot 2 was chosen because it was recognized by all guinea pig sera . The truncated unknown salivary protein [gi|148468913] , the salivary lipocalin [gi|149898816] and the salivary secreted protein [gi|149689094] of spots 3 and 4 were chosen because they were recognized by chicken sera . Furthermore , the salivary lipocalin and the salivary secreted protein from protein spot 4 were available as full-length cDNA clones from the T . infestans cDNA library [47] , increasing the likelihood that the expressed recombinant proteins would present the relevant epitopes . The salivary secreted protein was expressed successfully and named rTiSP14 . 6 . RTiSP14 . 6 was insensitive to sialidase A and O-glycanase , but sensitive to peptide-N-glycosidase F . Therefore , the apparent 28 kDa MW was due to glycosylation with N-linked oligosaccharides . ( In a subsequent MS analysis of rTiSP14 . 6 , two types of glycans were identified for one N-glycosylated site of rTiSP14 . 6 , biantennary and triantennary glycans linked to asparagine 105: NeuAcMan5GlcNac4 , Man8GlcNAc2 ) . Three N-glycosylated and no O-glycosylated sites were predicted for rTiSP14 . 6 using the NetGlyc 1 . 0 and NetOGlyc 3 . 1 Server ( detailed data not presented ) [67] , [68] . Such glycosylations are known for other T . infestans salivary proteins [16] . Differences in the glycosylation of rTiSP14 . 6 from the native salivary secreted protein [gi|149689094] are probably due to it being expressed in a mammalian system rather than in its native insect salivary gland cell . The suitability of rTiSP14 . 6 as a tool for the immunological assay of exposure to T . infestans was demonstrated by several results . 1 ) The serum reactivity to rTiSP14 . 6 increased with exposure time and gave a strong signal . 2 ) The sensitivity of rTiSP14 . 6 was high enough to react with sera from chickens in low-infested households . 3 ) The weaker antibody reaction to rTiSP14 . 6 in sera of guinea pigs than in chickens which also occurred in the field study , is no strong disadvantage , since chickens are found in all areas of the country , while guinea pigs only occur in limited areas restricted by the availability of alfalfa ( Medicago sativa ) which cannot be cultivated in the dry regions [69] . Since chickens are also present near or inside the houses in other countries of Latin America , rTiSP14 . 6 is an excellent exposure marker for T . infestans ( but see point 5 below ) . 4 ) Sera from hosts exposed to non-triatomine vectors ( mosquito and sand fly species ) failed to react with rTiSP14 . 6 , confirming the specificity of rTiSP14 . 6 as an immune marker for triatomine challenge . Since in the previous investigation , cross reactions were also not found using bed bugs and ticks , false positive results in an epidemiological survey of triatomine distribution are less likely [32] . 5 ) A major advantage is the reaction of rTiSP14 . 6 with sera of chickens challenged with T . brasiliensis , T . sordida , R . prolixus and P . megistus . Other triatomine species are capable of replacing T . infestans , and after control programs they may invade and colonize peridomestic sites; for example T . brasiliensis and P . megistus have replaced T . infestans in Brazil and T . guasayana may substitute for T . infestans in Argentina [1] , [70] , [71] . Although Rhodnius and Triatoma/Panstrongylus do not belong to the same tribe ( Rhodniini: monophyletic or paraphyletic origin and Triatomini: polyphyletic or paraphyletic origin ) within the polyphyletic Triatominae [72] , [73] , the native protein form of the cross reactive rTiSP14 . 6 seems to be ubiquitous in the different triatomine families and probably is an evolutionary conserved protein ( orthologous ) in the Triatominae . These results are consistent with previous Western blot results showing all these species elicited antibodies to a 14 kDa salivary protein [32] . Therefore , rTiSP14 . 6 is not T . infestans specific , but is suitable for detecting infestations by at least five species of triatomines . There seems to be a stronger reaction with sera challenged with T . infestans than with other triatomines which might allow a differentiation from these triatomine species using chicken sera . However , this needs to be verified in further investigations . Testing the more widespread use of rTiSP14 . 6 as an exposure marker for triatomines other than T . infestans can best be evaluated by sera from chickens living in regions where other triatomines are found . Within such a survey , sera from other animals including dogs and cats , which are frequent domestic reservoir hosts of T . cruzi , can also be tested [4] . In a recent study , T . infestans showed a feeding preference for dogs over chickens and cats [74] . In conclusion , the aim of this study was to develop an epidemiological marker to detect low-level infestation of T . infestans in endemic countries after insecticide control activities . Out of eight different highly immunogenic T . infestans salivary proteins , a 14 . 6 kDa salivary protein of unknown function was successfully expressed recombinantly . It reacted strongly with sera from chickens exposed to a low number of T . infestans in both the laboratory and field . This recombinant protein will enable the immunological detection of low numbers of different triatomines , not only T . infestans , and hence could be developed as a sensitive surveillance tool to warn of exposure risk to triatomines and thereby for Chagas disease . The development of recombinant salivary proteins as epidemiological tools may also be useful for the control of other vectors such as mosquitoes , ticks , tsetse flies and sand flies and to understand the epidemiology of the diseases that they transmit [22] , [75]–[78] .
|
Chagas disease , caused by Trypanosoma cruzi , is a neglected disease with 20 million people at risk in Latin America . The main control strategies are based on insecticide spraying to eliminate the domestic vectors , the most effective of which is Triatoma infestans . This approach has been very successful in some areas . However , there is a constant risk of recrudescence in once-endemic regions resulting from the re-establishment of T . infestans and the invasion of other triatomine species . To detect low-level infestations of triatomines after insecticide spraying , we have developed a new epidemiological tool based on host responses against salivary antigens of T . infestans . We identified and synthesized a highly immunogenic salivary protein . This protein was used successfully to detect differences in the infestation level of T . infestans of households in Bolivia and the exposure to other triatomine species . The development of such an exposure marker to detect low-level infestation may also be a useful tool for other disease vectors .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] |
[
"immunology/immune",
"response",
"infectious",
"diseases/epidemiology",
"and",
"control",
"of",
"infectious",
"diseases",
"biotechnology/protein",
"chemistry",
"and",
"proteomics"
] |
2009
|
Immunogenic Salivary Proteins of Triatoma infestans: Development of a Recombinant Antigen for the Detection of Low-Level Infestation of Triatomines
|
Missense mutant proteins , such as those produced in individuals with genetic diseases , are often misfolded and subject to processing by intracellular quality control systems . Previously , we have shown using a yeast system that enzymatic function could be restored to I278T cystathionine β-synthase ( CBS ) , a cause of homocystinuria , by treatments that affect the intracellular chaperone environment . Here , we extend these studies and show that it is possible to restore significant levels of enzyme activity to 17 of 18 ( 94% ) disease causing missense mutations in human cystathionine β-synthase ( CBS ) expressed in Saccharomyces cerevisiae by exposure to ethanol , proteasome inhibitors , or deletion of the Hsp26 small heat shock protein . All three of these treatments induce Hsp70 , which is necessary but not sufficient for rescue . In addition to CBS , these same treatments can rescue disease-causing mutations in human p53 and the methylene tetrahydrofolate reductase gene . These findings do not appear restricted to S . cerevisiae , as proteasome inhibitors can restore significant CBS enzymatic activity to CBS alleles expressed in fibroblasts derived from homocystinuric patients and in a mouse model for homocystinuria that expresses human I278T CBS . These findings suggest that proteasome inhibitors and other Hsp70 inducing agents may be useful in the treatment of a variety of genetic diseases caused by missense mutations .
Missense mutations are genetic alterations that result in the production of proteins with single amino acid changes and are an especially common cause of a variety of diseases [1] . Most disease causing missense mutations do not target key catalytic residues , but rather cause problems in protein folding . It is thought that missense mutations affect protein folding by “trapping” the protein in a non-functional intermediate state , preventing it from folding into its lowest-free energy native state . These trapped misfolded protein intermediates can either be degraded or form large molecular weight aggregates [2] . In theory , treatments that could reverse these protein-folding defects and promote proper folding would be of great utility in the treatment of a wide variety of genetic diseases . Three genetic diseases in which missense mutations are common include cystathionine β-synthase ( CBS ) deficiency , Li Fraumeni syndrome , and methylenetetrahydrofolate reductase deficiency . CBS deficiency is an inborn error of sulfur metabolism characterized by very high levels of plasma total homocysteine ( tHcy ) . CBS catalyzes the condensation of homocysteine with serine to form cystathionine and is the first step in the de novo production of cysteine . In healthy adults , tHcy concentration in plasma ranges from 5 to 15 µM , but untreated patients with CBS deficiency often have tHcy in excess of 200 µM [3] . CBS deficient patients suffer from various pathologies including arteriosclerosis , osteoporosis , mental retardation , and dislocated lenses [4] . The major cause of mortality in these patients is stroke . Treatments that lower tHcy such as B-vitamins , dietary methionine restriction , and betaine supplementation , can significantly reduce the incidences of vascular events in these patients despite the fact that post-treatment homocysteine levels are still several times higher than levels found in the normal population [5] , [6] , [7] . Mouse models for CBS deficiency also indicate that there is a threshold effect for tHcy toxicity and support the notion that a small increase in residual CBS activity may have large clinical benefits [8] . Li-Fraumeni syndrome is a dominant cancer susceptibility syndrome disorder caused by missense mutations in the TP53 tumor suppressor gene [9] . Li-Fraumeni patients suffer from a variety of cancers , including sarcomas , adrenocorticol carcinomas , breast cancer , leukemia , and brain tumors [10] . In general , TP53 behaves as a classic tumor suppressor gene , with the tumors losing or inactivating the wild-type copy of TP53 , resulting in expression of only the mutant form . Mutant forms of p53 tend to be stable , resulting in increased accumulation of protein [11] . Of the 165 mutations in the TP53 gene described in the Human Gene Mutation Database [12] , 110 are of the missense variety ( 68% ) . MTHFR is a critical enzyme in the remethylation of homocysteine to methionine . Its biochemical function is to catalyze the formation of 5-methyltetrahydrofolate , which is the methyl-group donor for the subsequent reaction catalyzed by methionine synthase . Mutations in MTHFR are known to cause MTHFR deficiency . MTHFR deficiency symptoms include developmental delay , motor or gait abnormalities , seizures , and premature vascular disease [13] . Thirty-four mutations have been described in MTHFR deficient patients , and 23 are predicted to encode missense mutations ( 67% ) [12] . Previously , work from our lab has shown that it is possible to restore significant enzymatic function to human CBS containing an isoleucine to threonine substitution at position 278 ( I278T ) by growth of cells in ethanol containing media [14] . This rescue was shown to require the induction of Hsp70 , a key cellular chaperone protein [15] . Hsp70 protein interacts with misfolded polypeptides along with co-chaperones and promotes refolding by repeated cycles of binding and release requiring the hydrolysis of ATP [16] . In the work described here , we have extended our observations by showing that treatments that induce Hsp70 can greatly increase enzymatic function from 17 additional mutant CBS proteins , as well as mutant forms of p53 and MTHFR . In addition we show that proteasome inhibitors can induce Hsp70 and that these drugs can greatly increase mutant CBS activity in human cells and in a mouse model of CBS deficiency . Our findings support the idea that drugs that induce Hsp70 may be useful in the treatment of genetic disorders caused by missense mutations .
We initially examined the effect of ethanol on a panel of 17 additional missense CBS mutations found in homocystinuric patients [17] . Each of these mutations was expressed in a yeast strain ( WY35 ) that is deleted for the endogenous CBS gene ( cys4Δ ) and growth was examined on cysteine-free media either lacking or containing 4% ethanol ( Figure 1A ) . In addition , we prepared total cellular lysates from the strains grown in cysteine-supplemented media with or without ethanol and measured both the steady state level of each mutant protein by Western blot and CBS enzyme activity ( Figure 1B and 1C , 1 , Table 1 ) . We found that 4/17 ( 24% ) of the mutants exhibited significant growth and greatly increased CBS enzyme activity ( 8 to 50-fold ) when grown in ethanol-containing media . Interestingly , like I278T , ethanol had a much more modest effect on steady state protein levels compared to enzyme activity , indicating that the treatment caused the specific activity of the enzyme to increase . Ethanol exposure also increased steady state Hsp70 levels , consistent with our previous observations ( Figure 1B ) . Previously , we had also shown that deletion of the small heat shock protein HSP26 ( hsp26Δ ) could effectively suppress the functional effects of I278T CBS mutation in yeast [14] . Therefore , we introduced our panel of additional missense mutants into a cys4Δ hsp26Δ double mutant strain and examined function . As with ethanol , hsp26Δ resulted in increased steady state levels of Hsp70 ( Figure 1E ) . Deletion of Hsp26 could rescue the cysteine growth auxotrophy and restore significant levels of enzymatic activity to 10/17 ( 59% ) of the mutants tested ( Figure 1D–1F , Figure S2 , Table 1 ) . Again , the increased level of enzyme activity in the suppressible mutants was impressive , ranging from an 8 to 55-fold increase resulting in levels that were between 25% and 58% of wild-type CBS . Like ethanol , most of the increased enzyme activity appears to be due to increased functionality of the mutant protein as opposed in increased protein levels . In our earlier work , we had shown that expression of I278T CBS drove down levels of Hsp26 protein and that I278T CBS , but not wild-type CBS , physically interacted with Hsp26 [14] . We tested two of the newer mutants ( G307S , D376N ) for these properties and found that they also caused decreased steady state levels of Hsp26 and co-immunoprecipitated Hsp26 ( Figure S3 ) . These results indicate that like I278T , G307S and D376N form a complex with Hsp26 . Since proteasome inhibitors have previously been shown to induce Hsp70 in both yeast and mammalian cells [18] , [19] , we tested if these agents might be able to restore function to missense mutant CBS proteins in vivo . Bortezomib ( also known as PS-341 or Velcade ) is a potent proteasome inhibitor that is currently used to treat humans with multiple myeloma [20] . We first tested the effects of bortezomib on mutant human I278T CBS function in yeast . We found that addition of bortezomib to yeast media strongly rescued , in a dose-dependent fashion , the cysteine auxotrophy of WY35 cells carrying a plasmid expressing human I278T CBS ( pI278T ) ( Figure 2A ) . After 24 hours , 75 µM bortezomib allowed WY35pI278T cells to achieve growth that was at least equivalent to WY35 expressing wild type human CBS ( phCBS ) . Examination of I278T protein and activity from treated vs . untreated cells indicated that bortezomib increased steady-state levels of I278T by 7-fold and increased I278T activity by 27-fold , to about 50% of that observed in untreated cells expressing wild-type CBS ( Figure 2B ) . These results show that bortezomib can both stabilize and restore function to I278T CBS expressed in yeast . Bortezomib treatment also caused a 2 . 8 to 3 . 6-fold increase in Hsp70 in WT and I278T expressing strains , respectively ( Figure 2B ) . To determine if this induction was essential for restoration of I278T CBS function , we examined the effect of bortezomib on yeast lacking the Hsp70-encoding gene SSA2 ( ssa2Δ; Figure 2C ) . As expected , steady state levels and the enzyme activity of I278T CBS were lower compared to wt CBS in both SSA2 and ssa2Δ cells in the absence of bortezomib . In the presence of bortezomib , we saw stabilization of I278T protein in ssa2Δ cells , but no increase in enzyme activity . In ssa2Δ cells expressing WT CBS , we did not observe any decrease in enzyme activity , indicating that SSA2 is not required generally for human CBS function . These results show that simply blocking proteolysis of I278T CBS is not sufficient to restore function to the mutant enzyme , and that SSA2 ( Hsp70 protein ) is required for bortezomib-induced rescue of function . We also investigated the effect of bortezomib on our panel of other patient-derived CBS mutants . We found that 9/17 of the alleles tested showed significant growth in cysteine-free media and had enzyme activity restored to between 37–68% of wild-type CBS ( Figure 2D ) . It should be noted that some of these mutants ( T353M and T262M for example ) produce stable but non-functional enzymes , indicating that bortezomib is not simply working by preventing protein degradation ( Figure S4 ) We next determined if chaperone manipulation could be used to restore function to disease causing mutants of p53 and MTHFR . To examine p53 function , we used a yeast strain ( yIG397 ) in which human p53 binds upstream of the ADE2 promoter , activates transcription , and results in correction of the strains adenine auxotrophy [21] . Into this strain we transfected plasmids that expressed either wild-type human p53 or one of three different patient-identified point mutations ( R175H , R273H , C277F ) and measured p53 function by examining growth in adenine deficient media . As expected , we found that cells expressing wild type human p53 grew well , while cells expressing the mutant alleles had almost undetectable levels of growth ( Figure 3A ) . However , when the media was supplemented with either 4% ethanol , 50 µM Bortezomib , or an hsp26Δ construct was introduced into the strain , each mutant exhibited significant growth rescue with at least one of the treatments . We performed more in depth analysis on two of the mutant alleles , R175H and R273H . In the absence of ethanol , steady state levels of both of these mutant proteins are slightly elevated compared to wild type CBS and cause elevations in the steady state levels of Hsp26 and Hsp104 ( Figure 3B ) . The elevation in Hsp104 was unexpected , as we had previously shown that I278T CBS had no effect on Hsp104 levels [14] . When ethanol is added to the cultures , Hsp70 levels are induced and p53 and Hsp26 levels are reduced to wild-type levels , while hsp104 levels remain elevated . The observation that p53 levels go down , while growth increases indicates that , like CBS , ethanol is enhancing the specific activity of p53 . We also examined the dependence of ethanol rescue on having a functionally intact SSA2 and HSP104 ( Figure S5 ) . Deletion of either gene resulted in total loss of the ability of ethanol to promote adenine independent growth . Immunoprecipitation experiments demonstrate that , similar to mutant CBS , Hsp26 recognizes mutant p53 but not wild-type p53 and that this interaction is lost upon ethanol treatment . ( Figure 3C ) . Taken together , these results show that p53 mutants behave similar , but not identical , to CBS mutants with regards to rescue by agents that perturb the cellular chaperone environment . We next determined the frequency by which defective p53 protein could be rescued by either ethanol or bortezomib treatment . We examined rescue using a panel of 22 single missense mutant p53 alleles that were obtained by random mutagenesis of the human p53 ( see Materials and Methods ) . Like CBS , we found that some mutants were rescuable by both ethanol and bortezomib ( 4/22 ) , some only by bortezomib ( 8/22 ) , some only by ethanol ( 4/22 ) , and some that were not rescuable by either treatment ( 6/22 ) . In total , we found that 16 out of 22 ( 72% ) of the randomly generated mutants were rescuable by at least one of the treatments ( Figure S6 ) . We have also examined the effect of ethanol and bortezomib on mutant alleles of human 5–10-methylene tetrahydrofolate reductase ( MTHFR ) expressed in yeast . In this assay , expression of a functional human MTHFR enzyme complements the methionine auxotrophy present in a met11Δ mutant [22] . Two missense mutant MTHFR proteins were tested for growth rescue by either addition of ethanol or bortezomib and we found that one , L323P , could be rescued by the addition of ethanol ( Figure 3D ) . We next determined if the restoration of function we observed in S . cerevisiae could also occur in mammalian cells . To test this , we examined the effect of proteasome inhibitors on mammalian cells expressing mutant human CBS protein . We obtained four primary fibroblast lines and one EBV transformed lymphoblastoid line from five patients with CBS deficiency . Four of the five lines were from patients that were homozygous for a particular mutation ( I278T , T353M , T262M , G307S ) , while one line was from a compound heterozygote ( A114V/E302K ) . We then examined both steady state CBS protein and CBS enzyme activity in untreated and in cells treated with the proteasome inhibitor MG132 ( Figure 4A ) . MG132 was used in these experiments because it was found to be a more potent inducer of Hsp70 than bortezomib for these cells ( Figure S7 ) . In the absence of drug , three of the five lines ( I278T , T253M , and A114V/E302K ) had undetectable levels of CBS protein and all five lines had <1% enzyme activity compared to a control primary human fibroblast line . Addition of MG132 resulted in restoration of steady-state CBS to wild-type levels in all five lines and caused a dramatic increase in enzyme activity ( over 100-fold increase to 28 to 66% of wt CBS ) in three of the five lines ( I278T , T353M , and T262M ) . Interestingly , the increase in enzyme activity in the T262M cells was not associated with increased T262M protein levels , again suggesting that the rescuing effect of proteasome inhibitors is not due to inhibition of mutant protein degradation . Consistent with this idea , we found that MG132 resulted in a two-fold increase in Hsp70 ( Figure 4A ) . With one exception , we note concordance in the behavior of mutants that were rescuable by proteasome inhibition in yeast and in patient fibroblasts . It is possible that the lack of rescue observed in the A114V allele may have to do with the cell line also having the E320K allele as well . In total , these results indicate that exposure of human cells to proteasome inhibitors can rescue enzymatic activity from several missense mutant CBS proteins . Previously we had developed a mouse model for homocystinuria caused by I278T CBS ( Tg-I278T Cbs−/− ) . This mouse contains a homozygous deletion of the mouse CBS gene , and expresses human I278T CBS under control of a zinc-inducible promoter [23] . In this animal it was demonstrated that human I278T CBS expressed in mouse liver had minimal enzyme activity [24] . To test the effect of bortezomib in vivo , we induced I278T expression by addition of zinc to the drinking water , and after one week injected 30 µg of bortezomib into the tail veins of four Tg-I278T Cbs−/− animals . After 17 hours , serum , liver and kidney were then harvested and analyzed . Examination of steady state CBS protein indicates that bortezomib greatly increased the amount of liver I278T protein and Hsp70 protein compared to a mock-injected control ( Figure 4B ) . In addition , CBS enzyme activity increased on average 4 . 3-fold in the livers of treated animals compared to untreated controls , to about 18 . 7% of that found in an animal that expresses wild-type human CBS ( Figure 4B ) . We also observed increased enzyme activity in kidney extracts in two of the four injected animals ( Figure 4B ) . Interestingly , the kidney does not show significantly increased steady-state I278T protein , indicating that the increased enzyme activity is due entirely to increased enzyme specific activity . Bortezomib treatment resulted in a 46% decrease in tHcy in serum , ( p<0 . 05 ) ( Figure S8 ) . Bortezomib also resulted in a 28% decrease in serum methionine , although this was not statistically significant ( Figure S8 ) . These results indicate that bortezomib can decrease tHcy in vivo by increasing residual I278T enzyme activity .
In the work described here , we have shown that it is possible to restore substantial amount of functional activity to a large percentage of missense mutant alleles of human CBS , p53 , and MTHFR expressed in S . cerevisiae by treatments that cause Hsp70 induction . These treatments include addition of ethanol or proteosome inhibitor to the media , or deletion of the small heat shock protein HSP26 . At least one of these treatments was able to dramatically increase enzyme or transcriptional activity in 17/18 ( 94% ) mutant CBS proteins , 19/25 mutant p53 proteins ( 76% ) , and 1/2 mutant MTHFR proteins ( 50% ) . Interestingly , different types of response to the different rescuing treatments were observed . Some mutants were rescued by all three treatments ( e . g . I278T CBS ) , some by two of three treatments ( e . g . T262M CBS ) , while others only responded to a single treatment ( e . g . G307S CBS ) . These differences are probably not due to differential Hsp70 induction , as all three treatments result in similar levels of induction . We speculate that each of these different treatments may have unique effects on other molecular chaperones and that each treatment may produce a unique intracellular folding milieu . With regards to CBS mutants , we found that both pyridoxine responsive mutants ( e . g . I278T ) and non-responsive mutants ( e . g . G307S ) could be rescued . Our findings support the view that most missense mutations primarily affect protein folding , as opposed to altering critical residues involved in specific catalytic or interaction sites . This agrees with biophysical studies of protein folding . As stated by DePristo et al . [25] “The overwhelming conclusion from 20 years of mutation studies on protein stability is that most amino-acid replacements , at all sites in a protein result in large effects on ΔG relative to the observed range of ΔG values themselves . ” It is also possible that in multi-domain proteins like CBS , mutations affect protein folding by trapping the protein in a non functional intermediate state due to the existence of a high kinetic barrier [2] . Therefore , a possible explanation for the effectiveness of chaperone manipulation is that chaperone complex either helps mutant proteins overcome rate-limiting kinetic constraints in the folding process or brings the trapped intermediate back to a folding competent state . Interestingly , in CBS , only one mutation , G116R , was not rescuable by any treatment . This mutation is located only three residues away from the lysine ( K119 ) that binds the active site pyridoxal phosphate at the base of an alpha helix [26] . Thus this change may be an example of a mutation that affects a key catalytic-site in the protein . Our results also show that this is not a yeast-specific phenomenon . We were able to restore function to three of six mutant alleles present in fibroblast cells from CBS deficient patients by treatment with the proteasome inhibitor MG-132 . In addition , we showed that a 4 . 3-fold increase in CBS activity could be achieved with a single injection of bortezomib into mice expressing I278T CBS . Consistent with our findings , Mu et al . showed that MG-132 could increase residual glucocerebrosidase enzyme activity four-fold in cell lines containing the L444P variant associated with Gaucher disease , and that a combination of MG-132 and a chemical chaperone , 2-Acetamido-2-deoxynojirimycin , could increase residual activity five-fold in cells containing the Tay-Sachs disease causing α-hexosaminidase G269S mutation [27] . In addition , the Hsp70-inducing drug arimoclomol was reported to delay disease progression in mice expressing a SOD1 mutant in which glycine is substituted with alanine at position 93 [28] . Taken together , these findings suggest that treatment with proteasome inhibitors and other Hsp70 inducers may be beneficial to individuals with severe genetic diseases caused by missense mutation . One potential concern with drugs that manipulate the molecular chaperone environment is they may have potential adverse effects on the normal proteome , and thus create side effects that would preclude their use as drugs . While this is a possibility , it should be noted that Bortezomib is already an FDA approved drug that is used to treat multiple myeloma . The major noted toxicity is peripheral neuropathy [29] . However , it is unknown if the levels of Bortezomib used to induce changes in the chaperone environment would need to be as high as those used in chemotherapy . Furthermore , there is another proteasome inhibitor in clinical trials ( PR-171 ) that does not exhibit a strong peripheral neuropathy effect [30] . In addition , there are at least two other drugs that are known Hsp70 inducers that have undergone successful phase I trials in humans including 17-allylamino-17-demethoxygeldanamycin and arimoclomol [31] , [32] . Therefore , while there may be some risk in this approach , there is no reason a priori to believe that the benefits of restoring function to a mutant protein would not out weigh the potential side effects . In addition to germline genetic diseases , the work described here may have relevance to the treatment of somatic genetic diseases such as cancer . Mutations in p53 are the most frequent genetic alteration in human cancer , and loss of p53 function is critical for tumorigenesis [33] . Several pharmacologic chaperones that restore function to mutant p53 protein have been identified and these have been shown to effectively induce apoptosis in certain cells that express mutant p53 [34] , [35] . Our findings suggest that proteasome inhibitors may also effectively restore function to certain p53 alleles . Although bortezomib has been shown to inhibit cell growth and cause apoptosis in a large number of cancer cell lines , the IC50 concentrations vary widely [36] . No systematic study of the relationship between bortezomib sensitivity and the p53 status of tumor cells has been reported . In summary , the data reported here shows that the functional effects of a majority of missense mutations in at least three human genes can be reversed by treatments that alter the intracellular molecular chaperone environment . Based on these results , we suggest that drugs that effect the intracellular chaperone environment may be useful in the treatment of a number of genetic diseases caused by missense mutations .
Yeast strains WY35 ( α leu2 ura3 ade2 trp1 cys4::LEU2 ) , LS1 ( α leu2 ura3 ade2 trp1 cys4::LEU2 hsp26::KanMX ) and LS3 ( α leu2 ura3 ade2 trp1 cys4::LEU2 ssa2::KanMX ) were generated as previously described [14] , [37] . Yeast strain yIG397 ( MATa ade2-1 leu2-3 , 112 trp1-1 his-11 , 15 can1-100 , ura3-1 URA3 3XRGC::pCYC1::ADE2 ) was obtained from Dr . Richard Iggo [21] . Strains LS3 ( yIG397+hsp26Δ ) , LS4 ( yIG397+ssa2Δ ) and LS5 ( yIG397+hsp104Δ ) were derived from this strain by transformation with appropriate deletion cassettes as previously described [14] . The MTHFR tester strain , XSY3-1a ( MATa ade2-1 , can1-100 , ura3 leu2 trp1 , his3 , trp1 met11Δ::TRP1 ) was created as previously described [22] . Plasmids expressing wild-type or mutant human CBS proteins were created by site-directed mutagenesis and gap-repair as described [38] , [39] . The plasmids expressing mutant p53 alleles were constructed using gap repair with pRDI-22 [21] . Additional p53 mutants were generated by amplification with TaqI polymerase , followed by gap-repair and subsequent screening on SC-ade media for non-functional alleles . Ninety-six clones were then isolated and sequenced . Twenty-three of these clones contained a single missense mutation and were used for the studies described here . Bortezomib ( Velcade™ , Millennium Pharmaceuticals , Inc ) was obtained from the Fox Chase Cancer Center outpatient clinic . Synthetic complete media lacking cysteine ( SC–Cys ) and ( SC–Cys+G418 ) were made as previously described [14] . SC+Cys media were made by adding glutathione to the indicated SC media at a final concentration of 30 µg/ml . All other chemicals were purchased from Sigma . For yeast cell growth assays , saturated cultures were diluted to an OD600 of 0 . 05 and treated with bortezomib in standard 15-ml borosilicate tubes containing 3 ml of media . Cells were then grown at 30°C with rotating aeration . After 24 hours OD600 was determined using a Milton Roy Spectronic 601 spectrophotometer ( Ivyland , PA ) . Cell lines used in this study were: WS1 ( WT ) [40] , F5889 ( I278T/I278T ) [41] , 382 ( T262M/T262M ) [42] , 2242 ( A114V/E302K ) [43] , 3161 ( T353M/T353M , JPK unpublished ) , 3079 ( G307S/G307S ) [44] . Fibroblasts and EBV transformed lymphoid cells were grown in MEM or RPMI medium supplemented with 15% FBS along with appropriate antibiotics , respectively , in 5% CO2 at 37°C . MG132 was purchased from Peptide International ( product number , IZL-3175v ) . MG132 or bortezomib was added to cells when they reached 80% confluency and after seven hours of exposure , the MG132 containing media was discarded , cells were rinsed with PBS , and then harvested for protein extraction . Saturated yeast were diluted to an OD600 of 0 . 05 and cells were harvested when OD was between 0 . 7 and 0 . 9 ( mid-log phase ) . Yeast extracts were prepared by mechanical lysis as described previously [45] . Total mammalian protein extracts were prepared using the mammalian protein extraction reagent obtained from Pierce , Rockford , IL ( product # 78501 ) . Protein concentration was determined by the Coomassie Blue protein assay reagent ( Pierce ) using bovine serum albumin as a standard . For immunoblot analysis , yeast extracts containing 20 µg or mammalian cell containing 50 µg of total protein were run on precast Tris-acetate gels ( Bio-Rad ) at 15 mA and transferred to polyvinylidene difluoride membrane as previously described [14] . Immunopurified mouse anti-CBS was purchased from ABNOVA ( catalogue # H00000875-A01 ) and was used at 1/10 , 000 dilution . Anti-p53 is a mouse monoclonal anti-body obtained by Calbiochem ( cat . #OP43 ) and was used at a 1/1000 dilution . Yeast Hsp70 monoclonal mouse serum was from AbCAM ( catalogue # ab5439 ) and used at a dilution of 1/5000 . Human Hsp70 monoclonal antibody ( product # SPA-810D ) was used at 1/1000 dilution . Α-Tubulin antibody used here is the same as described previously [14] . For rabbit antiserum , a horseradish peroxidase-conjugated anti-rabbit secondary anti-body was used at a 1/30 , 000 dilution ( Jackson ImmunoResearch , West Grove , PA ) . Signal was detected by chemiluminescence using SuperSignal West Pico chemiluminescent substrate ( Amersham Biosciences ) and Chemigenius station ( Syngene ) ; signal was quantitated by Alpha Innotech software . Immunopreciptiations were performed as previously described [14] . All animal studies were reviewed and approved by the Fox Chase Cancer Center Institutional Animal Care and Use Committee . Tg-hCBS Cbs−/− and Tg-I278T Cbs−/− mice were generated as previously described [23] , [46] . All mice were from C57BL6 strain background and were fed the rodent chow ( Teklad 2018SX; Harlan Teklad , Madison , WI , USA ) ad libitum . Adult mice ( 9–12 months old ) were injected with 30 µl of bortezomib ( 1 mg/ml ) via tail vein injection . After 17 hour , mice were sacrificed and liver , kidney and serum were extracted . Liver and kidney homogenates were prepared as previously described [46] . Western blotting was done under denaturing conditions as previously described [47] . CBS activity was measured in 30 mg of crude dialyzed protein extracts of liver and kidney , using a Biochrom 30 amino acid analyzer ( Biochrom , Cambridge , UK ) as described previously [47] . One unit of activity is one µM of cystathionine formed per hour per milligram of protein . The same instrument was employed to measure total homocysteine ( tHcy ) and methionine in serum as described [8] .
|
Genetic diseases are often caused by missense mutations: single nucleotide changes that cause a single incorrect amino acid to be substituted into the underlying protein . Most missense mutations cause the encoded protein to fold incorrectly and therefore not function properly . Examples of missense mutation diseases include CBS deficiency , Li-Fraumeni syndrome , and methylenetetrahydrofolate reductase deficiency , which are caused by alterations in the CBS , TP53 , and MTHFR genes , respectively . In the work presented here , we show that it is possible to restore substantial enzymatic function to disease-causing human mutant CBS , p53 , and MTHFR proteins by treatments that alter the intracellular folding environment . All of these treatments result in elevation of the Hsp70 protein , a molecular chaperone protein involved in helping proteins fold properly . One way to rescue mutant protein function is by using a drug that inhibits the function of the proteasome , the intracellular machine responsible for degrading misfolded proteins . Our findings suggest that drugs of this class may be potentially useful in the treatment of human genetic diseases caused by missense mutations .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"genetics",
"and",
"genomics/disease",
"models",
"genetics",
"and",
"genomics/genetics",
"of",
"disease",
"biochemistry/protein",
"folding"
] |
2010
|
Activation of Mutant Enzyme Function In Vivo by Proteasome Inhibitors and Treatments that Induce Hsp70
|
Tumourigenesis within the intestine is potently driven by deregulation of the Wnt pathway , a process epigenetically regulated by the chromatin remodelling factor Brg1 . We aimed to investigate this interdependency in an in vivo setting and assess the viability of Brg1 as a potential therapeutic target . Using a range of transgenic approaches , we deleted Brg1 in the context of Wnt-activated murine small intestinal epithelium . Pan-epithelial loss of Brg1 using VillinCreERT2 and AhCreERT transgenes attenuated expression of Wnt target genes , including a subset of stem cell-specific genes and suppressed Wnt-driven tumourigenesis improving animal survival . A similar increase in survival was observed when Wnt activation and Brg1 loss were restricted to the Lgr5 expressing intestinal stem cell population . We propose a mechanism whereby Brg1 function is required for aberrant Wnt signalling and ultimately for the maintenance of the tumour initiating cell compartment , such that loss of Brg1 in an Apc-deficient context suppresses adenoma formation . Our results highlight potential therapeutic value of targeting Brg1 and serve as a proof of concept that targeting the cells of origin of cancer may be of therapeutic relevance .
More than 90% of colorectal cancers ( CRC ) are characterised by aberrant activation of the canonical Wnt/β-catenin pathway , which is proposed to play a major role in the initiation and progression of CRC [1] . Despite this clear link between deregulated Wnt signalling and disease , therapies which target the Wnt pathway remain limited [2] . There is , therefore , a substantial demand for novel approaches to inhibit the Wnt pathway , preferably downstream of common aberrations such as mutations in APC , AXIN2 or β-catenin . The emerging plethora of epigenetic factors involved in DNA methylation , histone modifications and chromatin remodelling represent a set of new and relatively unexplored opportunities for such therapeutic intervention [3] . Brahma-related gene 1 ( BRG1 ) or SWI/SNF-related matrix-associated actin- dependent regulator of chromatin subfamily A member 4 ( SMARCA4 ) is one of the two mutually exclusive ATPase subunits of the 2-MD family of SWItch/Sucrose Non-Fermentable ( SWI/SNF ) class of chromatin remodelling complexes . BRG1 has been implicated in a variety of biological processes , in both normal and neoplastic tissues [4] , [5] . The majority of these studies , both in vitro and in vivo , suggest that BRG1 acts as a tumour suppressor . For example , it has been found to be mutated in numerous cancer cell lines and primary cancers [6]–[9] . In support of this , studies using Brg1 knock-out mouse models have shown that heterozygous loss of Brg1 increases susceptibility to both mammary gland and lung tumourigenesis [10] , [11] . By contrast , BRG1 has been shown to interact with β-catenin and facilitate trans-activation of Wnt-dependent reporter assays and endogenous Wnt target genes in cancer cell lines [12] , [13] . Therefore better understanding of the role of BRG1 in aberrant Wnt signalling may allow the development of novel Wnt intervention strategies . We have recently reported that loss of Brg1 in the small intestinal epithelium results in depletion of the intestinal stem cell population [14] . In this study we aimed to investigate the functional interaction between Brg1 and the Wnt pathway by generating mice , which carried floxed alleles of both Apc [15] and Brg1 [16] , [17] genes , thus placing Brg1 deficiency in the context of aberrant activation of the Wnt pathway . Using three different conditional approaches , we find that additional loss of Brg1 from the small intestinal epithelium attenuates the Wnt-dependent phenotype resulting from Apc deletion .
In order to assess the immediate consequences of Brg1 loss upon Wnt activation , we employed the ( Tg ( Vil-cre/ESR1 ) 23Syr ) transgene [18] driving expression of Cre-ERT2 recombinase under control of the Villin 1 promoter , further abbreviated as VillinCreERT2 . To achieve high penetrance inactivation of targeted genes , VillinCreERT2+Apcfl/fl and VillinCreERT2+Apcfl/flBrgfl/fl mice along with VillinCreERT2− controls were induced with four daily injections of 80 mg/kg Tamoxifen . We sacrificed the mice 4 days after the first induction and collected samples of the small intestinal epithelium . Immunohistochemical analysis of β-catenin and Brg1 expression in the jejunum epithelium at day 4 post induction revealed nuclear localisation of β-catenin in both VillinCreERT2+Apcfl/fl and VillinCreERT2+Apcfl/flBrgfl/fl mice as well as complete loss of Brg1 in double knock-out animals ( Figure 1A ) . Upon histological inspection , the small intestinal epithelium of both VillinCreERT2+Apcfl/fl and VillinCreERT2+Apcfl/flBrgfl/fl mice displayed an aberrant proliferative and apoptotic response synonymous with Wnt pathway activation [19] ( Figure 1A ) . Although no difference in crypt and villus length was observed between VillinCreERT2+Apcfl/fl and VillinCreERT2+Apcfl/flBrgfl/fl mice ( Figure 1B , p>0 . 05 , n≥4 ) , quantitative analysis of other histological parameters such as apoptosis and mitosis levels as well as 5-bromo-2′-deoxyuridine ( BrdU ) incorporation 2 hours after a BrdU pulse showed significant differences between the experimental groups ( Figure 1B–1C , Figure S1A–S1B ) . Quantification of cleaved Caspase3 positive cells showed reduced apoptosis in VillinCreERT2+Apcfl/flBrgfl/fl epithelium ( 1 . 85±0 . 51 and 1 . 0±0 . 40 Caspase3 positive cells per half-crypt , p = 0 . 04 , n = 4 , Figure 1B ) . This observation was supported by scoring of the number of apoptotic bodies in the jejunum of VillinCreERT2+Apcfl/fl and VillinCreERT2+Apcfl/flBrgfl/fl mice ( 4 . 09±0 . 34 and 2 . 67±0 . 45 , p<0 . 01 , n≥4 , Figure S1A ) . Quantification of Ki67 positive cells revealed significantly reduced proliferation in the small intestine of double knock-out mice compared to their Apc-deficient counterparts ( 46 . 18 and 31 . 91 positive cells per half-crypt , pooled standard deviation 4 . 54 , p = 0 . 0004 , n≥4 , Figure 1B ) . Consistent with this observation , scoring of BrdU positive cells 2 hours after labelling showed reduced BrdU incorporation in the jejunum of VillinCreERT2+Apcfl/flBrgfl/fl mice ( 32 . 51±6 . 38 and 23 . 91±0 . 99 , p = 0 . 038 , n≥4 , Figure S1B ) . Cumulative distribution analysis of both Ki67 and BrdU positive cells revealed that the expansion of the proliferative compartment resulting from Apc deletion was attenuated by additional loss of Brg1 ( Figure 1C and Figure S1C , for all comparisons Kolmogorov-Smirnov p<0 . 001 ) . Interestingly , immunostaining with lysozyme antibody showed a difference in the distribution of Paneth cells between VillinCreERT2+Apcfl/flBrgfl/fl and VillinCreERT2+Apcfl/fl mice . Small intestinal epithelium of VillinCreERT2+Apcfl/fl animals displayed mislocalisation of Paneth cells throughout the crypt ( Figure 1A , right panels ) , consistent with our earlier report of Apc loss [19] . Additional deletion of Brg1 in VillinCreERT2+Apcfl/flBrgfl/fl mice restored normal confinement of Paneth cells to the crypt base ( Figure 1A , right panels ) . Quantitative analysis of Paneth cell number and position confirmed this observation as both number and distribution of Paneth cells in VillinCreERT2+Apcfl/flBrgfl/fl epithelium were indistinguishable from normal mucosa ( Figure 2A–2B , for Paneth cell number p = 0 . 999 , for Paneth cell position Kolmogorov-Smirnov p = 0 . 17 ) . Unfortunately , high dose induction resulted in rapid health deterioration of both VillinCreERT2+Apcfl/flBrgfl/fl and VillinCreERT2+Apcfl/fl genotypes such that animals had to be killed by 4 days post induction . We therefore used the in vitro intestinal organoid system to investigate the fate of double knock-out small intestinal crypts beyond this time point . VillinCreERT2+Apcfl/fl and VillinCreERT2+Apcfl/flBrgfl/fl mice along with VillinCreERT2− controls were induced by 3 daily intraperitoneal injections of 3 mg Tamoxifen and sacrificed at day 3 post induction . Small intestinal crypts were isolated and placed in crypt organoid media lacking R-spondin ( crypts from control animals were grown in presence of R-spondin ) . By day 6 of culture ( day 9 post induction ) small intestinal crypts from control animals became complex organoids ( Figure 2C ) . At the same time , the majority of VillinCreERT2+Apcfl/fl crypts developed into large cystic structures ( Figure 2C ) consistent with a previous report of organoid culture under Wnt activated conditions [20] . In contrast , crypts derived from VillinCreERT2+Apcfl/flBrgfl/fl epithelium rarely developed into large cysts and instead the majority remained as small simple spherical organoids indicating that double knock-out crypts had limited proliferative capacity despite constitutive activation of Wnt signalling ( Figure 2C ) . Brg1 loss therefore attenuated the histological and physiological consequences of Apc loss , implicating a compromised Wnt pathway activity . Notably , we consistently observed fewer cells with nuclear β-catenin in the epithelium of double knock-out mice compared to the VillinCreERT2+Apcfl/fl animals . Since nuclear β-catenin is commonly used as a surrogate marker of β-catenin activation , we wished to investigate if Brg1 loss suppressed Wnt signalling upstream of β-catenin activation . Western-blotting analysis of ‘activated’ β-catenin ( dephosphorylated on Ser37 or Thr41 ) levels in the intestinal epithelium at day 4 after high-dose induction did not reveal any differences between the two experimental groups ( Figure 2D ) , suggesting that Brg1 deficiency was likely to suppress the Wnt pathway downstream of β-catenin activation . In order to further investigate the consequences of Brg1 loss on the Wnt pathway , we assessed expression levels of known Wnt target genes in the small intestinal epithelium of VillinCreERT2+Apcfl/fl and double knock-out mice . qRT-PCR analysis of CD44 , c-Myc , CyclinD1 and Ascl2 expression levels revealed significant down-regulation of these genes in VillinCreERT2+Apcfl/flBrgfl/fl mice compared to VillinCreERT2+Apcfl/fl epithelium ( Figure 2E , p<0 . 05 , n≥3 ) . Notably , we did not observe a significant difference in expression levels of Axin2 , which is commonly used as a readout of Wnt pathway activation [21] , indicating possible differential recruitment of Brg1 to distinct Wnt target genes . To further investigate the small intestine-specific role of Brg1 in the Wnt mediated transcriptional program , we performed genome-wide expression analysis of Apc-deficient and double knock-out ( DKO ) epithelium . VillinCreERT2+Apcfl/fl ( n = 3 ) and VillinCreERT2+Apcfl/flBrgfl/fl ( n = 4 ) mice along with VillinCreERT2− controls ( n = 4 ) were induced with four daily injections of 80 mg/kg Tamoxifen and whole epithelial extract was harvested at day 4 post induction . RNA was extracted from epithelial samples , labelled and hybridised to Mouse Ref8 v2 Illumina array . To determine the baseline effect of Brg1 loss , we performed similar analyses on wild type and induced VillinCreERT2+Brgfl/fl samples , which are described in our earlier publication [14] . Analysis of gene expression between Wnt activated VillinCreERT2+Apcfl/fl and control epithelium ( APCvsCTR gene set ) detected 548 unique differentially expressed ‘Wnt target genes’ including a number of previously reported Wnt target genes , such as Cd44 , Axin2 , Lgr5 , Tiam1 , Ephb2 , Efna4 and Sox9 [22] ( Figure 2G , Table S1 ) with four of these genes found to be down-regulated in DKO epithelium compared to Apc-deficient samples ( Table S2 ) . Cluster analysis of genes differentially expressed between Apc-deficient and control epithelium revealed clustering of DKO mice with control samples ( Figure 2F ) , indicating a greater similarity of the gene expression pattern between DKO and wild type samples compared to Apc-deficient intestine . Notably , the majority of gene expression values from DKO samples were intermediate between those of VillinCreERT2+Apcfl/fl and control samples ( Figure 2F ) , indicating attenuation of the Wnt pathway transcriptional signature regardless of whether Wnt activation induced or suppressed gene expression . We also observed a strong negative correlation pattern when directions of gene expression changes between Apc-deficient and wild type epithelium were contrasted with changes between double knock-out and Apc-deficient mice ( Figure S2A , ρ = −0 . 793 , p<0 . 0001 ) . In order to evaluate how many Wnt target genes were attenuated by additional Brg1 loss we compared the overlap between the sets of differentially expressed genes ( Figure 2G ) . We identified 197 unique differentially expressed genes between VillinCreERT2+Apcfl/fl and VillinCreERT2+Apcfl/flBrgfl/fl samples ( DKOvsAPC set ) ( Table S2 ) . Of these , half ( 99/197 ) appeared to be ‘Wnt targets’ defined as those present in APCvsCTR gene set ( Figure 2G , Table S1 ) , representing 18% ( 99/548 ) of the total number of genes that were deregulated following Apc loss ( Table S4 , Figure 2G ) . Since Brg1 loss was able to significantly attenuate deregulation of these Wnt targets , these genes could be designated as ‘highly dependent on Brg1’ ( Table S7 ) . Comparison of gene expression between VillinCreERT2+Apcfl/flBrgfl/fl and control samples ( DKOvsCTR gene set ) identified 200 unique differentially expressed genes ( Figure 2G , Table S3 ) . Comparison of this gene set to the APCvsCTR set revealed that 16% ( 87/548 ) of ‘Wnt target genes’ were present in both sets , but not in the DKOvsAPC set ( Table S5 ) . Since Brg1 loss failed to attenuate deregulation of these genes , they could be considered as ‘Brg1 independent Wnt targets’ ( Table S7 ) . Notably , Brg1 deficiency in double knock-out samples prevented deregulation of 84% ( 458/548 ) of Wnt target genes from their expression levels in control intestine . With the exception of the 99 highly Brg1 dependent Wnt targets these 66% ( 362/548 ) of genes could be designated as ‘Wnt targets with low Brg1 dependency’ ( Table S7 ) . This gene subset included such Wnt targets as Axin2 , Ephb2 and Sox9 . Overall , these data indicated that up to 84% of all Wnt target genes in the small intestinal epithelium are dependent to some extent on Brg1 for either their activation or suppression . It could be argued that gene expression changes upon additional deletion of Brg1 in the Apc-deficient epithelium could arise from the global effect of Brg1 loss on gene expression rather than its specific relevance for the Wnt-driven transcriptional programme . To address this caveat we also analysed the genes differentially expressed between Brg1-deficient and wild type epithelium . Despite a drastic effect on epithelial homeostasis , Brg1 loss induced a relatively moderate perturbation in gene expression at day 4 post induction with 86 genes differentially expressed between Brg1-deficient and control epithelium ( Table S8 ) [14] . Expectedly , a small proportion of the genes differentially expressed between Apc-deficient and double knock-out samples were also de-regulated between Brg1-deficient and wild type epithelium ( 23/197 genes ( 11 . 7% ) Table S9 , Figure S2B ) . This proportion was notably lower when Brg1 targets were compared to the list of Wnt targets with high Brg1 dependency ( 5/99 genes ( 5 . 1% ) , Table S9 ) . Together , these observations strongly suggested that gene expression changes following Brg1 loss in the context of Apc deficiency were largely a result of a specific effect of Brg1 deficiency on the Wnt pathway transcriptional programme rather than an impact of Brg1 deletion on global gene expression . 16 genes which overlapped between the DKOvsCTR and DKOvsAPC sets ( Table S6 , Figure 2G ) appeared deregulated by Brg1 deficiency regardless of Wnt activation and therefore most likely represented direct Brg1 targets independent of Wnt signalling . Most of these genes ( 11/16 ) were also present among genes differentially expressed between Brg1-deficient and control samples ( Table S9 , Figure S2C ) [14] . Notably , one of the Wnt targets , whose expression was attenuated following additional loss of Brg1 , was the proposed intestinal stem cell marker Lgr5 [23] . We therefore decided to explore the effects of Apc deletion and subsequent Brg1 loss on the expression of genes associated with the intestinal stem cell compartment . To this end we compared gene expression changes between our cohorts to the extensive stem cell signature . Using 3 distinct genome-wide expression platforms Muñoz et al . identified 510 genes that were preferentially expressed in the murine small intestinal Lgr5high cells [24] . We identified 460 of these stem cell-specific genes in our array , which corresponded to 721 probes ( Table S10 ) . We then employed ROAST function [25] from the limma Bioconductor package [26] to determine if stem cell-specific genes were enriched among genes differentially expressed between our cohorts . Consistent with the reported stem cell expansion following aberrant activation of Wnt signalling [27]–[29] , 50 . 5% of stem cell-specific genes were found to be up-regulated in VillinCreERT2+Apcfl/fl mice compared to the control cohort ( p<0 . 0001 ) . Double knock-out epithelium also displayed increased expression of stem cell-related genes compared to control samples , however only in 32 . 3% of genes ( p = 0 . 0002 ) . In contrast , 41 . 2% of all stem cell genes were suppressed in double knock-out epithelium compared to VillinCreERT2+Apcfl/fl mice ( p<0 . 0001 ) . We also queried sets of differentially expressed genes between our cohorts for the presence of stem cell-specific genes . While this approach was less sensitive than using ROAST function , we observed a substantial number of stem cell signature genes in our differentially expressed gene sets ( Tables S11 , S12 , S13 ) . In line with the proposed Wnt-driven stem cell expansion [27]–[29] , 10 . 6% ( 58/548 ) of all genes differentially expressed upon Apc deletion were found to belong to the stem cell signature . Of these 98% ( 57/58 ) were found to be up-regulated . In contrast , genes differentially expressed between double knock-out and Apc-deficient intestine contained 11 . 2% ( 22/197 ) of genes from the stem cell signature , all of which were down-regulated ( Tables S11 , S12 , S13 ) . Similar to the pattern of gene expression changes across all genes , stem cell related genes displayed strong negative correlation , when expression changes between Apc-deficient and wild type epithelium were contrasted with ones between double knock-out and Wnt activated intestine ( Figure S2D , ρ = −0 . 859 , p<0 . 0001 ) . Notably , the proposed Wnt-independent intestinal stem cell marker Olfm4 [27] was found to be down regulated in the double knock-out intestine when compared to both Apc-deficient and normal epithelium , indicating strong Brg1 control over Olfm4 expression . The apparent requirement of Brg1 for the expression of Wnt target genes following aberrant Wnt activation in the small intestinal epithelium raises the possibility that Brg1 loss may also prevent Wnt-driven tumour development . To explore this possibility we assessed the effects of Brg1 loss on Wnt-driven adenoma formation by driving recombination of the floxed Apc and Brg1 alleles with the Tg ( Cyp1a1-cre/ESR1 ) 1Dwi transgene , further abbreviated as AhCreERT [30] . This transgene encodes Cre-ERT recombinase under the control of the Cyp1A promoter . In contrast to VillinCreERT2 recombinase , which is expressed in the entirety of the intestinal epithelium , AhCreERT transgene's expression is confined to the stem cell and early progenitor compartments . Additionally , AhCreERT recombinase requires exposure to both β-naphthoflavone and tamoxifen for its activation , which results in tighter control over its activity . We used this approach to inactivate Apc and Brg1 in the small intestinal epithelium at a lower frequency than above , thus extending animal survival and enabling us to analyse the long-term effects of Brg1 loss on Wnt-driven adenoma formation . We induced 3 cohorts of mice ( AhCreERT− controls , AhCreERT+Apcfl/fl and AhCreERT+Apcfl/flBrgfl/fl ) with 5 bi-daily intraperitoneal injections of 80 mg/kg of β-naphthoflavone and 80 mg/kg of tamoxifen . Animals were aged and sacrificed either at day 10 post induction ( n = 4 ) or when they displayed signs of terminal illness ( n≥20 ) . Immunohistochemical analysis of β-catenin localisation in the small intestinal epithelium at day 10 revealed multiple aberrant foci with nuclear β-catenin in both AhCreERT+Apcfl/fl and double knock-out mice , indicating successful activation of the Wnt pathway in those lesions ( Figure 3A , left and central ) . At the same time , Brg1 immunostaining revealed clusters of Brg1 negative cells in AhCreERT+Apcfl/flBrgfl/fl intestinal epithelium ( Figure 3A , right ) . In contrast to the small intestine of VillinCreERT2+Apcfl/flBrgfl/fl mice , where the majority of cells after induction were deficient for Brg1 and displayed nuclear localisation of β-catenin , all the lesions with nuclear β-catenin in the AhCreERT+Apcfl/flBrgfl/fl intestine were Brg1-positive and we failed to detect any overlap between Brg1-deficient clusters and Wnt-activated lesions ( Figure 3A , central and right ) . This observation suggested that , when driven by AhCreERT recombinase , Brg1 loss was incompatible with long term activation of the Wnt pathway and the development of aberrant crypt foci , consistent with inability of crypts from double knock-out epithelium to form aberrant organoids in vitro ( Figure 2C ) . Analysis of the ageing cohorts revealed that combined deletion of Apc and Brg1 deletion provided a survival advantage in comparison to single Apc deletion ( Figure 3B ) . Whilst all the AhCreERT+Apcfl/fl mice became terminally ill within 20 days post induction ( median survival 9 days ) , the majority of AhCreERT+Apcfl/flBrgfl/fl mice survived substantially longer ( median survival 58 days ) with some mice surviving past 100 days ( Figure 3B , Log-rank test p<0 . 0001 , n≥20 for each cohort ) . No control animals ( n = 8 ) developed signs of ill health within the timeframe of the experiment . Histological inspection of the small intestine of the AhCreERT+Apcfl/flBrgfl/fl mice at late time points revealed numerous small lesions confined within the distended villi and identified as micro-adenomas ( Figure 3C ) , as well as rare large adenomas . Immunohistochemical analysis of β-catenin expression showed that all of these lesions were positive for nuclear β-catenin , confirming aberrant Wnt signalling activation ( Figure 3C , left ) . Notably , Brg1 staining demonstrated that all the lesions retained Brg1 expression ( Figure 3C , right ) , indicating that they were likely to originate from cells that recombined at the Apc , but not at the Brg1 loci . Along with the apparent lack of double mutant lesions at day 10 post induction , this indicated that long-term progression of Wnt-driven neoplasia in the small intestine was incompatible with Brg1 deficiency . We therefore explored the possibility that Brg1 loss in the context of activated Wnt signalling could reduce tumour burden and thus improve animal survival . We induced AhCreERT+Apcfl/fl and AhCreERT+Apcfl/flBrgfl/fl mice with two bi-daily injections of 80 mg/kg β-naphthoflavone and 80 mg/kg tamoxifen to achieve attenuated recombination and harvested the small intestinal epithelium at day 40 post induction . We then scored the number of lesions with nuclear β-catenin normalised to the number of normal crypt units contained within the analysed region of the small intestinal epithelium . Quantitative analysis of the tumour burden from two independent experiments revealed a 2 . 92-fold decrease in the number of lesions in the double knock-out small intestinal epithelium compared to AhCreERT+Apcfl/fl animals ( p = 0 . 026 , n≥5 ) indicating reduced tumour burden upon Brg1 loss . Prevalence of micro-adenomas and lack of advanced adenomas in the small intestines of the AhCreERT+Apcfl/flBrgfl/fl mice as late as 100 days post induction indicated that this genetic environment favoured the development of lesions with a limited growth potential . Barker et al . [28] suggested that adenomas originating from intestinal stem cells had a higher tumourigenic potential compared to those derived from transit amplifying cells . To test whether stem cell-specific Brg1 loss could attenuate Wnt-driven adenoma formation we intercrossed mice expressing the GFP-IRES-CreERT2 knock-in allele under control of the Lgr5 promoter ( further abbreviated as Lgr5-GFP-CreERT2 ) [23] with animals bearing targeted Apc and Brg1 alleles . Lgr5-GFP-CreERT2+Apcfl/fl and Lgr5-GFP-CreERT2+Apcfl/flBrgfl/fl mice were induced with four daily intraperitoneal injections of tamoxifen ( initial dose of 3 mg was followed by three doses of 2 mg ) . Animals were aged for 420 days after induction or until they developed signs of ill health . Survival analysis of the cohorts revealed a substantial increase in survival of double knock-out mice compared to their Lgr5-GFP-CreERT2+Apcfl/fl counterparts ( median survival 344 and 110 days post induction respectively , Figure 4A , Log-rank p<0 . 001 , n = 14 for each cohort ) . Immunohistochemical analysis of β-catenin expression in the small intestine of both Lgr5-GFP-CreERT2+Apcfl/fl and Lgr5-GFP-CreERT2+Apcfl/flBrgfl/fl mice revealed a mixture of micro and macro-adenomas with nuclear β-catenin at a range of time points ( Figure 4B ) . Immunohistochemical analysis of GFP expression in small intestinal tumours detected GFP positive cells in both macro and micro adenomas , implying activity of the Lgr5-GFP-CreERT2 transgene and stem cell origin of both types of lesions ( Figure 4B ) . This observation suggested that not all stem cell-derived adenomas were able to progress to advanced stages . Equally , we observed numerous micro adenomas devoid of GFP expression ( not shown ) . These lesions were likely to arise from early progenitor cells , which might have lost self renewal ability , but retained some Cre-ERT2 recombinase expression . Similar to the pattern observed in the small intestine of AhCreERT+Apcfl/flBrgfl/fl mice , adenomas in the small intestine of Lgr5-GFP-CreERT2+Apcfl/flBrgfl/fl mice retained Brg1 expression ( Figure 4B ) providing further support to the notion that Brg1 loss is incompatible with Wnt-driven adenoma formation .
We have demonstrated that inactivation of Brg1 resulted in reduced expression of Wnt target genes following activation of the canonical Wnt pathway via Apc deletion in the small intestinal epithelium . A number of studies have previously suggested that Brg1 facilitates trans-activation of Wnt target genes by activated β-catenin in cancer cell lines [12] , [13] , zebrafish [31] and during mammalian vascular development [32] . Our study provides the first evidence of the functional interaction between Brg1 and the Wnt pathway in the context of intestinal tumourigenesis using an in vivo system . Using transcriptome analysis , we identified sets of Wnt target genes which displayed differing levels of dependency on Brg1 function . Overall , we identified 548 genes that were deregulated upon Apc deletion in the small intestinal epithelium . Of these , 87 genes were Brg1 independent and 461 displayed a variable degree of dependency on Brg1 function , indicating that deregulation of the majority ( 85% ) of Wnt responsive genes in the small intestinal epithelium depended ( to differing degrees ) on the presence of functional Brg1 . This observation is in line with a previous report , which found a similar proportion ( 68% ) of Wnt targets to rely on Brg1 for their response to Wnt activation in the HEK293T cell line [33] . Consistent with the requirement for Brg1 in the maintenance of the small intestinal stem cell population [14] , we observed attenuated expression in a range of genes associated with the small intestinal stem cells following Brg1 loss in the context of aberrantly activated Wnt signalling . These genes included functionally validated stem cell markers such as Lgr5 , Ascl2 and Olfm4 , suggesting that Brg1 loss was able to impair the expansion of the ‘stem-like’ cell population characteristic of Wnt-driven intestinal tumourigenesis [27]–[29] . Furthermore , expression levels of Olfm4 in the small intestine of double knock-out mice were lower compared to that in both wild type and Apc-deficient epithelium , strongly suggesting a role of Brg1 in regulation of Olfm4 expression . Olfm4 is a secreted anti-apoptotic factor , which has been reported to be over-expressed in a variety of tumours [34] . Depletion of Olfm4 in gastric cancer cells has been reported to suppress proliferation and sensitise cancer cells to apoptosis [35] , [36] . Olfm4 thus constitutes a potentially attractive therapeutic target , especially so in view of its secreted nature , which makes it a feasible target for monoclonal antibody therapies . In addition to suppression of Wnt target genes , Brg1 loss also attenuated the physiological manifestations of Wnt activation in the intestinal epithelium , most notably the increased cell proliferation and mislocalisation of Paneth cells . Both Paneth cell quantity and position were restored to normal levels in the double knock-out epithelium , suggesting that Brg1 deletion was able to preserve physiological levels of EphB/Ephrin signalling in the context of Apc deficiency [37] . Consistent with this notion , our transcriptome analysis detected up-regulation of EphB2 expression in Apc-deficient intestine , but not in double knock-out epithelium , when compared to control samples . A similar effect of Wnt signalling attenuation on the Paneth cell mislocalisation has been previously demonstrated following loss of Mbd2 in the Apc-deficient small intestine [38] . Given the proposed role for Paneth cells as the intestinal stem cell niche [20] , this effect of Brg1 loss on Paneth cell mislocalisation could contribute to the attenuated stem cell expansion in double knock-out intestine . In the view of extensive role of Wnt signalling in tumourigenesis [1] , Brg1 mediated modulation of the Wnt pathway may have implications for the development of novel therapeutic approaches . Here we report that loss of Brg1 in the context of Apc deletion improved animal survival by preventing the formation of double mutant adenomas . All adenomas observed in the small intestine of double knock-out mice retained Brg1 expression indicative of their origin from cells that had lost Apc but had escaped Brg1 deletion . This implies an absolute requirement of functional Brg1 for Wnt-mediated tumourigenesis in this tissue . A similar relationship between Brg1 inactivation and a loss of tumour suppressor Snf5 ( Ini1 ) has been reported by Wang et al . [39] . Simultaneous inactivation of Brg1 and Snf5 under control of the T-cell lineage-specific Lck-Cre recombinase resulted in decreased tumour incidence and a longer disease onset . Similar to our observations , all the Snf5-deficient tumours , which developed in double-mutant animals , retained Brg1 expression [39] . Barker et al . [28] suggested a particular role for intestinal stem cells in Wnt-driven tumour initiation with non-stem cells giving rise to adenomas with limited growth capacity , while stem cell gene signature in human primary colorectal cancers was found to be associated with more aggressive phenotype [40] . In a similar fashion , Alcantara Llaguno et al . [41] reported neural stem cells as the cell of origin for malignant gliomas . Furthermore , a recent study has demonstrated a substantial survival advantage of genetic targeting of glioblastoma cancer stem cells using a neural stem cell marker Nestin [42] . Consistent with these reports , and in line with negative impact of Brg1 loss on long-term small intestinal stem cell survival , we observed a markedly improved survival upon stem cell-specific Brg1 deletion in the context of aberrant Wnt activation . Notably , the above study made use of stem cell-specific expression of the toxic thymidine kinase transgene to target the sub-population of cancer cells bearing normal stem cell markers [42] . In contrast to this , we report successful tumour suppression by targeting a gene required for the physiological small intestinal stem cell maintenance , a relationship , which to our knowledge has not been previously reported . It should be noted that mice with stem cell-specific Apc loss alone in the present report exhibited longer survival times compared to those in the study by Barker et al . [28] , which could be attributed to increased levels of silencing of the Lgr5-GFP-CreERT2 transgene in our mice . Tumours in the small intestine of animals bearing the Lgr5-GFP-CreERT2 transgene and targeted Apc alleles were mainly detected in the distal third of the small intestine with fewer lesions in the proximal part and very few tumours in between . Given that biallelic Apc deletion would fall within ‘high pathological Wnt’ scenario described in Leedham et al . [43] , this tumour distribution was consistent with the gradient of the Wnt signal and stem cell density in the murine intestine . It should be noted that given the concurrency of Brg1 and Apc deletion , results presented in this report pertain to cancer prevention rather than therapy . Additionally , Brg1 loss-driven elimination of intestinal stem cells is likely to be a major contributing factor to attenuated tumour burden in double knock-out mice and may therefore obscure the effects of Brg1 deletion on non-stem tumour cells . Importantly , this toxicity of Brg1 loss in respect to the small intestinal stem cell homeostasis constitutes a potential serious caveat to the use of Brg1 as a therapeutic target . However our data from the previous report [14] as well as from AhCreERT+Apcfl/flBrgfl/fl and Lgr5-GFP-CreERT2+Apcfl/flBrgfl/fl models in the present study demonstrate that partial Brg1 loss is well tolerated by the small intestinal epithelium , which is gradually repopulated with wild type cells . At the same time , partial loss of Brg1 was sufficient to reduce the tumour burden , suggesting that a therapeutic window may exist that would allow targeting Brg1 in the intestinal polyps , while allowing repopulation of the normal intestinal epithelium . A strategy that would allow for the efficient Brg1 deletion in existing adenomas or the use of a Brg1 inhibitor would be required to further address the effects of Brg1 in non-stem cell portion of established Wnt-driven tumours , as well as the potential therapeutic window of targeting Brg1 . In summary , we demonstrate using mouse models of intestinal cancer that Brg1 is essential for Wnt-driven tumourigenesis in the murine small intestine with attenuation of Wnt target gene expression and elimination of transformed stem cells as two likely mechanisms . Brg1 therefore constitutes a potential therapeutic target in cancers with an aberrantly activated Wnt pathway . Combined with our earlier observation that Brg1 is essential for stem cell maintenance in the small intestinal epithelium under the physiological conditions , these data may serve as a proof of concept that targeting the somatic stem cell as a cancer initiating cell may provide a valuable therapeutic approach , especially in the context of predisposition to Wnt-driven carcinogenesis , such as in Familial Adenomatous Polyposis patients .
All experiments were carried out in accordance with Animals ( Scientific Procedures ) Act 1986 under project licenses 30/2246 and 30/2737 issued by UK Home Office . The study was approved by the Cardiff University Research Ethics Committee . Mice were maintained on an outbred background and genotyped as described previously for targeted Apc allele [15] , Cre-ERT and Cre-ERT2 transgenes [19] , targeted Brg1 allele [16] . Detailed induction protocols and dissection procedures are described in Text S1 . Detailed description of protocols for tissue fixation , processing , immunohistochemistry , and quantitative analysis of tissue sections is available in Text S1 . All statistical tests except survival and cumulative distribution analyses were carried out using R software [44] . Scoring data were tested for normality using Shapiro-Wilk test and for equal variance using Levene's test . Normally distributed data with equal variance were tested for difference between means using one-way ANOVA . Where appropriate , p values were adjusted for multiple testing using TukeyHSD function in R . In cases of unequal variance between groups the difference between those groups was tested using t-test not assuming equal variance . Unless otherwise specified , pooled standard deviation from one-way ANOVA was used to represent error bars . Positional data were analysed for differences in distribution with Kolmogorov-Smirnov Z-test in SPSS ( version 16 . 0 . 2 ) . Kaplan-Meier survival curve and Log-rank survival analysis were carried out using GraphPad Prism ( version 6 ) . Detailed description of the protocols and statistical methods used for RNA extraction , qRT-PCR and transcriptome analysis is available in Text S1 . Microarray analysis was carried out using beadarray [45] and limma [26] packages from Bioconductor project . Microarray data for the study are publicly available from the GEO repository ( http://www . ncbi . nlm . nih . gov/geo ) under the series record GSE46129 . A detailed protocol for obtaining epithelial-enriched population of cells for protein extraction and western-blotting is provided in Text S1 . Protein was extracted from the epithelium-enriched small intestinal samples , separated , transferred and probed as described previously [46] . The primary antibodies used: mouse anti-active β-catenin ( 1∶2000; Millipore ) and mouse anti-β-actin ( 1∶5000; Sigma ) . Anti-mouse horseradish peroxidase conjugated secondary antibody ( 1∶3000; GE Healthcare ) and ECL or ECL Plus reagents ( Amersham Biosciences ) were used to detect the signal according to the manufacturer's manual . Small intestinal crypts from VillinCreERT2− controls , VillinCreERT2+Apcfl/fl and VillinCreERT2+Apcfl/flBrgfl/fl mice were isolated and cultured as described in Sato et al . [47] with minor adjustments . Detailed procedures are described in Text S1 .
|
Aberrant Wnt signalling is responsible for the majority of colorectal cancers , the third leading cause of cancer-related mortality in the UK . However , no therapies directly targeting Wnt signalling are currently available . Using mouse models of intestinal cancer , we demonstrate that deleting chromatin remodelling factor Brg1 in the context of Apc-deficient small intestinal epithelium attenuates Wnt-driven gene expression changes and prevents adenoma formation , which results in extended animal survival . We also demonstrate that Brg1 loss impairs the small intestinal stem cell expansion associated with aberrant activation of Wnt signalling . These findings highlight Brg1 as a potential therapeutic target in Wnt-driven intestinal tumourigenesis and illustrate the viability of targeting the somatic stem cell as the ‘cell of origin’ of cancer , which might be particularly valuable in patients with known predisposition to cancer .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"signal",
"transduction",
"medicine",
"and",
"health",
"sciences",
"cancer",
"genetics",
"cell",
"biology",
"chromosome",
"biology",
"gastroenterology",
"and",
"hepatology",
"gene",
"expression",
"genetics",
"biology",
"and",
"life",
"sciences",
"gastrointestinal",
"cancers",
"molecular",
"cell",
"biology",
"cell",
"signaling",
"chromatin"
] |
2014
|
Brg1 Loss Attenuates Aberrant Wnt-Signalling and Prevents Wnt-Dependent Tumourigenesis in the Murine Small Intestine
|
In cases of Dengue fever , late hospital admission can lead to treatment delay and even death . In order to improve early disease notification and management , it is essential to investigate the factors affecting the time of admission of Dengue cases . This study determined the factors associated with the time of admission among notified Dengue cases . The study covered the period between 2008 and 2014 in Region VIII , Philippines . The factors assessed were age , sex , hospital sector , hospital level , disease severity based on the 1997 WHO Dengue classification , and period of admission ( distinguishing between the 2010 Dengue epidemic and non-epidemic time ) . We analysed secondary data from the surveillance of notified Dengue cases . We calculated the association through chi-square test , ordinal logistic regression and linear regression at p value < 0 . 05 . The study included 16 , 357 admitted Dengue cases . The reported cases included a majority of children ( 70 . 09% ) , mild cases of the disease ( 64 . 00% ) , patients from the public sector ( 69 . 82% ) , and non-tertiary hospitals ( 62 . 76% ) . Only 1 . 40% of cases had a laboratory confirmation . The epidemic period in 2010 comprised 48 . 68% of all the admitted cases during this period . Late admission was more likely among adults than children ( p<0 . 05 ) . The severe type of the disease was more likely to be admitted late than the mild type ( p<0 . 05 ) . Late admission was also more likely in public hospitals than in private hospitals ( p<0 . 05 ) ; and within tertiary level hospitals than non-tertiary hospitals ( p<0 . 05 ) . Late admission was more likely during the non-epidemic period than the 2010 epidemic period ( p<0 . 05 ) . A case fatality rate of 1 or greater was significantly associated with children , severe diseases , tertiary hospitals and public hospitals when admitted late ( p<0 . 05 ) . Data suggests that early admission among child cases was common in Region VIII . This behavior is encouraging , and should be continued . However , further study is needed on the late admission among tertiary , public hospitals and non-epidemic period with reference to the quality of care , patient volume , out of pocket expense , and accessibility We recommend the consistent use of the 2009 WHO Dengue guidelines in order to standardize the admission criteria and time across hospitals .
Dengue fever is endemic in the Philippines . The Department of Health ( DOH ) reported 59 , 943 cases between January and September 2014 , with an incidence of 60 cases per 100 , 000 [1] . Sixty five percent of the cases in the country were admitted , while 35% were treated as outpatient [2] . Mortality in Dengue cases is often associated with treatment delay [3] . The delay arises mainly from late hospital admission . Multiple factors influence the time of admission–for example , distance from hospitals or poor recognition of warning signs and symptoms [4] . The ownership of hospitals ( ie . whether they are private or public ) also has a significant influence on health-seeking behaviour [5 , 6] . In private hospitals , early consultations are deterred by the immediate costs [5] . Data suggests that late admissions in public hospitals could be related to the lack of confidence in their services [6] . The presence of an epidemic can also cause a change in the time of admission . In a 2007/2008 epidemic in Rio de Janiero , Dengue cases frequently received primary medical care on the third day from the onset of illness [7] . This study assessed the effect of the 2010 epidemic in Region VIII [8] , Philippines on the time of admission . Differences in the admission time have also been associated with disease severity . Non-severe cases were usually admitted on the third and fourth day since the onset of the disease , while severe cases were admitted later [7] . In a 2006 study by Tomaschek in Puerto Rico , Dengue cases were seen by a clinician at least once , without appropriate assessment of the severe warning signs , before being admitted [9] . Late admission of severe cases among adults has been reported by their unusual clinical presentation , such as severe Dengue with no fever [7] . Early and improved management can reduce morbidity and mortality among Dengue patients ( 3 , 4 ) . In general , case fatality rate ( CFR ) of 1 may be considered a consequence of insufficient management , late diagnosis and hospital admission [10] . Therefore , understanding the factors related to the time of hospitalization can improve Dengue management . In this study , we determined the factors associated with hospital admission time among Dengue cases . The study period was from 2008 to 2014 in Region VIII , Philippines . The results of this study may provide recommendations for organizational policies and treatment protocols to improve the admission time of Dengue patients .
Dengue cases are reported weekly to the Philippines Integrated Disease Surveillance and Response ( PIDSR ) through passive surveillance and active sentinel surveillance [11] . We conducted an exhaustive retrospective sampling and analysis of the notified cases of Region VIII . The study period lasted from January 1 , 2008 to December 31 , 2014 . The case definition of Dengue was based on the 1997 World Health Organization ( WHO ) classification [11] . For this study , Dengue fever ( DF ) was operationally defined as a mild case . The Dengue hemorrhagic fever ( DHF ) and Dengue shock syndrome ( DSS ) were grouped together , and defined as a severe case . We measured the outcome variable as the time of admission , measured in the number of days between the onset of illness and the time of admission . The variable was categorized into early ( 0–2 days ) , regular ( 3–5 days ) and late ( 6 or more days ) admission . These categories were arbitrarily derived from the clinical phases and period of Dengue: febrile , critical and recovery [12] . We restricted the explanatory variables to the set of data available in the PIDSR , namely disease severity , age and sex of the patient , hospital sector , hospital level and period of admission . The hospital sector was categorized into public and private ownership . The hospital level was categorized into tertiary and non-tertiary hospitals [13] . The tertiary hospitals were generally located in the urban area of the region . The period of admission was divided into the epidemic ( 2010 ) and non-epidemic period ( 2008 , 2009 , 2011 , 2012 , 2013 , 2014 ) . The year 2010 was defined as the epidemic period because of the remarkable increase in the number of cases compared with other years . Only admitted cases were included in the study . Values with a missing date of admission and onset of illness were excluded . Cases with an admission time of over 90 days were excluded , since Dengue can only be confirmed serologically by IgM until 90 days [14] . We established the association of the explanatory variables to the outcome variable through chi-square test in the univariate analysis and ordinal logistic regression in the multivariable analysis . We also compared the case fatality rates ( CFR ) across varying times of admission by different factors using linear regression . The p-value was set at <0 . 05 . The model used was the proportional odds assumption for ordinal logistic regression [15] . It assumed that the coefficients between outcome categories ( time of admission ) were similar . We assumed the coefficients of early admission versus regular and late admission were similar , in regular admission versus late admission . This is called the proportional odds assumption or the parallel regression assumption . All statistical analyses were performed with RStudio Version 0 . 98 . 1103 and packages MASS , Hmisc , reshape2 , foreign , ggplot2 , rms , gridExtra ( The R Foundation for Statistical Computing , Vienna , Austria; http://www . r-project . org ) . The study was authorized and exempted from the bioethics approval though the DOH Region VIII director . The probability of physical , psychological , social , or economic harm occurring as a result of being included in the research study was minimal . The names of the patients were replaced with unique keys , ensuring the confidentiality of the information gathered .
There were 21 , 480 reported cases from 2008 to 2014 in Region VIII . The epidemic period ( 2010 ) accounted for 11 , 974 cases ( 56 . 68% ) . Only 1 . 40% of these cases were confirmed with laboratory tests . Most cases ( 85 . 74% ) were admitted in hospitals . A total of 16 , 357 admitted cases ( 76 . 15% ) were included in the study after excluding the repeated cases ( 1 . 38% ) , non-Region VIII cases ( 0 . 27% ) , outpatients ( 14 . 00% ) , missing values of the outcome variable ( 7 . 29% ) , cases with missing values of the outcome variable entries ( 1 . 38% ) , and no values for the explanatory variable ( 0 . 52% ) , ( Fig 1 ) . Males and females were equally distributed among admitted cases . Most of the admitted cases were children ( 70 . 09% ) , and those suffering from a mild disease ( 64 . 00% ) . Most of the cases were reported from the public sector ( 69 . 82% ) and non-tertiary hospitals ( 62 . 76% ) . The number of cases was equally proportional during the epidemic and non-epidemic period ( Table 1 ) . The univariate analysis ( Table 2 ) indicated significant association of the time of admission to disease severity , patient’s age , hospital sector , hospital level and period of admission ( p<0 . 05 ) . There was no association between sex and time of admission . Severe cases were admitted later than mild ones , and adult cases were admitted later than child cases . Although elderly cases ( aged 65 and above ) were also admitted later than child cases , there were only 64 elderly cases in this study . Public hospital cases were admitted later than private hospital cases . Similarly , tertiary hospital cases were admitted later than non-tertiary ones . Cases during the non-epidemic period were admitted later than cases during the 2010 epidemic . We stratified and described the time of admission by disease severity for sex and age of the patient , hospital level , hospital sector and period of admission ( S1 Table ) . Both mild cases and severe cases , indicated significant association of the time of admission with age , hospital sector , hospital level and period of admission ( p<0 . 05 ) but not with sex . Mild cases comprised 69% of the cases in public hospitals , as opposed to 52% in private hospitals . Mild cases made up 78% of the cases in non-tertiary hospitals , as compared with 40% in tertiary hospitals . Both mild and severe cases from public and tertiary hospitals were admitted later than those in private and non-tertiary hospitals . Both severe and mild cases have a significant association between the time of admission and age ( p<0 . 05 ) . Severe cases made up 37 . 86% of cases among children , as opposed to 31 . 62% in adults . There were only 22 severe cases among the elderly . Adults were admitted later than children . The time of admission for both severe and mild cases was also significantly associated to the epidemic or non-epidemic period . We found significant association between disease severity , age of patient , hospital sector , hospital level and period of admission in our multivariate analysis ( Table 3 ) . Late admission was more likely when the following factors occurred: a ) severe case of the disease ( p<0 . 05 ) in comparison with the mild type; b ) adult patients rather child patients ( p<0 . 05 ) ; c ) public hospital admission as opposed to private hospital ( p<0 . 05 ) ; d ) tertiary level hospital admission rather than non-tertiary ( p<0 . 05 ) ; and e ) non-epidemic period as opposed to an epidemic period ( p<0 . 05 ) . The factor with the highest influence was public sector hospital , with a higher probability of late admission in comparison with the private sector . There was no association between sex and the time of admission ( p = 0 . 62 ) . There was no significant difference in admission time between the elderly and children ( p = 1 . 37 ) . The results obtained in S3 Table were used to assess the proportional odds assumption of the model . The differences in the predicted coefficients in each level were only within the range of 0 . 01 to 0 . 4 . This suggests that the coefficients at different levels of the time of admission were similar in all the risk factors assessed , and that the proportional odds assumption was held in the model ( 15 ) . There were 106 deaths among the notified dengue cases with an overall CFR of 0 . 65 . There was a general increase in the CFR across time of admission . The CFR was equal or greater than 1 during late admission amongst children , severe disease , tertiary hospital , public hospital and females ( Fig 2 ) . However , the CFR at different times of admission was only significantly associated with age , severity , hospital level and hospital sector ( p<0 . 05 ) . It was not associated with sex ( p = 0 . 32 ) and epidemic period ( p = 0 . 15 ) ( S2 Table ) .
Severe cases were more likely to be admitted late in comparison to mild ones . The CFR of severe cases exceeded 2 during late admission . Patients with mild disease generally seek medical attention during the period 2–4 days after the onset of illness [14 , 21] . Under the 1997 classification , severe cases of Dengue are categorised as cases where hemorrhage or shock occurs [3 , 12] . These symptoms are generally present later than two days from the onset of illness , or at a late stage [14 , 16 , 18] which falls on the period of late admission . Hence , late admission may have been favoured among severe cases with high CFR based on the 1997 classification . In 2009 , WHO reclassified Dengue fever based on early warning signs to detect early on cases likely to deteriorate [22] . This new classification has increased sensitivity to severe cases of Dengue [23] . However , its use in the Region VIII surveillance was only initiated in 2013 , therefore it was not used in this study . Dengue infection severity can vary among individuals . It has been suggested that there are higher levels of viral load in severe cases correlating to the seriousnes of the disease [24] . The mechanism remains to be determined through viral serotyping , investigating susceptibility and tracing the sequence of the infection [25] . The symptoms of Dengue seemed to cause little alarm for adult patients , who were generally more likely to experience late admission than children ( Table 2 ) . On the other hand admission time amongst the elderly were similar to those found amongst children . A study in Southeast Asia found that cases of Dengue amongst adults showed mild signs and symptoms [25] . In contrast , a majority of the severe cases occur in children aged 2–15 years . Adults apparently acquire immunity from primary infection and avoid DHF [25] . S1 Table of this study is consistent with literature revealing a higher percentage of severe cases in children rather than adults [16 , 25] . Furthermore , the CFR among children during late admission exceeded 1% . This emphasizes the important practice of early admission among children who are likely to have severe Dengue . Dengue patients treated in public hospitals were more likely to be admitted late than those in private hospitals , particularly in mild cases . There was no difference in admission time between private and public sectors in both mild and severe cases ( S1 Table ) . The difference in admission time beween the hospital sectors suggests that there may be a disparity in the clinical practice between public and private hospitals . The CFR in public hospital during late admission approached 1% in comparison to the private hospital . The financial capacity and remote location of patients influence their health-seeking behaviour . Public hospitals are frequently visited by patients with more limited economic resources who may endure symptoms of disease until they are critically ill [26] . Three quarters of the patient load in our study were admitted by public hospitals . In these hospitals , the doctor’s fee and accommodation are government-subsidized [27] . However , public hospitals are often so full that there are no vacant beds to accommodate additional admissions . They also have meager medical supplies and insufficient personnel , which can lead to long waiting hours [27] . As a result of poor health facilities and the patients’ limited financial capacity , the late admission is more likely in public hospitals than private ones . In contrast , private hospitals experience a more reduced occupancy rate [27] . According to anecdotal descriptions from Region VII , private hospitals tend to have higher and faster admission rates , even in mild cases , in order to increase the number of clients for commercial purposes . In a 2013 report by the National Statistics Coordinating Board , 3 public hospitals and 7 private hospitals in Region VII were classified as tertiary hospitals , while 46 public hospitals and 21 private hospitals were classified as non-tertiary hospitals [28] . Most of the cases in this study were reported from non-tertiary hospitals ( 63% ) . In general , tertiary hospitals admit Dengue cases later than non-tertiary ones ( Table 3 ) . When we stratified the time of admission by disease severity , it appeared that late admission among tertiary hospitals was evident in both mild and severe cases ( S1 Table ) . There may be a disparity in the time of admission for mild cases . Most non-tertiary hospitals are funded by the private sector or local government . Non-tertiary hospitals are generally perceived as providing low quality services because of their meager resources [27] . Patients who bypass their services and seek treatment in better equipped tertiary hospitals have to travel to the urban area , where national public hospitals or large private hospitals are situated [27] . These tertiary hospitals have a higher likelihood of late admission than non-tertiary hospitals . The CFR in tertiary hospitals during late admission exceeded 1 . Tertiary hospitals are overburdened , because Dengue cases are more common in densely populated urban areas where vectors proliferate [10] . The 2010 epidemic accounted for the highest number of reported Dengue cases . The likelihood of late admission was higher during the non-epidemic period than the epidemic period . This is contrary to evidence suggesting that a high volume of patients can overburden and delay health services [29 , 30] . During the 2010 epidemic , the number of admissions increased to five times the average per year as compared to the non-epidemic period . The non-epidemic period covered 6 years which averaged 1 , 399 cases per year . The 2010 epidemic had 7 , 963 cases in a single year ( Table 1 ) . Anecdotal reports from the hospital staff revealed the use of the lobby and room extensions to accommodate the large number of patients . During the 2010 epidemic , public campaigns on early consultation were conducted . Individuals with fever lasting two or more days were advised to immediately seek consultation [8] . The community became more prone to suspecting any febrile condition to be Dengue fever . Anecdotal observations also suggest that increased awareness encouraged to promptly diagnose and admit Dengue cases . This may explain the lower likelihood of late admission during the epidemic period . However , the earlier admission may have occurred at the expense of the quality of services . The 1997 WHO Dengue guidelines classification was used in this study . The 1997 and 2009 Dengue guidelines use different criteria for categorizing clinical Dengue case severity [3 , 9] . The 1997 guidelines favour late admission for severe cases , since haemorrhagic and shock presentations naturally occur during the late phase of the disease [3] . The 2009 guidelines screen through warning signs that could occur during the early to late phase of the disease . However , the 2009 WHO classification was only adopted in Region VIII surveillance data in 2013 . As a result , our study only included a limited number of cases categorized under this new classification . The majority of Dengue cases were suspected and not laboratory confirmed , and thus based on clinical presentation . Due to the lack of serological laboratory confirmation , it is possible that cases that were not caused by Dengue were nonetheless included in the study . Other febrile illnesses may have been classified as Dengue in the study , especially during the 2010 epidemic . It is also worth bearing in mind that other diseases may have been classified as Dengue for insurance coverage reasons , seeing as a higher reimbursement is granted for cases of Dengue fever . An overestimation of the number of cases may have occurred , but there is no evidence that this affected the distribution of the cases in terms of admission time . Factors such as chronic diseases , concomitant diseases , immunologic status , Dengue virus serotype , secondary infection , body mass index , inter-hospital transfer , hospital accessibility , or socioeconomic level may also have influenced the time of admission . However , these factors were not included in the analysis as they are not part of the surveillance information collected . In Region VIII , the Philippines , late admission of Dengue cases was more likely among adults , public hospitals , tertiary hospitals and during non-epidemic period . Among the factors , the highest likelihood of late admission was in public hospitals than in private hospitals . Late admission was associated with hospital sector and level . These may be influenced by patients’ financial capacity , a high patient load and lack of resources in healthcare facilities , the geographic location of hospitals , and noncompliance with Dengue guidelines . The study also suggests a higher likelihood of early admission among children in comparison to adults . This behavior should be encouraged , as severe cases are more common among children than adults . The severe cases of Dengue were more likely to be admitted late than mild cases . However , severe diseases have late presentation which may have favoured their late admission . A CFR of 1 . 00 or greater was observed during the late admission of children , severe diseases , tertiary hospitals and public hospitals . These suggest that admission time and management should be improved in these factors to minimize death . Earlier admission was more likely during the 2010 Dengue epidemic in Region VIII than during the non-epidemic period , suggesting behavioural change attributable to increased awareness of the disease . Improvements in behaviour related to admission time can be brought about through increased knowledge , practice and compliance to the Dengue guidelines by the healthcare workers and population . We recommend the improvement and consistent implementation of hospital admission practices . Firstly , in order to facilitate early admission and prevent fatality , health care workers should be able to identify severe Dengue fever cases . The identification of severe cases may improve with the use of the 2009 WHO Dengue guidelines classification , which examines warning signs and indicates when the patient is admissible . Trainings should be conducted among healthcare professionals . The use of the 2009 WHO guidelines as the basis for admission , and the confirmation of the diagnosis through serological tests should be discussed and agreed on among doctors , surveillance staff and health insurance personnel . The guidelines are expected to standardize the diagnosis and hospital admission time across different hospital sectors and levels . Through public campaigns , the practice of early admission among children suffering from severe cases of the Dengue disease should be promoted among healthcare professionals . During epidemics , the provision of examining stations exclusively for children can ensure the on-going practice of early hospitalization . Secondly , the reasons for late admission in public and tertiary hospitals must be studied further in order to improve promptness in Dengue case management . Hence , further research will be needed to assess how admission time is affected by the volume of cases , accessibility and out-of-pocket expense of these hospitals The quality of health care services must also be further explored to understand how the health system adapted during the epidemic , and why admission time was earlier during the epidemic rather than non-epidemic period .
|
A variety of factors affect the time of admission of Dengue fever cases . These must be investigated , as delayed treatment of this disease can result in death . The authors of this study determined the factors associated with the time of admission among notified Dengue cases of Region VIII , Philippines , from 2008 to 2014 . The factors assessed were age and sex of the patient , hospital sector , hospital level , disease severity and the presence of Dengue epidemic . A secondary surveillance data of Dengue was used . The associations were determined using chi-square test and regression . Late admission was more likely amongst adults , severe cases of the disease , public hospitals , tertiary level hospitals , and during the non-epidemic period . In comparison , early admission was more likely in cases concerning children , mild cases of the disease , private hospitals , non-tertiary hospitals and during an epidemic period . Case fatality was significantly associated to children , severe diseases , public hospitals and tertiary hospitals when admitted late . The routine early admission of children should be promoted , as severe cases of Dengue fever are more likely among children . Consistent admission criteria for Dengue should be implemented across all hospital sectors and levels .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"medicine",
"and",
"health",
"sciences",
"elderly",
"tropical",
"diseases",
"geographical",
"locations",
"health",
"care",
"age",
"groups",
"chi",
"square",
"tests",
"adults",
"mathematics",
"statistics",
"(mathematics)",
"neglected",
"tropical",
"diseases",
"research",
"and",
"analysis",
"methods",
"infectious",
"diseases",
"dengue",
"fever",
"epidemiology",
"mathematical",
"and",
"statistical",
"techniques",
"hospitals",
"people",
"and",
"places",
"statistical",
"hypothesis",
"testing",
"philippines",
"health",
"care",
"facilities",
"asia",
"disease",
"surveillance",
"population",
"groupings",
"viral",
"diseases",
"geriatrics",
"physical",
"sciences",
"statistical",
"methods"
] |
2016
|
Factors Associated with the Time of Admission among Notified Dengue Fever Cases in Region VIII Philippines from 2008 to 2014
|
Central chemoreceptors are highly sensitive neurons that respond to changes in pH and CO2 levels . An increase in CO2/H+ typically reflects a rise in the firing rate of these neurons , which stimulates an increase in ventilation . Here , we present an ionic current model that reproduces the basic electrophysiological activity of individual CO2/H+-sensitive neurons from the locus coeruleus ( LC ) . We used this model to explore chemoreceptor discharge patterns in response to electrical and chemical stimuli . The modeled neurons showed both stimulus-evoked activity and spontaneous activity under physiological parameters . Neuronal responses to electrical and chemical stimulation showed specific firing patterns of spike frequency adaptation , postinhibitory rebound , and post-stimulation recovery . Conversely , the response to chemical stimulation alone ( based on physiological CO2/H+ changes ) , in the absence of external depolarizing stimulation , showed no signs of postinhibitory rebound or post-stimulation recovery , and no depolarizing sag . A sensitivity analysis for the firing-rate response to the different stimuli revealed that the contribution of an applied stimulus current exceeded that of the chemical signals . The firing-rate response increased indefinitely with injected depolarizing current , but reached saturation with chemical stimuli . Our computational model reproduced the regular pacemaker-like spiking pattern , action potential shape , and most of the membrane properties that characterize CO2/H+-sensitive neurons from the locus coeruleus . This validates the model and highlights its potential as a tool for studying the cellular mechanisms underlying the altered central chemosensitivity present in a variety of disorders such as sudden infant death syndrome , depression , and anxiety . In addition , the model results suggest that small external electrical signals play a greater role in determining the chemosensitive response to changes in CO2/H+ than previously thought . This highlights the importance of considering electrical synaptic transmission in studies of intrinsic chemosensitivity .
Central chemoreception is a neuronal sensory mechanism by which changes in CO2 and H+ levels in the brain are detected [1–3] . It occurs in specialized CO2/H+-sensitive centers in the brainstem that are involved in the neuronal network that regulates autonomic ventilation [4–10] . Regular ventilatory movements are controlled by respiratory neurons in the brainstem , which generate an appropriate respiratory rhythm and control the motor neurons that innervate the respiratory muscles [11–14] . Even small alterations in CO2/H+ levels in the blood and/or cerebrospinal fluid cause changes in ventilation . Brainstem neurons are considered the main sensory elements in the homeostatic regulation of respiratory gases [15 , 16] , and when these neurons are exposed to elevated CO2/H+ ( hypercapnia and/or acidosis ) , there is a noticeable increase in their firing rate . This change in firing rate can be triggered by several signaling pathways alone or in combination , such as a decrease in intracellular or external pH [17 , 18] , an increase in intracellular HCO3− [19] and/or a direct increase in CO2 [20] . The changes in firing rate of neurons from chemosensitive regions have been investigated under conditions of hypercapnic acidosis ( HA ) in some areas of the brainstem , such as the retrotrapezoid nucleus [21] , nucleus tractus solitarii [22 , 23] , locus coeruleus [24–26] and the medullary raphe [27] . It is often assumed that these neurons are intrinsically responsive to changes of CO2 , which means that they not simply respond to altered synaptic input from other chemosensitive neurons . In particular , neurons from LC have been demonstrated to be responsive when exposed to altered levels of CO2 in the presence of synaptic block media . However , it remains unclear whether their firing-rate response to increased levels of CO2/H+ can be completely attributed to intrinsic mechanisms of individual neurons or if the response is mediated in part by unnoticed inputs from chemical synapses or gap junctions [18 , 23 , 28] . Moreover , since there have been postulated a growing number of signals implicated in the chemosensitive response of individual neurons from the LC region [17–20] , the knowledge about the various signals , both excitatory and inhibitory , is getting so complex that our ability to understand the interactions between these differing components needs to be aided by mathematical and computational approaches . In particular , a single-cell neuron model is required to elucidate the effect of individual signals and their interaction on the chemosensitive response of individual neurons . Such an approach would help to elucidate whether their spike patterns and discharge frequencies are significantly affected by individual signals , by the combined effect of different stimuli , or by a contribution of small external inputs . Experimental evidence suggests that neurons from different chemosensitive regions are similar in their basal discharge frequencies and spiking patterns in response to current injection [27–29] . On the other hand , some mathematical models of the Hodgkin–Huxley type have been used to explore electrical behavior and neurophysiological characteristics in neurons from several regions of the brainstem [30–35] . For example , some well-known phenomena associated with the neuronal electrical behavior such as spike frequency adaptation ( SFA ) , post-stimulation recovery or delayed excitation , and postinhibitory rebound ( PIR ) have been related to specific ionic current mechanisms in medullary neurons [31 , 32] and in some cases these phenomena have been used to classify neurons by their electrophysiological behavior in response to electrical stimuli [30] . However none of these models examined specific electrical behaviors and related ionic mechanisms in individual neurons from chemosensitive regions , neither they have explored their intrinsic properties associated with their response to hypercapnic or acidotic stimuli . The chemosensitive response of these neurons to acidotic stimuli affects respiratory , arousal , emotional and memory circuits . It is therefore important to study the responses of individual neurons to changes in pH and CO2 , and to understand the cellular signaling mechanisms that govern autonomic and respiratory responses to such changes . This will , in turn , contribute to our understanding of the etiology of disorders with respiratory manifestations such as sudden infant death syndrome [12] , sleep apnea [36] , panic attacks [37–40] , or Rett syndrome [41] . Given the potential wide-ranging impact of understanding the cellular mechanisms governing CO2/H+ sensitivity in neurons from chemosensitive regions , it is important to develop a model that faithfully replicates the effect of CO2/H+ on the activity of individual chemoreceptors . In the current study we propose a single cell model framework for a CO2/H+-sensitive neuron that mimics the general electrophysiological discharge patterns of locus coeruleus neurons at rest , as well as in response to electrical and chemical stimuli . Our main goal in this study was to investigate 1 ) intrinsic neuronal activity in absence of all possible external input from electrical and chemical stimuli; 2 ) the effects of excitatory and inhibitory stimuli on neuronal activity; and 3 ) the role that individual signals and chemical stimuli may play in the chemosensitive response .
The model was used to examine the typical firing properties of individual neurons that are common among various chemosensitive regions of the brainstem [42 , 43] . To assess intrinsic activity in absence of all possible input from electrical and chemical stimuli , the stimulus current ( Is ) was initially set to zero and CO2/H+ levels established at normocapnic conditions . Simulations resulted in spontaneous regular pacemaker-like activity characteristic of chemosensory neurons [44–48] . Some other characteristics , such as membrane properties and action potential shapes , were also within the range observed for these chemosensitive neurons at normal and hypercapnic conditions ( Table 1 ) . In particular , simulations were run to predict some of the more important firing properties of individual chemosensitive neurons from the LC and to validate the model with intracellular recordings . As can be seen in Fig 1 the model results in spontaneous action potentials with a regular spiking pattern and with membrane properties ( action potential shape , AHP phase , and frequency ) similar to those observed in LC neurons . To simulate the basic electrochemical behavior of a chemosensitive neuron , we subjected the model to electrical ( from external stimulus current ) and chemical ( changes in CO2/H+ ) stimuli . Neuronal responses were distinguished by specific firing pattern phenomena , namely spike frequency adaptation , postinhibitory rebound , and post-stimulation recovery . To simulate the basic chemosensitive behavior of a locus coeruleus neuron , we subjected the model to different levels of CO2/H+ . As expected , neuronal responses were characterized by increased firing frequencies when related parameters where established to simulate hypercapnic and acidotic conditions in accordance with previous experiments in LC [49] . As illustrated in Fig 5 the predicted changes in frequency resulting from the chemosensitive response in the model are in agreement with intracellular recordings of LC neurons . We next investigated the effects of excitatory and inhibitory chemical stimuli ( i . e . hypercapnic acidosis followed by hypocapnic alkalosis ) on the tonic firing rate ( Fig 6 ) . In contrast with the electrical behavior in response to depolarizing and hyperpolarizing external stimuli , there was no evidence of postinhibitory rebound or post-stimulation recovery , and no depolarizing sag , when the stimulation was based on physiological CO2/H+ changes . Here , the neuronal response to inhibition by hypocapnic acidosis ( 2 . 5% CO2 ) resulted in a hyperpolarization of the membrane potential that prevented the neuron from firing . This hyperpolarizing effect was released by subsequent hypercapnic acidosis ( 15% CO2 ) , which made the neuron fire at higher frequencies and caused a new , elevated tonic firing rate compared with normocapnic conditions ( 5% CO2 ) . There was no evidence for postinhibitory rebound before hypocapnic inhibition , consistent with a lower accumulation of [Ca2+]i and inhibition of the L-type current during hypocapnic conditions . To understand the effect of individual signals , whether electrical or chemical , on the activity of the neuron , we calculated the sensitivity of the firing-rate response to each signal ( Table 2 ) . A sensitivity index ( SI ) was thus defined as the percentage of control firing rate to which the activity increases or decreases in response to a change in each signal ( Δs ) : SI=|FRs−FRc|FRc∙100 ( 1 ) where FRs is the firing rate in response to a specific change , FRc is the firing rate of the neuron in control ( normocapnic ) conditions ( 1 . 43 Hz ) , and Δs is the minimum increase or decrease of the signal necessary to produce an observable change in the firing rate . In other words , this value represents how much the neuron has to be stimulated with a particular signal in order to change its firing rate response from normal to altered activity . That is to say that in the case of chemical signals , a decrease in extracellular pH from 7 . 44 to 7 . 34 , or an increase in the CO2 level from 5% to 7% , is enough to produce an increased firing rate response ( see Table 2 ) . Overall , the effect of applied stimulus current on the increased firing-rate response exceeds the contribution of chemical signals . According to calculated SI values , a small change in applied current ( 0 . 02 nA ) can double the firing frequency , while a small reduction in pH ( 0 . 1 units ) increases the firing rate by less than 42% . Small changes in %CO2 constitute the signal for which the model results are less sensitive , causing an increase in firing rate of about 2% . This effect is schematized in Fig 7 . Here , the firing-rate response increases indefinitely with injected depolarizing current ( Fig 7C ) , but saturates for chemical stimuli ( Fig 7A and 7B ) . Fig 8 shows the effect of external stimuli on the simulated membrane potential of a spontaneously active neuron in response to different chemical stimuli and the firing frequency for a set of conditions ( normocapnia , isohydric hypercapnia , isocapnic acidosis , or hypercapnic acidosis ) at different values of the stimulus current . We observed the combined effect of the applied current and the chemical signals for three different stimulating external currents ( Is ) : depolarizing ( Is < 0 ) , no electrical stimulus ( Is = 0 ) , and hyperpolarizing ( Is > 0 ) . From Fig 8B we noticed that whereas the firing frequency remained unchanged for acidotic conditions ( i . e . IA and HA dark blue bars ) , it increased with excitatory input ( Is = -0 . 6 ) and decreased with inhibitory input ( Is = 0 . 6 ) at isohydric conditions ( i . e . grey and light blue bars ) . In addition , for each of the three acidotic conditions , we calculated the sensitivity of the firing rate at different stimulus currents using Eq 1 ( see Table 3 ) . Bars of the same color in Fig 8C represent the sensitivity of the firing-rate response to the same chemical condition but at different stimulus currents . In general , Fig 8C shows that the sensitivity of the model in each case increases with inhibitory input , as a consequence of an increase in the difference in firing rate between control ( dark grey ) and acidotic/hypercapnic conditions ( blue bars ) , as can be also observed from Fig 8B .
Modulation of spike frequency is a fundamental intrinsic property regulating repetitive firing dynamics in response to constant stimuli . Our results demonstrate that during the stimulation , in response to an external current or an excitatory chemical stimulus , there is an adaptation of the firing rate that regulates the initial activity in response to depolarization ( Figs 1 and 2 ) . Results from previous studies in chemosensitive brainstem regions suggest that such modulation of the action potential frequency is determined by the amplitude of the afterhyperpolarization phase of the action potential , which can be triggered by the inactivation of Ca2+-activated K+ currents [29 , 31] . In agreement with previous studies , our results suggest that the mechanism of decay of the firing frequency responsible for SFA ( Figs 2 and 3 ) and post-stimulation rebound ( Fig 4C ) involves an increase in calcium ion concentration inside the cell during depolarization . This accumulation of Ca2+ strengthens a calcium-activated K+ current ( in this case the SK current ) that transiently diminishes the initial firing rate ( Fig 2D and 2E ) . The same mechanism was observed upon termination of the rebound in Fig 4C , where the level of [Ca2+]i gradually started to decay . In both cases , once resting levels of [Ca2+]i had been restored , spiking resumed at an increased steady-state rate . In addition , we found that the modulation of the firing frequency during electrical stimulation was not only associated with intracellular calcium accumulation and corresponding SK current activation ( Fig 2C and 2D ) , but also with an inhibitory effect of the M-type K+ current ( Fig 2A–2C ) . In fact , there was less adaptation during chemical stimulation ( reflected by a lower SFA value ) where the M-type K+ current played no role . From these results , we propose that the M-type current plays a major role in modulating a high peak frequency during electrical stimulation and that the calcium-activated mechanism responsible for adaptation plays a more important role during chemical stimulation . Previous studies on respiratory neurons suggest that the T-type calcium current is the main current responsible for the initiation and evolution of the postinhibitory rebound that results when the neuron is released from the hyperpolarizing stimulus followed by stimulation with a depolarizing pulse . In agreement with Rybak [32] , we found that the dynamic response of the membrane potential under these conditions is associated with an increase in T-type Ca2+ current , which depolarizes the membrane and promotes accumulation of intracellular calcium . There is also an increase in A-type current that may limit the rebound spiking frequency ( Fig 4B ) . In the same way , T-type current activation is associated with the initial depolarization that causes the rebound of action potentials and modulates its frequency as it decreases to its initial state . During external electrical stimulation we also observed delayed excitation after the rebound ( Fig 4C ) due to a weaker activation of the SK current . In contrast with this response to depolarizing and hyperpolarizing external stimuli , there was no evidence of delayed excitation–not even a depolarizing sag was seen–when stimulation was based only on physiological CO2/H+ changes ( Fig 6 ) . This is in accordance with findings from a model of central CO2 chemosensitivity in Helix aspersa [44] , which postulates that the A-type K+ current is the primary sensor for chemosensitive response to hypercapnic acidosis . However , we found that it is the M-type , not the A-type , K+ current that is most involved initiating the increased firing-rate response when stimulated with external input . In any case , we have no evidence from the present study that the M-type or pH-sensitive A-type potassium currents initiate the chemosensitive response to hypercapnic excitation , although Li & Putnam ( 2013 ) show a major role for A-currents in the hypercapnic response of locus coeruleus neurons [45] . There are two interesting findings from the model . First , the increased firing-rate response to excitatory stimuli seems to saturate at increasing levels of CO2 and/or acidification . This is in accordance with previous observations from the locus coeruleus , where a saturation effect seems to govern the neuron’s chemosensitive behavior [46] . Second , a higher chemosensitivity is obtained with this model when the chemical stimulus involves decreased intracellular and extracellular pH but not an increase in CO2 level alone . It can be seen from the relatively high value of the SI calculated for the increased firing rate in isocapnic acidosis . This observation is also strengthened by the relatively low value of SI that was obtained in isohydric hypercapnia , when the stimulus involved an increase in CO2 but the pH remained at the control level . The fact that extracellular and not intracellular pH plays the major role in chemical signaling ( Fig 7D ) disagrees with a study by Filosa et al . ( 2002 ) , in which the increased firing rate of locus coeruleus neurons in response to acid challenges correlated most with the magnitude and the rate of fall in pHi [47] . This discrepancy can be explained in part because of the indirect limiting effect that pHo has in the model . In other words , pHo has the potential to inhibit the L-type calcium current , thus preventing the activation of the calcium-activated braking pathway and ultimately increasing the firing-rate response to acidotic challenges . Indeed , the indirect effect that chemical signals may exert on the model’s response to hypercapnic acidosis constitutes an important limitation of the model . We assumed an additive effect of signals on membrane protein activity , but it is clear that there is an intricate interaction between activated and inhibited currents . For example , acid inhibition of a K+ channel could depolarize the membrane , which itself may activate another K+ current , thus reducing the initial depolarizing effect of the pH-sensitive channel . This interaction , and many others , makes the precise combined effect of different signals difficult to predict . However , the assumption of separate and additive effects of signals on membrane protein activity is supported by the previous observation that contributions of each current to the overall effect of hypercapnic acidosis are approximately additive [48] . Importantly , the present work confirms that small external electrical signals play a greater role in determining the chemosensitive response to changes in CO2/H+ than previously thought . This implies that a neuron can be erroneously classified experimentally as a chemosensitive or non-chemosensitive cell if external electrical factors ( e . g . input resistance or gap junctions ) are not properly blocked . Furthermore , this observation supports the notion that synaptic transmission is very likely to modulate the responses of chemosensitive neurons [48] , and highlights the importance of considering synaptic transmission when defining and understanding key concepts such as intrinsic chemosensitivity .
The rates of activation and inactivation for voltage-sensitive channels are defined as first-order differential equations of the general form: dxdt=x∞−xτx , x∈{mi , hi} ( 5 ) with steady-state values x∞ x∞=11+e− ( V−Vx ) /kx , ( 6 ) where Vx and kx are half-activation voltage and slope factor , respectively ( Tables 6 and 7 ) . For each activation/inactivation variable x , time constant is defined by: τx={ax+bxe− ( ( V−Vτx ) /kτx ) 2ifx∈{mNa , hNa , hT}ax+bxcosh ( ( V−Vτx ) /kτx ) otherwise ( 7 ) We used the Hodgkin–Huxley formalism as a basis to include the effect of CO2 and H+ on the neuronal response . We implemented neuronal chemosensitivity in this ionic current model by quantifying the total CO2/H+ sensitivity of each channel and considering the excitatory and inhibitory role that hypercapnic and/or acidotic stimuli would have on individual channels due to the combined effect of independent signals . To approximate the total sensitivity of each channel , we assumed separate and additive effects of signals on membrane protein activity [22] . Thus , we expressed CO2/H+ sensitivity as a unitless function ( φi > 0 ) : φi=1−∑swi , sφi , s , sϵ{pHo , pHi , CO2} ( 8 ) where φi , s defines specific functions for pH-dependent inhibition/activation of the channel i , and wi , s is the contribution of the signal , s ( extracellular pH , intracellular pH and/or CO2 ) , to the present level of chemosensitivity . Here , a negative wi , s accounts for channel activation and a positive value for inhibition . For non-chemosensitive currents , the contribution of wi , s is set to zero ( see Table 8 ) . In accordance with [54] , we assumed that pH sensitivity ( due to changes in pHo or pHi ) relies on titrable amino acid residues in the channel protein . In the case of CO2-dependent activation , CO2 molecules are supposed to bind the channels and promote a second open state that enhances its activity [55] . As both situations describe a typical saturation effect , we express the relation between each signal and current inhibition/activation as a Hill equation: φi , s ( s ) =11+ ( s1/2/s ) hs ( 9 ) where s1/2 corresponds either to the pK ( pH level at the midpoint of current inhibition ) or to the percentage of CO2 required to achieve a half-maximal activation ( Table 9 ) . Hill coefficients hs for each curve represent the sensitivity to each stimulus . Parameter values for intracellular and extracellular pH responses ( titration curves ) are shown in Table 10 . CO2/H+-dependent inward rectification has been observed in adult and neonatal rat neurons from various chemosensitive regions . We therefore included an inwardly rectifying current ( IKir ) , with a reversal potential of −44 mV taken from [31]: IKir=g¯Kir ( V−EK−36 ) 1+exp{ ( V−EK+140 ) ZF/RT} ( 14 ) Assuming that there is a non-zero ion flux at rest , as the neuron fires spontaneously without afferent input , we added a small background current stated to be composed of chloride , sodium and potassium . Our reason for including these leak currents is because intracellular pH regulation by acid–base transporters results in altered neuronal excitability and their effect might be brought about by the pH modulation of Na+/H+ and Cl−/HCO3− exchangers [57] . In addition , pHo-sensitive K+ leak currents like the TASK-1 channel are expressed in chemosensitive neurons [58] and seem to be involved in the chemosensitive response of these neurons to hypercapnia . As the resting membrane potential will also be affected by the action of these factors , this change can be modeled by the following approximate formula , which assumes that sodium and chloride ion flux , from pHi-regulating proteins , balances leak K+ flux: IB=∑jIj , leakj∈{Na , K , Cl} ( 15 ) with different leak currents expressed as linear Ij , leak = gj , leak ( V–Ej ) . As the total leak conductance can be estimated as the reciprocal of the neuron’s input resistance we obtain: gK , leak=gtask ( 16 ) gCl , leak=gleak−gK , leak−gNa , leak ( 17 ) gNa , leak= ( Rin ) −1 ( ECl−Vr ) +gK , leak ( ECl−EK ) ECl−ENa ( 18 ) where Vr is the membrane potential that gives an equivalent leak equilibrium potential around −65 mV [36] and Rin is the input resistance ( Table 4 ) . The model was implemented in a MatLab programming environment using a variable-step differential equation solver . The resulting ODE system given by Eqs ( 2 ) , ( 5 ) and ( 12 ) was solved numerically using the stiff variable-step differential equation solver ode15s . For each simulation , the model was run until steady-state conditions , from which data and calculations were then collected .
|
The sensory mechanism by which changes in CO2 and H+ levels are detected in the brain is known as central chemoreception . Altered chemoreception is common to a wide variety of clinical conditions , including sleep apnea , sudden infant death syndrome , hyperventilation , depression , anxiety and asthma . In addition , CO2/H+-sensitive neurons are present in some regions of the brain that have been identified as drug targets for the treatment of anxiety and panic disorders . We are interested in understanding the cellular mechanisms that determine and modulate the behavior of these neurons . We previously investigated possible mechanisms underlying their behavior in rats to elucidate whether they respond to changes in intracellular or extracellular pH , CO2 , or a combination of these stimuli . To study the roles that signals and ion channel targets play in individual neurons we develop mathematical models that simulate their electrochemical behavior and their responses to hypercapnic and/or acidotic stimuli . Nowadays , we are focused on using computational tools to explore the firing pattern of such neurons in response to chemical ( CO2/H+ ) and electrical ( synaptic ) stimulation . Our results reveal significant effects of electrical stimulation on the responses of brainstem neurons and highlight the importance of considering synaptic transmission in experimental studies of chemosensitivity .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Model"
] |
[
"acidosis",
"medicine",
"and",
"health",
"sciences",
"action",
"potentials",
"pathology",
"and",
"laboratory",
"medicine",
"membrane",
"potential",
"brain",
"electrophysiology",
"neuroscience",
"surgical",
"and",
"invasive",
"medical",
"procedures",
"signs",
"and",
"symptoms",
"functional",
"electrical",
"stimulation",
"cellular",
"structures",
"and",
"organelles",
"animal",
"cells",
"locus",
"coeruleus",
"brainstem",
"cell",
"membranes",
"intracellular",
"membranes",
"cellular",
"neuroscience",
"diagnostic",
"medicine",
"cell",
"biology",
"anatomy",
"physiology",
"neurons",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"neurophysiology"
] |
2017
|
Theoretical perspectives on central chemosensitivity: CO2/H+-sensitive neurons in the locus coeruleus
|
Simian immunodeficiency virus ( SIV ) infection leads to AIDS in experimentally infected macaques , whereas natural reservoir hosts exhibit limited disease and pathology . It is , however , unclear how natural hosts can sustain high viral loads , comparable to those observed in the pathogenic model , without developing severe disease . We performed transcriptional profiling on lymph node , blood , and colon samples from African green monkeys ( natural host model ) and Asian pigtailed macaques ( pathogenic model ) to directly compare gene expression patterns during acute pathogenic versus non-pathogenic SIV infection . The majority of gene expression changes that were unique to either model were detected in the lymph nodes at the time of peak viral load . Results suggest a shift toward cellular stress pathways and Th1 profiles during pathogenic infection , with strong and sustained type I and II interferon responses . In contrast , a strong type I interferon response was initially induced during non-pathogenic infection but resolved after peak viral load . The natural host also exhibited controlled Th1 profiles and better preservation of overall cell homeostasis . This study identified gene expression patterns that are specific to disease susceptibility , tissue compartmentalization , and infection duration . These patterns provide a unique view of how host responses differ depending upon lentiviral infection outcome .
Natural reservoir hosts of simian immunodeficiency virus ( SIV ) do not develop AIDS in response to infection and live a normal lifespan . This is in contrast to non-natural hosts , such as Asian pig-tailed macaques ( PTs ) , which , when experimentally infected with SIV , develop AIDS in a similar manner to HIV-infected humans [1] , [2] . Pathogenic SIV infection is characterized by high viral replication , loss of CD4+ T cells , high immune activation , T cell apoptosis , and ultimately , AIDS [3]–[5] . SIV-infected natural hosts , such as African green monkeys ( AGMs ) , retain stable CD4+ T cell counts in peripheral blood even in the presence of viral loads ( VLs ) as high as those observed during pathogenic infection [6]–[8] . Given this dichotomy , insight into the mechanisms of HIV pathogenesis might be gained by comparing the transcriptional response of natural and non-natural hosts to SIV infection . High levels of viral replication and the loss of CD4+ T cells correlate with faster disease progression in pathogenic SIV/HIV infection [9]–[11] . Diseased hosts also exhibit substantial loss of uninfected CD4+ and CD8+ T cells as a result of activation-induced cell death [12] , [13] . It is thought that adaptive ( T and B cell ) antiviral immune responses may serve to control viral replication and spread [14] , [15] . Yet , high levels of CD8+ T cell activation are also predictive of disease progression [16]–[18] . Cytotoxic T cell ( CTL ) responses are associated with increased induction of inflammatory cytokines , which in turn also correlate with pathogenesis . Conversely , control of immune activation in non-pathogenic infection may relate to anti-inflammatory cytokine profiles ( reviewed in [19] ) . Dysregulation of the pro-inflammatory response has been reported in pathogenic lentiviral infections , often as a result of cytokine induction by viral proteins [20] , [21] . Secretion of CC-chemokines has been associated with an HIV-induced proinflammatory cytokine profile [22] , [23] , and molecules such as interferon gamma ( IFNγ ) are up-regulated in lymphoid tissues of SIV-infected rhesus macaques [24] , [25] . Although IFN responses are critical components of the innate immune response to viral pathogens , type 1 interferon expression is elevated in the acute phase of pathogenic SIV infection of non-human primates ( NHPs ) and persists in lymphoid tissue but fails to control viral replication [25] , [26] . Additionally , a strong innate response to SIV often includes induction of genes recognizing pathogen-associated molecular patterns ( PAMPs ) . It has been proposed that increased expression of inflammatory mediators may be a result of general immune activation caused by Toll-like receptor ( TLR ) induction by bacterial products escaping the gut [27] . Such bacterial translocation may be a result of epithelial damage from a sustained pro-inflammatory environment in the gut mucosa [28] . Increased levels of pro-inflammatory cytokines in HIV and pathogenic SIV infections have also been associated with T cell apoptosis via tumor necrosis factor ( TNF ) /FAS death receptor signaling pathways [29]–[31] . Lower levels of immune activation have been correlated with decreased FAS-induced cell death in HIV patients undergoing retroviral therapy [32] , and levels of immune activation and T cell apoptosis are low in natural SIV hosts with non-pathogenic infections [8] , [33] , [34] . Conversely , bystander activation induced cell death is associated with higher levels of T cell activation [12] , [13] , [35] . Excessive early T cell apoptosis has been examined in pathogenic SIV models [36] , [37] and has been proposed as a mechanism for pathogenesis . A positive correlation between primary acute-phase apoptotic levels in peripheral lymphoid tissue and disease progression in rhesus macaques has been demonstrated [36] . In addition to sustained immune activation , T cell apoptosis in pathogenic models may be linked to dysregulation of the cell cycle , as perturbations have been observed in lymph nodes of SIV-infected macaques but not in naturally-infected sooty mangabeys [38] . The mechanisms by which inflammatory responses , T cell activation , and apoptosis are induced , sustained , or suppressed differentially between pathogenic and non-pathogenic infections remain unclear . To delineate such mechanisms , we performed a longitudinal genomic analysis comparing the transcriptional responses of natural ( AGM ) and non-natural ( PT ) hosts in the setting of acute SIV infection with the same primary isolate ( SIVagm . sab92018 ) . For the microarray experiments , each animal was its own genetically matched control for assessing gene expression changes , with multiple animals for each experimental condition , enabling us to perform statistical tests to identify those gene expression changes that distinguish pathogenic from non-pathogenic SIV infection . Our observations revealed that SIV induced host-specific global gene expression patterns in lymph node ( LN ) , blood , and colon . In more than one respect , this study provides a unique view of lentiviral infection in vivo . First , we now have a global perspective of the host transcriptional response to SIV in pathogenic and non-pathogenic models . Second , several tissues that are known to be affected by lentiviruses were examined at multiple time points , thus providing significant insight into the dynamics of the acute phase of infection . Our analysis of these tissues separately and over time has revealed new insights into the kinetics and compartmentalization of interactions between virus and host .
A quantitative depiction of differential gene expression is shown in Figure 2A and 2B . For this and all ensuing analysis , data points for four animals per species/tissue/timepoint were used ( excluding AGM-3 and PT-3 at day 45+ in LNs , and PT-2 in colon at day 10 , due to sampling errors ) . Overall , the greatest differential regulation of genes occurred in the LNs of infected animals , and the total number of differentially expressed genes was comparable between AGMs and PTs . The number of differentially regulated genes in the peripheral blood was relatively consistent between day 10 and 45+ , though more changes overall were observed in PTs as compared to AGMs . In the colon of both species , the number of differentially expressed genes was greater at day 45+ ( a time when viral set point had been established ) than at day 10 ( peak VL ) . To identify genes whose expression was statistically different between the two species ( and which therefore might be related to disease outcome ) , we performed error-weighted statistical tests ( one-way ANOVA ) on data collected from all animals at each time point , directly comparing AGM to PT for each tissue . Figure 2C is a summary of the total number of genes that were significantly different by species . Despite having comparable numbers of differentially regulated genes at day 10 in LNs , there were over 600 genes expressed at significantly different levels in the AGMs vs . PTs at this time . In all tissues , it was evident that the most gene expression changes distinguishing species were found at viral peak ( day 10 ) , with fewer distinguishing expression profiles at day 45+ . In addition to comparing the species at each time-point by species ANOVA , we examined kinetics by comparisons of total gene expression changes between peak VL ( day 10 ) and viral set-points ( day 45+ ) in each species by time ANOVA ( Figure 2D ) . Significantly more time-dependent changes occurred in the colon of both species than in blood or LN . These results suggest that events which occur during the very acute phase of infection play a key role in disease outcome . As most SIV-specific gene expression regulation was found in lymph nodes , we focused much of our subsequent analysis on this tissue . Based on one-way species ANOVA comparing PTs to AGMs , we identified 618 genes whose expression levels at day 10 were significantly different between the two species ( Figure 2C ) . Gene ontology analysis revealed that the majority of these genes were associated with either immune responses or cell death ( Figure 3A ) , with redundancies in the categories of hematological system development , cell movement , and inflammatory disease . Because the top categories are very broad , genes were further subdivided based on available annotations for additional insight into possible differences in cell processes . Genes associated with cell death ( shown in Figure 3B and Table S1 ) included those mediating death receptor signaling , apoptosis , cell cycle arrest and progression , caspase activation , DNA damage response , or oxidative stress . Although there were numerous genes associated with death receptor signaling , there did not appear to be a species bias towards pro- or anti- FAS-mediated apoptosis . Genes more highly induced in AGMs included UNC5B and STK11 ( known to play a role in FAS-mediated apoptosis and associated with p53 signaling ) and TRADD and IER3 , which are positive or negative regulators of the FAS pathway , respectively . Genes that were more highly induced in PTs have functions more pertinent to the negative regulation of death receptor signaling ( e . g . , ASAH1 , BTK , and CFLAR ) and of apoptosis ( e . g . , IDE , CAMK2D , and LGALS3 ) . Genes associated with oxidative stress and DNA damage ( e . g . , SOD2 , DLST , GLRX , RRM2 , YWHAE , and RAD51L3 ) also showed higher levels of induction in PTs . Collectively , these patterns are consistent with the possibility that oxidative stress is more prominent in pathogenic infection , prompting the induction of cellular genes that serve a protective role against oxidative stress , DNA damage , and apoptosis . Cell death genes related to the cell cycle were also differentially regulated between species . In general , a greater proportion of these genes were induced in AGMs ( e . g . , AXL , CDN2D , and MYCN ) , including those known to be responsible for both progression and arrest of the cell cycle . Immune response genes that significantly distinguished the two species at peak VL ( day 10 ) in lymph nodes consisted mainly of those associated with T and B cell signaling and with chemotaxis of immune cells ( Figure 3C , Table S2 ) . Compared to AGMs , significantly more genes were highly induced in PTs and they exhibited a trend towards Th1 differentiation and proliferation , cytotoxic T cell activity , and IFNγ signalling . These genes included inflammatory mediators ( e . g . , CXCL9 and 10 , CCL3 , CTSC , and CCR2 ) as well as genes necessary for antigen presentation or CTL responses ( e . g . , CD86 , CD84 , CD8a , GZMA , GZMB , CLEC4A and KLRC1 ) . Although not identified by the ANOVA , PTs displayed a trend towards higher expression of additional genes for antigen-presentation ( e . g . , CCL17 , CD80 , GBP3 , and ICOS ) , dendritic cell recruitment/maturation ( e . g . , ADAMDEC1 , CCR7 , ISGF6 , and LAMP3 ) , and CTL responses ( e . g . , CD2 , CCL20 , CLEC7A , and SRGN ) . These patterns are consistent with other studies showing that pathogenic infections can be associated with robust Th1 responses [34] , [42] , [43] . In PTs , significantly higher expression of T cell chemokine genes from the day 10 species ANOVA was consistent with a generally higher level of immune activation in this species , as evidenced by the presence of significantly more CD8+ T cell activation in PT LNs ( by Ki67 levels ) at both day 10 and day 45+ [39] . High induction of pattern-recognition receptors ( e . g . , CLEC7A , TLR2 , and TL8 ) in PTs also suggest a strong innate immune response . Many genes regulating neutrophil chemotaxis and degranulation ( e . g . , LY75 , LYZ , FCGR3B , CSF2BA , CEBPA , and B4GALT1 ) were only induced in PTs . A greater number of genes associated with calcium signaling was also induced in PTs , possibly reflecting the use of calcium-signaling pathways by broadly induced chemokines . Interestingly , the HIV/SIV co-receptor CCR5 was induced in all animals but at significantly higher levels in PTs than AGMs , in support of previous observations [44] and with a proposed model of restricted CCR5 expression in natural SIV hosts [19] . From the above LN d10 ANOVA , a subset of 137 genes showed increased expression only in PTs but decreased expression or no change in AGMs ( Table S3 ) . These genes encode mediators of inflammation , apoptosis , proliferation , and IFNγ signaling . Gene ontology analysis of this subset identified a significant network linked to IFNγ signaling and included genes that positively regulate or are regulated by IFN-γ ( e . g . , IL12RB2 , RARRES1 , GPNMB and NUP98 ) ( Figure 4A ) . Others ( e . g . , CALU , HNRNPC , LMAN1 , and PDCD6 ) are linked to calcium signaling , consistent with the data shown in Figure 3C . As genes in this network were down-regulated in AGMs ( Figure 4B ) , these observations underscore the premise that T cell proliferation and strong Th1 responses are more specific to pathogenic than nonpathogenic SIV infection and , reciprocally , suggest an important role for the regulation of proliferation and inflammation in early stages of non-pathogenic SIV infection . In contrast to PTs , highly-expressed genes in AGMs ( day 10 species ANOVA , Figure 3 ) were associated with the control of an inflammatory/Th1 response and the development of hematopoietic progenitors . Notably , the AGMs showed up-regulation of the cytokine IL10 , a potent anti-inflammatory mediator and negative regulator of Th1 cytokine production , as well as NLRP3 , which is associated with prevention of inappropriate inflammatory responses [45] . This suggests a more active control of the inflammatory response in AGMs . CCL25 , a B-cell chemokine and promoter of IgA production , was also induced at higher levels in AGMs . Genes associated with mast cell accumulation and degranulation included COL17A1 , CCL2 , and LTC4S [46] . Many additional genes induced in AGMs , or down-regulated in PTs , were associated with eosinophil migration or myeloid chemotaxis and growth ( e . g . , CCL2 , IL10 , LTC4S , CCL11 , MADCAM1 , and PLA2G6 ) , or the development of hematopoietic progenitors ( e . g . , THY1/CD90 , CD33 , and CD34 ) . Genes associated with the proliferation or differentiation of bone marrow cells ( e . g . , RARA ) and of gamma delta T cells ( e . g . , JAG2 ) were also specifically induced in AGMs . Deficiencies in these functions have already been associated with disruption of homeostasis in pathogenic HIV/SIV infection [47] , [48] . These gene expression patterns are consistent with the possibility that the AGM may be better poised than the PT to sustain repopulation of the peripheral hematolymphoid system after SIV infection [4] . A key to uncovering AIDS pathogenesis could lie in discerning what is unique to the response of natural hosts to SIV infection . From the LN d10 ANOVA , 108 genes were identified as significantly induced only in AGMs ( Table S4 ) . The functions of these genes may serve to protect against disease progression in the natural host . In addition to those described above , many genes were associated with leukocyte movement and adhesion ( e . g . , DEFA3 and KCNA3 ) , and cell survival and growth ( e . g . , DNASE1L3 , AXL , and DUSP5 ) . A significant network linking functions to epidermal growth factor was also identified by Ingenuity Pathway Analysis ( not shown ) . The network genes encoded sulfotransferases for glucosamine signaling ( e . g . , HS3ST2 and HS3T3A1 ) , actin-binding proteins ( e . g . , FSCN2 and CDC42EP1 ) , and proteins required for cytoskeleton and extracellular matrix structure ( e . g . , EMILIN1 , TUBA1A , and COL17A1 ) . These patterns indicate that AGMs may be able to better maintain basic cellular functions and organ structure , consistent with reports that lymph node architecture is better preserved during non-pathogenic infections [49] . We similarly examined the kinetics of gene expression changes within each species by using ANOVA to compare expression data by day post-infection , identifying 297 genes whose expression significantly changed in PT lymph nodes between day 10 and day 45+ ( Figure S2A ) . Many genes induced at day 10 were linked to apoptosis and oxidative stress , with increased expression of DNA damage-inducible genes ( e . g . , SOD2 , YWAE , GZMB , GZMH , BAK1 , and PRF1 ) , some of which are specifically associated with apoptosis of T lymphocytes . This supports the day-10 species ANOVA , showing early and high expression of genes associated with stress responses in pathogenic infection and confirms that these responses are specific to viral peak and abate in LNs thereafter . About one third of genes identified by the PT time-dependent ANOVA showed either no change or decreased expression at peak VL ( day 10 ) , followed by strong induction at day 45+ . In general , these genes belonged to the functional categories of cellular assembly and tissue development . Of these genes ( blue bar in Figure S2A ) , 41 showed high expression in AGMs as early as day 10 , which persisted to day 45+ ( Figure S2B ) . The expression of these genes may contribute to a successful antiviral response or to the maintenance of proper cell homeostasis in non-pathogenic hosts , but they are induced too late to be effective during pathogenic infection . Gene ontology analysis showed that the genes induced earlier in AGMs are necessary for growth , differentiation , and structure ( e . g . , RYR2 , PCDHGA7 , MYLPF , EEF2K , and CAMK2N2 ) . The lymph node AGM time ANOVA identified 170 genes whose expression changed significantly between day 10 and day 45+ post-infection , less than 30 of which were common with the LN PT time ANOVA ( not shown ) . In contrast to what was observed in PT , most of these genes were highly induced in AGMs at day 10 but decreased by day 45+ , and many encoded inflammatory mediators specifically associated with the interferon response ( e . g . , IFIT2 , IFIT3 , STAT1 , IFITM2 , IFITM3 , CXCL10 , and IFIH1 ) . During this same time frame , high expression of these genes was sustained in PTs . These observations suggest that attenuation of interferon signaling at the transcriptional level may be an important characteristic of non-pathogenic SIV infection . Collectively , analysis of LN revealed that most differential gene expression patterns in PTs vs . AGMs occurred at peak VL ( day 10 ) and related to immune responses and cell death . During early pathogenic infection , species and time ANOVAs revealed high induction of genes associated with acute stress responses , DNA damage , and Th1 signaling . In AGMs , however , gene expression data suggested control of Th1 responses and maintenance of general cell homeostasis processes , including immune cell homeostasis . Though expression patterns in LNs of both species were more similar by day 45+ , there was a clear tendency for sustained inflammatory responses during pathogenic infection , while IFN signaling was significantly attenuated by day 45+ during non-pathogenic infection . The time- and species-specific gene expression changes in peripheral blood were very similar to those observed in lymph nodes . Time ANOVAs in peripheral blood , comparing AGM expression levels at day 10 to day 45+ , further confirmed evidence for an attenuated inflammatory response during non-pathogenic infection . Of 109 genes identified by AGM time ANOVA , one quarter were less induced or down-regulated by day 45+ ( Table S5 ) and most of these were involved in IFN-signaling ( e . g . , MX1 , MX2 , OAS1 , OAS2 , CXCl10 , IFI35 , IFI27 , and ISG20 ) . Expression patterns identified by blood day 45+ species ANOVA ( Figure S3 ) suggested the strong and sustained induction of genes associated with an inflammatory response only in PTs ( Figure S3 , blue bar; Table S6 ) , many of which can be regulated by IFN ( e . g . , MX1 , MX2 , OAS1 , and IFI27 ) . One unique pattern identified by blood day-10 species ANOVA that was not observed in LNs , was a marked decrease , only in PTs , in expression levels of most genes at peak viral load ( Table S7 ) . Overall , this broad “suppressive” trend in PTs was seen in genes associated with the regulation of basic cellular homeostasis [50] , including functions such as apoptosis regulation and the development of cells involved in the immune response . The decreased expression of these genes may in part be due to depletion of circulating CD4+ T cells in PTs at 10 days post-infection , either because of increased destruction , decreased production , or movement into tissues . Given the low proportion of CD4+ T cells in blood , however , it is unlikely that such depletion is responsible for all of these dramatic gene expression changes In contrast to LN and blood , we found that in the colon there were more genes that were significantly differentially expressed between time post-infection ( day 10 vs . day 45+ ) than between species . The AGM time ANOVA identified 658 genes whose expression distinguished day 10 from day 45+ . Most genes were unchanged or decreased at day 10 but were highly induced by day 45+ , very similar to the PT time ANOVA pattern . In fact , nearly 300 genes from the colon AGM and PT time ANOVAs were in common ( Figure S4 ) . Functions most highly represented by these genes included tissue development , cell movement , growth and proliferation , death , and cell-cell signaling . These included genes encoding scaffolding proteins , muscle filaments , extracellular matrix components , and cell cycle/apoptosis regulators . Ingenuity Pathway Analysis was used to identify interactions between nearly 30 genes associated with TGFβ signaling and fibronectin ( data not shown ) . These observations suggest that the expression of genes related to cell growth , proliferation , and apoptosis remain at baseline at peak VL and do not increase until late in acute infection . Collectively , the time ANOVA data suggest that the majority of SIV-induced transcriptional changes in the colon occur after peak VL in both species . Since most of the genes induced at viral set point were emblematic of broad tissue reconstruction , there appears to be a common response in the colon of both species that occurs in temporal relationship to the CD4+ T cell depletion found in each [40] , [41] , 51 . Of the nearly 700 genes identified by AGM time ANOVA in the colon , only a subset of 37 genes was induced at day 10 at high levels and significantly lower by day 45+ . These genes are shown in Figure 5A , along with the corresponding expression in PTs ( Figure 5B ) . Most are inflammatory mediators with many involved in type I interferon signaling , which is consistent with the patterns identified in the lymph nodes . Though expression levels consistently decreased in all AGMs , but not in PTs , from day 10 to day 45+ , the expression levels of many of these inflammatory mediators were actually higher in AGMs than PTs at day 10 . This suggests that AGMs not only show attenuation of an inflammatory response after acute infection , but that a more robust early inflammatory response in the colon may control viral replication and subsequent tissue damage . These results are consistent with a previous study that found an association between a stronger early IFN response and slower disease progression [52] . Perhaps the most surprising finding was that the colon showed the smallest number of genes statistically distinguishing the two species at peak VL . This and the time ANOVAs suggest that the host transcriptional response to SIV infection in colon is similar in both species , and is consistent with the fact that CD4+ T cells are lost in each . Similar to what was observed in peripheral blood and LN , more gene expression changes differed during pathogenic vs . non-pathogenic infection in the colon at day 10 than at day 45+ . Of 165 genes that distinguished PTs from AGMs in the day 10 colon ANOVA , 74 were highly induced only in the PTs ( Table S8 ) Pathway analysis revealed that the top pathways/networks represented by these up-regulated genes were associated with the acute phase response . Figure 6 shows the most significant network linking these genes and centers to the transcriptional regulator , NFκB . Most genes in the network linked important acute phase response regulators such as p38MAPK , IL1 , SOD2 , and CEBPB , which in turn positively influence the induction of additional inflammatory mediators . The Toll-like receptor pathway was also prominent , with increased expression of TLR2 and CD14 , possibly reflecting responses to bacterial products . This supports a report that TLR induction by bacterial products escaping the gut could trigger high expression of inflammatory mediators [27] . If so , it is evident that these pathways were not prominent in AGMs ( Figure 6B ) at this time , so it would seem unlikely that they were involved in the depletion of mucosal CD4+ T cells observed in this species . The end of acute infection ( day 45+ ) was associated with the expression of 106 genes in the colon that distinguished pathogenic and non-pathogenic infections ( Figure 2C and Table S9 ) . Here , tissue development and cell movement were identified as the top functional categories , followed by immune response , cell death , and proliferation . It has been reported that cell adhesion may play a role in maintaining the integrity of the mucosal barrier during pathogenic SIV infection [53] while dysregulation of growth factors and epithelial repair may compromise barrier integrity [28] , [54] , [55] . Therefore , it was interesting that most of the genes in the first category associated with membrane development , extracellular matrix , and cell adhesion ( RPL37A , CEACAM1 , CEACAM7 , COL4A2 , ALOX12 , DSC2 , WNT3 , and MAL ) were either highly induced only in AGMs or showed significantly decreased expression specifically in PTs . Notably , up-regulation of structural genes in this species ANOVA was negligible in PTs . Although both species appeared to initiate programs involving structural integrity genes by day 45+ ( as seen in the time ANOVAs , Figure S4 ) , subtle changes specific to AGMs may be part of a mechanism associated with the absence of disease progression . Evidence for bacterial translocation was again observed at day 45+ by induction of TLR 4 expression in PTs . Also , consistent with the observations in LN and peripheral blood , genes encoding inflammatory mediators were more highly induced at day 45+ in PT than in AGM colon . In sum , most gene expression patterns in the colon were common to both the PT and the AGM , and between the peak of viremia ( day 10 ) and day 45+ post-infection , animals of each species revealed significant increases in the expression of genes responsible for cell growth and development . Species-specific differences revealed a strong acute phase response in PTs at peak VL ( day 10 ) while induction of structural genes was most prominent in AGMs at viral set point ( day 45+ ) . Additionally , though the inflammatory response was high and sustained in PTs , there was a higher IFN response in AGM colon at day 10 , which was significantly attenuated by day 45+ post-infection .
This study used two NHP species to comprehensively analyze the host response associated with non-pathogenic and pathogenic SIV infection during the acute phase . Our primary findings are summarized in Table 1 . Most species-specific gene expression differences occurred in LN at the time of peak VL , and the majority of time-dependent changes for a given species were in the colon between day 10 and day 45+ . Overall , gene expression patterns in PTs ( pathogenic infection ) suggest a strong and sustained immune activation , including CTL activity and Th1 profiles . In contrast , the intensity and kinetics of the inflammatory response in AGMs ( non-pathogenic infection ) suggest strong interferon signaling at day 10 , which was then attenuated by day 45+ . The AGMs also exhibited fewer gene expression changes related to cell homeostasis ( defined here as genes associated with basic cellular functions , such as metabolism , growth , and continued hematopoietic regeneration ) . Although our study is the first to use global gene expression profiling to compare the host response during pathogenic and non-pathogenic SIV infection , a number of previous studies have also used NHP models in which infection outcome differs in an attempt to elucidate mechanisms associated with disease progression . Several themes have begun to emerge . For example , Cumont et al [56] also observed a link between dysregulation of immune activation and SIV pathogenesis , finding high levels of immune activation in LN T cell zones of SIV-infected macaques , but low levels of T cell activation and high levels of B cell activation in AGMs . This is consistent with our gene expression data . Early induction of T cell signaling genes during pathogenic infection supported a trend for Th1 cell differentiation and proliferation , cytotoxic T cell activity , and extensive IFNγ signaling . Increased IFNγ expression , an indicator of the Th1 response , has been detected in lymphoid tissue of macaques [24] , [43] , [57] , and dysregulation of Th1 agonists may contribute to pathogenesis [58] . In contrast , prominent Th2 cytokine profiles have been associated with SIV infection during non-pathogenic infection [59] . Levels of CD8+ T cell activation have been directly correlated with accelerated AIDS progression [16]–[18] . In our study , Ki67 levels were statistically higher in PT CD8+ T cells than in AGMs , and these levels correlated well with gene expression profiles indicative of a more pronounced CTL response in PTs ( e . g . , up-regulation of IFNγ , immunoproteasomes , and granzymes ) . In contrast , CD8+ T cell Ki67 expression and transcriptional profiles in the AGM indicated lower levels of T cell activation and less prominent Th1 patterns [34] , [60]–[62] . Although similar observations have been attributed to a low immune response of AGMs to core SIVagm protein [63] , we found that SIV-specific antigen responses were similar in both species [39] . This supports studies finding comparable SIV-specific cellular responses in both pathogenic and non-pathogenic models of SIV infection [64]–[67] . Therefore , the increased expression of genes associated with CTLs during pathogenic infection may be a result of improperly-regulated activation of bystander immune cells [34] . Loss of T regulatory cells ( Tregs ) [68] and cell cycle dysregulation [69] have been associated with pathogenic SIV-infection . For example , microarray analyses of macaque LN and jejunal biopsies [53] , [57] have shown dysregulation of cell cycle and increased induction of cytolytic activity as part of a localized host response . We also observed the acute loss of Tregs during pathogenic infection and cell cycle dysregulation in the transcriptional analysis of LNs at peak VL . Additionally , high expression levels of FOXO3 , which negatively regulates T cell activation and autoinflammation [70] , were unique to AGM blood ( data not shown ) . Taken together , this suggests that failure to control T cell activation/proliferation may contribute to poor outcome [56] , [62] . Numerous studies have demonstrated that varied levels of apoptosis , or the induction of apoptosis-related genes , are linked to species-specific differences in disease susceptibility [34] , [56] , [71] . Global gene expression patterns for apoptotic mediators formed a complex picture in our analysis . Cell death genes made up the most significant functional group that distinguished PT from AGMs at day 10 in LN . However , the number of induced genes associated with death receptor signaling were nearly equal in AGMs and PTs , while the number of genes associated with protection from apoptosis were higher in PTs . Higher levels of annexin staining [39] , loss of CD4+ T cell counts , and the induction of oxidative stress and DNA damage-inducible genes in PT lymph node and blood suggest an increased apoptotic response in PTs . The anti-apoptotic gene expression pattern in the LN species ANOVA may point to a negative feedback mechanism in PTs that attempts to protect the host from excessive cell death . However , because apoptosis is not generally a transcriptional event , it is difficult to interpret such gene expression changes and to directly correlate them with apoptotic events , except as they relate to increased expression of survival factors . The increased expression of genes associated with cell damage and death in PTs may be due in part to high and sustained levels of inflammation , and several studies have associated such a response with SIV/HIV pathogenicity [24] , [28] , [53] , [58] . However , it has been unclear whether there is an immediate anti-inflammatory environment in AGMs , or whether these animals are even capable of mounting a Type I IFN response to SIV . Our results demonstrate that acute SIV infection does indeed activate Type 1 IFN signaling in AGMs . For example , a group of eight IFNα genes were induced up to 10 fold over pre-infection levels in AGM LN at peak VL . We have previously shown that IFN can block SIV replication at early stages of infection in vitro [72] , and our current analysis indicates that early Type I and II IFN responses may help AGMs to limit SIV pathogenesis . Thus , AGMs and PTs both appeared to mount an early antiviral response , although the response of AGMs was stronger , earlier , and perhaps more effective . By day 45 , the IFN response was significantly attenuated in AGMs , but less so in PTs . It is possible that IFN signaling was actively suppressed in AGMs shortly after peak VL , thereby limiting the induction of inflammatory mediators that would otherwise hasten disease progression . Alternatively , the lower induction of IFN genes may be a consequence of lower viral loads in the LN of these animals , although the same pattern of attenuation was observed in colon , where viral loads were similar to those observed in PTs . The strong and early IFN signaling in AGM colon is consistent with reports suggesting that a strong innate immune response in mucosal tissue is a mechanism for limiting SIV pathogenesis [52] , [53] , [73] . However , it is also notable that many of the time-dependent gene expression changes in the colon were common to both species and were associated with the regulation of structural genes . This finding indicates that normal cell homeostasis was disrupted in the context of both pathogenic and non-pathogenic infection at the primary site of SIV infection [74] . Although the initial inflammatory environment could lead to damage of the gut mucosa in both PTs and AGMs [27] , the early attenuation of the response in AGMs may help to reduce such damage . In contrast , the loss of mucosal integrity in PTs may contribute to pathogenesis [28] , [75] , [76] . In LN , genes encoding growth or hematopoietic factors and gamma-delta T cell differentiation were significantly more induced in AGMs than in PTs . Genes encoding proteins implicated in B cell homing and differentiation , as well as the development and chemotaxis of mast cells and eosinophils , were also induced in AGMs . Although this could be due to changes in lymphoid cell distribution , it is also consistent with the hypothesis that certain cell subpopulations in the AGM immune system are better adapted to control SIV infection than their counterparts in the PT . Superior immune function may also be linked to the unique induction of cell survival and structural genes observed in AGM LN , since the destruction of LN architecture is associated with rapid disease progression in SIV-infected macaques [49] . Although we did not observe LN destruction at the relatively early time points studied in this experiment , immunological data showed that PTs had higher levels of immune activation and apoptosis in lymphoid tissues than did AGMs . AGM-specific induction of structural genes during early acute infection may also lead to better preservation of LN architecture at later stages of infection . Therefore , our gene expression data suggest that AGMs may be able to maintain overall immune homeostasis through LN preservation , chemotaxis of certain immune cell subsets , and the general ability to maintain normal cellular processes despite continued viral replication . Global gene expression profiling is increasingly being used to study lentiviral infection of cell lines , NHPs , and human patients . Similar to the findings reported here , many of these studies have also observed differences in the regulation of genes related to apoptosis and the immune response , including those associated with IFN signaling , disruption of immune cell homeostasis , and the preservation of LN architecture [47] , [57] , [77]–[81] . Most intriguing are similarities in the expression patterns we observed in SIV-infected AGMs to those identified in lymphoid tissue of HIV patients on HAART [82] . In both cases , genes associated with cytotoxic T cells showed a decrease in expression compared to that observed in SIV-infected PTs or untreated HIV-infected humans . In contrast , signs of immune activation are attenuated in patients on HAART , as we observed in AGMs . Genes that contribute to extracellular matrix structure and that support hematopoietic cell growth are also induced in patients on HAART . These striking similarities suggest that mechanisms independent of viral suppression—possibly the quality of immune activation or resolution—are different in the context of pathogenic vs . non-pathogenic infection . These findings also underscore the relevance and reproducibility of genomic SIV studies for elucidating mechanisms of AIDS pathogenesis . In summary , our findings emphasize the importance of tissue compartmentalization and kinetics in SIV pathogenesis as well as the unifying pattern of IFN induction and attenuation in the natural host . Because the greatest number of gene expression changes distinguishing pathogenic from non-pathogenic infection occurred in lymph node at peak viral load , we propose that this may be a promising site and time point for further study . To our knowledge , this is the first time that such a comprehensive data set—derived using the macaque genome to directly compare gene expression profiles during acute stages of pathogenic and non-pathogenic SIV infection—has been made available to the public .
The original SIVagm . sab92018 primary isolate [6] was kindly provided by Dr . O . Diop at the Pasteur Institute in Dakar , Senegal and is CCR5- and CXCR4-duo-tropic [83] . Our viral stock was obtained by infecting one SIV negative C . sabaeus ( Caribbean origin from the colony at the Center for Primate Neuroethology and Neuropsychiatric Institute , University of California , Los Angeles ) with 300 50% tissue culture infectious doses ( TCID50 ) of SIVagm . Sab92018 ( Dakar ) , and by collecting plasma at day +10 following IV inoculation . The virus stock ( pure plasma ) was titrated on SupT1 cells at a titer of 1250 TCID50/ml , corresponding to 1 . 06×109 RNA copies/ml ( as measured by qRT-PCR; see below ) and 7 ng/ml p27 ( as measured by p27 ELISA , Zeptometrix , Buffalo , NY ) as described in 2 and in Methods . Ethics Statement: All animal and in vitro procedures were performed using standard protocols and according to guidelines approved by the University of Washington Environmental Health and Safety Committee , the Occupational Health Administration , the Primate Center Research Review Committee , and the Institutional Animal Care and Use Committee . The four adult male AGMs ( 10 years old , weighing 6 . 0 kg , interquartile range 5 . 8–6 . 2 ) and four adult male PTs ( 11 years old , weighing 16 kg , IQR 15 . 3–17 . 0 ) included in this study were housed at the Washington National Primate Research Center ( WaNPRC ) . At day 0 , animals were inoculated intravenously with 600 50% tissue culture infectious doses ( TCID50 ) of SIVagmSab92018 and followed until necropsy at day 45 or day 49 ( subsequently noted as day 45+ ) . Blood samples were collected at days −14 , 10 , and 45+ . At days −14 , 10 , and 45+ , biopsies were obtained from the colon and lymph nodes ( inguinal ) . For microarray and VLs on tissues stored in RNAlater ( colon and lymph node biopsies ) or in OCT ( lymph node biopsies ) , tissues were homogenized using a Polytron PT2100 tissue grinder ( Kinematica , Switzerland ) and RNA was extracted using TRIZOL Reagent according to manufacturer's instructions ( Invitrogen ) . Whole blood was collected into PAXgene tubes for RNA extraction with the PAXgene RNA blood kit ( Qiagen , Valencia , CA ) . RNA concentrations were quantified on an ND-1000 UV-Vis spectophotometer ( Nanodrop , Wilmington , DE ) and controlled for integrity and purity on a capillary electrophoresis system ( Agilent 2100 Bioanalyzer; Agilent Technologies , Santa Clara , CA ) and processed for microarray as described ( Kash et al . , 2006a; Kash et al . , 2006b; Kobasa et al . , 2007 ) RNA VLs in plasma and tissues were measured by real-time quantitative PCR ( ABI PRISM 7700 , Applied Biosystem Foster City , CA ) in a one step RT-PCR reaction ( HoTaq One Step-RT PCR Mix; Molecular Cloning Laboratories , South San Francisco , CA ) , using primers and probes as previously described ( Diop et al . , 2000 ) . Gene expression profiling on whole blood , colon , and LN biopsies at days 10 and 45+ was compared to the baseline time point ( day −14 ) for each tissue and each animal . Probe labeling was performed using Low RNA Input Fluorescent Linear Amplification kit ( Agilent Technologies , Santa Clara , CA ) . Slide hybridizations were performed with rhesus macaque ( Macaca mulatta ) oligonucleotide microarrays containing 18 , 000 rhesus probes ( Agilent ) , representing over 17 , 000 unique rhesus genes . The platform used here was based on the 3′-UTR sequences of rhesus transcripts , as identified in the 6×-coverage rhesus genome ( MMUL1 . 1 ) . Based on available sequences , the probes of the rhesus array have 98% sequence identity to corresponding sequences in M . nemestrina or M . fascicularis [84] . Extending this to C . sabaeus , we estimate this to also have better than 97% identity to M . mulatta . Each microarray experiment was done with two technical replicates by reversing dye hybridization for experimental and reference samples . Slides were scanned with an Agilent DNA microarray scanner , and image data were processed using Agilent Feature Extractor Software ( Agilent Technologies , Palo Alto , CA ) , which also performed error modeling . All data were subsequently uploaded into Rosetta Resolver 7 . 0 ( Rosetta Biosoftware , Kirkland , WA ) and Spotfire Decision Suite 8 . 1 ( Spotfire , Somerville , MA ) for data analysis . In accordance with proposed MIAME ( minimum information about a microarray experiment ) standards [85] , all data described in this report , including sample information , intensity measurements , gene lists , error analysis , microarray content , and slide hybridization conditions , are publically available at http://viromics . washington . edu . The Resolver system performs a squeeze operation that creates ratio profiles by combining replicates while applying error weighting . The error weighting consists of adjusting for additive and multiplicative noise . A P value is generated that represents the probability that a gene is differentially expressed . The Resolver system then combines ratio profiles to create ratio experiments using an error-weighted average as described in Roland Stoughton and Hongyue Dai , Statistical Combining of Cell Expression Profiles ( US Patent #6 , 351 , 712 , February 26 , 2002 ) . For each microarray experiment , the calculation of mean ratios between expression levels of each gene in the analyzed sample pair , standard deviations , and P values was performed using Resolver . In this study , genes selected for data analysis had a greater than 95% probability of being differentially expressed ( P≤0 . 05 ) and fold change ≥2 . Ingenuity Pathway Analysis ( IPA ) software and Entrez Gene ( www . ncbi . nlm . nih . gov/sites ) were used for gene ontology analysis . Analysis of Variance ( ANOVA ) , a factorial-based analysis method for determining statistical differences between means of different populations , was used to identify genes that statistically distinguished either species or time points . Complete blood counts ( CBC ) were determined at the WaNPRC Clinical Lab in Seattle as well as by absolute counting on 50 µl whole blood , using Trucount absolute counting tubes ( BD Biosciences , San Jose , CA ) . Phenotyping was performed by cell surface staining . FACS data were analyzed using FlowJo software with standard gating strategies and then transferred into analysis and graphic software ( Excel , StatView ( Abacus Concepts , Berkeley , CA ) ) . Multifunctional cytokine analysis was performed after stringent gating of each cytokine positive population and subsequent Boolean gating ( FlowJo ) .
|
Simian immunodeficiency virus ( SIV ) does not cause disease in African green monkeys ( a natural host for the virus ) , whereas experimentally infected Asian macaques ( a non-natural host ) develop a progressive disease that is similar to that which occurs in HIV-infected humans . Insight into how HIV causes disease and leads to development of AIDS may therefore be gained by comparing the response of natural and non-natural hosts to SIV infection . To this end , we examined changes that occurred in gene expression levels over time and in multiple tissues derived from African green monkeys and Asian macaques experimentally infected with SIV . Infection leads to host-specific gene expression patterns in lymph nodes , blood , and colon . The natural and non-natural hosts differed with respect to the timing , intensity , and duration of infection-induced gene expression changes associated with inflammation and response to stress .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"genetics",
"and",
"genomics/gene",
"expression",
"genetics",
"and",
"genomics/functional",
"genomics",
"virology/immunodeficiency",
"viruses",
"virology/animal",
"models",
"of",
"infection",
"immunology/innate",
"immunity",
"virology/mechanisms",
"of",
"resistance",
"and",
"susceptibility,",
"including",
"host",
"genetics",
"infectious",
"diseases/hiv",
"infection",
"and",
"aids",
"virology/effects",
"of",
"virus",
"infection",
"on",
"host",
"gene",
"expression",
"immunology/immunity",
"to",
"infections",
"virology/host",
"antiviral",
"responses"
] |
2009
|
Transcriptional Profiling in Pathogenic and Non-Pathogenic SIV Infections Reveals Significant Distinctions in Kinetics and Tissue Compartmentalization
|
Dengue is the most prevalent arthropod-borne viral illness in humans with half of the world’s population at risk . During early infancy , severe dengue can develop after a primary dengue virus infection . There has been a clinical observation that severe dengue during the first year of life is seen only in chubby infants . We examined the associations between the development of severe dengue and adipose tissue accumulation patterns during the first year of life in a prospective observational clinical study of infants and dengue virus infections . We found that adipose tissue contains two potential targets for dengue virus infection and production- adipocytes and adipose tissue macrophages . During the first year of life , total body adiposity and visceral adipose tissue stores were at their highest levels in early infancy . Early infancy was also characterized by a relative decrease in alternatively activated ( anti-inflammatory ) macrophages , and a relative increase in circulating pro-inflammatory cytokines . The data has been used to propose a model where the adipose tissue accumulation pattern and pro-inflammatory environment during early infancy provide the conditions for the potential development of severe dengue in immune-susceptible infants .
Dengue is the most prevalent arthropod-borne viral illness in humans with half of the world’s population at risk . The global burden of symptomatic dengue is on the order of 100 million cases/year [1] . The dengue viruses ( DENVs ) are single-stranded , positive-sense , RNA-containing enveloped viruses belonging to the Flavivirus genus within the Flaviviridae family [2] . There are four serotypes of DENVs ( DENV1-4 ) . DENV infections produce a wide spectrum of clinical illness . It ranges from asymptomatic or mild illness , to classic dengue fever ( DF ) , to a severe and potentially life threatening disease , dengue hemorrhagic fever ( DHF ) /dengue shock syndrome ( DSS ) ( severe dengue ) . The majority of severe dengue is characterized by a transient vascular leakage , and is associated with high viral loads and a pro-inflammatory cytokine storm [3 , 4] . Most severe dengue is seen in older children and adults with a heterologous ( secondary ) DENV infection [3 , 5] . However , DHF/DSS is also seen during early infancy among infants with a primary DENV infection [3 , 6] . There has been a clinical observation that DHF/DSS during the first year of life is seen only in chubby infants . We therefore examined the associations between the development of infant DHF/DSS and adipose tissue accumulation patterns during the first year of life in a prospective observational clinical study of infants and DENV infections that we have been conducting in San Pablo , Laguna , Philippines [7 , 8] . Total body adiposity is comprised of visceral adipose tissue and subcutaneous adipose tissue [9] . We found that adipose tissue contains two potential targets for DENV infection and production- adipocytes and adipose tissue macrophages . During the first year of life , total body adiposity and visceral adipose tissue stores were at their highest levels in early infancy . Early infancy was also characterized by a relative decrease in alternatively activated ( anti-inflammatory ) macrophages , and a relative increase in circulating pro-inflammatory cytokines . The data has been used to propose a model where the adipose tissue accumulation pattern and pro-inflammatory environment during early infancy provide the conditions for the potential development of DHF/DSS in immune-susceptible infants .
The clinical study protocol was approved by the institutional review boards of the Research Institute for Tropical Medicine , Philippines , and the University of Massachusetts Medical School . Mothers and their healthy infants were recruited and enrolled after providing written informed consent . The prototype DENV2 strain NGC was used for the adipocyte infection experiments . De-identified matched pairs of visceral ( omental ) and subcutaneous pre-adipocytes were purchased from Zen-Bio ( Research Triangle Park , NC ) . Pre-adipocytes were seeded into wells and terminally differentiated into mature adipocytes , according to the manufacturer’s instructions . The adipocytes were adsorbed with DENV2 NGC at a multiplicity of infection ( MOI ) = 3 x 2 h , washed x 3 , and supernatant was collected for DENV quantitative ( q ) RT-PCR from days 0–4 . Experiments were performed in quadruplicate . Viral RNA was extracted from the supernatants ( Qiagen ) , and DENV qRT-PCR was performed , as previously described [10] . Details regarding the study protocol have been previously described [8] . In brief , study enrollment began in 2007 in San Pablo , Laguna , Philippines . Healthy infants and their mothers were enrolled when the infant was around 2 months old . A subset of infants returned approximately every 2 months over the first year of life for study visits . Blood samples , clinical , and epidemiological information were collected at the study visits . We conducted surveillance year-round for hospitalized acute febrile illnesses in study infants across the seven hospitals serving San Pablo . During the rainy season ( June-November ) , mothers were also encouraged to bring their infants to the San Pablo City Health Office for evaluation of outpatient febrile illnesses . A DENV infection was identified in febrile infants by serotype-specific RT-PCR in acute-phase sera [11] and DENV IgM/IgG ELISA [12] in paired acute and convalescent phase sera . Primary or secondary DENV infections were identified by previously established serologic criteria for the paired IgM/IgG ELISA results [12] At each study visit , weight was measured to the nearest tenth of a kg , length was measured to the nearest tenth of a cm , and subscapular skinfold thickness was measured to the nearest mm by calipers . Body mass index ( BMI ) was calculated as weight ( kg ) / ( length ( m ) ) 2 , and was used as an estimate of total body adiposity . The estimated subcutaneous adipose tissue mass was calculated as the subscapular skinfold thickness ( mm ) x body surface area ( BSA ) ( m2 ) ( √ ( length ( cm ) ) x ( weight ( kg ) ) /3600 ) . The estimated visceral adipose tissue mass was calculated as BMI—estimated subcutaneous adipose tissue mass . sCD163 , tumor necrosis factor-α ( TNF-α ) , interleukin-1β ( IL-1β ) , and IL-6 levels were measured in longitudinally collected sera over the first year of life from 176 healthy infants using a multiplex assay , according to the manufacturer’s instructions ( Bio-Plex , Bio-Rad ) .
Paired visceral and subcutaneous pre-adipocytes were differentiated to mature adipocytes and infected with DENV2 strain NGC at MOI = 3 . 60–80% of the cells in each well were mature adipocytes . DENV2 NGC was able to productively infect visceral and subcutaneous adipocytes equally well ( Fig 1 ) . Adipose tissue consists of approximately 80% adipocytes and 20% stromal cells . The majority of stromal cells are macrophages . Thus , adipose tissue contains two potential target cells for DENV infection and production- adipocytes and adipose tissue macrophages [3 , 13] . We performed longitudinal anthropometric measurements in 427 healthy infants over the first year of life . BMI was used as an estimate of total body adiposity , and BMI peaked between the ages of 4–7 months old ( S1 Fig ) . Subscapular skinfold thickness was used as a measure of subcutaneous fat . The equations used to estimate the subcutaneous and visceral adipose tissue stores in the infants are described in the Methods section . Over the first year of life , the estimated subcutaneous adipose tissue mass gradually increased , while the estimated visceral adipose tissue mass reached its highest level around ages 4–7 months old , and then decreased ( Fig 2a ) . Primary DENV infections in infants that led to DHF/DSS clustered between the ages where visceral adipose tissue stores were around their highest levels ( Fig 2b ) . We measured in a longitudinal manner the serum levels of sCD163 in 176 healthy infants over the first year of life . The mean circulating levels of sCD163 decreased after birth , reached their lowest levels between ages 4–7 months old , and then gradually increased ( Fig 3 ) . sCD163 is a macrophage activation marker that can be increased by a pro-inflammatory state or an increase in alternatively activated macrophages [14] . During infancy , the nadir of circulating sCD163 levels occurred when circulating pro-inflammatory cytokines were at their highest levels . This suggests that circulating sCD163 levels during infancy reflected the alternative activation state of macrophages . We measured in a longitudinal manner over the first year of life the serum levels of TNF-α , IL-1β , and IL-6 in the same 176 infants with sCD163 determinations . The mean circulating levels of these pro-inflammatory cytokines increased after birth , reached their highest levels between the ages of 4–7 months old , and then decreased ( Fig 4 ) .
In adipose tissue , DENV can productively infect adipocytes and macrophages . We observed that infants who developed unambiguous DHF/DSS after a primary DENV infection clustered between the ages of 3–10 months old . In this age range , estimated visceral adipose tissue mass was also around its highest level during the first year of life . Mean circulating levels of a marker for alternatively activated macrophages were relatively decreased during this time period , and mean circulating levels of pro-inflammatory cytokines were relatively increased . This data has led us to propose a model for infant DHF/DSS risk factors that explains the clinical observation that DHF/DSS during the first year of life is seen only in chubby infants . First , we propose that adipose tissue ( specifically adipocytes and adipose tissue macrophages ) is an important source of DENV production . During the first year of life , severe dengue ( DHF/DSS ) can develop with a primary DENV infection when total body adiposity and visceral adipose tissue stores are around their highest levels . When visceral adipose tissue mass is around its highest level in early infancy , there is also a relative decrease in alternatively activated ( anti-inflammatory ) macrophages [15] and there exists a pro-inflammatory cytokine environment . When infants become immune-susceptible ( maternally-acquired anti-DENV antibody titers fall below protective levels ) , these conditions provide an environment where there can be high viral loads with the development of a pro-inflammatory cytokine storm- the risk factors for developing DHF/DSS . Childhood and adult obesity are associated with increased visceral adipose tissue stores , macrophage infiltration of the visceral adipose tissue , a relative paucity of alternatively activated visceral adipose tissue macrophages , and a pro-inflammatory cytokine environment [16–18] . We suggest that a similar pattern of adipose tissue accumulation and macrophage infiltration and activation is seen during early infancy , and this pattern creates the conditions for the potential development of severe dengue in immune-susceptible infants after a primary DENV infection .
|
Dengue is the most prevalent arthropod-borne viral illness in humans with half of the world’s population at risk . During early infancy , severe dengue can develop after a primary dengue virus infection . There has been a clinical observation that severe dengue during the first year of life is seen only in chubby infants . We therefore examined the associations between the development of severe dengue and adipose tissue accumulation patterns during the first year of life in a prospective observational clinical study of infants and dengue virus infections . We found that adipose tissue contains two potential targets for dengue virus infection and production- adipocytes and adipose tissue macrophages . During the first year of life , total body adiposity and visceral adipose tissue stores were at their highest levels in early infancy . Early infancy was also characterized by a relative decrease in alternatively activated ( anti-inflammatory ) macrophages , and a relative increase in circulating pro-inflammatory cytokines . The accumulation of visceral adipose tissue , the relative decrease in alternatively activated macrophages , and the relative increase in circulating pro-inflammatory cytokines during early infancy provide the environment for the development of severe dengue in immune-susceptible infants .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[] |
2015
|
The Pattern of Adipose Tissue Accumulation during Early Infancy Provides an Environment for the Development of Dengue Hemorrhagic Fever
|
Viral invasion triggers the activation of the host antiviral response . Besides the innate immune response , stress granules ( SGs ) also act as an additional defense response to combat viral replication . However , many viruses have evolved various strategies to suppress SG formation to facilitate their own replication . Here , we show that viral mRNAs derived from human parainfluenza virus type 3 ( HPIV3 ) infection induce SG formation in an eIF2α phosphorylation- and PKR-dependent manner in which viral mRNAs are sequestered and viral replication is inhibited independent of the interferon signaling pathway . Furthermore , we found that inclusion body ( IB ) formation by the interaction of the nucleoprotein ( N ) and phosphoprotein ( P ) of HPIV3 correlated with SG suppression . In addition , co-expression of P with NL478A ( a point mutant of N , which is unable to form IBs with P ) or with NΔN10 ( lacking N-terminal 10 amino acids of N , which could form IBs with P but was unable to synthesize or shield viral RNAs ) failed to inhibit SG formation , suggesting that inhibition of SG formation also correlates with the capacity of IBs to synthesize and shield viral RNAs . Therefore , we provide a model whereby viral IBs escape the antiviral effect of SGs by concealing their own newly synthesized viral RNAs and offer new insights into the emerging role of IBs in viral replication .
Human parainfluenza virus type 3 ( HPIV3 ) , a member of the Paramyxoviridae family , can cause acute respiratory tract diseases such as pneumonia and bronchitis in infants and children [1] . The genome of HPIV3 contains a non-segmented negative-strand RNA , which encodes 6 proteins: the nucleoprotein ( N ) , phosphoprotein ( P ) , matrix protein ( M ) , fusion protein ( F ) , hemagglutinin/neuraminidase ( HN ) , and large RNA-dependent RNA polymerase ( L ) [2] . The virus enters host cells through a membrane fusion process carried out by HN and F [3] . P bridges L to the N-encapsidated genomic RNA to initiate viral transcription and replication [4–6] , which generates 6 capped and polyadenylated messenger RNAs ( polyA+ mRNAs ) and a full-length anti-genomic RNA [7] . The new synthesized viral proteins and genomic RNA assemble into virions , which are released from the cell through the budding process [8 , 9] . Viral invasion triggers host antiviral defense responses , including the well-known innate immune response and the protein kinase R ( PKR ) -dependent stress response [10 , 11] . On one hand , viral RNAs synthesized during the infection are usually regarded as non-self , pathogen-associated molecular patterns ( PAMPs ) . The PAMPs can be detected by retinoic acid-inducible gene I ( RIG-I ) and melanoma differentiation-associated protein 5 ( MDA5 ) [12–16] , and this detection activates the interferon ( IFN ) signaling pathway and results in the expression of IFN-α/β . Secreted IFN-α/β induces the transcription and expression of IFN-stimulated genes ( ISGs ) , which play a critical role in restricting viral replication and propagation in host cells [17–19] . On the other hand , particular RNA species such as double-stranded RNA ( dsRNA ) or secondary-structured RNA generated during viral infection can also be detected by PKR [20 , 21] . This recognition activates PKR auto-phosphorylation , which results in the phosphorylation of eukaryotic initiation factor 2 α subunit ( eIF2α ) . Phosphorylated eIF2α prevents the assembly of the ternary pre-initiation complex and inactivates global protein synthesis . Stalled translation initiation complexes are then recruited into the cytoplasmic aggregates , which are termed SGs [22] . Thus , SGs typically consist of stalled mRNAs , 40S ribosomal subunits , various translation initiation factors such as eIF3 , eIF4A , eIF4E , and eIF4G , and several SG-nucleating factors , including G3BP [23–25] . As viral protein synthesis must rely on the host translation machinery , SG-mediated translational arrest usually plays an antagonistic role in the viral life cycle [26] . Given the inhibitory effect of SGs on viral infection , many viruses have evolved various strategies to block SG formation to ensure efficient viral replication . Sendai virus ( SeV ) and measles virus ( MV ) encode a C protein to limit the accumulation of dsRNA to inhibit SG formation [27 , 28] . Influenza A virus ( IAV ) NS1 protein prevents viral dsRNA from being detected by PKR , thus resulting in the inhibition of SG formation [29 , 30] . Poliovirus infection induces SG formation at the early stage of viral infection; then , SGs gradually disappear due to the cleavage of G3BP by viral 3C protease [31] . Respiratory syncytial virus ( RSV ) sequesters p38 and OGT into viral IBs to suppress SG assembly [32] . Ebola virus ( EBOV ) IBs sequester various SG markers and inhibit SG formation during infection [33 , 34] . In this report , we show that viral mRNAs derived from HPIV3 infection induce eIF2α phosphorylation and trigger SG formation in a PKR-dependent manner . HPIV3-induced SGs play an inhibitory role in HPIV3 replication by sequestering viral mRNAs . Disruption of SG formation by different approaches dramatically increases viral protein expression and virion production without influencing IFN production . Furthermore , we show that HPIV3 IBs formed during infection block SG formation by shielding viral RNAs .
To determine whether HPIV3 infection induces SG formation , we infected HeLa cells with HPIV3 for up to 36 hour ( h ) and analyzed the distribution of SG marker proteins TIA-1 and G3BP at different time points post-infection ( pi ) . An antibody against HPIV3 was used to evaluate viral protein expression and viral replication ( Fig 1A ) . In mock-infected cells , TIA-1 was homogeneously distributed in both the cytoplasm and the nuclei and G3BP was homogeneously distributed in the cytoplasm . In contrast , TIA-1 translocated from the nuclei to the cytoplasm and formed aggregates co-localizing with G3BP during viral infection , which was similar to the results after sodium arsenite ( AS ) stimulation ( Fig 1A ) . As infection progressed , the number of cells containing SGs increased to nearly 80% at 24 hpi and this ratio remained stable until 36 hpi ( Fig 1B ) . To determine whether HPIV3-induced SGs contain other typical SG marker proteins , we also examined the distribution of eIF4A , eIF4E , and eIF4G and found that all these marker proteins were re-distributed into SGs and co-localized with G3BP during HPIV3 infection ( S1A Fig ) . To determine whether HPIV3-induced SGs are cell-specific , we also examined the distribution of TIA-1 and G3BP in HEp-2 cells and MK2 cells infected with HPIV3 and found that both TIA-1 and G3BP were re-distributed into SGs and co-localized with each other , suggesting that SG formation is a general process during HPIV3 infection ( S1B Fig ) . Since the phosphorylation of eIF2α is critical for SG formation , we evaluated the phosphorylation status of eIF2α in HPIV3-infected cells and found that the level of phosphorylated eIF2α increased at 12 hpi , peaked at 24 hpi , and remained stable until 36 hpi . ( Fig 1C ) . Furthermore , we examined the distribution of phosphorylated eIF2α via immunofluorescence and found that little phosphorylated eIF2α was detected in mock-treated cells . In contrast , both HPIV3 infection and AS stimulation obviously increased the level of phosphorylated eIF2α , and phosphorylated eIF2α was recruited into SGs , which co-localized with G3BP ( S1C Fig ) , suggesting that HPIV3 infection activates eIF2α phosphorylation and subsequently induces SG formation . Assembly of SGs is a highly dynamic process , and phosphorylation of eIF2α inhibits translational initiation , allowing mRNA transcripts release from the elongating ribosomes and accumulate at SGs under environmental stress . Drugs that stabilize polysomes inhibit the assembly of SGs , whereas drugs that destabilize polysomes promote the assembly of SGs . cycloheximide ( CHX ) stabilize polysomes by freezing ribosomes on translating mRNAs and inhibit SG formation[35] . Previous studies reported that some viruses trigger the formation of atypical SGs that do not disassemble in the presence of CHX [36 , 37] . We next sought to determine whether HPIV3 infection- triggered SGs disassemble in response to CHX . CHX was added to cells in the presence of AS or HPIV3 infection . SGs induced by AS were cleared via incubation with CHX for 1 h ( Fig 2A and 2B ) . Similarly , after incubation with CHX for 3 h , HPIV3-induced SGs completely disappeared ( Fig 2C and 2D ) . CHX treatment results in global inhibition of protein synthesis . We examined the viral HN protein level via western blotting and found that HN protein expression was inhibited by incubation with CHX ( Fig 2E and 2F ) . To rule out possible effect of CHX on viral RNA synthesis , we also examined the viral RNA level after treatment with CHX , and found that neither viral genomic RNA nor mRNA synthesis was significantly affected in the presence of CHX ( Fig 2G and 2H ) . These data suggest that HPIV3 infection induces the formation of canonical SGs . HPIV3 infection induces SG formation , suggesting that either the viral RNAs or the viral proteins or both of them induce SG formation . N , P , M , F , and HN were separately expressed , but G3BP was always homogeneously distributed in the cytoplasm ( S2A Fig ) , suggesting that none of them triggered SG formation . Furthermore , over-expression of HPIV3 viral proteins could not induce the phosphorylation of eIF2α either ( S2B Fig ) , suggesting that SGs triggered by HPIV3 infection is not due to the accumulation of viral proteins . Then , we speculated that viral RNAs generated from viral replication and transcription might be the critical factors for inducing SG formation in host cells . We extracted RNAs from mock- or HPIV3-infected cells . These RNAs were subsequently transfected into HeLa cells that were immunostained with antibodies against TIA1 and G3BP to detect SG formation . We found that RNAs extracted from mock-infected MK2 cells failed to induce TIA-1 and G3BP aggregation into SGs . In contrast , RNAs extracted from HPIV3-infected MK2 cells induced the formation of a large number of SGs ( Fig 3A , compare panel “-HPIV3 MK2 RNA” to “+HPIV3 MK2 RNA” ) . As a positive control , Poly I:C ( pIC ) also stimulated SG formation ( Fig 3A ) , suggesting that RNAs derived from HPIV3-infected cells trigger SG formation . The time course experiment showed that small SGs appeared at 4 h post-transfection . As time extended , the SGs gradually became larger and reached the maximum at 12 h post-transfection ( S2C and S2D Fig ) . Several RNA viruses have been reported to activate eIF2α phosphorylation through activation of the upstream kinase PKR [20] . Having found that HPIV3 infection activated eIF2α phosphorylation , we sought to determine whether RNAs derived from HPIV3-infected cells could activate PKR/eIF2α phosphorylation . pIC transfection induced PKR and eIF2α phosphorylation ( Fig 3B ) . Similarly , RNAs extracted from HPIV3-infected cells also activated PKR/eIF2α phosphorylation . In contrast , PKR/eIF2α phosphorylation was not detected using RNAs extracted from mock-infected cells ( Fig 3B ) . As expected , AS treatment failed to activate PKR phosphorylation , but significantly activated eIF2α phosphorylation ( Fig 3B ) , which is consistent with previous research results showing that AS induces SG formation in a PKR-independent manner [38 , 39] . Taken together , these results show that HPIV3 viral RNAs activate PKR/eIF2α phosphorylation , thus inducing SG formation . Because viral replication and transcription generate genome RNAs , anti-genome RNAs , mRNAs and dsRNA , we sought to determine which RNAs play a critical role in inducing SG formation . For this purpose , we isolated PolyA+ RNAs ( viral mRNAs ) from RNAs derived from HPIV3-infected cells , and the remaining fraction was termed as PolyA- RNAs . Both fractions were evaluated for the induction of SG formation . PolyA+ RNAs had a much greater ability to induce SG formation than PolyA- RNAs ( Fig 3C and 3D ) . Furthermore , we also found that in vitro transcribed HPIV3 N mRNA could also induce SG formation ( S2E Fig ) . Therefore , we concluded that mRNAs of HPIV3 trigger SG formation . Next , we sought to know whether the SGs induced by transfection of PolyA+ RNAs shares the same property with the SGs induced by HPIV3 infection . CHX were added to cells transfected with PolyA+ RNAs derived from HPIV3-infected cells , and results showed that SGs induced by PolyA+ RNAs from HPIV3-infected cells also gradually disassembled in the presence of CHX ( Fig 3E and 3F ) , which is in accordance with the result of HPIV3 infection . We next sought to confirm whether HPIV3 infection induces SG formation in a PKR-dependent manner . We created cell lines to induce stable knockdown of PKR expression by transducing HeLa cells with lentiviral shRNA transduction vector . HPIV3 infection obviously induced SG formation in mock-knockdown cells ( Fig 4A , panel “sh-ctrl , ” and 4B ) . In contrast , PKR-knockdown cells failed to form SGs upon HPIV3 infection ( Fig 4A , panel “sh-PKR , ” and 4B ) . Similarly , neither RNAs from HPIV3-infected cells nor pIC induced SG formation in PKR-knockdown cells ( Fig 4C and 4D; S3A and S3B Fig ) , suggesting that HPIV3 induced SG formation in a PKR-dependent manner . Because SGs were still able to form upon AS treatment in PKR-knockdown cells ( S3C and S3D Fig ) , the inhibition of SG formation in PKR-knockdown cells during HPIV3 infection or pIC treatment was not due to an intrinsic defect of cells in response to stress . Next , we detected the phosphorylation level of PKR and eIF2α in both mock- and PKR-knockdown cells during HPIV3 infection . In mock-knockdown cells , HPIV3 infection obviously induced PKR/eIF2α phosphorylation , whereas in PKR-knockdown cells , phosphorylated PKR/eIF2α was not detected ( Fig 4E ) , suggesting that HPIV3-induced eIF2α phosphorylation depends on PKR phosphorylation . Interestingly , knockdown of PKR expression significantly increased viral HN protein expression and viral titers ( Fig 4E and 4F ) , indicating that HPIV3-induced SGs may play an inhibitory role in viral replication . Several viruses have been reported to trigger the formation of SGs that capture viral RNAs and suppress viral replication [26] . We next sought to evaluate the distribution of viral RNAs in HPIV3-infected cells . A cell line stably expressing GFP-G3BP was used to perform the RNA fluorescent in situ hybridization ( RNA-FISH ) assay . GFP-G3BP in HPIV3-infected cells was recruited into SGs ( Fig 5A ) . Viral positive-strand RNAs ( +vRNAs , including the mRNA , the most abundant viral RNA and the antigenome RNA ) and genomic RNAs ( -vRNAs ) were detected with specific probes in HPIV3-infected cells ( Fig 5A ) . Clearly , +vRNAs co-localized well with SGs , but -vRNAs were enriched adjacent to and encircle the SGs ( Fig 5A ) , suggesting that HPIV3-induced SGs can capture +vRNAs . Since viral mRNAs are the major component of +vRNAs , we hypothesized that SGs sequester viral mRNAs to suppress viral protein translation , thereby inhibiting viral replication . Having established that knockdown of PKR expression significantly increases HN expression and HPIV3 titers , and HPIV3-induced SGs capture viral mRNAs , we sought to determine whether SG formation inhibits HPIV3 replication . Because G3BP is a core factor in SG formation , we first knocked down G3BP expression by shRNA and found that minimal G3BP could be observed in G3BP-knockdown cells , while TIA-1 expression was substantial ( Fig 5B ) . The percentage of G3BP-knockdown cells containing SGs decreased markedly upon HPIV3 infection ( Fig 5B and 5C ) or AS treatment ( S3E and S3F Fig ) , suggesting that knockdown of G3BP expression significantly impairs SG formation . We then detected HN expression and HPIV3 titers and found that inhibition of SG formation significantly enhanced HN expression ( Fig 5D ) and viral production ( Fig 5E ) , indicating that SGs play an antiviral role . To exclude the possibility that the antiviral activity of G3BP is inherent , we over-expressed an HA-tagged non-phosphorylatable mutant of eIF2α , eIF2α-S51A ( HA-eIF2α-S51A ) , in HeLa cells , and found that eIF2α-S51A expression resulted in deficient SG formation , but eIF2α expression had no effect on SG formation upon HPIV3 infection ( Fig 5F and 5G ) or AS stimulation ( S3G and S3H Fig ) . Similarly , inhibition of SG formation caused by eIF2α-S51A expression significantly enhanced HN expression and viral production , but eIF2α expression had no effect on either HN expression or viral production ( Fig 5H and 5I ) . Taken together , these data suggest that HPIV3-induced SG formation has an antiviral effect , and inhibition of SG formation enhances HPIV3 replication . HPIV3 infection generates various viral RNAs in which dsRNA , the 5’-triphosphate of genome and the anti-genome RNA are generally considered as an inducer of IFN production [13 , 40] . We confirmed that HPIV3 infection and RNAs extracted from HPIV3-infected cells indeed efficiently induced IFN expression ( Fig 6A and 6B ) , suggesting that viral RNAs derived from HPIV3 infection can not only induce SG formation but also activate IFN production . Next , we sought to know whether there is a link between IFN induction and SG formation . First , we over-expressed the caspase activation and recruitment domain ( CARD ) of RIG-I ( RIG-I-N ) or VISA ( also known as MAVS , IPS-1 or cardif ) to activate the RLR pathway , and found that over-expression of RIG-I-N or VISA induced potently IFN expression , but failed to induce SG formation ( S4A and S4B Fig ) . Furthermore , to rule out the possibility that activation of RLR pathway is required for SG formation , MEF cells-knocked out RIG-I or VISA were infected with HPIV3 . The results showed that HPIV3 infection induced SG formation in MEF cells-knocked out RIG-I or VISA as efficiently as in wild-type MEF cells ( S4C–S4E Fig ) . Taken together , these data suggest that IFN induction is not required for SG formation which is consistent with previous results that activation of the RLR pathway does not induce and is not required for SG formation[38] . Several studies have suggested that SGs may serve as a platform for IFN production [41–43] , while others have pointed out that SGs are dispensable for the induction of innate immune response [38 , 44] . To determine whether HPIV3-induced SGs are involved in IFN response , we measured IFN response when SG formation was disrupted . We found that HPIV3 infection still potently induced IFN expression in spite of the inhibition of SG formation by over-expression of eIF2α-S51A ( Fig 6C ) or knockdown of G3BP expression ( Fig 6D ) . Furthermore , HPIV3 infection equally induced IRF3 translocation into the nucleus in both mock-knockdown and G3BP-knockdown cells ( Fig 6E ) , suggesting that SG formation is dispensable for the activation of IFN response . Since SG formation plays an antiviral role in HPIV3 infection , we next sought to determine whether viral proteins can inhibit SG formation to escape the antiviral response of SGs . First , we found that over-expression of M , F , or HN separately had no effect on SG formation in HPIV3-infected cells ( S5A and S5B Fig ) . We also expressed N or P individually in HPIV3-infected cells ( S5C Fig ) and examined distribution of N or P ( Fig 7A ) . We found that there was no significant difference for HPIV3-induced SG formation in the presence or absence of N or P , and over-expressed N or P distributed homogeneously in >90% of HPIV3-infected cells ( Fig 7A–7C ) , but , to our surprise , we did find that ~8% of cells over-expressing N or ~3% of cells over-expressing P formed IBs in HPIV3-infected cells ( due to over-expressed N interaction with P derived from HPIV3 infection or over-expressed P interaction with N derived from HPIV3 infection ) , and all the cells containing IBs appear to have no SGs ( Fig 7A–7C; “+”implies a cell forming IBs ) , indicating that over-expression of N or P does not inhibit HPIV3-induced SG formation unless N or P form IBs , which also implies that IBs may play an inhibitory function on SG formation . We previously reported that when co-expressed N and P of HPIV3 could form IBs , which were the center of HPIV3 viral RNA synthesis [7] . To confirm that IBs indeed inhibit SG formation , we over-expressed N and P of HPIV3 to force the formation ability of IBs in HPIV3-infected cells . We found that ~60% cells formed IBs , all the IBs-containing cells had no SGs and the percentage of cells containing SGs decreased from ~80% to ~20% . ( Fig 7D , panel “GFP-P+N-Myc” and Fig 7E and 7F ) . However , co-expression of GFP-P with NL478A-Myc , a point mutant of N ( we previously showed that NL478A was unable to form IBs with P [7] ) , failed to inhibit SG formation in HPIV3-infected cells ( Fig 7D , panel “GFP-P+NL478A-Myc” and Fig 7E and 7F ) , suggesting that the formation of IBs by over-expression of N and P indeed efficiently inhibit SG formation , and disruption of IB formation abolishes the inhibitory effect on SG formation in HPIV3-infected cells . We previously showed that N and P were the critical components of IB formation of HPIV3 [7] . Because it is hard to reveal the formation of IBs during wild type HPIV3 infection due to lack of specific antibody against N or P , we used a recombinant virus HPIV3HA-P , ( a HA tag fused to the N-terminal of P ) to detect the expression of P and the formation of IBs in HPIV3HA-P-infected cells [45] . At 24–36 hpi , HA-P homogeneously distributed throughout the cytoplasm , and ~80% of cells formed SGs . ( Fig 7G , panel 24 h and 36 h ) . As the time increased pi , HA-P gradually formed IBs , and staining with the antibody against HPIV3 also showed the IBs structures ( Fig 7G , panel at 48 h and 60 h , cells marked with “+” represent the cells that form IBs ) . At 48 hpi , ~20% cells form IBs and SGs are always excluded in these IBs-containing cells; at 60 hpi , the number of IBs-containing cells increased up to ~40% and simultaneously , the number of SGs-containing cells decreased to ~20% ( Fig 7G–7I ) . Similar results were also obtained when HEp-2 and MK2 cells were infected with HPIV3HA-P ( S6A and S6B Fig ) , suggesting that IBs have an inhibitory effect on SG formation . To confirm that HPIV3HA-P has similar replication behavior , we also performed a complementary experiment in wild type HPIV3-infected cells and did find wild type HPIV3 induced IB formation in ~20% of cells at 48 hpi and ~40% of cells at 60 hpi ( S6C Fig , cells marked with “+” represent the cells that form IBs , and S6D and S6E Fig ) . Similarly , SGs decreased along with the increase of IBs and all the cells containing IBs were deficient in SG formation . Taken together , these data show the inhibitory effect of IBs on SGs formation . Having established that IB formation can inhibit SG formation during HPIV3 infection , we next sought to determine whether the IBs formed by N and P of HPIV3 could inhibit SG formation induced by other stress stimuli . We over-expressed N and P of HPIV3 as well as AS or pIC stimuli and found that IB formation failed to inhibit AS- and pIC-induced SG formation ( Fig 8A and 8C ) . The percentage of cells containing SGs was not affected in the presence of HPIV3 IBs ( Fig 8B and 8D ) , suggesting that HPIV3 IBs do not disrupt typical stress response pathways of SG formation and inhibition of HPIV3-induced SG formation by HPIV3 IBs is specific . Our previous results have showed that IBs are the center of RNA synthesis [7] . Having found that mRNAs of HPIV3 induced the formation of SGs , which subsequently sequestered viral mRNAs to restrict viral replication , and that IBs inhibited SG formation in an HPIV3-infection specific manner , we sought to determine whether HPIV3 IBs could shield newly synthesized viral RNAs of HPIV3 to escape the inhibitory effect of SGs on viral replication . We co-expressed GFP-P and N-Myc in HPIV3-infected cells and examined viral RNAs via RNA-FISH assay . The results revealed that both–vRNAs and +vRNAs obviously co-localized with IBs ( Fig 9A and 9B , panel “GFP-P+ N-Myc” ) . To further demonstrate that the accumulation of viral RNAs in IBs is essential for the inhibition of SG formation , we identified a truncated mutant of N , NΔN10 ( lacking N-terminal 10 amino acids of N ) , that could still form IBs when co-expressed with P , but was function-defective and failed to support viral RNA synthesis ( Fig 9C ) . We found that neither -vRNAs nor +vRNAs co-localized with these function-defective IBs ( Fig 9A and 9B , panel “GFP-P+ NΔN10-Myc” ) . Subsequently , IBs formed by GFP-P and NΔN10-Myc also failed to inhibit SG formation in HPIV3-infected cells ( Fig 9D , panel “GFP-P+ NΔN10-Myc” ) , suggesting that viral RNAs which are synthesized and shielded in IBs are critical to avoid the induction of SG formation . Furthermore , previous research showed that co-expression of N and P of RSV can also form IBs [45] . Thus , we co-expressed GFP-PRSV and Myc-NRSV to form RSV IBs in HPIV3-infected cells and found that neither -vRNAs nor +vRNAs co-localized with RSV IBs ( Fig 9A and 9B , panel “GFP-PRSV+ Myc-NRSV” ) . Again , RSV IBs were unable to inhibit HPIV3-induced SG formation ( Fig 9D , panel “GFP-PRSV+ Myc-NRSV” and 9E ) . Taken together , these data show that HPIV3 IBs synthesized viral RNAs and simultaneously shielded viral RNAs to avoid the formation of SGs . Since transfection of RNAs derived from HPIV3-infected cells was sufficient to induce SG formation , we next sought to determine whether HPIV3 IBs could inhibit the SG formation induced by the transfection of viral RNAs . We co-expressed N and P of HPIV3 as well as viral RNAs extracted from HPIV3-infected cells . As shown in the RNA-FISH assay , HPIV3 IBs neither captured +vRNAs of HPIV3 transfected into HeLa cells ( Fig 10A , panel “+HPIV3 MK2 RNA” ) nor suppressed SG formation induced by viral RNAs from HPIV3-infected cells ( Fig 10B and 10C ) , suggesting that HPIV3 IBs specifically shield newly synthesized HPIV3 viral RNAs , thus resulting in the inhibition of HPIV3-induced SG formation .
In this study , we showed that mRNAs of HPIV3 induced SG formation in a PKR and eIF2α phosphorylation-dependent manner and found that HPIV3-induced SG formation is a general phenomenon in at least two cell types . Furthermore , we demonstrated that HPIV3 IBs efficiently suppressed the SG formation by shielding newly synthesized viral RNAs to escape recognition of PKR . Previous studies reported that some viruses-induced SGs were distinct from canonical SGs in both composition and function . Such as vesicular stomatitis virus ( VSV ) and rabies virus ( RABV ) , two members of the Rhabdoviridae family , can trigger noncanonical SG formation in infected cells [36 , 37] . These SGs do not disassemble in the presence of CHX , a polysome-stabilizing drug that traps stalled mRNAs in polysomes , and VSV-induced pseudo-SGs lack several SG markers such as eIF3 and eIF4A . In our study , we observed that multiple canonical SG markers such as TIA-1 , G3BP , eIF4A , eIF4E , and eIF4G , were all recruited into HPIV3-induced SGs ( Fig 1 and S1 Fig ) . CHX treatment resulted in the disassembly of HPIV3- and AS-induced SGs ( Fig 2 ) , suggesting that HPIV3 infection induces canonical SG formation . But the disassembly of HPIV3-induced SGs in response to CHX is somewhat delayed compared to AS-induced SGs . Three possibilities may contribute to delayed disassembly of HPIV3-induced SGs . 1 ) HPIV3-induced SG formation is PKR dependent , but AS-induced SG formation is not ( Fig 4A and 4B; S3C and S3D Fig ) ; 2 ) HPIV3-induced SGs are rich of viral mRNAs , while AS-induced SGs are not ( Fig 5A ) ; 3 ) HPIV3-induced SGs go through a longer period of time during infection ( 24 h ) than AS treatment ( 1 . 5 h ) ( Fig 2A–2D ) . Taken together , these differences may delay the disassembly of HPIV3-induced SGs in response to CHX . Transfection of RNAs extracted from HPIV3-infected cells instead of the expression of each viral protein could induce PKR/eIF2α phosphorylation , thus resulting in SG formation in host cells ( Fig 3A and 3B and S2A and S2B Fig ) , suggesting that vRNAs are the major factors triggering SG formation . Three major categories of vRNAs are generated during HPIV3 infection: viral genomic RNAs , anti-genomic RNAs and a series of PolyA+ RNAs . We isolated PolyA+ RNAs from the vRNAs and found that PolyA+ RNAs had a greater ability to induce SG formation than PolyA- RNAs ( Fig 3C and 3D ) . Since most PolyA+ RNAs within the vRNAs are viral mRNAs , we concluded that the mRNAs of HPIV3 are the major factors that induce SG formation during HPIV3 infection . But the exact mechanism how viral mRNAs activate phosphorylation of PKR and induce SG formation remains to be further explored . Since SGs usually play inherent antiviral activity , what is the relationship between SGs and vRNAs during viral infection ? In RABV- and Newcastle disease virus ( NDV ) -infected cells , vRNAs are synthesized in viral replication complexes ( VRC ) , then +vRNAs , including viral mRNAs or/and anti-genomic RNAs , are selectively transported from VRC to SGs , whereas the viral genomic RNAs are excluded [37 , 43]; in IAV-infected cells , viral genomic RNAs localized within SGs [42] , while in encephalomyocarditis virus ( EMCV ) -infected cells , viral dsRNAs are sequestered into SGs [41] , suggesting that the vRNAs in SGs differ with viruses types . In this study , +vRNAs of HPIV3 were sequestered into HPIV3-induced SGs , while–vRNAs were enriched adjacent to and encircle the SGs ( Fig 5A ) . Since SGs are considered storage sites of translational initiation trapped mRNAs , and viral mRNAs are the major component of +vRNAs , it is easy to understand that +vRNAs are selectively recruited into the core of HPIV3-induced SGs . Because in the process of HPIV3 replication , -vRNAs serve as the template for the synthesis of +vRNAs ( include viral anti-genome RNA and mRNA ) , –vRNAs are closely coupling with +vRNAs and encircle SGs , suggesting that HPIV3-induced SGs can capture viral mRNAs and may restrict viral replication through inhibition of viral translation . HPIV3 infection induces SG formation in a PKR-dependent manner . Knockdown of PKR expression significantly inhibits eIF2α phosphorylation , thus blocking SG formation ( Fig 4 ) , which is in accordance with the critical role of PKR in SG formation triggered by many viruses [20 , 28 , 42 , 46] . Furthermore , knockdown of PKR expression increases HPIV3 protein expression and viral production ( Fig 4E and 4F ) , implying that HPIV3-induced SGs have an antiviral role . We further disrupted SG formation via knockdown of G3BP expression and over- Inhibition of SG formation via expression of an eIF2α non-phosphorylatable mutant , eIF2α-S51A , dramatically increased viral protein expression and viral production of HPIV3 ( Fig 5 ) , but had no effect on the production of IFN ( Figs 5 and 6 ) , suggesting that HPIV3-induced SGs indeed play an antiviral role in spite of the IFN induction . These results are consistent with previous findings that SGs are dispensable for induction of innate immune response [38 , 44] . How does HPIV3 escape from antiviral activity of SGs ? Some viruses appear to inhibit SG formation during infection [31 , 34 , 47–56] , and some viruses may tolerate or exploit SGs to facilitate their own replication [37 , 57 , 58] . In our study , we showed that N or P expressed individually could form viral IBs in some HPIV3-infected cells , and all these IB-containing cells were devoid of SG formation ( Fig 7A–7C ) , suggesting that HPIV3 IBs have a novel inhibitory effect on SG formation . Our previous work showed that over-expression of HPIV3 N and P proteins can form IBs , while a point mutant of N , NL478A , failed to interact with P , thus , resulting in failure to form IBs when co-expressed with P [7] . Therefore , we co-expressed P with N or NL478A to detect their inhibitory effect on SG formation . Our results showed that N and P formed large numbers of IBs which efficiently inhibited SG formation , while NL478A failed to form IBs with P and did not inhibit SG formation ( Fig 7D–7F ) . We confirmed this result by infecting cells with HPIV3HA-P and found that all the HPIV3HA-P-infected cells containing IBs were deficient in SG formation ( Fig 7G–7I and S6A–S6B Fig ) , suggesting that HPIV3 IBs inhibit SG formation during HPIV3 infection . It should be noted that IB formation was not observed prior to approximately 48 hpi in HPIV3HA-P-infected cells , which is later than IB formation prior to 24 hpi as described in our previous studies [45] . The possible reason for the difference is that a new batch of cell lines were used in this assay . Previous studies showed that interplay between viral IBs and SGs is complicated , and there is no common conclusion of how viral IBs influence SG formation . RABV-induced SGs located closely to viral factories which are IB-like structures , termed as Negri bodies ( NBs ) , and viral RNAs are synthesized in NBs and viral mRNAs are specially transported from NBs to SGs [37]; VSV-induced SG-like structures co-localize or even share the same structure with viral cytoplasmic IBs [36]; IBs of RSV sequester p38 and OGT into viral IBs to suppress SG assembly [32 , 57]; EBOV sequesters many SG maker proteins to form SG-like structures inside viral IBs , probably resulting in the inhibition of antiviral role of SGs [34] . How do HPIV3 IBs inhibit SG formation during HPIV3 infection ? We found that SGs still robustly formed in IB-containing cells once the cells were treated with AS or pIC ( Fig 8 ) , suggesting that HPIV3 IBs do not inhibit AS- or pIC-induced SG formation and the inhibition of HPIV3-induced SG formation by IBs is virus-specific . Since IBs are the sites for RNA synthesis , we then sought to determine whether HPIV3 IBs could shield viral RNAs of HPIV3 to inhibit SG formation . Heinrich BS , et al . showed that VSV synthesizes its primary viral RNA throughout the cell cytoplasm by input RNP in the absence of viral IBs . Accumulation of viral proteins results in the formation of viral IBs that contain the RNA synthesis machine and become the predominant sites of viral RNA synthesis[59] . It is likely that this model also applies to HPIV3 infection in which the synthesis of viral RNA is prior to the formation of IBs and is initially unrestricted , but subsequently is shielded to avoid the SG formation . We then evaluated the vRNA-holding capacity of HPIV3 IBs and found that IBs formed by HPIV3 N and P redirect viral RNA synthesis from cytoplasm into IBs which shield viral RNAs and avoid the formation of SGs ( Fig 9 ) . However , IBs formed by RSV N and P failed to shield viral RNAs of HPIV3 and were therefore unable to inhibit HPIV3-induced SG formation ( Fig 9 ) . Similarly , IBs formed by NΔN10 and P of HPIV3 were replication-deficient and unable to shield viral RNAs or inhibit HPIV3-induced SG formation ( Fig 9 ) , suggesting that the vRNAs-holding capacity is rather critical for HPIV3 IBs to inhibit SG formation . Furthermore , to our surprise , we found that HPIV3 IBs also failed to capture the transfected vRNAs extracted from HPIV3-infected cells , suggesting that IBs of HPIV3 only specifically shield newly synthesized viral RNAs to inhibit SG formation ( Fig 10 ) . It is conceivable that HPIV3 IBs synthesize and hold viral mRNAs temporarily , then release viral mRNAs slowly or in an unknown way , which avoid accumulation of mRNAs instantly and SG formation . However , the exact mechanism how IBs shield newly synthesized viral RNAs and escape from recognition of PKR need to be further explored . To our knowledge , this is first report to describe a novel inhibitory effect of IBs on virus-induced SG formation by specifically capturing its own newly synthesized viral RNAs , thus helping the virus to escape antiviral SG formation and facilitate its own replication .
HeLa ( Human cervical cancer epithelial cells and were obtained from China Center for Type Culture Collection ) , HeLa-GFP-G3BP ( stable GFP-G3BP expression , derived from HeLa ) , LLC-MK2 ( monkey kidney cell line and were obtained from China Center for Type Culture Collection ) , HEp-2 ( Human laryngeal carcinoma epithelial cells and were originally obtained from American Type Culture Collection ) , HEK293T ( Human embryonic kidney 293 cells and were obtained from China Center for Type Culture Collection ) , RIG-I-/- and VISA-/- MEF ( mouse embryonic fibroblast cells and obtained from Shu HB lab , Wuhan university , China ) cells were cultured in Dulbecco’s modified Eagle’s medium ( DMEM , Gibco ) supplemented with 10% fetal bovine serum ( FBS , Gibco ) and 1% penicillin-streptomycin ( Gibco ) . PKR or G3BP stable knockdown cells were made via transduction with lentiviral shRNA transduction particles . Knockdown cells were cultured in DMEM maintenance media with puromycin ( 1 ug/ml , Sigma ) . Recombinant HPIV3 carrying an HA tag fused with the N-terminal of viral P ( HPIV3HA-P ) was constructed by our laboratory as described previously [45] . Both HPIV3 and recombinant HPIV3HA-P were propagated in LLC-MK2 cells by inoculation at a multiplicity of infection ( MOI ) of 0 . 1 . AS ( Sigma ) and CHX ( MCE ) were used at concentrations of 0 . 5 mM and 100ug/ml , respectively , for the indicated times . pIC was purchased from InvivoGen . Cells were cultured in 6-well or 24-well plates at a density of 70%-80% at 37°C overnight and incubated with HPIV3 at an MOI of 1 plaque-forming unit/cell for 2h at 37°C and 5% CO2 . Then the medium was replaced with fresh medium with 10% FBS . For the plaque assay , virus stock was serially diluted 10-fold up to 105 , and MK2 cells grown in 24-well plates were infected with 400 ul of the dilutions for 2h at 37°C and 5% CO2 . Then , the medium was replaced with methylcellulose , and the cell plates were incubated at 37°C and 5% CO2 for an additional 3–4 days until visible viral plaque was detected . After being stained with crystal violet , the plaques were counted to calculate the viral titers . Cells were harvested and lysed with lysis buffer ( 150 mM NaCl , 50 mM Tris-HCl [pH 7 . 4] , 1% Triton X-100 , 1 mM EDTA [pH 8 . 0] and 0 . 1% sodium dodecyl sulfate [SDS] ) for 30 min on ice . The supernatants were collected via centrifugation at 12000 g at 4°C for 30 min . The protein concentration was determined using the Bradford assay method ( Bio-Rad ) . Samples were boiled with SDS-PAGE loading buffer at 100°C for 10min and resolved via 10% SDS polyacrylamide gel electrophoresis ( SDS-PAGE ) . Proteins were then transferred to nitrocellulose membranes . The membrane was blocked with 5% milk in phosphate buffered saline ( PBS ) with 0 . 1% Tween 20 ( PBST ) for 1h before being incubated with primary antibodies overnight and then incubated with secondary antibodies for another 1h . The primary antibodies used were as follows: mouse anti-HN ( 1:1000 , Abcam ) , mouse anti-GAPDH ( 1:2500 , Santa Cruz ) , rabbit anti-eIF2α ( 1:1000 , CST ) , rabbit anti-phosphorylated eIF2α ( 1:1000 , CST ) , rabbit anti-PKR ( 1:1000 , Abcam ) , rabbit anti-phosphorylated PKR ( 1:1000 , Abcam ) , mouse anti-G3BP ( 1:1000 , BD Biosciences ) , mouse anti-HA tag ( 1:10000 , Sigma ) , rabbit anti-IRF3 ( 1:1000 , Abcam ) , and mouse anti-LMNB1 ( 1:1000 , Applygen ) . HRP-conjugated goat anti-mouse immunoglobulin ( IgG ) ( 1:5000 ) and goat anti-rabbit IgG ( 1:5000 ) were used as secondary antibodies . HeLa cells were cultured on coverslips in 24-well plates overnight . After transfection or/and infection , cells were harvested at the indicated times . Cells were fixed with 4% paraformaldehyde and permeabilized with 0 . 2% Triton X-100 for 20min at room temperature . After being blocked with 3% bovine serum albumin ( BSA ) for 30min , cells were incubated with primary antibodies diluted in 1% BSA at 4°C overnight and secondary antibodies diluted in 1% BSA at room temperature for another 1h . Cells were mounted with Fluoroshield ( Sigma ) and examined by using a Leica confocal microscope after staining with 1 mg/ml 4’ , 6-diamidino-2-phenylindole ( DAPI ) in PBS . The primary antibodies used were as follows: goat anti-TIA-1 ( 1:200 , Santa Cruz ) , goat anti-HPIV3 ( 1:1000 , Abcam ) , rabbit anti-TIA-1 ( 1:500 , ABclonal ) , rabbit anti-G3BP ( 1:500 , ABclonal ) , rabbit anti-eIF4A ( 1:500 , ABclonal ) , rabbit anti-eIF4E ( 1:500 , ABclonal ) , rabbit anti-eIF4G ( 1:200 , CST ) , rabbit anti-phosphorylated eIF2α ( 1:200 , CST ) , mouse anti-G3BP ( 1:500 , BD Bioscience ) , mouse anti-HA tag ( 1:2000 , Sigma ) , mouse anti-Flag tag ( 1:1000 , Sigma ) , and mouse anti-Myc tag ( 1:200 , Santa Cruz ) . The secondary antibodies used were as follows: Alexa Fluor 647 donkey anti-goat IgG ( 1:1000 , Invitrogen ) , Alexa Fluor 488 donkey anti-rabbit IgG ( 1:1000 , Invitrogen ) , and Alexa Fluor 594 donkey anti-mouse IgG ( 1:1000 , Invitrogen ) . The RNA-FISH assay was performed according to the manufacturer’s instructions for the QuantiGene ViewRNA ISH Cell Assay kit ( Affymetrix ) . Cells were fixed in 4% paraformaldehyde solution at room temperature for 30 min and permeabilized with detergent solution at room temperature for 5 min . Protease solution was added to the cells at a suitable dilution ( 1:1000–1:4000 ) in PBS , and the cells were incubated at room temperature for 10 min . After being washed with PBS , the cells were incubated with a special probe set at a 1:100 dilutions at 40°C for 3 h . Then a pre-amplifier , amplifier , and label probe were sequentially added to the cells all at a 1:25 dilution , and the cells were incubated at 40°C for 30 min . Cells were mounted with Fluoroshield ( Sigma ) and examined by using a Leica confocal microscope after being stained with 1 mg/ml DAPI in PBS . The probe sets targeted to HPIV3 +vRNA and HPIV3 –vRNA were purchased from Affymetrix . Total RNA from mock- or HPIV3-infected cells was solated using TRIzol reagent and precipitated by isopropanol . PolyA+ RNA was subsequently isolated from the total RNA according to the manufacturer’s instructions for the GenElute mRNA Miniprep Kit ( Sigma ) . We added 15 ul oligo ( dT ) polystyrene beads to 500 ug total RNA samples . The mixture was incubated at 70°C for 10 min and at room temperature for another 10 min . After centrifugation , supernatants were subjected to ethanol precipitation to obtain a concentrated PolyA- RNA fraction . The beads were washed twice with 500 ul wash solution through a GenElute spin filter . The PolyA+ RNA fraction was subsequently eluted and collected . Purification was repeated twice to yield a pure PolyA+ RNA fraction . cDNA of HPIV3 N was constructed into PBS vector downstream of T7 promoter and template was linearized and purified by phenol: chloroform and ethanol precipitation . mRNA was transcribed in vitro according to the manufacturer’s instructions for the Transcript Aid T7 High Yield Transcription Kit ( Thermo Scientific ) and purified using the RNeasy Mini Kit ( QIGEN ) . We transfected 0 . 5 ug of the RNA samples into HeLa cells in 24 wells using Lipofectamine 2000 reagent ( invitrogen ) . At 12 h post-transfection , the cells were harvested and subjected to immunofluorescence analysis . HEK293T cells were transfected with pWPI-GFP-G3BP and the package plasmids pCMV-VSV-G , pCMV-Tat and pCMV-R8 . 91 . At 48 h post-transfection , supernatants containing lentiviruses were collected to infect HeLa cells . Then the HeLa cells stably expressing GFP-G3BP were passaged . The shRNA lentiviruses were packaged via transfection of PKR shRNA or G3BP shRNA with the package plasmids psPAX2 and pMD2 . G into HEK293T cells . At 48 h post-transfection , lentiviruses were collected to infect HeLa cells to induce stable knockdown of PKR or G3BP expression . Cells were harvested , and then the nuclei fraction and the cytosol fraction were separated and extracted according to the manufacturer’s instructions for the Nuclear-Cytosol Extraction kit ( Applygen ) . Samples were boiled with SDS-PAGE loading buffer for western blot analysis . HEK 293T cells were transfected with 50 ng IFNβ-Luc reporter and 20 ng TK-Luc reporter for 12 h and then infected with HPIV3 for another 24 h . The firefly luciferase activity was assayed and normalized by that of Renilla luciferase . Experiments were performed in triplicate . Total RNA were isolated for qPCR analysis to measure the indicated RNA abundance . Data shown are the relative abundance of the indicated RNA normalized to that of GAPDH . The following Primers were used: HPIV3 M protein forward: 5’-AGAAGAACAGTCAAAGCGAAAG-3’; HPIV3 M protein reverse: 5’-CTCCAACTAATCCCAAAG-3’; Human GAPDH forward: 5’-GAGTCAACGGATTTGGTCGT-3’; Human GAPDH reverse: 5’-GACAAGCTTCCCGTTCTCAG-3’; Mouse IFNB1 forward: 5’-AGTTACACTGCCTTTGCC-3’; Mouse IFNB1 reverse: 5’-TGAGGACATCTCCCACGT-3’; Mouse GAPDH forward: 5’-GCATTGTGGAAGGGCTCA-3’; Mouse GAPDH reverse: 5’-AGGCGGCACGTCAGATC-3’ . HeLa cells were infected with vTF7-3 for 1h , then transfected with pGADT7-P ( 125ng ) , pGEM4-L ( 100ng ) , HPIV3 minigenome plasmid carrying the luciferase reporter gene ( 50ng ) together with pCDNA3 . 0-N-Myc ( 100ng ) or pCDNA3 . 0-NΔN10-Myc ( 100ng , 200ng or 400ng ) for 24h . Cell lysates were collected to determine the luciferase activity or analyzed via western blot .
|
Human parainfluenza virus type 3 ( HPIV3 ) is one of the major causes of acute respiratory tract diseases such as pneumonia and bronchitis in infants and children . Virus invasion activates cellular stress responses . One of these responses is the formation of SGs which counteract viral replication . However , many viruses have evolved various strategies to suppress SG formation , thus facilitating their own replication . We sought to determine if ( and how ) HPIV3 modulates SG formation to facilitate its replication and found that the viral messenger RNAs ( mRNAs ) of HPIV3 trigger SG formation in infected cells . As time increased post-infection , the number of cells containing SGs increased as well . To escape this response , HPIV3 forms IBs that shield viral RNAs , thereby preventing SG formation and allowing the virus to replicate and survive—and potentially invade other cells .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] |
[
"phosphorylation",
"medicine",
"and",
"health",
"sciences",
"nucleic",
"acid",
"synthesis",
"nuclear",
"staining",
"hela",
"cells",
"biological",
"cultures",
"messenger",
"rna",
"brain",
"microbiology",
"red",
"nucleus",
"cell",
"cultures",
"rna",
"synthesis",
"chemical",
"synthesis",
"research",
"and",
"analysis",
"methods",
"specimen",
"preparation",
"and",
"treatment",
"staining",
"proteins",
"cell",
"lines",
"viral",
"replication",
"biosynthetic",
"techniques",
"biochemistry",
"rna",
"dapi",
"staining",
"anatomy",
"nucleic",
"acids",
"post-translational",
"modification",
"virology",
"biology",
"and",
"life",
"sciences",
"cultured",
"tumor",
"cells"
] |
2018
|
Inclusion bodies of human parainfluenza virus type 3 inhibit antiviral stress granule formation by shielding viral RNAs
|
Urogenital schistosomiasis due to Schistosoma haematobium is a serious underestimated public health problem affecting 112 million people - particularly in sub-Saharan Africa . Microscopic examination of urine samples to detect parasite eggs still remains as definitive diagnosis . This work was focussed on developing a novel loop-mediated isothermal amplification ( LAMP ) assay for detection of S . haematobium DNA in human urine samples as a high-throughput , simple , accurate and affordable diagnostic tool to use in diagnosis of urogenital schistosomiasis . A LAMP assay targeting a species specific sequence of S . haematobium ribosomal intergenic spacer was designed . The effectiveness of our LAMP was assessed in a number of patients´ urine samples with microscopy confirmed S . haematobium infection . For potentially large-scale application in field conditions , different DNA extraction methods , including a commercial kit , a modified NaOH extraction method and a rapid heating method were tested using small volumes of urine fractions ( whole urine , supernatants and pellets ) . The heating of pellets from clinical samples was the most efficient method to obtain good-quality DNA detectable by LAMP . The detection limit of our LAMP was 1 fg/µL of S . haematobium DNA in urine samples . When testing all patients´ urine samples included in our study , diagnostic parameters for sensitivity and specificity were calculated for LAMP assay , 100% sensitivity ( 95% CI: 81 . 32%-100% ) and 86 . 67% specificity ( 95% CI: 75 . 40%-94 . 05% ) , and also for microscopy detection of eggs in urine samples , 69 . 23% sensitivity ( 95% CI: 48 . 21% -85 . 63% ) and 100% specificity ( 95% CI: 93 . 08%-100% ) . We have developed and evaluated , for the first time , a LAMP assay for detection of S . haematobium DNA in heated pellets from patients´ urine samples using no complicated requirement procedure for DNA extraction . The procedure has been named the Rapid-Heat LAMPellet method and has the potential to be developed further as a field diagnostic tool for use in urogenital schistosomiasis-endemic areas .
Human schistosomiasis , a parasitic freshwater snail transmitted disease caused by several species of genus Schistosoma trematode worms , is one of the 17 neglected tropical diseases ( NTDs ) considered by World Health Organization ( WHO ) [1] . It is estimated that 732 million persons are at risk of infection worldwide and over 200 million people are infected with this disease in 74 different countries , especially in sub-Saharan Africa [2–4] , where both associated morbidity and mortality are a significant barrier to social and economic development [5–7] . It must be also observed that the prevalence of imported schistosomiasis is increasingly a problem in non-endemic areas due to the growing number of international travellers to endemic areas , expatriates and immigrants from endemic countries [8–10] . Although humans are mainly infected by five species of schistosomes , namely Schistosoma mansoni , S . haematobium , S . japonicum , S . mekongi , and S . intercalatum , the main burden of disease in sub-Saharan Africa is usually attributed to two species referred to as the major human schistosomes: S . mansoni , causing hepatic and intestinal schistosomiasis and S . haematobium , the chief cause of urogenital schistosomiasis [3] . More people are infected with S . haematobium than with the other schistosomes; it is estimated that 112 million people suffer from urogenital schistosomiasis [11–14] . The infection typically results in haematuria , anaemia , dysuria and genital and urinary tract lesions , but in severe cases it may also lead to kidney damage . It is well known that the deposition of S . haematobium eggs eventually leds to squamous cell carcinoma of the bladder in many chronically infected individuals [15 , 16] the International Agency for Cancer Research ( IACR ) in association with WHO classified S . haematobium as a Group 1 biological carcinogen [17] . Moreover , most of women infected with S . haematobium suffer from female genital schistosomiasis of the lower genital tract [13]; which impairs fertility [18] and also increases susceptibility of the woman to HIV [19] . For the diagnosis of urogenital schistosomiasis , the gold standard remains microscopic detection of excreted ova in urine samples [20] after using either sedimentation/centrifugation or filtration methods [21] . These conventional methods are inexpensive , easy to perform under field conditions and relatively rapid . However , parasitological diagnosis has classically low sensitivity , especially in low-grade infections and may be affected by day-to-day variability in egg excretion , often missing diagnosis by microscopy [22 , 23] . In addition , egg count-based criteria cannot be carried out in the acute phase of the disease since the parasite have not yet started to produce eggs . The collection of a larger number of urine samples per individual on consecutive days instead of a single one may increase the sensitivity of microscopic detection , but is more expensive and also time-consuming [23] . Identifying blood in the urine-micro or macrohaematuria- has been widely and successfully used as a good indicator of S . haematobium infection , mainly in a high prevalence situation . However , haematuria is a nonspecific symptom of urogenital schistosomiasis in areas of low endemicity and can be incorrectly estimated depending on the infection prevalence in an area [24 , 25] . Antibody-based assays are useful to confirm S . haematobium infections , but do not distinguish active infection from past exposure , and so low sensitivity and specificity results frequently occur . Moreover , antibody tests are usually negative during acute symptomatic urogenital schistosomiasis . On the other hand , assays that detect circulating antigens seem very promising in the early phase of infection but still lack sensitivity in the diagnosis of light infections [20 , 26 , 27] . To overcome the drawbacks of both classical parasitological and immunological diagnostic methods , the development of new , more sensitive and specific molecular diagnostic tools for the diagnosis of urogenital schistosomiasis are desirable and still needed . In recent years , several studies have reported the utility of polymerase chain reaction ( PCR ) -based assays for sensitive and specific detection of S . haematobium DNA in human urine [28–30] and serum [31] samples . However , the PCR-based technologies are not widely used in low-income S . haematobium endemic countries because skilled operators and costly equipment are needed . In this way , the loop-mediated isothermal amplification ( LAMP ) assay [32] offers a field-friendly alternative to PCR-based technologies as it is less time consuming than PCR and can be performed using a simple heating block or water bath , with results read by the naked eye under natural or UV light [33 , 34] . Additionally , LAMP reagents can be storage at room temperature for weeks [35] , the reaction shows low susceptibility to typical inhibitory compounds occurring in samples [36–38] , its robustness against variation of reaction conditions such as pH and temperature has been described [39] and it can operate with minimal handling and processing of DNA samples for amplification [40] , [41–43] , or even without prior DNA extraction [36] . Thereby , considering these salient advantages over most DNA-based amplification tests , LAMP technology shows a potential use in clinical diagnosis and surveillance of infectious diseases , particularly under field conditions for most NTDs [44 , 45] . Several successful approaches for LAMP assay for Schistosoma spp . detection have been recently reported in laboratory settings using experimentally infected animals , such as S . japonicum in rabbits [46 , 47] or S . mansoni in mice [48] , as well as in field settings for monitoring infected snails with S . mansoni , S . haematobium [49 , 50] and S . japonicum [51 , 52] . Additionally , a LAMP to detect S . japonicum in human sera has been also reported [53] . Thus , with the aim to develop new , applicable and cost-effective molecular tools for the diagnosis of urogenital schistosomiasis , in our work we have developed a new sensitive and specific LAMP assay for detection of S . haematobium in human urine samples . In this study , the effectiveness of the LAMP assay was evaluated in a number of patients´ urine samples with parasitological proven infection with S . haematobium . Different fractions of urine samples ( whole urine , supernatants and pellets ) as well as different methods for DNA extraction were used to compare results and cost-effectiveness . To the best of our knowledge , this is the first report using LAMP assay for detection of S . haematobium in human urine samples .
Human urine samples used in this study were obtained as part of public health activities at Hospital Universitario Insular , Las Palmas de Gran Canaria , Spain . Later , samples were sent and stored at CIETUS , University of Salamanca , Spain , for further analyses . Human urine samples were not collected specifically for this study and all were obtained under written informed consent and coded and tested as anonymous samples . Participation of healthy urine donors for obtaining simulated artificial urine samples was voluntary . All participants were given detailed explanations about the aims , procedures and possible benefit of the study . The study protocol was approved by the institutional research commission of the University of Salamanca . Ethical approval was obtained from the Ethics Committee of the University of Salamanca ( protocol approval no . 48531 ) . In the first procedure for DNA extraction we used the i-genomic Urine DNA Extraction Mini Kit ( Intron Biotechnology , UK ) following the manufacturers´ instructions . DNA samples thus obtained were stored at -20°C until use in LAMP reactions . In the second procedure , we used the hot NaOH extraction method [54] with minimal modifications in the standard protocol by adding sodium docecyl sulfate ( SDS ) to ensure disruption of the S . haematobium eggs to release the DNA . Briefly , an equal volume of a 50 mM NaOH solution containing 0 . 1% of SDS was added to urine aliquots of 100 μl and then heated at 95°C for 30 min . Subsequently , the tubes were centrifugated at 5000 rpm for 5 min and a volume of 50 μL of supernatant was recovered in a new clean tube and mixed with an equal volume of a 1 M Tris-HCl solution at pH 8 . 0 . Each new solution thus obtained was stored at -20°C until further use as template in LAMP assays . In the third procedure , -named the “Rapid-Heat LAMP method”- , each aliquot of whole urine , supernatant and pellet obtained from each urine sample was heated at 95°C for 15–20 min and then briefly spun to pellet the debris . After this , 2 μL of the supernatant were used immediately as template for each LAMP reaction . The remaining volume of each sample was stored at -20°C . To obtain DNA to be used as template in LAMP reactions to test the remaining 76 clinical urine samples included in the study , we firstly obtained the urinary sediment ( pellet ) as already indicated and , subsequently , the Rapid-Heat LAMP method was applied . A set of six oligonucleotide primers were used for the LAMP assay , targeting eight regions in the 2522 base pair ( bp ) sequence of S . haematobium ribosomal intergenic spacer ( IGS ) DNA retrieved from GenBank ( Accession No . AJ223838 ) [55] . The outer forward primer ( F3 ) , outer backward primer ( B3 ) , forward inner primer ( FIP ) , backward inner primer ( BIP ) , and loop forward ( LF ) and backward ( LB ) primers were designed using the online Primer Explorer V4 software ( Eiken Chemical Co . Ltd , Tokyo , Japan; http://primerexplorer . jp/elamp4 . 0 . 0/index . html ) according to the general criteria described by Notomi et al . [32] and finally selected based on the criteria described in “A Guide to LAMP primer designing” ( http://primerexplorer . jp/e/v4_manual/index . html ) . The location and nucleotide sequences of the six primers are shown in Fig 1 . All the primers were of HPLC grade ( Thermo Fisher Scientific Inc . , Madrid , Spain ) . To confirm the specificity for the designed primers in annealing exclusively with the S . haematobium DNA correct target sequence , a BLASTN local search and alignment analysis [56] was carried out in different online databases against currently available nucleotide sequences for other organisms ( NCBI; http://blast . ncbi . nlm . nih . gov/Blast . cgi ) as well as specifically against human , murine ( Ensembl; http://www . ensembl . org/Multi/Tools/Blast ) and other related Schistosoma species genomes ( Sanger Institute; http://www . sanger . ac . uk/resources/software/blast/ ) . The outer LAMP primer pair , designated F3 and B3 , was initially tested for the amplification of S . haematobium DNA by a touchdown-PCR ( TD-PCR ) to verify whether the correct target was amplified . The PCR assay was conducted in 25 μL reaction mixture containing 2 . 5 μL of 10x buffer , 1 . 5 μL of 25 mmol/L MgCl2 , 2 . 5 μL of 2 . 5 mmol/L dNTPs , 0 . 5 μL of 100 pmol/L F3 and B3 , 2 U Taq-polymerase and 2 μL ( 1 ng ) of DNA template . Conditions for TD-PCR amplification were as follows: an initial denaturation was conducted at 94°C for 1 min , followed by a touchdown program for 15 cycles with successive annealing temperature decrements of 1 . 0°C every 2 cycles . For these 2 cycles , the reaction was denatured at 94°C for 20 s followed by annealing at 58°C-55°C for 20 s and polymerization at 72°C for 30 s . The following 15 cycles of amplification were similar , except that the annealing temperature was 54°C . A final extension was performed at 72°C for 10 min . The specificity of PCR using outer primers F3 and B3 was also tested with 20 heterogeneous DNA samples from other parasites included in the study . The sensitivity of the PCR was also assayed to establish the detection limit of S . haematobium DNA with 10-fold serial dilutions ranging from 0 . 5 ng/μL to 0 . 5 atg/μL prepared as mentioned above . The assays were performed with 2 μL of the diluted template in each case , thus resulting a final concentration of DNA ranging from 1 ng/μL to 1 atg/μL . Negative controls ( ultrapure water instead of DNA template ) were included in each run . The PCR products ( 5–10 μL ) were subjected to 2% agarose gel electrophoresis stained with ethidium bromide and visualized under UV light . To evaluate the LAMP primer set designed in S . haematobium DNA amplification , we set up the reaction mixture using Bst 2 . 0 WarmStart DNA polymerase ( New England Biolabs , UK ) combined with different betaine ( Sigma , USA ) and MgSO4 ( New England Biolabs , UK ) concentrations . Thus , LAMP reactions mixtures ( 25 μL ) contained 1 . 6 μM of each FIP and BIP primers , 0 . 2 μM of each F3 and B3 primers , 0 . 4 μM of each LB and LF primers , 1 . 4 mM of each dNTP ( Bioron ) , 1x Isothermal Amplification Buffer -20 mM Tris-HCl ( pH 8 . 8 ) , 50 mM KCl , 10 mM ( NH4 ) 2SO4 , 2 mM MgSO4 , 0 . 1% Tween20- ( New England Biolabs , UK ) , betaine ( ranging 0 . 8 , 1 or 1 . 2 M ) , supplementary MgSO4 ( ranging 4 , 6 or 8 mM ) and 8 U of Bst 2 . 0 WarmStart DNA polymerase with 2 μL of template DNA . To establish the standard protocol for LAMP reactions mixtures assayed , a range of temperatures ( 61 , 63 and 65°C ) was tested in a heating block for 30 , 50 or 60 min and then heated at 80°C for 5–10 min to inactivate the enzyme and thus to terminate the reaction . Then , both optimal temperature and incubation time were determined and used in the following tests . Positive ( S . haematobium DNA ) and negative ( no DNA template ) controls were always included in each LAMP assay . To estimate the accuracy of the LAMP assay as a diagnostic test , the percentages of sensitivity , specificity , positive predictive value ( PPV ) and negative predictive value ( NPV ) were calculated using the MedCalc statistical program version 15 . 2 . 2 ( MedCalc Software , Ostende , Belgium ) according to the software instruction manual ( www . medcalc . org ) .
To confirm that the expected target was amplified , a PCR reaction was performed using outer primers F3 and B3 to amplify S . haematobium DNA . Thus , a 199 bp amplicon was successful obtained ( Fig 2A ) . In order to determine the lower detection limit of the PCR reaction , a 10-fold serial dilution ranging from 10−1 to 10−9 of S . haematobium DNA was amplified . The minimum amount of DNA detectable by PCR using outer primers was 1 ng ( Fig 2B ) . According to specificity , when DNA samples obtained from other parasites included in the study were subjected to this PCR assay , amplicons were never amplified ( Fig 2C ) . To establish a standard procedure for the LAMP assay we used the Bst 2 . 0 WarmStart DNA polymerase applying a range of temperatures ( 61 , 63 and 65°C ) for testing different mixtures containing variable concentrations of betaine ( ranging 0 . 8 , 1 or 1 . 2 M ) combined with supplementary variable concentrations of MgSO4 ( ranging 4 , 6 or 8 mM ) in a heating block for 30 , 50 and 60 min . The best amplification results were obtained when the reaction mixture contained 1 M of betaine combined with supplementary 6 mM of MgSO4 ( resulting a final concentration of 8 mM MgSO4 in 1x Isothermal Amplification Buffer ) and was incubated for 50 min at 63°C in a heating block ( Fig 3A ) . Once the most favourable conditions and molecular components were established for the LAMP assay , all positive results in subsequent reactions could be clearly visually observed by the naked eye by inspecting the colour change after adding SYBR Green I as well as the typical ladder of multiple bands after electrophoresis on agarose gels . To determine the specificity of the primers designed , a panel of 20 DNA samples from other parasites were subjected to the LAMP assay . As shown in Fig 3B , only LAMP products were amplified when S . haematobium DNA was used as template and no false positive amplification was observed , thus indicating the high specificity of the established LAMP assay . Regarding to the sensitivity of the LAMP assay , a 10-fold serial dilution of S . haematobium genomic DNA was amplified by LAMP . The results indicated that the detection limit for the LAMP reaction was 100 fg ( Fig 3C ) . This suggested that the LAMP assay is 104 times more sensitive than the PCR using outer primers F3 and B3 ( see Fig 2B ) . On the other hand , the sensitivity of LAMP assay in simulated fresh human urine samples artificially contaminated with DNA from S . haematobium was also examined . In this case , the detection limit of LAMP assay was 10 fg/μL when performing the DNA extraction with the commercial kit ( Fig 4A ) , whereas the detection limit was established in 1 fg/μL using the Rapid-Heat LAMPellet method for DNA extraction ( Fig 4B ) . Comparative LAMP results obtained when testing aliquots of whole urine , supernatants and pellets from patients´ urine samples with parasitological confirmed S . haematobium infection after using the three different DNA extraction methods attempted in our study are shown in Figs 5 , 6 and 7 , respectively . In LAMP tests using a starting volume of whole patients´ urine samples of 100 μL/each we obtained 15/18 positive results when performing DNA extraction using the i-genomic Urine DNA Extraction Mini Kit ( Fig 5A ) , 11/18 when using the NaOH/SDS extraction method ( Fig 5B ) and 12/18 when the Rapid-Heat LAMP method was applied ( Fig 5C ) . In LAMP tests for supernatant fraction of patients´ urine samples we obtained only 3/18 positive results when performing DNA extraction using the i-genomic Urine DNA Extraction Mini Kit ( Fig 6A ) , 4/18 when using the NaOH/SDS extraction method ( Fig 6B ) and 9/18 when the Rapid-Heat LAMP method was applied ( Fig 6C ) . Finally , in LAMP tests for the urinary sediment ( pellet ) obtained from the urine samples we obtained 17/18 positive results when performing DNA extraction using the i-genomic Urine DNA Extraction Mini Kit ( Fig 7A ) , 15/18 when using the NaOH/SDS extraction method ( Fig 7B ) and a total of 18/18 when the Rapid-Heat LAMP method was applied ( Fig 7C ) . Thus , in general , the higher effectiveness in LAMP amplification of S . haematobium DNA in patients´ urine samples was obtained when the urinary sediment ( pellet ) was used for DNA extraction; moreover , the simple Rapid-Heat LAMP method provided the best results of the three methods assayed for extracting DNA detectable by LAMP . Thereby , the minimal pellet obtained from urine samples , in addition to the Rapid-Heat LAMP method for DNA detection-hereafter "Rapid-Heat LAMPellet method"- , was set up as the most advantageous procedure to be used in successive LAMP reactions to detect S . haematobium DNA in urine samples and to test all the clinical samples included in our study . The results of all 94 patients´ urine samples evaluated by duplicated for S . haematobium DNA detection by using the Rapid-Heat LAMPellet method are presented in Table 1 . We obtained LAMP positive results in 18/18 confirmed S . haematobium infected urine samples , in 1/9 urine samples with other helminths species confirmed infections ( specifically a patient infected with a "hookworm" ) , in 1/5 urine samples with other agents confirmed infections ( specifically a patient infected with Trichomonas vaginalis ) , in 1/15 urine samples from patients with eosinophilia without a confirmed diagnosis and , finally , in 5/24 urine samples from patients without either eosinophilia and none apparent disease . The seven parasitological S . mansoni-positive urine samples as well as the 16 urine samples from healthy non-endemic donors ( used as negative controls samples ) were all negative by LAMP . All positive results could be visually observed in tubes by color change after adding SYBR Green I and also after electrophoresis on agarose gels as a ladder of multiple bands of different sizes ( S1 Fig ) . Considering the results obtained , diagnostic parameters for sensitivity and specificity were calculated for our LAMP assay , 100% sensitivity and 86 . 67% specificity , and also for microscopy detection of eggs in urine samples , 69 . 23% sensitivity and 100% specificity . The PPV and NPV for both LAMP assay and microscopy were also calculated; all statistic data obtained are showed in Table 1 .
Urogenital schistosomiasis due to S . haematobium remains a serious underestimated public health problem , particularly in sub-Saharan Africa . Frequency of urogenital schistosomiasis in travellers , expatriates and migrants is in the same range to that of intestinal schistosomiasis due to S . mansoni [57] . As there is no vaccine to protect against schistosomal infection , mass praziquantel treatment of populations at risk of infection is being conducted routinely in many endemic areas; however , reinfections rapidly occur because of recurrent direct contact with water infected with parasites [58] . Considering the current problems of parasitological , serological and molecular methods in detecting schistosomal infections [59] , new , simple , accurate and affordable diagnostic tools are essential for providing specific treatment and for maximizing the success of control of urogenital schistosomiasis in endemic areas; as well as for monitoring drug effectiveness . Point-of-care tests are being developed as economic evaluation diagnostic technologies for infectious diseases control strategies as they are easy to use and interpret , require minimal laboratory infrastructure , are cost-effective , reduce patient waiting time and potentially therefore reduce loss to follow-up , and may have comparable or higher sensitivity to microscopy [60] . The LAMP technology-as a DNA amplification method- combines rapidity , simplicity and high specificity [32] and has a wide range of possible applications , including point-of-care testing in developing countries [61 , 62] . We have developed a LAMP assay for rapid , sensitive , specific and cost-effective detection of S . haematobium in human urine samples , even in the absence of parasites eggs in excreta , as a basis for a potential field diagnostic tool for use in schistosomiasis endemic areas . Besides its excellent performance , the most striking results of this study are the simplicity to perform the whole process without requiring DNA extraction from a small volume of starting urine to get the urinary sediment ( pellet ) to carry out the molecular analysis . We have named this simple procedure the "Rapid-Heat LAMPellet method" . To accomplish its development , we designed a specific set of six primers targeting eight regions in a species specific sequence of S . haematobium ribosomal IGS [55] . The ribosomal IGS regions within Schistosoma species generally contain unique sequence motifs which are specific to that group of organisms . In addition , the IGS target locus has been already used for successful detection of Schistosoma spp . infection in freshwater snails by real-time PCR and oligochromatographic dipstick rapid technology ( PCR-OC ) [63] . Several other advantages of these sequences to be use in molecular studies have been already reported elsewhere [55 , 64] . Once the primer set was designed , in silico comparisons of the expected 199 bp sequence with the on line available genomes showed the higher homology in alignment length with S . haematobium and no cross-reaction was found , specifically with S . mansoni; this result is especially important as these two species are the main schistosomes producing co-infections in most areas of sub-Saharan Africa [58] . Specificity results obtained in in silico were later verified by PCR using outer primers F3-B3 . After this , we attempted to establish the most suitable reaction mixture for the six specific primers in the LAMP assay . We used the Bst polymerase 2 . 0 WarmStart as this warm-start version has several advantages compared to wild-type Bst DNA polymerase large fragment , such as faster in obtaining amplification signals [65] and increased stability at room temperature [66] . These features are important when testing a large number of samples under field conditions in endemic areas where limited resources for the maintenance of a cold chain exists . As the LAMP reaction might be facilitated by the addition of loop primers [67] our LAMP assay designed was accelerated by the addition of a pair of loop primers , thus allowing to amplify successfully S . haematobium DNA in only 50 min , whereas a previously described LAMP assay to amplify S . haematobium DNA in freshwater snails takes 120 min to complete the reaction [49 , 50] . The specificity of the LAMP assay was determined using a panel of heterogeneous control DNA samples of a number of parasites . The assay specifically produced typical ladder patterns from the target sequence only for S . haematobium DNA . The sensitivity of the LAMP resulted 104 times greater than that of PCR using outer primers ( 100 fg vs . 106 fg or 1ng , respectively ) . It is usually considered that LAMP is highly sensitive compared to conventional PCR methods and other studies also found a higher sensitivity when comparing LAMP results in contrast to PCR in amplification of DNA from Schistosoma species , including S . japonicum [47] , S . haematobium and S . mansoni [49 , 48] . The effectiveness of our LAMP assay was assessed in patients´ urine samples with confirmed S . haematobium infection by microscopic examination . Bearing in mind a potential easy and cost-effective large-scale application in field conditions , we evaluated different DNA extraction methods for their ability to isolate DNA from small volumes of different fractions of human urine samples , including whole urine , urine supernatant and urinary sediment ( pellet ) to compare results . A simple , quick and economically DNA extraction method for use in combination with small volumes of clinical urine specimens could greatly reduce the infrastructure requirements of collecting , handling , storing and processing the patients´ samples in schistosomiasis endemic areas where limited resources exist . The three different DNA extraction methods tested in our work were much more efficient in extracting detectable DNA by LAMP when using aliquots of whole urine and pellets than supernatants . This seems to be logical since after centrifugation to remove and retain supernatants , both potential free S . haematobium DNA and parasite eggs-and therefore containing DNA- found in patients´ urine samples should be concentrated at the bottom of the tube , thus improving the sensitivity of the DNA molecular detection methods , as previously described [68] . When using the pellets , the simple rapid-heating method allowed us to obtain a very good-quality detectable DNA that did not compromise LAMP amplification and all the S . haematobium-positive urine samples tested were successfully amplified . The consistent results in DNA obtained from aliquots of whole urine and pellets when applying a commercial kit may be due to the well-known effectiveness of this procedure to isolate genomic DNA from urine samples suitable for further molecular analyses [69] . Urine specimens contain many inhibitors which may interfere in DNA amplification [70] , so removing inhibitors as much as possible by using a kit is convenient to ensure that DNA will be subsequently efficiently amplified . However , since this procedure could be very expensive to use when a large number of samples must be tested , an inexpensive and simple rapid-heating method is much more advantageous . It is also known that DNA purification from samples could be omitted in LAMP reactions , since LAMP assays have shown a significant tolerance to inhibitor substances derived from a number of biological samples [71] , [72] , [73] . Additionally , other LAMP assays with high sensitivity and no complicate requirement procedure for DNA extraction have been developed for molecular detection and diagnostic of bacterial [74] and parasitic [75] diseases in urine samples . Moreover , a simple heating DNA obtaining method has been also successfully applied with other clinical samples , such us blood [41] and swaps [42] in LAMP amplification of both Plasmodium and Leishmania species nucleic acids , respectively . To really establish the sensitivity of our LAMP assay in urine samples that most closely resembled the patients´ urine specimens analyzed , we used a panel of simulated human urine samples artificially spiked with S . haematobium genomic DNA . For these samples , to extract DNA as template in LAMP we used both the commercial kit and the rapid-heat methods since these procedures showed the highest efficiency to obtain detectable DNA by LAMP in S . haematobium-positive clinical samples . After extracting DNA with the commercial kit , LAMP detection limit resulted tenfold higher than that obtained using S . haematobium genomic DNA 10-fold serially diluted without DNA extraction ( 10 fg vs . 100 fg , respectively ) . Unexpectedly , when heating the simulated samples , we obtained a limit of detection tenfold higher than that obtained when using purified DNA samples by the commercial kit ( corresponding to 1 fg vs . 10 fg , respectively ) . An increased sensitivity has been also reported when using crude DNA extraction methods compared with a commercial method ( i . e . DNazol ) for template preparation from the pellets or supernatants of nasopharyngeal aspirates for LAMP detection of adenovirus [76] . Thus , the sensitivity value of 1 fg was considered as the lower limit of the detection threshold of the LAMP assay in detecting S . haematobium DNA in human urine samples . By reference , as S . mansoni genome contains approximately 580 fg of DNA [77] , theoretically our LAMP assay would detect S . haematobium diluted DNA in urine samples corresponding to less than the equivalent to a single parasite cell . Such sensitivity is a feature of great value to overcome the difficulties of detecting urogenital schistosomiasis in areas of low transmission or in individual cases with a very low worm burden . Then , taking into account both the high sensitivity and the good-quality detectable S . haematobium DNA by LAMP in easy to obtain and handling heated pellets from clinical urine samples , we tested the remaining 76 specimens included in our study by the Rapid-Heat LAMPellet method . We obtained negative results by LAMP in all parasitologically S . mansoni-positive urine samples tested ( corroborating again that no cross-reaction with that schistosome species occurs ) and also in urine samples from healthy non-endemic donors used as negative controls . Nevertheless , eight LAMP positive results were obtained when testing patients´ urine samples from other groups which were formerly microscopy-confirmed as S . haematobium-negative . It may be rational to consider that those eight LAMP positive results are truly S . haematobium-infected samples which were undetected in the microscopic analysis since this method is very low sensitive , especially in low-grade infections and high day-to-day variable . Regarding the two LAMP positive results in patients´ urine samples with other microscopy-confirmed infectious diseases ( i . e . hookworm and T . vaginalis ) , it is not uncommon to find co-infections of S . haematobium with other organisms such as bacteria , protozoa and helminths , including the hookworms [78] . It is unlikely that this result is due to a cross-reaction with hookworm since we obtained LAMP negative results in other three patients´ urine samples with microscopy-confirmed infection with this geohelminth . One eosinophilic without confirmed diagnosis patient as well as five non-eosinophilic without apparent pathologic disease individuals had S . haematobium-positive results by LAMP . The presence of absence of eosinophils is usually used as a biomarker for helminthic infections , including schistosomiasis [79]; however , it is not predictive of Schistosoma species infection and may generate inconsistent results [80] . Thus , application of our LAMP method may improve the identification of cases with low-intensity infections as well as in cases which did not pass eggs in urine samples , thus revealing infections in people frequently presumed to be uninfected . Finally , although all patients´ urine samples were tested in duplicate with the same result , it would be very interesting to know how reproducible the technique is when testing in field settings as well . In conclusion , we have demonstrated that simply rapid-heating urinary pellets for good-quality DNA extraction was effective for use in LAMP assays with regard of detecting S . haematobium in clinical urine samples . This procedure has been named the Rapid-Heat LAMPellet method and it would be well-suited to diagnose urogenital schistosomiasis in resource-limited endemic regions because of its rapidity , easy handling , cost-effectiveness and both high detection specificity and sensitivity . The next step for refining the assay by conducting a field evaluation in an endemic setting should be desirable .
|
Human schistosomiasis is a disease caused by several species of parasitic worms of the genus Schistosoma that is affecting 200 million people , especially in sub-Saharan Africa . Most people are infected with Schistosoma haematobium , the species that causes urogenital schistosomiasis and also bladder cancer in many chronic infections . The definitive diagnostic test is based on microscopic examination of urine samples to detect parasite eggs . This method has low sensitivity , high day-to-day variability and cannot be carried out in the acute phase of the disease since the parasite has not started yet to lay eggs . New high-throughput diagnostic tools would be desirable , permitting early treatment and preventing the pathology associated with chronic infections . An interesting approach is the loop-mediated isothermal amplification ( LAMP ) technique because of its simplicity in operation and potential use in clinical diagnosis and surveillance of infectious diseases . In this study , we developed and evaluated a LAMP assay for detection of S . haematobium DNA in patients´ urine samples using heated pellets with no complicated requirement procedure for DNA extraction , namely the Rapid-Heat LAMPellet method . This is a new , easy , rapid and cost-effective LAMP method that should prove useful for mass screening in limited-resource settings in urogenital schistosomiasis-endemic areas .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[] |
2015
|
The Rapid-Heat LAMPellet Method: A Potential Diagnostic Method for Human Urogenital Schistosomiasis
|
Neuronal cargos are differentially targeted to either axons or dendrites , and this polarized cargo targeting critically depends on the interaction between microtubules and molecular motors . From a forward mutagenesis screen , we identified a gain-of-function mutation in the C . elegans α-tubulin gene mec-12 that triggered synaptic vesicle mistargeting , neurite swelling and neurodegeneration in the touch receptor neurons . This missense mutation replaced an absolutely conserved glycine in the H12 helix with glutamic acid , resulting in increased negative charges at the C-terminus of α-tubulin . Synaptic vesicle mistargeting in the mutant neurons was suppressed by reducing dynein function , suggesting that aberrantly high dynein activity mistargeted synaptic vesicles . We demonstrated that dynein showed preference towards binding mutant microtubules over wild-type in microtubule sedimentation assay . By contrast , neurite swelling and neurodegeneration were independent of dynein and could be ameliorated by genetic paralysis of the animal . This suggests that mutant microtubules render the neurons susceptible to recurrent mechanical stress induced by muscle activity , which is consistent with the observation that microtubule network was disorganized under electron microscopy . Our work provides insights into how microtubule-dynein interaction instructs synaptic vesicle targeting and the importance of microtubule in the maintenance of neuronal structures against constant mechanical stress .
Microtubule and molecular motors mediate polarized transport of neuronal proteins to either axons or dendrites [1] . Microtubules are oriented uniformly with their plus ends towards the distal end of the axon , which facilitates kinesin-dependent targeting of presynaptic proteins [2] . By contrast , targeting of postsynaptic molecules to the dendrite , such as glutamate receptors , requires the minus end-oriented dynein motors [3] , consistent with the fact that many microtubules in the dendrites orient the minus end distally [4] . Mislocalization of presynaptic proteins to the dendrite occurs when this polarized pattern of axon-dendritic microtubule arrays is disrupted [5] , [6] , when kinesin function is compromised [7] , [8] , or when dynein activity is inadvertently increased [3] , [7] , [8] . Synaptic vesicle ( SV ) precursors are generated in the neuronal cell body and transported to the synapses by the unidirectional motor Kinesin 3/KIF1A [1] . On the other hand , the dynein motor complex mediates retrograde SV transport in the axon [1] , [9] . Since SVs are cargos for both KIF1A and dynein , it is intriguing that they are exclusively targeted to the axon and prevented from entering the dendrites . Previous biochemical and structural studies suggest that kinesin and dynein share an overlapping binding region at the C-terminus of α-tubulin [10] . The N-terminus of the H12 helix of the α-tubulin contains a stretch of absolutely conserved acidic residues ( 414EEGE , equivalent to 415EEGE in the yeast α-tubulin ) and interacts with ATP-bound KIF1A [11] . A recent study on dynein structures also implicates this region in the interaction between microtubule and the microtubule-binding domain ( MTBD ) of dynein [12] , although validation of this model in the context of in vivo , eukaryotic system is still lacking . Mutations of any of the three glutamic acids in the yeast α-tubulin to alanine dramatically reduced the frequency of kinesin binding to the microtubules [13] . Mutations of several conserved , acidic residues in the H12 helix of β-tubulin ( E410 , E412 , D417 ) to alanine similarly reduced microtubule affinity for kinesins . Interestingly , E410K , D417H and D417N in the human β-tubulin TUBB3 , among other point mutations , had been found in patients with congenital neurological syndrome with ophthalmoparesis and peripheral neuropathy [14] . In particular , TUBB3 ( E410K ) and TUBB3 ( D417H ) , but not other disease-related TUBB3 mutations , were shown to impair the bindings of Kinesin 1/KIF5 , Kinesin 3/KIF1A and KIF21 when expressed in cultured mammalian neurons [15] . The interaction between dynein and microtubule was not affected by these TUBB3 mutants [15] . These studies established the critical importance of negative charge in the H12 helix of the α- and β-tubulins in mediating microtubule-kinesin interaction , but the molecular mechanisms governing microtubule-dynein interaction and its physiological significance remain unexplored . Here we describe a novel mutation of G416 in the α-tubulin MEC-12 of C . elegans to glutamic acid ( G416E ) . Homologous mutations at this site of α-tubulin had not been reported in human diseases or tested in genetic model organisms . In C . elegans , mec-12 is highly expressed in the six touch receptor neurons that detect gentle mechanical stimulation on the worm cuticle [16] , [17] . MEC-12 and the touch neuron-specific β-tubulin MEC-7 are required to form the unusual , 15-protofilament giant microtubules in these neurons [16] . These giant microtubules had been implicated in the transduction of mechanosensation , although the mechanisms remain enigmatic [18] . We show that this gain-of-function G416E mutation redirects SVs to non-axon compartment in the C . elegans mechanosensory neuron PLM , and it does so by increasing microtubule affinity for dynein .
In wild-type C . elegans , the bilaterally symmetric touch receptor neurons ALM and PLM develop a single anterior process that forms synapses in the nerve ring and in the ventral nerve cord , respectively ( Figure 1A ) . Touch neuron synapses are enriched in RAB-3- ( + ) synaptic vesicles ( SVs ) , the active zone protein SYD-2/Liprin-α , and mitochondria ( Figure 1B , 1D , and S1 ) [19] , [20] . The PLM neurons also have a short posterior process that does not form synapses . In an EMS mutagenesis screen ( see Materials and Methods ) , we identified gm379 , a mutant with prominent SV phenotypes in the touch neurons ( Figure 1B′-1G′ ) . gm379 animals lacked RAB-3- ( + ) SVs at the touch neuron synapses , and instead SVs accumulated in the neuronal soma ( Figure 1B′-1E′ , 1H ) . We refer to these as SV transport defects for the rest of the paper . Surprisingly , SVs were also redirected to the PLM posterior process , a phenotype that we call SV mistargeting ( Figure 1E′ , 1I ) . These results were confirmed using two other SV reporters , jsIs37 ( Pmec-7::SNB-1::GFP ) that marked the SV membrane protein synaptobrevin/SNB-1 , and jsIs219 ( Psng-1::SNG-1::GFP ) labeling another SV protein synaptogyrin/SNG-1 , with GFP ( Figure 1F-1G′ ) . These ectopic SVs showed very limited motility , and many were stationary ( Figure 1J ) . We followed gm379 mutants through development , and confirmed that SV transport defects and mistargeting were present at early larval stages and progressively worsened ( Figure 1H ) . The transport and targeting of synaptic active zone protein SYD-2 was affected to a much milder degree , and surprisingly , SYD-2 failed to mistarget to the PLM posterior process ( Figure S1 ) . The dissociation in the mutant phenotypes of SV and active zone proteins indicates that the gm379 mutation caused relatively specific defects in SV targeting rather than generally impaired axon transport or induced ectopic synapse formation . In addition to SV transport defects , the gm379 mutant touch neurons had progressive neurite swelling and misshapen soma ( Figure 2A and S2A ) . Neurite defects evolved from small beadings at early larval stages into triangular-shaped swellings in L4 and adult animals , and mitochondria were frequently found to be present at the swellings ( Figure 2A , 2B and S2A ) . These swellings were dynamic in morphology , as movements often induced reversible buckling of the neurite and changed the width and height of the swellings ( Figure 2C and Video S1 ) , which was similar to what had been described earlier for the tubulin acetyltransferase mutant mec-17 [21] . Neurite buckling or swelling were never seen in the wild type even under maximal muscle contraction induced by levamisole . This observation suggests that the gm379 mutation rendered the touch neurite susceptible to deformation under mechanical strain , a phenotype that was also seen when the membrane skeleton protein UNC-70/β-spectrin was lost [22] . We therefore test whether genetic paralysis of the animals suppresses neurite defects of the gm379 mutant . Mutation in the muscle myosin gene unc-54 almost completely paralyzed the animals , and it significantly reduced the number of neurite swellings in the gm379 mutant touch neurons ( Figure 2D and 2E ) . This result implies that the gm379 mutation compromises the ability of the touch neurites to cope with mechanical stress . We were curious whether massive neurite swellings in the gm379 mutant predispose touch neurons to degeneration . We first performed longitudinal imaging of individual ALM and PLM neurons through adulthood , and found that touch neurites in the gm379 mutant underwent progressive disorganization ( Figure 2F ) . Touch neuron degeneration , characterized by swelling of neuronal soma , neurite interruption and extensive beadings , began to emerge in the gm379 mutant at D9 and progressively increased ( Figure 2G and 2H ) . Touch neuron degeneration was extremely rare in the wild type at comparable age ( Figure 2G and 2H ) [23] . Interestingly , touch neuron degeneration of the gm379 mutant was suppressed by the unc-54 mutation ( Figure 2G and 2H ) . These results indicate that recurrent mechanical strain imposed on the touch neurons during locomotion is an important precipitating factor for late-onset neurodegeneration in the gm379 mutant . Under serial thin-section electron microscopy , we found that the characteristic 15-protofilament microtubules of C . elegans touch neurons were preserved in the gm379 mutant , including those in the PLM posterior process ( Figure 3A-C , S2B-D ) . Mitochondria could be found where touch neuron processes swelled abnormally ( Figure 3B ) , consistent with our light microscopic observation ( Figure 2B ) . In longitudinal sections , in contrast to the wild type , where neuronal microtubules formed long straight bundles , touch neuron microtubules in the gm379 animals curved focally at sites of organelle accumulation ( Figure 3D ) . We observed bending and splitting of neuronal microtubule bundles at sites of mitochondria accumulation in focal axonal swellings ( Figure 3D4 ) . These ultrastructural studies suggest that the microtubule network of the touch neurons is abnormal in the gm379 mutant . We cloned gm379 by single nucleotide polymorphism ( SNP ) mapping , complementation test and DNA sequencing , and found that it contained a missense mutation of the touch neuron-specific α-tubulin mec-12 that alters an absolutely conserved C-terminal glycine residue to glutamate ( G416E ) [19] , [17] ( Figure 4A and 4B ) . Interestingly , gm379 animals were touch-insensitive ( 30% touch-sensitive , compared to wild type , 89%; and mec-12 ( e1607 ) null , 12% , n>35 ) . We found another intron mutation in gm379 that was distant to exon-intron junctions ( nucleotide 828 of unspliced transcript , G to A mutation ) . Two null mutants of mec-12 , e1607 and tm5083 , also had SV transport defects , but not SV mistargeting in the PLM or axon swelling in the touch neurons ( Figure 4C , S3 ) . RNAi against mec-12 in the gm379 mutant almost completely abolished axon swelling or SV mistargeting , indicating that these two phenotypes were neomorphic ( Figure 4C , 4D ) . Expression of the MEC-12 ( G416E ) mutant tubulin in the mec-12 null mutants recapitulated the SV mistargeting phenotypes of the mec-12 ( gm379 ) mutant ( 67% of transgenic animals showed SV mistargeting , n = 21 v . s . 6 . 5% of array-loss siblings showing SV mistargeting , n = 77 ) , confirming that SV mistargeting and neurite swelling were indeed caused by the mec-12 ( gm379 ) rather than other unidentified mutations in the background . To gain further insights into the genetic nature of the mec-12 phenotypes , we analyzed various heterozygous mec-12 mutants as well as trans-heterozygotes between these alleles ( Figure 4E ) . Heterozygous animals containing the e1607 , tm5083 or gm379 mutations all showed moderate SV transport defects that were less severe than the homozygous animals . These observations suggest that loss of mec-12 functions causes semi-dominant SV transport defects due to haploinsufficiency . Heterozygous gm379 mutants did not display neurite swelling or SV mistargeting . By contrast , mec-12 ( gm379 ) /mec-12 ( e1607 ) and mec-12 ( gm379 ) /mec-12 ( tm5083 ) trans-heterozygous mutants had SV mistargeting without neurite swelling , implying that the presence of wild-type MEC-12 somehow prevents MEC-12 ( G416E ) from mistargeting SVs or disrupting microtubule organization . In conclusion , our genetic analysis indicates that gm379 causes both neomorphic ( neurite swelling and SV mistargeting ) and semi-dominant loss-of-function ( SV transport defects ) phenotypes of mec-12 . Stable microtubules formed in the gm379 touch neurons , based on the EM data and the preserved lysine 40 ( K40 ) acetylation of α-tubulin [17] , [24] , [25] ( Figure S4A ) . Moreover , a mec-7/β-tubulin null mutation completely suppressed neurite swelling and dramatically reduced SV mistargeting of mec-12 ( gm379 ) ( Figure S5 ) , suggesting that SV mistargeting and neurite swelling phenotypes require intact microtubules . Labeling touch neuron microtubules with the plus end-binding protein EBP-2::GFP showed that in the wild type , microtubules oriented plus-end distally in the anterior PLM process ( Figure S6 ) [26] . In the PLM posterior process , microtubules showed mixed polarity ( Figure S6 ) . These patterns of microtubule polarity were preserved in the mec-12 ( gm379 ) mutant ( Figure S6 ) . A recent study reported that mutations in the tubulin acetyltransferase mec-17 caused neurite degeneration with SV mislocalization in the touch neurons [27] . To test whether altered microtubule posttranslational modifications are responsible for SV mistargeting in the mec-12 ( gm379 ) mutant , we performed immunostaining experiments , but did not observe gross difference in microtubule acetylation or tyrosination in the touch neurons between the wild type and the mec-12 ( gm379 ) mutant ( Figure S4A ) . Tubulin polyglutamylation signal was restricted to the amphid and phasmid sensory cilia as in the wild type , and was not ectopically expressed in the mutant touch neurons ( Figure S4B ) . Moreover , we performed feeding RNAi to knock down the tubulin polyglutamylase ttll-4 , the tubulin deglutamylase ccpp-6 , and a few genes ( ttll-5 , ttll-12 , ttll-15 and ccpp-1 ) that bear sequence homology to human tubulin amino acid ligases ( Wormbase at http://www . wormbase . org ) [28] , [29] , in both wild type and the mec-12 ( gm379 ) mutant . None of these RNAi resulted in SV mistargeting in the wild type or suppressed SV mistargeting in the mec-12 ( gm379 ) animals . Based on these results , we conclude that SV mistargeting in the mec-12 ( gm379 ) is not a consequence of altered microtubule posttranslational modifications . It is possible that the interaction between KIF1A and microtubule was altered by the G416E mutation . We found that the strong loss-of-function unc-104 ( rh43 ) /KIF1A mutation caused completely penetrant SV transport defects and , surprisingly , low percentage of SV mistargeting in the PLM ( Figure 5A and 5B ) . Moreover , this unc-104 mutation enhanced SV mistargeting of mec-12 ( gm379 ) rather than suppressing the phenotype , with more SVs mistargeted to the PLM posterior process and distributed more distally ( Figure 5C , 5D ) . This result suggests that SV mistargeting in the mec-12 ( gm379 ) mutant is not caused by aberrant UNC-104 activity . Furthermore , overexpression of UNC-104 significantly rescued SV transport defects and mistargeting in the mec-12 ( gm379 ) mutant , with more SVs reaching synapses in the nerve ring or entering the anterior ALM and PLM processes ( Figure 5E-5H , Figure S7A , S7B ) . While these data are consistent with the interpretation that UNC-104 activity was reduced in the mec-12 ( gm379 ) mutant , resulting in severe SV transport defects , they also indicate that SV mistargeting in the mutant requires the activity of an unknown molecule . We wondered whether increased activity of the minus end motor dynein is responsible for SV targeting to the PLM posterior process in the mutant , based on the presence of minus end-out microtubules in the PLM posterior process and the unc-104 effects . dhc-1 encodes the heavy chain for cytoplasmic dynein in C . elegans [30] . If enhanced dynein activity is responsible for SV mistargeting in the mutant , elimination of dynein function should suppress it . We could observe SV mistargeted to the PLM posterior process as early as 2-3 fold embryos , before the animal hatched . With the available dhc-1 mutant alleles , it was not possible to lose DHC-1 functions at such early stages without compromising animals' viability . To eliminate DHC-1 functions as early as possible , and to circumvent lethality due to widespread DHC-1 loss , we specifically knocked down dhc-1 in the touch neurons , but not in other somatic tissues , by simultaneously expressing sense and antisense dhc-1 from the mec-7 promoter , which we named transgenic dhc-1 RNAi . Strikingly , transgenic dhc-1 RNAi significantly suppressed SV mistargeting of the mec-12 ( gm379 ) mutant , with about one third of the transgenic animals completely devoid of mistargeted SVs ( Figure 6A , 6B ) . This result was confirmed by another independently generated dhc-1 RNAi array ( Figure S8A ) . Transgenic dhc-1 RNAi also significantly reduced SV mistargeting in the unc-104; mec-12 ( gm379 ) mutant ( Figure 6C ) . In the wild type , transgenic dhc-1 RNAi had little effects on the intensity of GFP::RAB-3 or SNB-1::GFP in the PLM soma or synapses ( Figure S8B ) . These data indicate that SV mistargeting in the mec-12 ( gm379 ) mutant is mediated by the dynein motor . The neurite swelling phenotypes of the mutant , by contrast , were not changed by dhc-1 RNAi , suggesting that SV mistargeting and neurite defects are mechanistically distinct . We noted that eliminating DHC-1 functions in the mec-12 ( gm379 ) mutant significantly restored synaptic targeting and anterograde transport of SVs , with concomitant reduction of SV accumulation in the touch neuron soma ( Figure 6A and 6D ) . These effects were similar to those caused by excess UNC-104 ( Figure 5E-H , S7A , S7B ) . Indeed , synaptic targeting and anterograde transport of SVs were completely abolished in the dhc-1; unc-104; mec-12 ( gm379 ) triple mutant , suggesting that the phenotypic rescue caused by the dhc-1 mutation requires UNC-104 ( Figure 6D ) . Although significantly increased compared to those in the mec-12 ( gm379 ) mutant , SV signals in the nerve ring synapses of the dhc-1; mec-12 ( gm379 ) were still much weaker than the wild type . Together with the severe reduction of presynaptic SVs in the mec-12 ( gm379 ) mutant , these observations suggest that UNC-104 was not able to display a full range of activity on the mutant microtubule scaffolds . To test whether the MEC-12 ( G416E ) mutant microtubules have increased affinity for dynein , we performed microtubule sedimentation in transgenic strains expressing MEC-12 and MEC-7 pan-neuronally , and assayed for the amount of DHC-1 or UNC-104 associated with microtubules by blotting with anti-DHC-1 and anti-UNC-104 antibodies , respectively . The presence of MEC-12-containing , stable microtubules was verified by detecting 6-11B-1 antibody immunoreactivity in neurons other than the touch receptors ( ) . After normalization to tubulin , the amount of DHC-1 co-sedimented with MEC-12 ( G416E ) -containing microtubules dramatically increased , compared to that co-sedimented with wild-type microtubules ( Figure 6E ) . In support of this conclusion , we found that GFP::DHC-1 , expressed from the low-copy integrated transgene orIs17 ( Pdhc-1::gfp::dhc-1 ) , which drives DHC-1 expression from the endogenous promoter , was accumulated in the PLM posterior process in the mec-12 ( gm379 ) , but not in the wild type ( Figure 6F ) . Together these results indicate that mutant microtubules have increased affinity for DHC-1 . To our surprise , we did not detect a change in the amount of UNC-104 co-sedimented with mutant microtubules ( Figure 6E ) or a change of UNC-104 protein localization in the PLM neuron . This implies that UNC-104 can still associate with the mutant microtubules , although its function is somehow disrupted . The aforementioned data indicate that G416 of MEC-12 plays a critical role in determining the relative affinity of microtubules for dynein . To further decipher the mechanisms that instruct microtubule-dynein affinity , we systemically replaced G416 with acidic ( aspartic acid/D ) or basic ( lysine/K , arginine/R ) residues , as well as alanine ( A ) and glutamine ( Q ) , the latter being similar to glutamic acid in side chain length but did not carry charges ( Figure 7A , 7B ) . These MEC-12 species were expressed in the touch neurons of the mec-12 ( e1607 ) null mutant . SV mistargeting was seen only with the expression MEC-12 ( G416D ) , but not other G416 substitutions ( Figure 7A , 7B ) . These results suggest that SV mistargeting in the gm379 mutant was caused by the increased negative charges at the EEGE cluster of MEC-12 . Is the arrangement of the acidic residues important for dynein affinity of microtubules ? To answer this question , we moved the glycine to residue 414 , 415 or 417 , with reciprocal glutamic acid substitution at 416 ( E414G/G416E , E415G/G416E , G416E/E417G , referred as GEEE , EGEE and EEEG , respectively; Figure 7A , 7B ) , so that the arrangement , but not the sum , of negative charges was altered , and asked whether this manipulation affects SV targeting . While expression of MEC-12 ( GEEE ) or MEC-12 ( EGEE ) did not result in significant SV mistargeting , expression of MEC-12 ( EEEG ) triggered SV mistargeting in about 30% of the mec-12 ( e1607 ) animals ( Figure 7A , 7B ) . These observations indicate that SV targeting critically depends on the magnitude and the spatial arrangement of negative charges in the EEGE cluster of the H12 helix . Previous structural studies suggest that dynein binds the H12 helix of α-tubulin [10] , [12] . Redwine et al . proposed that E3378 and R3382 of the dynein microtubule-binding domain ( MTBD ) form intramolecular salt bridge , and upon approaching the tubulin dimer , negative charges of the α-tubulin H12 disrupt this salt bridge by attracting R3382 , which carries positive charges ( Figure 8C ) . An E3378K mutation of MTBD disrupted this intramolecular salt bridge and increased both the affinity and the run length of dynein on the microtubules [12] . We speculate that the E3378K mutation facilitates the electrostatic interaction between MTBD and the negative charges of the α-tubulin H12 domain . To test this , we expressed and purified a fragment of C . elegans DHC-1 MTBD , and showed that this MTBD precipitated with microtubules synthesized from purified bovine tubulin in the in vitro sedimentation experiment ( Figure 8A and S9 ) . Moreover , D3323K mutation , which is equivalent to E3378K mutation of the yeast dynein MTBD , enhanced MTBD-microtubule interaction across a range of tested concentrations ( Figure 8A , 8B and S9 ) . This result supports our hypothesis that exaggerated microtubule affinity for dynein critically depends on the electrostatic interaction between the EEGE-containing H12 helix of α-tubulin and the dynein MTBD ( Figure 8C ) .
In the present study , we characterized a gain-of-function mec-12/tubulin mutant that displayed synaptic vesicle mistargeting and neurite swelling phenotypes , which were absent in the mec-12 null mutant . In this mutant , single amino acid substitution augments microtubule-dynein interaction , thereby mistargeting synaptic vesicles to non-axon compartments . This observation extends previous structural studies on the charged cluster of the α-tubulin H12 helix and provides a biological context for such charge-based coupling between microtubule and dynein . The neurite swelling and degeneration phenotypes could bear important implications for human neurological diseases associated with missense tubulin mutations , as discussed below . Previous structural studies suggest that dynein binds the H12 helix of α-tubulin , and the interaction between dynein and microtubule is not as strong as that between kinesins and microtubule [10] , [12] . One advantage for this suboptimal dynein-microtubule affinity is flexibility and dynamic range for dynein processivity on the microtubule scaffolds . The hypothesis that dynein-microtubule interaction is not optimized also suggests that it is possible to further enhance dynein-microtubule affinity by mutating residues at this interface . Here , we show that this is indeed the case: by increasing negative charges of the α-tubulin H12 helix or augmenting positive charges of the dynein MTBD , microtubule-dynein interaction was strengthened . More importantly , we show that this aberrantly high affinity of microtubule for dynein redistributed synaptic vesicles to ectopic compartments in the neurons . We speculate that the suboptimal affinity of microtubule for dynein prevents initial SV entry into the dendrite , yet allows for retrograde SV transport within the axon . The G416E mutation of MEC-12 increased microtubule affinity for dynein probably at the expense of dynein processivity: we found that motility of the mistargeted SVs in the PLM posterior process was profoundly compromised ( Figure 1J ) . Consistent with this view , Redwine et al . showed that E3378K mutation of MTBD caused about 40% reduction in the velocity of dynein movements on microtubules . By contrast , none of the G416 substitutions restored the UNC-104/KIF1A-dependent , anterograde SV transport in the mec-12 ( e1607 ) null mutant . This is also consistent with the observation by Redwine et al . that microtubule-kinesin interaction was molecularly optimized , therefore changes to the tubulin residues at the microtubule-kinesin interface will only compromise but not compensate for or even further improve this interaction . Rather than merely serving as a track , mounting evidence indicates that microtubules play an active role in directing axon transport [2] . One such example is the various posttranslational modifications on microtubules , which regulate axon transport by restricting different kinesin motors to discrete subcellular compartments [31] , [32] . The intrinsic organization of microtubule is another critical factor regulating polarized axon transport . UNC-33 and UNC-44 , the C . elegans homologs for the Collapsin Response Mediator Protein 2 ( CRMP2 ) and neuronal ankyrin , respectively [33] , [34] , regulate axon transport by maintaining uniform microtubule polarity in the axon and the dendrite [5] . In C . elegans head neurons , microtubules are uniformly oriented with their minus-ends towards the distal of the dendrite . In the unc-33 and the unc-44 mutants , dendritic microtubules showed mixed polarity , which resulted in aberrant sorting of axonal proteins into the dendrite [5] . In these mutants , the dendritic localization of SVs requires UNC-104 . Of note , mistargeting of axonal proteins in the unc-33 and unc-44 mutants occurred to the SVs and multiple active zone components . By contrast , SYD-2 was not mistargeted in the mec-12 ( gm379 ) mutant , and we found that SV transport in the touch neurons was not affected in unc-33 and unc-44 mutants . These observations indicate that while perturbation to the gross architecture of microtubule causes extensive mistargeting of multiple axonal cargos , changes at a restricted , yet functionally critical site of microtubule could lead to targeting defects of specific axonal proteins . Many tubulin mutations associated with human diseases are point mutations that alter protein function rather than eliminating protein products [35]–[37] . Mutant tubulins are still incorporated into microtubule polymers; phenotypes presumably arise from altered microtubule dynamics or disrupted interactions with molecular motors or microtubule-associated proteins [14] , [15] , [38] . The G416E mutation of MEC-12 caused extensive neurite swellings that eventually led to degeneration of the touch neurons . Ultrastructurally , there were unbundling of microtubules and accumulation of mitochondria at focal neurite swellings . Interestingly , these neurite defects were independent of dynein activity , but could be suppressed by genetic paralysis of locomotion . In light of two recent studies that showed touch neurite buckling or swelling when unc-70/β-spectrin or mec-17/tubulin acetyltransferase was mutated [21] , [22] , this suggests that microtubules confer resistance of touch neurites to deformation imposed by constant muscle activity . When this structural resilience is compromised , touch neurons are likely to degenerate presumably in a wear-and-tear fashion induced by the animal's constant movements . Supporting this notion , commissural axons in unc-70 mutants also increasingly break as the animals grow and move , and paralysis of the animals prevents axon interruption [39] . Unlike the mec-17 mutant , in which defective tubulin acetylation was associated with abnormal microtubule protofilament number , tubulin acetylation and microtubule protofilaments were unaffected in the mec-12 ( gm379 ) mutant . We hypothesize that the G416E substitution may alter the association of microtubule-binding proteins with microtubules , changing the stability or structure of microtubule lattice and rendering the neurites susceptible to mechanical strain . It would be interesting to test whether any of the tubulin mutations found in human diseases generates similar axon defects and could also be ameliorated by reduced activity of neighboring musculature . The molecular mechanism by which G416E mutant microtubules cause axon degeneration awaits future investigation .
Strains were cultured as described [40] . The following alleles were used in this study: N2 ( Bristol strain ) , CB4856 , LG I: dhc-1 ( or283ts ) , unc-54 ( e190 ) ; LG II: unc-104 ( rh43 ) ; LG III: mec-12 ( e1607 ) ( a gift from Martin Chalfie , Columbia University ) , mec-12 ( tm5083 ) , mec-12 ( gm379 ) ; LG V: sid-1 ( pk3321 ) ; him-5 ( e1490 ) ; LG X: mec-7 ( ok2152 ) . Transgenes used in the current study are: jsIs37 ( Pmec-7::SNB-1::GFP ) /IV , jsIs219 ( Psng-1::SNG-1::GFP ) /II , jsIs821 ( Pmec-7::GFP::RAB-3 ) /X , jsIs973 ( Pmec-7:mRFP ) /III , jsIs1111 ( Pmec-4::UNC-104::GFP ) , jsIs1238 ( Pmec-7::SYD-2::GFP ) ( jsIs821 , jsIs973 , jsIs1111 , and jsIs1238 are gifts from Michael Nonet , Washington University ) , juIs76 ( Punc-25::GFP ) /II , Punc-17::RFP/V ( a gift from Joshua Kaplan , Massachusetts General Hospital ) , orIs17[Pdhc-1::GFP::DHC-1 , unc-119 ( + ) ] ( a gift from Bruce Bowerman , University of Oregon ) , otIs118 ( Punc-33::GFP ) /IV , uIs71 ( Pmec-18::SID-1 , Pmyo-2::mCherry ) , zdIs5[Pmec-4::GFP , lin-15 ( + ) ]/I , twnEx8 ( Pmec-7::TOMM20::mCherry , Pmyo-2::gfp ) ( “Pmec-7::mito::mCherry” ) , twnEx40 ( Pmec-7::GFP::EBP-2 , Pdpy-30::dsRed ) , twnEx42[Pmec-7::dhc-1 ( RNAi ) , Pmyo-2::GFP] , twnEx55[Pmec-7::MEC-12 ( G416E , E417G ) , Pdpy-30:: NLS::dsRed] , twnEx73[Pmec-7::MEC-12 ( G416E ) 5 ng/µl , Pdpy-30::NLS::dsRed] , twnEx74[Pmec-7::MEC-12 ( G416D ) , Pdpy-30:: NLS::dsRed] , twnEx75[Pmec-7::MEC-12 ( G416Q ) , Pdpy-30:: NLS::dsRed] , twnEx76[Pmec-7::MEC-12 ( G416A ) , Pdpy-30:: NLS::dsRed] , twnEx77[Pmec-7::MEC-12 ( G416K ) , Pdpy-30:: NLS::dsRed] , twnEx78[Pmec-7::MEC-12 ( G416R ) , Pdpy-30:: NLS::dsRed] , twnEx79[Pmec-7::MEC-12 ( E414G , G416E ) , Pdpy-30:: NLS::dsRed] , twnEx80[Pmec-7::MEC-12 ( E415G , G416E ) , Pdpy-30:: NLS::dsRed] , twnEx88[Pmec-7::MEC-12 ( G416E ) 10 ng/µl , Pdpy-30::NLS::dsRed] , twnEx89[Pmec-7::dhc-1 ( RNAi ) , Pdpy-30:: NLS::dsRed] , twnEx98[Punc-119::MEC-12 , Punc-119::MEC-7 , unc-119 ( + ) ] , twnEx99[Punc-119::MEC-12 ( G416E ) , Punc-119::MEC-7 , unc-119 ( + ) ] , Ex ( Punc-104::UNC-104::GFP ) , Ex ( Punc-104::UNC-104::mRFP ) ( both from Oliver Wagner , National Tsing-Hua University , Taiwan ) . mec-12 ( e1607 ) is a G to A point mutation at nucleotide 430 of mec-12 cDNA , resulting in glycine to serine mutation at amino acid 144 . Germ line transformation was performed by microinjection of purified DNA of interest as described [41] . Initially , an EMS mutagenesis screen was performed in the zdIs5; cwn-1 ( ok546 ) animals to identify mutations that cause synthetic polarity defects of the touch neurons [42]–[44] . While causing no defects in neuronal polarity , gm379 was recovered due to its prominent axonal defects . The cwn-1 mutation was then removed from the mutant . The SV transport defects , SV mistargeting and axonal swellings were all independent of the cwn-1 ( ok546 ) mutation in the background , and the cwn-1 mutant displayed none of the gm379 phenotypes . Single nucleotide polymorphism mapping was performed as described [45] . zdIs5 was included in this mapping to assist the identification of the homozygous gm379 mutants . In brief , male animals of the Hawaiian strain CB4856 were crossed to zdIs5; gm379 , and gm379 homozygotes were later recovered from F2 progeny . F3 animals from individually cloned F2 animals were washed off plates and genomic DNA extracted by proteinase K treatment . To map gm379 , we selected 48 SNPs from the 5 autosomes and the sex chromosome , and semi-quantitatively determined the ratio of N2/Hawaiian SNP for each locus using the restriction enzyme DraI . Our SNP mapping located gm379 to a region between -12 and +7 MU of Chromosome III , a region that contains the mec-12 locus . Feeding RNAi was performed as described [46] , with 1 mM IPTG pre-induction for 2 hours . For touch neuron-specific RNAi , we used jsIs973 ( Pmec-7::mRFP ) mec-12 ( gm379 ) ; sid-1 ( pk3321 ) ; jsIs821 ( Pmec-7::GFP::RAB-3 ) ; uIs71 ( Pmec-18::SID-1 , Pmyo-2::mCherry ) animals [47] . The only RNAi-sensitive cells in the sid-1; uIs71 background are the six touch receptor neurons . Five L4 animals were placed on the RNAi plates and cultured at 20°C . The F1 progeny of were then transferred to another freshly prepared RNAi plates at L4 , and their progeny ( F2 ) scored for axon and synaptic vesicle phenotypes . Each RNAi experiment was repeated three times to confirm the results . Efficiency of feeding RNAi against neuronal genes in this genotype was confirmed by mec-12 RNAi , which resulted in 55% animals losing the PLM branch ( n = 22 ) , with control RNAi having no effects ( 0% , n = 27 ) . This penetrance was similar to what was observed in the mec-12 ( e1607 ) null mutant ( 58% , n = 60 ) . Additional RNAi control included mec-7 and rho-1 and all showed results comparable with mutant analysis . Cloning and construction of plasmids were performed with standard molecular biology techniques . All expression constructs in the twnEx series transgenes were in the pPD95 . 77 Fire vector backbone , which contains the unc-54 3′-UTR for optimized expression in C . elegans . Primer sequence information is available upon request . Neurite swelling of ALM and PLM was scored in live animals with the integrated GFP reporter zdIs5 ( Pmec-4::GFP ) , which is expressed in the six mechanosensory neurons: ALMs , PLMs , AVM and PVM . Beading is defined as oval or round swelling along primary axons . Neurite swelling is defined as triangular protrusion or looping of axonal membrane . Neurodegeneration is define as swelling and round-up of the neuronal soma with neurite interruption , thinning and large beading formation . To characterize the evolution of neurite defects in mutants , wild type and gm379 animals were synchronized by hatching and arresting early L1 in M9 at 20°C . Animals were then allowed to feed on regular E . coli plates with axon morphology scored at different time points ( 6 hr , 12 hr , 24 hr , 36 hr , 48 hr , and 60 hr post hatching ) that correspond to distinct larval and adult stages . Because unc-54 animals are defective in egg laying and die from progeny hatched inside their bodies , 5-fluoro-2′-deoxyuridine ( FUdR ) was added to the plate at the final concentration of 50 µM to stop progeny production . FUdR was applied to the mec-12 ( gm379 ) mutant and the wild type in experiments where unc-54 was also tested . Synaptic vesicles in the touch neurons were visualized with the integrated GFP reporter jsIs821 ( Pmec-7::GFP::RAB-3 ) which labels synaptic vesicles in the six touch neurons . The authenticity of synaptic vesicle defects in the gm379 mutant was confirmed with another GFP reporter , jsIs37 ( Pmec-7::SNB-1::GFP ) . Touch neurons were simultaneously labeled by the RFP reporter jsIs973 ( Pmec7::mRFP ) . Animals were synchronized and jsIs821-labeled synaptic vesicles quantified at distinct developmental stages . Images were acquired using the 63x Carl Zeiss Apochromat objective and the Zeiss AxioImager M2 imaging system . Because the posterior PLM process was very thin ( less than 0 . 5 µm , see Figure S2B and S2C ) , we did not take confocal z-axis image stacks for pixel quantification . Pixel density was derived using ImageJ by quantifying total pixel number divided by the area marked by the neuronal marker jsIs973 . We excluded the neuronal nucleus when quantifying pixel density of the soma . For Figure 5G , because the UNC-104-overexpression array carries mCherry fused to UNC-104 , which may complicate determination of neurite area by the jsIs973 marker , we decided to quantify total pixel number of fluorescence on the entire PLM posterior process . Mistargerted SVs often formed GFP::RAB-3 aggregates of variable size , and we therefore did not quantify GFP punctum number or individual punctual intensity . The length of the PLM posterior process was measured by the software Axio Vision Rel . 4 . 8 . The distance of synaptic vesicle distribution was determined as the fraction of the PLM posterior process marked with synaptic vesicle GFP . All image quantification was done blind to avoid bias . Worms were high pressure frozen in either a Bal-Tec HPM 010 ( Bal-Tec AG , Liechenstein ) or Leica HMP 100 ( Leica Microsystems , Vienna ) high pressure freezer and freeze substituted in 1% osmium tetroxide and 0 . 1% uranyl acetate in acetone over a period of 2 hours by the SQFS method of McDonald and Webb [45] . Infiltration of Epon epoxy resin was carried out by 15 minute incubations in 25 , 50 , and 75% acetone-resin mixtures on a rocker , then three 15 minute incubations in pure resin . Polymerization of resin was for 2 hours in a 100°C oven . Sections of 70 nm thickness were post-stained with 2% uranyl acetate in 70% methanol for 4 minutes and lead citrate ( Reynolds , 1963 ) for 2 minutes . Images were viewed on a Tecnai 12 ( FEI Inc . , Hillsboro , OR , USA ) transmission electron microscope operating at 120 kV , and images recorded with a Gatan Ultrascan 1000 CCD camera ( Gatan Inc . , Pleasanton , CA , USA ) . Some high magnification views of microtubule were taken out of focus in order to highlight protofilament patterns [48] , [49] . EBP-2 comets were barely visible in wild type touch neurons , which could be attributed to the very stable microtubule structures in these cells . Therefore we devised an assay in which low-dose ( 0 . 125 mM ) colchicine was applied to the worms to generate a moderate level of microtubule perturbation . L2 worms with twnEx40 ( Pmec-7::EBP-2::GFP , Pmyo-2::GFP ) were grown on colchicine-containing NGM plates for 8 hours , picked off the plates and imaged one hour later . Under such treatment , a significant percentage of touch neurons displayed variable degree of microtubule growth with EBP-2 comets . Imaging acquisition was performed with the Zeiss AxioImager M2 imaging system . Worm immunostaining was performed as described [50] . Briefly , mixed-stage animals were flash-frozen in liquid nitrogen and fixed in 2% paraformaldehyde on ice for at least 4 hours , permeabilized by Tris-Triton , and subjected to series of reduction and oxidation by sequential β-mercaptoethanol , dithiothreitol ( DTT ) and hydrogen peroxide treatment in 1% borate base buffer , and stained with primary antibodies in PBST-A . The following primary antibodies were used in this study: 6-11B-1 ( mouse monoclonal anti-K40 acetylated α-tubulin , 1∶500 , Santa Cruz Biotech ) , GT335 ( mouse monoclonal anti-polyglutamylated tubulin , 1∶100 , Enzo Life Sciences ) , rabbit polyclonal anti-detyrosinated tubulin ( 1∶200 , Millipore ) , YL1/2 ( rat monoclonal anti-tyrosinated tubulin , 1∶200 , Santa Cruz Biotech ) , and rabbit polyclonal anti-GFP ( 1∶250 , Santa Cruz Biotech ) . Secondary antibodies are goat anti-rabbit , goat anti-rat or goat anti-mouse IgG conjugated with Alexa488 or Alexa568 used at 1∶100 ( Molecular Probes ) . Animals were counterstained with DAPI at 1∶1000 diluation in 2% n-propylgallate ( NPG ) and observed with the Zeiss AxioImager M2 imaging system . For fluorescence confocal microscopy , L4 to young adult hermaphrodite animals were anesthetized with 1% sodium azide , mounted on agar pad , and observed under Zeiss LSM700 confocal imaging system . A fragment of the microtubule-binding domain ( MTBD , amino acids 3207-3372 ) of C . elegans DHC-1 was cloned into the KpnI site of the pET30α vector with the primers: 5′ GGTACCCTCGCAGAGCAGCTGAAG 3′ ( forward ) and 5′ GGTACCTTATTCCTGGGTCTTCTTTGCAGC 3′ ( reverse ) , and tagged at the N-terminus with 6xHis . Expression in E . coli was induced by 0 . 5 mM IPTG at 16°C for 4 hours , for avoiding protein aggregate formation in subsequent steps of purification . Bacterial pellet was collected by centrifugation and resuspended in the lysis buffer containing 50 mM NaH2PO4 , 500 mM NaCl , 10 mM imidazole , 0 . 1% lysozyme , protease inhibitor cocktail and were homogenized by sonication . Cell extract was centrifuged at 15 , 000 g for 30 min at 4°C and MTBD was purified by passing the supernatant through HisPur™ Ni-NTA resin ( Thermo Fisher Scientific , Walthem , USA ) . Purified MTBD was dialyzed in HEPES buffer ( 80 mM HEPES pH 7 . 0 , 2 mM MgCl2 , 0 . 5 EGTA ) for microtubule binding protein spin-down assay . The microtubule sedimentation assay was performed as described with modifications [51] . In brief , unc-119; twnEx98[Punc-119::MEC-12 , Punc-119::MEC-7 , unc-119 ( + ) ] and unc-119; twnEx99[Punc-119::MEC-12 ( G416E ) , Punc-119::MEC-7 , unc-119 ( + ) ] transgenic animals were grown to gravid adults on standard NGM plates . Animals were collected by washing and centrifugation in 0 . 1 M PIPES ( pH 6 . 94 ) , 4 . 0 mM MgCl2 , 5 mM EGTA , 0 . 1 mM EDTA , 0 . 9 M glycerol , 1 mM PMSF , and 1 mM DTT ( PMEG ) at 4°C and resuspension in cold PMEG with protease inhibitor cocktail . Worms were then manually homogenized and centrifuged at 20 , 000xg for 45 min and the pellet discarded . The supernatant was centrifuged at 150 , 000xg for 60 min and the pellet discarded . The translucent supernatant was then supplemented with 2 mM GTP , 10 pM taxol , 1 U/ml hexokinase , 50 mM glucose , and 25–50 nm AMP-PNP and incubated on ice for 90 min for microtubule polymerization . Microtubule polymers were sedimented through 20% sucrose cushion made in PMEG with10 pM taxol by centrifugation at 20 , 000xg for 90 min . The pellet was resuspended in 1 ml PMEG containing 10 nM taxol and 50 mM NaCl . Microtubules and associated proteins were pelleted at 20 , 000xg for 40 min and dissolved in water for further analysis . In vitro microtubule sedimentation assay was performed by using microtubule binding protein spin-down assay kit ( Cytoskeleton , Denver , USA ) . In brief , microtubules were synthesized from purified bovine tubulins and incubated with purified MTBD at room temperature for 30 minutes . Microtubule-associated proteins were pelleted in sucrose cushion by centrifugation at 20000xg , resuspended and analyzed by SDS-PAGE . Coomassie blue-stained signals of tubulins and MTBD were quantified using ImageJ . The signal intensity of MTBD was normalized to respective tubulin signals . Protein lysate from microtubule sedimentation assay was boiled with SDS lysis buffer and separated by SDS-PAGE ( 6% acrylamide ) . Proteins were transferred to a nitrocellulose membrane and probed with mouse anti-UNC-104 monoclonal antibody ( 1∶60 , a gift from Dr . S . Koushika ) [52] , rabbit anti-DHC-1 polyclonal antibody ( 1∶200 , a gift from Dr . P . Gonczy ) [30] or 6-11B-1 ( 1∶1000 , Santa Cruz Biotech ) for acetylated α-tubulin , followed by HRP-based chemiluminescence detection . Signal intensity was quantified using ImageJ . All experiments were repeated at least three times . To quantify the amount of DHC-1 or UNC-104 co-sedimented with microtubules , the pixel intensity of individual bands was first quantified using ImageJ . We first normalized the amount of bound DHC-1 or UNC-104 relative to sedimented microtubules in respective experiments . Next , we normalized the DHC-1/microtubule values to the averaged UNC-104/microtubule value , and data from five independent experiments were expressed as a fold change relative to the UNC-104/microtubule ratio .
|
Axons and dendrites are two classes of neuronal process that differ in their functions and molecular compositions . Proteins important for synaptic functions are mostly synthesized in the cell body and sorted differentially into the axon or dendrites . Microtubules in the axon and dendrite maintain their structural integrity and regulate polarized protein transport into these compartments . We identified a novel α-tubulin mutation in C . elegans that caused mistargeting of synaptic vesicles and induced progressive neurite swelling , which resulted in late-onset neurodegeneration . We showed that this tubulin mutation weakened microtubule network and abnormally increased microtubule affinity for dynein , a motor protein responsible for cargo sorting to the dendrite . This enhanced microtubule-dynein affinity is due to augmented negative charge at the carboxyl terminus of α-tubulin . Neurite swelling and neurodegeneration could be ameliorated by reduced physical activity , suggesting that recurrent mechanical strain from muscle contraction jeopardized neurite integrity in the long run . Mutations in α- and β-tubulins are found in human neurological diseases; our findings therefore contribute to understanding the pathogenic mechanism of human neurological diseases associated with tubulin mutations .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"developmental",
"biology",
"cell",
"biology",
"genetics",
"biology",
"and",
"life",
"sciences",
"neuroscience"
] |
2014
|
Genetic Analysis of a Novel Tubulin Mutation That Redirects Synaptic Vesicle Targeting and Causes Neurite Degeneration in C. elegans
|
The actions of the RIG-I like receptor ( RLR ) and type I interferon ( IFN ) signaling pathways are essential for a protective innate immune response against the emerging flavivirus West Nile virus ( WNV ) . In mice lacking RLR or IFN signaling pathways , WNV exhibits enhanced tissue tropism , indicating that specific host factors of innate immune defense restrict WNV infection and dissemination in peripheral tissues . However , the immune mechanisms by which the RLR and IFN pathways coordinate and function to impart restriction of WNV infection are not well defined . Using a systems biology approach , we defined the host innate immune response signature and actions that restrict WNV tissue tropism . Transcriptional profiling and pathway modeling to compare WNV-infected permissive ( spleen ) and nonpermissive ( liver ) tissues showed high enrichment for inflammatory responses , including pattern recognition receptors and IFN signaling pathways , that define restriction of WNV replication in the liver . Assessment of infected livers from Mavs−/−×Ifnar−/− mice revealed the loss of expression of several key components within the natural killer ( NK ) cell signaling pathway , including genes associated with NK cell activation , inflammatory cytokine production , and NK cell receptor signaling . In vivo analysis of hepatic immune cell infiltrates from WT mice demonstrated that WNV infection leads to an increase in NK cell numbers with enhanced proliferation , maturation , and effector action . In contrast , livers from Mavs−/−×Ifnar−/− infected mice displayed reduced immune cell infiltration , including a significant reduction in NK cell numbers . Analysis of cocultures of dendritic and NK cells revealed both cell-intrinsic and -extrinsic roles for the RLR and IFN signaling pathways to regulate NK cell effector activity . Taken together , these observations reveal a complex innate immune signaling network , regulated by the RLR and IFN signaling pathways , that drives tissue-specific antiviral effector gene expression and innate immune cellular processes that control tissue tropism to WNV infection .
Acute virus infection induces host innate immune defense programs that serve to control virus replication , prevent virus-mediated pathology , and aid in developing sterilizing immunity ( i . e . humoral and cell-mediated immunity ) . During RNA virus infection , intracellular viral RNA is recognized as a non-self pathogen associated molecular pattern ( PAMP ) by the RIG-I like receptors ( RLR ) , RIG-I and MDA5 [1] , [2] . Upon binding virus-specific RNA structures and target nucleic acid sequences , RIG-I and MDA5 undergo conformational change and interact with the adaptor protein MAVS , leading to activation of NF-κB and interferon regulatory factor ( IRF ) , including IRF-3 and IRF-7 , that drive transcription of antiviral target genes , interferon-β , pro-inflammatory cytokines , and hundreds of interferon-stimulated genes ( ISGs ) [1] , [3] . This host response is further amplified by signaling through the type I interferon ( IFN ) receptor that drives the formation of the multimeric interferon-stimulated gene factor 3 ( ISGF3 ) , consisting of IRF-9 , STAT2 and/or STAT1 , that binds to interferon-stimulated response promoter elements ( ISRE ) and further amplifies the transcription of ISGs . While several studies have identified key innate immune host factors in controlling virus replication and protection , little is known about context ( specific cell types and organs ) and how these processes regulate innate immune responses to restrict tropism of virus infection West Nile virus ( WNV ) is an emerging neurotropic flavivirus that is the leading cause of mosquito-borne encephalitis in humans in the United States . The WNV pathogenesis model of infection in mice provides a platform to study immune response processes and pathways that regulate infection . After subcutaneous footpad inoculation in mice , WNV initially replicates in the skin at the inoculation site and the draining popliteal lymph node , resulting in a primary viremia and spread to the spleen . Within the spleen the virus is amplified and following a secondary viremia , WNV invades the central nervous system tissues ( e . g . , brain and spinal cord ) , thus recapitulating infection and pathogenesis of human infection transmitted by mosquito vector [4] . WNV replication is typically restricted to the skin , draining lymph node , spleen , and central nervous system in humans and WT mice [4] , [5] . Low levels of infectious virus can be recovered from the lung , kidney , heart , pancreas and other peripheral tissues but not the liver , of WT infected mice [6] . Most peripheral organs , including the liver , are not typically associated with WNV replication in humans . However , reported cases of kidney , liver , and heart organ transplant-transmitted WNV infections have been described with outcomes ranging from asymptomatic infections to death in the recipients [7] . These clinical observations suggest that peripheral organs in humans are also capable of being infected by WNV but infection is restricted or controlled by immune defense programs . The RLR and the type I IFN signaling pathways are essential in eliciting innate immune responses that restrict WNV infection , control tissue tropism , and serve to program downstream adaptive immune responses [8]–[11] . During WNV infection , RIG-I is essential for triggering early antiviral immune defenses , whereas MDA5 serves to enhance and sustain this response [12] . Known target cells of WNV infection , including dendritic cells , macrophages , and neurons rely exclusively on RLR signaling to control WNV replication and drive innate antiviral immune defenses [8] . In the absence of the RLR or type I IFN signaling , WNV displays enhanced virus replication and tissue tropism [8] , [13] , indicating that the RLR and type I IFN signaling pathways impart a host response that restricts tissue permissiveness to WNV . Recent studies have identified several antiviral effector genes that control specific stages in the life cycle of WNV infection . Most notably , the interferon-induced protein with tetratricopeptide repeats ( IFIT ) family members [14]–[16] , interferon induced transmembrane proteins ( IFITM ) 1 and 2 [17] , [18] , radical S-adenosyl methionine domain containing 2 ( RSAD2; commonly known as viperin; [19] , ISG20 [17] , and interferon-inducible double stranded RNA activated kinase ( PKR; [20] ) have been shown to directly inhibit WNV infection . In addition , a high-throughput antiviral effector gene screen revealed hundreds of ISGs whose expression corresponds with IFN-mediated restriction of WNV replication , however , the biological significance of these genes in regulating pathogenesis is not defined [21] . Moreover , little is known of how these ISGs function in a context-dependent ( e . g . in specific cells and tissues ) manner to limit virus replication and spread in vivo . The production of and response to type I IFN during the acute stage of virus infection is a major linkage point between innate and adaptive immunity . IFN-α and IFN-β have been shown to sustain B cell activation and differentiation [22]–[24] , support the expansion of antigen-specific CD4+ and CD8+ T cells [25]–[28] , and enhance the activation of natural killer ( NK ) cell responses against infection [29] . Moreover , recent studies have implicated the RLR signaling pathway in regulating inflammation , humoral immune responses , and T cell responses during viral infection [8] . Specifically , the absence of MAVS leads to excessive inflammation characterized by enhanced pro-inflammatory cytokine and chemokine production , enhanced virus-specific T cell responses , dysregulated humoral responses , and defective expansion of regulatory T cells during WNV infection [8] . In addition , the RLR LGP2 was found to be essential in regulating T cell survival and effector functions during virus infection [30] . These findings implicate the RLRs and their signaling as well as the induction of the IFN response as integral processes that restrict infection by containing acute virus replication and spread while bridging innate and adaptive immune responses to systemically control infection . Here , we interrogated WNV infection in mice utilizing a systems biology approach comprising of genomics analyses , immunologic assessments , computational modeling , and hypothesis testing to define the nature of a protective innate immune response in controlling flavivirus tissue tropism and restriction of acute infection . We directly compared in vivo innate immune signaling and effector functions within the spleen , a permissive tissue for WNV infection , and the liver , a nonpermissive tissue to WNV infection , among WT and innate immune-deficient mouse lines . Our results define RLR and type I IFN signaling programs as pivotal immunologic regulatory nodes of WNV restriction . In addition , these studies demonstrate that hepatic tropism of WNV infection is controlled through a novel axis of RLR and type I IFN signaling linked to antiviral and innate immune cellular signaling pathways .
To evaluate the roles of the RLR and type I IFN signaling pathways in protection against WNV infection , wild-type ( C57BL/6 ) , Mavs−/− , Ifnar−/− , and Mavs−/−×Ifnar−/− ( herein referred to as DKO ) mice were challenged with WNV-TX , a virulent lineage I and emergent West Nile virus strain ( Figure 1A ) . All mouse lines were derived directly in pure C57BL/6 genetic background or were fully backcrossed to derive a pure C57BL/6 strain . Mavs−/− mice displayed a significant increase in mortality ( from 29 . 2% to 100% mortality; P<0 . 0001 ) as compared to WT infected mice , demonstrating the importance of the RLR signaling pathway in protection against WNV infection ( Figure 1B; [8] ) . We also observed a significant increase in mortality of Ifnar−/− mice from WNV infection ( from 29 . 2% to 100% mortality; P<0 . 0001 ) , with mice succumbing to WNV within 2 . 7 days post infection ( p . i . ) , consistent with other studies demonstrating the importance of type I IFN signaling in protection against WNV infection [31] . DKO mice also displayed a significant increase in susceptibility to WNV infection over WT mice ( 100% mortality compared to 29% mortality , P<0 . 0001 ) . Remarkably , we observed a significant increase in average survival time of WNV-infected DKO mice as compared to Ifnar−/− infected mice . Together , these findings demonstrate that the RLR and type I IFN signaling pathways mediate protection against WNV infection but may also serve to modulate the immune response to infection . The innate immune response is critical for controlling WNV replication , both in peripheral organs and the CNS , which ultimately leads to protection from virus challenge . Thus , the lack of virus control can directly lead to reduced survival . To define the roles of the RLR and type I IFN signaling pathways alone and in combination for controlling virus replication , spread , and tropism , we evaluated viral burden in the spleen and liver after WNV challenge . These tissues represent a permissive ( spleen ) and a nonpermissive ( liver ) tissue in WT mice and humans . As compared to WT infected mice , spleens from Mavs−/− , Ifnar−/− , and DKO infected mice overall exhibited higher peak viral burden ( Figure 2A ) . Virus replication was detected as early as day 1 p . i . in the spleens from Mavs−/− and Ifnar−/− infected mice , with spleens from DKO infected mice displaying significantly higher viral loads . Thus , even in the context of the permissive spleen tissue environment , the RLR and type I IFN signaling pathways impart innate immune responses that restrict early virus replication and limit peak viral burden . Similarly , Mavs−/− , Ifnar−/− , and DKO infected mice exhibited liver infection ( Figure 2B ) , wherein livers from DKO infected mice displayed the highest peak viral burden and infectious virus production as compared to Mavs−/− and Ifnar−/− infected mice . Interestingly , Mavs−/− infected livers showed significantly reduced viral titers as compared to either Ifnar−/− and DKO infected mice , suggesting RLR-independent signaling pathways control WNV replication in the liver . Importantly , no infectious virus was detected in livers from WT infected mice , consistent with other studies that have shown resistance of WNV infection in the livers of WT mice [6] , [32] . Through the use of a highly-sensitive qRT-PCR assay , we found that WNV RNA levels in the liver peaked on day 4 p . i . ( Figure 2C ) . While we cannot entirely rule out the possibility of blood contamination attributing to the detection of WNV RNA in the liver , these findings are consistent with a study that showed immunofluorescence staining of WNV antigen in hepatocytes of livers from WT infected mice [33] . These observations suggest that livers from WT mice are exposed to WNV , but infection is either prevented or controlled through restrictions imposed by innate immune programs mediated by RLR and type I IFN-dependent signaling . Moreover , these observations indicate that RLR and type I IFN actions serve to control the extent of WNV infection in permissive/tropic tissue . To define the gene expression signatures that control permissiveness to WNV infection , we applied mouse whole-genome microarrays to profile global gene expression changes between permissive- ( spleen ) and nonpermissive- ( liver ) tissues from WT infected mice on day 4 p . i . Differentially expressed ( DE ) genes ( defined as ≥1 . 5-fold change in expression with paired Student's t-test P≤0 . 01 ) were identified by comparing infected spleen and liver to their respective tissues from strain-matched mock-infected controls . Venn analysis defined 3 categories of DE genes whose expression was linked with viral burden in WT infected tissues: 1 ) Spleen-specific- representing non-restrictive genes in controlling tropism; 2 ) Common- representing genes that are necessary but not sufficient for controlling tropism; and 3 ) Liver-specific- representing genes that restrict and control WNV infection tropism ( Figure 3A ) . These analyses yielded 945 spleen-specific DE genes , 179 common DE genes , and 479 liver-specific DE genes . In addition , we performed transcriptional profiling from Mavs−/− , Ifnar−/− , and DKO infected mouse liver to identify genes whose expression is regulated by RLR signaling , type I IFN signaling , or both pathways . We visualized the 658 DE genes from WT infected liver tissues and found that the absence of RLR or type I IFN receptor signaling resulted in reduced gene expression , with the greatest differences in gene expression observed from DKO infected livers ( Figure 3B ) . Hierarchical clustering and functional annotation of liver infected genes revealed that RLR and type I IFN signaling have a significant impact on expression of genes related to G-protein coupled receptor protein signaling pathway ( GO:0007186 ) , cell division ( GO:0051301 ) and immune response ( GO:0006955 ) biological processes . Additionally , we identified 359 RLR signaling-dependent DE genes , 281 IFN signaling-dependent DE genes , and 464 RLR and type I IFN signaling-dependent DE genes ( Table S1 ) from WNV infected livers . These gene sets define a signature of RLR and type I IFN signaling target genes that correspond to restricted tissue tropism to WNV . We next performed computational modeling of gene expression data to define gene networks and their canonical regulatory pathways responsive to WNV infection within WT spleen and liver . DE gene lists were analyzed with Ingenuity Pathway Analysis ( IPA ) . The top five biological categories from this analysis are presented in Table 1 . WT infected spleen displayed cell cycle and cancer-related pathways as top-ranking biological processes . This signature is consistent with initiation of cell-mediated innate and adaptive immune responses and their dependence on immune cell proliferation beginning on day 4 p . i . [8] , [22] . In contrast , WT infected liver tissues displayed inflammatory response and cell-to-cell signaling and interaction as top ranking biological processes , reflecting the essential nature of the innate immune response in restricting tissue permissiveness to WNV infection . Network analysis of genes comprising the inflammatory response biological function from WT infected livers ( Figure 4A ) revealed a highly interconnected network consisting of pattern recognition receptors ( DDX58 ( RIG-I ) , IFIH1 ( MDA5 ) , DHX58 ( LGP2 ) , TLR3 , and TLR7 ) , IRF-3 and ISRE-target genes with known antiviral effector function against WNV and other RNA viruses ( MX1 [34] , IFIT1 [15] , [16] , [35] , IFIT2 [36] , ISG15 [37] , [38] , GBP2 [39] , and GBP3 [40] , innate immune transcription factors ( IRF-7 , IRF-9 , STAT1 , and STAT2 ) , proteins related to antigen presentation ( CD86 , TAP1 and PSMB8 ) , and pro- and anti-inflammatory cytokine and chemokine related genes ( TNF-α , CXCL10 and IL-10RA ) . As a proof of principle , we also independently identified innate immune genes that have previously been shown to control WNV replication and tissue tropism . These include IRF-7 [41] , TLR7 [33] , [42] , TLR3 [43] , RSAD2 [19] , MDA5 [12] , RIG-I [12] , IFIT1 [15] , IFIT2 [14] , [15] , ISG20 [17] , and OAS [44] . Through network analysis , we defined MDA5 , RIG-I , IRF-7 , IRF-9 , STAT2 , STAT1 , and TNF-α as regulatory nodes of gene expression ( as determined by ranking the total number of edges to a given node ) within infected livers . Additionally , we observed distinct tissue-specific gene expression patterns , in which essential innate immune signaling components RIG-I , MDA5 , IRF-7 , IRF-9 , STAT1 , and STAT2 were expressed in spleen and liver tissues , whereas TLR3 and TNF-α were only expressed in infected liver tissue . These findings support our hypothesis that distinct tissue-specific innate immune programs restrict tissue tropism to WNV . To identify the innate immune signaling pathways that directly regulate gene expression and restrict liver tropism of WNV , we compared genes within this inflammatory network with expression profiles from Mavs−/− , Ifnar−/− , and DKO infected liver tissues . The absence of MAVS resulted in the loss of expression for a select number of genes , including CD180 , IFI203 , GM11428 , IL10RA , TLR3 , and TNF-α ( Figure 4B ) . Surprisingly , in the absence of type I IFN signaling , few genes were found to be reduced or downregulated in expression . Rather , many genes were found to be increased or upregulated in expression as compared to genes from WT or Mavs−/− infected liver tissues . The combined absence of RLR and type I IFN signaling pathways displayed markedly downregulated gene expression as compared to WT or Mavs−/− and Ifnar−/− infected liver tissues . Genes that were expressed to high levels in the Ifnar−/− infected liver tissues were strongly reduced in DKO infected liver tissues ( Figure 4C–G ) , indicating that a major component of the host response to WNV infection triggered in Ifnar−/− infected liver is mediated through RLR signaling . These findings demonstrate that RLR and type I IFN signaling pathways are required to confer complete expression of host response genes for innate immune protection and the restriction of infection within liver tissue . To reveal the relationship between the known cellular pathways enriched in WT infected spleen and liver we applied IPA and computational modeling of curated response pathways . We found that pathways specific to the role of pattern recognition receptors in recognition of bacteria and viruses , activation of IRFs by cytosolic pattern recognition receptors , and interferon signaling were among the top scoring enriched canonical pathways in infected tissue ( Tables S2 and S3 ) . To define overlapping gene-sets and network clusters among infected tissues , we visualized the top-ranked IPA pathways from WT infected liver through a network-based Enrichment Map method within Cytoscape [45] , in which canonical pathways were organized into networks grouped by major biological function . In this analysis , canonical pathways are represented by a node with edges represent overlapping genes between sets ( Figure 5 and Table S4 ) . This visualization method revealed a highly interconnected network of biological processes that are regulated within the WT infected liver . The IFN signaling biological function displayed the highest enrichment score ( ES = 30 . 14 ) , demonstrating that IFN-mediated responses play a strong role in regulating early host responses to WNV infection . Furthermore , the IFN signaling intra-connected canonical pathways ( comprised of: 1- role of pattern recognition receptors in recognition of bacteria and viruses , 2- activation of IRF by cytosolic pattern recognition receptors , 3- interferon signaling , 4- toll-like receptor signaling , 5- role of PKR in interferon induction and antiviral response , and 6- role of RIG-I like receptors in antiviral innate immunity ) displayed mutually overlapping gene-sets among the major biological functions listed in Figure 5 , including gene sets of the innate cellular immune response ( ES = 24 . 31 ) , adaptive cellular immune response ( ES = 16 . 61 ) , disease-specific signaling ( ES = 13 . 04 ) , cytokine signaling ( ES = 10 . 91 ) , humoral immune response ( ES = 3 . 50 ) , and intracellular and secondary messenger signaling ( ES = 1 . 09 ) . Using enrichment map analysis , we next investigated additional signaling pathways that may regulate liver tropism to WNV by examining the innate cellular immune response regulatory nodes . We identified NK cell signaling ( node 2; P = 3 . 39×10−6 ) and NK cell-related signaling pathways , including communication between innate and adaptive immune cells ( node 1; P = 2 . 14×10−6 ) and crosstalk between dendritic cells and natural killer cells ( node 3; P = 2 . 4×10−4; Table S4 ) as top-scoring canonical pathways . Among these pathways , the NK cell signaling canonical pathway was exclusively enriched in the liver , suggesting that NK cells and related signaling pathways play an important role in the control of WNV tissue tropism and restriction of infection in the liver . Moreover , our data sets indicated that additional signaling pathways , including TREM1 signaling ( P = 6 . 17×10−7 ) , Fcγ Receptor-mediated Phagocytosis in Macrophages and Monocytes ( P = 1 . 26×10−3 ) , and other top scoring liver-specific canonical pathways might impart further innate immune actions that contribute to restriction of WNV tissue tropism . We also compared gene expression data from KO infected liver tissues to determine what role the RLR and type I IFN signaling pathways play in regulating NK cell signaling . The NK cell signaling pathway was enriched in livers from Mavs−/− and Ifnar−/− infected mice , but not from DKO infected mice , suggesting an important connection between the RLR and type I IFN signaling axis and NK cell regulation ( Table 2 ) . NK cells are important components of the innate immune response that regulate viral infection by killing infected cells and regulating inflammatory cytokine milieu [46] . However , the role of NK cells in regulating tissue tropism to WNV infection or in controlling WNV replication has not been defined . In the context of WT infected livers ( Figure 6 ) , analysis of the NK cell signaling network revealed upregulated expression of molecules that mediate inflammatory cytokine production ( REL , STAT1 , DAP12 , and CD300 ) , detection of “missing self” ( MHC class I complex , TAP1 , and PSMB8 ) , NK cell activation ( CD44 , VAV1 , BTK , and LCP2 ) , activation of antibody-dependent cell cytotoxicity ( FCER1G , FCGR3A ) , secretion of cytokines ( TNF-α , and IL10 ) and regulatory receptors ( KLRK1 , KLRA12 , and NCR1 ) . As expected , Mavs−/− or Ifnar−/− infected livers displayed minor alterations in gene expression within the NK cell signaling network as compared to WT infected livers . Strikingly , DKO infected livers displayed significantly reduced expression for nearly every gene within the NK cell signaling pathway . These findings further suggest that NK cell responses within the liver could be directly regulated by the RLR and type I IFN signaling axis . We next performed biological validation studies to determine the extent of NK cell expansion and activation in livers from WNV-infected mice . On day 4 p . i . , livers from WT infected mice displayed significantly increased immune cell infiltrates and NK cells ( CD3ε−NK1 . 1+ ) as compared to mock-infected WT mice ( Figure 7A–B ) , indicating that WNV infection triggers a liver NK cell response . Livers from Mavs−/− and Ifnar−/− infected mice also showed significant increases in NK cell numbers as compared to their respective mock infected mice . Remarkably , DKO infected mice displayed dramatically reduced numbers of NK cells in the liver as compared to mock infected mice , demonstrating that RLR and type I IFN signaling is essential in regulating liver NK cell recruitment and/or expansion during WNV infection . Phenotypic analysis revealed enhanced proliferation ( KLRG1+ expression; Figure 7C ) and maturation ( CD11b+ expression; Figure 7D ) on liver NK cells recovered from WT infected mice as compared to gene-knockout infected mice . Furthermore , liver NK cell subsets from WT infected mice displayed enhanced IFN-γ and/or CD107a ( a sensitive marker for NK cell functional activity , including cytokine secretion and NK cell-mediated lysis [47] ) positive phenotype as compared to NK cells from Ifnar−/− and DKO , but not Mavs−/− , infected mice ( Figures 7E–G ) . These findings corroborate our computational and pathway analyses that found specific enrichment for the NK cell signaling pathway within WT infected livers as compared to DKO infected livers . Overall , these findings provide biological validation that the RLR and type I IFN signaling axis is important in regulating NK cell recruitment , proliferation , and effector functions within the livers of WNV infected mice . Crosstalk between dendritic cells and natural killer cells is essential for NK cell activation during viral infection [48] . Our computational modeling revealed specific enrichment for NK cell processes , including the canonical pathway for DC/NK cell crosstalk ( see Figures 5 and 6A ) . As DCs are a target cell for WNV infection in vivo , we sought to establish whether the RIG-I like receptor or IFN signaling pathways function in a cell-intrinsic or cell-extrinsic manner to regulate NK cell effector functions through DC/NK cell crosstalk during infection . We evaluated this using a coculture assay wherein NK cells are cultured together with bone-marrow derived DCs and assayed for NK cell effector function [48] . Culture of WT WNV-infected BM-DCs with syngeneic purified WT NK cells ( 1∶2 ratio ) triggered NK cell effector function , as determined by NK cell expression of intracellular IFN-γ ( Figure 8A ) , CD107a ( Figure 8B ) or induction of both IFN-γ and CD107a ( Figure 8C ) as compared to NK co-culture with naïve control DCs . Cocultured Mavs−/− NK cells with WT WNV-infected DCs exhibited similar enhancement of IFN-γ and CD107a expression as compared to WT NK cells , demonstrating that MAVS does not act in a cell-intrinsic manner to regulate NK cell effector function during WNV infection . In contrast , Ifnar−/− and DKO NK cells cocultured with WT WNV-infected DCs exhibited an attenuated effector phenotype , demonstrating that IFN functions in a cell-intrinsic manner to regulate NK cell effector function during WNV infection . In a reciprocal set of cocultures , Mavs−/− , and Ifnar−/− WNV-infected DCs exhibited reduced ability to trigger IFN-γ secretion , but not degranulation , in NK cells ( Figures 8D–F ) . DKO WNV-infected DCs exhibited a significant reduction in the ability to activate NK cell effector functions , including IFN-γ secretion and CD107a expression ( Figures 8D–F ) . Combined with our in vivo NK cell analysis , we conclude that both the RLR and type I IFN signaling pathways act in a NK cell-extrinsic manner , whereas IFN acts in a NK cell-intrinsic manner , to activate NK cell effector functions during viral infection . Overall , these observations reveal a complex innate immune signaling network , regulated by the RLR and type I IFN signaling pathways , that drives tissue-specific innate immune programs to control tissue tropism to WNV infection . Furthermore , these studies provide novel insight into the regulation of hepatic tropism to WNV infection through the expression of antiviral effector genes and innate immune cellular processes .
Innate immunity is our first line of defense against viral pathogens and serves to protect us from infection on a daily basis . The proper coordination of innate immune signaling pathway function and efficient expression of antiviral effector genes are key to governing the outcome of virus infection and immunity . In this study , we demonstrate that the RLR and type I IFN signaling axis plays an essential role in protection against WNV infection and functions in a tissue-specific manner to control virus replication and restrict tissue tropism . We demonstrate that WNV can indeed infect liver tissue but this infection is rapidly cleared in WT infected mice but not RLR-deficient and/or IFN signaling-deficient mice . In order to define the genomic signature and components of a protective innate immune response that restricts WNV infection , we applied an integrated systems biology approach to uncover novel connections between gene networks and pathways . Functional genomics analysis and pathway modeling revealed that inflammatory response programs were highly enriched in WT WNV-infected liver compared to the permissive spleen . As a proof of principle , we identified genes encoding known innate antiviral effectors against WNV [14]–[20] , including pattern recognition receptor and interferon signaling pathways [8] , [13] . In addition , we revealed that the NK cell signaling network is an important feature of innate immune restriction of WNV infection . Through pathway modeling and biological studies , the NK cell responses were found to be directly regulated by the RLR and type I IFN signaling axis . Our observations support a model in which the gene networks within the RLR and type I IFN signaling axis , including IRFs , STATs , and ISG effectors , impart restriction of virus replication while linkage with innate immune cellular process , such as with NK cell signaling , facilitate the innate immune response . Ultimately , this complex signaling network controls WNV infection and prevents virus spread within the liver ( Figure 9 ) . While NK cells from livers of WNV infected WT mice were fully mature , proliferating , and functional , NK cells from DKO infected liver tissues were dysregulated , exhibiting reduced expression of maturation markers , reduced proliferation , and attenuated effector functions . In contrast to infected livers , gene network analysis revealed that the permissive environment of the spleen for WNV infection displayed high enrichment for IFN and related signaling pathways . Interestingly , we observed significant enrichment of the canonical pathway involved in crosstalk between DCs and NK cells in the spleen and this enrichment was consistent with an increase in splenic NK cells in WT infected mice ( data not shown ) . However , we did not observe spleen-specific enrichment for NK cell interactions or related signaling pathways , suggesting that NK cell-dependent signaling is essential in hepatic but not splenic control of WNV infection . Overall , this study identified host factors and their biological processes that control liver tropism , and defined key linkages between the RLR and type I IFN signaling axis and signaling of NK cell effector function . An unexpected and exciting finding from this study was that DKO mice lived longer than Ifnar−/− mice . This increase in average WNV infection survival time correlated with stark differences in gene expression profiles in permissive and nonpermissive tissues from these knockout mice , wherein Ifnar−/− infected tissues displayed a robust inflammatory response as compared to DKO infected tissues that showed a dramatic loss in inflammation . This finding suggests that Ifnar−/− infected mice are likely succumbing to infection from a combination of inflammation-mediated and virus-mediated pathology . In contrast , DKO infected mice are likely succumbing from virus-mediated pathology as uncontrolled virus replication linked with a stark reduction in overall inflammatory response . An intriguing question is why do DKO infected tissues show such reduced inflammatory responses as compared to either Mavs−/− or Ifnar−/− infected tissues , despite the presence of RLR-independent signaling pathways , including TLR , Nod-like receptor , and C-type lectin receptors ? Despite the increased viral burden within Mavs−/− infected livers , we observed minor changes in inflammation , as defined by expression of inflammatory and antiviral effector genes ( Figure 4 ) , as compared to WT infected livers . While this response is unable to effectively control WNV replication within the liver , these findings do suggest that RLR-independent signaling pathways participate in triggering a host inflammatory response to WNV infection . Indeed , analysis of Ifnar−/− infected livers suggest that RLR-dependent and -independent signaling pathways are responsible for inducing an inflammatory response , despite the lack of type I IFN signaling . Remarkably , livers from DKO mice showed abrogated inflammation , revealing novel crosstalk between RLR-dependent , RLR-independent , and type I IFN signaling pathways that impart an antiviral inflammatory response . Previous studies of conventional infection-based analysis of knockout mice [4] and gene over-expression or silencing studies in vitro [17] , [18] identified specific restriction factors that play an important role in controlling WNV infection and pathogenesis . These genes include those encoding pattern recognition receptors ( RIG-I , MDA5 , Toll-like receptor ( TLR ) 3 , TLR7 , PKR ) , antiviral effector proteins ( ISG20 , IFIT1 , IFIT2 , and RSAD2 ) , transcription factors ( IRF-7 , IRF-9 , STAT1 , and STAT2 ) , and pro-inflammatory cytokines ( type I IFNs , IFN-γ , TNF-α , CXCL10 , and IL10 ) that inhibit various steps of WNV infection [3] , [4] , [12] , [14] , [15] , [17] , [19] , [49] . Our unbiased systems approach now confirms the tissue-specific role for these factors , while importantly revealing novel candidates of virus restriction , including IFI203 , IFI27 , GBP2 , GBP3 , GBP5 , GBP6 , GBP8 , GBP9 , FPR2 , and IRG1 . When categorizing putative antiviral effector genes based on tissue-specific expression patterns ( permissive versus nonpermissive tissue ) , we found that expression of the aforementioned restriction factors are linked directly to RLR signaling or type I IFN signaling pathways [17] , [18] . Moreover , as our observations reveal a novel linkage of NK cell signaling with the RLR and type I IFN signaling axis , we conclude that efficient innate immune response requires coordinated signaling through multiple innate immune pathways that extend well beyond any single host restriction factor . NK cells are an important component of the innate immune response and are essential in bridging the innate and adaptive immune responses [50] . These cells have been shown to be crucial for immune defense against viral pathogens , including mouse cytomegalovirus [46] , lymphocytic choriomeningitis virus [51] , WNV [52] , Herpes viruses [53] , Hepatitis C virus [54] , and Influenza virus [55] . We note that a direct role for NK cells in controlling WNV infection in vivo has yet to be demonstrated , though early control of virus replication and tissue permissiveness has been linked to perforin and IFN-γ , two host factors that are actively secreted by NK cells and other innate immune cells [56] , [57] . We attempted to deplete NK cells using an antibody-based approach , however , we found that NK cells can only be fully depleted from spleens , but not liver at early times post-WNV infection ( data not shown ) . These results confirm our informatics analysis that NK cells or NK cell signaling are not required for controlling WNV replication in the spleen . At this time , we are unable to conclusively demonstrate an essential role for NK cells in controlling WNV liver tropism in vivo . Nonetheless , West Nile and Dengue virus E protein directly bind to the natural cytotoxicity receptor NKp44 , expressed on human NK cells , leading to IFN-γ secretion and cytolytic activity [58] . Our pathway modeling suggested that WNV infection triggers liver-specific NK cell activation that is directly regulated by the RLR and type I IFN signaling axis . In fact , our biological validation studies found that infection of WT mice triggered expansion and activation of NK cells within the liver but this response was markedly reduced in DKO infected mice . Furthermore , in vitro coculture studies demonstrated that IFN signaling , rather than RLR signaling , function in an NK cell-intrinsic manner to regulate effector functions . It is remains plausible that the RLR signaling pathway supports NK cell activation and maturation during WNV infection since NK cell responses from livers of DKO mice were strongly reduced as compared to NK cells from Ifnar−/− infected mice . With respect to IFN , our findings are consistent with previous studies that have shown that IFN functions in a NK cell-intrinsic manner to regulate NK cell activation , proliferation , and maturation [29] . These findings are consistent with other independent studies which found that MAVS functions in a NK cell-extrinsic manner , specifically in dendritic cells and diverse accessory cells , to regulate NK cell activity . This suggests that type I IFN production by DCs imparts signaling crosstalk to NK cells that drives their activation [59]–[61] . Our findings expose a novel linkage between the RLR and type I IFN signaling axis with the regulation of NK cell expansion and effector activity during WNV infection . The use of our integrated systems biology approach has uncovered tissue-specific host restriction factors that control permissiveness to WNV infection . These studies have uncovered novel linkages between the RLR and type I IFN signaling axis while identifying key innate immune regulatory nodes that can be considered as targets for pharmacological interventions . Our findings have significant implication for the design of strategies to target host restriction factors and pathways for antiviral therapies against flavivirus infections .
This study was carried out in accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health . Animal experiments were approved and performed in accordance to the Institutional Animal Care and Use Committee at the University of Washington ( Protocol Number: 4158-01 ) . BHK-21 cells were cultured in Dulbecco's modified Eagle medium ( DMEM ) supplemented with 10% fetal bovine serum ( FBS ) , HEPES , L-glutamine , sodium pyruvate , antibiotic-antimycotic solution , and nonessential amino acids . WNV isolate TX 2002-HC ( WNV-TX ) was described previously [62] and tittered by a standard plaque assay on BHK-21 cells . Working stocks of passaged WNV-TX [8] or WNV-TX infectious cloned virus [63] were generated in BHK-21 cells as previously described and used in both in vitro and in vivo experiments . Sti−/− mice ( C57BL/6 background; referred to in text as Mavs−/− ) were generated in the Gale laboratory [30] . Ifnar1−/− mice were generously provided by Murali-Krishna Kaja ( C57BL/6 background; referred to in the text as Ifnar−/− ) . C57BL/6 wild type inbred mice were commercially obtained ( Jackson Laboratories , Bar Harbor , ME ) . All mice were genotyped and bred in specific pathogen-free conditions in the animal facility at the University of Washington . The methods for mice use and care were performed in accordance with the University of Washington Institutional Animal Care and Use Committee guidelines . Age-matched six to twelve week old mice were inoculated subcutaneously in the left rear footpad with 100 PFU of WNV-TX in a 10 µl inoculum diluted in Hanks balanced salt solution ( HBSS ) supplemented with 1% heat-inactivated FBS . Mock infected mice were inoculated in a similar manner with diluents alone . Mice were monitored daily for morbidity and mortality . For in vivo studies , infected mice were euthanized , bled , and perfused with 20 ml of phosphate-buffered saline ( PBS ) . Spleen , kidney , and livers were removed , weighed , and homogenized in 500 ul of PBS containing 1% heat-inactivated FBS using a Precellys 24 at 1500 RPM for 20 seconds ( Bertin Technologies , France ) . Sample homogenates were tittered by plaque assay on BHK-21 cells . For qRT-PCR analysis of viral load , RNA was extracted from tissues as described below and WNV RNA copy number was measured by RT-quantitative PCR ( RT-qPCR ) as previously described [8] . Expression oligonucleotide arrays were performed on RNA isolated from spleen and liver tissues from strain and time-matched mock infected mice ( n = 2 ) and WNV-TX infected WT ( n = 3; day 4 p . i . ) , Mavs−/− ( n = 3; day 4 p . i . ) , Ifnar−/− ( n = 3; day 2 p . i . ) , and Mavs−/−×Ifnar−/− ( n = 3; day 4 p . i . ) mice . RNA was extracted from mock-infected and WNV-TX infected tissues using a combination of TRIzol ( Life Technologies ) with RNAlater solution ( Ambion ) according to the manufacturer's instructions . RNA was further purified using RNeasy columns ( Qiagen ) . RNA samples were spectroscopically verified for purity , and the quality of the intact RNA was assessed using an Agilent 2100 Bioanalyzer . cRNA probes were generated from each sample by the use of an Agilent one-color Low Input Quick Amp labeling kit ( Agilent Technologies ) . Individual cRNA samples were hybridized to Agilent mouse whole-genome oligonucleotide 4-by-44 microarrays ( G4122F; approximately 39 , 000 unique mouse genes ) according to the manufacturer's instructions . Slides were scanned with an Agilent DNA microarray scanner , and the resulting images were analyzed using Agilent Feature Extractor version 8 . 1 . 1 . 1 . This software was used to perform image analysis , including significance of signal and spatial detrending and to apply a universal error model . For these hybridizations , the most conservative error model was applied . Raw data were then loaded into a custom-designed laboratory information management system ( LIMS ) . Data were warehoused in a Labkey system ( Labkey , Inc . , Seattle , WA ) and analyzed using GeneData Analyst 2 . 2 . 1 software ( GeneData Solutions In Silico , San Francisco , CA ) , and Spotfire DecisionSite for Functional Genomics 9 . 1 software ( Tibco Spotfire , Somerville , MA ) . Raw microarray data have been deposited in NCBI's Gene Expression Omnibus under GEO Series accession number GSE39259 and are also accessible through the Katze Lab website ( www . viromics . washington . edu ) in accordance with proposed Minimum Information About a Microarray Experiment ( MIAME ) standards . A Student's t-test ( P≤0 . 01 ) was performed to determine the genes that had significantly different expression levels with infection compared to levels in mock infections for each of the four mouse strains ( 1 . 5 fold change ) . Functional analysis of statistically significant gene expression changes was performed with the DAVID Bioinformatics Resources [64] , [65] and Ingenuity Pathways Analysis ( IPA; Ingenuity Systems ) . These softwares analyze expression data in the context of known biological response and regulatory networks as well as other higher-order response pathways . Ingenuity functional analysis identified biological functions and/or diseases that were most significant . For all analyses , a Benjamini-Hochberg test correction was applied to the IPA-generated P value to determine the probability that each biological function assigned to that data set was due to chance alone . In the functional networks , genes are represented as nodes , and the biological relationship between two nodes is represented as an edge ( line ) . All edges are supported by at least one published reference or from canonical information stored in the Ingenuity Pathways Knowledge Base . Age-matched six to twelve week old mice were inoculated subcutaneously in the left rear footpad with either diluents alone or 100 PFU of WNV-TX and mice were euthanized and perfused with 20 ml of cold phosphate-buffered saline ( PBS ) . Whole livers were dissected and placed in a 2 ml of cold complete RPMI media ( cRPMI; 10% fetal bovine serum , L-glutamine , Non-essential amino acids , sodium pyruvate , and antibiotics/antimycotic ) . For isolation of liver immune cells , whole livers were mechanically disrupted with glass slides in cRPMI media , triturated , and digested with 0 . 25 mg/ml Collagenase B ( Roche ) and 1 U/ml type I DNase in cRPMI media at 37°C for 45 min . Immune cells were isolated after gradient centrifugation from a 44/56% Percoll ( Sigma Aldrich ) interface , performed red blood cell lysis , washed twice with FACS ( PBS with 1% heat-inactivated fetal bovine serum ) staining buffer and counted . Immune cells were stained with directly-conjugated antibodies specific to APC-CD3ε ( eBiosciences ) , PE-NK1 . 1 ( eBiosciences ) , PE-Cy7 CD11b ( Biolegend ) , FITC-KLRG1 ( Biolegend ) , and APC-Cy7-CD107a ( Biolegend ) . Intracellular IFN-γ staining was performed as previously described [8] . Briefly , lymphocytes were washed and stained with cell surface markers followed by permeabilization-fixation using the Cytofix-Cytoperm Kit ( BD-PharMingen ) and stained with either a Pacific Blue- ( ebiosciences ) or PE-Cy7-conjugated IFN-γ antibody ( Biolegend ) at 4°C for 30 min , washed and analyzed by flow cytometry . Flow cytometry was performed on a BD LSRII machine using BD FACSDiva software . Cell analysis was performed on FlowJo ( v . 9 . 5 . 2 ) software . NK cells were isolated directly from naïve C57BL/6 ( WT ) , Mavs−/− , Ifnar−/− , or DKO splenocytes using the NK cell negative selection Isolation kit ( Miltenyi Biotec ) , according to the manufacturer's instructions ( 2 rounds of purification; greater than 85% purity ) . Bone-marrow derived DC were generated as described previously [8] . Briefly , bone marrow cells from WT , Mavs−/− , Ifnar−/− , or DKO mice were isolated and cultured for 7 days in RPMI-1640 supplemented with granulocyte-macrophage-colony stimulating factor ( 20 ng/ml ) to generate DC . On day 7 , cells were collected and analyzed for CD11b+ and CD11c+ expression and used when the purity was greater than 90% . 1×105 DCs were infected with WNV-TX at an MOI of 5 . 0 for 2 hours , washed with serum-free DMEM , resuspended in cRPMI , and cocultured with purified NK cells at a ratio of 1∶2 in a 96-well U-bottom plate ( 200 µl total volume ) . After 18 hours , cells were removed from the cultures , stained as described above , and analyzed by flow cytometry . Kaplan-Meier survival curves were analyzed by the log-rank test . For in vivo viral burden analysis , Mann-Whitney analysis was used to determine statistical differences . For immune cell analysis , an unpaired Student's t-test was used to determine statistical differences . A P≤0 . 05 was considered statistically significant . All data were analyzed using Prism software ( GraphPad Prism5 ) .
|
West Nile virus ( WNV ) , a mosquito-transmitted RNA flavivirus , is an NIAID Category B infectious agent that has emerged in the Western hemisphere as a serious public health threat . The innate immune effectors that impart restriction of WNV infection are not well defined . WNV infection is sensed by the host RIG-I like receptors ( RLR ) , a class of pattern recognition receptors , to trigger type I interferon ( IFN ) and related innate immune defense programs . Using a systems biology approach , we evaluated the contribution of the RLR and type I IFN signaling pathways in controlling tissue tropism . WNV infection triggers tissue-specific innate immune responses , specifically antiviral effector genes and natural killer ( NK ) cell signaling related genes , which are directly regulated by the combined actions of the RLR and type I IFN signaling pathways . Cocultures of dendritic and NK cells revealed that RLR and type I IFN signaling pathways are essential in promoting NK cell activation during WNV infection . Our observations indicate that combined RLR- and type I IFN-dependent signaling programs drive specific antiviral effector gene expression and programs NK cell responses that , together , serve to restrict WNV tissue tropism .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"immune",
"cells",
"immune",
"activation",
"antigen-presenting",
"cells",
"immunity",
"to",
"infections",
"immunology",
"microbiology",
"emerging",
"viral",
"diseases",
"host-pathogen",
"interaction",
"mechanisms",
"of",
"resistance",
"and",
"susceptibility",
"emerging",
"infectious",
"diseases",
"immune",
"defense",
"animal",
"models",
"of",
"infection",
"inflammation",
"biology",
"immune",
"response",
"systems",
"biology",
"nk",
"cells",
"virology",
"immunity",
"innate",
"immunity",
"antivirals",
"genetics",
"and",
"genomics"
] |
2013
|
A Systems Biology Approach Reveals that Tissue Tropism to West Nile Virus Is Regulated by Antiviral Genes and Innate Immune Cellular Processes
|
While influenza viruses are a common respiratory pathogen , sporadic reports of conjunctivitis following human infection demonstrates the ability of this virus to cause disease outside of the respiratory tract . The ocular surface represents both a potential site of virus replication and a portal of entry for establishment of a respiratory infection . However , the properties which govern ocular tropism of influenza viruses , the mechanisms of virus spread from ocular to respiratory tissue , and the potential differences in respiratory disease initiated from different exposure routes are poorly understood . Here , we established a ferret model of ocular inoculation to explore the development of virus pathogenicity and transmissibility following influenza virus exposure by the ocular route . We found that multiple subtypes of human and avian influenza viruses mounted a productive virus infection in the upper respiratory tract of ferrets following ocular inoculation , and were additionally detected in ocular tissue during the acute phase of infection . H5N1 viruses maintained their ability for systemic spread and lethal infection following inoculation by the ocular route . Replication-independent deposition of virus inoculum from ocular to respiratory tissue was limited to the nares and upper trachea , unlike traditional intranasal inoculation which results in virus deposition in both upper and lower respiratory tract tissues . Despite high titers of replicating transmissible seasonal viruses in the upper respiratory tract of ferrets inoculated by the ocular route , virus transmissibility to naïve contacts by respiratory droplets was reduced following ocular inoculation . These data improve our understanding of the mechanisms of virus spread following ocular exposure and highlight differences in the establishment of respiratory disease and virus transmissibility following use of different inoculation volumes and routes .
Despite reports of conjunctivitis following infection with numerous respiratory pathogens ( including influenza , adenovirus , respiratory syncytial virus , and others ) , research investigating the role of ocular infection in virus pathogenicity and transmissibility has been underrepresented [1]–[4] . Influenza virus represents a highly transmissible respiratory pathogen , resulting in >200 , 000 hospitalizations in the United States annually [5] . While ocular disease is generally rare following influenza virus infection in humans , viruses within the H7 subtype have demonstrated an apparent ocular tropism , with the majority of human infections with H7 influenza viruses associated with conjunctivitis [6] . Moreover , ocular complications have been sporadically documented following seasonal , 2009 H1N1 pandemic , and avian H5N1 virus infections in humans [7]–[13] . Numerous properties allow the eye to serve as both a potential site of influenza virus replication as well as a gateway for the establishment of a respiratory infection . Similar to epithelial cells within the human respiratory tract , human ocular tissue and secreted mucins express sialic acids , the cellular receptor for influenza viruses [14]–[16] . The anatomical proximity between the eye and nasal passages , notably the linkage of both systems via the nasolacrimal duct , facilitates aqueous exchange and provides shared lymphoid tissue between these sites [17] , [18] . Influenza virus can rapidly spread between ocular and respiratory tissues , as was demonstrated in a recent study which detected by RT-PCR live attenuated influenza vaccine ( LAIV ) in nasal washes in humans within 30 minutes of experimental ocular exposure to LAIV-containing aerosols [19] . Well-characterized mammalian models to study extraocular spread following ocular inoculation with influenza virus have been limited to the mouse [20] . The ferret , widely used to study influenza pathogenesis and transmission following intranasal inoculation , has also been recognized as an appropriate experimental model for studies involving the visual system [21]–[23] . A previous study demonstrated H7N3 virus replication in nasal , ocular , and rectal tissue following ocular inoculation in ferrets , but did not comprehensively examine the ability of multiple subtypes to use the eye as a portal of entry or examine virus transmissibility following inoculation by this route [24] . It is clear from epidemiological and laboratory data that ocular exposure to influenza virus can manifest in both ocular and respiratory disease . However , the properties that contribute towards the ocular tropism of select influenza virus subtypes , and the mechanisms of virus spread from ocular to respiratory tract tissue following ocular exposure to influenza virus , are not well understood . Here , we present a ferret model where influenza virus in a liquid suspension is placed on the surface of the eye and massaged across the surface of the eye within the conjunctival sac ( ocular inoculation ) to study the ability of human and avian influenza viruses to cause disease and transmit to naïve animals . We found that both human and avian influenza viruses can mount a productive virus infection following ocular inoculation , attributable to replication-independent drainage of virus inoculum from the site of inoculation to respiratory tract tissue . The viral infection following ocular inoculation was capable of causing severe and fatal disease ( in the case of H5N1 virus ) , but was less transmissible by respiratory droplets ( in the case of seasonal influenza viruses ) compared with infection following inoculation by the traditional intranasal route .
Due to a high degree of similarity in lung physiology and distribution of sialic acids in respiratory tract tissues , the ferret model is frequently utilized to model the kinetics and severity of respiratory disease following administration of human and avian influenza viruses by the intranasal route [23] , [25] . To determine if this homology extends to ocular tissue , we examined the composition of sialic acids on the ferret cornea , which represents a potential site of influenza virus replication following ocular inoculation . Staining of ferret corneal epithelial sheets with biotinylated lectins specific for α2–3 and α2–6 sialic acids revealed a predominance of α2–3-linked sialic acids with relatively weak expression of α2–6 sialic acids on the epithelial surface , an expression pattern similar to human corneal and conjunctival tissue ( data not shown ) [2] , [20] . To assess the pathogenicity of influenza viruses of multiple subtypes following ocular inoculation ( i . o . ) in ferrets , we administered 106 EID50 of each indicated virus in a volume in 100 µl on the corneal epithelial surface of the right eye of an anesthetized ferret and massaged the inoculum across the surface of the eye with the eyelid ( Table 1 ) . Ferrets were observed daily for two weeks for clinical signs of illness ( including fever , weight loss , nasal or ocular discharge ) . Nasal wash ( NW ) and rectal swab ( RS ) samples were collected on alternate days post-inoculation ( p . i . ) and were titered for infectious virus , while conjunctival wash ( CW ) samples were collected alternate days p . i . to measure the incidence and kinetics of infectious virus replication and levels of viral RNA from inoculated eyes ( Tables 1 and 2 , Figures 1 and 2 ) . All virus subtypes tested replicated in ferrets following ocular inoculation , as measured by detectable virus in NW samples as early as day 1 p . i . ( Table 2 and Figure 1 ) . The duration of virus shedding from NW samples and transient fever and weight loss generally mirrored that seen following intranasal ( i . n . ) inoculation for each virus [26]–[29] . However , in comparison to i . n . inoculation , the incidence of nasal discharge was reduced following i . o . inoculation with all influenza viruses tested , and sneezing was less frequent in ferrets following infection with human influenza viruses ( Table 1 ) [30] , [31] . Infectious virus was detected in CW samples from all influenza virus subtypes tested , with levels of viral RNA generally correlating with virus titers ( Figure 3 ) . Ocular inoculation with all H7 influenza viruses tested resulted in elevated peak mean viral titers in NW samples and a higher incidence of detection in CW samples compared with H5N1 viruses ( 45% positive among H7 virus samples compared with 25% of H5 virus samples ) ( Table 2 , Figure 2 ) . However , despite reduced viral replication in NW samples , H5N1 viruses were capable of causing >20% weight loss and lethal disease in ferrets following i . o . inoculation ( Table 1 ) . Interestingly , all seasonal H1N1 and H3N2 viruses evaluated were detected at high titer in both NW and CW samples; i . o . inoculation with Mex/4108 , Brisbane , and Panama viruses , along with the H7N7 NL/230 virus , resulted in the highest peak mean titers in CW samples ( >103 EID50/ml ) compared with all viruses examined ( Table 2 , Figure 2 ) . All viruses with the exception of Thai/16 virus were also detected at low titer in RS samples , with peak titers 101 . 8–102 . 7 EID50/ml observed days 3–7 p . i . ( Table 2 ) . In summary , we found that both avian and human influenza viruses were capable of mounting a productive infection in ferrets following i . o . inoculation , with virus replication observed in samples collected from both ocular and respiratory tract locations regardless of virus subtype . H7 influenza viruses replicated to peak titers >2 logs higher compared with H5 influenza viruses in NW samples following i . o . inoculation , yet H5N1 influenza viruses were capable of maintaining a lethal phenotype following introduction by the ocular route . Seasonal and 2009 H1N1 pandemic influenza viruses efficiently used the eye as a portal of entry to replicate efficiently in the upper respiratory tract as well as ocular tissue . To examine the capacity of influenza viruses to cause severe disease following ocular inoculation , and to better identify those features specific to ocular inoculation , we inoculated ferrets by either the traditional i . n . route ( using a 1 mL inoculation volume ) or the i . o . route ( using a 100 µl inoculation volume ) with 106 EID50 of NL/219 , NL/230 , or Brisbane virus and collected systemic tissues on day 3 p . i . ( Table 3 and 4 ) . While i . n . inoculation with the H7N7 viruses tested in this study results in high virus titers throughout the respiratory tract of ferrets , H7N7 virus dissemination following i . o . inoculation was generally restricted to the upper respiratory tract , with a >3 log reduction in titers in nasal turbinates ( p<0 . 05 ) and only sporadic virus isolation in trachea and lung samples compared with intranasal inoculation ( p<0 . 005 ) ( Table 3 ) . A similar pattern of virus dissemination following H7N7 i . o . virus infection was observed when ferrets were inoculated by the i . n . route using a 100 µl and not 1 mL inoculation volume ( Table 3 ) . The H1N1 virus Brisbane replicated with comparable efficiency in nasal turbinate samples regardless of the inoculation route or volume chosen , but similar to H7N7 virus ocular infections , did not consistently replicate to high titers in lower respiratory tract tissues . Unlike virus dissemination to the respiratory tract , virus spread to the intestinal tract was not contingent on the route or volume of inoculation . Despite restriction of virus following i . o . inoculation to upper respiratory tract tissues compared with traditional 1 mL intranasal inoculation through day 3 p . i . , virus introduced by the ocular route was still capable of causing lethal disease , as ferrets inoculated with the HPAI H5N1 virus Thai/16 by the ocular route required euthanasia days 7–8 p . i . due to development of neurological signs ( Table 1 ) . Ferrets which succumbed to Thai/16 virus infection following ocular inoculation exhibited pronounced lymphopenia in peripheral blood and systemic spread of virus to all regions of the respiratory tract and brain comparable to 1 mL intranasal inoculation , albeit with reduced lethargy and a delayed time-to-death ( [32] and data not shown ) . These data suggest that ferrets inoculated by the ocular route succumb to a similar course of disease as intranasally inoculated ferrets , however following i . o . inoculation there is a delay in both the kinetics of virus dissemination and the development of neurological signs and severe disease , potentially owing to differences in virus inoculum reaching lower respiratory tract tissues at the time of ocular inoculation . Ocular tissue is not routinely titrated following i . n . inoculation of influenza virus in ferrets , making it difficult to elucidate if viral titers in ocular tissue are a function of i . o . inoculation or are detected regardless of the inoculation route . Therefore , we collected both left and right whole ferret eyes and all surrounding conjunctiva/eyelid for virus titration from ferrets inoculated by the intranasal or ocular route with NL/219 , NL/230 , or Brisbane viruses ( Table 4 ) . Surprisingly , sporadic viral titers from both left and right eye and conjunctival tissue were detected following HPAI H7N7 virus infection by both i . n . ( using either a 1 mL or 100 µl inoculation volume ) or i . o . inoculation routes ( Table 4 ) . While the magnitude of viral titers and viral RNA was generally similar between intranasal and ocular routes of inoculation , real-time RT-PCR detected CW-positive samples with a greater sensitivity compared with viral culture . Isolation of virus from ocular tissue may be a reflection of the ability of these HPAI viruses to spread to extra-pulmonary tissues post-inoculation as previously described [26] . However , virus was also detected in left and right conjunctival tissue following i . n . or i . o . inoculation of the H1N1 virus Brisbane , a virus which lacks a high capacity for systemic spread [28] . Comparable levels of viral RNA were isolated from CW samples from ferrets inoculated with Brisbane virus by either intranasal or ocular routes , although infectious virus was only detected in CW samples collected from the eyes of ferrets inoculated by the ocular route . To confirm that virus detected in the eye and conjunctiva was associated with tissue-specific virus replication , immunohistochemistry ( IHC ) was performed to visualize the presence of influenza A nucleoprotein ( NP ) in ferret ocular tissues . As shown in Figure 3 , influenza virus antigen was detected in epithelial cells from both the lacrimal glands in the conjunctiva and the ciliary processes in the eye collected day 3 p . i . from ferrets inoculated by the ocular route . These results indicate that the route of virus inoculation in ferrets can affect the extent of virus dissemination in respiratory tract tissue , but extra-pulmonary spread , notably to ocular tissue , is present regardless of the point of entry once an infection is established . The detection of high viral titers in NW samples as early as day 1 p . i . following i . o . inoculation suggests replication-independent spread of virus from the eye to the respiratory tract ( Figure 1 ) ; this has been similarly hypothesized in previous studies , but has yet to be proven experimentally [20] , [33] , [34] . Reduced viral titers in the lungs of ferrets on day 3 p . i . following ocular compared with intranasal administration further indicates differential patterns of virus spread following inoculation ( Table 3 ) . To visualize the deposition of virus immediately following different routes of inoculation , we labeled NL/219 virus with an AF680 fluorescent tag ( NL/219-FL ) and inoculated ferrets with equal quantities of NL/219-FL virus by the ocular ( 100 µl total volume ) or intranasal ( 1 ml total volume diluted in PBS ) route ( Figure 4 ) . Ferrets were euthanized 15 minutes following virus inoculation for ex vivo imaging . In ferrets inoculated by the traditional intranasal route , the majority of virus was deposited in the nasal turbinates and lungs , consistent with a previous study demonstrating virus dissemination throughout upper and lower respiratory tract tissue following this route of inoculation [32] . In contrast , virus deposition in ferrets inoculated by the ocular route ( right side only ) was primarily localized in the nasal turbinates and right conjunctiva . Lower relative quantities of virus inoculum were present in the upper trachea and esophagus following either intranasal or ocular inoculation . These findings demonstrate that , following i . o . inoculation in ferrets , influenza virus rapidly spreads to the nasal turbinates and upper trachea in a replication-independent manner , but in contrast to i . n . inoculation , does not immediately deposit in peripheral lung tissue . Furthermore , initial deposition of virus inoculum following ocular inoculation occurs not on the corneal surface of the eye but is rather concentrated in the surrounding conjunctival tissue . To determine if ocular exposure to influenza virus results in a transmissible respiratory infection , we inoculated ferrets by the ocular route with selected influenza viruses known to transmit following traditional i . n . inoculation to naïve contacts in the presence of direct contact or by respiratory droplets ( Figure 5 ) . Transmission was assessed by the detection of virus in NW samples and seroconversion of contact ferrets . To assess virus transmission in the presence of direct contact , ferrets were inoculated by the ocular route with the H7N2 virus NY/107 or the H7N7 virus NL/230 , both viruses which transmit efficiently by this route following i . n . inoculation in ferrets [27] . Twenty-four hours later , a naïve ferret was placed in the same cage as each inoculated ferret to assess transmission . Both NY/107 and NL/230 viruses replicated efficiently in the inoculated ferrets following ocular inoculation as expected , and spread to 2/3 and 3/3 contact ferrets by day 7 post-contact ( p . c . ) , respectively ( Figure 5A ) . In addition to high titers of virus in the NW of contact ferrets , NY/107 contact ferrets with detectable virus in NW samples also had detectable infectious virus in CW samples ( 2/3 ferrets , peak titers 100 . 98–2 . 25 EID50 ) , and NL/230 contact ferrets had detectable infectious virus in CW ( 3/3 ferrets , peak titers 100 . 98–2 . 25 EID50 ) and RS ( 2/3 ferrets , peak titers 101 . 98–2 . 75 EID50 ) samples . All NL/230 and NY/107 DC contact ferrets seroconverted by the end of the observation period ( data not shown ) . These results indicate that virus transmission in the presence of direct contact can occur following exposure to ferrets which exhibit a respiratory infection generated by i . o . inoculation , with virus recovery from contact ferrets in both respiratory and ocular samples . To assess virus transmissibility by respiratory droplets in the absence of direct contact , ferrets were inoculated by the ocular route with the H1N1 virus Brisbane or the H3N2 virus Panama , both viruses which transmit efficiently by this route following traditional i . n . inoculation [28] , [30] . Twenty-four hours following i . o . inoculation , a naïve ferret was placed in an adjacent cage with modified side walls , so that air exchange was permitted between inoculated and contact ferrets in the absence of direct or indirect contact . Unlike the efficient transmission observed with these viruses following traditional i . n . inoculation , ferrets inoculated by the ocular route with either Brisbane or Panama only transmitted virus by respiratory droplets to 1/3 contact ferrets ( Figure 5B ) . Virus was not detected in CW or RS samples from the infected Brisbane contact ferret; the infected Panama contact ferret had a peak CW titer of 102 . 75 EID50 day 5 p . c . and peak RS titer of 102 . 5 EID50 day 7 p . c . While RD contacts with detectable virus in NW samples seroconverted to homologous virus at the end of the experimental period , contact ferrets which did not have detectable virus in NW samples did not exhibit seroconversion ( data not shown ) . Ferrets inoculated intranasally with 106 EID50 of Brisbane virus in a reduced 100 µl volume and tested for their ability to transmit virus by respiratory droplets exhibited a similar pattern of virus transmissibility as ferrets inoculated by the ocular route , with virus shedding and seroconversion detected in only 1/3 contact ferrets ( data not shown ) . These findings suggest that despite high titers of virus in NW samples , the respiratory infection resulting from inoculation of ferrets with a reduced volume , by either ocular or intranasal inoculation routes , is distinct from that following traditional i . n . inoculation , characterized by a diminished incidence of sneezing and nasal discharge and resulting in reduced transmission of virus by respiratory droplets .
While the ferret has proved essential for the study of influenza virus pathogenesis and transmission , the use of this species to examine alternate inoculation methods has been limited [32] , [35]–[38] . Characterizing the progression of disease following alternate routes of inoculation with influenza virus will assist in the better understanding of unique features and the relative severity and risk associated with different exposure routes . In this study , we established an in vivo model using the ferret to assess the ability of influenza viruses of multiple subtypes to use the eye as a gateway to establish a productive infection . Both human and avian influenza viruses were capable of mounting a respiratory virus infection in ferrets following i . o . inoculation . The detection of virus in ocular samples collected from ferrets inoculated by either ocular or intranasal routes demonstrates the importance of studying ocular involvement in respiratory virus infection . Divergent patterns of virus transmissibility by respiratory droplets following use of different inoculum volumes and routes of inoculation highlights the complexity of properties which govern virus transmission . The high similarity of respiratory tract tissue between humans and ferrets makes the ferret model an attractive one for modeling human respiratory disease and investigating the role of receptor specificity in influenza virus pathogenesis , providing an advantage over murine models [25] . We found that the sialic acid composition of ferret corneal epithelial sheets more closely mimics that of humans compared with a mouse model , demonstrating another physiological parallel between ferrets and humans [20] . Bridging the α2–3 rich corneal and conjunctival epithelial surfaces with α2–6 rich upper respiratory tract tissue is the lacrimal duct , which expresses both α2–3 and α2–6 linked sialic acids [14] , [34] . Characterization of the distribution of sialic acids in the ferret conjunctiva and lacrimal duct , in addition to the composition of ferret ocular mucins , will allow for a better understanding of virus attachment and replication in these locations . However , in a previous study , we demonstrated that the ability of influenza viruses to bind to or replicate in ocular tissue cannot be explained by sialic acid binding specificity alone [20] . Our detection in ferret ocular tissue of both human and avian influenza viruses with distinct binding specificities further underscores this point ( Table 4 , Figure 3 ) . Macroscopic signs of ocular disease in ferrets were not observed during the course of infection with any virus tested , similar to prior observations in mouse and rabbit models following deposition of influenza virus on the corneal surface [20] , [39] . A previous study in ferrets reported mild conjunctival inflammation following i . o . inoculation with an H7N3 virus , however this may be attributable to strain-specific differences or the use of younger ( 3–5 month old ) ferrets [24] . Despite the absence of visible ocular complications , virus was consistently detected in CW samples from ferrets inoculated by the intranasal or ocular route ( Table 2 , Figure 2 ) . Levels of viral M1 RNA generally correlated with the magnitude of virus isolation , and were a more sensitive detection method compared with virus isolation in CW samples , similar to that observed in human eye swabs ( Table 4 , Figure 2 ) [40] . The presence of virus in RS samples following both i . n . and i . o . inoculation with influenza virus has been previously reported and likely originates from virus swallowed during inoculation [24] , [32] , [41] , as indicated by deposition of virus in the esophagus following initial virus inoculation by both inoculation routes ( Table 3 , Figure 3 ) . Unlike in a murine model , the ferret model supported virus replication of both human and avian influenza viruses following i . o . inoculation [20] . In this ferret model , H7 viruses were detected at higher titer in NW samples and with a higher frequency in ocular CW samples compared with H5N1 viruses , suggesting a recapitulation of the tropism of the H7 virus subtype observed in humans ( Table 2 ) . However , the permissiveness of multiple virus subtypes to cause a productive infection following i . o . inoculation in the ferret points to a greater capacity of influenza viruses to use the eye as a portal of entry in an experimental in vivo setting , just as previous in vitro studies have demonstrated that numerous human ocular cell types distributed throughout the ocular area support infection and replication with both avian and human influenza viruses [42]–[44] . Cumulatively , these previous in vitro studies suggest that the apparent ocular tropism associated with viruses of the H7 subtype is not due to a superior ability to replicate in ocular cells compared with other virus subtypes . Future studies evaluating potential fine receptor specificity differences on the ocular surface and the composition of ocular mucins which may restrict exposure to the ocular epithelial surface to selected virus subtypes may provide a greater understanding of this property . Consistently high titers of human influenza viruses in ocular samples following both i . n . and i . o . inoculation indicates that these H1N1 and H3N2 viruses are not exhibiting a preferential tropism for ocular infection but more likely are a reflection of the high titers observed in the upper respiratory tract in these tissues independent of the initial inoculation route . Specifically , the nasolacrimal duct which links the ocular lacrimal sac to the nasal meatus could serve as a conduit for virus-containing fluid exchange between ocular and respiratory tract tissue [18] . Numerous reports have documented drainage of vaccine or immunizing agents to nasal tissue following topical ocular administration as well as the spread of intranasally administered solutions to the conjunctival mucosal surface [18] , [45] . However , spread of virus from respiratory tract tissue to ocular tissue following i . n . inoculation with human or avian influenza viruses has not been observed previously in the ferret and only sporadically reported in the mouse , possibly due to relatively low titers of virus in nasal tissue or other anatomical differences [20] , [24] , [26] , [46] . Detection of both human and avian influenza viruses in ferret eyes and conjunctival tissue following i . n . inoculation indicates that virus can circulate proximal to the nasal cavity and nasolacrimal canal more readily than previously considered and that more routine collection of ocular tissue during standard virus pathotyping in mammalian models is warranted to better understand the extent of viral ocular dissemination ( Table 4 ) . While it is unlikely that ferret grooming practices are solely responsible for virus spread between these locations , human studies with numerous respiratory viruses have demonstrated the ability to self-inoculate between ocular and respiratory sites , and it is possible that self-inoculation could be further contributing to virus spread in this model [47] , [48] . Inoculation of ferrets by the ocular route with 106 EID50 of selected human and avian influenza viruses in a 20 µl volume resulted in comparable results to those obtained employing a 100 µl inoculation volume , indicating that replication-independent drainage of virus to respiratory tract tissue and subsequent virus detection in NW and CW samples reported in this study was not contingent on the inoculum volume ( data not shown ) . Visualization of virus deposition using fluorescently-tagged virus has allowed for a new understanding of dissemination patterns following in vivo inoculation . Using this technique , we confirmed previously hypothesized reports of replication-independent drainage from ocular to respiratory tract tissue [20] , [33] . Viral load measured in respiratory tract tissues day 3 p . i . following i . n . or i . o . inoculation largely mirrored initial virus deposition patterns , with tissues possessing the greatest quantity of virus reflecting those sites of greatest initial virus deposition during virus inoculation ( Table 3–4 , Figure 4 ) . Both i . o . and i . n . inoculation using a 100 µl volume resulted in detection of virus at high titers in upper and not lower respiratory tract tissues day 3 p . i . , indicating that inoculation with a reduced volume leads to limited initial virus deposition in respiratory tract tissues regardless of inoculation route . The delay in high virus titer recovery from lower respiratory tract tissues during HPAI virus infection after i . o . inoculation and the delay in onset of severe disease and lethal outcome with Thai/16 virus is likely due to replication-dependent spread ( and not deposition ) of virus to the lower respiratory tract , suggesting that during inoculation , the majority of virus was retained in conjunctival tissues , drained to the nasal turbinates , or swallowed and diverted away from the lower respiratory tract; future study of ferret lacrimal tissue in this role is warranted . Comparable delays in onset of severe disease compared with traditional i . n . inoculation were observed in a murine model of i . o . inoculation and a ferret model of aerosol inoculation , despite ultimately similar lethal outcomes [20] , [32] . The finding here of H5N1 subtype viruses using the eye as a portal of entry to initiate a lethal infection , shown previously in a mouse model , underscores the risk of ocular exposure to influenza viruses , even those subtypes not typically considered to have a tropism for this tissue [20] . Despite efficient replication of seasonal H1N1 and H3N2 viruses in the upper respiratory tract of ferrets following i . o . inoculation , these viruses did not result in frequent detection of sneezing and nasal discharge , and did not transmit efficiently to naïve contacts by respiratory droplets ( Table 1 , Figure 5 ) . Infrequent sneezing is commonly observed among influenza viruses which do not transmit efficiently by respiratory droplets and could be contributing to the reduced transmissibility seen here [28] , [30] , [31] . Further research is needed to better understand the virologic and immunologic properties which confer the incidence of sneezing and nasal discharge and the role of these properties in virus transmissibility [49] . Additionally , the efficiency of virus transmission by respiratory droplets following i . o . inoculation was likely influenced by the reduced initial virus deposition and subsequent limited replication in the ferret trachea , as reduced virus transmissibility by respiratory droplets was observed following i . n . inoculation when using a 100 µl but not 1 mL volume ( Table 3 , Figure 4 , and data not shown ) . Despite similarly high virus titers and duration of virus shedding in NW samples , the presence of expelled virus particles originating from tracheal replication which was present during traditional i . n . , but not i . o . or i . n . inoculation using a reduced volume , may have contributed to differing transmission efficiencies between inoculation routes and may represent a previously unrecognized role for virus replication in tracheal tissue in virus transmissibility by respiratory droplets . In contrast , virus transmission in the presence of direct contact did not differ between inoculation routes . The reduced transmissibility of virus following i . o . inoculation is in agreement with epidemiological studies which demonstrate that the majority of human cases of conjunctivitis following H7 influenza virus exposure are self-limiting [50] . Further study evaluating the shedding of virus into the environment among persons infected with influenza viruses which cause respiratory or ocular disease will shed light on potential differences in transmission dynamics independent of virus subtype . The diversity of potential exposures to influenza virus underscores the importance of studying the development of respiratory disease resulting from alternate exposure routes . This knowledge is critical for both a greater understanding of the establishment of influenza virus respiratory disease as well as differences in virus transmission dynamics following differing exposure routes . The facile dissemination of virus inoculum from ocular to nasal tissue , and the detection of virus in both NW and CW samples throughout the acute phase of ferret infection , highlights the ability for concurrent ocular and respiratory disease following influenza virus infection; not surprisingly , reports of conjunctivitis and influenza-like illness in the same individual have been documented during H7 outbreaks resulting in human infection [40] , [51] . While much regarding the properties which regulate the ocular tropism of influenza viruses remains to be determined , our results highlight the potential for a range of influenza A subtypes to initiate infection through the eye and support the use of eye protection during occupational exposure to aerosols containing influenza viruses [19] , [52] , [53] .
This study was carried out in strict accordance with recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health . All ferret procedures were approved by Institutional Animal Care and Use Committee ( IACUC ) of the Centers for Disease Control and Prevention and in an Association for Assessment and Accreditation of Laboratory Animal Care International-accredited facility . Animal studies were performed in accordance with the IACUC guidelines under protocol #2195TUMFERC-A3: “Studies on the Pathogenesis and Transmission of Recombinant Influenza Viruses in Ferrets” . Influenza A viruses of the H7 , H5 , and H1 subtype used in this study are shown in Table 1 . Virus stocks were propagated in the allantoic fluid cavity of 10 day old embryonated hens' eggs as previously described [26]; virus stocks were confirmed by sequencing to be free of mutations . The 50% egg infectious dose ( EID50 ) for each virus stock was calculated by the method of Reed and Muench [54] following serial titration in eggs . Fluorescent-tagged virus ( NL/219-FL ) was generated using formalin-inactivated NL/219 virus and a SAIVI Antibody Alexa Fluor 680 Labeling kit ( Invitrogen ) per manufacturer's instructions as previously described [32] . All experiments with HPAI viruses were conducted under biosafety level 3 containment , including enhancements required by the U . S . Department of Agriculture and the Select Agent Program [55] . Male Fitch ferrets ( Triple F Farms ) , 7 to 10 months old and serologically negative by hemagglutination inhibition to currently circulating influenza viruses , were used in this study . Ferrets were housed in a Duo-Flow Bioclean mobile clean room ( Lab Products ) for the duration of each experiment . Intranasal ( i . n . ) inoculations were performed under anesthesia as previously described using 106 EID50 of virus diluted in PBS in a 1 ml or 100 µl volume [29] . Ocular ( i . o . ) inoculations were performed under anesthesia using 106 EID50 of virus diluted in PBS in a 100 µl or 20 µl volume . Virus inoculum was administered dropwise to the surface of the right eye of each ferret and massaged over the surface of the eye by the eyelid . Ferrets were monitored daily post-inoculation for morbidity and clinical signs of infection as previously described [29] . Any ferret which lost >25% of its pre-inoculation body weight or exhibited neurological dysfunction was euthanized . Virus shedding was measured on alternate days post-inoculation ( p . i . ) in nasal washes ( NW ) , conjunctival washes ( CW ) , and rectal swabs ( RS ) . NW and RS samples were collected as previously described [28] , [29] . CW were obtained by bathing the inoculated ( right ) ferret eye three times with 500 µl wash solution ( PBS containing Pen/Strep , Gentamycin , and BSA ) and collecting the run-off in a small petri dish , then swabbing the surface and surrounding conjunctival tissue of the right eye with a pre-wettened cotton swab for 5 seconds , and placing the swab in a collection tube containing the run-off liquid . All samples were immediately frozen on dry ice and stored at −70°C until processed . To assess virus dissemination following i . n . or i . o . inoculation , three ferrets per group were inoculated with indicated viruses and euthanized 3 days p . i . for postmortem examination and collection of tissues for virus titration as previously described [29] . Tissue specimens were collected for virus titration were immediately frozen on dry ice and stored at −70°C until processed . Blood samples collected during necropsy were subjected to complete blood counts ( CBCs ) and serum chemistry analyses performed per manufacturer's instructions as previously described [56] . Virus transmissibility following i . o . inoculation was assessed by inoculating ferrets by the ocular route with indicated viruses and , 24 hrs p . i . , placing a naïve ferret in the same cage as an inoculated ferret [to assess transmission in the presence of direct contact ( DC ) ] or in an adjacent cage with modified side-walls to allow air exchange between inoculated and contact animals via perforations but inhibiting direct or indirect contact between animals [to assess transmission by respiratory droplets ( RD ) ] as previously described [30] . For each i . o . transmission experiment , an aliquot of each virus stock used to characterize transmissibility in previous publications by the i . n . route was tested . NW , CW , and RS samples were collected on alternate days p . i . /post-contact ( p . c . ) to assess kinetics of virus shedding . Serum was collected days 17–21 p . i . /p . c . to measure seroconverison . Animal research was conducted under the guidance of the Centers for Disease Control and Prevention's Institutional Animal Care and Use Committee in an Association for Assessment and Accreditation of Laboratory Animal Care International-accredited animal facility . NW , CW , and RS samples were serially titrated in eggs , starting at a 1∶10 dilution ( NW , RS; limit of detection , 101 . 5 EID50/ml ) or 1∶2 dilution ( CW; limit of detection , 100 . 8 EID50/ml ) . Virus infectivity for all samples was calculated by the method of Reed and Muench [54] . At the time samples were thawed for virus titration , RNA was extracted from CW samples using a QIAamp Viral RNA kit ( Qiagen ) . Real-time RT-PCR was performed with a QuantiTect SYBR Green RT-PCR kit ( Qiagen ) using an influenza A virus M1 gene primer set to determine viral load [32] . Viral RNA copy numbers were extrapolated using a standard curve based on samples of known virus as previously described [32] , [57] . Baseline levels were determined by collecting CW samples from uninfected ferrets . Tissue specimens were homogenized in 1 ml cold PBS using disposable sterile tissue grinders and clarified by centrifugation before serial titration in eggs , starting at a 1∶10 dilution . Ferret eye and conjunctival tissues were rinsed with PBS prior to virus titration . Eye , conjunctival , and nasal turbinate tissues are expressed as EID50/ml , while all other tissues are expressed as EID50/g . Uninfected ferret corneal epithelial sheets were dissociated from excised whole ferret eyes following incubation in tetrasodium EDTA for determination of expression of surface sialoligosaccharides as previously described [20] , [58] . To assess virus dissemination , ferrets were inoculated with NL/219-FL virus either i . o . ( 100 µl ) or i . n . ( 1 ml ) as previously described [32] . Fifteen minutes p . i . , ferrets were euthanized and respiratory and ocular tissues were excised for ex vivo imaging using a Spectrum in vivo imaging system and Living Image 4 . 0 Software ( Caliper Life Sciences ) . All ex vivo imaging was performed in triplicate . To quantify the presence of NL/219-FL virus in excised tissues , regions of interest were drawn around each tissue using Living Image 4 . 0 Software to obtain maximum relative efficiency values for each tissue , expressed as photons/second/cm2/steradian , the mean of which was generated from three ferrets per tissue as expressed in Figure 3 . Tissues for immunohistochemistry ( IHC ) were collected day 3 p . i with the viruses indicated . or from uninfected ferrets , fixed by submersion in 10% neutral buffered formalin for 3 days , routinely processed , and embedded in paraffin . Immunohistochemical detection of influenza A virus nucleoprotein was performed as described previously [59] . The Pearson product-moment correlation coefficient was generated to measure the correlation between viral titer and viral RNA copy number in CW sample using GraphPad Prism 5 . 0 ( GraphPad Software , Inc . ) . Statistical significance for all other experiments was determined using Student's t test .
|
Most infections with influenza virus result in respiratory disease . However , influenza viruses of the H7 subtype frequently cause ocular and not respiratory symptoms during human infection , demonstrating that the eye represents an alternate location for influenza viruses to infect humans . Using a ferret model , we studied the ability of influenza viruses to cause disease following ocular inoculation . We found that both human and avian influenza viruses could use the eye as a portal of entry to establish a respiratory infection in ferrets . Influenza viruses were also detected in ocular samples taken from ferrets during virus infection . We identified that influenza viruses spread to different tissues in ferrets when inoculated by ocular or respiratory routes , and that these differences affected the transmissibility of influenza viruses in this model . This study is the first to confirm that virus can spread from the eye to the respiratory tract in a replication-independent manner , and offers greater insight in understanding the ability of influenza viruses of all subtypes to cause human infection by the ocular route .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"virology",
"biology",
"microbiology"
] |
2012
|
Influenza Virus Respiratory Infection and Transmission Following Ocular Inoculation in Ferrets
|
Viral infection leads to induction of pattern-recognition receptor signaling , which leads to interferon regulatory factor ( IRF ) activation and ultimately interferon ( IFN ) production . To establish infection , many viruses have strategies to evade the innate immunity . For the hepatitis B virus ( HBV ) , which causes chronic infection in the liver , the evasion strategy remains uncertain . We now show that HBV polymerase ( Pol ) blocks IRF signaling , indicating that HBV Pol is the viral molecule that effectively counteracts host innate immune response . In particular , HBV Pol inhibits TANK-binding kinase 1 ( TBK1 ) /IκB kinase-ε ( IKKε ) , the effector kinases of IRF signaling . Intriguingly , HBV Pol inhibits TBK1/IKKε activity by disrupting the interaction between IKKε and DDX3 DEAD box RNA helicase , which was recently shown to augment TBK1/IKKε activity . This unexpected role of HBV Pol may explain how HBV evades innate immune response in the early phase of the infection . A therapeutic implication of this work is that a strategy to interfere with the HBV Pol-DDX3 interaction might lead to the resolution of life-long persistent infection .
Hepatitis B virus ( HBV ) is the prototypic member of the hepadnavirus family and a major cause of liver diseases . An estimated 400 million people are persistently infected with HBV worldwide . A significant subset of these HBV carriers progresses to severe liver disease , such as hepatocellular carcinoma , which may cause up to one million deaths per year . Interferon and nucleoside analogs such as lamivudine and adefovir are used to treat chronic hepatitis B patients but have limited utility due to the adverse effect and the emergence of drug-resistant variants , respectively [1] . Thus , there is a clear medical need for new therapeutic strategies . Viral infection leads to the initiation of antiviral innate immune responses resulting in the expression of type I interferons ( IFNs ) , IFNα and IFNβ , and pro-inflammatory cytokines [2] . Recently , the cellular mechanisms used to detect viruses and elicit production of IFNs and pro-inflammatory cytokines have been described in detail . It is now well-established that viruses , similar to bacteria and fungi , are initially recognized by host pattern-recognition receptors ( PRRs ) [2] , [3] . Viral nucleic acids ( both RNA and DNA ) are the most important pathogen-associated molecular patterns ( PAMPs ) recognized by PRRs [3] . Two families of PRRs have been defined . The first is a subfamily of Toll-like receptors ( TLRs ) that include TLR3 , TLR7 , TLR8 , and TLR9 , which are mainly expressed in the endosomes of some cell types , especially plasmacytoid dendritic cells . Recognition by TLRs of viral PAMPs initiates TLR-mediated signaling pathways that culminate in the activation of transcription factors NFκB , IRF3 , and IRF7 . Specifically , TLRs recruit signaling adaptors , including TIR-domain-containing adaptor protein inducing IFNβ ( TRIF ) . This activates TANK-binding kinase 1 ( TBK1 ) /IκB kinase-ε ( IKKε ) to phosphorylate and activate the transcription factors IFN-regulatory factors ( IRF ) 3 and 7 [2] , [4] . The second family of PRRs are comprised of the retinoic-acid inducible gene I ( RIG-I ) -like receptors ( RLRs ) , including RIG-I and melanoma differentiation-associated gene 5 ( MDA-5 ) [3] . Similar to TLRs , the recognition of viral nucleic acids by RLRs leads to a cascade of signaling events that result in activation of NF-κB , IRF3 , and IRF7 . Specifically , RLRs recruit the signaling adaptor protein IFNβ-promoter stimulator 1 ( IPS-1 , also known as MAVS , VISA , or Cardif ) , activating the downstream TBK1/IKKε kinases , which then phosphorylate and activate IRF3 and IRF7 [2] . The capacity of both signaling pathways to restrict viral replication is consistent with their downstream convergence at the TBK1/IKKε kinases responsible for activation of IRF-3 . As stated above , viral infection leads to activation of cellular signaling such as IRF signaling , which culminates in IFN production . Infection by HBV appeared to be an exception . In an acute HBV-infected chimpanzee model , Chisari and colleagues have reported that HBV fails to induce transcription of any cellular genes that relate to the entry and expansion of the virus , implicating the lack of innate immune response upon HBV infection [5] , [6] . By contrast to the earlier report , evidence was accumulating , which indicated that the innate immune system is , in fact , able to sense HBV infection . An early induction of innate immune response , as shown by the early development of natural killer cell and natural killer T cells response , was observed in two patients with acute viral infection [7] . More recently , by using a HepaRG cells , a permissive hepatocyte cell line for HBV infection , Zoulim and colleagues found that HBV infection elicits a strong innate antiviral response that leads to a significant reduction of HBV DNA synthesis [8] . Taken together , it is conceivable that one of viral proteins could impair innate immune response early in infection . Two recent reports have shown that DDX3 DEAD box RNA helicase , which is known to be involved in diverse steps of RNA metabolism , could augment IRF signaling via its interaction with IKKε or TBK [9] , [10] . In other words , DDX3 augments TBK/IKKε activity , which phosphorylates IRF3 and IRF7 . Interestingly , our group has shown that DDX3 binds to HBV Pol ( P protein ) and inhibits viral reverse transcription [11] . Since DDX3 is essential for augmentation of IRF signaling , we postulated that HBV Pol impairs antiviral innate immune responses by inhibiting IRF signaling via its interaction with DDX3 . Consistent with the notion , our results demonstrate that IRF signaling is significantly inhibited by HBV Pol , a finding that defines HBV Pol as a viral protein that counteracts antiviral pattern recognition receptor signaling .
To determine if HBV proteins could impair IRF signaling and counteract host innate immune responses to HBV infection , we tested the ability of HBV proteins to inhibit interferon ( IFN ) β promoter activity . Synthetic dsRNA mimic polyinosine-polycytidylic acid ( poly I∶C ) is recognized by TLR3 in endosomes when added to medium [12] . To induce IRF signaling , human hepatoma HepG2 cells , which support viral genome replication , were transfected with one of three viral protein expression constructs — core , polymerase ( Pol ) , or HBx — , a TLR3 expression construct , and a reporter construct expressing luciferase under control of the IFNβ promoter ( Fig . 1A ) . Cells were complemented with the TLR3 expression construct because HepG2 cells are deficient in TLR3 expression [13] . To ensure that the physiological levels of viral proteins were attained , the amount of the three viral protein expression constructs for transfection were determined by cotransfection with the corresponding gene-null HBV replicon: for HBV Pol ( see Fig . S2 ) , for core protein ( data not shown ) , and for HBx [14] . Two days following transfection , cells were treated with poly I∶C . Eight hours post-poly I∶C treatment , cells were harvested and luciferase activity was measured . As anticipated , the luciferase activity was significantly induced when cells were complemented with TLR3 , compared to that induced by poly I∶C only ( Fig . 1A ) . The data indicated that HepG2 cells were able to recognize extracellular dsRNA and induce IFN production when complemented with TLR3 . More importantly , the data showed that HBV Pol , but not other viral proteins including HBx , significantly suppressed IFN promoter activity ( Fig . 1A ) . The inhibition of IFN promoter activity by HBV Pol was somewhat unexpected , since current literature suggests that almost all of its known function is confined to the functions – encapsidation and viral reverse transcription – occurring inside of nucleocapsids [15] . Further , reverse transcriptase activity of HBV Pol appeared not to be involved in the inhibition of IFN promoter activity , as the YMHD mutant of HBV Pol , reverse transcriptase activity deficient mutant , remained to inhibit IFN promoter activity ( Fig . 1A ) . In addition , we found that three viral envelope glycoproteins– L-HBsAg , and M-HBsAg , and S-HBsAg– had no impact on IFN promoter activity ( Fig . 1B ) , indicating that HBV Pol is the only viral protein that has an inhibitory effect on IFNβ production . To substantiate the above results in the context of viral life cycle , the impact of viral proteins on IFNβ production was examined by using a viral replicon , which could lead to viral genome replication when transfected [14] . Three mutants were made in which one viral gene was inactivated per construct: ( i ) P ( Pol ) -null , ( ii ) C ( core ) -null , and ( iii ) X-null ( Fig . S1 ) . An increase in IFNβ production by one of the HBV mutant constructs would point out that particular gene in the inhibition of IFN production . Luciferase activity was monitored following transfection as described above . The data revealed that only cells transfected with the HBV P-null replicon construct induced a higher level of IFN production , whereas the other three replicons , including the wild-type , induced modest level of IFN production ( Fig . 1C ) . These results suggested that HBV Pol derived from the HBV replicon – wild-type or X-null , or C-null replicon – decreased IFN production , implicating the physiological relevance of the findings in the viral life cycle . Intriguingly , IFN promoter activity induced by HBV P-null replicon was significantly higher than what was achieved by poly I∶C ( Fig . 1C ) . The implication is that some yet-to-be known viral PAMPs derived from HBV P-null replicon could contribute to the augmented IFN promoter activity ( see Discussion ) . To examine whether HBV Pol could suppress diverse PAMP-mediated signaling , poly I∶C was given to cells by two distinct routes: ( i ) addition to medium , and ( ii ) transfection via lipofectin . The poly I∶C is recognized by TLR3 in endosomes by adding it to the cell medium , whereas it is recognized by MDA5 when it is delivered to the cytoplasm by transfection via lipofectin [16] , [17] . Cells were transfected with incremental dose of the HBV Pol and TLR3 expression constructs along with the reporter construct . Following treatment with poly I∶C in the medium , the reporter assay result showed a dose-related decrease of IFN promoter activity by HBV Pol expression , corroborating the above conclusions ( Fig . 2A ) . Likewise , when poly I∶C was given by lipofectin transfection , IFN promoter activity was similarly decreased by HBV Pol ( Fig . 2B ) . Next , innate immune response was triggered by Sendai virus ( SenV ) , a potent stimulus of the RIG-I pathway [17] , [18] . Cells were transfected with the HBV Pol expression construct and the IFN-luciferase construct . Two days after transfection , cells were treated with 100 HA U/mL of SenV for 6 h before harvest . The data indicated that HBV Pol diminished IFN promoter activity in a dose-related manner ( Fig . 2C ) . Overall , the data presented suggest that HBV Pol suppresses RIG-I mediated IFN production as well as TLR3-mediated IFN production . To substantiate the above findings , we wanted to eliminate a possibility that the reduction of the IFN promoter activities by HBV Pol is ascribed to the over-expression of HBV Pol . To assess whether the HBV Pol we expressed ranges the physiological level , we sought to show that the amount of viral DNAs synthesized via complementation by the HBV Pol in the HBV P-null replicon transfected cells is comparable to that of WT HBV replicon transfected cells . To this end , HepG2 cells were transfected with the same four increasing amount of the HBV Pol expression construct as above to complement the P-null construct ( i . e . , P- ) for the viral genome replication . Viral DNAs extracted from cytoplasmic capsids were measured by Southern blot analysis ( Fig . S2 ) . The data showed that the amount of viral DNA synthesized by complementation was less than that of WT HBV replicon ( Fig . S2 , lane 1 versus lanes 3 to 5 ) , suggesting that the HBV Pol expressed in the Fig . 2A to Fig . 2C was not exceeding physiological level . In addition , by using HepG2 . 2 . 15 cell line that stably expresses viral proteins and support HBV replication [19] , we consistently found that IFN promoter activity induced upon SenV ( Sendai virus ) infection was pronouncedly diminished in HepG2 . 2 . 15 cells ( Fig . 2D ) , validating the impact of HBV Pol in a more physiological setting . Phosphorylation and nuclear translocation of IRF3 represent hallmarks of antiviral innate immunity . To examine whether HBV Pol inhibits the phosphorylation of IRF3 , cells were transfected with TLR3 and HBV Pol construct as indicated . Eight hours before harvest , cells were then treated with poly I∶C . Endogenous IRF3 and its phosphorylated counterparts were detected by Western blot analysis with anti-IRF3 and anti-phosphorylated IRF3 ( Ser396 ) antibodies , respectively ( Fig . 3A ) . As anticipated , Western blot analysis indicated that the higher molecular weight bands of endogenous IRF3 , which represent phosphorylated IRF3 , appeared when cells were treated with poly I∶C ( Fig . 3A , lane 2 ) . However , the phosphorylated IRF3 was undetectable following HBV Pol expression ( Fig . 3A , lane 3 ) , suggesting that HBV Pol inhibits phosphorylation of IRF3 . ( see below for the explanation for lane 4 ) . Nuclear translocation of IRF3 was examined by fluorescence microscopy . Cells were transfected with the IRF3-GFP fusion protein construct to monitor IRF3 nuclear localization ( Fig . 3B ) . Cells were also transfected with HBV Pol and TLR3 constructs and then treated with poly I∶C . As previously demonstrated [20] , IRF3-GFP predominantly found in the cytoplasm was induced to undergo nuclear localization only when cells were treated with poly I∶C ( Fig . 3B , panel a versus panel b ) . By contrast , IRF3-GFP remained localized in cytoplasm when cells were transfected by HBV Pol and treated with poly I∶C ( Fig . 3B , panel c ) , suggesting that HBV Pol prevents the nuclear translocation of IRF3-GFP . Overall , these data are consistent with the conclusion that HBV Pol inhibits the IRF signaling . ( see below for the explanation for panel d ) . To further elucidate the specific mechanism by which viral HBV Pol interferes with IRF activation , the impact of HBV Pol on the signaling pathway leading to IRF activation was investigated . To trigger RIG-I mediated IRF3 signaling , RIG-I was over-expressed . On the other hand , TRIF is an adaptor for the TLR3 receptor and mimics TLR3 signaling when over-expressed . To trigger TLR-mediated IRF3-signaling , TRIF over-expression was employed . Cells were transfected with the IRF3 reporter construct and an incremental dose of HBV Pol expression construct , along with either RIG-I or TRIF ( Fig . 4A and Fig . 4B ) . The data indicated that the luciferase activity was decreased by HBV Pol in a dose-dependent manner ( Fig . 4A and Fig . 4B ) . Thus , we concluded that HBV Pol inhibited both TLR-mediated and RIG-I-mediated IRF3 signaling . The TBK1/IKKε complex represents the effecter protein kinase of IRF signaling , phosphorylating IRF3/7 , when activated by appropriate recruitment of IPS/MAVS or TRIF ( Fig . 5A ) . The results shown above indicated that both TLR3- or RIG-I-mediated IRF signaling was inhibited by HBV Pol , and the phosphorylation of IRF3/7 was blocked by HBV Pol . Thus , it is conceivable that HBV Pol inhibits TBK1/IKKε activity . To gain further insight into the mechanism by which HBV Pol interferes with TLR3-mediated and RIG-I-mediated IRF signaling , we determined if TBK1 or IKKε triggered IRF signaling could also be blocked by HBV Pol . Cells were transfected with either TBK1 or IKKε , and IRF3 reporter construct along with incremental doses of the HBV Pol construct . Results indicated that IRF3 signaling was triggered either by TBK1 or IKKε over-expression and was blocked by HBV Pol in a dose-dependent manner ( Fig . 5B and 5C ) . Essentially identical data were obtained with an IRF7 reporter assay performed in parallel ( data not shown ) . Thus , the data suggested that HBV Pol exerts its inhibitory effect on TBK1/IKKε or another downstream point in this pathway . An ISRE reporter assay was then carried out , which indicated that HBV Pol inhibited IKKε-triggered ISRE activation , but not IRF3-triggered ISRE activation ( Fig . 5D and 5E ) . These results pointed to TBK1/IKKε as the molecular target of HBV Pol for the inhibition of IRF signaling . Next , we examined whether HBV Pol inhibits TBK1/IKKε activity directly or indirectly . Recently , two independent groups demonstrated that DDX3 enhanced TBK1/IKKε activity via its interaction with TBK1 or IKKε [9] , [10] . Additionally , we have shown that DDX3 binds to HBV Pol [11] . Thus , we hypothesized that HBV Pol would inhibit TBK1/IKKε activity via interaction with DDX3 . One prediction of the hypothesis is that over-expression of DDX3 would restore IRF signaling that has been inhibited by HBV Pol . Previous studies were limited to HEK293 and RAW264 . 7 macrophages; therefore , we confirmed the importance of DDX3 for the activation of IRF signaling in HepG2 cells following downregulation of DDX3 ( Fig . S3A and Fig . S3B ) . To determine if DDX3 could restore IRF3 signaling suppressed by HBV Pol , cells were transfected by either TBK1 or IKKε constructs along with a maximal level of HBV Pol to obtain the highest level of IRF signaling inhibition . Cells were also cotransfected with increasing doses of the DDX3 construct to determine if DDX3 could restore the diminished IRF signaling . The luciferase data revealed that the ectopic expression of DDX3 rescued the IRF signaling in a dose-dependent manner ( Fig . 6A and 6B ) , indicating that DDX3 antagonizes the inhibitory effect of HBV Pol on IRF signaling . Consistently , ectopic expression of DDX3 also rescued the phosphorylation and nuclear translocation of IRF3 ( Fig . 3A , lane 4; Fig . 3B , panel d ) . Given the functional interaction between HBV Pol and DDX3 [11] , it is possible that HBV Pol disrupts the IKKε-DDX3 interaction . To assess this possibility , a co-immunoprecipitation ( co-IP ) following transfection of Flag-IKKε , HA-DDX3 , and HBV Pol expression constructs was performed ( Fig . 6C ) . Similar to previous reports [9] , an interaction between IKKε and DDX3 was observed ( Fig . 6C , lane 3 ) . Importantly , when the HBV Pol construct was cotransfected , the IKKε-DDX3 interaction was significantly diminished , consistent with an interpretation that HBV Pol disrupts the IKKε-DDX3 interaction ( Fig . 6C , lane 4 ) . It was also noted that DDX3 level was enhanced and a higher molecular weight form of DDX3 was evident when IKKε was ectopically expressed , which might represent a phosphorylated form ( Fig . 6C , lanes 3 and 4 ) ; intriguingly , the similar changes of DDX3 were also observed in the previous report [9] . It is conceivable that the phosphorylation of DDX3 by IKKε stabilizes DDX3 . Based on the antagonistic activity of DDX3 on the inhibitory effect of HBV Pol on IRF signaling and the disruption of IKKε-DDX3 interaction by HBV Pol , we concluded that HBV Pol suppresses IRF signaling by disrupting the IKKε-DDX3 interaction .
HBV , that has been dubbed a “stealth virus” , efficiently evades antiviral innate immune responses early in infection [6] . However , the underlying immune evasion mechanism remained enigmatic . The data presented here revealed that HBV evades host innate immune response via the inhibition of pattern recognition receptor signaling by one of the viral proteins . Results presented here clearly show that HBV Pol is the viral protein that blocks the TLR3-and RIG-I-induced pattern recognition receptor signaling in a physiologically relevant setting ( Fig . 2 ) . Further , evidence suggests that HBV Pol suppresses IRF signaling by inhibiting TBK1/IKKε activity , the effector protein kinase of IRF3/7 signaling ( Fig . 3 ) . Importantly , we also demonstrated that HBV Pol inhibits TBK1/IKKε activity by disrupting the interaction between IKKε and DDX3 ( Fig . 6C ) . Overall , besides its inherent catalytic role in viral reverse transcription , our results here confer a novel immune-regulatory role to HBV Pol . Recent evidence obtained in cultured cells showed that HBV is capable of inducing innate host response [8] , [21] . However , it remained to be learned how HBV abrogates innate immune response . The data presented here indicate that HBV Pol blocks both TLR-mediated and RIG-I-mediated IRF3 signaling ( Fig . 1 and Fig . 2 ) . Importantly , two experiments presented here supported the physiological relevance of the impact of HBV Pol on innate immune response . First , the data in Fig . 1C showed that the inhibition of IRF signaling by HBV Pol is physiologically relevant , since the inhibitory effect by HBV Pol was observed under the condition , where the physiological level of viral proteins was attained by using either wild-type or mutant HBV replicon in HepG2 cells ( Fig . 1C ) . This finding is further strengthened by the evidence that IFN production was only marginally induced upon SenV infection in HepG2 . 2 . 15 cell line compared to a parental HepG2 cells ( Fig . 2D ) . Further , the data shown in Fig . 1C is significant in a few respects . It was not unexpected that the extent of inhibition attained by the X-null and C-null construct was comparable to that by the wild-type replicon ( Fig . 1C ) , since similar level of HBV Pol would be expressed in transient transfection setting , regardless of whether viral genome replication is induced . Secondly , the data clearly ruled out a possibility that either X protein ( i . e . , HBx ) or core protein is related to the inhibition , since IFNβ production in either X-null or C-null transfected cells was comparable to that of WT transfected cells ( Fig . 1C ) . Thirdly , the cells transfected by the P-null replicon mounted significantly and reproducibly higher IFN production than that by mock DNA , suggesting that a viral molecule derived from the P-null replicon caused augmentation of IFNβ production , in addition to what has already been induced by poly ( I∶C ) ( Fig . 1C ) . It is tempting to speculate that the viral molecule , perhaps viral RNAs , may represent viral PAMPs that caused enhanced IFNβ production . Although the liver is an important site for chronic viral infection , little is known about how innate immune response is initiated in hepatocytes [13] , [22] . We chosen HepG2 cell for the most experiments , a hepatoma cell line , that supports HBV genome replication in a manner depending on HBx expression [14] . Since TLR3 expression is deficient in HepG2 cells , IFNβ reporter assays were carried out following complementation of TLR3; note that HepG2 cell is proficient in RIG-I mediated signaling , which senses SenV [13] . On the other hand , unlike to most other viruses , the PAMP signature of HBV , which are sensed by pattern recognition receptors such as TLRs or RIG-I , has not been defined [2] . Consequently , our analysis was limited to use of poly I∶C or SenV to trigger pattern recognition receptor signaling . However , it should be noted that the utilization of heterologous inducers such as poly I∶C or SenV in this work does not invalidate our conclusions , since the HBV Pol inhibits TBK/IKKε kinase , which lies downstream of PAMPs in IRF signaling ( Fig . 5A ) . Nonetheless , natural viral infection , rather than transfection , involving susceptible human hepatocytes , merit further investigation , as recently demonstrated in HCV infection [23] . In terms of the impact of HBV Pol on IFN signaling , three observations are worthy of mentioning . First , our preliminary data indicated that in addition to its inhibitory effect on IRF signaling , HBV Pol inhibits , to a lesser extent , NF-kB signaling , which also contributes to IFNβ production ( data not shown ) . If it holds , it would suggest an intriguingly possibility in that HBV Pol blocks IFNβ production by interfering two distinct signaling pathways , leading to IFNβ production . Work is in progress to obtain the mechanistic details . Second , relevantly , abundant detection of HBV Pol in nonencapsidated state has implicated that HBV Pol could contributes to viral pathogenesis or immune evasion [24] , [25] . Lastly , in fact , HBV Pol has been previously identified as one of viral proteins that confer the resistance to IFN treatment [26] , [27] . It should be noted , however , that we found its inhibitory role in IFN induction ( i . e . , IRF signaling ) , whereas the published work found its inhibitory effect in IFN action ( i . e . , JAK/Stat signaling ) . DEAD-box RNA helicases constitute a large family of proteins that comprises at least 38 members in human genome [28] . DEAD-box RNA helicases exhibit multiple roles in diverse aspects of RNA metabolism such as transcription , pre-mRNA splicing , RNA export , translation , and RNA decay [29] . Not surprisingly , DDX3 has also been implicated in multiple distinct cellular processes as well . First , DDX3 has been reported to act as a transcriptional factor in the nucleus [30] . Secondly , DDX3 was shown to bind to eIF4E , a translation initiation factor with cap-binding properties , effectively suppressing translation [31] . Finally , DDX3 has been implicated in various viral life cycles . For instance , DDX3 was shown to be essential for nuclear export of human immunodeficiency virus-1 ( HIV-1 ) RNA through the Rev/RRE pathway [32] . In addition , DDX3 supports viral replication of hepatitis C virus ( HCV ) genome via its interaction with the HCV core protein [33] , [34] . In contrast , DDX3 was shown to inhibit HBV genome replication via its interaction with HBV Pol [11] . Besides its roles in RNA metabolism , a novel function relevant to innate immunity was recently reported by two groups [9] , [10] . Although the augmentation of IRF signaling by DDX3 was found independently by two groups , discrepancies have been noted regarding the specific mechanism for DDX3-mediated IRF3 activation . Specifically , Bowie and colleagues [9] demonstrated that the IKKε-DDX3 interaction is significantly enhanced upon SenV infection with concomitant IRF3 phosphorylation , indicating that DDX3 stimulates the protein kinase activity of IKKε . In contrast , Decker and colleagues [10] concluded that DDX3 acts as a transcription factor of the IFNβ promoter , which is in agreement with the transcriptional role of DDX3 reported by another study [30] , [35] . Interestingly , we have demonstrated that HBV Pol disrupts the IKKε-DDX3 interaction ( Fig . 6C ) , which is consistent with the conclusions of Bowie and colleagues ( 7 ) . However , in-depth analyses are needed to clarify this mechanistic issue . Throughout evolution , viruses have developed strategies to evade host immune response [36] . Bowie and colleagues [9] revealed that the vaccinia virus K7 protein interferes with IRF signaling by inhibiting TBK1/IKKε activity via a mechanism involving its interaction with DDX3 . Here , we demonstrated that HBV Pol impairs IRF signaling by inhibiting TBK1/IKKε activity via the HBV Pol-DDX3 interaction . Although diverse viral proteins including structural proteins and nonstructural proteins have evolved to evade immune response [36] , it is intriguing that viral polymerase , besides its inherent catalytic contribution , has evolved to interfere innate immune response . More importantly , it is interesting that vaccinia virus and HBV , two unrelated viruses , acquired the ability to evade the immune response by subverting DDX3 during evolution . Administration of TLR ligands was shown to inhibit HBV replication in a transgenic mouse model , implying that pattern recognition receptor signaling could be exploited for the treatment of chronic HBV infections [37] . Subsequent studies corroborated the above findings either by transfection studies using hepatoma cell lines [38] or by using nonparenchymal liver cells from mice [39] . Along this line , we speculate that disruption of the HBV Pol-DDX3 interaction by therapeutic intervention could invoke sustained antiviral immune responses leading to resolution of chronic viral infections . In this regard , structural elucidation of the HBV Pol-DDX3 interaction merits further investigation . This work was presented at the International Meeting on Molecular Biology of Hepatitis B Viruses , which was held in Tours , France from August 30 to September 3 of 2009 [40] and the result similar to ours will be reported by others [41] .
HepG2 , HepG2 . 2 . 15 , and HEK293 cells were grown in Dulbecco's Modified Eagle's medium supplemented with 10% fetal bovine serum ( GIBCO-BRL ) and 10 µg/mL gentamycin at 37°C in 5% CO2 and were passaged every third day . Cells were transfected using polyethylenimine ( 25 kDa , Aldrich , St . Louis , MO ) as described [42] . The amount of plasmid DNA transfected ( 12 µg per 60-mm plate and 30 µg per 100-mm plate ) was kept constant by inclusion of vector DNA , pcDNA3 . HEK293 cells were employed for the experiment shown in Fig . 3 and Fig . 6 , because of its higher transfection efficiency ( >80% ) compared to that of HepG2 cells ( <10% ) ( data not shown ) . Polyinosine-polycytidylic acid ( poly I∶C ) was purchased from Amersham . Sendai virus ( SenV ) was kindly provided by Prof . Moon-Jung Song ( Korea University ) . All DNA constructs were generated by overlap extension PCR protocols as previously described [43] . HBV Pol expression construct encoding HBV Pol with three copies of Flag tag at its N-terminus has been previously described [42] . The HBV replicon construct ( i . e . , WT ) and its X-null version has been described previously [14] . The HBV C-null and P-null replicons have been described [42] . YMHD mutant of HBV Pol was made by substitution of the aspartic acid residue ( i . e . , 540D ) , a constituent of YMDD motif critical for catalysis , to histidine ( H ) , as previously described [44] . Three constructs that was used to express the viral envelope glycoproteins– L-HBsAg , and M-HBsAg , and S-HBsAg , was made by inserting the respective ORFs into pcDNA3 ( Invitrogen ) . The sources of the remaining plasmids are as follows: IRF3-GAL4 , IRF7-GAL4 , pFR luciferase reporter , IFNβ-Luc reporter ( Bowie , Trinity College Dublin , Ireland ) , TBK1-Flag , IKKε-Flag ( Dr . Fitzgerald , University of Massachusetts Medical School , Worcester , MA ) , Flag-RIG-I ( Dr . Fujita , Kyoto University ) , Flag-TRIF ( Akira , Osaka University ) , HA-TLR3 ( InvivoGen ) , pISRE-Luc reporter ( Stratagene ) , HA-DDX3 , AS-DDX3 ( K . T . Jeang , N . I . H . ) , and IRF3-GFP ( Garcia-Sastre , Mt . Sinai School of Medicine ) . Western blot analysis was performed as described [42] . For the detection of the Flag tagged Pol protein , mouse anti-FLAG M2 antibody ( Sigma , 1∶5000 ) was used . Anti-IRF antibody ( Invitrogen ) and phospho-specific IRF3 ( Ser396 ) antibody ( Cell Signaling ) were used to detect IRF3 and phosphorylated IRF3 , respectively . Promoter induction and transcription factor activation were measured using HepG2 cells seeded onto 24-well or 6-well plates and transfected after 24 h with expression vectors and luciferase reporter gene . For the IRF3/7 assay , an IRF3/7-GAL4 fusion vector was used in combination with the pFR luciferase reporter , as previously described [45] . Southern blot analysis was performed as previously described [46] . Briefly , the extracted viral DNA was separated by electrophoresis through a 1 . 3% agarose gel in a 0 . 5× Tris-acetate-EDTA buffer and then transferred onto a nylon membrane . The nylon membrane was prehybridized and hybridized with a 32P-labeled full-length HBV DNA probe in a hybridization solution for 16 h at 65°C . Images were obtained using the phosphoimager ( BAS-2500; Fujifilm ) . Immunoprecipitation was performed as described with modifications [47] . Briefly , after transient transfection , the medium was removed and the cells were rinsed twice in cold PBS , incubated for 30 min at 4°C in lysis buffer [50 mM Tris-HCl ( pH 7 . 4 ) , 150 mM NaCl , 1 mM EDTA , 1 mM DTT , 0 . 2 mM PMSF , and 1% NP-40] , and collected by scraping . Cell debris was removed through centrifugation at 10 , 000×g for 10 min at 4°C . Extracts were pre-cleared with protein G–agarose beads for 1 h at 4°C . The primary antibody was added for 1 h at 4°C , and immunoglobulin complexes were collected on protein G–agarose beads for 1 h at 4°C . The beads were washed five times with 1 ml of lysis buffer . Protein complexes were recovered by boiling in Laemmli sample buffer and analyzed by SDS–PAGE . HEK293 cells were grown on 18-mm coverslips in 12-well plates and transfected with 2 . 5 µg of total DNA . At 48 h after transfection , cells were fixed with 3 . 7% formaldehyde . Slides were mounted onto glass slides with ProLong Antifade Kit ( Molecular Probes ) and examined by confocal microscopy ( LSM 510 Meta; Carl Zeiss , Germany ) . HBV ayw subtype: V01460 J02203 DDX3: accession NM_001356 . 3
|
Viral infection is sensed by the host innate immune system , which acts to limit viral infection by inducing antiviral cytokines such as the interferons . To establish infection , many viruses have strategies to evade the innate immunity . For the hepatitis B virus ( HBV ) , which causes chronic infection in the liver , the evasion strategy remains mysterious . An earlier study using the chimpanzee as a model suggested that the host innate immune system failed to detect HBV . As a result , it was dubbed “stealth virus” . In contrast , subsequent studies performed in vitro have suggested that HBV is , in fact , detected by the innate immune system but can effectively counteract this response . Whether HBV is detected by the innate immune system remains controversial; however , it is widely accepted that , regardless of detection , HBV effectively inhibits the host innate immune response early in infection through an unknown mechanism . The data presented here indicate that HBV Pol ( polymerase or reverse transcriptase ) blocks the innate immune response . This unexpected role of HBV Pol may explain why HBV appears to act as a “stealth virus” in the early phase of the infection .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"virology/viral",
"replication",
"and",
"gene",
"regulation",
"virology/immune",
"evasion"
] |
2010
|
Hepatitis B Virus Polymerase Blocks Pattern Recognition Receptor Signaling via Interaction with DDX3: Implications for Immune Evasion
|
In both prokaryotic and eukaryotic cells , gene expression is regulated across the cell cycle to ensure “just-in-time” assembly of select cellular structures and molecular machines . However , present in all time-series gene expression measurements is variability that arises from both systematic error in the cell synchrony process and variance in the timing of cell division at the level of the single cell . Thus , gene or protein expression data collected from a population of synchronized cells is an inaccurate measure of what occurs in the average single-cell across a cell cycle . Here , we present a general computational method to extract “single-cell”-like information from population-level time-series expression data . This method removes the effects of 1 ) variance in growth rate and 2 ) variance in the physiological and developmental state of the cell . Moreover , this method represents an advance in the deconvolution of molecular expression data in its flexibility , minimal assumptions , and the use of a cross-validation analysis to determine the appropriate level of regularization . Applying our deconvolution algorithm to cell cycle gene expression data from the dimorphic bacterium Caulobacter crescentus , we recovered critical features of cell cycle regulation in essential genes , including ctrA and ftsZ , that were obscured in population-based measurements . In doing so , we highlight the problem with using population data alone to decipher cellular regulatory mechanisms and demonstrate how our deconvolution algorithm can be applied to produce a more realistic picture of temporal regulation in a cell .
Recent technological advances have made feasible studies of biological systems at the single-cell level [1]–[4] . However , our current understanding of single-cell biochemistry and physiology has been largely inferred from averaged population measurements that often mask individual cell dynamics and lead to a distorted picture of cell behavior . Such cell population data can be difficult to reconcile with single-cell models , such as those that attempt to describe cell-cycle-dependent gene expression kinetics [5]–[7] . In particular , mathematical models of single cells that rely on population data for constraints on biochemical parameters may arrive at incorrect conclusions . Among the properties hidden by population averaging is cell-to-cell variability , such as that found in gene expression and protein production [8]–[11] . We refer to the natural variation found between cells at the same position in their cell cycles as synchronous variability . A population experiment in which synchronous variability is the only source of variability can at most yield the average of the observable of interest ( e . g . , gene expression levels ) . However , in addition to the inherent synchronous variability , typical time-series experiments on cells contain a significant asynchronous variability: even if cells have been physically or chemically synchronized , individual cells within a synchronized population exist at variable points in their respective cell cycles . As a result , the extraction of ‘true’ temporal data from such populations is difficult , since contributions from cells in different stages of the cell cycle are averaged . From a mathematical perspective , population asynchrony may be modeled as a kernel function that maps the average of an observable in the absence of asynchronous variability to the value measured at the population level . Population asynchrony has been modeled in yeast as both a time-dependent [12] , [13] and time-independent [14] source of variability . With an accurate asynchrony model , extracting the average of an observable becomes an inverse problem for which established regularization methods can be used . These computational methods can effectively remove from population data artifacts that are due solely to asynchrony , or uncover features that are masked by population averaging [13]–[15] . The resulting data is thus better suited for comparison with single-cell models and parameter estimation . Population asynchrony characterization is most easily done with a synchronizable system such as the dimorphic bacterium Caulobacter crescentus . Caulobacter begins its cycle as a motile ‘swarmer’ ( SW ) cell and differentiates to a non-motile ‘stalked’ ( ST ) cell just prior to the initiation of DNA replication . The SW stage is thus analogous to the G1 phase of the eukaryotic cell cycle , and the ST stage is analogous to the S and G2 phases [16] . At the SW-to-ST transition , the flagellum is released and a narrow cylindrical extension of the cell envelope ( the ‘stalk’ ) is grown in its place . A new flagellar assembly is constructed at the pole opposite the stalk as the cell cycle progresses , and on cell division , a new motile , chemotactic SW cell is spawned . The remaining ST cell immediately commences another round of DNA replication and division while the SW cell begins the full cell cycle ( Fig . 1 ) . Centrifugation of a mixed culture of Caulobacter in Ludox or Percoll separates SW cells from all other cell types , so that nearly pure cultures of SW cells can be easily obtained [17] , [18] . However , even a perfectly pure culture of SW cells includes a mixture of new and old SW cells , and variance in the cell cycle times of individual cells within this synchronized population leads to a further increase in the heterogeneity of the population as time-series experiments progress . Additional heterogeneity is introduced following cell division , as each dividing cell results in both a SW and ST cell . Thus , even a perfectly synchronized population develops a significant and time-dependent population asynchrony . We propose a simple model for the time-dependent distribution of Caulobacter cell types in a population during synchronized growth . Our model accurately matches observed distributions of synchronized Caulobacter cells during a time-series experiment , and may be extended to any organism for which the synchrony state can be characterized—particularly those that undergo asymmetric division . We then combine a generalization of deconvolution with our Caulobacter distribution model to extract the “single-cell”-like synchronous average of gene expression profiles from published cell cycle microarray data . The resulting expression profiles more accurately predict the cell-cycle position and size of gene expression peaks , display new features not evident in the original microarray data set , and demonstrate robustness to uncertainty in model parameters . This represents a new advance in the study of cell-cycle dependent gene expression in Caulobacter . The deconvolution method presented herein can be generally applied to characterize time-dependent processes in a variety of biological model systems .
To effectively remove the effects of population asynchrony from measured data , we must first establish a model describing the temporal position of cells within their own cell cycles and how they are distributed in the population . In this section we develop this model in the context of Caulobacter , however , the modeling framework and deconvolution procedure remain generally applicable to other model systems . We refer to the position of a cell within its own cell cycle as the cell's phase , and define it to be a number between zero and one . By our definition represents a new SW cell and is a predivisional cell at the instant before cell division ( Fig . 1 ) . In addition to and , other phases of interest are the phase at which the cell transitions from SW to ST , from ST to early predivisional cell ( EPD ) , and from early predivisional to late predivision cell ( LPD ) . The concept of a cell cycle phase has been used previously , referred to as either the cell division unit or cell cycle unit [19]–[21] . At time t following synchronization , we assume that each cell of a large population of cells is described by three variables: All three of these cell-specific quantities are random variables; and do not change with time , and is time dependent . Therefore , a probability density function ( PDF ) may be written to describe the distribution of these parameters in a population of cells at a given time t ( 1 ) The variables and are assumed to be independent and normally-distributed ( and ) . The Caulobacter cell cycle time coefficient of variation ( COV ) was previously determined to be 0 . 13 [1] , i . e . . We assume that has the same COV and a mean value of , consistent with previous reports [17] , [22] . For notational simplicity , we let and rewrite the Eq . ( 1 ) as , with given as the products of the two independent normal distributions just described . The conditional distribution is based on a phase evolution model that is firmly rooted in experimental observations . We begin by considering a single cell ( indexed k ) described by the variables and . This cell progresses through the phases of its own cell cycle with a ‘velocity’ of as experiment time passes; that is , for . When , and the cell reaches the end of its cycle , two daughter cells emerge at different cell cycle phases: the new SW ( characterized by ) cell begins at and the new ST cell ( now characterized by ) begins at the SW-to-ST transition phase . The new SW-to-ST transition phases and cell cycle times , , are redrawn from their respective distributions . Having constructed a model for the distribution of cell types , we now show how this distribution can be used to map gene expression at the single-cell level to the expression data derived from cellular populations . The signal intensity measured in a typical microarray experiment is proportional to the population-level concentration of the measured species [23] . Thus , for each gene j in an RNA expression assay , the signal intensity at measurement time t is ( 2 ) where is the number of RNA transcripts in the population and is the total cellular volume . For a large number of cells , the total population volume is ( 3 ) where is the volume of a cell with at phase , and is the expectation of a single cell's volume over . Similarly , the total number of RNA transcripts at time t for a given gene j is ( 4 ) where is the synchronous average cycle-dependent expression of gene j , i . e . , the average expression of all cells at the exact same phase . The expression level has units ( # transcripts/volume ) . Note that we may substitute the synchronous average expression function for the true single-cell function in the above equation because the synchronous cell-to-cell variability is independent of ( see supplementary Text S1 for more details ) . It has been previously shown that the Caulobacter division plane is not located at the center of the cell , rather the cell volume is partitioned 40% SW cell to 60% ST cell [24] . We use this fact to construct a simple piecewise linear approximation for the volume of cell k , with parameters , as a function of cell cycle phase ( 5 ) where is the cell volume at just prior to division . We have assumed that the variance of the final cell size distribution is small so that is effectively constant across all cells . Using the above approximations , the total concentration of gene j transcripts at time t ( Eq . ( 2 ) ) can then be written as an integral transform ( 6 ) where is the kernel of the transform , and has the intrepretation of a fractional volume density . That is , represents the fraction of the total population volume at time t that exists in ( a small interval around ) phase . The kernel mapping function depends on , where the volume and probability are known functions . However , the functional form of is complicated by the facts that cells evolve at different rates and that new cells are being generated at different phases . We therefore resort to simulation methods in order to evaluate and . The rule-based Caulobacter cell-type phase evolution model described above enables us to simulate cell populations and growth . An initial population of cells was subjected to simulated growth for a length of time equal to 10 average cell division times . We observe , empirically , that this amount of time is sufficient in order to obtain a steady state population of cells whose phase distribution is independent of the initial seed population . The synchronized population is then drawn from the steady state population by keeping only those cells in the SW state and rejecting all others . The steady state distribution is shown in Fig . 2A , and the distribution of synchronized cells is shown in Fig . 2B . After synchronization , time is declared , and the expression experiment begins . Our results utilized 106 synchronized cells at . Rewriting as ( 7 ) we see that is the product of i ) the probability ( density ) of observing at time t and ii ) the average cell volume at time t conditioned on phase . These two quantities are evaluated through the simulation by allowing the synchronized cells to evolve until a desired time t is reached and the current population of cells , , can be used to evaluate . For a desired , let the denote the indices of the cells with phases approximately equal to ( 8 ) where is a small interval . The marginal probability density is approximated as ( 9 ) with denoting the cardinality of set . The expected volume is similarly calculated as ( 10 ) he integral may be approximated using quadrature methods on a sampled version of or by observing that the integral is the expected volume over all cells at time t , which is calculated by substituting for in the right hand side of Eq . ( 10 ) . Hence , Eq . ( 9 ) and Eq . ( 10 ) , combined with a rule-based model of the evolution of cell types within a population enable us to compute the kernel transformation needed to invert population measurements into single-cell data . The kernel is shown for six different times following synchronization in Fig . 3 . The time evolution of is also shown with 0 . 5 minute resolution in supplementary Video S1 . We observe that the kernel structure is highly time dependent and not well-modeled by any common form . As such , any attempts to reconstruct expression functions by deconvolving with fixed kernels , e . g . a Gaussian kernel , will lead to poor results . With the complete noiseless measurement model given as the integral equation in Eq . ( 6 ) , extracting average single-cell information involves solving the integral equation for given a set of concentration measurements ( the j subscripts on and are dropped for notational clarity ) . Because the number of measurements is finite and small , the inversion process is ill-posed and requires a degree of regularization , i . e . , the introduction of additional information . Since is a physical process , we expect it to be a smooth continuous function and model it as a natural cubic spline . That is , we assume can be well-modeled by a number of piecewise cubic polynomials with boundary constraints ensuring that the entire function is smooth . Cubic splines have been previously used to regularize and simplify ill-posed integral equations [25] , [26] and to represent gene expression profiles [13] . Under the cubic spline model , the expression function may be written ( 11 ) where form a set of basis functions for the natural cubic splines with a particular set of knots . See , e . g . , [27] , [28] , for a discussion of splines and methods of constructing the basis functions . The coefficients determine the particular realization of from within the family of functions spanned by the natural cubic spline basis . We choose a dense sampling of knots uniformly spread over the [0 , 1] domain of . With an matrix , is an representing evaluated at the knot values . In order to estimate the expression function , which is solely specified by in our model , we minimize the following cost criterion ( 12 ) where . The first term is a data fidelity measure that quantifies the closeness of the model-predicted measurements to the actual measurements , weighted by the inverse of the measurement variance of each particular measurement , ( see supplementary Text S1 ) . The second term in Eq . ( 12 ) , a second derivative cost , is a regularization term that penalizes solutions containing rapid fluctuations and is commonly used in regularizing natural smooth systems [28]–[31] . The constant is a smoothness parameter that establishes a tradeoff between data fidelity and smoothness enforced by the second derivative norm . The smoothness parameter is chosen though cross-validation ( described in the next section ) . The cost function is minimized subject to two constraints The final optimization problem is to minimize subject to the two constraints ( 15 ) As illustrated in the supplementary Text S1 , the cost function may be written as a quadratic form . For the results presented in this paper Eq . ( 15 ) was solved using the quadprog function of MATLAB's Optimization Toolbox version 4 . 0 . The sampled estimated expression function is then given as , or the elements of may be used in Eq . ( 11 ) to evaluate for any value of . The solution to the optimization problem ( Eq . ( 15 ) ) depends on the value of the smoothness parameter : small favor data fidelity and are susceptible to overfitting , large may oversmooth the estimated expression function . Cross-validation provides a principled method to select an appropriate value of . The results in this paper utilize leave-one-out cross-validation [28] , [32] as follows . For a fixed value of , the optimization is first performed using all the data except for measurement m ( with value ) . Denote the resulting estimated expression function as . The process is repeated , excluding a different measurement each time . The total cross-validation measure ( 16 ) is then minimized over to obtain , which is then used in Eq . ( 15 ) with all the data in order to obtain the optimal which , in turn , produces the desired expression estimate .
The cell-type distribution model enables us to mathematically determine the probability that a cell taken from a synchronized population is in a given phase . For example , the probability that a single Caulobacter taken from a population minutes following synchronization is in the SW phase is ( 17 ) However , because is difficult to compute directly , we may alternatively calculate various probabilities from the simulation described in the previous section . Our simulated distribution , with cells grouped broadly into the SW , ST , EPD , and LPD types , is shown alongside the experimentally-determined distribution in Fig . 4 . The ST-EPD and EPD-LPD transition phases were fixed at 0 . 69 and 0 . 87 respectively , with the mean cell-cycle time taken to be with COV = 0 . 13 . Experimental data was reproduced from Judd et al . [33] . As can be seen in Fig . 4 , our cell-type distribution model predicted highly similar fractions of SW , ST , EPD , and LPD cells . Experimentally , distinguishing between ST and EPD cells and EPD and LPD cells is difficult as the morphological differences between them are subtle , thus our assignment of those transition phases is somewhat arbitrary . The difference between SW and ST is more easily observed experimentally . Overall , our model predicted a distribution of cells that is , on average , only a few percent different from experimental observation at all time points and for all cell types . There are over 500 cell cycle-regulated genes in the Caulobacter genome [34] . In this paper we apply our deconvolution method to analyze the expressions of a subset of these: genes that are essential for cell viability or proper development and have been included in previous models of the Caulobacter cell cycle control network [6] , [7] , [35]–[37] . Microarray data for 10 cell cycle-regulated genes ( ctrA , dnaA , ccrM , gcrA , cckA , chpT , pleC , divJ , divK , and ftsZ ) was taken from a cell-cycle Affymetrix expression data set published by McGrath et al . [38] . The original microarray measurements , model-predicted measurements , and spline-predicted profiles are shown in Fig . 5 . The regularization parameters used , as determined by cross-validation , are listed in supplementary Table S1 . In general , the deconvolution procedure yielded expression profiles with peaks shifted to later times relative to the population data , and recovered details lost in the population averaging . For example , the deconvolved expression profile for ctrA remains flat until the SW-to-ST transition , and shows an expression ‘shoulder’ before the main peak around the phase of cell compartmentalization ( transition from EPD-LPD ) . The transcription of chpT , pleC , and ftsZ is similarly delayed until the SW-to-ST transition . Both ccrM and divK are highly repressed until just prior to the EPD stage . Many of the genes also show a narrowing of the expression peaks . An extended analysis of these 10 deconvolved gene profiles is left for the Discussion section .
While population-level experimental techniques typically allow for high-throughput and fast data collection , they are unable to capture many of the details present at the level of single cells . This is an unavoidable consequence of population averaging; population-based data are in fact transforms of organism- and condition-specific population asynchrony kernels with single-cell data . Thus , an assumption of equivalence of population and single-cell data is an assumption of a non-physical delta function integral kernel . Recognizing this , cell distribution models have been proposed with the aim of extracting more detailed information from biological time-series data . Perhaps the simplest improvement on the delta function model is a fixed kernel such as a Gaussian . Further improvements have been made by allowing for a Gaussian kernel whose width increases with time ( e . g . , [13] ) . However , a normal distribution of this kind is not sufficient to describe the complex cell-phase distribution of organisms that undergo asymmetric division , and attempts to deconvolve single-cell expression for such organisms will lead to unreliable results . As a result , we have developed an intuitive mathematical model of the cell-type ( or , alternatively , the cell-phase ) distribution of asymmetrically-dividing cells as a function of time following synchronization , using Caulobacter as a specific example . Our model takes into account the initial population asynchrony and , similar to the yeast cell cycle phase probability density model presented in Orlando et al . [12] , captures the phase variability resulting from asymmetric cell division and differences in cell cycle times . An appealing aspect of our model is its simplicity; a knowledge of three easily-measured parameters—namely the mean SW-to-ST transition phase ( or equivalent ) , division time COV , and SW/ST cell total volume fraction ( or equivalent ) —and the initial synchronization state ( i . e . , the cell-type distribution at the outset of an experiment ) are all are that is required to describe the time-dependent cell-type distribution . The aforementioned parameters and initial synchronization state are specific to a given model system and experimental condition . For a synchronized population of Caulobacter under normal growth conditions , we use a mean SW-to-ST transition phase of ∼0 . 25 , division time COV of 0 . 13 , cell volume partitioned 40% SW to 60% ST , and a simulated initial cell cycle phase distribution that accurately models the real synchronization process . But Caulobacter is not the only synchronizable model system to which our cell-type distribution model can be applied . Indeed , synchronizable model systems are found across the tree of life , including E . coli [42] , S . cerevisiae [43] , and mammalian cells [44] . A 1957 review by Campbell describes synchronization methods for 11 microbial species [45] . For the symmetrically dividing E . coli , the equivalent of the SW-to-ST transition phase would be set to zero , and the two daughter cells would ( on average ) have the same volume . In the case of S . cerevisiae , the SW-to-ST transition phase equivalent is equal to the average fraction of the cell cycle that the budded daughter cell remains in the early G1 stage [46] , with the average size of the budded cell being smaller than that of its mother [47] . The division time COVs for a number of commonly studied systems have already been published ( a compilation of these values can be found in [1] ) . Initial cell distributions for many of these organisms have to be determined . We note that we have assumed a perfect Caulobacter synchrony , i . e . , exactly 100% of the cells at the beginning of the experiment are SW cells . In real cell synchrony experiments , SW fractions are close but not necessarily equal to 100% ( see , e . g . , [22] ) . However , minor differences in the purity of a synchronized population are not expected to significantly alter our results . That our cell-phase distribution model is consistent with experimental observations of the time-dependent state of a Caulobacter population ( Fig . 4 ) supports this assumption . Along with characterization of cell distribution , there has been considerable interest in recent years in extracting “single-cell”-like information from population data using deconvolution-type algorithms [13]–[15] , [48] , [49] . Although all algorithms of this kind are somewhat limited in the level of detail they can provide about biological systems—at best , only synchronous average information , and not the full stochastic variability between cells at identical phases , can be determined—they have been highly effective at uncovering features not visible in the population data . The model-based deconvolution method presented here is an extension to these previous methods and a powerful tool for the analysis of biological data , requiring no more information than the parameters described previously , and is applicable to any time-series data set for which the state of the synchrony is known or can be predicted . In particular , our method can be applied to time-series gene expression data to identify additional cell cycle-regulated genes not previously discovered and to complete meta-analyses across multiple platforms ( i . e . competitive hybridization oligo arrays or non-competitive hybridization arrays such as Affymetrix ) . Although the differences in the data obtained from different platforms may require modifications to the kernel function , the method itself is independent of the experimental and biological details; indeed , the method supports arbitrary kernel functions . Even with a detailed and accurate kernel and an accepted deconvolution-type algorithm , the precise shape of a deconvolved function is in general highly sensitive to the value of the regularization parameter ( in this work; see Eq . ( 12 ) ) . To objectively address this problem , we employ a cross-validation routine that provides a sensible and well-established criterion for determining the appropriate amount of regularization . Our use of cross-validation in deconvolution of time-series gene expression data thus represents an improvement over methods that use arbitrary regularization based only on visual inspection of the estimated profiles . By construction , the model-based deconvolution method presented in this paper mitigates the effects of synchronization loss in expression experiments . However , as with all time series experiments , the estimates remain dependent on the sample rate of the data . If the sample rate is insufficiently high to capture salient gene activity , important events in the expression profile may be missed . In principal , lower sampling rates may be accommodated by increasing the number of assumptions made about the expression profile to be estimated . In this paper , smoothness ( Eq . ( 12 ) ) , positivity ( Eq . ( 13 ) ) , and continuity ( Eq . ( 14 ) ) were all used to decrease the effective degrees of freedom and supply a maximal , yet realistic , amount of a priori information . The cubic splines support a broad class of potential expression functions , however more restrictive models could be used to supply stronger assumptions and support lower sampling rates—at the cost of potentially being overly restrictive and not capturing the true gene expression profile . See , e . g . , [50] for further consideration of sample rates in temporal data . The synchronous average expression profiles extracted using our generalized deconvolution algorithm are , with the effects of population asynchrony removed , a much-improved reflection of biological reality . We demonstrated this with Caulobacter , calculating deconvolved expression profiles for 10 genes previously found to be cell cycle-regulated and essential for cell viability or polar cell development ( Fig . 5 ) . As mentioned in Results , the deconvolved expession profiles generally have their peaks shifted to later times relative to the population data . This is to be expected , since even a perfectly-synchronized population at the outset of an experiment contains both young SW cells ( ) and old SW cells ( and all cells in between ) . Many of the genes analyzed here also show a narrowing of their expression peak ( s ) following deconvolution , although this is not universally true . The expression profile of divJ , for example , is shifted to later times but not otherwise fundamentally changed; the peak , located just after the SW-to-ST transition in the deconvolved profile , is as broad as in the population measurement . Thus , expression peak narrowing is not an artifact of the deconvolution method , but rather a property of an individual gene's expression profile . Here we highlight some of our Caulobacter-specific results that also demonstrate the power of combining an organism-specific kernel with a generalized deconvolution routine: ctrA . As the master regulator of the Caulobacter cell cycle [51] , ctrA is arguably the most well-characterized of Caulobacter genes . It has been shown that ctrA expression is controlled by two promoters ( P1 and P2 ) that are differentially-regulated by phosphorylated CtrA ( CtrA∼P ) : the weaker P1 is negatively-controlled by CtrA∼P and the stronger P2 is positively-controlled ( Fig . 7A ) . The P1 promoter is activated in the early ST cell , immediately following replication of the chromosomal ctrA locus . Activation of the weak P1 promoter leads to an increase in the CtrA∼P concentration , which then activates the stronger P2 and represses P1 [52] . The differential regulation can be seen in Fig . 7B , left panel ( data reproduced from Reisenauer and Shapiro [53] ) . Although these details are not visible in the population-level microarray data , they are revealed in the deconvolved expression profile ( Fig . 7B , middle and right panels ) . For example , in the deconvolved profile , ctrA expression remains flat until DNA replication is initiated at the SW-to-ST transition . Perhaps most interestingly , the initial expression ‘shoulder’ is consistent with expression from P1 , and the main peak beginning around the phase of cell compartmentalization ( transition from EPDLPD ) , is consistent with expression from P2 . The shape of the deconvolved ctrA profile is thus validated by our previous knowledge of the mechanism of ctrA regulation . ftsZ . The tubulin homolog FtsZ is essential for bacterial cell division . It has been shown that transcription of ftsZ is repressed in SW cells and activated only when the DNA replication begins [20] . However , this regulation is not clear from the microarray data alone . Specifically , the raw microarray data show no delay in ftsZ transcription from the time the experiment begins ( Fig . 8 , left panel ) . In contrast , the deconvolved expression profile reveals the delay in transcription initiation until the beginning of the ST stage ( Fig . 8 , right panel ) , consistent with our understanding of ftsZ regulation . divK and ccrM . DivK is an essential single-domain response regulator that is transcriptionally-activated by CtrA∼P and plays a role in the cell cycle-regulated proteolysis of CtrA [54] . The essential ccrM DNA methyltransferase gene [55] has an expression profile similar to that of divK . In both cases , deconvolution reveals that expression begins in the EPD cell , and that the change from zero to maximal expression happens over a much shorter time ( i . e . , the response is more switch-like ) than is evident from the population data . cckA . One of the more interesting results is the predicted transcription profile of cckA , which encodes an essential histidine kinase responsible for CtrA phosphorylation [56] . The population-level microarray measurements show a single expression peak approximately half-way through the cell cycle , while the deconvolved profile shows two peaks: one beginning at the SW-to-ST transition and another peaking in the EPD cell . Although this result has not been previously reported , it does suggest the interesting possibility that cckA is under the control of additional and unknown layers of transcriptional regulation during the cell cycle . These deconvolution results appear to be relatively insensitive to changes in model parameters . Of the parameters used in the cell cycle phase distribution model , the mean SW-to-ST transition phase is the one that is known with the least certainty . However , we found that precise knowledge of the mean transition phase under a given condition is not absolutely necessary for extraction of average single-cell data with our deconvolution algorithm . Even a substantial change in the assumed SW-to-ST transition phase had only a small effect on the deconvolved profiles . With respect to the single-cell volume model employed in the deconvolution algorithm , even the extreme and false assumption of fixed cell volume had an insignificant effect on the shape of the deconvolved expression profile . One Caulobacter-specific result that merits further discussion is the SW-to-ST transition phase . Although accepted as around 0 . 25 , or even up to 0 . 33 , for standard growth in a rolling tube or shaken flask [17] , [22] , [39] , [40] , it can change under other conditions . We present data showing that the mean transition phase is reduced to 0 . 15 in a microfluidic environment in which the cells are rapidly growing . We recognize that a possible explanation for this low value may be that the timing of the SW-to-ST transition in our microfluidic growth experiments is skewed by a division control system in which ST cells that have just transitioned from the SW stage divide on a different time scale than ST cells that follow from cell division . However , we are not aware of any data that would suggest that this is the case . Indeed , the morphology of ST cells after the transition from SW cells appears to be the same as the morphology of ST cells after division , and a single mean SW-to-ST transition phase in our model is consistent with experimental observations ( Fig . 4 ) . Furthermore , given that a population of Caulobacter cells starved for carbon or nitrogen tend to arrest during the SW phase [57] , [58] , it is likely that the SW-to-ST transition phase can both increase and decrease , and be well above 0 . 33 under less-favorable environmental conditions . That the timing of this cell cycle ‘checkpoint’ may vary with growth conditions is a fascinating result that deserves more detailed study . To our knowledge , our deconvolution method is the first to specifically deal with the unique analytical challenges posed by dimorphic organisms . Although this method can be applied to any time-series measurement made on a cellular population , we have demonstrated its utility with an analysis of cell-cycle regulated gene expression in Caulobacter . Certainly , directly measuring the concentration of individual transcripts in real time in single cells remains the gold standard in quantifying the gene expression behavior of single cells; the insights provided by such real-time , single-cell studies of mRNA have been profound [59]–[62] . Still , despite recent progress and a number of successes , the real-time measurement of mRNA in single cells remains a challenging problem . Our method allows for the simple analysis of mRNA concentrations measured with common laboratory tools and advances the performance of population-level methods closer to that of single-cell studies . Thus , combining high-throughput experimental expression data with novel computational algorithms can provide new and exciting insights into the function of cellular systems .
|
Time-series analyses of cellular regulatory processes have successfully drawn attention to the importance of temporal regulation in biological systems . A number of model systems can be synchronized such that data collected on cell populations better reflect the dynamic properties of the individual cell . However , experimental synchronization is never perfect , and the degree of synchrony that does exist at the outset of an experiment is quickly lost over time as cells grow at different rates and enter different developmental or physiological states on cell division . Thus , data collected from a population of synchronized cells can lead to incorrect models of temporal regulation . Here we demonstrate that the problem of relating population data to the individual cell can be resolved with a computational method that effectively removes the effects of both imperfect synchrony and time-dependent loss of synchrony . Application of this deconvolution algorithm to a cell cycle time-series data set from the model bacterium Caulobacter crescentus uncovers critical temporal details in the expression of essential genes that are not evident in the raw population-based data . The deconvolution routine presented here is a robust and general tool for extracting biochemical parameters of the average single cell from population time-series data .
|
[
"Abstract",
"Introduction",
"Model",
"Results",
"Discussion"
] |
[
"mathematics",
"computational",
"biology/systems",
"biology",
"cell",
"biology/gene",
"expression"
] |
2009
|
Model-Based Deconvolution of Cell Cycle Time-Series Data Reveals Gene Expression Details at High Resolution
|
The way in which single neurons transform input into output spike trains has fundamental consequences for network coding . Theories and modeling studies based on standard Integrate-and-Fire models implicitly assume that , in response to increasingly strong inputs , neurons modify their coding strategy by progressively reducing their selective sensitivity to rapid input fluctuations . Combining mathematical modeling with in vitro experiments , we demonstrate that , in L5 pyramidal neurons , the firing threshold dynamics adaptively adjust the effective timescale of somatic integration in order to preserve sensitivity to rapid signals over a broad range of input statistics . For that , a new Generalized Integrate-and-Fire model featuring nonlinear firing threshold dynamics and conductance-based adaptation is introduced that outperforms state-of-the-art neuron models in predicting the spiking activity of neurons responding to a variety of in vivo-like fluctuating currents . Our model allows for efficient parameter extraction and can be analytically mapped to a Generalized Linear Model in which both the input filter—describing somatic integration—and the spike-history filter—accounting for spike-frequency adaptation—dynamically adapt to the input statistics , as experimentally observed . Overall , our results provide new insights on the computational role of different biophysical processes known to underlie adaptive coding in single neurons and support previous theoretical findings indicating that the nonlinear dynamics of the firing threshold due to Na+-channel inactivation regulate the sensitivity to rapid input fluctuations .
How do pyramidal neurons transform their input into output spike trains ? The answer to this question is of fundamental importance because it determines potential coding strategies of the brain [1–5] . Theoretical studies of the Integrate-and-Fire model responding to in vivo-like fluctuating currents concluded that , depending on the average strength μI ( DC component ) of the input current , single neurons can operate in two different regimes known as the fluctuation driven regime and the mean driven regime [6] . While in the fluctuation driven regime output spikes are exclusively evoked by transient excursions of the membrane potential ( such as those caused by a volley of synchronous inputs ) , a neuron operating in the mean driven regime is continuously active and its output firing rate encodes the average intensity of the input current . This view has recently been challenged by in vitro recordings demonstrating that the output firing rate of pyramidal neurons from rat prefrontal cortex ( PFC ) [7] , somatosensory cortex ( SSC ) [8] and hippocampus [9] always increases with the amplitude σI of rapid input fluctuations , independent of the DC component μI . These results indicate that pyramidal neurons are not static entities , but adapt their intrinsic dynamics in order to maintain selective sensitivity to rapid input fluctuations over a broad range of input statistics . Experimental evidence supporting the view that the input-output transformation performed by single neurons generally depends on the input statistics has also been provided by in vitro measurements of the frequency-response properties of different neuronal types [10–13] . Because of this complex adaptive behavior , it remains a major challenge to design a spiking neuron model that is at the same time simple enough to be understood from a computational perspective and flexible enough to predict spikes over an extended range of input statistics [14] . Enhanced sensitivity to rapid input fluctuations has been initially linked to slow adaptation mechanisms [7 , 8 , 15] and has been qualitatively reproduced in silico using Hodgkin-Huxley models featuring either decreased Na+-conductance , increased K+-conductance , increased leak conductance , slow Na+–channel inactivation or low-threshold K+-channels [7 , 15–17] . Consistent with the fact that different biophysical mechanisms can lead to enhanced sensitivity to rapid input fluctuations , theoretical studies based on the Morris-Lecar model related this phenomenon to fundamental aspects of the spike initiation dynamics [5 , 18 , 19] . While these studies indicate that enhanced sensitivity to rapid input fluctuations is essentially mediated by subthreshold-activating currents [19] , recent theoretical studies demonstrate that responsiveness to rapid signals can also be enhanced by a nonlinear dynamics of the firing threshold due to fast Na+–channel inactivation [20 , 21] . In the present study , we recorded the in vitro response of layer 5 pyramidal ( L5 Pyr ) neurons to in vivo-like fluctuating currents of different offsets μI and standard deviations σI . In agreement with previous results from distinct neuronal cell types , we found that: i ) the average rate response remained sensitive to rapid input fluctuations over a broad range of offsets μI; ii ) the effective timescale of somatic integration was progressively reduced with increasing μI; iii ) the membrane potential at which spikes originated was correlated positively with μI and negatively with σI . To explain these seemingly different phenomena within a single mathematical framework , we introduced a new spiking model—called the inactivating Generalized Integrate-and-Fire ( iGIF ) model—in which the firing threshold is nonlinearly coupled to the subthreshold membrane potential and depends linearly on the spike history . Despite its relative complexity , the iGIF model remains amenable to analytical treatment and , while being linked to known biophysical processes , its parameters can be efficiently extracted from intracellular recordings using a new likelihood-based method . The iGIF model is first shown to capture enhanced sensitivity to rapid input fluctuations , account for firing threshold variability and predict the spiking activity of L5 Pyr neurons over an extended range of input statistics . To study the computational role of the firing threshold dynamics , the iGIF model is then analytically mapped to a Generalized Linear Model ( GLM ) [22 , 23] in which both the input filter and the spike-history filter dynamically adapt to the input statistics . In agreement with the theoretical predictions of Platkiewicz and Brette [21] and recent experimental findings from the barn owl cochlear nucleus [24 , 25] , our experimental and theoretical results demonstrate that the effective timescale over which pyramidal neurons integrate their inputs is not entirely controlled by the membrane timescale , but adapts to the input statistics as a result of the nonlinear dynamics of the firing threshold . This adaptive behavior promotes detection of rapid signals over an extended range of input statistics , thus explaining enhanced sensitivity to input fluctuations in L5 Pyr neurons .
To study single-neuron computation over a broad range of input statistics , we intracellularly recorded the response of cortical neurons evoked in vitro by a set of 5-second currents generated by independently varying the mean μI and the standard deviation σI of the fluctuations ( Fig 1A–1C ) . In vivo-like fluctuating currents were generated with a filtered Gaussian process and injected at the soma of L5 Pyr neurons of mouse SSC ( see Materials and Methods ) . Mimicking the activity levels observed in awake mice [31] , neurons responded by emitting action potentials at rates between 0 and 20 Hz . In agreement with previous results from rat PFC [7 , 32] , SSC [27] and hippocampus [9] , augmenting the magnitude of input fluctuations significantly increased the output firing rate over the entire range of depolarizing offsets that were tested ( Fig 1D and 1E ) . This result is at odds with theories based on standard Integrate-and-Fire models , stating that rapid input fluctuations affect the average firing rate only in the fluctuation driven regime , where the mean input by itself is not sufficient to evoke action potentials [27 , 33] . To characterize the mechanisms underlying enhanced sensitivity to input fluctuations , we fitted several Generalized Linear Models ( GLM , see [22 , 23] ) to datasets of different μI ( see Materials and Methods ) . In our GLM , spikes are generated stochastically with a firing intensity that depends on the input current as well as on previous action potentials ( Fig 2A ) . Briefly , the input current is first passed through a linear filter κGLM ( t ) and then transformed into a firing intensity by an exponential nonlinearity . Each time an action potential is fired , an adaptation process hGLM ( t ) is triggered . In contrast to a Linear-Nonlinear-Poisson ( LNP ) model ( which can be considered as a GLM without spike-history filter hGLM ( t ) ) , a GLM with spike-history filter is equivalent to a Generalized Integrate-and-Fire model [6 , 34] . When LNP models are used to relate an external input ( e . g . , a visual input ) to the spiking activity of a neuron recorded in vivo , the input filter—generally measured by spike-triggered averaging ( STA ) —is interpreted as a receptive field [35] . When a GLM with spike-history filter is used instead [23] ( i . e . , when non-Poissonian firing statistics are accounted for ) , then the input filter κGLM ( t ) generally differs from the STA , but provides a better estimator of the neuron’s receptive field [35 , 36] . Here , a GLM with spike-history filter is used to model the spiking activity of a neuron responding in vitro to somatic current injections [34] . In this case , κGLM ( t ) describes the temporal window over which the neuron effectively integrates its input to generate action potentials ( cf . , Eq 10 ) . Since the GLM is equivalent to a Generalized Integrate-and-Fire model [6 , 34] , the input filter κGLM ( t ) is commonly interpreted as the membrane filter ( that is , the filter linking the input current to the subthreshold membrane potential ) . However , since GLM parameter extraction entirely relies on spiking data and does not exploit the information contained in the subthreshold membrane potential fluctuations , κGLM ( t ) can in principle also capture other mechanisms that affect the spiking probability without altering the membrane potential [34] . In the following we refer to κGLM ( t ) as effective integration filter and use its timescale τGLM to quantify the size of the temporal window over which the input is effectively integrated to generate spikes . While the input filters of LNP models are notoriously sensitive to input statistics ( see , e . g . , [36–39] ) , the filters of a GLM are generally thought to be more stable across a broad variety of inputs . However , by comparing the GLM filters extracted under different stimulus conditions , we found that both κGLM ( t ) and hGLM ( t ) changed with μI ( Fig 2B and 2C ) . In particular , increasing μI resulted in shorter integration filters ( Fig 2B ) . We quantified this result by fitting κGLM ( t ) with a single-exponential function and found that the effective timescale of integration τGLM was drastically reduced from 27 . 1 , s . d . 3 . 6 ms to 4 . 9 , s . d . 1 . 0 ms ( Fig 2D ) . These results are in line with recent experimental findings from rat motorneurons [39] and suggest that L5 Pyr neurons enhance their sensitivity to rapid input fluctuations by shortening their effective timescale of integration [21] . One of the central aims of the present study is to understand why the GLM filters are input-dependent , and to relate this complex form of adaptation to the phenomenon of enhanced sensitivity to rapid input fluctuations . Neurons can shorten their timescale of integration by increasing their total membrane conductance [40 , 41] . To check whether the observed decrease in τGLM was due to conductance changes , we measured the effective membrane timescale τ m eff as a function of μI by fitting an Adaptive Leaky Integrate-and-Fire model to the subthreshold membrane potential fluctuations evoked by input currents with different depolarizing offsets ( see Materials and Methods ) . Importantly , while the effective timescale of integration τGLM quantifies the size of the optimal filter linking the input current to output spikes ( cf , Eq 10 ) , the effective membrane timescale τ m eff quantifies the size of the optimal filter linking the input current to the membrane potential fluctuations . We found that changes in the effective membrane timescale only accounted for part of the reduction observed in τGLM ( Fig 2D ) . We therefore have to search for an additional mechanism of timescale reduction that does not affect the subthreshold membrane potential dynamics . Previous studies indicate that enhanced sensitivity to rapid input fluctuations are mediated by a mechanism that reduces the number of Na+-channels available for spike initiation [7 , 8 , 15 , 17] . In particular , a theoretical study of Platkiewicz and Brette [21] recently showed that fast Na+-channel inactivation could enhance sensitivity to rapid input fluctuations by making the firing threshold nonlinearly dependent on the subthreshold membrane potential . A nonlinear coupling between membrane potential and firing threshold could additionally explain the experimental discrepancy between τ m eff and τGLM [21] . Using standard methods , we extracted from intracellular recordings the voltage at which individual action potentials were initiated ( see Materials and Methods ) . Consistent with earlier results [41–44] , the voltage threshold for spike initiation always increased with the mean input μI ( Fig 3A , 3E and 3F ) and was reduced in the presence of input fluctuations σI ( Fig 3A , 3B , 3G and 3H ) . By further analyzing our raw data , we found that at a given firing rate , the average subthreshold membrane potential always decreased with increasing levels of input fluctuations ( Fig 3C ) . Consistent with the hypothesis that the firing threshold reduction observed under large input fluctuations was mediated by a firing threshold dependence on the subthreshold membrane voltage [21 , 43] , we found that , on average , these two quantities were nonlinearly related ( Fig 3D ) . Overall , the results reported in Fig 3 confirm previous experimental findings and are in line with the theoretical predictions of Platkiewicz and Brette [21] . In the next section , a new Generalized Integrate-and-Fire model is introduced that will allow us to provide direct evidence for a nonlinear coupling between membrane potential and firing threshold . By analyzing the model , we will then explain: i ) why the voltage threshold for spike initiation depends on the input statistics ( Fig 3 ) ; ii ) why the effective timescale of somatic integration is shorter than the membrane timescale and adapts to the input statistics ( Fig 2 ) ; iii ) how L5 Pyr neurons maintain sensitivity to rapid input fluctuations over a broad range of input statistics ( Fig 1 ) . To explain within a single framework the experimental findings reported in Figs 1–3 , we fitted a new spiking model to intracellular recordings . The model was obtained by extending our previous Generalized Integrate-and-Fire model ( GIF; [44 , 45] ) with a nonlinear coupling between firing threshold and membrane potential , which possibly arises from fast Na+-channel inactivation [21] . We refer to this model as iGIF , where i stands for inactivating ( Fig 4A , see Materials and Methods ) . In the iGIF model , spikes are generated stochastically according to a firing intensity which exponentially depends on the instantaneous difference between the membrane potential V and firing threshold VT [29 , 46]: λ = λ 0 exp V - V T Δ V , ( 1 ) where λ0 = 10 kHz is a constant and the parameter ΔV regulates the level of stochasticity . When ΔV → 0 , the model becomes deterministic and spikes are emitted reliably whenever the voltage reaches the firing threshold . When ΔV > 0 , the membrane potential can cross the firing threshold without emitting a spike and spikes can be emitted even when the membrane potential is below threshold . Consequently , when ΔV is large , the iGIF model features strong trial-to-trial variability and generates action potentials with poor temporal precision . The dynamics of the membrane potential are deterministic and modeled as a leaky integrator augmented with an adaptation current IA: C V ˙ = - g L ( V - E L ) + I - I A , ( 2 ) where I denotes the input current , C , gL and EL define the passive properties of the membrane and the dynamics of the adaptation current IA are described by the following conductance-based model [47]: I A ( t ) = ∑ t ^ j < t η ( t - t ^ j ) · ( V - E R ) , ( 3 ) where ER is a reversal potential and η ( t - t ^ j ) describes the time course of the conductance change triggered by the emission of an action potential at time t ^ j . The dynamics of the firing threshold VT can be derived mathematically from an Hodgkin-Huxley model featuring fast and slow Na+-channel inactivation ( see refs . [21] and Materials and Methods ) and writes: V T ( t ) = θ ( t ) + ∑ t ^ j < t γ ( t - t ^ j ) , ( 4 ) where γ ( t ) describes threshold changes induced by a previous action potential and θ ( t ) implements a nonlinear coupling between VT and V , which is governed by the differential equation [21]: τ θ θ ˙ ( t ) = - θ ( t ) + V T * + θ ∞ ( V ) , ( 5 ) with V T * being the threshold baseline . Depending on the functional shape of the steady-state function θ∞ ( V ) , Eq 5 can make the firing threshold sensitive to the depolarization rate of the membrane potential [21 , 48] . Indeed , if θ∞ ( V ) is an increasing function of V , membrane depolarizations occurring on a slower rate than the characteristic timescale τθ will raise the firing threshold . Consequently , compared to fast inputs , slow currents will evoke action potentials that are initiated at larger voltages [21 , 48] . The threshold mechanism described in Eq 5 can thus shorten the effective timescale over which the input current is integrated to generate spikes , without affecting the subthreshold voltage dynamics [21 , 24] . Importantly , the functional shape of η ( t ) , γ ( t ) and θ∞ ( V ) , along with all the other iGIF model parameters , are extracted from intracellular recordings using a two-step fitting procedure ( see Materials and Methods; a Python implementation is freely available at https://github . com/pozzorin/GIFFittingToolbox ) . In the first step , the parameters governing the subthreshold dynamics of the membrane potential—including the functional shape of the spike-triggered conductance η ( t ) —are extracted by minimizing the sum of squared errors on the subthreshold voltage derivative . In the second step , the parameters governing the dynamics of the firing threshold—including the functional shape of the voltage-coupling θ∞ ( V ) and the spike-history filter γ ( t ) —are obtained with a maximum likelihood approach similar to the one used to fit GLMs to spiking data [22 , 23] . In what follows , we refer to the iGIF model estimated using this method as iGIF-free , where free indicates that the functional shape of the above-mentioned functions is not defined a priori , but is extracted from data . We found that the passive properties of the membrane were characterized by a timescale τm = 35 . 3 , s . d . 8 . 6 ms ( Fig 4B ) . Spike-triggered conductance changes were always positive ( Fig 4C ) and associated with a low reversal potential ER = -57 . 0 , s . d . 3 . 9 mV . When displayed on log-log scales , the decay of the spike-triggered threshold movement γ ( t ) was approximatively linear over several orders of magnitude ( Fig 4D ) . This result is in agreement with a previous finding that , in Pyr neurons , spike-frequency adaptation does not have a preferred timescale , but is characterized by a power-law decay [45 , 49] . By extracting the shape of θ∞ ( V ) directly from intracellular recordings , we found that firing threshold and subthreshold membrane potential were indeed nonlinearly coupled ( Fig 4E , black ) . Moreover , the nonparametric estimate of θ∞ ( V ) was in striking agreement with the smooth-linear rectifier predicted by a systematic reduction of the Hodgking-Huxley model [21] . This result indicates the threshold-voltage coupling arises from fast Na+-channels inactivation . Since the value of the coupling timescale τθ = 8 . 6 , s . d . 3 . 0 ms ( Fig 4F ) was also consistent with previous measurements of fast Na+-channel inactivation ( see , e . g . , ref . [50] ) , we used the intracellular recordings to fit a simpler iGIF model , referred to as iGIF-Na , in which θ∞ ( V ) was assumed a priori to be the smooth linear rectifier function θ ∞ Na ( V ) predicted in ref . [21]: θ ∞ Na ( V ) = k a log 1 + exp V - V i k i , ( 6 ) where Vi is the half-inactivation voltage of Na+-channels , and where the asymptotic slope of θ ∞ Na ( V ) is determined by the ratio between the activation slope ka and the inactivation slope ki of Na+-channels ( see Materials and Methods ) . Note that , in the iGIF-Na model , the firing threshold is effectively coupled to the membrane potential only when the subthreshold voltage V is close to or larger than Vi . Moreover , an asymptotic slope ka/ki equal or close to 1 implies that , at large voltages V ≫ Vi , the spiking probability becomes independent of the average value of the membrane potential and is only affected by voltage fluctuations which are fast compared to the characteristic timescale τθ [21] . Both the spike-triggered movement of the firing threshold γ ( t ) ( Fig 4D , red ) and the nonlinear coupling θ ∞ Na ( V ) ( Fig 4E , red ) extracted by fitting the iGIF-Na model to data confirmed the results obtained with the nonparametric ( i . e . , free ) method ( Fig 4D and 4E , black ) . In agreement with the fact that Na+-channels expressed in central neurons have similar activation and inactivation slopes ( i . e . , ka ≈ ki ) [21] , the asymptotic slope of θ ∞ Na ( V ) was very close to 1 ( Fig 4G ) . Again in line with previous biophysical measurements [21] , the half-inactivation voltage Vi obtained by taking into account the fact that recordings were not corrected for the liquid junction potential of 14 . 5 mV ( see Materials and Methods ) were comprised between -60 and -55 mV ( Fig 4G ) . Despite the smaller number of parameters compared to the free search , the log-likelihood of the iGIF-Na model was not significantly different from that of the iGIF-free model ( Fig 4E , inset ) . This result provides additional evidence for the hypothesis that the biophysical mechanism underlying the nonlinear coupling between firing threshold and membrane potential is fast Na+-channel inactivation . In the followings , we just work with the iGIF-Na model , which , for simplicity , will be referred to as iGIF model . To verify whether the iGIF model was able to capture enhanced sensitivity to rapid input fluctuations , we repeated our experimental paradigm in silico by testing the iGIF model with a set of 5-second currents generated by systematically varying the input parameters μI and σI ( Fig 5A–5C ) . We compared the average firing rate response of the model against experimental data and found that , despite its relative simplicity , the iGIF model captured the behavior of Pyr neurons over a broad range of input statistics ( Fig 5A ) . In particular , the iGIF model exhibited enhanced sensitivity to input fluctuations throughout the entire set of depolarizing currents μI that were tested and reproduced the average firing rate response with an accuracy of ϵrate = 1 . 0 , s . d . 0 . 2 Hz ( defined as the root-mean-square error between model and data ) . Notably , the iGIF model also captured the complex dependence of the firing threshold on input statistics . In particular , the voltage threshold at which spikes were initiated was positively correlated with μI ( Fig 5B ) and negatively correlated with σI ( Fig 5B and 5C ) . To appreciate the importance of modeling the nonlinear coupling between firing threshold and membrane potential , we also fitted our previous GIF model [44 , 45] to the same experimental data . The GIF model differs from the iGIF model simply because its firing threshold dynamics only depend on the spike-history and not on the membrane potential ( see Materials and Methods ) . As expected , the GIF model could not capture the firing rate dependence on σI , was less accurate in reproducing the firing rates observed at steady-state ( ϵrate = 1 . 7 , s . d . 0 . 3 Hz; see Fig 5D and 5F ) and was unable to explain the firing threshold dependence on the input statistics ( Fig 5E and 5F ) . Finally , the results obtained by computing the cross-validated log-likelihood ( see Materials and Methods ) of the iGIF and the GIF model as a function of the input strength μI confirmed that a spiking model in which the firing threshold dynamics simply depend on previous action potentials is not sufficient to capture the spiking activity of Pyr neurons over a wide range of input statistics ( Fig 5G and 5H ) . Beyond the firing rates , the iGIF model also reproduces the fine temporal structure of the spiking response ( Fig 6 ) , one of the aims of single-neuron modeling [51] . To take into account the fact that single neurons are stochastic [52] and to avoid problems related to overfitting , we assessed spike-timing prediction on a new experimental dataset ( test dataset ) . This dataset was collected by performing nine repetitive injections of a new fluctuating current Itest ( t ) ( frozen-noise ) that was not used for parameter extraction . In order to test the model’s ability of predicting spikes evoked by different levels of input fluctuations , the standard deviation of the current Itest ( t ) was modulated by a slow sinusoidal function ( Fig 6A , see Materials and Methods ) . On average , the iGIF model with parameters extracted from the dataset used to compute the f−μI curves ( f-I dataset , see Fig 1 ) was able to predict 75 . 1 , s . d . 3 . 2% of the spikes with a precision of ±4 ms ( Fig 6B and 6E ) . The iGIF model performed well also in predicting the slow fluctuations of the firing rate induced by the sinusoidal input modulation ( Fig 6C and 6F ) as well as the rapid dynamics of the subthreshold membrane potential ( Fig 6D ) . As expected , the performance achieved by the GIF model were significantly lower with 48 . 8 , s . d . 9 . 2% of predicted spikes ( Fig 6B–6F ) . In previous studies [44 , 45] , we found that the GIF model was able to predict around 80% of the spikes observed in Pyr neurons responding to quasi-stationary inputs . At first glance , the low performance achieved here might therefore seem surprising . This result can however be understood by comparing the degree of stochasticity of the GIF model and the iGIF model ( Fig 6G ) . In both models , the parameter ΔV regulates the level of stochasticity of the spiking process ( see Eq 1 ) . In the ideal case of a perfect model , ΔV is optimally tuned to capture trial-to-trial variability . In reality , a lack of flexibility in the model can bias the estimation of ΔV towards large values . The reason for this is that , in an oversimplified model , signals mediated by those single-neuron features that the model cannot describe are interpreted as noise [53] . While the level of stochasticity observed in the iGIF model was weak ( ΔV = 0 . 59 mV , s . d . 0 . 13 mV ) , the values obtained by fitting the GIF model to the f-I dataset were always very high ( ΔV = 2 . 74 mV , s . d . 0 . 54 mV ) , indicating that the GIF model is not sufficiently flexible to capture neuronal activity over a broad range of input statistics . Consequently , the GIF model emitted spikes with low temporal precision and achieved a poor performance in predicting individual spikes . To make sure that the success of the iGIF model does not simply result from the aberrant level of stochasticity in the GIF model , we reassessed spike-timing prediction in both models by extracting parameters from a third dataset ( training dataset , see Materials and Methods ) obtained by injecting a 120-s current having the same statistics as the test dataset ( Fig 6E–6G , open bars ) . Indicating that the GIF model is sufficiently flexible to describe the neuronal activity evoked by simpler stimuli ( i . e . , by stimuli that are similar to those used in our previous studies [44 , 45] ) , its level of stochasticity dramatically decreased ( Fig 6G ) , leading to a spike prediction reliability of M d * = 76 . 2 % , s . d . 5 . 1% . Nevertheless , the iGIF model with parameters extracted from the training dataset still outperformed the GIF model by predicting 82 . 8% of spikes ( M d * = 82 . 8 % , s . d . 2 . 0%; Fig 6E ) . Overall , these results demonstrate that the iGIF model is an excellent spiking neuron model capable of predicting individual spikes with millisecond precision and capturing the activity of Pyr neurons over a wide range of input statistics . But one central question remains unsolved: How do Pyr neurons adapt their effective timescale of integration to maintain sensitivity to rapid input fluctuations , regardless of μI ? To answer this question , we analyzed the iGIF model response to three fluctuating currents with different offsets μI and a fixed standard deviation σI ( Fig 7 ) . In response to a weak input μI = 90 pA , the firing rate f ≈ 2 Hz was low and threshold movements induced by different action potentials did not accumulate significantly ( Fig 7A , top ) . Given the modest contribution of spike-dependent threshold adaptation , spikes were initiated with relatively low firing thresholds and subthreshold membrane potential fluctuations were mostly confined to low voltages V < Vi , where the coupling between firing threshold and subthreshold membrane potential is not active ( Fig 7A , bottom ) . However , even at low firing rates , the threshold-voltage coupling was recruited during large positive fluctuations of the membrane potential ( Fig 7A , top red ) . Increasing the input strength to μI = 230 pA resulted in an output firing rate of f ≈ 10 Hz ( Fig 7B , top ) . In this regime , spikes were initiated at larger thresholds , subthreshold membrane fluctuations occurred at voltages V ≈ Vi ( Fig 7B , bottom ) and the nonlinear coupling between firing threshold and membrane potential was constantly active . Further increasing the mean input to μI = 450 pA made the iGIF model fire at f ≈ 18 Hz ( Fig 7C ) . In this regime , threshold movements triggered by different spikes accumulated significantly making it possible for the subthreshold membrane potential to reach more depolarized voltages V > Vi , where the threshold coupling reaches its maximal strength ( Fig 7C , bottom ) . Overall , the results reported in Fig 7 provide evidence for the existence of a non-trivial interplay between spike-dependent and voltage-dependent threshold movements . In particular we found that , at large firing rates , the increased contribution of spike-dependent adaptation allowed the subthreshold membrane potential to attain voltages at which the strength of the nonlinear threshold coupling is maximal ( compare voltage distributions in Fig 7A–7C , bottom ) . Thus , increasing μI progressively strengthened the threshold sensitivity to subthreshold voltage fluctuations ( compare red traces in Fig 7A–7C , top ) . As a result of the threshold dynamics , the membrane potential distribution always peaked below the average firing threshold ( Fig 7A–7C , bottom ) , a characteristic signature of the subthreshold regime in which neurons are sensitive to rapid input fluctuations [54] . To study the functional implications of the progressive activation of the coupling between firing threshold and membrane potential , we systematically reduced the iGIF model to a GLM ( see Fig 2A and Materials and Methods ) . More precisely , we analytically computed the GLM filters κ ^ GLM ( t ) and h ^ GLM ( t ) that best approximate the iGIF model response to a set of stationary currents with different average intensities μI ( see Materials and Methods ) . First , because of the spike-triggered conductance , the effective membrane filter κ m eff ( t ) ( cf , Eqs 28 and 29 ) becomes shorter with increasing μI ( Fig 7D–7F black; see Materials and Methods ) . Second , because of the threshold-voltage coupling , the shape of the effective integration filter κ ^ GLM ( t ) is affected by the threshold dynamics according to the following equation: κ ^ GLM ( t ) = κ m eff ( t ) - G ¯ θ · ∫ 0 ∞ 1 τ θ exp - s τ θ · κ m eff ( t - s ) d s , ( 7 ) where G ¯ θ is the average activation level of the threshold-voltage coupling Gθ ( V ) computed with respect to the voltage distribution ( see Fig 7A–7C , bottom and Materials and Methods ) . In response to a low input μI , the strength of the threshold coupling was weak on average and somatic integration was mostly controlled by the effective membrane filter ( Fig 7D ) . Increasing μI progressively shifted the membrane potential distribution towards voltages where the threshold coupling becomes more and more important ( see Fig 7B and 7C ) . Consequently , the effective integration filter κ ^ GLM ( t ) became shorter than κ m eff ( t ) ( Fig 7E and 7F red ) . These theoretical results , whose accuracy was confirmed by extracting the effective integration filter κGLM ( t ) directly from the spiking activity generated by the iGIF model ( Fig 7D–7F gray ) , indicate that both the spike-triggered conductance and the firing threshold dynamics actively control the timescale of somatic integration . Importantly , since only the first mechanism affects the membrane voltage , the impact of the threshold dynamics on somatic integration cannot be seen in the subthreshold dynamics of the membrane fluctuations . Using the iGIF model parameters extracted from experimental data , we applied our theoretical results and systematically computed the effective membrane filter κ m eff ( t ) ( Fig 8A ) and the effective integration filter κ ^ GLM ( t ) ( Fig 8B ) for different input strengths μI . Notably , the timescales of the theoretical filters κ m eff ( t ) and κ ^ GLM ( t ) explained the experimental discrepancy between the effective membrane timescale τ m eff and the effective timescale of somatic integration τGLM ( Fig 8C ) . Finally , an interaction between the threshold-coupling and the adaptation current IA captured by the iGIF model explained why the GLM spike-history filter hGLM ( t ) was shortened at increasing input strengths μI ( Fig 8D; see Materials and Methods ) . Overall , these results indicate that the iGIF model can be interpreted as an enhanced GLM in which both the input filter κGLM ( t ) and spike-history filter hGLM ( t ) adapt to the input statistics . In order to study the temporal dynamics of single neuron adaptation , we finally performed a switching experiment in which the iGIF model was stimulated with a fluctuating current , whose mean μI periodically switched between a low and a high value , with cycle period Tcycle = 10 s ( Fig 8E ) . In response to a sudden increase in μI , the output firing rate transiently increased and then decayed over multiple timescales , confirming that in the iGIF model the combined action of the spike-triggered conductance and the spike-triggered movement of the firing threshold mediates spike-frequency adaptation . Similarly , in response to a sudden decrease of μI , the output firing rate initially dropped and then partially recovered . By computing κ ^ GLM ( t ) at different moments in time relative to the cycle , we found that , in contrast to spike-frequency adaptation , adaptive changes in the effective timescale of somatic integration τ ^ GLM were almost instantaneous ( Fig 8E , red and 8F ) . The reason for this is that the effects induced by the threshold coupling depends on the momentary voltage , rather than on the mean firing rate . The results presented in Fig 8E and 8F are reminiscent of the adaptive behavior previously observed in retinal ganglion cells [55] and in motion sensitive neurons [56] responding in vivo to external stimuli with varying statistics and show that an intrinsic nonlinearity , such as the one resulting from fast Na+-channel inactivation , can mediate a rapid and seemingly complex form of adaptation [16 , 57] . Overall , our results indicate that L5 Pyr neurons respond to a sudden change in the input statistics by adapting both their output firing rate and the temporal window over which the input current is effectively integrated . The high speed at which the timescale of somatic integration adapts indicates that , regardless of the input statistics , L5 Pyr neurons respond preferentially to rapid input fluctuations resulting , e . g . , from coincident spike arrival .
The membrane potential at which spikes are initiated is highly variable both in vitro and in vivo [24 , 58 , 59] . Many studies have demonstrated that the voltage threshold for spike initiation correlates not only with the average value of the membrane potential [43 , 59] , but also with the duration of previous interspike intervals [29 , 41 , 42 , 44 , 60] and with the speed at which the firing threshold is approached [42 , 58 , 59 , 61] . When single neurons are stimulated with current ramps of different slopes , rapid rates of depolarization are often associated with lower thresholds [48 , 62] . While in rat pyramidal neurons this phenomenon has been linked to the activation of Kv1 channels [48] , those are apparently not expressed in L5 Pyr neurons of the mouse SSC [63 , 64] . As indicated by theoretical studies based on the standard Hodgkin-Huxley model [20 , 21] and by in vivo recordings from barn owl auditory neurons [24] , spike-threshold sensitivity to the membrane depolarization rate can alternatively be mediated by a nonlinear coupling between firing threshold and membrane potential due to fast Na+-channel inactivation . Providing indirect evidence for this hypothesis , we found that the somatic voltage at which action potentials originated was highly variable , depended nonlinearly on the average membrane potential , was positively correlated with the DC component of the input current μI and decreased with increasing input fluctuations σI ( Fig 3 ) . Dual patch-clamp recordings have shown that the membrane potential recorded at the soma does not necessarily reflect the membrane potential at the axon initial segments ( AIS ) , where spikes are initiated [65 , 66] . These results questioned whether somatic threshold variability , and more generally the somatic spike shape , reflects real integrative properties of the neuron or are just an epiphenomenon of spike back-propagation from the AIS [66 , 67] , but see [21 , 68–70] . To investigate the origin of spike-threshold variability , we fitted our intracellular recordings with a new spiking neuron model , in which the firing threshold is dynamically coupled to the membrane potential via a state variable θ and depends linearly on previous spikes ( see Eqs 4 and 5 ) . A spike-triggered movement of the firing threshold can in principle be implemented by incrementing the value of θ after the emission of each action potential [25] . However , the timescale over which spike-triggered effects occur can in principle differ from the timescale τθ of the threshold-voltage coupling . For this reason , our iGIF model accounts for spike-dependent threshold adaptation with an independent process γ ( t ) . Importantly , the functional shape of γ ( t ) and that of the steady-state function θ∞ ( V ) of the threshold-voltage coupling were not assumed a priori , but were extracted from intracellular recordings using a nonparametric maximum-likelihood procedure . By fitting the iGIF model to data we found that spike initiation is characterized by several timescales . First , the firing threshold is nonlinearly coupled to the subthreshold voltage on a rapid timescale τθ ∈ [5 , 15] ms . Second , spike emission leads to a quasi-instantaneous increase of the firing threshold . Third , the threshold increase triggered by a spike decays over multiple timescales . In agreement with the hypothesis that the coupling between firing threshold and membrane potential results from fast Na+-channel inactivation [21] and confirming the results from a previous study in which a similar model was shown to account for spike-threshold variability in vivo [24] , we found that θ∞ ( V ) was correctly described by a smooth rectifier function ( Fig 4E ) . Moreover , the coupling timescale τθ , the half-inactivation voltage Vi and the asymptotic slope of θ∞ ( V ) were consistent with the biophysical features of Na+-channels expressed in central neurons [21] . Na+-channel inactivation also occurs on slow timescales [71–75] . Slow Na+-channel inactivation has been previously linked to enhanced sensitivity to rapid input fluctuations [15 , 17] and , in our iGIF model , is phenomenologically captured by the spike-triggered component of the firing threshold dynamics . Confirming our earlier results [45] and consistent with the fact that spike-frequency adaptation does not have a preferred timescale [49] , we indeed found that threshold movements induced by previous spikes lasted for several seconds and were characterized by a power-law decay ( see Fig 4D ) . Overall , as discussed in ref . [21] and reviewed in the Materials and Methods , the firing threshold dynamics featured by the iGIF model can be interpreted as phenomenological description of Na+-channels in which inactivation is independently controlled by one fast and several slow gating variables [15 , 71] . While being inferred by maximizing spike-timing prediction ( rather than by directly fitting the somatic voltage at spike onset ) , our model did not simply account for the spike-threshold dependence on input statistics ( see Fig 5 ) . Indeed , it also allowed us to: i ) improve spike-timing prediction ( Fig 6 ) and ii ) explain why the effective timescale of somatic integration is not entirely controlled by the effective membrane timescale and adapts to the input statistics ( see Fig 8 ) . In line with the findings of Fontaine et al . [24] , these results indicate that somatic spike-threshold variability is not a measurement artifact , but a genuine feature of cortical action-potential generators [68] . Comparing the GLM filters κGLM ( t ) extracted by fitting the spiking responses to different current injections revealed that the effective timescale over which neurons integrate their inputs decreases at increasing input strengths μI ( Fig 5D ) . Previous studies in which the response properties of spiking neurons have been analyzed using the Linear Nonlinear Poisson model [76] already suggested that the integration properties of single neurons—as measured by the spike-triggered average of the input [77]—depend on the input statistics ( see , e . g . , refs . [39 , 78 , 79] ) . However , since the spiking activity of single neurons responding in vitro to external currents is strongly non-Poissonian ( see , e . g . , ref . [39] ) , the STAs reported in the above-mentioned studies can not be directly interpreted as the temporal window over which the input current is somatically integrated . Most importantly , input-dependent changes in STA could merely reflect changes in spiking statistics , which are unrelated to changes in somatic integration [36–38] . In contrast to LNP models , GLMs account for non-Poissonian statistics by means of a specific spike-history filter hGLM ( t ) . Consequently , the input filter κGLM ( t ) can characterize changes in intrinsic neuronal properties which are separable from changes in spiking statistics . To test whether the timescale reduction revealed by the GLM-based analysis was due to a conductance increase resulting , e . g . , from the progressive recruitment of a subthreshold adaptation current [18 , 19] , we quantified the effect of μI on the effective membrane filter κ m eff ( t ) . While κGLM ( t ) , together with the spike-history filter hGLM ( t ) , describes how the input current is transformed into a spiking probability , κ m eff ( t ) transforms input currents into subthreshold membrane voltages . Although the membrane filters extracted from the data did shorten at increasing μI , the membrane timescale reduction was not sufficient to explain the timescale reduction revealed by the GLM-based analysis ( Fig 2D ) . This result demonstrates that κGLM ( t ) does not simply reflect the subthreshold membrane response properties , but also accounts for additional mechanisms capable of regulating the effective timescale of integration without affecting the membrane voltage . To elucidate this point and better understand the biophysical meaning of the GLM input filter κGLM ( t ) , we analytically reduced the iGIF model to a GLM [22 , 23] and found that the effective timescale of somatic integration is controlled by: i ) the ratio between the cell capacitance C and leak conductance gL , ii ) the conductance-based spike-triggered adaptation η ( t ) and iii ) the dynamic coupling θ ( t ) between firing threshold and membrane potential ( see Eq 7 ) . Importantly , while the first two neuronal features affect the subthreshold membrane potential and explain the observed membrane timescale reduction ( see Fig 8A–8C ) , the threshold-voltage coupling only acts on the effective timescale of integration , thus explaining the experimental discrepancy between τGLM and τ m eff ( Fig 8C ) . The consequences of fast Na+-channel inactivation for threshold-voltage coupling and shortened effective timescale of somatic integration have been previously studied in refs . [21 , 24] . Briefly , assuming that any postsynaptic spike has already been emitted and that the membrane potential is resting at V0 , a presynaptic spike inducing a postsynaptic current IEPSC ( t ) = ϵδ ( t ) of weak amplitude ϵ will evoke an excitatory postsynaptic potential δVEPSP ( t ) = ϵκm ( t ) , where κm ( t ) is the passive membrane filter . If the firing threshold is coupled to the membrane potential according to Eqs 4 and 5 , this EPSP will in turn evoke a firing threshold increase δ V T , EPSP ( t ) ≈ G θ ( V 0 ) ∫ 0 t κ θ ( s ) δ V EPSP ( t - s ) d s , where κ θ ( t ) = 1 τ θ exp ( - t τ θ ) is a low-pass filter with timescale τθ and the approximation comes from the fact that θ∞ ( V ) has been linearized around V0 ( see Eq 5 ) . Since in our model the spike emission probability depends on the difference between membrane potential and firing threshold , increasing the firing threshold has the same effect as decreasing the membrane potential . Consequently , a model in which the firing threshold is coupled to the membrane potential can be seen as a model with constant firing threshold in which every EPSP is accompanied by an inhibitory postsynaptic potential δVIPSP ( t ) = δVT , EPSP ( t ) with characteristic rise time τθ [21] . In such a model , presynaptic spikes will thus be integrated via an effective postsynaptic potential δVeff ( t ) = δVEPSP ( t ) −δVIPSP ( t ) that decays on a shorter timescale compared to δVEPSP ( t ) . Since the steady-state function θ∞ ( V ) of the threshold-voltage coupling is well described by a smooth rectifier ( see Fig 4E ) , the magnitude of δVIPSP ( t ) depends on the postsynaptic membrane potential V0 via a sigmoidal gain function G θ ( V ) = d θ ∞ d V ( V ) ( see Fig 7A–7C ) . When the postsynaptic potential V0 at spike arrival is smaller than the half-inactivation voltage Vi , Gθ ( V ) is small and the threshold increase ( i . e . , the inhibitory signal ) becomes negligible . However , when V0 > Vi , Gθ ( V ) increases and the effective postsynaptic potential δVeff ( t ) shortens . The fact that the average subthreshold membrane potential increases with the input strength μI finally explains why strong inputs are effectively integrated on shorter timescales ( Fig 7 ) . While confirming that fast Na+-channel inactivation enhances sensitivity to rapid inputs , our results indicate that slow Na+-channel inactivation acts as an homeostatic mechanisms by increasing the mean firing threshold in response to strong inputs . In agreement with the suggestion of Platkiewicz and Brette [21] , we found that slow and fast Na+-channel inactivation interact in a nontrivial way . Indeed , by increasing the mean firing threshold , slow Na+-channel inactivation allows for subthreshold voltage fluctuations to occur at more depolarized voltages V ≫ Vi , where Gθ ( V ) is large ( see Fig 4E ) . Dynamic-clamp experiments have demonstrated that CA1 Pyr neurons operating in high-conductance state switch their behavior from integrators to differentiators [11 , 78] . This complex form of adaptation has been qualitatively reproduced in a phase-plane model according to which a shunting-induced increase of the firing threshold allows for M-currents to activate at subthreshold voltages [78] . The interplay between shunting and subthreshold adaptation reported in ref . [78] shares some similarities with the nonlinear interaction we found between slow , spike-dependent and fast , voltage-dependent threshold adaptation . Two important differences are however to be noticed . First , the subthreshold adaptation process in ref . [78] results from the activation of an M current , which , in contrast to the threshold-voltage coupling featured by our iGIF model , modifies the spike response properties of the cell by altering the effective membrane filter κ m eff ( t ) . In agreement with our result that enhanced sensitivity to rapid input fluctuations is mediated by a dynamic coupling between firing threshold and membrane potential , direct measurements of the subthreshold membrane impedance of CA1 Pyr neurons [11] have shown that the differentiating behavior observed in high-conductance state is not mediated by subthreshold resonance [80] . Second , while in ref . [78] the threshold increase triggering the switch from temporal integration to differentiation is gated by the synaptic input ( i . e . , by the conductance increase resulting from synaptic bombardment ) , the recruitment of the threshold-voltage coupling in the iGIF model is gated by the postsynaptic firing rate via slow , spike-dependent threshold adaptation ( see Fig 4E ) . Overall , our results explain why L5 Pyr neurons maintain sensitivity to rapid input fluctuations regardless of their working regime , confirm theoretical predictions about the firing threshold dyanmics , demonstrate that the firing threshold plays a crucial role in determining the integration properties of single neurons , and shed new light on in vivo studies where sensitivity to rapid signals has been linked to the voltage threshold for spike initiation [58 , 59 , 61] . The statistical properties of sensory stimuli ( such as , e . g . , visual inputs or whisker movements ) are complex and vary over time . Given their limited dynamic ranges , sensory systems must thus constantly adapt their coding strategies in order to provide an efficient representation of the external world [81 , 82] . Adaptation is a hallmark of virtually all sensory systems and occurs over multiple timescales . Many sensory systems adapt their input-output transformation not only to the mean , but also to the variance and even to higher-order moments of the input statistics [83] . Over the last decades , sensory adaptation has been repeatedly investigated using an experimental paradigm known as the switching experiment [83] . In a switching experiment , the spiking activity of a neuron is recorded in vivo , while the animal is presented with a controlled stimulus that rapidly fluctuates over time . To assess adaptation to local input statistics , the stimulus is generated according to a Gaussian process whose mean ( or variance ) is periodically switched between two values . In response to a sudden change in the variance ( i . e . , the contrast ) of a visual input , both retinal ganglion cells and motion-sensitive neurons in the fly feature two forms of adaptation [55 , 56] . Right after a change in input contrast , these neurons rapidly modify the shape of their receptive field , thereby adapting the stimulus feature to which they are preferentially responsive . While this mechanism is very fast , the same neurons also feature a slower form of adaptation that manifests itself in a decay of the output firing rate over multiple timescales . This second mechanism , known as spike-frequency adaptation , does not induce further changes in the receptive field , but simply reduces the overall excitability of the neuron . Since both the timescale and the net effect of these two adaptation processes are different , it has been hypothesized that changes in feature selectivity and output firing rate are controlled by two independent mechanisms [83] . Whether and how both forms of adaptation can be mediated by intrinsic neuronal properties remains unclear . To investigate this issue , we performed a switching experiment in silico by presenting our iGIF model with an in vivo-like fluctuating current , whose mean μI periodically switched between two values ( Fig 8E–F ) . Our results predict that , in response to a sudden change in the mean input , L5 Pyr neurons rapidly modify their receptive field , that is , the temporal window according to which the input current is somatically integrated . Moreover , the spiking response predicted by our iGIF model was characterized by spike-frequency adaptation that occurred on much slower timescales . Overall , our results suggest that the forms of sensory adaptation reported in refs . [55 , 56] do not required network effects , but can in principle be implemented by two qualitatively distinct cellular mechanisms: a fast , nonlinear coupling between firing threshold and subthreshold voltage—possibly mediated by fast Na+-channel inactivation—and a slow , spike-triggered processes—possibly mediated by slow Na+-channel inactivation and by other ion-channels mediating afterhyperpolarization currents . Cortical neurons feature a strong nonlinear behavior , making single-neuron computation dependent on the input statistics [10–12 , 39] . During the last decade , a number of simplified threshold models [22 , 29 , 41 , 53 , 84–86] , including our previous GIF model [44 , 45] , have been shown to accurately predict the spiking response evoked in vitro by stationary ( or quasi-stationary ) currents [51 , 87 , 88] . Simplified threshold models are usually obtained by partially linearizing the dynamics of conductance-based biophysical models [21 , 89 , 90] . Thus , when assessed on a broad range of input statistics , their performance generally drops [14] ( see also Fig 5 ) . For the same reason , model parameters extracted from responses to different inputs generally differ [13] ( see also Fig 2 ) . These results reflect the inability of simplified threshold models of capturing the nonlinear dynamics underlying complex forms of single neuron adaptation . Overall , designing and fitting to data a simplified spiking model capable of predicting the electrical activity of cortical neurons operating in different regimes remains a big challenge [14] . Indeed , increasing the complexity of a spiking neuron model rapidly makes parameter estimation a difficult problem [91] . To solve this problem , we introduced the inactivating Generalized Integrate-and-Fire ( iGIF ) model which extends the standard Leaky Integrate-and-Fire model in three directions . First , noise is introduced by the escape-rate model for stochastic spike generation [22 , 29 , 46] . Second , a spike-triggered conductance η ( t ) [47] and a spike-triggered movement of the firing threshold γ ( t ) [85 , 92 , 93] are included for spike-frequency adaptation . Third , a nonlinear coupling θ between membrane potential and firing threshold [21 , 25 , 48] is added for enhanced sensitivity to input fluctuations [7] . The iGIF model can be related to a conductance-based model in which: i ) the combined activity of voltage-activated and Ca2+-activated K+-channels [94–96] results in the spike-triggered conductance increase η ( t ) [47 , 97] , ii ) Na+-channels are gated by a fast inactivation variable [98] implementing the nonlinear coupling θ between firing threshold and membrane potential [20 , 21] and iii ) Na+-channels are also gated by an additional set of slow inactivation variables [71 , 99] , whose combined activity results in a spike-triggered threshold movement γ ( t ) decaying over multiple timescales [20 , 21] . While the biophysical interpretation of the spike-triggered conductance η ( t ) is well established ( see , e . g . , refs . [47 , 97] ) , the link between the inactivation properties of Na+-channels and the firing threshold dynamics featured by the iGIF model is more involved and thus deferred to the Materials and Methods section . Would such a conductance-based model perform better than our iGIF model ? Since fitting detailed biophysical models to electrophysiological data is cumbersome [51] , especially in situations where the ion-channel dynamics are not known a priori , answering this question is not trivial . In contrast to conductance-based models , the iGIF model accounts for the intricate dynamics of different ion-channels with simple phenomenological descriptions . Consequently , its parameters can be efficiently extracted from intracellular recordings with a new two-step procedure , which extends previous methods [22 , 44 , 45 , 84] . In contrast to previous studies [27 , 100] , where optimal parameters of spiking neuron models have been inferred directly from the f-μI curves , our fitting procedure exploit the information contained in the subthreshold membrane potential fluctuations , thus allowing for the characterization of adaptation mechanisms resulting from the firing threshold dynamics [34] . Despite its relative simplicity , the iGIF model captures complex forms of single neuron adaptation , providing a good description of L5 Pyr neurons over an extended range of input statistics . Despite its relative complexity , the iGIF model can be analytically mapped to a GLM with input-dependent filters . We conclude that the iGIF model provides an accurate , yet intuitive description of single-neuron computation over a broad range of input statistics .
All procedures in this study were conducted in conformity with the Swiss Welfare Act and the Swiss National Institutional Guidelines on Animal Experimentation for the ethical use of animals . The Swiss Cantonal Veterinary Office approved the project following an ethical review by the State Committee for Animal Experimentation . Somatic whole-cell in vitro current clamp recordings were performed on 300 μm thick parasagittal acute slices from the right hemispheres of male P13-P15 C57Bl/6J wild-type mice . Brains were quickly dissected and sliced ( HR2 vibratome , Sigmann Elektronik , Germany ) in ice-cold artificial cerebrospinal fluid ( ACSF ) ( in mM: NaCl 124 . 0 , KCl 2 . 50 , MgCl2 10 . 0 , NaH2PO4 1 . 25 , CaCl2 0 . 50 , D- ( + ) -Glucose 25 . 00 , NaHC03 25 . 00; pH 7 . 3 , s . d . 0 . 1 , aerated with 95% O2 / 5% CO2 ) , followed by a 15 minute incubation at 34°C in standard ACSF ( in mM: NaCl 124 . 0 , KCl 2 . 50 , MgCl2 1 . 00 , NaH2PO4 1 . 25 , CaCl2 2 . 00 , D- ( + ) -Glucose 25 . 00 , NaHC03 25 . 00; pH 7 . 40 , aerated with 95% O2 / 5% CO2 ) . To ensure intact axonal and dendritic arborisation , electrophysiological recordings were conducted in slices cut parallel to the apical dendrites . Recordings in Layer 5 of the primary somatosensory cortex were performed at 32 , s . d . 1°C in standard ACSF with an Axon Multiclamp 700B Amplifier ( Molecular Devices , USA ) using 5–7 MΩ borosilicate pipettes , containing ( in mM ) : K+-gluconate 110 . 00 , KCl 10 . 00 , ATP-Mg2+ 4 . 00 , Na2-phosphocreatine 10 . 00 , GTP-Na+ 0 . 30 , HEPES 10 . 00 , biocytin 5 . 00 mg/ml; pH 7 . 30 , 300 mOsm . Cells were visualized using infrared differential interference contrast video microscopy ( VX55 camera , Till Photonics , Germany and BX51WI microscope , Olympus , Japan ) . Data were acquired with sampling frequency ΔT−1 = 10 kHz using an ITC-18 digitizing board ( InstruTECH , USA ) controlled by a custom-written software module operating within IGOR Pro ( Wavemetrics , USA ) . Voltage signals were low-pass filtered ( Bessel , 10 kHz ) and not corrected for the liquid junction potential of +14 . 5 mV estimated from Igor XOP Patcher’s Calculator ( courtesy of Drs . F . Mendez and F . Würriehausen , MPI for Biophysical Chemistry , Göttingen , Germany ) . Consequently , the membrane potentials and the firing thresholds reported in this study are positively biased by 14 . 5 mV . Only cells with an access resistance ≤ 20 MΩ ( 17 . 7 , s . d . 2 . 3 MΩ , n = 6 ) were retained for further analysis . In all the experiments included in this study , neurons were stimulated with fluctuating currents I ( t ) generated according to an Ornstein-Uhlenbeck process: τ I I ˙ ( t ) = - I ( t ) + μ I + 2 τ I σ I · ψ ( t ) , ( 8 ) where ψ ( t ) is a Gaussian white-noise process with zero mean and unitary variance , τI is the correlation timescale , μI is the mean current and σI defines the magnitude of the fluctuations ( that is , the standard deviation of the current ) . The temporal correlation of the input was fixed to τI = 3 ms and input currents I ( t ) were generated at a sampling rate ΔT−1 = 10 kHz . To measure the impact of input fluctuations on the single-neuron input-output transfer function ( i . e . , the f-μI curve ) , we somatically injected a set of 5-second currents with different means μI and standard deviations σI . To let the cell recover , injections were performed with interstimuli intervals of 25 seconds . Similar protocols have already been applied in previous studies [7 , 8 , 27] . Here , to broadly explore the parameter space ( μI , σI ) and to accurately estimate the experimental f-μI curves , we considered four different standard deviations σI ∈ {0 , 50 , 100 , 150} pA and eight different means μI ∈ [0 , μmax] nA , with μmax begin cell-dependent . Each neuron was stimulated with 32 different inputs that were presented randomly . This protocol was repeated 3 times giving a total number of 96 current injections . When stimulated with strong inputs , pyramidal neurons undergo spike failure and can not sustain repetitive firing for long periods of time ( see , e . g . , ref . [71] ) . At the beginning of each experiment , the maximum current μmax was defined in such a way as to reach saturation of the steady-state firing rate while preventing spike failures . For that , neurons were tested with 6-s-long noiseless currents ( i . e . , σI = 0 ) of increasing magnitude μI . Cells that could not sustain continuous firing for a DC current μI < 0 . 4 nA were not further considered . The maximal mean input μmax was comprised between 0 . 4 and 0 . 55 nA . To evaluate model performance in predicting the occurrence of individual spikes , a different set of experiments was performed . Currents were generated according to Eq 8 , but in this case , the stochastic process used to generate the input was made non-stationary by modulating the standard deviation σI with a sinusoidal function of time: σ I ( t ) = σ 0 1 + 1 2 · sin 2 π T · t , ( 9 ) with T = 5 s being the modulation period . For each cell , input parameters were calibrated to obtain an average firing rate of 10 Hz oscillating between 7 and 13 Hz , approximatively . After calibration , input parameters were in the following ranges: μI ∈ [120 , 190] pA , σ0 ∈ [120 , 190] pA . Since the spiking responses of both neurons and GIF models are stochastic , spike-timing prediction was quantified on a test set obtained by 9 repetitive injections of the same ( i . e . frozen-noise ) 20-s current generated according to Eqs 8 and 9 . For parameter extraction , a training set was used in which single neurons were stimulated with a single 120-s current having the same statistics as the test set , but in which a different realization of the white-noise process ψ ( t ) was used . All the injections were performed with inter-stimuli intervals of 25 seconds . When acquired with the same electrode used to inject the external input I ( t ) , current-clamp recordings Vrec ( t ) are biased versions of the membrane potential Vdata ( t ) [41] . This bias can in principle be removed using series resistance or bridge balance compensation . However , perfect calibration of these methods is technically difficult to achieve . Moreover , during long experiments , the electrode properties , and in particular the series resistance Re , are subject to change [45] . Quantitative comparison between membrane potentials evoked by input currents having different offsets μI requires accurate electrode compensation . Indeed , a non-neutralized series resistance R ˜ e would lead , on average , to a mean input-dependent bias V bias ( μ I ) = R ˜ e μ I ( see , e . g . , ref . [45] ) . To avoid this and others problems , for all the in vitro recordings included in this study , online series resistance compensation was complemented by Active Electrode Compensation ( AEC ) [41 , 101] . For that , the same procedure applied as in ref . [45] was used . In case of long experiments , estimating the electrode properties at different moments in time can improve the quality of the data by removing drifts due to slow changes in the electrode properties [45] . For this reason , electrode filters used for AEC were extracted from 10-s subthreshold injections performed before the training set , before the test set and every sixteen injections in the protocol used to measure the f-μI curves . Subthreshold input currents were generated according to Eq 8 with μI = 0 nA , σI = 75 pA and τI = 3 ms . Following ref . [7] , we estimated the voltage threshold for each spike in the dataset by measuring the membrane potential Vdata at which the depolarization rate dVdata/dt became larger than 10 mV/ms ( see Fig 3E–3H ) . With this definition , threshold crossing always occurred less than 1 ms before the membrane potential reached 0 mV . Since at the soma spike initiation is very sharp ( see ref . [69 , 102] and Fig 3F and 3H ) and since our analysis is only based on relative variations between the voltage threshold in different conditions , rather than on absolute values , the exact definition does not matter [20] . In Fig 3C and 3D , the average subthreshold membrane potential was computed by discarding all the data points { t | t ∈ [ t ^ j - 2 ms , t ^ j + 10 ms ] } that were too close to action potentials { t ^ j } . In the GLM [22 , 23] , spikes are generated stochastically with firing intensity λGLM ( t ) defined as: λ GLM ( t ) = λ 0 · exp ∫ 0 ∞ κ GLM ( s ) I ( t - s ) d s + ∑ t ^ j < t h GLM ( t - t ^ j ) , ( 10 ) where λ0 is a constant , κGLM ( t ) is an arbitrarily-shaped filter through which the input current is effectively integrated and hGLM ( t ) accounts for all spike-triggered processes that make the single-neuron activity history-dependent [46] . GLM parameter extraction is performed using the standard maximum likelihood method described in refs . [22 , 23] . For that , both κGLM ( t ) and hGLM ( t ) were expanded in linear combinations of rectangular basis functions . In the main text , the GLM filters analytically derived from the iGIF model ( see below ) are denoted κ ^ GLM ( t ) and h ^ GLM ( t ) . In the iGIF model , spikes are produced stochastically according to the conditional firing intensity λ ( t ) defined by the exponential escape-rate model [29 , 46]: λ ( t ) = λ 0 exp V ( t ) - V T ( t ) Δ V , ( 11 ) where V ( t ) is the membrane potential , VT ( t ) is a dynamic threshold , ΔV defines the level of stochasticity and , without loss of flexibility , we fixed λ0 = ΔT−1 = 10 kHz . In the limit ΔV → 0 , the iGIF model becomes deterministic and action potentials are fired reliably each time t ^ the firing threshold is reached ( i . e . , when V ( t ^ ) = V T ( t ^ ) ) . When ΔV > 0 , the iGIF model is stochastic , and the probability of emitting an action potential at time t ^ ∈ [ t , t + d t ] is given by [6]: p ( t ^ ∈ [ t , t + d t ] ) = 1 - exp - ∫ t t + d t λ ( s ) d s , ( 12 ) meaning that , when V = VT , a spike is emitted during a time step of ΔT = 0 . 1 ms with probability p = 0 . 63 . When ΔV > 0 , spikes can be emitted even if the membrane potential is lower than the firing threshold . Similarly , the membrane potential can cross the firing threshold without evoking a spike . With increasing ΔV , spike emission becomes more stochastic and progressively loses sensitivity to V ( t ) −VT ( t ) . The maximal level of stochasticity is reached when ΔV → ∞ . In this limit , the iGIF model becomes formally equivalent to an homogeneous Poisson process with stochastic intensity λ0 . In the iGIF model , probabilistic spike emission is the only source of stochasticity . Importantly , the level of stochasticity ΔV is not assumed a priori , but is extracted from experimental data along with all the other parameters ( see below ) . The dynamics of the subthreshold membrane potential are modeled as a leaky integrator augmented with a spike-triggered conductance η ( t ) that describes the time course of the conductance change after a spike . More precisely , the membrane potential evolves according to the following differential equation: C V ˙ = - g L ( V - E L ) + I - ∑ t ^ j < t η ( t - t ^ j ) · ( V - E R ) , ( 13 ) where C , gL and EL describe the passive properties of the membrane , τm = C/gL is the passive membrane timescale , { t ^ 1 , t ^ 2 , t ^ 3 , … } are the spike times , ER is a reversal potential and I is the external input . Conductance changes triggered by different spikes accumulate and produce spike-frequency adaptation ( or facilitation ) . The functional shape of η ( t ) is not assumed a priori , but is extracted from experimental data ( see below ) . After each spike , the membrane potential is reset to Vreset and the numerical integration only restarts after an absolute refractory period Tref . The absolute refractory period was set to Tref = 4 ms and the voltage reset was estimated by computing the average membrane potential after a spike ( i . e . V reset = 〈 V ( t ^ j + T ref ) 〉 j ) . Since a period of absolute refractoriness can also be implemented by setting the first milliseconds of the spike-triggered threshold movement to high values , the particular choice of Tref is not crucial . The dynamics of the firing threshold VT are given by: V T ( t ) = θ ( t ) + ∑ t ^ j < t γ ( t - t ^ j ) , ( 14 ) where γ ( t ) describes the movement of the firing threshold after the emission of an action potential . Similar to η ( t ) , the functional shape of γ ( t ) is not assumed a priori , but is extracted from the data . Since γ ( t ) can only account for spike-dependent effects , the model is augmented with an additional state variable θ ( t ) , which couples the firing threshold to the subthreshold membrane potential . Based on theoretical results obtained by a systematic reduction of the Hodgkin-Huxley model ( see next section ) , this coupling is expected to be nonlinear [20] . In the iGIF model , the dynamics of θ ( t ) are given by τ θ θ ˙ = - θ + V T * + θ ∞ ( V ) , ( 15 ) where V T * is a constant , τθ is the characteristic timescale on which the threshold reacts to changes in the membrane potential and θ∞ ( V ) is the voltage-dependent steady-state towards which θ converges . To avoid a priori assumptions on the biophysical processes underlying the threshold-voltage coupling , θ∞ ( V ) is defined in the iGIF-free model as an arbitrary function of the membrane potential and is extracted from experimental data using a novel non-parametric maximum likelihood approach ( see below ) . Spike-dependent movements of the firing threshold are modeled by γ ( t ) and the state variable θ is reset to V T * after each spike . The iGIF-Na model is defined exactly as the iGIF-free model except for the fact that the dynamics of θ ( t ) are predefined based on the biophysics of Na+-channels [21] and is given by Eqs 5 and 6 . The reasons for this choice are reviewed now . In order to extract from experimental data the effective membrane timescale τ m eff ( see Fig 2D ) , intracellular recordings were split in different datasets according to μI and independently fitted with a Leaky Integrate-and-Fire model equipped with a spike-triggered current η C ( t - t ^ j ) , as opposed to a spike-triggered conductance η ( t - t ^ j ) . In other words , we used a model obtained by dropping the term ( V−ER ) from Eq 13 and replacing gL by an effective conductance g ˜ L eff and replacing η ( t ) by ηC ( t ) . By fitting this model to data using a linear regression similar to Eq 19 , the average conductance increase mediated by spike-dependent processes is included in the effective leak conductance g ˜ L eff . The effective conductance-induced membrane timescale is thus given by τ m eff = C / g ˜ L eff . In the main text , the effective membrane timescale analytically derived from the iGIF model ( see below ) is denoted τ ^ m eff . In order to understand how different adaptation processes captured by the iGIF model affect single-neuron computation , we systematically reduced the iGIF model to a GLM . The reduction is computed in three steps . To avoid problems related to overfitting and to allow for a comparison between models that differ in the total number of parameters , the performance reported in this study were , unless specified otherwise , evaluated on separate data sets that were not used for parameter extraction . A quantitative measure of the quality of both the GIF and the iGIF model is provided by the log-likelihood [23]: L L model = ∑ t ∈ { t ^ } log λ model ( t ) - ∫ 0 T λ model ( t ) d t ( 36 ) where λmodel ( t ) is the conditional firing intensity of the model after parameter optimization , { t ^ } is the experimental spike train and T is the total duration of the experiment on which the model performance were evaluated . All of the log-likelihoods reported in this study were normalized with respect to a homogenous Poisson process with constant intensity defined by the experimental firing rate r ¯ = N spikes / T , as well as with respect to the total number of spikes Nspikes [23]: L L = 1 log ( 2 ) · N spikes L L model - N spikes ( log r ¯ - 1 ) , ( 37 ) such that units are in bit per spike . Spike-timing prediction was quantified using the spike-train similarity measure M d * [105] . As in our previous studies [44 , 45] , M d * was computed using the Kistler coincidence window with a temporal granularity of Δ = ±4 ms .
|
Over the last decades , a variety of simplified spiking models have been shown to achieve a surprisingly high performance in predicting the neuronal responses to in vitro somatic current injections . Because of the complex adaptive behavior featured by cortical neurons , this success is however restricted to limited stimulus ranges: model parameters optimized for a specific input regime are often inappropriate to describe the response to input currents with different statistical properties . In the present study , a new spiking neuron model is introduced that captures single-neuron computation over a wide range of input statistics and explains different aspects of the neuronal dynamics within a single framework . Our results indicate that complex forms of single neuron adaptation are mediated by the nonlinear dynamics of the firing threshold and that the input-output transformation performed by cortical pyramidal neurons can be intuitively understood in terms of an enhanced Generalized Linear Model in which both the input filter and the spike-history filter adapt to the input statistics .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"action",
"potentials",
"medicine",
"and",
"health",
"sciences",
"membrane",
"potential",
"electrophysiology",
"neuroscience",
"systems",
"science",
"mathematics",
"forecasting",
"statistics",
"(mathematics)",
"computational",
"neuroscience",
"cellular",
"structures",
"and",
"organelles",
"research",
"and",
"analysis",
"methods",
"computer",
"and",
"information",
"sciences",
"animal",
"cells",
"mathematical",
"and",
"statistical",
"techniques",
"statistical",
"methods",
"nonlinear",
"dynamics",
"biophysics",
"cell",
"membranes",
"physics",
"intracellular",
"membranes",
"cell",
"biology",
"physiology",
"neurons",
"single",
"neuron",
"function",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"physical",
"sciences",
"computational",
"biology",
"neurophysiology"
] |
2016
|
Enhanced Sensitivity to Rapid Input Fluctuations by Nonlinear Threshold Dynamics in Neocortical Pyramidal Neurons
|
Very little is known about how vector-borne pathogens interact within their vector and how this impacts transmission . Here we show that mosquitoes can accumulate mixed strain malaria infections after feeding on multiple hosts . We found that parasites have a greater chance of establishing and reach higher densities if another strain is already present in a mosquito . Mixed infections contained more parasites but these larger populations did not have a detectable impact on vector survival . Together these results suggest that mosquitoes taking multiple infective bites may disproportionally contribute to malaria transmission . This will increase rates of mixed infections in vertebrate hosts , with implications for the evolution of parasite virulence and the spread of drug-resistant strains . Moreover , control measures that reduce parasite prevalence in vertebrate hosts will reduce the likelihood of mosquitoes taking multiple infective feeds , and thus disproportionally reduce transmission . More generally , our study shows that the types of strain interactions detected in vertebrate hosts cannot necessarily be extrapolated to vectors .
Interactions between pathogen strains within hosts can be profound and affect many aspects of infectious disease biology , including disease severity and infectiousness , as well as the evolution of virulence and the spread of drug resistance [1–7] . Yet for medically important vector-borne diseases , very little is known about the nature and implications of strain interactions within the vector . This is in striking contrast to what is known about strain interactions in the vertebrate host . For example , malaria parasites in mixed strain infections experience significant competitive suppression within the vertebrate host [8–17] . Whether competitive suppression also occurs in their mosquito host is unknown . The progression through the vector is relatively long and complex [18] and involves severe population bottlenecks [19] . Parasite density also influences both the development of the parasite and the probability of the vector surviving for long enough to infect a new host [20–22] . Therefore , strain interactions that increase or decrease parasite density are likely to alter the probability of transmission to a new vertebrate host . Mixed strain ( genotype ) infections in mosquitoes are common [23 , 24] and there are three distinct but non-exclusive routes by which they could arise . First , multiple parasite strains could be taken up from a host during a single blood meal . Mixed strain infections are the norm in areas of high transmission [25] , and multiple parasite strains can be transmitted to a vector from a single infective feed [26] . Second , mosquitoes that are disturbed during feeding may move to a new host , resulting in multiple hosts contributing blood to one feeding cycle [27–29] . Finally , mosquitoes could feed on different hosts in successive blood feeding cycles . Studies on human and bird malaria parasites have suggested that mosquitoes that take multiple infective feeds have higher oocyst burdens and parasites at different stages of development , which is suggestive of the accumulation of infections over multiple feeding cycles [30–33] . What impact this has on parasite development or on vector survival not been previously tested . If secondary infections are equally likely to be acquired , then of the mosquitoes surviving to become infectious , up to ~40% of infectious mosquitoes could have oocysts , and up to ~17% could have sporozoites originating from multiple feeds ( Fig A in S1 Text ) . The possibility that mosquitoes can acquire mixed infections from multiple feeds is interesting in its own right , but experimentally , infection from successive blood meals would also provide a way to analyse the competitive interactions between strains without the confounding problems of strain recombination . Parasites in the same blood meal freely recombine in the mosquito gut . There can be no recombination between strains acquired in different feeding cycles because zygotes are formed within a few minutes of a blood meal . When a successive meal takes place several days later , all gametes from the first meal are gone [34] . Here we show that mosquitoes can accumulate mixed strain infections from feeding on multiple hosts , and that the presence of oocysts from an existing parasite infection make subsequent infections more likely and more productive . Additionally , we show that vector mortality was no higher for double infections than for infections with a single parasite strain .
An initial study ( experiment 1 ) was conducted to test whether it is possible for mosquitoes to pick up multiple infections from multiple bloodmeals . Six cages , each containing ~100 three to five day old Anopheles stephensi female mosquitoes were used . Half of the cages fed on mice infected with the rodent malaria parasite Plasmodium chabaudi ( strain ER ) , and half received an uninfected blood meal ( control ) . Four days after their initial feed , all cages of mosquitoes received a second blood meal containing P . chabaudi strain AJ parasites . This 4 day schedule corresponds to the preferred blood-feeding frequency for female mosquitoes [35–37] . Seven days after the second blood meal ( experimental day 11 ) when parasites from the second feed were expected to have established as mature oocysts , ~30 mosquitoes per cage were removed , dissected and tested for the presence and density of each of the parasite strains by genotype specific PCR on infected midguts ( see Table 1 for treatment groups and sample sizes ) . A comparison with mosquitoes dissected four days earlier confirmed that our ability to detect infections from the first feed did not decline over this time period ( S1 Fig ) . We found that mosquitoes become doubly infected with parasites from successive blood meals . A total of 31% ( ±5 . 3 SEM ) of mosquitoes became infected with ER parasites during their first feed and of these infected mosquitoes 50% ( ±10 . 4 SEM ) additionally became infected with AJ parasites during their second feed ( Fig 1 ) . We then conducted a second larger study ( experiment 2 ) with 21 cages , again each containing ~100 female mosquitoes . Six cages received two infective blood meals with one each of our two parasite strains ( 3 x AJ-ER and 3 x ER-AJ ) , six cages received an infective blood meal only on their first feed ( 3 x AJ-C and 3 x ER-C ) , six cages received an infective blood meal only on their second feed ( 3 x C-AJ and 3 x C-ER ) , and finally three cages received two uninfected blood-meals ( C-C ) ( Table 1 ) . All cages received two blood meals with mosquitoes in single infection treatments being given an uninfected feed in place of one of the infective blood meals . This was done in order to control for any effect of a second blood meal on parasite replication [38] . This fully factorial study design allowed us to examine how the presence of a co-infecting strain affects parasites that enter the vector first and second , and to test whether co-infection impacts vector survival . The six cages that received two infective feeds were all found to contain mosquitoes infected with parasites of both strains . For cages which fed on ER first and AJ second , 75% ( ±4 . 6 SEM ) of mosquitoes became infected on their first feed , and of these 25% ( ±5 . 3 SEM ) additionally became infected with AJ . For cages that fed on AJ first and ER second , 25% ( ±4 . 7 SEM ) of mosquitoes became infected on their first feed , and of these 78% ( ±8 . 8 SEM ) additionally became infected with ER ( Fig 1 ) . Therefore both parasite strains were able to establish in already infected vectors . It was not possible to determine which feed individual oocysts originated from , but by using quantitative PCR we were able to determine the genome count ( total number of potential sporozoites produced ) for each of our strains within each infected mosquito midgut . The production of sporozoites within the oocyst requires the acquisition of ( presumably limited ) nutrients from the mosquito [27 , 39] and has previously been shown to be negatively related to oocyst density [20] . Due to anaemia and immune factors from the vertebrate host , infective bloodmeals are also likely to be lower quality . Therefore , we predicted that the host infection status and/or the establishment of a new infection during oocyst development would negatively impact parasite replication ( competitive suppression ) . However , the host infection status of the second bloodmeal ( infective or control ) did not affect the number of genomes from the first infection for either of our focal strains ( Treatment ( infective or control ) : χ2 = 0 . 01 , p = 0 . 77; Treatment*Focal strain: χ2 = 0 . 20 , p = 0 . 66; Fig 2; Table B in S2 Text ) . When we split our infective treatment group by whether the second infection established or not , we found no effect of secondary infection on AJ ( Control vs . Infected: z = 0 . 24 , p = 0 . 99; Fig 2; Table B in S2 Text ) , and for ER infections genome numbers were actually slightly higher in mosquitoes which were subsequently infected with AJ ( Control vs . Infected: z = 3 . 49 , p = 0 . 01; Fig 2; Table B in S2 Text ) . This suggests that the development of established malaria infections is not negatively impacted by secondary infections . In our first experiment , AJ was used as our focal strain and was more than five times as likely to infect mosquitoes already infected with ER than mosquitoes which had previously received a control feed or had been exposed to ER on the first feed but had not become infected ( previous infection status Χ2 = 21 . 38 , p<0 . 0001; Fig 3; Table C in S2 Text ) . In our second experiment , we measured how previous infection affected the establishment of parasites received during the second bloodmeal for both our strains . In agreement with experiment 1 , mosquitoes which had become infected during their first feed were much more likely to then become infected on their second feed ( infection with focal strain ~ previous infection status Χ2 = 7 . 09 , p<0 . 01; Fig 3; Table C in S2 Text ) . Infection probabilities varied with focal strain and experiment ( Fig 3 ) , which was likely due to mice having lower gametocyte densities for AJ infections in experiment 2 ( Table A in S2 Text ) . However , the relative increase in infection probability during a second feed for previously infected mosquitoes remained consistent ( previous infection status*focal parasite strain in experiment 2: Χ2 = 0 . 44 , p = 0 . 80; previous infection status*experiment for AJ: Χ22 , 7 = 0 . 99 , p = 0 . 32 ) . Therefore the presence of parasites from a previous infection increased the probability of a new infection for both our focal strains and in replicate experiments . The observed increase in infection probability during the second bloodmeal for mosquitoes infected during the first could be due to ( 1 ) mosquito variation in susceptibility , so that some individuals had a higher likelihood of infection during both feeds , ( 2 ) blood-meal quality of the first feed having knock on effects for the second feed ( for example , feeding on an anaemic mouse for the first blood-meal could result in mosquitoes taking a larger second blood-meal ) , or ( 3 ) the first infection facilitating the establishment of the secondary infection ( either through physical damage to the midgut , changes in resource availability , or immune depletion ) . In each of our experiments , mosquitoes where randomly allocated to experimental cages from the same cohort of inbred mosquitoes . It is therefore unlikely that there would be variation in susceptibility between cages , although it is possible that there could be variation in susceptibility between mosquitoes within cages . If there were a subset of mosquitoes refractory to infection in each cage we would expect i ) the total number of mosquitoes in each cage to remain constant ii ) mosquitoes which failed to become infected during their first feed would be less likely than controls to become infected during a second feed . In both our experiments , cages which received two infectious feeds had an overall higher prevalence of infection from the second feed than in control cages ( Χ21 , 4 = 6 . 07 , p = 0 . 034 ) , suggesting the increase in susceptibility in these cages was occurring over and beyond the background level of infection . Additionally , previously exposed but uninfected mosquitoes were just as likely to become infected on their second bloodmeal as mosquitoes from control cages ( Experiment 1: X2 = 2 . 04 , p = 0 . 1; Experiment 2: X2 = 2 . 05 , p = 0 . 2; Fig 3; Table C in S2 Text ) and therefore did not represent a refractory subset of individuals . Differences in blood-meal quality per se are also unlikely to explain increased transmission to already infected mosquitoes: mosquitoes that had previously received a control feed or had received an infective feed but remained uninfected were equally likely to become infected during their second bloodmeal ( Control vs . exposed: Experiment 1: X2 = 2 . 04 , p = 0 . 1; Experiment 2: X2 = 2 . 05 , p = 0 . 2; Fig 3 ) , and there was no effect of the mouse red blood cell density on probability of infection ( Experiment 1: Χ2 = 0 . 01 , p = 0 . 99; Experiment 2: Χ2 = 0 . 10 , p = 0 . 75 ) . By a process of elimination , it seems most likely that the presence of a primary infection directly increases the chance of a secondary infection establishing . In order to determine how this occurs ( e . g . whether through interactions with vector immunity , resources , or physical damage to the mosquito midgut ) more experiments are needed . As expected , overall oocyst burdens were higher in mosquitoes that were infected during both bloodmeals compared to mosquitoes infected only on their second bloodmeal . However , the magnitude of this effect depended on the order of strains in the double infections . The highest oocyst burdens were found in mosquitoes with AJ infections followed by ER infections ( oocyst density ~ infection status*focal parasite: X2 = 9 . 22 , p<0 . 005; Fig 4; Table D in S2 Text ) . It was not possible to reliably determine which infection individual oocysts resulted from , but we were able to compare genome counts for our focal infections developing in double infections those in matched single infections ( controls ) . Infections that established in already infected mosquitoes had higher genome counts than those that established in previously uninfected ( naïve ) mosquitoes ( X2 = 8 . 15 , p<0 . 005; Fig 5; Table E in S2 Text ) . The magnitude of this effect depended on the focal strain ( genome count 6 x higher for ER but over 300 x higher for AJ; Fig 5 ) . Higher genome counts in already infected mosquitoes could have been due to some mosquitoes being more susceptible to both infections , but genome counts from the first and second infections for double infected mosquitoes were unrelated ( Χ21 , 8 = 0 . 002 , p = 0 . 97; S2 Fig ) . Therefore , the presence of parasites from a prior infection increases both the chances that subsequent infection will establish , and the density that subsequent infection will reach in the mosquito . The probability that parasites will be transmitted to a new vertebrate host depends both on the ability of the parasite to establish and replicate within the vector and the potential number of infective bites a vector can take , which will depend on how many blood feeding cycles the mosquito survives for . We performed a comprehensive examination of the impact of infection status on vector survival . A total of 1631 mosquitoes across 21 cages were monitored twice daily until death ( our longest lived mosquito died 72 days after receiving its first bloodmeal ) . Three cages fed on uninfected mice during both blood meals ( C-C ) , 12 cages fed on control mice for one bloodmeal and infective mice for the other ( C-AJ , C-ER , AJ-C or ER-C ) , and 6 cages fed on infective mice during both bloodmeals ( AJ-ER or ER-AJ ) ( Table 1 ) . Dead mosquitoes were tested for the presence of infection and identity of the infecting strain ( s ) using PCR . There was no significant difference in survival between control uninfected mosquitoes and exposed but uninfected mosquitoes ( Χ21 , 615 = 0 . 003 , p = 0 . 96 ) , therefore these groups were analysed together giving us 4 groups for comparison ( uninfected; infected with AJ; infected with ER; infected with both strains ) . While PCR of mosquito cadavers allowed us to directly determine infection status ( uninfected , infected with AJ , infected with ER , or double infection ) for mosquitoes used in survival analysis oocyst counts from dead mosquitoes are not possible . Therefore , a mean oocyst density was calculated from a subset of ~30 mosquitoes per cage which were removed and dissected 7 days after each infective bloodmeal . Dissected mosquitoes were counted as censored points in the survival analysis . Total gametocyte densities were taken as the summed gametocyte density from the two feeds taken by each mosquito and red blood cell density was the mean of the two feeds . Across all groups there were no significant relationships between mosquito survival and red blood cell density in the blood-meals ( Χ2 = 0 . 001 , p = 0 . 97 ) , mean oocyst density ( Χ2 = 0 . 84 , p = 0 . 36 ) , or gametocyte density ( Χ2 = 3 . 04 , p = 0 . 08 ) , therefore these factors were dropped from the statistical models ( Table F in S2 Text ) . There was a significant effect of infection status on mosquito survival ( 4 level factor; uninfected , AJ infection , ER infection , double infection; Χ23 , 891 = 9 . 53 , p = 0 . 024 ) . However the only significant pairwise comparison was between uninfected mosquitoes and those infected with AJ alone ( AJ vs . uninfected: Χ21 , 673 = 6 . 5 , p = 0 . 01; ER vs . uninfected: Χ21 , 810 = 1 . 05 , p = 0 . 31; AJ vs . ER: Χ21 , 253 = 0 . 24 , p = 0 . 62; Double infection vs . uninfected: Χ21 , 638 = 0 . 002 , p = 0 . 99; Double infection vs . AJ: Χ21 , 81 = 0 . 15 , p = 0 . 70; Double infection vs . ER: Χ21 , 218 = 0 . 002 , p = 0 . 97; Fig 6; Table F in S2 Text ) , and so we conclude that while there was some evidence of clone differences in virulence , there was no evidence that double infections had a greater virulence to the mosquito than single infections ( Fig 6 ) . While this initially seems surprising , given that double infections contained more parasites than single infections , it is likely that all the densities within our experiment where low enough to not have a detectable impact on vector survival , particularly under laboratory conditions with ad libitum access to glucose and water [20–22 , 40] .
So far as we are aware , our experiments provide the first conclusive evidence that mosquitoes are capable of accumulating multiple infections over successive blood meals . We found that they are ( Fig 1 ) , and furthermore that the presence of parasites from a previous infection facilitates both the establishment and density of subsequent malaria parasite infections ( Fig 3 , Fig 5 ) without negatively impacting the replication of the primary infection ( Fig 2 ) or mosquito survival ( Fig 6 ) . Facilitation of establishment and density of secondary infections contrasts with the competitive suppression seen during mixed strain infections in the vertebrate host [9 , 41] . Previous studies have shown negative density dependence in the production of sporozoites by oocysts , presumably due to resource limitation or apparent competition mediated by the vector immune response [20] . However , parasites in our study are unlikely to have reached the threshold for negative density dependence to impact development ( estimated at ~200 oocysts [20] ) . It is possible that the facilitation we observed is because primary infection leads to structural changes in the mosquito midgut making it easier for a second infection to invade , and/or that the vector’s anti-parasite immune response may be depleted or suppressed by the primary infection , thereby leading to lower ookinete mortality . Another interesting possibility is that parasites respond to cues signalling the presence of another genotype and alter their replication schedules , as can apparently occur in vertebrate infections [9 , 42] . Changes in vector biting behaviour induced by the primary infection [36] , or trade-offs between the duration of oocyst development and sporozoite production , may mean that the fitness-maximizing intrinsic incubation period for malaria parasites is different for parasites sharing the vector with parasites from an existing infection . If this were the case , the higher genome counts from secondary infections could be due to parasites speeding up their replication when entering an already infected mosquito , in order to maximise representation in the salivary glands when the mosquito bites new hosts . Further experiments are required in order to determine how the within-vector environment changes with the establishment of a previous infection and why this increases the probability of a new infection and its density . A good first step would be to track the ookinetes invasion and establishment of oocysts , using fluorescently marked parasites within a previously infected mosquito , and therefore determine at which stage facilitation occurs . At first glance , our discovery that a primary malaria infection facilitates a subsequent infection contrasts with the finding by Rodrigues et al . that midgut bacteria introduced into the mosquito haemolympth by invading ookinetes prime the vector immune response , reducing the density of subsequent malaria parasite infections [43] . Several differences in experimental protocols may account for the apparent contradiction . For example , overall oocyst loads in our experiments were close to natural infection densities [27 , 44 , 45] and much lower than those of Rodrigues et al . ( mean ~5 oocysts per midgut in our single infections compared to means of ~15 & ~60 [43] ) . Perhaps a large number of ookinetes must cross the midgut to generate sufficient bacterial infection to prime a protective anti-Plasmodium effect . Alternatively , our challenge infections were four days after our primary infections . Rodrigues et al . [43] challenged their mosquitoes 7 and 14 days later; perhaps anti-malaria immunity elicited by bacterial invasion takes a week or more to develop . The elegant experimental protocols of Rodrigues et al . were not designed to look at direct interactions between the priming and challenge parasites because they induced early death of primary infections . Some combination of their protocols and ours would make possible the analysis of the outcomes of co-infections initiated further apart in time and at higher parasite densities . We concentrated on infections acquired from successive blood meals because mosquitoes rarely live long enough to transmit infections acquired two or more gonotrophic cycles after the first [35 , 46 , 47] . Combined , our results suggest that mosquitoes taking multiple infective bites will disproportionally contribute to onward malaria transmission of individual strains . How often mosquitoes would be expected to take multiple infective feeds in natural transmission settings depends on many other parameters ( e . g . biting rate , proportion of infectious hosts , vector survival ) . Using parameters from Killeen et al . [35] we estimate that without facilitation , ~10–41% of infectious vectors would have oocysts originating from more than one feeding cycle and ~8–17% of infectious mosquitoes would have salivary gland sporozoites originating from multiple blood meals ( S1 Text ) . These estimates are lower bounds; with facilitation these proportions could be much higher . They will be even higher if mosquitoes feed on multiple hosts within a gonadotrophic cycle [27–29] , if infected mosquitoes are more likely to blood feed [48] , and if infected hosts are more attractive to mosquitoes [49] , as has been recorded . Our data are in keeping with the observation that mixed species infections in the field appear to be higher in mosquitoes than would be expected from the single constitutive species prevalence’s , or from the prevalence of mixed infections in humans [4] . Additionally , accumulation of infections multiple feeds could partially explain the lower than expected rates of heterozygous oocysts observed in field studies of P . falciparum [45] ( as parasites from multiple feeds will not be able to mate ) . The controlled experiments reported here are not feasible in natural transmission settings as they require replicate infections in vertebrate hosts with known infection densities , matched time since infection ( to control for transmission blocking immunity ) and parasite strains which can be tracked by PCR through the mosquito . However , if mosquitoes in the field are accumulating multiple infections over the course of their lives , we predict that older mosquitoes would have a higher prevalence of mixed infections than younger mosquitoes [4] . With tools now available for determining infection diversity [25 , 45] and rapid estimation of age of field caught mosquitoes [50] , this can be tested . If the facilitation we have demonstrated here occurs in natural transmission settings , there could be significant epidemiological consequences . Control measures reducing prevalence in the vertebrate host , and therefore reducing the likelihood of mosquitoes taking multiple infective feeds , could disproportionally reduce transmission of individual strains – for example of drug resistant parasites . By increasing the proportion of infectious mosquitoes with mixed strain infections it is also likely that the facilitation reported here will increase the rates of mixed infections in vertebrate hosts which could have implications for infection virulence and the spread of resistant strains [1 , 51] . More generally , our results point to contrasting effects of mixed strain infections during the malaria lifecycle – while different parasite strains competitively suppress each other in the vertebrate host [6 , 9 , 52 , 53] , we have found that they facilitate each other in the mosquito . The potential epidemiological and evolutionary consequences of this antagonism and synergy could be investigated using mathematical models of malaria populations .
The two wild type Plasmodium chabaudi parasite strains ( AJ and ER ) used here were originally collected from thicket rats ( Thamnomys rutilans ) in the Congo [54] , maintained as part of the WHO Registry of Standard Malaria Parasites ( The University of Edinburgh ) before transportation to Penn State University where they are stored in liquid nitrogen . Mice in our experiments were 6–10 week old female C57Bl/6 kept on a 12:12 L:D cycle . The mice were fed on Laboratory Rodent Diet 5001 ( LabDiet; PMI Nutrition International , Brentwood , MO , USA ) and received 0 . 05% PABA-supplemented drinking water to enhance parasite growth [55] . Infections were established via intraperitoneal ( IP ) injection with 5x105 parasites . For each transmission , double the number of mice needed were infected 14 , 15 or 16 days prior to mosquito bloodmeal . On the day of transmission gametocytemia ( proportion of red blood cells containing gametocytes taken from thin blood smears ) and red blood cell density ( from 2 μL of blood examined by Flow Cytometry , Beckman Coulter Counter; see [56] ) was used to calculate the gametocyte density per μL of blood . The mice with infections containing the highest density of gametocytes were selected and anaesthetized with a 5μL IP injection of Ketamine ( 100 mg/kg ) and Xylazine ( 10 mg/kg ) and placed on top of individual mosquito cages for 30 minutes . One mouse was used per feed per cage ( experiment 1: 12 mice used for 6 cages; experiment 2: 42 mice used for 21 cages; see Table 1 for treatment groups ) . As each cage was fed on a different mouse , the density of transmission stages in the blood of each mouse was compared across treatment groups within each experiment , confirming that focal gametocyte densities did not significantly differ ( AJ in experiment 1: F1 , 4 = 2 . 22 , p = 0 . 21; AJ in experiment 2: F1 , 4 = 0 . 05 , p = 0 . 84; ER in experiment 2: F1 , 4 = 0 . 71 , P = 0 . 44; see Table A in S2 Text for gametocyte densities in each of the relevant pairwise comparisons ) . In order to maximise power without increasing the number of animals used , mosquitoes from the cages receiving two infective feeds were used to examine both the effect double infections on both the first and second infection to establish ( see Table 1 ) . Anopheles stephensi larvae were reared under standard insectary conditions at 26°C , 85% humidity and a 12L:12D photo-period . Eggs were placed in plastic trays ( 25 cm × 25 cm × 7 cm ) filled with 1 . 5 L of distilled water . To reduce variation in adult size at emergence , larvae were reared at a fixed density of 400 per tray . Larvae were fed on ground TetraFin fish flakes and from 10–11 days after egg hatch , pupae were collected daily and placed in emergence cages . The adults that emerged were fed ad libitum on a 10% glucose solution supplemented with 0 . 05% paraaminobenzoic acid ( PABA ) . Adult female mosquitoes between 3 and 5 days old were equally distributed across all experimental cages with 100–120 female mosquitoes per cage . Experimental cages were given Ad lib access to 10% glucose solution supplemented with 0 . 05% paraaminobenzoic acid ( PABA ) apart from in the 24 hours prior to feeding on mice where they were deprived of glucose to increase propensity to blood feed . After both blood-feeds , any visibly unfed females were removed and discarded and mosquitoes were provided with bowls for oviposition . Sample sizes in Table 1 reflect the number of mosquitoes that took full bloodmeals on both occasions they were offered a host . In order to ensure densities were comparable our focal infections were always assessed after 7 days . This means that when we were testing for an impact on the first infection mosquitoes were dissected at experimental day 7 and when we were testing for an impact on the second infection mosquitoes were dissected at experimental day 11 ( 7 days after the second bloodmeal on experimental day 4 ) . To determine infection status and density ~30 mosquitoes per cage were removed , killed with chloroform and dissected . Midguts were examined for oocyst presence and intensity and infected guts were then placed individually into 30 μL of chilled PBS within 1 . 5 mL microtubes . Tubes were maintained on ice prior to storage at -80°C . DNA was extracted from individual mosquito midguts using the E . Z . N . A MicroElute Genomic DNA kit ( Omega Bio-Tek ) as per manufacturer’s instructions , eluted in a total volume 20 μL and stored at -80°C . Clone specific genome numbers were determined by PCR following the methods in [57] . Cages were checked for dead mosquitoes twice daily until all mosquitoes had died ( 72 days after receiving their first blood meal ) . Mosquito cadavers were stored individually in 1 . 5mL microtubes and immediately frozen at -20°C for short-term storage before being moved to -80°C within two weeks . Parasite DNA was extracted for the mosquito cadavers and the presence and genome count for each strain was quantified using the same methodology as for dissected midguts except for the addition of 2 . 5μL of BSA per reaction well prior to PCR analysis ( 10mg/mL Bovine Serum Albumin , New England BioLabs Inc . ) . BSA was used as pigment found in the eyes of insects has previously been shown to inhibit DNA amplification [58] . A pilot study confirmed previous studies [59] , showing BSA was successful at preventing this inhibition . Infection prevalence in dead mosquitoes from each cage strongly correlated with prevalence from dissected mosquitoes confirming our ability to reliably detect parasite infection through this method ( R2 = 0 . 99 for AJ; R2 = 0 . 96 for ER prevalence; R2 = 0 . 95 for the mean number of strains per mosquito; S3 Fig ) . All analysis was performed using R version 3 . 0 . 2 ( R core team ( 2013 ) http://www . R-project . org ) . Gametocyte densities in the mice used for transmission were calculated by multiplying the gametocytemia by the red blood cell density and were log10 transformed and analyzed using general linear models . The proportion of mosquitoes infected with the focal strain for each group was analyzed using generalized mixed effect models ( glmer ) with a binomial error structure and cage fitted as a random effect ( lme4 . R package version 1 . 0–6 ) . For analysis of infection density within the mosquito , only infected mosquitoes were included and host gametocyte density was fitted as a random effect in models . Oocyst densities were analysed using glmer with a poisson error structure and sporozoite densities were log10 transformed and analysed using lmer models . Survival analysis was performed using Cox proportional hazard mixed effect models ( Terry Therneau ( 2012 ) coxme: Mixed Effects Cox Models . R package version 2 . 2–3 ) with experimental cage fitted as a random effect and infection status , estimated total red blood cells in bloodmeals and the mean oocyst density from mosquitoes dissected from the same cage fitted as fixed effects . Total red blood cell density in bloodmeals was estimated from red blood cell densities in the two mice each cage fed on ( one per feed ) and was included to account for any variation in the quality of bloodmeals received . For all analyses we followed model simplification by sequentially dropping the least significant term and comparing the change in deviance with and without the term to Chi-square distributions until the minimum adequate model was reached . Full details of statistical models can be found in S2 Text and data are deposited in the Dryad repository: ( doi:10 . 5061/dryad . 8nr13 ) [60] . This study was carried out in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health . The protocol was approved by the Animal Care and Use Committee of the Pennsylvania State University ( Permit Number: 35790 ) .
|
Very little is known about how malaria parasite strains interact with each other inside mosquitoes . In this study we show that mosquitoes that have already been infected with one strain of malaria parasites are more likely to become infected with a new strain . Moreover , the presence of an existing infection enhances the replication of malaria parasites with no obvious impact on mosquito survival . Our results illustrate that interactions between strains are important factors in parasite survival and transmission across the whole of their life cycle .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[] |
2015
|
Existing Infection Facilitates Establishment and Density of Malaria Parasites in Their Mosquito Vector
|
Insects rely primarily on innate immune responses to fight pathogens . In Drosophila , antimicrobial peptides are key contributors to host defense . Antimicrobial peptide gene expression is regulated by the IMD and Toll pathways . Bacterial peptidoglycans trigger these pathways , through recognition by peptidoglycan recognition proteins ( PGRPs ) . DAP-type peptidoglycan triggers the IMD pathway via PGRP-LC and PGRP-LE , while lysine-type peptidoglycan is an agonist for the Toll pathway through PGRP-SA and PGRP-SD . Recent work has shown that the intensity and duration of the immune responses initiating with these receptors is tightly regulated at multiple levels , by a series of negative regulators . Through two-hybrid screening with PGRP-LC , we identified Rudra , a new regulator of the IMD pathway , and demonstrate that it is a critical feedback inhibitor of peptidoglycan receptor signaling . Following stimulation of the IMD pathway , rudra expression was rapidly induced . In cells , RNAi targeting of rudra caused a marked up-regulation of antimicrobial peptide gene expression . rudra mutant flies also hyper-activated antimicrobial peptide genes and were more resistant to infection with the insect pathogen Erwinia carotovora carotovora . Molecularly , Rudra was found to bind and interfere with both PGRP-LC and PGRP-LE , disrupting their signaling complex . These results show that Rudra is a critical component in a negative feedback loop , whereby immune-induced gene expression rapidly produces a potent inhibitor that binds and inhibits pattern recognition receptors .
Insects rely primarily on innate immune responses to fight pathogens . The Drosophila immune response has proven to be an experimentally powerful and conserved model system for the study of innate immunity [1] , [2] , [3] , [4] . In particular , the insect immune response relies on evolutionary conserved NF-κB signaling cascades for the control of inducible antimicrobial peptide ( AMP ) gene transcription . This antimicrobial peptide response is critical for protection against many microbial pathogens [5] , [6] . In Drosophila , two signaling pathways regulate the production of these antimicrobial peptides - the IMD and Toll pathways [7] . The Toll pathway responds to many Gram-positive bacterial and fungal infections [8] , while the IMD pathway is potently activated by DAP-type peptidoglycan ( PGN ) from Gram-negative bacteria and certain Gram-positive bacteria [9] , [10] . Two receptors , PGRP-LC and PGRP-LE , are able to recognize DAP-type PGN at the cell surface or in the cytosol , respectively , and trigger the IMD pathway [11] , [12] , [13] , [14] , [15] , [16] . Upon binding DAP-type PGN , both PGRP-LC and PGRP-LE multimerize and signal via a common motif in their N-terminal domains , known as the RHIM-like domain [15] , [17] , [18] . The RHIM-like domain is critical for signaling by either receptor , but the mechanism ( s ) involved remain unclear [15] . Genetic experiments suggest that the imd protein functions immediately downstream of PGRP-LC and upstream of all other known components of the pathway [19] . IMD associates with both PGRP-LC and -LE , although the PGRP-LC RHIM-like motif is not required for this interaction [15] . Nonetheless , the complexes formed on these receptors are likely to be critical to trigger further signal transduction . Recent work has shown that the intensity and duration of the immune response is tightly regulated in Drosophila . As in mammals , over-exuberant immune responses can be detrimental , and the proper down modulation of immunity is critical for health and fecundity [20] , [21] , [22] . In order to keep the immune response properly modulated , the Toll and IMD pathways are controlled at multiple levels by a series of negative regulators . For example , the amidases PGRP-LB and PGRP-SC reduce the immunostimulatory activity of PGN by digesting it [23] , [24] . Intracellularly , the IMD signaling pathway is further down–regulated by Dnr1 , POSH , Caspar and the E3-ligase complex containing SkpA , dCullin and Slimb [25] , [26] , [27] , [28] . Additionally , the JNK and Relish branches of the IMD pathway are thought to mutually inhibit each other [29] , [30] , [31] . In this study , we identify and characterize a negative feedback regulator of the IMD pathway , dubbed rudra . Expression of rudra was rapidly induced following immune challenge . Moreover , in flies and cells , rudra is critical for controlling immune-induced gene expression . Following infection , rudra mutant flies hyper-activated antimicrobial peptide gene expression resulting in increased resistance to microbial infection . Using various biochemical and genetic techniques , Rudra was found to interact with the receptors PGRP-LC and PGRP-LE and disrupt the signaling complex assembled on these receptors . Due to its ability to destroy this receptor signaling complex and inhibit immune responses , rudra was named for Shiva , the Indian god of destruction , who in his Rudra phase of mind causes inhibition and destruction of all life on earth .
In order to identify potential partners and regulators of the IMD pathway receptors , a yeast two-hybrid screen was performed with the cytoplasmic domain of PGRP-LC as bait [32] , [33] . 25 strongly interacting clones were further analyzed with a set of baits that carried mutations in the RHIM-like domain of PGRP-LC ( or irrelevant control baits ) . One clone interacted strongly with the wild-type cytoplasmic domain of PGRP-LC but weakly with the RHIM-like mutant baits ( Table 1 ) . This clone encoded amino acids 30–197 of CG15678 , and will be referred to as rudra from hereafter . To confirm the yeast two-hybrid results , co-immunoprecipitation experiments were performed . Using epitope tagged constructs and transient transfection in Drosophila S2* cells , both PGRP-LE and PGRP-LC were found to associate with Rudra ( Figure 1A , E ) . In a heterologous system ( HEK cells ) , similar robust associations were observed between Rudra and PGRP-LE or −LC ( Figure 1B , C ) . The interaction between Rudra and PGRP-LE was also readily detectable , by co-immunoprecipitation , when these proteins were produced in a rabbit reticulocyte in vitro translation system ( Figure S1 ) . These data demonstrate that Rudra interacts directly with the receptors PGRP-LC and PGRP-LE . In order to determine which domain ( s ) of the receptors interact with Rudra , co-immunoprecipitation assays were performed with various mutant versions of PGRP-LC or PGRP-LE . Consistent with the yeast two-hybrid data , which indicated involvement of the RHIM-like domain for interaction , a mutant form of PGRP-LE lacking the RHIM motif ( Δ98-113 ) showed little interaction with Rudra ( Figure 1A , B ) . Using a set of large deletions ( Figure 1D ) , the N-terminal cytoplasmic domain of PGRP-LC was found to be essential for association with Rudra . Removal of the first 144 amino acids decreased Rudra interaction , while removal of nearly the entire cytoplasmic ( Δ1-253 ) domain abolished interaction . The PGRP-LC extracellular domain was not involved in the interaction ( Figure 1E ) . We then attempted to map the PGRP-LC interaction more finely with a set of mutants that span the entire cytoplasmic domain with sequential 50 amino acid deletions . However , Rudra co-immunoprecipitated with all of these deletion mutants , suggesting some redundancy in the interaction mechanism ( Figure S2 ) . The yeast two-hybrid data suggest that some of the interacting activity involves the PGRP-LC RHIM domain , while the larger deletions suggest another interaction motif likely lies in the first 144 amino acids ( Figure 1D , E ) . Overall , we conclude that Rudra directly interacts with the signaling domains of PGRP-LC and PGRP-LE . The interaction with PGRP-LE is largely mediated by the RHIM motif while the interaction with PGRP-LC appears to involve multiple , partly redundant , mechanisms . Previous microarray studies have suggested that rudra is a target of the IMD signaling pathway [29] , [34] , [35] . In order to confirm and extend these findings , the expression of rudra was analyzed at various times after immune stimulation of S2* cells , by qRT-PCR . rudra transcript was rapidly induced , peaking in 30–60 minutes and returning to near baseline levels within 24 hours ( Figure 2A ) . The kinetics of rudra expression were markedly faster and more transient than the expression of AMP genes . For example , Diptericin mRNA levels , as measured by Northern blotting , did not peak until 6 hours after PGN stimulation , and then remained elevated for at least 24 hours ( Figure 2A ) . Even though the expression profiles of rudra and AMP genes are distinct , they both require the NF-κB factor Relish [35] , [36] . Next , RNAi was used to characterize the function of rudra in the IMD pathway . S2* cells were transfected with dsRNA for rudra , and then stimulated with PGN for various times . As monitored by Northern blotting , antimicrobial peptide genes Diptericin ( Dpt ) , Attacin ( Att ) and Cecropin ( Cec ) were induced to markedly higher levels in cells treated with rudra RNAi , compared to cells transfected with a control lacZ dsRNA ( Figure 2B ) . These data suggest that rudra is a negative regulator of IMD signaling . To further test if rudra is a negative regulator of the IMD pathway , stable cell lines expressing rudra from a copper-inducible promoter were selected . These cell lines were treated with copper for 1 . 5 hours , to induce rudra expression , and then stimulated with PGN for 5 hours , to stimulate the IMD pathway . rudra over-expression potently inhibited the induction of Dpt ( Figure 3A ) . Also , to test if rudra negatively regulates the Toll pathway , stable cell lines expressing rudra from the actin promoter were selected . These cell lines were treated with SPZ-C106 for 18 hours to stimulate the Toll pathway . rudra over-expression did not robustly inhibit the induction of Drosomycin , as compared to its ability to inhibit PGN-induced Diptericin expression ( Figure S3 ) . These data demonstrate that rudra is potent inhibitor of the IMD pathway but has little effect on Toll signaling . Using the UAS system and a heat shock Gal4 ‘driver’ , transgenic flies that ectopically express rudra were also characterized . rudra expression was induced with a 1 . 5 hour heat shock and then flies were challenged with E . coli . In two independent UAS-rudra lines , IMD signaling was strongly inhibited by rudra expression , as monitored by Northern blotting for Dpt induction ( Figure 3A ) . These results are consistent with the data from cultured cells , and argue that rudra is a potent negative regulator of the IMD pathway in vivo . In order to phenotypically characterize the loss of rudra , a strain carrying a P-element at position 123 in the 5′ UTR of rudra ( EY00723 ) was analyzed [37] , [38] , [39] . First , the level of rudra transcript in this strain was compared to an isogenic white strain , by qRT-PCR ( Figure 4A ) . [To isogenize mutant and wild-type strains , EY00723 was backcrossed with the white strain for six generations prior to these analyses] . Similar to the cell culture data , rudra transcription was rapidly induced following infection in wild-type flies . Again , the induction of rudra expression occurs more rapidly , and is resolved more quickly , than does AMP gene expression ( compare Figure 4A to 4B ) . The transposon insertion in the 5′ UTR markedly inhibited rudra expression , with nearly undetectable levels at all time points , demonstrating that this allele of rudra is a strong hypomorph . Also , a transgenic rescue strain was constructed , using a 4 . 5 Kbp genomic fragment ( rudrarescue ) . This genomic rescue construct partially restored immune-inducible expression of rudra , but it did not completely return to wild-type levels ( Figure 4A ) . Next , the immune response of wild-type , rudraEY00723 , and the rudrarescue strains were compared . Diptericin expression , as monitored by Northern blotting at various times following septic E . coli infection , was elevated at all time points in rudraEY00723 compared to the isogenic wild-type strain ( Figure 4B ) . The rudrarescue transgenic line restored Diptericin to levels between that observed in the wild-type and rudra mutant flies , consistent with partially restored levels of rudra expression observed in this line . rudra heterozygotes also displayed elevated AMP gene expression ( data not shown ) . These results , together with the data from ectopic expression , demonstrate that rudra is a potent negative regulator of the IMD pathway in flies , as well as in cultured cell lines . We then asked what consequence these elevated AMP levels might have during an infection . To this end , wild-type and rudraEY00723 flies were infected with the Gram-negative pathogen Erwinia carotovora carotovora ( Ecc ) . As reported previously , Ecc is a mildly pathogenic infection in wild-type animals , such that most flies succumb over the course ∼10 days ( Figure 4C ) [27] , [40] . As expected , PGRP-LE; PGRP-LC double mutant flies , which lack both receptors involved in detecting DAP-type PGN , were rapidly killed by this infection ( P = 0 . 0252 , compared to wild-type animals ) . On the other hand , rudra mutants showed significantly improved survival compared to wild-type flies ( P = 0 . 0052 ) . These results show that loss of rudra , and the ensuing increase in AMP levels , enhances resistance to this Gram-negative pathogen . We next sought to determine the molecular mechanism ( s ) used by Rudra to control signal transduction . Relish , the NF-κB precursor protein essential for IMD triggered gene expression , is regulated by immune-induced cleavage and phosphorylation ( [41] , [42] , unpublished data D . E-H . and N . S ) . Rudra expression prevented both the cleavage and phosphorylation of Relish ( Figure 5A ) . Recently , we also discovered that imd protein is rapidly cleaved following immune stimulation ( unpublished data , N . P . and N . S ) and expression of rudra potently inhibits this cleavage ( Figure 5A ) . These results suggest that Rudra functions upstream of Relish activation and IMD cleavage . AMP gene expression can be triggered by ectopically expressing certain components of the IMD pathway . In particular , over-expression of either of the receptors , PGRP-LC or PGRP-LE , or imd is sufficient to drive AMP gene expression . Likewise , over-expression of the caspase Dredd is sufficient to drive Relish cleavage . To further analyze the position that Rudra acts in the IMD pathway , it was over-expressed with these signaling components in doubly selected stable cell lines . Rudra potently inhibited signaling induced by over-expression of the receptors PGRP-LC or PGRP-LE , but had no effect on the induction of Diptericin expression caused by IMD over-expression ( Figure 5B ) . Likewise , Rudra did not inhibit Relish cleavage caused by over-expressing the caspase Dredd ( Figure 5C ) . These results suggest that Rudra functions upstream of Dredd and IMD , but downstream of the receptors , and is consistent with binding data demonstrating an association between Rudra and either PGRP-LC or PGRP-LE . In addition to interacting with the receptors , Rudra avidly bound to IMD . The IMD association was detected by transient transfection/co-immunoprecipitation assays , in either S2* cells ( data now shown ) or HEK cells ( Figure 6A ) . On the other hand , Rudra did not associate with dFADD , another factor known to interact with IMD . In all , these data argue that Rudra directly interacts with both IMD and the receptors PGRP-LC and PGRP-LE . These results suggest two possible models for the inhibition of IMD signaling by Rudra: ( 1 ) Rudra may associate with both the receptor and its signaling adaptor ( IMD ) , holding them together in an inactive confirmation; or ( 2 ) Rudra may interact with both PGRP-LC and IMD separately , disrupting the association between the receptor and its adaptor . To probe these possibilities , co-immunoprecipitation experiments were performed with lysates from cells co-transfected with PGRP-LC ( T7 tag ) , imd ( FLAG tagged ) and/or rudra ( also FLAG tagged ) . In assays with just the receptor and either IMD or Rudra , PGRP-LC interacted with either the adaptor or the inhibitor , in both Drosophila and human cells ( Figure 6B , C ) . However , when all three proteins were simultaneously co-expressed , PGRP-LC and Rudra still robustly co-precipitated , but the association between IMD and the receptor was markedly reduced . These data suggest that Rudra interferes with the interaction between PGRP-LC and IMD , and this disruption provides a molecular mechanism explaining how Rudra down-modulates IMD signaling at the level of the receptor , consistent with the functional and binding data presented .
Recent work has shown that the intensity and duration of the immune response is tightly regulated in Drosophila [23] , [24] , [25] , [27] , [28] . Over-exuberant immune responses can be dangerous and the proper down modulation of immunity is important for health and fecundity [20] , [22] . To keep the immune response properly modulated , the Toll and IMD pathways are controlled at multiple levels by multiple negative regulators . In this study , we have characterized a new negative feedback regulator of the IMD pathway . rudra transcript is rapidly induced following septic infection , and rudra mutant flies or rudra knockdown cells over-express antimicrobial peptides . In the case of Erwinia carotovora carotovora infection , this elevated level of AMP production leads to increased survival . A similar phenotype was reported for mutants lacking Caspar , which is thought to inhibit downstream signaling events [27] . The results presented here , in cells and flies , demonstrate that rudra is a key component in a negative feedback loop that keeps the IMD pathway in check . In addition to these loss-of-function results , over-expression of rudra potently blocked signaling through the IMD pathway , both in cells and in flies . Moreover , we exploited this activity to analyze which steps in the IMD pathway are inhibited by Rudra . Using various molecular assays to monitor different PGN-induced events in the IMD pathway , we found that Rudra interfered with cleavage of IMD . Signaling mediated by receptor over-expression was also inhibited by Rudra , but this was not the case for signaling induced by over-expression of downstream components . Together , these data strongly support the notion that Rudra interferes with receptor function and is consistent with the association between Rudra and the receptors PGRP-LC or PGRP-LE . Using assays in yeast , Drosophila , human cells and in vitro , Rudra was shown to interact directly with PGRP-LC and PGRP-LE . The interaction between PGRP-LE and Rudra required the RHIM-like domain of PGRP-LE , which is also critical for signaling by this receptor . However , the region through which PGRP-LC interacts with Rudra is less clear and likely involves multiple , partly redundant interfaces . Rudra also interacted with the imd protein . Moreover , Rudra interfered with the interaction between the receptor PGRP-LC and IMD , destabilizing the receptor signaling complex . From these results , we propose that Rudra is a negative feedback regulator that down modulates the IMD pathway by binding the receptors and interrupting the associations with their cognate signaling adaptor IMD . This regulatory loop is critical to properly regulate the immune response . In agreement with the data presented here , Kleino et al . ( 2008 ) recently reported that rudra/CG15678 is a negative regulator of the IMD pathway , although they refer to this gene as poor Imd response upon knock-in ( pirk ) . They showed that rudra/pirk is rapidly induced following infection , similar to the data presented here , and further demonstrated that rudra induction is dependent on Relish , both in cells and in flies . Using reporter assays in S2 cells , they found that Pirk inhibits IMD signaling but not the Toll pathway . With transgenic RNAi fly lines , they also found that knockdown of pirk caused the hyper-expression of the antimicrobial peptide genes . Also , flies over-expressing Pirk blocked the activation of the IMD pathway and were more susceptible infection . These results are consistent with the data presented here , although we have characterized a mutant allele of rudra and additionally show that this mutant exhibits enhanced protection against Erwinia infection . The data presented here also expand on the findings of Kleino et al . ( 2008 ) by showing that Rudra not only interacts with both PGRP-LC and IMD , but also that these interactions with Rudra disrupt the direct association between PGRP-LC and IMD . Kleino et al . ( 2008 ) reported that central portion of Rudra consists of two repetitive amino acid elements of unknown function and structure , which they named the Pirk domain . The Pirk domain is required for the interaction with IMD , but not with PGRP-LC . Rudra does not contain obvious homology to any other protein motifs , and no mammalian homologs are readily detected . [36] . Recently , multiple mechanisms involved in regulating the Drosophila immune response have come to light . Given that it is well-established that immune activation in flies has a cost , such as reduced fecundity [20] , [22] and hypersensitivity to infection [23] , [24] , [27] , [43] , [44] , it is not surprising that multiple negative regulatory circuits control the immune response . Similarly , in mammals , innate and adaptive immune responses are held in check by multiple mechanisms , in order to prevent inflammatory and autoimmune diseases while at the same time allowing an effective response to infection . Future studies will address the possible negative consequences of the lack of proper IMD regulation observed in the rudra mutant animals .
Insoluble PGN from E . coli was purchased from Invivogen . rudra mutant line , EY00723 , was originally isolated by the Drosophila Genome Project gene disruption consortium and provided by the Bloomington Drosophila Stock Center . The flies were backcrossed for six generations to a w1118 strain in order to isogenize . In all experiments , rudraEY00723 mutants were compared to isogenic w1118 animals . PGRP-LE112;;PGRP-LCΔE , double mutant flies were reported previously [45] . Survival experiments were performed with 60 flies at 29°C , following infection by pricking in the abdomen with a microsurgery needle dipped into a concentrated pellet of Erwinia carotovora carotovora 15 [24] . Surviving flies were transferred to fresh vials and counted daily , until all wild-type flies died . Kaplan-Meier plots are presented and P-values were calculated by log-rank test using GraphPad Sigma Plot . Total RNA from flies or cultured cells was isolated with the TRIzol reagent ( Invitrogen ) as described previously [33] . Expression of Diptericin , Attacin , Cecropin and the control rp49 ( ribosomal protein ) was analyzed by Northern blotting [33] . Northern blots were quantified with a phosphoimager ( Fuji ) and AMP gene expression was normalized to rp49 levels . For qRT-PCR , RNA was DNase treated and re-extracted with phenol-chloroform . cDNA was synthesized using Superscript II ( Invitrogen ) and quantitative PCR analysis was performed on a DNA engine Opticon 2 cycler ( MJ Research , Watertown MA ) using SYBR Green ( Biorad ) . The specificity of amplification was assessed for each sample by melting curve analysis and relative quantification was performed using a standard curve with dilutions of a standard . The quantified data was normalized to rp49 levels . In all S2*-based cell experiments , cells were pre-treated with 1 mM 20-hydroxyecdysone for 24 to 40 hr before treatment with 500 mM CuSO4 and/ or PGN ( 100 ng/ml ) . dsRNA was generated and purified as reported previously [46] . Cells were split 24 hours after transfection to 1 . 0×106/mL and then were treated with 1 mM 20-hydroxyecdysone . After 24 hours , cells were treated ( or left untreated ) with PGN ( 100 ng/ml ) for various time , as indicated . In vitro translation was performed following the protocol of the manufacturer ( Promega ) . Immunoprecipitations were carried out with rabbit anti-T7 ( Bethyl labs ) in lysis buffer ( 20 mM Tris at pH 7 . 6 , 150 mM NaCl , 2 mM EDTA , 10% Glycerol , 1% Triton X-100 , 1 mM DTT , NaVO4 , glycerol 2-phosphate and protease inhibitors ) . For immunoprecipitation from cells , Schneider S2* cells were first transfected by calcium phosphate method with appropriate expression plasmids . Cells were split 24 hours after transfection to 1 . 0×106/mL and 24 hours later , were treated with 500 µM copper sulphate for 5 hr , when necessary , for expression from the metallothionein promoter . Immunoprecipitations were performed in lysis buffer and analyzed by SDS-PAGE followed by immunoblot analysis with anti-T7 MAb ( Novagen ) , anti-V5 ( Sigma ) , anti-IMD ( gift of J . -M . Reichhardt ) or anti-Flag ( Sigma ) antibodies . Stable cell lines and immunoblotting were performed as described previously [33] . The generation and characterization of phospho-specific Relish antibody will be detailed elsewhere ( D . E . -H . and N . S . , unpublished data ) . For the UAS transgenic , the rudra ORF was amplified by PCR and subcloned into the EcoRI and BglII sites of pUAST . For genomic rescue , a BAC clone ( Drosophila Resource Center [47] ) was used as a template to amplify a 4 . 5 Kbp genomic fragment containing the complete rudra locus plus flanking sequences , which was then cloned into the EcoRI and BamHI sites of pCaSpeR [48] . After sequence verification , standard techniques were used for P-element–mediated transformation at the MGH Drosophila transgenics facility . For immune stimulation assays , adults ( males and females in equal numbers ) , were infected by pricking in the abdomen with a microsurgery needle dipped into a concentrated pellet of E . coli ( 1106 ) , RNA was extracted 8 h later , and assayed by Northern blotting . The rudra gene was cloned into pRmHa3 vector by standard methods to create constructs expressed from the metallothionein promoter . The constructs were then transfected into S2* cells in conjunction with pHs-Neo at a ratio of 50∶1; stable transfectants were then selected with G418 ( 1 mg/ml ) . For double stable cell lines , the rudra expression plasmid was transfected into S2* cell lines that were previously selected to carry plasmids expressing either PGRP-LC , PGRP-LE , IMD or Dredd . The rudra plasmid was selected with a second selectable marker , either G418 ( 1 mg/ml ) or hygromycin ( 20 U/ml ) , as appropriate .
|
The innate immune system controls the immediate response to infection . Innate immunity relies on germline encoded receptors , receptors that are present at birth , to recognize germs and trigger a protective response . Invertebrates ( i . e . , insects ) rely on innate immunity to survive in microbial-rich environments , such as rotting fruit . However , uncontrolled innate immune responses are dangerous , leading to severe pathologies like sepsis , inflammatory bowel diseases , and lupus . Therefore , the intensity and duration of the innate immune response is kept in-check by multiple regulatory mechanisms . Here , we have identified a new feedback regulator of the Drosophila ( the fruit fly ) immune response , which we call Rudra . Using various approaches , we show that in the absence of Rudra the innate immune system is hyper-activated . This elevated immune response leads to better protection against bacterial infection . On the other hand , when present in excess , Rudra prevents the activation of the immune response . Furthermore , we show that Rudra turns off the immune response by binding to the receptors that are responsible for detecting bacteria , thereby preventing downstream responses .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"infectious",
"diseases/bacterial",
"infections",
"immunology/genetics",
"of",
"the",
"immune",
"system",
"immunology/immunity",
"to",
"infections",
"immunology/innate",
"immunity"
] |
2008
|
Rudra Interrupts Receptor Signaling Complexes to Negatively Regulate the IMD Pathway
|
Most species are superbly and intricately adapted to the environments in which they live . Adaptive evolution by natural selection is the primary force shaping biological diversity . Differences between closely related species in ecologically selected characters such as habitat preference , reproductive timing , courtship behavior , or pollinator attraction may prevent interbreeding in nature , causing reproductive isolation . But does ecological adaptation cause reproductive incompatibilities such as hybrid sterility or lethality ? Although several genes causing hybrid incompatibilities have been identified , there is intense debate over whether the genes that contribute to ecological adaptations also cause hybrid incompatibilities . Thirty years ago , a genetic study of local adaptation to copper mine soils in the wildflower Mimulus guttatus identified a locus that appeared to cause copper tolerance and hybrid lethality in crosses to other populations . But do copper tolerance and hybrid lethality have the same molecular genetic basis ? Here we show , using high-resolution genome mapping , that copper tolerance and hybrid lethality are not caused by the same gene but are in fact separately controlled by two tightly linked loci . We further show that selection on the copper tolerance locus indirectly caused the hybrid incompatibility allele to go to high frequency in the copper mine population because of hitchhiking . Our results provide a new twist on Darwin's original supposition that hybrid incompatibilities evolve as an incidental by-product of ordinary adaptation to the environment .
Adaptation to local environmental conditions by natural selection is the primary cause of evolutionary change in natural populations . Ecological adaptation can cause reproductive isolation when selection acts on traits that influence the likelihood of intermating in nature , such as habitat preference , reproductive timing , courtship behavior , or pollinator attraction [1] , [2] . However , it is controversial whether adaptation to local environmental conditions promotes the development of reproductive incompatibilities such as hybrid sterility or lethality . As Darwin [3] first discussed , reproductive incompatibility should not evolve directly via natural selection ( but see [4] ) and would only arise as an incidental by-product of interspecific divergence . Darwin realized that reproductive isolation could evolve if it is not expressed as lineages diverge , perhaps due to populations inhabiting geographically distinct regions , and it would only manifest in cases in which these populations hybridized subsequent to divergence . Classic models of the evolution of hybrid incompatibility , independently developed by Bateson [5] , Dobzhansky [6] , and Muller [7] ( BDM ) , predict that alleles at different loci may accumulate within distinct lineages , and although they may be neutral or adaptive in an ancestral population , they will produce deleterious interactions when brought together by hybridization . This model is now thoroughly supported by genetic mapping studies of hybrid incompatibility loci . The role of natural selection in driving the evolution of incompatibility alleles continues to be an area of rich investigation [8]–[10] . Selection may drive the evolution of BDM incompatibility alleles in a variety of ways . With the molecular genetic dissection of multiple hybrid incompatibility systems , researchers have determined that many of the underlying loci have experienced rapid evolutionary divergence consistent with natural selection [9] , [10] , but there is little evidence of what may be causing this rapid divergence . Evidence for the traditional view that hybrid incompatibility evolves as a by-product of adaptation to divergent environments is supported by only a few lab-based experimental evolution studies [11]–[13] . Of the hybrid incompatibility genes identified in natural populations , it seems unlikely they function in classical ecological adaptation . Instead researchers have speculated that these incompatibilities evolved due to intragenomic conflict [14] , [15] . For example , the incompatibility Overdrive locus causes both segregation distortion and hybrid male sterility in Drosophila psuedoobscura [16] . Ultimately , it is considerably difficult to determine the evolutionary forces acting upon hybrid incompatibility alleles during the process of divergence because these studies focus on species reproductively isolated for thousands of generations . In this study , we investigate whether recent adaptation to a copper mine habitat by the wild flower Mimulus guttatus caused the evolution of hybrid incompatibility . Extreme soil environments , either naturally occurring or created anthropogenically , characterized by low nutrient and high heavy-metal concentrations , impose strong selection on plant communities [17] , [18] . M . guttatus has colonized multiple mine sites in the Sierra Nevada foothills region of central California , the largest of which is located at the Keystone Union mine complex in the town of Copperopolis [19] . A survey of tolerance to high copper concentration in plants from Copperopolis and adjacent off-mine populations suggests this phenotype is under strong selection in the mine habitat . Copper tolerance is nearly fixed in the mine population ( 99 . 77% , N = 2 , 796 ) and segregating at very low frequency in a sample of 15 off-mine populations located with 40 km of Copperopolis ( 8 . 7% , N = 1 , 440 ) [20] . An investigation into the genetic basis of this trait initiated over 30 years ago identified a locus that appeared to cause both copper tolerance and hybrid lethality in crosses to other off-mine populations [21] , [22] . Copper tolerance is controlled by a dominant allele at a single Mendelian locus , when measured as a threshold character in lab-based root growth assays . Crosses between tolerant lines from a recombinant backcross population ( Figure S1 ) and the off-mine population , Cerig-y-drudion , Wales , United Kingdom ( hereafter referred to as Cerig ) , yielded various levels of F1 hybrid lethality , whereas crosses between nontolerant lines from the same recombinant backcross population always produced viable offspring [22] . Additional crosses demonstrated there is a single incompatibility locus in the Copperopolis population and that variation in F1 hybrid lethality is caused by multiple loci segregating in the Cerig population [22] . Cerig was used in the initial crosses by M . R . Macnair and continues to be a focal population in this study , because it was the first population identified to produce inviable hybrid offspring in crosses with Copperopolis [22] . Subsequent experiments have demonstrated that the Cerig hybrid lethality factors are geographically widespread and segregate in three California populations . In contrast , the Copperopolis lethality factor is geographically restricted; no incompatible plants were identified in test crosses with plants from 15 other populations ( N = 10 plants/population ) [23] . The lethality factor is at high frequency in the Copperopolis population; 16 Copperopolis genotypes ( five of which are from this study , see details below ) have been tested and all of them produce lethal offspring when crossed to Cerig tester lines [22] , [23] . Hybrid lethality in this system is consistent with the BDM model and was not observed to be genetically de-coupled from the copper tolerance phenotype [22] , [24] . One attractive hypothesis is that they are controlled by the same locus and F1 hybrid lethality evolved as a pleiotropic by-product of the locally adaptive copper tolerance . An alternative hypothesis is that the two phenotypes are not pleiotropic and are instead controlled by two distinct loci in tight genetic linkage . The physiological manifestation of BDM hybrid incompatibility in this system reveals possible genetic mechanism for inviability , while maintaining support for both the pleiotropy and linkage hypotheses . Hybrid lethality manifests in F1 plants as tissue yellowing and death in the early stages of development and prevents the plants from reaching reproductive maturity [22] , [24] . This lethality is similar to the hybrid necrosis phenotype known for many years in the plant agriculture literature [25] . Recently , it has been demonstrated that hybrid necrosis is an autoimmune-like response due to negative interactions between divergent pathogen-resistance proteins , often involving nucleotide-binding leucine-rich repeats ( NB-LRRs ) and their interacting partners [26] , [27] . Although there is no support for NB-LRR's dual function in disease resistance and trafficking of heavy-metal ions , this could be the case for their partner proteins . NB-LRR protein–protein interactions are difficult to examine , and most partners remain unknown [28] . Additionally , there is strong evidence of pleiotropy between disease resistance and heavy metal tolerance from an independent class of pathogen-resistance proteins , defensins , which contribute to enhanced zinc tolerance in Arabidopsis halleri [29] . It remains to be determined whether selection on the copper tolerance allele in M . guttatus resulted in the near fixation of a hybrid incompatibility factor because these phenotypes are the result of pleiotropy at a single locus . The primary aim of this study is to investigate the genetic basis of the copper tolerance and hybrid necrosis phenotypes to determine whether they are pleiotropic or controlled by distinct genes in tight genetic linkage . We also use population genetic approaches to infer the nature of natural selection acting on this copper tolerance and hybrid necrosis genomic region .
To determine whether copper tolerance and hybrid lethality are controlled by the same locus , we conducted a high-resolution genetic mapping experiment . Our mapping approach focused on first identifying the location of copper tolerance locus because , of the two phenotypes , it is easier to measure , and then screen a subset of plants with recombination events near the tolerance locus for hybrid inviability . In order to genetically map these phenotypes with high resolution , we used a near isogenic line ( NIL ) created through recurrent backcrossing of a Copperopolis by Cerig35 F1 ( this line is compatible with Copperopolis ) to lines derived from Stinson Beach , California . For seven generations , only tolerant plants were selected to be backcrossed to Stinson Beach ( Figure S1 ) . The Stinson Beach population is nontolerant and lacks any intrinsic incompatibilities with Copperopolis [23] . We used multiple outbred Stinson Beach lines to avoid the development of inbreeding depression in our backcross lines . Each backcross generation segregated tolerant and nontolerant progeny in equal ratio , in accordance with predictions for a single , dominant tolerance locus . We created our mapping population by crossing a single tolerant ( BC7T ) and a single nontolerant ( BC7NT ) line to produce F1BC7 progeny ( Figure S1 ) . We set about identifying the introgressed tolerance region in the NIL lines ( BC7T , BC7NT ) by screening 541 genetic markers distributed across the genome ( mimulusevolution . org ) . We identified 42 markers that were homozygous in the BC7NT line and heterozygous in the BC7T line , possibly indicating a region of introgression from the tolerant parent . To establish which heterozygous markers are linked to the copper tolerance locus , we tested for an association between phenotype and genotype in 80 F1BC7 lines . We measured copper tolerance in the F1BC7 plants as a threshold character in accordance with previously published hydroponic methods [20] . We found the genotypes of one marker , MgSTS242 , located on linkage group ( LG ) 9 , were very tightly correlated with tolerance phenotype ( Figure 1 ) . We henceforth refer to this copper tolerance locus as Tol1 . To fine map the genomic position of Tol1 , we created a mapping population of 4 , 340 F1BC7 plants , phenotyped each plant for copper tolerance , and genotyped each plant with the MgSTS242 marker . Initially , we identified 83 recombinant plants , however we repeatedly phenotyped and genotyped these lines and we were able to confirm 36 true recombinant plants , indicating that this marker is 0 . 83 cM from the Tol1 ( Figure 2 ) . To fine map Tol1 , we designed additional genetic markers near MgSTS242 using the currently unpublished M . guttatus genome assembly ( www . phytozome . net ) . This assembly is not contiguous on the scale of chromosomes; the genome is instead composed of many , relatively small , genomic scaffolds that do not assemble into large chromosome-length blocks . The assembly in the Tol1 genomic region is particularly poor because this is a highly repetitive pericentromic region of LG 9 and none of the scaffolds that map to this region in the reference are longer than 1 MB . We attempted to determine the location of 10 target scaffolds relative to MgSTS242 and Tol1 by designing multiple new markers per scaffold and screening them in our recombinant lines . We identified three scaffolds that are in tight linkage to Tol1 ( Figure 2 ) . We determined that a marker on scaffold 84 is in tightest association with Tol1; we found 14 plants have a recombination event between Sc84_37kb and Tol1 , indicating this marker is located within 0 . 32 cM of the tolerance locus ( Figure 2 ) . We identified three additional scaffolds that map to this region ( 97c , 157 , and 238 ) , but they do not reside within the Sc84_37kb–Tol1 interval ( Text S1 ) . We have not included these scaffolds in our map because we have a limited amount of genetic material and we could not test all 36 recombinants . We attempted to identify a flanking marker for Tol1 by designing markers in the 3′ end of scaffold 63b , which is predicted to be adjacent to the 5′ end of scaffold 84 ( unpublished data , Uffe Hellsten , Joint Genome Institute , Walnut Creek , CA ) . However , we found no evidence of linkage between our markers and the tolerant phenotype , likely indicating that scaffold 63b is outside of the NIL introgression region or there are errors in the reference genome assembly ( Figure 2 and Text S1 ) . We observed the same pattern of no linkage between Tol1 and markers located in two additional scaffolds ( 460 and 925 ) that are predicted to be located in this region in the reference genome assembly . In order to determine if the tolerance and incompatibility phenotypes are controlled by the same locus , we crossed 18 ( nine tolerant and nine nontolerant ) recombinant plants to an incompatible genotype , Cerig10 ( Figure S2 ) . Crosses were conducted in each direction to determine if there is asymmetry in the incompatibility . We measured hybrid inviability as the percentage of offspring with a majority of tissue being yellow or necrotic in a block of 60 plants ( Figure 3A and 3B ) . We scored hybrid inviability in 2–8 replicate blocks per line . We used Cerig10 as our tester line because this line produced the highest level of hybrid lethality in crosses to Copperopolis lines . In crosses between Cerig10 and five Copperopolis genotypes , we found an average of 83% of the offspring were inviable ( Figure S2 ) . Our F1BC7×Cerig10 crosses produced a bimodal distribution of hybrid inviability . The majority of crosses with nontolerant lines produced 5%–15% inviable progeny , whereas the majority of crosses with tolerant lines produced 45%–62% inviable progeny ( Figure 3C and 3D ) . We identified two plants , 25_E01 and 25_E11 , with recombinant phenotypes ( Figure 3C ) . Hybrid inviability of nontolerant line 25_E01 is significantly elevated compared to all nontolerant control lines ( Wilcoxon test , z = 2 . 46 , p<0 . 014; Table S1 ) except 46_B12 , which had only two replicate measurements ( Wilcoxon test , z = 1 . 84 , p = 0 . 065; Table S1 ) . Hybrid inviability of tolerant line 25_E11 is significantly lower than the two tolerant control lines ( Wilcoxon test , z = 2 . 46 , p<0 . 014; Table S1 ) . This establishes that copper tolerance and hybrid inviability are controlled by two distinct , but tightly linked , loci and that hybrid necrosis is not a pleiotropic by-product of copper tolerance . We henceforth refer to the hybrid inviability locus as Nec1 . To map the genomic location of Nec1 , we compared the Tol1 genotyping information for these lines to our lethality data and determined that Nec1 maps to scaffold 84 . To fine map Nec1 , we designed markers in this region and genotyped plants with informative recombination events . We found that the incompatibility locus maps to an intergenic region between 283 kb and 293 kb on scaffold 84 ( Figure 4 ) . This region in the M . guttatus genome ( www . phytozome . net ) contains a single Gypsy3 transposable element ( TE ) ( Figure 4 ) . We are prevented from continuing to narrow this region , despite identifying three lines ( 26_C07 , 25_E01 , and 48_B01 ) with recombination events within this locus ( Figure 4 ) because of the repetitive nature of this element . We attempted to sequence across this genomic interval using long-range PCR , but these attempts were unsuccessful . Our long-range PCR may have failed because repetitive DNA in this TE element or because this genomic interval is much larger in the Copperopolis genome compared to the reference genome . Nec1 maps to an intergenic region containing a Gypsy3 TE . Unregulated TE replication can cause hybrid sterility in Drosophila; in this system , hybrid sterility manifests in crosses in a single direction because hybrids lack the maternally inherited TE replication-suppression mechanisms [30] . It is unlikely that this is the cause of inviability in our system because significant levels of hybrid lethality occur in both crossing directions , although it is higher when Cerig is the paternal parent ( Table S2 ) . Instead , the functional changes at Nec1 likely result from altered gene expression or gene function . The Copperopolis allele at Nec1 may cause hybrid lethality by altering the cis-regulatory expression of a gene within this region , and the functional gene may flank our mapped interval or reside even further away along the chromosome . The nearest gene , MGV1A004323M , resides 5 kb from Nec1 and encodes a glycosyltransferase metabolic enzyme ( www . phytozome . net ) . This protein has greatest homology to A . thaliana gene AT3G18170 , which attaches glyosyl groups to N-linked glycan molecules critical for construction of plant cell membranes [31] . The second flanking gene , MGV1A026627M , is located 41 kb from Nec1 and encodes a Jumonji-C histone demethylation protein . The Jumonji-C proteins are part of the histone demethylase family , whose function reverses the epigenetic silencing of transposons and genes in eukaryotic genomes [32] . Neither of these genes , nor is any gene within 100 kb of Nec1 , is a known participant in plant disease-resistance pathway . Alternatively , Nec1 may involve another gene that is absent from the reference genome . This gene may have undergone recent functional or expression changes to cause negative interactions with a member of the plant disease-resistance pathway , producing the hybrid necrosis phenotype . This hypothesis is especially attractive because the NB-LRRs often exhibit large variation in copy number within species [33] . Although additional work is required to distinguish between the possible mechanisms of Nec1 function , our results clearly demonstrate that hybrid inviability did not evolve as a pleiotropic by-product of the evolution of copper tolerance . What evolutionary force caused the lethality allele at Nec1 to go to high frequency in the Copperopolis population ? The hybrid lethality phenotype reached high frequency in the Copperopolis population because of genetic drift or natural selection . Genetic drift may have been the primary evolutionary mechanism causing the lethality allele at Nec1 to rise to high frequency if the initial population of mine colonists experienced a bottleneck event . This scenario would cause a genome-wide reduction in genetic variation in the mine population relative to neighboring off-mine populations [34] . Alternatively , the hybrid lethality allele at Nec1 may have been directly selected because this allele confers some fitness advantage in the mine habitat . A variant of the selection hypothesis is that the lethality allele at Nec1 may have hitchhiked to high frequency because of selection on the tightly linked Tol1 locus . In this case , the tolerance and incompatibility alleles would have resided on the same haplotype in the founders of the mine population . Selection on this genomic region would leave a molecular genetic signature of reduced genetic variation in the mine population and increased genetic differentiation between mine and off-mine populations in this genomic region , compared to the rest of the genome [35]–[38] . Studies that attempt to identify these molecular signatures of selection in natural populations have been used to identify putatively selected loci in many different systems [39]–[44] . The power of population genetic studies to detect signatures of selection can be diminished by multiple interdependent processes: selection on standing genetic variation [45] , population structure [46] , and time since a selective sweep [34] . These studies can most readily identify a molecular signature of selection in the classic case of a hard sweep , in which strong selection on a new mutation causes a single haplotype to rapidly go to fixation and tightly linked alleles hitchhike to high frequency [34]–[38] . However , if this mutation is segregating in the ancestral population and has recombined onto multiple haplotypes prior to selection , hitchhiking of tightly linked alleles is reduced [45] . This event , termed a “soft sweep , ” will not decrease genetic variation or increase population differentiation at linked alleles to the same degree as a hard sweep . The number of generations since a selected allele goes to high frequency affects the molecular signature of selection because each generation there is new opportunity for mutation , migration , and recombination to add genetic variation to the derived population [34] . The young age of the mine habitat ( 150 y ) suggests that if a sweep occurred , the molecular signature of this event would still be apparent . The mine and off-mine populations are in close geographic proximity , and there is opportunity for gene flow , suggesting there will be little effect of population structure . Gene flow will increase the amount of shared genetic variation at selectively neutral sites in the genome , but strong selection in the mine environment could maintain the original haplotype ( or haplotypes ) containing the tolerance allele in the face of gene flow . In order to evaluate the selection hypothesis , we compared genetic variation at markers linked to Tol1 and markers randomly distributed throughout the genome . We sampled individual plants , collected as seed from unique maternal families in 2005 and 2007 , from the mine population at Copperopolis ( N = 108 ) and from two off-mine populations: O'Byrnes Ferry Road ( N = 33 ) , located 2 km from Copperopolis , and Hunt Road ( N = 39 ) , located 9 km from Copperopolis ( Figure S3 ) . The geographic proximity of these populations and recent divergence of the mine population , a maximum of 150 generations , suggests any signal of selection will not be lost in a haze of population structure . Furthermore , this population sampling is designed to explicitly test for an effect of habitat-mediated selection on a single genomic region with known phenotypic effect , Tol1/Nec1 loci , and not a genome-wide analysis . Lastly , this design will not be confounded by the effect of hybrid inviability because these off-mine populations lack the incompatible allele found in the Cerig population [23] . We measured genetic variation of eight loci in the Tol1 fine mapped region ( Tol1Link ) and 11 unlinked loci ( Tol1UnLink ) for the Copperopolis and two off-mine populations . The Tol1UnLink markers serve as a control to estimate the average level of genetic variation and population differentiation . We used co-dominant DNA fragment length polymorphism markers [47] to estimate the within-population genetic variation , which we report as number of alleles ( Na ) and expected heterozygosity ( He ) [48] . We report both values , but these are often correlated when using fragment length markers . We calculated genetic differentiation between populations as Fst for two marker classes: Tol1Link and Tol1UnLink using Fdist [36] implemented in the program LOSITAN [49] . To evaluate whether Fst is significantly elevated in the Tol1 fine mapped region , we compared the observed levels of Fst at the Tol1Link markers to a null distribution of Fst generated from coalescence simulations of population divergence using the Tol1UnLink dataset in the program LOSITAN . Our hypothesis is that Fst for the Tol1Link markers will be significantly higher than the Tol1UnLink markers . The coalescence simulations assume two populations and use Wright's symmetrical Island Model of migration [50] to calculate the probability that two randomly chosen chromosomes in a population have a most recent common ancestor within that population without an intervening migration or mutation event [36] , [49] . This model also assumes that the two populations are at equilibrium with constant effective population size . Our populations likely violate this assumption because the mine population was only recently established . However , Beaumont and Nichols found that a recent colonization event had no significant effect on their estimates of the Fst distribution [36] . We observe low Fst for all Tol1UnLink markers , and for all but a single marker , there is no significant differentiation for the three pairwise population comparisons ( Table S3 ) . Using the Tol1UnLink dataset , we also estimate the relatedness between the three populations using the program STRUCTURE version 2 . 3 [51] . In this analysis , the two off-mine populations cluster more tightly with each other than with Copperopolis ( Figure S4 ) ; thus , we combined the two off-mine populations into a single sample for subsequent analyses . We found mine and off-mine populations have no difference in Na and only slightly reduced He for the Tol1UnLink markers ( Table 1 ) . This suggests that the mine population did not undergo a dramatic population bottleneck during colonization , or if it did , this signal has been erased by ongoing gene flow . Conversely , we find that Na and He of the Tol1Link markers is sharply reduced in the Copperopolis population compared to the off-mine population , and this pattern is strongest at the four markers in tightest linkage to Tol1 and Nec1 ( Table 1 ) . We found that the four markers that are most tightly linked to Tol1 and Nec1 have significantly elevated Fst when compared to the Tol1UnLink distribution ( Figures 5 and S5 ) . Fst values varied between the four markers; the markers with the two highest values are Sc84_364kb ( Fst = 0 . 546 ) and Sc84_37kb ( Fst = 0 . 317 ) are in tightest linkage to the Nec1 and Tol1 loci , respectively . These data support the two predictions arising from the selection hypothesis: low genetic variation within Copperopolis and high differentiation between populations at markers linked to the tolerance and hybrid lethality loci .
We demonstrate that copper tolerance is not pleiotropic with hybrid inviability . Instead , both phenotypes are controlled by distinct loci that are in tight genetic linkage . We genetically mapped the copper tolerance locus to a highly repetitive pericentromeric genomic region . The large amount of repetitive DNA and relatively rare euchromatic sequence has inhibited the assembly large genomic scaffolds in this region of M . guttatus reference genome ( unpublished data , Uffe Hellsten , Joint Genome Institute , Walnut Creek , CA ) . Our attempts to fine map the Tol1 locus to a genomic scaffold , or even identify a flanking marker , have been stymied by the complex nature of this genomic region despite our large mapping population . In contrast , we were able to fine map the hybrid lethality locus to a 10 kb intergenic region with a contiguous genomic scaffold . The Copperopolis allele at Nec1 may cause hybrid lethality by altering the cis-regulatory expression of a neighboring gene , or this region in the Copperopolis genome may harbor genes that are missing in the reference M . guttatus genome . We attempted to amplify the 10 kb region from the Copperopolis genome using long-range PCR , but these attempts were unsuccessful . The failed PCR attempts may be caused by repetitive DNA within the gypsy-3 TE or because this region may be expanded or rearranged in the Copperopolis genome compared to our reference . We observed a large amount of variance between independent replicates of the high-lethality lines ( Figures 3C , 3D , and S2 ) . This variance is likely caused by two factors—segregating variation at multiple incompatibility loci in our tester line , Cerig10 , and variation in temperature in the greenhouses in which the F1BC7×Cerig10 offspring were grown . There are multiple incompatibility loci segregating within the Cerig population that interact with the Copperopolis allele at the locus we have now identified as Nec1 [22] , [23] . Our tester line , Cerig10 , was chosen because it gave the highest level of incompatibility , but it may be heterozygous at some incompatibility loci , inflating variation in hybrid lethality . An additional source of variation may be the temperature at which the offspring were grown; research on hybrid necrosis in A . thaliana has demonstrated that this phenotype manifests at 16°C but is absent at 23°C [27] . Our plants were grown under controlled greenhouse conditions , but daily or seasonal temperature fluctuations may have exceeded the capability of the greenhouse to maintain constant temperature . We find strong support for the hypothesis that adaptation to the mine habitat imposed selection on the Nec1/Tol1 genomic region in the Copperopolis population . The high level of genetic variation in the Tol1UnLink dataset in the Copperopolis population argues against a dramatic bottleneck event during mine colonization . The four markers in tight linkage to Tol1 and Nec1 have reduced Na and He in the mine population , as well as significantly elevated Fst . We interpret these data as evidence of a sweep at Tol1 , driven by selection for copper tolerance , causing the Nec1 BDM incompatibility allele to hitchhike to high frequency in the mine population . This model predicts that the marker in tightest linkage to Tol1 , Sc84_37kb , would have the highest Fst and sharpest reduction in He . Contrary to this prediction , we found that marker Sc84_364kb , the marker nearest Nec1 , has the highest Fst and sharpest reduction in He . This finding suggests that adaptation to the mine environment was more complex than the classic hard selective sweep at Tol1 . Although we cannot completely rule out the possibility that the incompatibility allele at Nec1 has also experienced selection in the mine population , the physiological effect of the tolerance allele and the biogeographic distribution of copper tolerant plants , which are nearly fixed in multiple mine populations and are segregating at 8 . 3% of neighboring off-mine populations , is strong evidence that copper tolerance is under selection in the mine habitat [21] . Because the lethality allele at Nec1 has only been found in a single , very recently derived population [23] and has no discernible phenotypic effect beyond causing hybrid lethality , it is less likely this allele is experiencing selection in the mine environment . We suggest multiple factors that may explain why the patterns of He and Fst we observe do not conform to the predictions of a classic hard selective sweep . Variation in the mutation rate between markers Sc84_37kb and Sc84_364kb may explain the differences in He and Fst . Additionally , we note that we have been unable to fine map the precise location of Tol1 because this is a highly repetitive pericentromeric region and the Sc84_367kb marker may actually be closer to Tol1 in the Copperopolis genome than its position in the reference genome suggests . Finally , we suggest that the tolerance allele at Tol1 may not have experienced a hard selective sweep . If the tolerance allele was segregating at low frequency in the ancestral population and experienced a soft selective sweep , tightly linked sites would be affected differently depending on the amount of genetic variation at these sites in the ancestral population and the level of linkage disequilibrium ( LD ) between the selected allele and tightly linked sites when selection first acted upon this population [45] . Tolerant plants are segregating at 8 . 3% in off-mine populations [23] , and although we are unable to determine whether this is caused by segregating ancestral variation or recent migration from mine to off-mine populations , it indicates that the tolerant allele may predate the development of the copper mines and recently rose to near fixation via a soft selective sweep . We investigate whether a soft sweep at Tol1 could have caused the lethality allele at Nec1 to hitchhike to high frequency using a quantitative model of genetic hitchhiking . To simulate hitchhiking on tightly linked sites following a hard or soft selective sweep , we use the two-locus model of genetic hitchhiking described by Maynard Smith and Haigh [35] and Barton [38] . In the classic model of a hard selective sweep , the beneficial allele resides on a single haplotype and the selected and linked alleles are initially in complete LD . Because copper mining was initiated in the 1860s , we focused on strong selection situations that would cause a beneficial allele to become nearly fixed within 150 generations . We model the effects of hitchhiking on neutral allele segregating at low frequency ( 5% ) in the ancestral population because the hybrid lethality allele has not been found in neighboring off-mine populations [23] . Additional details of our hitchhiking model are provided in Text S1 . Consistent with previous findings [38] , we find that during a hard selective sweep , tightly linked alleles within 0 . 5 cM hitchhike to high frequency in the derived population ( Figure S6A , S6D , and S6G ) . We next model the soft sweep scenario , in which the selected allele is segregating at low frequency ( 5% and 10% ) in the ancestral population . We find that a soft selective sweep can cause sites within 0 . 5 cM of the beneficial allele to hitchhike to high frequency for a wide range of parameter values ( Figure S6B , S6C , S6E , S6F , S6H , and S6I ) . The primary determinant of hitchhiking effects is the initial level of LD between the selected and neutral alleles . In the hard sweep model , LD is 1 , however this parameter can vary when the focal beneficial mutation and a tightly linked allele are segregating in the ancestral population . We find that when we set initial levels of LD to high levels ( 0 . 8–1 . 0 ) , a tightly linked allele within 0 . 5 cM can hitchhike to high frequency in the derived population ( Figure S6 ) . Overall , these findings demonstrate that for this situation , in which a M . guttatus population has rapidly adapted to the copper mine environment , a soft or hard selective sweep can produce large hitchhiking effects on tightly linked sites that happen to be in strong LD in the ancestral population . Although this hitchhiking model makes assumptions that are likely violated in natural populations—deterministic selection on an allele , bi-allelic loci , and constant population size—it provides a useful quantitative framework to describe the effects of genetic hitchhiking on loci in tight genetic linkage . The variation we observe in of He and Fst between Tol1Link markers suggests that adaptation in the mine environment may have occurred in a manner more complex than the classic hard sweep model . A soft selective sweep may not uniformly reduce genetic variation or increase genetic differentiation at tightly linked sites if there were differences in the amount of ancestral genetic variation at these sites or variation in the levels of LD between the selected allele and linked sites in the ancestral population [45] . In a soft sweep , the primary determinant of the strength of hitchhiking effects is the initial level of LD between the selected allele and tightly linked alleles . Unfortunately , the extent of LD in the ancestral population for this pericentromeric region is not known . Although we cannot conclusively state whether there was a hard or soft sweep at the Tol1/Nec1 region , we do find strong evidence of selection on this region in the mine population , and the results of our hitchhiking simulations of a hard or soft sweep are consistent with the hypothesis that selection at Tol1 caused the hybrid lethality allele at Nec1 to hitchhike to high frequency in mine population . In summary , we have determined that the copper tolerance and hybrid inviability phenotypes are controlled by two distinct , tightly linked loci . Hybrid inviability is not a pleiotropic by-product of adaptation to the mine environment , as originally proposed [22] . We find that Tol1 maps to a highly repetitive , unassembled region of the M . guttatus genome , whereas Nec1 maps to a 10 kb region containing a single TE . Furthermore , we demonstrate that there has been strong positive selection on the Nec1/Tol1 genomic region . We interpret these data as evidence that the Nec1 lethality allele rose to high frequency because of genetic drift induced by selection on the tightly linked Tol1 . We demonstrate that natural selection on one locus can cause a tightly linked hybrid lethality allele to hitchhike to high frequency , providing empirical evidence for a new twist on Darwin's hypothesis that reproductive isolation can evolve as an incidental by-product of adaptation to novel environments .
Our mapping population was created with repeated backcrossing and phenotypic selection for copper tolerance [20] , [21] . The initial cross between Copperopolis and Cerig was conducted in 1981 , but after the discovery of hybrid inviability loci segregating in this population , all subsequent crosses were made to the Stinson Beach population . The Tol1/Nec1 mapping population was grown and maintained at the University of Exeter greenhouse from 2005–2009 . Tolerance was scored as a threshold character as previously described [20] . Tolerance was measured repeatedly ( 2–6 times ) for every putative recombinant line , although a few lines did perish in the greenhouse before they could be re-tested . To score hybrid lethality , we grew 60 seedlings from a focal cross and scored the number of plants with majority yellow or necrotic tissue after 3 wk of growth [22] . Hybrid lethality was scored in multiple ( 2–8 ) growouts per line from 2008–2009 . Genetic mapping of Tol1 and Nec1 was conducted at Duke University using tissue samples shipped from University of Exeter . Genomic DNA was extracted from plant tissue using a modified hexadecyl trimethyl-ammonium bromide chloroform extraction protocol [52] . Most of the genetic markers used in this analysis are DNA fragment-length polymorphism markers analyzed using capillary electrophoresis on an ABI 3730×l DNA Analyzer [53] . These genetic markers score variation in PCR fragment length created by insertion-deletions . For most markers , primers reside in conserved exon sequence and amplify intron sequence [47] . The size of the amplified fragments was scored automatically by the program GENEMARKER ( SoftGenetics , 2005 , State College , PA ) and was confirmed by eye . For additional fine-mapping markers , we used di-deoxy sequencing of PCR products to identify SNPs distinguishing Copperopolis and Stinson Beach alleles ( Markers: Sc84_180kb , 234kb , 252kb , 277kb , 281kb , 283kb , 293kb , 297kb; Sc86_144kb , 400kb; Sc341_99kb ) . SNPs were scored manually using Sequencher ( Gene Codes Corp . , Ann Arbor , MI ) . To be sure of genotypes for markers defining the Nec1 locus , Sc84_283kb , 293kb , and 297kb , we ran 2–3 independent reactions for each genetic line . We sampled Copperopolis , O'Byrnes Ferry Road , and Hunt Road populations in 2005 and 2007 . We collected seed from unique maternal plants and grew a single plant from each maternal line at Duke University greenhouse for genetic testing . We measured genetic variation using fragment-length polymorphism markers [47] . We calculated the number of alleles and heterozygosity using the program Arlequin [48] . We calculated observed levels of Fst for Tol1Link and Tol1UnLink markers and generated null distribution of Fst using Tol1UnLink markers using coalescence simulations implemented in the program LOSITAN [49] with the following settings: two populations in the Island Model , 1 , 000 , 000 simulations , neutral mean Fst , and infinite allele mutation model as suggested for microsatellite markers [34] . To conduct numerical simulations of the two-locus hitchhiking model based on Maynard Smith and Haigh [35] and Barton [38] , we wrote a program in C . Constant variables in each simulation were the population size , N , set at 1 , 000 , and the allele frequency of tightly linked neutral allele , U0 , set at 0 . 05 . We ran simulations varying the strength of selection , initial frequency of selected allele , and LD between the selected allele and tightly linked sites ( Figure S6 ) .
|
Adaptive evolution by natural selection is the primary force generating biological diversity . A critical question is whether the evolution of hybrid incompatibility , which is essential for the maintenance of species diversity , is caused by adaptive evolution . In this article , we investigate one of the most widely cited examples of ecological divergence driving the evolution of reproductive incompatibility , the strong association between hybrid lethality and copper tolerance in a copper mine population of the wildflower Mimulus guttatus . Hybrid lethality and tolerance of high levels of copper co-segregate as a single Mendelian locus . While copper tolerance and hybrid lethality are nearly universal in the mine population at Copperopolis , California , they are absent from adjacent off-mine populations , suggesting that reproductive isolation evolved rapidly as a pleiotropic by-product of recent adaptation to the mine environment . We find that copper tolerance and hybrid lethality are controlled by distinct loci , in tight genetic linkage . We also demonstrate that this genomic region has experienced strong recent selection and conclude that ecological selection for copper tolerance indirectly caused the neighboring hybrid lethality allele to hitchhike to high frequency . To our knowledge , this is the first case to demonstrate that reproductive isolation factors can evolve as an incidental by-product of adaptation to novel environments through genetic hitchhiking .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"evolutionary",
"ecology",
"plant",
"biology",
"population",
"genetics",
"quantitative",
"traits",
"plant",
"science",
"linkage",
"(genetics)",
"plant",
"genetics",
"biology",
"evolutionary",
"genetics",
"genotypes",
"trait",
"locus",
"plant",
"evolution",
"phenotypes",
"heredity",
"genetics",
"gene",
"flow",
"evolutionary",
"biology",
"evolutionary",
"processes",
"genetics",
"and",
"genomics",
"complex",
"traits"
] |
2013
|
Indirect Evolution of Hybrid Lethality Due to Linkage with Selected Locus in Mimulus guttatus
|
We study apo and holo forms of the bacterial ferric binding protein ( FBP ) which exhibits the so-called ferric transport dilemma: it uptakes iron from the host with remarkable affinity , yet releases it with ease in the cytoplasm for subsequent use . The observations fit the “conformational selection” model whereby the existence of a weakly populated , higher energy conformation that is stabilized in the presence of the ligand is proposed . We introduce a new tool that we term perturbation-response scanning ( PRS ) for the analysis of remote control strategies utilized . The approach relies on the systematic use of computational perturbation/response techniques based on linear response theory , by sequentially applying directed forces on single-residues along the chain and recording the resulting relative changes in the residue coordinates . We further obtain closed-form expressions for the magnitude and the directionality of the response . Using PRS , we study the ligand release mechanisms of FBP and support the findings by molecular dynamics simulations . We find that the residue-by-residue displacements between the apo and the holo forms , as determined from the X-ray structures , are faithfully reproduced by perturbations applied on the majority of the residues of the apo form . However , once the stabilizing ligand ( Fe ) is integrated to the system in holo FBP , perturbing only a few select residues successfully reproduces the experimental displacements . Thus , iron uptake by FBP is a favored process in the fluctuating environment of the protein , whereas iron release is controlled by mechanisms including chelation and allostery . The directional analysis that we implement in the PRS methodology implicates the latter mechanism by leading to a few distant , charged , and exposed loop residues . Upon perturbing these , irrespective of the direction of the operating forces , we find that the cap residues involved in iron release are made to operate coherently , facilitating release of the ion .
Functional proteins are complex structures , which may remain mainly unmodified as a result of a multitude of mutations [1] , yet may have their energy surface go through significant changes upon perturbing highly specific regions [2]–[4] . The various accessible states populated may be manipulated by inducing short and long-range conformational changes in the structure [5]; alternatively , a dynamical control may take place without any significant structural variation [6] , [7] . To explore the presence or the absence of such “shifts in the energy landscapes , ” [8] one needs to perturb the protein structure , and observe the response [9] . The perturbation may be in the form of changing the environmental factors ( e . g . changes in ionic concentration [10] ) , or may target specific locations on the structure itself , either through chemically modifying the residues ( inserting mutations ) [11] or by inducing site-specific perturbations ( e . g . as is done in single molecule experiments [12] , or through ligand binding ) . Ubiquitous post-translational modifications are also possible . The response may be measured directly , as a change in the overall conformation of the protein [13] , or indirectly , e . g . , through determining the kinetic parameters , and proposing kinetic models that explain the observations . [14] , [15] The purpose in such work is to understand and therefore control the response of the protein for a plethora of reasons , including , but not limited to , the design of efficient drugs [16] , [17] , or to tailor enzymes serving as “materials . ”[18] Linear response theory ( LRT ) has been recently used to study conformational changes undergone by proteins under selected external perturbations [19] . This approach has recently been applied to the study of the conformational switching upon phosphorylation [20] . In this study , we develop a toolkit that we term perturbation-response scanning ( PRS ) which is based on sequential application of LRT to study the origins of structural changes undergone by protein molecules . Similar approaches have been adopted in other work , whereby the perturbations on residues are introduced by modifying the effective force constants [21] or distances [22] between contacting pairs . PRS relies on systematically applying forces at singly selected residues and recording the linear response of the whole protein . The response is quantified as both the magnitude of the displacements undergone by the residues , and their directionality . Closed form expressions that summarize the theoretical implications of the PRS technique in the limit of a large number of perturbations introduced at a given residue are provided . We note that we have previously studied the stability of proteins using a similar sequential perturbation-response approach , based on inserted displacements followed by energy minimization of the system [9] , [23] . Therein we have also shown that the response of the system is within the linear regime for local distortions of atoms up to ca . 1 . 5 Å , despite the large local forces brought about [9] . Using PRS , we analyze the ferric binding protein A ( FBP ) as an example system , and describe alternative approaches that may have evolved in the structure to control function . The validity of the methodology is supported by molecular dynamics ( MD ) simulations . FBP is involved in the shuttling of Fe+3 from the mammalian host to the cytoplasm of pathogenic bacteria . To make iron unavailable to such pathogens , host organisms have iron transport systems such as the protein transferrin that tightly sequester the ion . Pathogens have developed strategies to circumvent this approach , one of them being the development of receptors for the iron transport proteins of the host . FBP resides in the periplasm , and receives iron from these receptors to eventually deliver it to the cytosol [24] . The protein is made up of two domains characteristic of periplasmic Fe+3 binding family as well as the host protein transferrin . These host/pathogen iron uptake proteins are thought to be distantly related through divergent evolution from an anion binding function . Fe+3 is bound to FBP with remarkable affinity , with association constants on the order of 1017–1022 M−1 depending on the measurement conditions [25] . It was recently shown that a relatively high affinity of iron binding is required for the removal of iron from transferrin , and its transport across the periplasm [26] . Yet , this high affinity poses a Fe+3 transport dilemma , suggesting another necessary step for the release of the ion . It is of interest to understand how Fe+3 is eventually released from the binding site for subsequent use by the pathogen . One mode of action that has been suggested involves the control of the Fe+3 release kinetics by the exchange of synergistic anions forming relatively stable intermediates [25] , [27] , [28] . Another involves the direct action of chelators on the ion [25] . It has also been shown that mutants of FBP that are defective in binding the synergistic anion are still capable of donating iron , suggesting the possibility of still other alternative mechanisms for the process [29] . FBP is referred to as bacterial transferrin due to the similarities with transferrin in the structural folds , the highly conserved set of iron-coordinating residues , and their usage of a synergistic anion [30] . They do , however , differ in size , transferrin being made up of two-lobes having high sequence identity with each other ( e . g . 45% in human transferrin ) . Each lobe itself is comparable to FBP in size , fold , and iron binding location . In transition from the open to the closed form , only one of the sub-domains in each lobe undergoes significant reorientation , similar to FBP [31] . Despite the resemblance , the iron binding/release kinetics in the two lobes differ . It has been implicated that there may be several approaches used for iron release in transferrins , including chelating agents and synergistic anions acting directly around the ferric binding site [32]–[35] . Additionally , it has been shown that chloride and other ion concentrations are effective on the kinetics , and it has been proposed that allosteric anion binding sites that trigger large conformational changes exist [10] , [36] , [37] . Based on the similarities between FBP and transferrin , it is of interest to find out if these routes also exist for FBP , and if they do , what the details of the mechanisms are . It is also of significance to determine possible binding locations on the surface as well as to understand the physical origin of such control . In the current work , we study FBP in detail due to an extensive literature on the iron uptake mechanisms of this and evolutionarily related proteins; moreover , the molecular dynamics ( MD ) results of the apo structure have previously been analyzed by perturbing a singly selected residue with linear response theory [19] . We develop the PRS scheme such that , ( i ) we systematically apply LRT by scanning every residue on the protein so as to discriminate between residues that have major contributions to the biologically significant displacements , measured by x-ray experiments; ( ii ) we provide closed-form expressions for both the magnitude and the directionality of the response; ( iii ) we carry out further analysis of the response to uncover regions in the protein where coherent responses occur . Our findings are in agreement with a model where iron uptake by the protein is a favored process in the fluctuating environment , while iron release is specifically managed through several mechanisms including chelation and allosteric control . Furthermore , our findings suggest additional locations on the protein surface , far from the binding site , for allosteric control of iron release . The observations fit the “conformational selection” model whereby the existence of a weakly populated , higher energy conformation that is stabilized in the presence of the ligand is proposed [38] , [39]; the hypothesis has recently been supported by NMR experiments by studying the protein structural ensemble of up to microseconds [40] .
In this study we introduce the PRS methodology based on systematically perturbing all residues of a protein , and classifying both the magnitude and the directionality of the recorded linear response . The approach is unique in that the cross-correlations between residues are processed as input for further predictions on the system behavior , unlike other methods where they are the final results . Closed-form expressions for the magnitude and the directionality of the response are also provided . The protein FBP is studied in detail , and results using PRS for additional cases are presented in Text S1 . All computer programs used in the analyses are available upon request . The findings on FBP are also supported by MD simulations . For the particular case of FBP , our analysis suggest the existence of two alternative mechanism of Fe ion release in FBP: ( i ) Local control of the ion by synergistic anions and chelators acting in the binding groove , and ( ii ) remote control by ions acting on distant charged residues located in solvent exposed loops ( e . g . , D47 , D52 , K234 ) due to their observed ability to mechanically control the cap over the ligand binding region . For FBP , the former type of control has been evidenced by a plethora of experiments where the exchange of synergistic anions forming relatively stable intermediates or the direct action of chelators on the ion have been observed [25] , [27] , [28] . There are no mutational studies on FBP directly implicating the distant residues mentioned in latter scenario . However , it was recently shown that H . influenza strains expressing mutant proteins that are defective in binding the phosphate anion are capable of donating iron , calling for mechanisms of iron transport that do not involve a synergistic anion . [29] Furthermore , the kinetic effect of chloride and perchlorate ( which does not coordinate to Fe+3 ) has called for anion binding sites on the surface of FBP , similar to those found in transferrin [27] . Allosteric anion binding sites that trigger large conformational changes located at the surface [10] , [36] , [37] have been determined in the structurally and functionally similar protein transferrin . In particular , R124 located in the N-lobe of the latter has been found to control iron release rate by anchoring synergistic anions [32] . Structural alignment [56] of transferrin with FBP shows that this position is equivalent to F142 located in the helix supporting the Fe binding region; the latter is amongst the residues that result in the highest correlations with the experimental data upon perturbation . Similarly , K206 which provides anion binding sites in human transferrin N-lobe holds an equivalent position to that of E193 in FBP , the latter also showing high displacement correlation following its perturbation ( figure 6 ) . The results resolve the so-called Fe+3 transport dilemma: The protein is ready to uptake the ion in the apo structure ( figure 6a ) , but it is necessary to perturb highly specific locations along the chain for its release in holo FBP ( figures 6b and 7 ) . This mode of action provides a mechanism for recent NMR observations of ligand binding whereby the energy landscape of the free protein is made-up of a set of coupled low-free energy states [5] , [40] , [57] , [58] . Therein , ligand binding is considered as the source for shifts in the landscapes [38] , [39] . PRS confirms this view for the particular case of FBP , and further provides the mechanisms on how the ligand binding region may be manipulated , as outlined by scenarios ( i ) and ( ii ) . In summary , the PRS methodology introduces an efficient approach to determine regions in the protein that mechanically moderate binding region motions . It may therefore be used to determine candidate sites for mutational studies . In a forthcoming study , the biological implications from the results of PRS will be presented on set of twenty proteins that display various types of motions such as shear , hinge , allosteric , partial refolding as well as more complex protein motions , as classified in the Database of Macromolecular Motions [59] . Although studies such as the current one help estimate allosteric sites , they do not provide information on pathways along which the remote communication takes place . As future work , it is of interest to locate these paths . Robust techniques to predict them using evolutionary information [60] or the specificity of the interactions within the residue networks [61] have been developed . Cross-correlation information from MD simulations have been used with residue network properties [62] to generate information on remote communication pathways [60] , [63] , [64] . Complemented with point mutation studies , such analyses will not only aid the protein design process , but will also uncover the physics of remote communication between residues .
|
Upon binding ligands , many proteins undergo structural changes compared to the unbound form . We introduce a methodology to monitor these changes and to study which mechanisms arrange conformational shifts between the liganded and free forms . Our method is simple , yet it efficiently characterizes the response of proteins to a given perturbation on systematically selected residues . The coherent responses predicted are validated by molecular dynamics simulations . The results indicate that the iron uptake by the ferric binding protein is favorable in a thermally fluctuating environment , while release of iron is allosterically moderated . Since ferric binding protein exhibits a high sequence identity with human transferrin whose allosteric anion binding sites generate large conformational changes around the binding region , we suggest mutational studies on remotely controlling sites identified in this work .
|
[
"Abstract",
"Introduction",
"Discussion"
] |
[
"physics/condensed",
"matter",
"biophysics/theory",
"and",
"simulation",
"biophysics/biomacromolecule-ligand",
"interactions",
"mathematics/statistics",
"computational",
"biology/molecular",
"dynamics",
"physics/interdisciplinary",
"physics"
] |
2009
|
Perturbation-Response Scanning Reveals Ligand Entry-Exit Mechanisms of Ferric Binding Protein
|
Ferroportin ( FPN ) is the only known cellular iron exporter in mammalian cells and plays a critical role in the maintenance of both cellular and systemic iron balance . During iron deprivation , the translation of FPN is repressed by iron regulatory proteins ( IRPs ) , which bind to the 5′ untranslated region ( UTR ) , to reduce iron export and preserve cellular iron . Here , we report a novel iron-responsive mechanism for the post-transcriptional regulation of FPN , mediated by miR-485-3p , which is induced during iron deficiency and represses FPN expression by directly targeting the FPN 3′UTR . The overexpression of miR-485-3p represses FPN expression and leads to increased cellular ferritin levels , consistent with increased cellular iron . Conversely , both inhibition of miR-485-3p activity and mutation of the miR-485-3p target sites on the FPN 3′UTR are able to relieve FPN repression and lead to decreased cellular iron levels . Together , these findings support a model that includes both IRPs and microRNAs as iron-responsive post-transcriptional regulators of FPN . The involvement of microRNA in the iron-responsive regulation of FPN offers additional stability and fine-tuning of iron homeostasis within different cellular contexts . MiR-485-3p-mediated repression of FPN may also offer a novel potential therapeutic mechanism for circumventing hepcidin-resistant mechanisms responsible for some iron overload diseases .
While iron is an essential nutrient for all cells , high levels of iron can lead to toxicity . Therefore , cellular iron homeostasis is carefully maintained by an exquisite system of iron regulatory proteins ( IRPs ) that respond to iron levels and coordinate the expression of targets important for balancing iron export and uptake with intracellular storage and utilization [1] , [2] . Ferroportin ( FPN ) functions as the only known iron exporter in mammalian cells and plays a critical role in the maintenance of both cellular and systemic iron balance [3]–[5] . Although ubiquitously expressed , FPN is most abundant in cell types known to absorb , process , recycle , and export significant amounts of iron , including duodenal enterocytes , hepatocytes , erythroid cells and reticuloendothelial macrophages [5]–[7] . Given the important regulatory role of FPN , it is not surprising that FPN is regulated at multiple levels–transcriptionally by heme [8] , [9] , post-transcriptionally by the IRP system [10]–[12] , and post-translationally by the iron regulatory hormone hepcidin [13] , [14] . During iron deficiency , IRPs inhibit the translation of FPN by binding to the iron regulatory element ( IRE ) located in the 5′ untranslated region ( UTR ) of FPN messenger RNA ( mRNA ) , leading to lower FPN protein levels , decreased export of iron , and cellular iron retention [11] . Hepcidin targets membrane-bound FPN for degradation and decreases FPN-mediated iron export . Defects in this FPN ‘off-switch’ as a result of hepcidin deficiency or hepcidin resistance due to FPN gain-of-function mutations can eventually lead to systemic iron overload in the form of hemochromatosis [15]–[18] , resulting in significant tissue damage and multi-organ failure with limited therapeutic options [19] . Therefore the identification of novel mechanisms for the post-transcriptional regulation of FPN that can bypass these pathogenic defects will be an important step in the development of novel interventions to ameliorate iron overload and improve clinical outcomes for these patients . Although there is a greater understanding of transcriptional and IRP-mediated regulation of FPN under various stresses such as heme , nitric oxide , oxidative stress , and hypoxia [8] , [9] , [20]– , it is not clear whether an IRP-independent mechanism exists for post-transcriptional FPN regulation . One potential class of post-transcriptional regulators are microRNAs— endogenous non-coding small RNAs that bind to complementary sites in the 3′UTR of target mRNAs and drive translational repression or mRNA degradation [23]–[26] . MicroRNAs have been found to play roles as important mediators in various stress responses from flies , worms , and zebrafish to mammals [27] . Recently , the liver-enriched miR-122 has been found to be critical for the control of systemic iron homeostasis in mice by targeting Hfe and Hjv , which encode proteins important for the hepcidin hormone response to systemic iron availability [28] . While this landmark study focused on miR-122 and captured its role in systemic iron homeostasis , it is unknown whether there are microRNAs that respond to intracellular iron levels and play a role in cellular iron homeostasis , particularly those that can potentially regulate FPN expression . In this study , we examine the potential role of microRNAs in the IRP/IRE-independent post-transcriptional regulation of FPN . We identify microRNAs with altered expression under cellular iron deprivation and find that the microRNA miR-485-3p directly targets the 3′ UTR of FPN . Through gain-of-function and loss-of-function studies , we provide compelling evidence to support a role for miR-485-3p as an important post-transcriptional regulator of endogenous FPN expression and modulator of cellular iron homeostasis .
To investigate the iron-responsive regulation of the FPN expression , we first demonstrated the known iron-dependent IRP-mediated regulation of the FPN 5′UTR using a luciferase reporter construct with the full FPN 5′UTR placed upstream of luciferase ( Figure 1A ) . The iron-replete and iron-deficient conditions were created by addition of the iron supplement ferric ammonium citrate ( FAC ) and the iron chelator deferoxamine ( DFE ) , respectively . When normalized to the activity of a control empty reporter under identical treatment , we found that FPN 5′UTR luciferase activity was responsive to iron levels in the human HepG2 hepatocyte cell line— significantly decreased during iron depletion ( 0 . 681 fold baseline ±0 . 029 , p<0 . 0001 ) and significantly increased ( 2 . 554 fold baseline ±0 . 099 , p<0 . 0001 ) during iron supplementation ( Figure 1B ) . Similar results were seen in the human K562 human erythroid cell line ( Figure S1 ) . To determine whether the FPN 3′UTR , which lacks IRE , could also be a target of iron-dependent regulation , we used a reporter construct with the full FPN 3′UTR placed downstream of luciferase and analyzed reporter activity in HepG2 cells under different iron conditions . Surprisingly , we found that iron depletion led to significant inhibition of FPN 3′UTR reporter activity ( 0 . 437 fold baseline ±0 . 041 , p = 0 . 012 ) ( Figure 1B ) . Additionally , iron supplementation led to significant increase ( 1 . 463 fold baseline ±0 . 024 , p<0 . 0001 ) in FPN 3′UTR reporter activity . Similar results were seen in K562 cells ( Figure S1 ) . Since the 3′UTR lacks the IRE region , it is unlikely that these changes are a result of IRP-mediated regulation . Collectively , these data show that both the FPN 5′UTR and 3′UTR can be regulated by iron concentration and indicate an unexpected regulatory role for the 3′ UTR in iron-dependent regulation of FPN . Both RNA-binding proteins and microRNAs are known to function as post-transcriptional regulators via the 3′UTR . To identify microRNAs that could play a role in this regulation , we performed microRNA profiling to identify iron-responsive microRNAs in the K562 erythroid cell line , a well-characterized model for the study of cellular iron metabolism [7] , [29] , [30] . We treated K562 cells with FAC ( iron-rich condition ) , DFE ( iron-deficient condition ) , or mock ( baseline ) treatment . Following DFE treatment , we noted increased levels of both IRP2 protein ( Figure 1C ) and transferrin receptor mRNA ( Figure 1D ) , as expected under iron depletion . We then used quantitative Real-Time PCR Taqman Low Density Arrays ( TLDA ) to measure the expression of 754 microRNAs under these different iron conditions . Threshold cycle ( Ct ) values were obtained by the RQ Manager v1 . 2 software with automatic threshold settings . Of the 300 microRNAs considered to be expressed under these conditions , we identified 44 microRNAs which were differentially expressed from baseline by log2 expression of at least 0 . 5 in either the iron-deficient or iron-rich condition ( Figure 1E ) . To prioritize the iron-responsive microRNAs with potential regulatory roles in iron homeostasis , we first analyzed the microRNAs ( Figure S2A–S2B ) with predicted mRNA targets in the cellular iron homeostasis gene ontology ( GO: 006879 ) ( Table S1 ) , using the microRNA . org and TargetscanHumanv6 . 0 databases [31] , [32] . Notably , 7/8 repressed and 16/21 induced microRNAs under iron deficiency have predicted iron-related targets ( Figure S2A–S2B ) . Two induced microRNAs , miR-485-3p and miR-194 , are predicted to target the FPN 3′UTR ( Figure 2B ) . We used individual TaqMan microRNA Real-time assays to confirm the induction of miR-485-3p , miR-194 , and three additional microRNAs ( miR-30a* , miR-149 , and miR-502-3p ) in independent biological replicates in K562 under iron deprivation ( Figure 2A and Figure S2C ) . To determine if these results could be seen in other cell types , we also measured the expression of these microRNAs in response to iron deprivation in HEL ( human erythroid ) , HEK293 ( human embryonic kidney ) , and HepG2 ( human hepatocyte ) cell lines ( Figure 2A and Figure S2C ) . We found that miR-485-3p exhibited the most uniform and significant induction in response to iron deprivation across all four tested cell lines . To further verify these findings in primary cells , we subjected human primary macrophages to iron depletion and found a similar degree of miR-485-3p induction ( Figure S2D ) . To date , miR-485-3p has shown only one confirmed target , which is involved in the expression of DNA topoisomerase II in human lymphoblastic leukemia cells [33] . Another study identified an allele variant in functional miR-485-3p target sites of the neurotrophin-3 receptor gene ( NTRK3 ) as a susceptibility factor for anxiety disorders [34] . While most of the targets predicted by TargetScan 6 . 2 ( Table S2 ) have not been functionally validated , these targets contain many genes involved in G-protein coupled receptor protein signal ( GO:0007186 ) , response to external stimulus ( GO:0009605 ) and regulation of metabolism ( GO:0019222 ) . The FPN 3′UTR has predicted target sites for both miR-194 and miR-485-3p ( Figure 2B ) . To determine whether the FPN 3′UTR is targeted by these microRNAs , we used expression constructs encoding the precursor hairpin sequences for miR-194 ( pc-miR-194 ) or miR-485 ( pc-miR-485 ) to overexpress miR-194 and miR-485 and measure their respective effects on FPN 3′UTR reporter activity . Enforced miR-485 expression led to significant inhibition of FPN 3′UTR reporter activity ( 0 . 639 fold control ±0 . 044 , p<0 . 001 ) ( Figure 2C and Figure S2E ) , while miR-194 overexpression did not inhibit ( 1 . 186 fold control ±0 . 05 , p = . 001 ) reporter activity ( Figure S2E ) . These data indicate that miR-485-3p , but not miR-194 , can act to repress the FPN 3′UTR . We then used antisense-mediated 2′-O-methyl oligonucleotides ( AMOs ) specific for miR-485-3p to determine the effect of inhibition of endogenous miR-485-3p-mediated RNA-induced silencing complex ( RISC ) activity [35] on the FPN 3′UTR reporter . Treatment with AMO-485-3p led to significantly increased FPN 3′UTR luciferase reporter activity compared to control in HepG2 ( Figure 2D ) and K562 ( Figure S2F ) cells . Next , we mutated the sequence of the only predicted canonical 8mer miR-485-3p binding site on the FPN 3′UTR , given the high confidence for microRNA-mediated repression with this predicted seed match type [36] , and created mutant FPN 3′UTR reporters with mutation in either one predicted site ( MT-448 or MT-618 ) or both predicted sites ( MT-448+618 ) ( Figure 2E ) . While both individual mutations led to significantly increased reporter activities compared to the wild type FPN 3′UTR reporter at baseline ( Figure 2E ) , the change caused by MT-618 ( 1 . 536 fold control ± . 065 , p<0 . 001 ) was more than that caused by MT-448 ( 1 . 359 fold control ± . 066 , p< . 0001 ) . Mutation of both sites led to even higher ( 1 . 690 fold control ± . 086 , p< . 0001 ) reporter activity ( Figure 2E ) , indicating that both predicted miR-485-3p binding sites contribute significantly to regulation of the FPN 3′UTR . To determine the effect of the predicted miR-485-3p binding sites on the FPN 3′UTR during the iron-deficient state , we measured luciferase activity of the mutant MT-618 FPN 3′UTR reporter compared to the wild type FPN 3′UTR reporter during iron deprivation under increasing concentrations ( 0–150 µM ) of DFE . We found that both the mutant and wild type FPN 3′UTR luciferase reporter activities were decreased under all tested DFE concentrations compared to empty vector control ( Figure 2F ) , however the MT-618 FPN 3′UTR reporter demonstrated significantly higher expression compared to wild type ( Figure 2F ) under 100 and 150 µM DFE , indicating that this miR-485-3p binding site is a significant contributor in the regulation of the FPN 3′UTR under iron deprivation . Collectively these studies identify FPN as a direct and physiologically relevant target of miR-485-3p . Next , we assessed the effect of miR-485-3p on endogenous FPN protein expression and intracellular iron regulation . A previously published FPN antibody [37] identified a ∼68 KDa protein with reduction in intensity following the silencing of FPN via pooled siRNAs ( Figure S3A ) . Enforced expression of miR-485 repressed endogenous FPN protein in both HepG2 ( Figure 3A ) and K562 ( Figure S3B ) cells and increased intracellular ferritin levels ( Figure 3B and Figure S3C ) . Transfection with increased concentrations of miR-485 led to dose-dependent increased miR-485-3p expression ( Figure 3C and Figure S3D ) and corresponding decreases in transferrin receptor ( TFRC ) mRNA levels ( Figure 3D and Figure S3E ) , consistent with an increase in cellular iron . Importantly , these changes occurred without significant changes in FPN mRNA levels ( Figure S3F ) . Since TFRC is a predicted target of miR-485-3p ( Figure S2A and S3G ) , we tested the potential regulatory relationship using reporter constructs with the wild type ( TFRC-3UTR-WT ) or with a mutated miR-485-3p binding site ( TFRC-3UTR-MT1937 ) . Co-transfection of miR-485-3p did not affect the reporter activities of either reporter constructs ( Figure S3H ) . Therefore , miR-485-mediated changes in TFRC mRNA are likely secondary to the changes in the cellular iron status instead of a result of direct regulation . The specific inhibition of miR-485-3p activity in HepG2 cells by AMOs led to significant and reproducible increase in FPN protein levels in response to iron depletion ( Figure 3E and Figure S3I–S3J ) . We demonstrate that these cells with loss of miR-485-3p function are in a greater state of iron deficiency , as evidenced by increased levels of IRP2 protein ( Figure 3E ) , increased TFRC mRNA expression ( Figure 3F ) , and decreased ferritin light chain ( FTL ) mRNA expression ( Figure 3G ) . The observed increase in the expression of ferroportin despite increased IRP2 protein level suggests that miR-485-3p activity is necessary for the response of FPN expression to iron depletion . This observation is consistent with the possibility that the microRNA activity of miR-485-3p plays an important role that is separate and distinct from the regulation by IRP2 . Collectively , these gain-of-function and loss-of-function data strongly support a role of miR-485-3p as an important post-transcriptional regulator of endogenous FPN expression . Finally we sought to mimic the regulation of endogenous FPN mRNA by constructing a luciferase reporter ( FPN-5UTR-LUC-3UTR ) containing both the FPN 5′UTR and FPN 3′UTR placed upstream and downstream of luciferase , respectively ( Figure 4A ) . We measured luciferase reporter activity of the FPN-5UTR-LUC-3UTR reporter compared to the FPN 3′UTR reporter during iron deprivation under increasing concentrations ( 0–150 µM ) of DFE ( Figure 4B ) . When normalized to control empty reporter under identical conditions , we found that although both reporters exhibited significantly decreased activities under all tested DFE concentrations , the effect of iron deprivation on the FPN-5UTR-LUC-3UTR reporter was slightly , but significantly more decreased under 50 , 100 , and 150 µM DFE compared with that of the FPN 3′UTR reporter . We used the FPN-5UTR-LUC-3UTR , FPN-5′UTR , and FPN 3′UTR reporters to further characterize the effect of both regulatory regions on post-transcriptional regulation of FPN in response to miR-485 overexpression ( Figure 4C–4E ) or inhibition of miR-485-3p activity ( Figure 4F–H ) . Enforced expression of miR-485 significantly inhibited both the FPN-5UTR-LUC-3UTR and FPN 3′UTR reporters , leading to 0 . 442 fold control ( ± . 027 , p<0 . 0001 ) and 0 . 639 fold control ( ± . 044 , p<0 . 0001 ) reporter activities , respectively ( Figure 4C–4D ) . Enforced expression of miR-485 significantly increased FPN 5′UTR activity ( 1 . 283 fold control ± . 022 , p = 0 . 0004 ) ( Figure 4E ) . Since the FPN 5′UTR contains the IRP binding site and does not have predicted miR-485 target sites , this increase likely reflects the increased endogenous cellular iron retention due to miR-485-mediated decreased endogenous FPN levels . Inhibition of miR-485-3p-mediated RISC activity by AMO-485-3p led to an increased activity of the FPN 5UTR-LUC-3UTR ( 1 . 488 fold control ± . 053 , p<0 . 0001 ) and FPN 3′UTR ( 1 . 282 fold control ± . 023 , p<0 . 0001 ) reporters compared to control inhibitor ( Figure 4F–4G ) . Treatment with AMO-485-3p led to significantly decreased FPN 5′UTR activity ( 0 . 638 fold control ± . 064 , p = 0 . 010 ) ( Figure 4H ) . With the AMO-485-3p-mediated potentiation of FPN levels and the subsequent continued export of iron , this decrease likely reflects the binding of IRPs in response to decreased endogenous cellular iron levels . In summary , we demonstrate the post-transcriptional regulation of FPN during the iron-deficient condition by miR-485-3p via the 3′UTR in addition to the well-recognized regulation by IRPs via the 5′UTR ( Figure 4I ) . These findings support a model that includes both IRPs and miR-485-3p as concurrent modulators of mRNA stability and translation in the post-transcriptional regulation of FPN expression ( Figure 4J ) and in the fine-tuning of cellular iron homeostasis .
Given the crucial role of FPN in iron metabolism , extensive regulation of FPN occurs at multiple levels , including the transcriptional [8] , [9] , [38] , post-transcriptional [10]–[12] , and post-translational ( hepcidin ) levels [13] , [14] , [16] , [39] . This study , for the first time , establishes the 3′UTR of FPN as an important regulatory region and miR-485-3p as a post-transcriptional regulator in response to iron deprivation to reduce FPN expression and iron export in the maintenance of cellular iron homeostasis . The discovery of iron-responsive microRNAs and microRNA-mediated regulation of FPN in several cell lines and primary macrophages illustrates the complexity of regulatory mechanisms for the precise and dynamic regulation of cellular iron . However , these findings were mainly obtained from the cellular response to varying iron levels in vitro . The use of primary macrophages and several cell types studied , and their longstanding use in this field offers a broad baseline and physiologically relevant context to indicate the potential relevance of individual microRNAs and their functional target ( s ) in cellular iron regulation . But it will be important to further establish the in vivo relevance of these findings using clinical samples from individuals with iron overload and iron deficiency conditions or receiving treatments to correct conditions of iron deficiency or overload . Several microRNAs have been found to regulate targets with key roles in iron homeostasis . The hypoxia-induced miR-210 is known to directly target ISCU1/2 , which play a role in the biogenesis and integrity of iron-sulfur clusters [40] . Repression of iron-sulfur clusters increases the functionality of IRP1 as an RNA-binding protein and indirectly alters IRP1-dependent regulation [40] , [41] . The liver-specific microRNA miR-122 is known to directly target Hjv and Hfe , both important for hepcidin expression , and has been shown to play an important role in the control of murine systemic iron homeostasis [28] . However , no studies to date have sought to identify a potential repertoire of microRNAs whose expression levels are associated with changes in cellular iron concentration in mammalian cells . Using an unbiased approach , we have identified iron-responsive microRNAs with predicted mRNA targets associated with cellular iron homeostasis . These microRNAs are expected to play an integral role in the cellular iron response . The regulation of FPN by microRNAs is likely to be distinct from other well-established mechanisms in several important ways . Unlike the systemic regulation of FPN by circulating hormone hepcidin , the monitoring of local iron levels by cellular microRNAs can lead to a more dynamic response to spatial and temporal fluctuations . While both microRNAs and IRPs are iron-responsive and target a group of mRNAs , they may also respond to different sets of non-iron environmental conditions and regulate distinct sets of target mRNAs to allow for diversity and fine-tuning of gene regulation . The targeting of FPN by microRNAs in the 3′ UTR allows for the possibility of iron-dependent regulation of subsets of FPN mRNAs known to lack the 5′ UTR [42] , [43] . Given that the specific composition of microRNAs can differ among cell types , distinct and coordinated responses of iron-responsive networks may exist within different cell types . Thus , it will be important to extend this study to other relevant cell types and validate the interaction of iron-responsive microRNAs with predicted iron-related targets . Additionally , high-throughput techniques to probe the microRNA-mRNA interactome [44] , [45] offer powerful complementary approaches to identify the in vivo target mRNAs associated with Ago2 during different iron states . Such exploration of iron-responsive microRNAs and their respective targets will lead to a more comprehensive pathway demonstrating an integrated role for microRNAs in the regulation of cellular iron homeostasis . Since FPN is known to be repressed by the IRP/IRE system under the iron-deficient condition , our findings suggest a potential cooperative relationship between RNA-binding proteins ( RBPs ) and microRNAs in the regulation of FPN . The cooperative contribution of RBPs , including both IRPs and the microRNA-guided RISC , to the post-transcriptional regulation of target RNAs constitutes a major regulatory layer of gene expression [46] , [47] . RBPs can function to promote or inhibit microRNA target availability and binding , leading to the enhancement or inhibition of mRNA stability and translation [27] . In the case of FPN , it is possible that the IRP/IRE 5′UTR interaction can be further stabilized and fine-tuned by the microRNA-mediated RISC on the 3′UTR to enable a more dynamic and fine-tuned expression over a wide range of iron conditions . Finally , microRNA-mediated regulation by miR-485-3p may offer a novel alternative means to target intracellular FPN and alter cellular iron status . Successful proof-of-concept studies supporting the use of therapeutic microRNA mimics have been demonstrated with microRNAs identified as functional tumor suppressors in mouse models of cancer [48]–[50] . Therapeutic inhibition of miR-122 with locked nucleic acid–modified oligonucleotides [51] has recently been shown to successfully lower hepatitis C virus ( HCV ) replication in chronically infected primate models and lead to long-lasting suppression of HCV viremia and improvement of HCV-induced liver pathology [52] . Manipulation of the miR-485-3p-FPN regulatory axis can potentially be used as a tool to bypass hepcidin deficiency or hepcidin resistance due to FPN gain-of-function mutations , mechanisms that lead to systemic iron overload pathology . Since FPN expression and cellular iron levels can control the growth of Salmonella [53] , [54] , the miR-485-3p-FPN relationship may also prove relevant for antimicrobial resistance strategies .
K562 cells were maintained in RPMI; 293 and HepG2 cells were maintained in DMEM . All cells were incubated in humidified atmosphere of 5% CO2 at 37°C and supplemented with 10% fetal bovine serum ( HyClone ) and 1% penicillin and streptomycin . Ferric ammonium citrate ( FAC ) and deferoxamine ( DFE ) were purchased from Sigma . For iron depletion , cells were treated with DFE ( 100 µM unless otherwise indicated ) diluted in PBS and added to the media for indicated time intervals ( 16–24 hours ) . For iron supplementation , cells were treated with FAC ( 500 nM ) diluted in PBS and added to the media for indicated time intervals ( 16–24 hours ) . Human peripheral blood mononuclear cells ( PBMC ) were isolated by Ficoll Paque ( GE healthcare ) density centrifugation from whole blood . Monocytes were enriched from freshly isolated PBMC through plastic adherence for 1–2 hours . To differentiate monocytes into macrophages , cells were plated into RPMI 1640 media with 2 mM glutamine ( Gibco ) containing 10% fetal bovine serum ( Hyclone ) , 100 µg/ml streptomycin , 100 U/ml penicillin , 1% Na-pyruvate , 1% NEAA ( Non-Essential Amino Acids ) and 50 ng/ml rHu M-CSF . Cells were allowed to differentiate over a course of seven days , and then treated . Luciferase reporters were constructed using the psi-CHECK2 vector ( Promega ) . The 5′UTR of FPN was amplified using primers ( forward: CAGCTAGCCCGACTCGGTATAAGAGCTG; reverse: CAGCTAGCAACAGGAGTGCAAGGAACTG ) and cloned upstream of Renilla luciferase to form FPN-5UTR-LUC . The 3′UTR of FPN amplified using primers ( forward: TTTAACTGTTGCTATCCTGTTACT; reverse: CCTTTTTACAAAGATTTTACAACATAG ) and cloned into the MCS downstream of Renilla luciferase ( FPN-3UTR-LUC ) . The 3′UTR of TFRC amplified using primers ( forward: GCAAAATGCATGCCCTGTA; reverse: AAGCATTGGGTGGGTAAATTC ) and cloned into the MCS downstream of Renilla luciferase ( TFRC-3UTR-LUC ) . Mutant reporters were constructed using primer-based overlapping PCR with the following primers: FPN3UTRmt448-F- TGCATCTTAGTTATTTTTAAAAACAAATTCTTCAAGTTAGAAGACTAAATTTTGATAACTAATATTATCCTTATTG , FPN3UTRmt448-R CAATAAGGATAATATTAGTTATCAAAATTTAGTCTTCTAACTTGAAGAATTTGTTTTTAAAAATAACTAAGATGCA , FPN3UTRmt618-F- ACATCAAGAGCTTCGTGGAG , FPN3UTRmt618-R- CTCGAGTTACAAAGATTTTACAACATAG , FPN3UTRmt618-mutF- GGCAACATATTTGTTAGAAGCA , FPN3UTRmt618-mutR- CTAACAAATATGTTGCCCCCATC , TFRC3UTRmt1937-F-GATGGTTCACTCACGGAGCTTCGAACTTATTGTAACCTACATTTAATTGATC , TFRC3UTRmt1937-R-GATCAATTAAATGTAGGTTACAATAAGTTCGAAGCTCCGTGAGTGAACCATC . All reporters were transfected into K562 and HepG2 cells using Lipofectamine 2000 or Lipofectamine LTX ( Invitrogen ) . Reporter assays were conducted using the Dual Luciferase Reporter Assay System ( Promega ) and the Tecan Infinite F200 reader according to manufacturer's protocol . Expression constructs encoding miR-485 ( pc-miR-485 ) and miR-194 ( pc-miR-194 ) were created by insertion into a cytomegalovirus-based pcDNA3 cloning vector ( Invitrogen ) using the following primers: pc-miR-485-forward:TCATGTGTGGTACTTGGAGA; pc-miR-485-reverse: AAAAGAAGTCAGCCATGTGT; pc-miR-194-forward: GAATTCCCATGATGAGCAAAAGGAATC; pc-miR-194-reverse: CTCGAGATCAAAAGTAACAGCATCTC; Antisense-2′O-methyl-modified nucleotides were purchased from ( Dharmacon ) . AMO-485-3p: AGAGAGGAGAGCCGTGTATGAC; AMO-CNTL1: AAGGCAAGCUGACCCUGAAGU; AMO-CNTL2: CCAUCUUUACCAGACAGUGUUA . Control ( Ambion ) and FPN siRNA ( SMARTpool from Dharmacon ) were transfected into K562 and HepG2 cells using nucleofection ( Amaxa ) and lipofection ( Invitrogen ) methods . Quantitative real-time RT-PCR ( qRT-PCR ) analysis of microRNA expression using TaqMan Low Density Arrays ( TLDA ) Human microRNA Panel ( Applied Biosystems ) was conducted according to manufacturer's instructions using the ABI 7500 real-time PCR system ( Applied Biosystems ) . The Ct data were obtained by the RQ Manager v1 . 2 software using automatic threshold settings and normalized to RNU48 endogenous control . MicroRNAs with a log2 expression change of at least 0 . 5 in either the iron-rich or iron-deficient condition when compared to baseline were considered to be iron-responsive . Individual qRT-PCR Taqman mature microRNA assays ( Applied Biosystems ) were used for validation of results from TLDA . qRT-PCR analysis of mRNA expression using Power SYBR Green ( Applied Biosystems ) was conducted as described previously [55] with primers specific for ferroportin , transferrin , and ferritin light chain , with beta-tubulin as an internal control . Western blots were performed as described in [55] using the following antibodies: IRP2 ( Santa Cruz Biotechnology ) , alpha tubulin ( Sigma ) , GAPDH ( Santa Cruz Biotechnology ) , anti-mouse IgG-HRP ( R&D Systems ) and anti-rabbit IgG-HRP ( R&D Systems ) . The FPN antibody was kindly provided by Dr . Tomasa Barrientos de Renshaw of the Andrews Laboratory as described [37] . Relevant protein band was identified by performing siRNA experiments using control ( Ambion ) or SMARTpool FPN-targeting siRNA ( Dhamarcon ) . Ferritin ELISA was performed using the Human Ferritin ELISA assay ( Abnova ) according to manufacturer's instructions . Statistical analyses were performed using the Student's t test . Results were considered statistically significant at a p-value <0 . 05 ( * ) , <0 . 01 ( ** ) , or < . 0001 ( *** ) . n . s = nonsignificant . Graphs were generated using Prism 5 software ( GraphPad software , Inc . ) QPCR-Beta tubulin-F- GCACATAGTAGGCGCTCAAT QPCR-Beta tubulin-R- ATCTGGAGACCCAGCTTCTT QPCR-FPN-F- GACATGAGCAAGAGCCTAC QPCR-FPN-R- AGGCTGGTTGTAGTAGGAGA QPCR-TFRC-F- AAAATCCGGTCTAGGCACAG QPCR-TFRC-R- CCTTTAAATGCAGGGACGAA QPCR-FTL-F- GGGTCTGTCTCTTGCTTCAAC QPCR-FTL-R- GGTTGGCAAGAAGGAGCTAA QPCR-GAPDH-F- AGCAAGAGCACAAGAGGAAG QPCR-GAPDH-R- GGTTGAGCACAGGGTACTTT
|
Cellular iron homeostasis is maintained by a sophisticated system that responds to iron levels and coordinates the expression of targets important for balancing iron export and uptake with intracellular storage and utilization . Ferroportin is the only known cellular iron exporter in mammalian cells and plays a critical role in both cellular and systemic iron balance . Thus the ability to regulate cellular iron export is of great interest in the search for therapeutic strategies to control dysregulated iron homeostasis , iron overload disorders , and conditions affected by cellular iron concentrations such as antimicrobial resistance . During iron deprivation , repression of ferroportin levels reduces iron export and preserves cellular iron . Ferroportin translation is known to be repressed by iron regulatory proteins that bind to the 5′UTR , yet alternative mechanisms that can post-transcriptionally regulate ferroportin have not been previously reported . Here , we find that miR-485-3p is induced during iron deficiency and represses ferroportin by directly targeting its 3′UTR , and further experimental evidence supports a model that includes both iron regulatory proteins and microRNAs as post-transcriptional regulators of ferroportin . These findings demonstrate a novel role for microRNAs in the cellular response to iron deficiency and can have therapeutic implications for various diseases of iron homeostasis .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"cellular",
"stress",
"responses",
"molecular",
"cell",
"biology",
"cell",
"biology",
"gene",
"regulation",
"genetics",
"molecular",
"genetics",
"biology",
"genetics",
"and",
"genomics"
] |
2013
|
Iron-Responsive miR-485-3p Regulates Cellular Iron Homeostasis by Targeting Ferroportin
|
The epidermis is the largest organ of the body for most animals , and the first line of defense against invading pathogens . A breach in the epidermal cell layer triggers a variety of localized responses that in favorable circumstances result in the repair of the wound . Many cellular and genetic responses must be limited to epidermal cells that are close to wounds , but how this is regulated is still poorly understood . The order and hierarchy of epidermal wound signaling factors are also still obscure . The Drosophila embryonic epidermis provides an excellent system to study genes that regulate wound healing processes . We have developed a variety of fluorescent reporters that provide a visible readout of wound-dependent transcriptional activation near epidermal wound sites . A large screen for mutants that alter the activity of these wound reporters has identified seven new genes required to activate or delimit wound-induced transcriptional responses to a narrow zone of cells surrounding wound sites . Among the genes required to delimit the spread of wound responses are Drosophila Flotillin-2 and Src42A , both of which are transcriptionally activated around wound sites . Flotillin-2 and constitutively active Src42A are also sufficient , when overexpressed at high levels , to inhibit wound-induced transcription in epidermal cells . One gene required to activate epidermal wound reporters encodes Dual oxidase , an enzyme that produces hydrogen peroxide . We also find that four biochemical treatments ( a serine protease , a Src kinase inhibitor , methyl-ß-cyclodextrin , and hydrogen peroxide ) are sufficient to globally activate epidermal wound response genes in Drosophila embryos . We explore the epistatic relationships among the factors that induce or delimit the spread of epidermal wound signals . Our results define new genetic functions that interact to instruct only a limited number of cells around puncture wounds to mount a transcriptional response , mediating local repair and regeneration .
The development of a specialized epidermal barrier layer represents a key step during the evolution of multi-cellular organisms . This outer integument provides protection from the environment and helps maintain cellular homeostasis . Epidermal barriers consist of epithelial cells that are tightly joined by adherens and other types of junctional complexes , as well an apical extracellular matrix layer that is highly variable . The mammalian epidermal barrier is constructed from a constantly renewing multicellular layer , in which cells follow a complex process of cell division and differentiation to form the stratum corneum [1] . In arthropods like Drosophila melanogaster , a single epidermal cell layer secretes a multilayered matrix of cross-linked lipid , protein , and chitin to generate a largely impermeable cuticle barrier [2] , [3] . Despite the great differences between the components and physical makeup of their epidermal barriers , both mammals and arthropods make use of conserved cellular mechanisms , transcriptional regulators , and signaling pathways during the generation of epidermal barriers as well as during their regeneration after wounding [4] , [5] , [6] , [7] . There are many complex processes that contribute to epidermal wound healing; these include clot formation , reepithelialization , cellular proliferation , inflammation , and barrier replacement [8] . Drosophila is a genetically tractable system for discovering evolutionarily conserved genes involved in such epidermal wound healing processes , as it has been for discovering genes that regulate animal septic wound responses [9] . One useful system for elucidating cellular mechanisms involved in wound healing has been Drosophila dorsal closure—where sheets of embryonic epidermal cells migrate to join at the dorsal midline—which uses some of the same cellular processes that are used to heal wounds [10] , [11] . For example , both dorsal closure and wound healing involve the recruitment of an actin-cytoskeleton “purse-string” to help close the edge of the wound or the edge of a gap in a migrating dorsal epidermal sheet [12] . Several evolutionarily conserved transcriptional regulatory pathways have been linked to developmental control of barrier formation as well as wound healing [13] . For example , Grainy head ( Grh ) transcription factors are required in a variety of animals for the development of impermeable epidermal barriers as well as normal wound repair [5] , [6] , [14] , [15] , [16] , [17] . In Drosophila , Grh accomplishes these functions in part through regulation of the Dopa Decarboxylase ( Ddc ) and Tyrosine hydroxylase ( ple ) genes , which encode enzymes that produce cuticle protein cross-linkers [5] , [18] . Other transcription factors with conserved roles in wound repair are those in the JUN family , which are required for wound reepithelialization in both mammals and Drosophila [7] , [19] . Upstream of JUN , the JUN amino-terminal kinase ( JNK ) signaling pathway [20] is required in the Drosophila epidermis for dorsal closure and wound reepithelialization [4] , [21] , [22] . misshapen ( msn ) , which encodes a Drosophila JNK-kinase-kinase-kinase , is distinctive because it is transcriptionally activated around embryonic , larval and adult epidermal wounds [4] , [21] , [22] , [23] , [24] , [25] . Recent reports have shown that JNK signaling is also required during Drosophila wing imaginal disc regeneration [26] , [27] . Another signaling pathway involving the gene stitcher ( stit ) , which encodes a receptor tyrosine kinase , is also activated around epidermal wound sites in Drosophila embryos , and is required for normal wound reepithelialization , and activation of some epidermal barrier repair genes [28] . There have been a few focused genetic screens for Drosophila mutants required for normal epidermal wound repair . One was a screen of 665 P-element insertional mutants for abnormal wound reepithelialization phenotypes after laser wounding during embryogenesis . The mutations with the most severe defects were in the genes for the JUN transcription factor and ßHeavy Spectrin [7] . Interestingly , wound closure defects were not observed in several genes that are required for dorsal closure and epithelial migration during Drosophila development , indicating that wound closure and dorsal closure are , to some extent , under the control of distinct genetic systems [7] . Another Drosophila genetic screen used a combination of dominant negative and RNAi-mediated knockdowns to test about 180 genes , focusing on Receptor Tyrosine Kinases ( RTKs ) , JNK signaling components , and cytoskeletal components after pinch or puncture wounding of the larval epidermis [22] , [29] . The knockdown or knockout of function in about 20 genes showed defective reepithelialization after larval wounding . Genes required for normal larval reepithelialization include those encoding components of the JNK signaling pathway like the transcription factors JUN ( Drosophila Jra ) and FOS ( Drosophila kay ) , as well as the Drosophila PDGF/VEGF-like receptor ( Pvr ) , and some proteins that regulate or remodel the actin cytoskeleton [22] , [29] . A few of the tested genes ( encoding JUN , JNK , and JUNKK , respectively ) were also required for the transcriptional activation of a wound response reporter gene ( misshapen-lacZ ) in larval epidermal cells surrounding sterile wound sites [22] . We have initiated a large , unbiased , genetic screen to identify mutations that are required for localized activation of epidermal genes around clean puncture wounds in Drosophila embryos . At this point , there are cis-regulatory wound enhancers identified from the Ddc , ple , msn , kkv , and stit genes [25] , [28] . These enhancers , when attached to fluorescent reporter genes ( hereafter called wound reporters ) provide a visible readout of wound-induced gene activation after epidermal wounding , and can be used to identify mutations that are required to activate or localize ( delimit ) this response . A few hours after wounding late stage embryos , fluorescent signal from these epidermal wound reporters can be observed in a zone that extends ∼5–10 cells from puncture sites . Some particularly interesting regulatory genes that we discuss in this paper are those required to delimit or localize the activity of wound reporters to a zone within a few cell diameters from wound sites . Mutations in such genes result in a global activation of wound reporters in most or all epidermal cells after wounding . One of the wound localization genes we identified is reggie-1/Flotillin-2 ( referred to as flotillin-2 or Flo-2 in Flybase ) , which was originally isolated as a gene that is activated in wounded , regenerating goldfish neurons [30] , and as a protein enriched in lipid rafts [31] . At the cellular level , Flo-2 appears to be involved in a variety of cell signaling and adhesion functions , at least in part via its role in clathrin-independent vesicular trafficking [32] , [33] , [34] , [35] , [36] . Analogous to wounded fish neurons , the Flo2 gene in Drosophila embryos is transcriptionally activated in cells surrounding epidermal wounds . We find that Flo-2 interacts in a pathway involving Drosophila Src42A to delimit epidermal wound responses , and that overexpression of either Flo-2 or activated Src42A can inhibit wound reporter activation , whether it is triggered locally by epidermal puncture , or globally by injection of trypsin or hydrogen peroxide .
Monitoring the activity of epidermal wound reporters in late stage Drosophila embryos [25] provides an in vivo assay that can be used to screen for mutations that are required to activate or localize the expression of genes that respond to epidermal wounding . We began such a screen using a collection of well-defined small chromosomal deletions of the Drosophila melanogaster genome [37] , searching for regions containing zygotic functions required for normal activation of the wound reporter ple-WE1 [25] . One advantage of this screen is that most Drosophila zygotic mutants survive to late stages of embryogenesis , differentiate their epidermis ( which can be assayed by the activation of an anal pad specific enhancer that exists alongside the wound enhancer within the ple-WE1 sequence ) , and can still be assayed for wound reporter activity . At this point , we have screened 300 deletions that include approximately 4 , 600 genes on the X and 2nd chromosomes ( Figure 1A ) . Sixteen of these deletions had abnormal ple-WE1 expression , and therefore contained putative epidermal wound regulatory genes . Analysis of the genes within the collection of deletions indicates that the zygotic functions of many signaling pathways , transcription factors , and other regulators of cellular properties have no effect on the activation of the ple-WE1 epidermal wound reporter ( Table 1 ) . For example , deletion mutants of patched ( Hedgehog pathway ) , shaggy/GSK3 or disheveled ( Wingless pathway ) , Notch ( Notch signaling pathway ) , Pvr ( PDGF/VEGF signaling pathway ) , domeless ( JAK/STAT pathway ) , and wengen ( TNF pathway ) all showed normal ple-WE1 expression after wounding . It is possible that the maternal contribution of some of these genes is sufficient to rescue potential effects on wound reporter activation in zygotic mutants . Within the 16 regions required for normal epidermal wound reporter expression , we have currently defined 7 single gene mutations with new functions in wound reporter activation or localization ( Figure 1B ) . The functions of two of these genes ( ghost/stenosis , and Duox ) are required for wound gene activation , and five of the genes ( flotillin-2 , wurst , varicose , Src42A , and the Drosophila homolog of yeast MAK3 ) are required to localize wound reporter activity to the immediate vicinity of wounds . One focus of this paper is on flotillin-2 ( Flo-2 ) . The Flo-2 protein has been well characterized at the cellular and biochemical levels , but the genetic interactions of Flo-2 are still enigmatic in the diverse cellular processes in which it participates [38] . The Drosophila genome encodes only one Flo-2 ortholog [39] . Null mutants that eliminate Drosophila Flo-2 protein also accumulate little or none of the related Flo-1 protein , since it is apparently destabilized in the absence of Flo-2 [40] . Flo-2 mutant animals have normal morphology and are viable and fertile [40] . Despite having normal adult morphology , Flo-2 mutants show a reduced spread of Wnt and Hedgehog signals in wing imaginal discs [40] , [41] . Flo-1 mouse mutants are viable and fertile under standard lab conditions and the mutants have somewhat reduced Flo-2 protein levels [42] . We tested a Drosophila deletion mutant that eliminated the Flo-1 gene , but ple-WE1 expression was normal after wounding ( Table 1 ) . Normal epidermal activation of our fluorescent protein wound reporters can be easily detected 4 hours after wounding wild type stage 16 embryos ( Figure 2A , 2B ) , although transcripts from the same endogenous wound-activated genes in a narrower zone of cells can be detected in fixed embryos within 30 minutes after wounding [5] , [25] . The fluorescent protein reporters have the great advantage of being easily detectable in living embryos or larvae , whereas nucleic acid or antiserum probe permeability into late stage embryos or larvae with partially or fully differentiated cuticle is labor intensive at best . However , the fluorescent protein reporters typically represent a delayed version ( by our estimate , a few hour delay ) of the transcriptional response in cells surrounding wounds . This is due to the requirement of enhanced-GFP or enhanced-dsRed proteins to oxidatively mature to fluorescence . We estimate that the fluorescent reporter proteins we used [5] , [25] , [43] . have a half-time to maturation of about an hour in fly embryos after the reporter gene RNAs are translated . Additionally , time is required to accumulate detectable levels of fluorescent protein , and this is dependent on the strength of the epidermal wound enhancer being tested . For example , the Ddc . 47 wound enhancer appears to be slightly stronger than the ple-WE1 enhancer [5] , [25] . In Flo-2 mutants , fluorescent wound reporter proteins driven by the ple-WE1 and Ddc . 47 wound enhancers are detected at 4 hours after wounding , but reporter expression is significantly expanded compared to wild type embryos ( Figure 2A , 2B ) . By 6 hours post-wounding reporter expression in Flo-2 mutants spreads to include most epidermal cells in late embryos ( Figure 2A , 2B ) . Using in situ hybridization or protein immunodetection , we also tested the effect of Flo-2 mutants on the activation pattern of several endogenous wound response genes ( ple , Ddc , msn , stit , and Src42A ) , and found comparable expanded expression domains after wound induction ( Figure S1 , and data not shown ) . Nearly all ( ∼90% ) of wounded Flo-2 mutant embryos survive , hatch to become larvae , and progress to adulthood , which was similar to that observed in wounded wild type embryos ( data not shown ) . We did not detect any characteristics of abnormal epidermal wound healing [26] , or ectopic melanization in the punctured Flo-2 mutants , although we did not carefully examine the kinetics of healing in the mutants . Activation of wound reporters in Flo-2 mutants was wound-dependent; in no instances was constitutive expression of reporters detected in mutant embryos . Flo-2 transcripts normally accumulate in all cells of the embryo , including the epidermis ( Figure 3A ) , with higher apparent levels in the central nervous system [40] , [41] . The insertional mutant into Flo-2 ( Flo-2[KG00210] ) used in this study fails to accumulate transcripts ( Figure S1A; [40] ) . A 500 bp region within the largest of the Flo-2 introns contains predicted high affinity binding sites for the Grh , AP-1 , and ETS transcription factors . Such sites are found in clusters in previously characterized epidermal wound response enhancers from the ple , Ddc , msn , and stit genes [25] , [28] . To test whether Flo-2 is transcriptionally activated around epidermal wounds , we carried out in situ hybridizations on wounded embryos using a Flo-2 probe . As seen in Figure 3B , Flo-2 transcripts are expressed at higher levels in cells surrounding epidermal wound sites , overlapping with the wound induced activation of the endogenous Ddc gene . Previous studies showed the activation of some epidermal wound response genes depends on the function of the Grh transcription factor [25] , [28] . As seen in Figure 3C , Flo-2 transcriptional activation around epidermal wound sites is dramatically reduced in grh mutant embryos compared to wild type siblings ( Figure 3B ) , consistent with Flo-2 being activated in a Grh-dependent manner . With Flo-2 loss of expression resulting in widespread activation of epidermal wound genes , we wished to test whether overexpression of Flo-2 would have an effect on wound gene activation . This was accomplished using a fly line in which the Flo-2 cDNA was fused to a UAS promoter , in combination with an arm-GAL4 driver [44] . armadillo ( arm ) , the Drosophila homolog of ß-catenin , is expressed ubiquitously in embryos , and the arm-GAL4 driver can induce high levels of UAS-Flo-2 expression beginning at stage 10 of embryonic development ( Figure S2A , [45] ) . Ubiquitous high levels of Flo-2 inhibit the activation of the Ddc . 47 and ple-WE1 epidermal wound reporters around wound sites ( Figure 4A , Figure S3A ) . To test whether the inhibition caused by overexpression of Flo-2 was cell autonomous , we used en-GAL4 to drive high levels of Flo-2 in the engrailed ( en ) , posterior compartment of each embryonic segment ( Figure S2B , [46] ) . Overexpression of Flo-2 using en-GAL4 is sufficient to silence the activation of the Ddc . 47 and ple-WE1 epidermal wound reporters in all cells near a wound , even in those that do not produce Flo-2 at higher levels ( Figure 4C , data not shown ) . The lack of any activation of the epidermal wound reporter in the en>Flo-2 overexpression experiments suggests that Flo-2 can act cell non-autonomously , at least at short range , to inhibit the ability of cells to respond to wound signals . Higher levels of Flo-2 do not appear to be toxic , as overexpression of Flo-2 with either the arm-GAL4 or the en-GAL4 drivers does not obviously alter embryonic development , and animals so treated survive to produce viable and fertile adults ( data not shown ) . Although we do not know the signaling mechanisms that allow cells 5–10 cell diameters from a wound site to sense the presence of an epidermal break and activate wound gene transcription , one system that might be involved is an activation of serine protease cascades . Serine proteases regulate the production of some localized developmental signals [47] , infectious innate immune signals [9] , and activate localized melanization around wounds [48] . We tested whether a serine protease , trypsin , would be sufficient to induce widespread activation of epidermal wound reporter genes when injected into late stage embryos . Injection of trypsin into the body cavity ( or into the perivitelline space ) of stage 16 embryos results in a global activation of Ddc . 47 and ple-WE1 epidermal wound reporters ( Figure 4B , Figure S3B ) . This trypsin treatment does not appear to result in widespread epidermal cell death , nor is the epidermal paracellular barrier—which prevents diffusion of all but very small molecules through epithelia—breached when trypsin is injected into the perivitelline space of stage 16 embryos ( R . P . , unpublished results ) . Although trypsin is sufficient to activate the ple and Ddc wound reporter genes , as yet we have no current evidence that a specific endogenous serine protease is required to activate epidermal wound-induced transcriptional responses . Strikingly , overexpression of Flo-2 under the control of either arm-GAL4 or en-GAL4 is sufficient to inhibit trypsin-induced activation of Ddc . 47 or ple-WE1 wound reporters throughout the entire embryonic epidermis ( Figure 4B , 4D; Figure S3C , S3D ) . This finding suggests that puncture-induced and trypsin-induced activation of wound genes might act through a common pathway that can be inhibited by overexpression of Flo-2 . The inhibition of protease-induced wound reporter gene activation observed with en>Flo-2 overexpression is consistent with the idea that Flo-2 can act cell non-autonomously . Similar to Flo-2 mutants , mutants in Drosophila Src42A show more widespread activation of the ple wound reporter or Ddc transcription in epidermal cells after localized punctures ( Figure 5A , 5B ) . Since previous research has uncovered functional and biochemical interactions between Flotillins and Src family tyrosine kinases [33] , [49] , we decided next to focus on the relationships between Flo-2 and Src42A in the regulation of epidermal wound response genes . Src42A is the Drosophila homolog of vertebrate c-Src [50] . Src family kinases were found to play important roles in several signaling pathways [51] . Like Flo-2 , Src42A is itself a wound response gene; Scr42A transcripts accumulate to high levels in cells surrounding wound sites in wild type embryos ( Figure 5C ) . Src42A transcription is also globally activated in all epidermal cells in wounded Flo-2 mutant embryos , although the converse is not true , as Flo-2 transcript levels are unchanged in Scr42A mutants ( data not shown ) . In this sense at least , Flo-2 wound-dependent transcriptional activation is not behaving as other wound-induced genes like Ddc and ple , which show widespread wound-induced transcription in Src42A mutants . Similar to Flo-2 , overexpression of Src42A . CA ( a constitutively activated form , [52] ) with arm-GAL4 inhibits both the local puncture , as well as trypsin-induced , activation of the Ddc . 47 and ple-WE1 epidermal wound reporters ( Figure 5D , 5F; data not shown ) . In contrast to Flo-2 , overexpression of Src42A . CA in stripes using en-GAL4 inhibits the trypsin-induced activation of the ple-WE1 and Ddc . 47 epidermal wound reporters only in the cells where en>Src42A . CA is over-expressed ( Figure 5E , 5G ) . Thus , overexpressed Src42A . CA acts cell autonomously to inhibit epidermal wound reporter activity . Embryos with deletion mutations that eliminate other Src family tyrosine kinases , e . g . Btk29A ( Tec homolog ) , shark ( Syk homolog ) , hopscotch ( JAK homolog ) , and minibrain ( DYRK homolog ) all had normal , localized ple-WE1 wound reporter activity ( Table 1 ) . This suggests that the Src42A inhibition of epidermal wound reporter activity is specific , and not a general property of Src family tyrosine kinases . To test whether chemical inhibition of Src function would result in widespread wound reporter activation , we simultaneously wounded and injected one such chemical inhibitor , SU6656 [53] , into the body cavity . This treatment induced widespread , patchy expression of the Ddc . 47 and ple-WE1 wound reporters throughout the embryonic epidermis ( Figure S4A , S4B ) . This widespread activation of wound reporter genes after inhibition of Src kinase function was not suppressed by overexpression of Flo-2 using the arm-GAL4 driver ( Figure S4A , S4B ) . Flo-2 is associated with , and may stabilize the Flotillin-dependent fraction of lipid rafts in membranes [33] . Therefore , we were prompted to test whether chemicals that disrupt lipid rafts might influence the spread of wound reporter activation in the epidermis . One chemical that inhibits lipid raft formation is methyl-ß-cyclodextrin ( MßCD ) , which depletes cholesterol ( and other similar lipids ) from cell membranes , and can influence many intracellular signaling pathways [54] , [55] , [56] . Simultaneous wounding and injection of MßCD into the body cavity of Drosophila embryos is sufficient to activate the Ddc . 47 and ple-WE1 wound reporters throughout the entire embryonic epidermis ( Figure 6A and data not shown ) . This result suggests that the effect of reducing functional Flo-2 on the wound response can be mimicked by more severe disruptions in lipid raft organization and membrane composition . One wound-induced signal that can function to attract blood cells to the site of clean wounds in zebrafish and Drosophila embryos is Hydrogen Peroxide ( H2O2 ) [57] , [58] , [59] . We wished to test whether injection of H2O2 into the body cavity of Drosophila embryos was sufficient to activate the Ddc . 47 and ple-WE1 epidermal wound reporters in embryos , and found that a wide range of concentrations of H2O2 could activate the Ddc . 47 and ple-WE1 wound reporters in most or all epidermal cells ( Figure 6B , and data not shown ) . Interestingly , Flo-2 or Src42A . CA overexpression is sufficient to inhibit the both MßCD and H2O2 activation of the Ddc . 47 and ple-WE1 epidermal wound reporters ( Figure 6C , 6D; Figure S4C , S4D , and data not shown ) . These results suggest that high levels of Flo-2 or Src42A . CA are potent inhibitors of chemically-induced transcriptional activation of these epidermal wound reporters . To determine whether H2O2 production was required for the induction of epidermal wound reporters , we tested a Drosophila mutant in the gene for the Dual oxidase protein ( Duox ) , the enzyme responsible for the production of H2O2 , [57] . We found that Duox mutant embryos show a dramatically decreased activation of the ple-WE1 epidermal wound reporter ( Figure 6E ) . However , in the Duox mutant background , trypsin injection is still sufficient to activate the ple-WE1 reporter throughout the epidermis ( Figure 6F ) , suggesting H2O2-induced wound signaling might be upstream of , or in parallel to , a serine protease-dependent activation of epidermal wound reporters .
Drosophila melanogaster wound healing is an example of a regenerative process , which requires localized epidermal cytoskeletal changes , and localized wound-induced changes in epidermal transcriptional activity [60] . Our genetic screen with wound-dependent reporters has allowed us to identify novel components that regulate the localized transcriptional response to wounding in epidermal cells . This research identifies seven genes that are required to either activate ( Duox and ghost/stenosis ) or localize ( Flo-2 , Src42A , wurst , varicose , and Drosophila homolog of yeast Mak3 ) the expression patterns of epidermal wound reporters . The number of new functions involved in the delimitation of epidermal wound response near wound sites was unexpected , but indicates that considerable genetic effort is devoted to localizing the activity of transcriptional wound responses during regeneration . One of the genes that limits the spread of epidermal wound reporters after clean epidermal punctures is Flo-2 , as mutants of this gene show a broad expansion of epidermal wound gene activation . Drosophila Flo-2 is itself transcriptionally activated around epidermal wound sites , consistent with an evolutionarily conserved role in regeneration after wounding . In vertebrates , reggie-1/Flo-2 gene expression is activated in wounded fish optic neurons [30] , and reggie-1/Flo-2 and reggie-2/Flo-1 morpholino knockdowns in wounded zebrafish retinal explants reduced axon outgrowth compared to controls [61] . Flo-2 transcriptional activation around Drosophila epidermal wound sites is dependent on the grh genetic function , which is required to activate at least a few other epidermal wound response genes [5] , [25] , [28] . Flo-2 thus appears to act in the same pathway as grh , although it may act both downstream and upstream of grh , since overexpression of Flo-2 can inhibit the activation of other grh-dependent wound response genes . In this respect , Flo-2 resembles the stit receptor tyrosine kinase gene , which is both transcriptionally activated by Grh , as well as required for grh-dependent activation of other downstream wound genes [28] . Amazingly , overexpression of Flo-2 can even inhibit the global activation of the Ddc . 47 and ple-WE1 wound reporters that are induced by the serine protease trypsin , or by hydrogen peroxide . The inhibitory function of overexpression of Flo-2 on wound induced transcription is cell non-autonomous , at least over the range of a few cell diameters , as shown by the ability of striped overexpression of Flo-2 to silence puncture or trypsin-induced gene activation throughout the epidermis . The only animal where Flo-2 null mutants have so far been characterized is Drosophila , where Flo-2 has been shown to regulate the spread of Wingless ( Wg ) and Hedgehog ( Hh ) signals in the wing imaginal discs [40] , [41] . In the wing discs , both the secretion rate and the diffusion rate of these two lipid-modified morphogens were increased when Flo-2 was overexpressed , and decreased when Flo-2 and Flo-1 proteins were not expressed [41] . Despite the reduced spread of Wg and Hh morphogen proteins in Flo-2 mutant imaginal discs , adult morphology of mutants was normal , presumably because of compensatory mechanisms that occur later in development . Whereas a reduced range of activation of wg and hh long range transcription target genes was observed in Flo-2 mutant imaginal discs , we observe a greatly increased range of wound-induced gene activation in Flo-2 mutant embryos . This apparent discrepancy could be explained if one invokes of a long-range wound-induced inhibitory signal that in wild type embryos diffuses faster and farther than a wound activating signal , and thereby functions to limit the wound response to nearby epidermal cells [62] , and that in Flo-2 mutants this potential inhibitory signal has reduced secretion , concentration , and/or diffusion range . This notion is consistent with the cell non-autonomous effect of overexpressed Flo-2 on inhibiting wound- or trypsin-induced gene activation . A similar scheme of controlling signal spreading has been seen in the way that Mmp2 acts cell non-autonomously to limit FGF signaling during Drosophila tracheal development and branch morphogenesis [63] . It's also possible that Flo-2 normally is required to set a global threshold that wound-induced signals must overcome in order to activate wound transcription , for example via Flo-2-dependent endocytosis/degradation of a diffusible wound signal and its receptor ( perhaps the Stit RTK [28] ) , and that signal strength normally surpasses the Flo-2 threshold only in the vicinity of a wound . In this model , loss of Flo-2 would result in all epidermal cells being able to exceed the wound signal threshold , and overexpression of Flo-2 would prevent any cells from exceeding the wound signal threshold . The cell non-autonomous effects of Flo-2 overexpression under this model might be explained by an increase in Flo-2-dependent endocytosis/degradation that rapidly depletes an activating signal from the extracellular space . Many previous studies have documented biochemical , molecular biological , and cell biological interactions between Src family kinases and Flotillins [33] , [34] , [49] , [64] . In Drosophila , lack of Src42A and Flo-2 leads to expanded spread of wound gene activation , and overexpression of Flo-2 or activated Src42A can inhibit wound gene activation , which is consistent with an interaction between the two functions during the process of wound gene regulation . In cultured mammalian cells , Flo-2 can be phosphorylated by Src family kinases in an extracellular signal-dependent fashion . This phosphorylation is associated with changes in the normal intracellular trafficking of Flotillin-containing membrane microdomains and vesicles [33] , [35] , [49] . Since overexpressed Flo-2 in Drosophila can act in a cell non-autonomous fashion to inhibit wound gene activation , and overexpressed Src42A acts in a cell autonomous fashion to inhibit wound gene activation , one interpretation is that Flo-2 lies genetically upstream of Src42A in the epidermal wound response . This hypothesis appears to be inconsistent with the vertebrate biochemical data indicating that Src kinases phosphorylate Flotillins to activate their diverse functions . However , an observation that is consistent with Src42A activating Flo-2 protein function , is that even when Flo-2 is overexpressed , addition of chemical inhibitors of Src family kinases to wounded embryos , results in widespread Ddc . 47 or ple-WE1 wound reporter activation . One interpretation of this results Flo-2 protein , no matter the level of expression , is inactive in the absence of Src42A function . Complex feedback loops involving signaling proteins being regulated by a transcription factor , while the activity of the same transcription factors is regulated by the same signaling pathway , have been observed in the control of Drosophila epidermal wound gene expression and reepithelialization [22] , [28] , so there may be similar dynamic cross-regulatory interactions between Flo-2 and Src42A in the localization of the epidermal wound response , interactions not easily captured in linear genetic pathway diagrams [65] . The inhibitory effect of Src42A on wound gene activation suggests that it might antagonize a signaling cascade that leads to the epidermal wound response . A good candidate for such a signaling cascade is the RTK pathway involving the Stit kinase . Stit is a RET-family RTK that is required for robust activation of the Ddc and stit wound reporter genes in wounded embryos [28] . Other evidence consistent with RTK pathway importance in wound gene activation is that phosphotyrosine accumulates persistently around wound sites [5] , [28] , and that ERK kinase function is required for robust activation of the Ddc wound reporter gene [5] . Interestingly , Src42A has been shown to act as an inhibitor of some Drosophila RTK proteins ( those encoded by the torso , Egfr , and sevenless genes ) in a few different tissues during Drosophila development [66] . The Flo-2 and Src42A functions in epidermal wound localization after clean wounding are reminiscent of the role of Drosophila WntD during infectious wounding . WntD mutants show higher levels of some antimicrobial peptide genes after septic injury of adults [67] . Previous evidence suggested that H2O2 and Duox could provide wound-induced inflammatory signals and antimicrobial activities [58] , [59] , [68] , [69] , [70] , [71] . Our studies show that Duox is required to activate wound reporter genes after epidermal wounding , and that injected exogenous H2O2 is sufficient to activate widespread epidermal wound gene expression . Overexpression of either Flo-2 or Src42A . CA can inhibit the H2O2 -dependent wound reporter expression , suggesting that all of these components are in a common pathway controlling the activation of epidermal wound reporters . However , the ability of trypsin injection to activate the Ddc . 47 and ple-WE1 wound reporters in Duox mutants suggests that a serine protease might act downstream of , or in parallel to , H2O2-dependent wound signals . A recent report showed that in cultured mammalian cells , a Src kinase phosphorylates and inhibits a Flo-2-associated enzyme , peroxiredoxin-1 , which results in increased stability of H2O2 [72] . This is consistent with our results placing Flo-2 , Src42A , and H2O2 in a common wound signaling pathway . Like H2O2 , the injection of methyl-ß-cyclodextrin ( MßCD ) into wounded embryos triggers a global wound response in the epidermis . MßCD strongly depletes cholesterol and other sterols from membranes and disrupt lipid rafts [73] , [74] , but was also shown to remove sphingolipid-associated proteins such as Src-Family Kinases [54] . The effects of MßCD , in combination with the effects of loss of Flo-2 , suggests that the integrity of lipid rafts and associated proteins are required to inhibit epidermal wound signals . In cultured cells , MßCD treatments trigger a release of EGF receptors from membrane microdomains , which increases EGFR , and perhaps other RTK , signaling in a ligand-independent manner [75] . Interestingly , in cultured keratinocytes , MßCD treatment can induce the expression of involucrin [76] , which encodes a protein , analogous to Drosophila Ple/tyrosine hydroxylase , which is required for the formation of an epidermal barrier . Similarly , MßCD injections into Drosophila embryos might also cause an increase the levels of a wound signal produced or released from cells adjacent to the wound site , allowing more widespread transcriptional activation of wound reporter genes . Our observations that overexpression of Src42A or Flo-2 can inhibit the MßCD -triggered activation of epidermal wound reporter genes suggest that high levels of these proteins might overcome lipid raft-inhibitory effects on wound signaling pathways . Other genes ( wurst and varicose ) identified in the screen have phenotypes similar to Flo-2 and Src42A mutants ( Figure 1 ) . wurst encodes an evolutionarily conserved trans-membrane protein , containing a heat shock cognate protein 70 binding domain and a clathrin binding motif [77] . wurst is ubiquitously expressed in embryonic epithelial cells , strongly up-regulated during endocytosis-dependent luminal clearance , and mislocalized in mutants with endocytosis defects [77] . wurst mutant embryos have tortuous tracheal tubes , due to a failure to properly endocytose matrix material from the tracheal lumen [77] . varicose encodes an evolutionarily-conserved septate junction scaffolding protein , in the Membrane Associated GUanylate Kinase ( MAGUK ) family [78] , [79] , [80] . varicose is expressed in epidermally-derived cells ( including the hindgut and trachea ) and co-localizes with the septate junction proteins , Coracle and Neurexin4 [80] . varicose mutant embryos develop permeable tracheal tubes and paracellular barrier defects in epithelia [79] , [80] . Like wurst mutants , varicose mutants also have abnormal matrix composition in the tracheal lumen , and may also have abnormal extracellular matrix composition produced by other epidermal cells . Another gene ( ghost , also known as stenosis ) identified in this screen is required for wound reporter activation like Duox or grh ( Figure 1 ) . ghost encodes the Drosophila Sec24CD homolog , a coat protein of COPII vesicles in the ER/Golgi trafficking pathway [81] , [82] . Transport of cargo from the ER to the Golgi via COPII vesicles is required to achieve normal amounts of secretion of extracellular matrix proteins into the developing Drosophila tracheae and normal apical-basal localization of membrane proteins [81] , [82] , [83] . Presumably , similar secretion and membrane localization defects occur in non-tracheal epidermal cells , which account for the severe cuticle deposition defects in ghost ( Sec24CD ) mutants . It is fascinating to note that our finding that ghost ( Sec24CD ) is required for transcriptional activation of epidermal wound reporter genes is consistent with the finding that RNAi knockdowns of Sec24C in a planaria ( Schmidtea mediterranea ) interfered with normal regeneration after amputation wounds [84] . It is possible that the ghost mutants do not secrete enough wound signals , or the protein matrix necessary for the propagation of a wound signal . Another gene required for the activation of wound reporters is shroud ( sro ) . Based on a previous paper by Giesen et al . ( 2003 ) [85] , we believed sro to be an allele in the Drosophila Fos-D isoform [25] , and hypothesized that one of the Drosophila kayak/Fos transcription factors was required for the activation of some epidermal wound gene reporters [25] . However , as Niwa et al . ( 2010 ) [86] recently discovered , sro[1] and other sro point mutant alleles do not map in the kayak/Fos gene , but in an immediately adjacent transcription unit ( Nm-g/sro ) that encodes an enzyme in the sterol metabolic pathway that is necessary for production of ecdysone hormone . At first glance , the requirement of sro to activate some wound reporters suggested that these reporters rely on ecdysone signaling . This is possible , although we have tested deletions that eliminate zygotic functions of the ecdysone receptor gene , as well as of the phantom gene ( which encodes another enzyme in the ecdysone synthesis pathway ) , and embryos that are zygotic mutants in either gene show normal activation of the ple-WE1 wound reporter after puncture wounding . In summary , from our large unbiased screen , we have identified several genes that add to our understanding of the complex pathways that control the signals that activate wound response transcription near puncture wounds . At the cellular level , there appears to be a correlation between genetic functions required to localize wound-induced gene activation , and cellular functions required for endocytosis and/or apical-basal polarity . For example , one function of Flo-2 is in signal-dependent endocytosis , although Flo-2 also plays other roles in vesicular trafficking [32] , [33] , [34] , [35] , [36] . There have been many studies showing that endocytosis can regulate extracellular signaling strength and duration [87] . For example , one study found that tagged-FGF8 showed increased accumulation , spread , and target gene activation when Rab-5-mediated endocytosis was reduced in zebrafish embryos [88] . We believe that further studies on wound response signaling may provide new insights into how membrane microdomains , endocytosis of membrane receptors , and the composition and organization of the extracellular matrix , regulates the transmission of wound signals .
Fluorescent Balancers , Deficiencies , and Mutant alleles were obtained from the Bloomington Drosophila Stock Center: FKG = FM7c , P{GAL4-Kr . C}DC1 , P{UAS-GFP . S65T}DC5 , CKG = CyO , P{GAL4-Kr . C}DC3 , P{UAS-GFP . S65T}DC7 , Flo-2{KG00210} , Src42A-E1 , UAS-Src . CA , arm-GAL4 , en-GAL4 , Duox{KG07745} , wurst{G814} , varicose{03953b} , and ghost{KG029061} . UAS-Flo-2 was provided by Vladimir Katanaev . Ddc . 47 and ple-WE1 were previously described [25] . Embryos were collected on apple juice agar plates and aged to 15–17 h at 25°C . Embryos were washed into mesh baskets , dechorionated in bleach for 1 min , then washed copiously with water . Embryos were then transferred to a clean slab of apple juice agar and aligned for 30–60 min at 18°C , transferred to slides with double-sided tape , then covered in a 1∶1 ratio of 700∶27 weight halocarbon oil . Embryos were then wounded bilaterally with fresh microinjection needles made from an automated puller mounted on a micromanipulator , allowed to recover for 3–8 h at room temperature , and visualized under fluorescent light in a compound microscope to determine wound reporter activity . At least 3 independent experiments with at least 50 successfully wounded embryos were performed . Assays involving homozygous deletion or mutant embryos were performed in parallel to heterozygous-balancer embryos . A Kr-GFP fluorescent marker on the balancer chromosome [89] , was used to determine the genotype of the embryos . Assays involving UAS-GAL4 overexpression were performed in parallel to UAS-non-GAL4 controls . All embryos were impaled using a micromanipulator so that the needle protruded 1 embryo-width from the exit wound . Wound reporter responses were rated on a scale of “no activity , localized activity , or global activity . ” Images were obtained by wounding embryos with microinjection needles and imaged on a Leica SP2 confocal microscope , selecting representative embryos to image . Images were resized while constraining proportions to maintain resolution . Adobe Photoshop adjustment functions were used equally on images to enhance clarity , but not to obscure , eliminate , or misrepresent any information . Original images are available on request . Individual embryos were simultaneously wounded and injected by using a syringe to expel the various solutions into the body cavity of the embryo . A Pipetman was used to load the solutions to be injected into the pulled capillary microinjection needles . Needles were broken on the side of a glass cover slip on a glass slide . Serine Protease-Trypsin from bovine pancreas was solubilized in 1 mM HCl pH 3 . 0 to 2 mg/mL ( Sigma ) . Src Inhibitor-SU6656 was solubilized in 50% DMSO to 100 µM ( Calbiochem ) . Methyl-ß-cyclodextrin ( MßCD ) was solubilized in 1 mM NaOH to 3 mM ( Sigma ) . Hydrogen Peroxide-H2O2 was diluted in H2O to 0 . 6 M ( Fisher ) . Chemical-wounded embryos were simultaneously wounded and injected with a 1∶4 ratio of 1% toluidine blue dye and solubilized compounds . Toluidine blue dye allowed for visual confirmation of solubilized compounds being injected into the body cavity . Control embryos were wounded with a broken needle containing 1∶4 ratios of 1% toluidine blue dye and solute without chemical . A wide range of chemical concentrations was tested to obtain optimal activation of the epidermal wound reporter and maintain high levels of embryo survival after body cavity injection . Probes were generated from partial or full cDNA clones from the Drosophila Gene Collection [90] , [91] . anti-Stitcher antibody was provided by Christos Samakovilis . Probe labeling and hybridization protocol was as described in Dave Kosman's multiplex FISH protocol [92] .
|
An epidermal wound provides signals that initiate a variety of localized responses , some of which act to regenerate and repair the breach in the epidermal barrier . The Drosophila melanogaster embryonic epidermis provides an excellent system to discover new genes that regulate wound-healing processes . Using fluorescent epidermal “wound” reporters that are locally activated around wound sites , we have screened almost 5 , 000 Drosophila mutants for functions required to activate or delimit wound-induced transcriptional responses to a local zone of epidermal cells . Among the seven new genes required to delimit the spread of wound responses are Flotillin-2 and Src42A . These two genes are also sufficient , when overexpressed at high levels , to inhibit wound-induced transcription in epidermal cells . One new gene required to activate epidermal wound reporters encodes Dual oxidase , an enzyme that produces hydrogen peroxide . We also find that four biochemical treatments ( a serine protease , a Src kinase inhibitor , methyl-ß-cyclodextrin , and hydrogen peroxide ) are sufficient to globally activate epidermal wound response genes in Drosophila embryos . Our results define new genetic functions , and the interactions among them , which regulate the local transcriptional response to puncture wounds .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"model",
"organisms",
"genetics",
"biology",
"molecular",
"cell",
"biology",
"genetics",
"and",
"genomics"
] |
2011
|
Duox, Flotillin-2, and Src42A Are Required to Activate or Delimit the Spread of the Transcriptional Response to Epidermal Wounds in Drosophila
|
Malaria in pregnancy remains a substantial public health problem in malaria-endemic areas with detrimental outcomes for both the mother and the foetus . The placental changes that lead to some of these detrimental outcomes have been studied , but the mechanisms that lead to these changes are still not fully elucidated . There is some indication that imbalances in cytokine cascades , complement activation and angiogenic dysregulation might be involved in the placental changes observed . Nevertheless , the majority of studies on malaria in pregnancy ( MiP ) have come from areas where malaria transmission is high and usually restricted to Plasmodium falciparum , the most pathogenic of the malaria parasite species . We conducted a cross-sectional study in Cruzeiro do Sul , Acre state , Brazil , an area of low transmission and where both P . vivax and P . falciparum circulate . We collected peripheral and placental blood and placental biopsies , at delivery from 137 primigravid women and measured levels of the angiogenic factors angiopoietin ( Ang ) -1 , Ang-2 , their receptor Tie-2 , and several cytokines and chemokines . We measured 4 placental parameters ( placental weight , syncytial knots , placental barrier thickness and mononuclear cells ) and associated these with the levels of angiogenic factors and cytokines . In this study , MiP was not associated with severe outcomes . An increased ratio of peripheral Tie-2:Ang-1 was associated with the occurrence of MiP . Both Ang-1 and Ang-2 had similar magnitudes but inverse associations with placental barrier thickness . Malaria in pregnancy is an effect modifier of the association between Ang-1 and placental barrier thickness .
In countries endemic for Plasmodium spp . , malaria in pregnancy ( MiP ) remains an important health burden that can result in maternal anaemia , small-for-gestational-age babies , low birth weight , and even miscarriages and stillbirths [1–3] . Most of the studies looking at possible mechanisms for these outcomes have been conducted in areas where P . falciparum is either the only parasite species present or the main species in circulation . Some of the associations found with these outcomes , which are not necessarily mutually exclusive , include the sequestration of parasites in the placenta , the presence of monocytes and inflammatory cytokines in the placenta , activation of the complement cascade , dysregulation of hormonal and glucose pathways , compromised amino acid uptake by the placenta and impaired angiogenesis [4–11] . Some of the inflammatory cytokines associated with MiP include ( tumour necrosis factor ( TNF ) -α , interferon ( INF ) -γ ) , interleukin ( IL ) -10 , IL-8 , IL-6 and macrophage inflammatory protein ( MIP ) -1α [12–17] . The angiogenic factors angiopoietin ( Ang ) -1 , Ang-2 and Tie-2 ( which serves as a receptor for both Ang-1 and Ang-2 ) are involved not only in vasculogenesis and angiogenesis but also play distinct roles in mediating inflammation in infectious diseases [8 , 18–21] . Despite the fact that P . vivax infections can also result in severe outcomes for both mother and child , there are relatively few studies conducted in areas where P . vivax is the most prevalent Plasmodium species [22–24] . The Amazon region , in Brazil , is one such area . We are beginning to understand some of the basic pathological processes associated with P . vivax during pregnancy and have recently shown that the placentas of women exposed to P . vivax during pregnancy have increased barrier thickness , syncytial knots and monocyte numbers compared to uninfected placentas , even in the absence of evidence for parasite sequestration ( a hallmark for P . falciparum-associated pathology ) [25] . What contributes to these histopathological observations is not known . Other studies have observed modifications in the vasculature within the placental villi suggesting that this parasite impacts angiogenic processes [26] . In this study we aimed to measure plasma levels of angiogenic factors and cytokines in women exposed to Plasmodium spp . during pregnancy in an area where P . vivax is predominant and to associate those levels with pathological features of the placenta .
The region where our study was conducted has been described elsewhere . This region presents a higher prevalence of P . vivax infections than P . falciparum infections [27] . A cross-sectional study of primigravidae at delivery was conducted in the maternity unit of the Hospital da Mulher e da Criança do Juruá in Cruzeiro do Sul , Acre , Brazil from December 2012 to August 2013 . A study team member approached every eligible woman who arrived at the maternity and questionnaires were applied to every woman who agreed to participate in the study to obtain epidemiological data . Women with a history of smoking during pregnancy , drug use and who presented with syphilis , HIV or hepatitis were excluded from the study . Due to the extremely high percentage of C-sections performed in Brazilian maternity units , women who underwent a C-section were not excluded from the study; however this was controlled for in the analysis . Peripheral blood and placental blood ( from the maternal side of the placenta ) were collected in heparin tubes . Thin and thick smears were made and stained with Giemsa , and two drops of blood were spotted on filter paper for assessment of malaria status by Nested-PCR . Biopsies of placental tissue were collected , fixed in 10% neutral buffered formalin at 4°C and then kept in ethanol 70% until they could be sent to São Paulo University for processing . Paraffin-embedded sections of placental tissue were stained with Haematoxilin-Eosin ( H&E ) or Masson’s Trichrome Stain ( MTS ) for histological examination . A Zeiss Axio Imager M2 light microscope equipped with a Zeiss Axio Cam HRc camera was used to capture images of the placentas . Some of the parameters were evaluated and analysed using Image J ( Image J 1 . 46c , Wayne Rasband , National Institutes of Health , USA , http://imagej . nih . gov/ij ) . Additionally , medical data was collected concerning blood-pressure , haemoglobin and haematocrit levels and axillary temperature . Malaria during pregnancy was diagnosed by microscopy by the endemic surveillance team of Cruzeiro do Sul , Acre , Brazil . Malaria at delivery was diagnosed by evaluation of a peripheral/placental blood smear and/or by molecular methods ( nested-PCR ) . For molecular detection , DNA was obtained from dried-blood spots ( placental and peripheral ) on filter paper through the use of a commercially available extraction kit ( QIAmp DNA Micro , Qiagen ) , following the manufacturer’s instructions . The nested-PCR reaction was conducted as previously published by dos Santos et al . [28] . Histopathological parameters were analysed using the Tissue Microarray technique , conducted at the AC Camargo Hospital , with the exception of the evaluation of immune cells , which was performed on slides containing the full length of the tissue . Two individuals performed all measurements . Cases that proved to be contradictory between observers were re-evaluated until consensus was reached . All histological evaluation methods were optimised by our group and are described elsewhere [25] . The angiogenic factors angiopoietin ( Ang ) -1 ( 1:20 dilution ) and Ang-2 ( 1:10 dilution ) and their associated soluble receptor Tie-2 ( 1:20 dilution ) , were measured using the commercially available DuoSet ELISA development kits from R&D systems , according to the manufacturer’s instructions . Levels of cytokines in both placental and peripheral plasma were measured with the commercially available Millipore kit HCYTOMAG-60K-07 ( IL-1β , IL-10 , IL-6 , IL-8 , MIP-1α , TNF-α ) , using Luminex technology and following the manufacturer’s instructions . All plasma samples were processed and kept at -80°C in Cruzeiro do Sul until they were sent to University of São Paulo . All available samples of placental blood were evaluated while for peripheral blood a random subset of uninfected women ( n = 12 ) and infected women ( n = 28 ) were chosen . Malaria in pregnancy was defined as evidence of Plasmodium infection during pregnancy or at term by microscopy . Current infection was defined as a Plasmodium infection detected at term by microscopy and/or histology and/or PCR . Anaemia was defined as a haemoglobin level lower than 11 g/dL . Low birth weight was defined as an infant weight of less than 2 , 500 g . Data were analysed using Stata 12 software ( StataCorp , College Station , TX , USA ) and GraphPad Prism ( GraphPad Prism version 5 for Mac OX , GraphPad Software , San Diego CA , USA , www . graphpad . com ) . Variables with normal distributions were analysed using the means and standard deviation , and the variables that were non-normally distributed were analysed using the medians and interquartile range . Differences in the mean values between groups were evaluated using Student’s t-tests or Mann-Whitney U-tests accordingly . Categorical data and proportions were analysed using chi-square tests . All placental parameters evaluated were ln-transformed before statistical analysis was performed . For determining the effects of one or more infections on the histopathological parameters measured and MiP infection status on angiogenicfactors/cytokines one-way ANOVA tests with Bonferroni correction were used . Multivariate linear regressions with placenta parameters as outcome variables were used to look for associations with levels of angiogenic factors and cytokines . Gestational age and C-section were found to be confounders and were included in the models . To uncover the role of Plasmodium spp . infection during pregnancy as an effect modifier of the association between angiogenic factors and cytokines with placental parameters , we introduced an interaction term in the multilinear regression model . Ethical clearance was provided by the committees for research of the University of São Paulo and the Federal University of Acre ( Plataforma Brasil , CAAE: 05736812 . 0 . 0000 . 5467 and 05736812 . 0 . 3001 . 5010 , respectively ) . All the study participants gave written informed consent or had their legal guardians do so , if they were minors .
During the eight months of recruiting , 147 women were enrolled into the study . Of those , 10 women were excluded from the analysis due to use of cigarettes and/or illicit drugs ( 7 women ) during pregnancy or presence of infections ( syphilis , hepatitis B , hepatitis C or HIV ) . The present study comprised 137 women ( Table 1 ) . A total of 92 out of the 137 women were defined as uninfecteds after microscopic , histologic or molecular examination revealed no evidence of Plasmodium spp . infection either during pregnancy or at delivery . The women without evidence of Plasmodium infection tended to be older than the women who presented an infection during pregnancy [mean ( standard deviation ( SD ) ]: 19 . 7 ( 4 . 2 ) vs 18 . 5 ( 2 . 7 ) , p = 0 . 064 , Table 1 . Clinically , both groups of women seemed to differ only in the axillary temperature ( uninfected vs infected , mean ( SD ) : 36 . 2 ( 0 . 4 ) vs 36 . 7 ( 1 . 2 ) , p<0 . 001 , Table 1 ) . Interestingly , both groups of women had similar levels of haemoglobin at delivery . The babies from both groups were also similar in terms of gestational age at delivery and birth weight ( Table 1 ) . A history of malaria prior to pregnancy was more often reported by those in the infected group vs women in the uninfected group ( 88 . 9% vs 60 . 9% , p = 0 . 001 , Table 1 ) . A total of 19 women were identified as having had P . vivax-only infections during pregnancy . Twenty-one percent of those women presented with infections at delivery ( Table 1 ) , with no detectable parasites in the placenta . P . falciparum infections accounted for a third ( n = 14 ) of the single species infections detected during pregnancy , with 5 women showing infections at delivery ( Table 1 ) . Of these 5 women , two showed abundant evidence of parasite and parasite pigment in the placenta ( Fig 1 ) . The placentas of the women who experienced a Plasmodium infection during pregnancy did not differ in weight from those who remained uninfected ( Table 2 ) . Similarly , there was no difference between both groups of women regarding parameters of placental histopathology ( syncytial knots , thickness of the placental barrier and presence of monocytes ) ( Table 2 ) . However , when the women who experienced more than one infection during their pregnancy were compared to uninfected women , there was a small but significant increase in both the thickness of the placental barrier ( median [IQR] ) 4 . 52 [4 . 25 , 5 . 12] vs 4 . 17 [3 . 56 , 6 . 61] , p = 0 . 023 and the percentage of intervillous monocytes 3 . 13 [2 . 19 , 4 . 22] vs 2 . 02 [1 . 20 , 3 . 12] , p = 0 . 025 ( Table 2 ) . Despite the small numbers , it is possible to realise that the increase in the thickness of the placental barrier observed is at least partially driven by P . vivax , while the increase in the percentage of intervillous monocytes seems independent of it ( S1 Fig ) . Levels of cytokines and chemokines ( with the exception of IFN-γ and IL-10 ) were found to be higher in the placenta than in the periphery , although this result should be taken with caution due to the small number of samples evaluated ( Fig 2 ) . Similarly , the levels of Ang-1 , Ang-2 and Tie-2 were increased in the placenta compared to the periphery ( Fig 3A , 3B and 3C ) . Contrary to what was observed for cytokine levels , the levels of Ang-1 and Tie-2 were significantly reduced ( p = 0 . 025 and p = 0 . 017 respectively ) in those women who underwent a caesarean ( Fig 3D ) , possibly reflecting the nature of the hormonal changes that occur during the normal course of a vaginal delivery . Placental and peripheral levels of both cytokines and chemokines did not generally vary between women with Plasmodium spp . infection during pregnancy and those without ( Table 3 ) . Similarly , placental levels of angiogenic factors were also identical between the women with Plasmodium spp . infection during pregnancy and those without ( Table 3 ) . However , the women with evidence of a current malaria infection had increased levels of cytokines in the periphery and of IL-10 in the placenta , which was also seen in vivax-only infected women ( Table 3 and S2 and S3 Figs ) . Additionally , we observed a decrease of peripheral Ang-1 in women with a current infection and a significantly higher ratio of Tie-2:Ang-1 in both the total women who had MiP or only the women who had a current infection , relative to the uninfected women ( Table 3 ) . Though the numbers are small , if we segregate current malaria in pregnancy by P . vivax and P . falciparum the data shows that both the decrease of Ang-1 and the increase in the ratio of Tie-2:Ang-1 are more closely associated with P . falciparum than with P . vivax ( Ang-1 ( median [min , max] ) : Uninfected ( n = 88 ) : 12 . 67 [3 . 71 , 32 . 16] , vivax ( n = 4 ) : 9 . 19 [9 . 14 , 9 . 85] , p = 0 . 975 , falciparum ( n = 5 ) : 8 . 89 [8 . 56 , 9 . 51] , p = 0 . 098; ratioTie2:Ang-1: Uninfected ( n = 88 ) : 0 . 89 [0 . 20 , 2 . 56] , vivax ( n = 4 ) : 0 . 76 [0 . 50 , 1 . 51] , p = 0 . 766 , falciparum ( n = 5 ) : 1 . 42 [0 . 86 , 2 . 65] , p = 0 . 096 ) . Associations between levels of angiogenic factors and cytokines with placental parameters were assessed for factors and cytokines measured in the placenta . Multilinear regression models , controlling for possible confounders ( caesarean and gestational age ) ( S1 Table ) revealed similar strength but opposite associations between placental barrier thickness and increased levels of Ang-2 ( coeficient [95% CI] , p-value ) 0 . 30 [0 . 10 , 0 . 51] , p = 0 . 004 and Ang-1–0 . 30 [-0 . 60 , -0 . 01] , p = 0 . 045 . Increased Ang-1 was also associated with a decrease in placental weight -36 . 34 [-69 . 76 , -2 . 91] , p = 0 . 033 . Rather then segregating the women into MiP positive or negative , thus reducing the power of our analysis , we investigated whether these associations were modified by the presence or absence of malaria during pregnancy by adding Plasmodium spp . infection as an interaction term in the model . Interestingly , the association between Ang-2 and placental barrier thickness was not modified by the presence of Plasmodium infection; however , there was some evidence that malaria during pregnancy altered the effect of Ang-1 on placental barrier thickness ( Fig 4 ) .
In this study , we were able to measure and compare cytokine and angiogenic factors levels at delivery between women who had malaria during pregnancy and those who did not . Additionally , we associated the levels of these molecules with the occurrence of histopathological changes in the placenta . Malaria during pregnancy in this region of Brazil has previously been associated with detrimental outcomes for both mother and child [25 , 29–31] . The placenta , in its role as a bridge between mother and foetus , will either suffer or mediate some of the injuries to both the mother and the foetus , which eventually lead to these adverse outcomes [32 , 33] . In Africa , where virtually all malaria infections are caused by P . falciparum , the placental events and mechanisms that contribute to adverse obstetric outcomes have been extensively studied and include adhesion of the parasite to the placenta [4] , accumulation of inflammatory cells with production of cytokines [5] , histological modifications to the villi [6] , complement activation [7–9] , disruption of nutrient transport and disturbance of the angiogenesis process [10 , 11] . In areas where P . vivax is also present , both P . falciparum and P . vivax infections are able to significantly impact placental development and foetal outcomes [34–36] , but the mechanisms involved have been insufficiently studied [26] . Because both our recently published results and those of others show that P . vivax and P . falciparum seem to have a similar magnitude of effects on the placenta in areas of low malaria transmission and where treatment is readily available [25 , 26] , we grouped all Plasmodium spp . infections together for evaluation . In this study , the Plasmodium-infected and uninfected women had very similar epidemiological characteristics . Additionally , apart from an increase in the axillary temperature in those who were infected , no other clinical features were significantly different between the groups of women , including haemoglobin levels , prevalence of anaemia , newborn birth weight and proportions of prematurity and low birth weight . These results may be a positive reflection of the active and prompt diagnosis and treatment policies in place in Acre state , where our study site is located and where women are often diagnosed and treated under supervision within 48 h [37] . Placental parameters between infected and uninfected women differed when the number of infections during pregnancy was taken into account . Consistent with our previous report [25] , the women who experienced more than one infection during pregnancy had significantly increased placental barrier thickness and mononuclear cells in the intervillous space , compared to uninfected women . This finding highlights the accumulation of insults to the placenta when multiple infections occur . The observation that placental levels of both cytokines and angiogenic factors were significantly higher in the placenta than in the periphery , independently of malaria infection status , may reflect the different time-points of collection of samples ( peripheral blood collected before delivery; placental blood collected after delivery ) but is also a clear indication of the distinct nature of both these compartments . When studying placental parameters , one should focus on the placental milieu and not rely on that of the periphery . Similar to research findings described in other populations [11 , 38] , we observed a decrease in peripheral Ang-1 levels in the pregnant women with an active infection at delivery as well as an increase in the ratio between Tie-2:Ang-1 in the women who had malaria during pregnancy . Interestingly , this seemed to be associated mainly to P . falciparum infections . Neither the level of placental Ang-1 nor the placental ratio Tie-2:Ang1 were different between women with and without infection . This may be a reflection of the low grade placental infections or , most likely , it may reflect different regulatory mechanisms in the peripheral and the placental compartments regarding the levels of these factors . Peripheral levels of all cytokines and chemokines ( with the exception of IL-6 ) were increased in the women who had malaria at delivery , and this appears to be true for those women with P . vivax infection only , but the low numbers from whom these levels were available do not allow for a more substantial analysis . In the placenta , levels of IL-10 were significantly increased in the women who had malaria at delivery compared to the uninfected women . Once more , low numbers dictated the analysis but women with P . vivax only also appeared to have higher levels of placental IL-10 . This finding substantiates the role of IL-10 not only as a marker of MiP but of malaria infection in general [39] . Increased levels of Ang-1 were associated with decreases in placental weight and with a decrease in placental barrier thickness in the entire cohort of patients . An analysis of the effect of malaria during pregnancy on this association revealed that the burden of the interaction occurred in the women who had a Plasmodium infection . In uninfected women , levels of Ang-1 did not alter placental barrier thickness while there is a negative association between levels of Ang-1 and placental barrier thickness in women who experienced malaria during pregnancy . A study with a larger number of samples from infected women is necessary in order to verify this observation . Decreased levels of peripheral and placental Ang-1 have previously been associated with the occurrence of malaria in pregnancy in African women , in an area of high P . falciparum transmission and where the detrimental effects of MiP are substantial [11 , 38 , 40] . Additionally , an association between levels of Ang-1 and complement activation ( responsible for vascular insufficiency in the placenta ) was also observed and postulated to contribute significantly to the occurrence of low birth weight [38] . In this study , we did not observe a significant decrease in placental Ang-1 in infected women but we were able to detect an association between Ang-1 levels and placental barrier thickness in women with malaria during pregnancy , which may constitute a physical mechanism for the occurrence of vascular insufficiency . Additionally , in our study , there was no association between MiP and the occurrence of low birth weight , and so it is not possible to evaluate the real impact that this increase in thickness of the placental barrier may have . In conclusion , in this area where diagnosis and treatment of malaria are readily available and where the impact of malaria during pregnancy is mild , we were able to detect a link between decreased levels of Ang-1 and an increase in the placental barrier thickness of Plasmodium-infected women .
|
Plasmodium infections during pregnancy represent a substantial health burden for mothers and their neonates leading to underweight babies , preterm deliveries , abortions , stillbirths , and even maternal mortality . The placenta , in its role as a bridge between mother and foetus , will either suffer or mediate the injuries which eventually lead to these adverse outcomes . Plasmodium infections during pregnancy can affect the metabolism of the placenta , its functions and even its architecture . These changes are well described in Sub-Saharan Africa , where P . falciparum infections predominate , but little is known in areas where P . vivax is predominant . The state of Acre in Brazil is one such area . We have recently shown that malaria in pregnancy is associated with placental pathology in this region . In the present study we sought to find out the association between the placental and peripheral levels of angiogenic factors , which are known and important players in the formation and function of the placenta , and the occurrence of placental pathology in Plasmodium-infected women . We found that , lower levels of Angiopoietin-1 in women who experienced malaria during pregnancy were associated with specific modifications occurring in the placenta . This provides a possible pathway through which these pathological changes occur .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] |
[] |
2015
|
Malaria in Pregnancy Interacts with and Alters the Angiogenic Profiles of the Placenta
|
The mosquito resistance to the insecticides threatens malaria control efforts , potentially becoming a major public health issue . Alternative methods like ivermectin ( IVM ) administration to humans has been suggested as a possible vector control to reduce Plasmodium transmission . Anopheles aquasalis and Anopheles darlingi are competent vectors for Plasmodium vivax , and they have been responsible for various malaria outbreaks in the coast of Brazil and the Amazon Region of South America . To determine the IVM susceptibility against P . vivax in An . aquasalis and An . darlingi , ivermectin were mixed in P . vivax infected blood: ( 1 ) Powdered IVM at four concentrations ( 0 , 5 , 10 , 20 or 40 ng/mL ) . ( 2 ) Plasma ( 0 hours , 4 hours , 1 day , 5 , 10 and 14 days ) was collected from healthy volunteers after to administer a single oral dose of IVM ( 200 μg/kg ) ( 3 ) Mosquitoes infected with P . vivax and after 4 days was provided with IVM plasma collected 4 hours post-treatment ( 4 ) P . vivax-infected patients were treated with various combinations of IVM , chloroquine , and primaquine and plasma or whole blood was collected at 4 hours . Seven days after the infective blood meal , mosquitoes were dissected to evaluate oocyst presence . Additionally , the ex vivo effects of IVM against asexual blood-stage P . vivax was evaluated . IVM significantly reduced the prevalence of An . aquasalis that developed oocysts in 10 to 40 ng/mL pIVM concentrations and plasma 4 hours , 1 day and 5 days . In An . darlingi to 4 hours and 1 day . The An . aquasalis mortality was expressively increased in pIVM ( 40ng/mL ) and plasma 4 hours , 1 , 5 10 and 14 days post-intake drug and in An . darlingi only to 4 hours and 1 day . The double fed meal with mIVM by the mosquitoes has a considerable impact on the proportion of infected mosquitoes for 7 days post-feeding . The oocyst infection prevalence and intensity were notably reduced when mosquitoes ingested blood from P . vivax patients that ingested IVM+CQ , PQ+CQ and IVM+PQ+CQ . P . vivax asexual development was considerably inhibited by mIVM at four-fold dilutions . In conclusion , whole blood spiked with IVM reduced the infection rate of P . vivax in An . aquasalis and An . darlingi , and increased the mortality of mosquitoes . Plasma from healthy volunteers after IVM administration affect asexual P . vivax development . These findings support that ivermectin may be used to decrease P . vivax transmission .
The 2016 World Malaria Report ( WHO ) estimated 212 million cases of malaria worldwide , leading to 429 , 000 deaths , which illustrates that malaria remains an important public health problem . In the Americas , 389 , 390 cases and 87 deaths were reported in 2016 with Brazil reporting 24% and Peru 19% of the malaria cases [1] . The majority ( 92 . 5% ) of the malaria cases occurred in the Amazon sub-region with 69% being Plasmodium vivax [1–4] . Despite considerable efforts , the majority of South American countries are still far from achieving vivax malaria elimination . Current strategies to combat malaria transmission in South America include diagnosis and treatment with artemisinin-based combination therapy ( ACT ) [5–7] and long-lasting insecticidal nets [8] , supported by indoor-residual spraying of insecticide IRS [9 , 10] . However , widespread insecticide resistance in vectors threatens the effectiveness of LLINs and IRS [1–15] . Anopheles aquasalis and Anopheles darlingi are known to be susceptible to P . vivax infection ( [16–18] ) . Anopheles aquasalis is considered the primary vector in coastal areas of Central and South America and has been used as a neotropical anopheline vector for laboratory model to evaluate host-parasite interactions [18–20] . Anopheles darlingi is the primary vector in the Amazonian region of South America [21 , 22] . The reemergence of attention on transmission blocking strategies for Plasmodium [23 , 24] have raised research efforts in attempt to find vaccines [25–27] , drugs [28 , 29] or microorganisms [30–32] able to disrupt the life cycle of the parasite in the mosquito vector . In this context , the endectocide ivermectin ( IVM ) has arisen as a new promising tool to be added to malaria control programs . Ivermectin is a safe drug with activity against a wide range of internal and external parasites and it is used widely in both veterinary and human medicine [33] . Ivermectin as a single oral dose ( 150–200 μg/kg ) is effective for the treatment or control of Onchocerca volvulus , Wuchereria bancrofti and Strongyloides stercoralis . Ivermectin has been widely distributed to humans via mass drug administration ( MDA ) campaigns against onchocerciasis and lymphatic filariasis in Africa and Latin America [34–36] . Ivermectin has a secondary effect on ectoparasites that feed on recently treated individuals [37] , including activity against Anopheles vectors at concentrations present in human blood after standard doses [38 , 39] . Consequently , IVM has emerged as a potential tool for malaria control [40 , 41] . Ivermectin MDA provides a unique insecticide dissemination route via mosquito ingestion of the compound through a blood meal rather than physical contact as most insecticides are delivered . In Anopheles vectors , IVM acts as an agonist of glutamate-gated chloride channels , causing flaccid paralysis and eventual death [42] . Both in vitro and in vivo studies have shown that a blood meal containing IVM cause a significant reduction in the adult Anopheles lifespan , with secondary sub-lethal effects delaying time to re-feed [38 , 43] , a sporontocidal effect [38 , 44 , 45] , reductions in fecundity [39 , 46 , 47] and egg hatch rate [39] , and even reduced locomotor activity [48] . Field studies of the effect of MDA with IVM on malaria transmission showed that a single dose of 150–200 μg/kg reduced the survivorship of wild Anopheles gambiae , with an accompanying reduction in the Plasmodium falciparum sporozoite rate [9 , 49 , 50] . These studies demonstrated that IVM could be a potential additional tool for malaria control . Several studies have demonstrated that powdered IVM p ( IVM ) inhibits Plasmodium development in the vector , including P . falciparum in An . gambiae [44 , 45] and more recently P . vivax in Anopheles dirus and Anopheles minimus [38] and An . darlingi [43] . Recent evidence suggests the presence of long-lived IVM metabolites that may have mosquito-lethal activity [38] . Thus , it is critical to determine the sporontocidal effect of IVM and potential metabolites in human plasma/blood after administration of IVM at clinically relevant doses . Ivermectin has also been shown to inhibit the development of asexual blood stages of P . falciparum in vitro and Plasmodium berghei in vivo [51] . Additionally , it has been reported that IVM has liver stage inhibition in a P . berghei rodent model [52] . It remains to be determined if these blood and liver stage effects of IVM have a prophylactic personal protective effect for a human . In this study , for the first time , we evaluated the IVM effect and its possible metabolites against P . vivax in An . aquasalis and An . darlingi . Additionally , we investigated the potential effects of IVM against asexual blood stages of P . vivax . A better understanding of the effects of IVM and possible metabolites on parasite maturation and transmission to mosquito vectors would contribute to the development of IVM for malaria control strategies .
The procedures were approved by the Fundação de Medicina Tropical Dr . Heitor Vieira Dourado ( FMT-HVD ) Ethics Review Board ( ERB ) ( Approval Number: 296 . 723 CAAE: 14148813 . 7 . 0000 . 0005 ) and by the U . S . Naval Medical Research Unit No . 6 ( NAMRU-6 ) and Walter Reed Army Institute of Research Institutional Review Boards ( NMRCD . 2008 . 0004 and WRAIR#2175 ) . All subjects provided and signed written informed consent . This study was conducted at FMT-HVD , a tertiary care center for infectious diseases in Manaus , Amazonas State , Brazil and NAMRU-6 in Iquitos ( Loreto ) , Peru . In order to obtain P . vivax samples , adult patients ( ages ≥18 years ) infected with P . vivax were recruited from the areas surrounding Manaus and Iquitos . P . vivax infection was determined by light microscopy of Giemsa stained blood samples . In Brazil , P . vivax infected patients were identified at FMT-HVD , enrolled and venous blood ( 15 ml ) was collected . In Peru , P . vivax infected patients were identified at Ministry of Health Centers and hospitals in Iquitos , transported to NAMRU-6 , enrolled and venous blood ( 15 ml ) was drawn on site for the ivermectin sporogony experiments . The age , gender , history of previous malaria episodes , and place of residence were obtained from the patients and it was verified no signs of severe disease and no previous antimalarial treatment during the preceding 4 weeks . After blood collection , all patients were treated for P . vivax infection following guidelines of the Brazilian Health Ministry or Peruvian Ministry of Health [53] . In order to obtain metabolized ( mIVM ) plasma samples , healthy volunteers were recruited in Manaus , Brazil . Inclusion criteria were: adults ≥18 years old , history of at least 3 month-negative malaria , and were confirmed as healthy volunteers by health works . Anopheles aquasalis were reared at Laboratory of Medical Entomology at FMT-HVD in Manaus , Brazil and An . darlingi were reared at NAMRU-6 in Iquitos , Peru . These colonies were kept at a constant temperature ( 24–26°C ) and relative humidity ( 70–80% ) on a 12:12 light-dark cycle . Larvae were hatched in room temperature water and fed ground TetraMin fish food for An . aquasalis and rodent diet for An . darlingi provided daily . The larvae were allowed to pupate and emerge into adults in an enclosed mesh-covered cage with water and 10% sucrose available [18 , 54] . Adult mosquitoes used for experiments were between 3–5 days post-emergence . Powdered pIVM and powdered chloroquine ( pCQ ) reference material were obtained from Sigma Aldrich ( St . Louis , MO , USA ) . pIVM was dissolved in dimethyl sulfoxide ( DMSO ) to a concentration of 10 mg/ml and pCQ in Roswell Park Memorial Institute ( RPMI ) 1640 medium ( Sigma Aldrich , St . Louis , MO , USA ) to a concentration of 1 mg/ml and aliquots were frozen at -20°C . IVM was serially diluted in PBS to achieve experimental concentrations . All experimental drug regimens were managed at the FMT-HVD in Manaus , Brazil . IVM tablets ( Abbot Laboratórios do Brasil , State , Brazil ) were administered at a single dose of 200 μg/kg . Vivax patients received chloroquine ( CQ ) tablets ( Farmaguinhos Laboratórios do Brasil , Rio de Janeiro State , Brazil ) , administered as a daily dose for three days ( i . e . , 600 mg in the first day and 450 mg in the second and third day ) . Primaquine ( PQ ) tablets ( Med Pharma , São Paulo State , Brazil ) were administered as a daily dose of 30 mg for 7 days following the Brazilian MoH guideline for vivax malaria treatment [53] . To evaluate the effect of mIVM on P . vivax development , five volunteers , all healthy men from 18 to 50 years , were recruited for the experiments . They received one single dose ( 200 μg/kg ) of IVM . Blood was collected in heparinized tubes at 0 and 4 hours , and 1 , 5 , 10 and 14 days after drug intake . For the evaluation of the in vivo mIVM effect on P . vivax , 15 patients with confirmed P . vivax malaria infection were recruited . The patients were divided into four groups and different treatment regimens were provided: ( 1 ) IVM plus CQ , ( 2 ) CQ alone , ( 3 ) PQ plus CQ and ( 4 ) IVM plus PQ plus CQ . Before and after 4 hours of drug intake , blood samples were collected . The patients that receive IVM plus CQ or CQ alone have received the first PQ dose after blood was drawn at 4 hours past CQ intake . All patients were treated with PQ and CQ dosage following the Brazilian Ministry of Health Guidelines . Four experiments were performed to determine the effect of IVM on P . vivax in either An . aquasalis or An . darlingi ( Fig 1 ) : P . vivax infected blood from patients was prepared as described above . The plasma was removed and packed red blood cells ( RBCs ) were washed with RPMI 1640 medium , repeated twice , and reconstituted to 40% hematocrit with non-immune human AB serum ( experiment 1: pIVM ) or with plasma from drug-treated volunteers or patients ( experiments 2–4: mIVM , double feed mIVM , and in vivo mIVM ) . Adult female mosquitoes were sugar starved overnight prior to infection via membrane feeding assay ( MFA ) . Blood meals ( 1ml ) prepared as described above were offered to groups of at least 100 mosquitoes for 30 minutes via membrane feeder devices at 37°C as described in detail elsewhere [18 , 20] . The fully engorged mosquitoes were separated into different cages and kept until 7 or 14 days post-feeding . Mosquito mortality was monitored on day 7 or 14 . Seven days after P . vivax blood meal ingestion midguts from all experimentally infected mosquito groups were dissected in PBS under a stereo microscope . The midguts were stained with 0 . 1% commercial Mercurochrome ( Merbromin , Sigma- Aldrich , USA ) , placed under a cover slip and examined for the presence of oocysts with a compound microscope ( Optical Microscopy , Olympus , Germany ) . Infection prevalence was expressed as percentage of mosquitoes with at least one oocyst . Infection intensity was determined as the arithmetic mean of oocysts counted per dissected midgut . Plasma samples from healthy volunteers and patients were shipped on dry-ice to the Department of Clinical Pharmacology , Mahidol-Oxford Tropical Medicine Research Unit , Bangkok , Thailand for IVM drug concentration measurements . Plasma concentrations of IVM were determined by a newly developed and validated method using solid-phase extraction and liquid chromatography with tandem mass spectrometry ( manuscript in preparation ) . The linear range for quantification was 0 . 97–384 ng/m . Three replicates of quality control samples at low , middle , and high concentrations were included in the analysis to ensure precision and accuracy . The observed total assay coefficient of variation was <10% in all quality control samples in accordance with US Food and Drug Administration requirements [56] . The P . vivax schizont maturation assay was performed as previously described [57] with five P . vivax samples from Manaus , Brazil . The parasitemia was determined by counting the number of parasites per 200 leukocytes . The plasma and buffy coat were separated from RBCs by centrifugation and packed RBCs were washed with RPMI and this was repeated twice . Leukocytes were removed by passing samples in a cellulose column [58] and afterward , the RBC pellet was suspended in McCoy 5A medium supplemented with 3 . 2% glucose and 20% human AB serum . Drug plates were freshly prepared to avoid drug degradation . The pIVM and pCQ stock solutions were subsequently diluted in McCoy’s medium to obtain 9 serial dilutions of the drug ( 1000–3 . 9 ng/ml ) . To determine IC50s , 50 μl of pIVM and pCQ solutions were added into 3 wells of 96 well plates , and 9 serial dilutions of each drug were done in triplicate . Three wells were free of drug and served as a control . Additionally , 50 μl of plasma samples from each IVM-treated healthy volunteer 0 and 4 hours after IVM administration were added at 4 different dilutions ( 1:2 , 1:4 , 1:8 and 1:16 ) in triplicate . Then 50 μl of parasite solution was added to each well to achieve a 4% hematocrit . The plates were incubated in a modular incubator chamber containing 5% CO2 , 5% O2 , and 90% N2 at 37°C . Incubation was stopped when at least 40% of parasites on the ring stage had matured to schizonts in the drug-free control wells . After incubation , the plates were allowed to stand for 30 minutes in a semi-vertical position . The supernatant was removed , erythrocytes suspended in the remaining fluid , and a thick blood film was made from each well . Thick blood films were stained with Giemsa stain . The number of schizonts containing more than three nuclei per 200 asexual stage parasites was determined in each blood film . Data for analyses prepared as follows: Schizont maturation in relation to control ( % ) = 100 x ( number of schizonts in treated well/number of schizonts in control wells ) . Data were entered in Prism 7 . 03 ( GraphPad , USA ) and subsequently analyzed by Stata 11 . 2 ( Data Analysis and Statistical Software , Texas , USA ) . Mosquito oocyst prevalence and schizont proportions were compared by one-way ANOVA and post-hoc analysis using paired T-test to compare each concentration with respect to the control . Mosquito oocyst intensity was compared by Kruskal-Wallis test and post-hoc analysis using Kolmogorov-Smirnov to compare each concentration with respect to the control . P-values <0 . 05 were considered statistically significant . The P . vivax mosquito-stage oocyst inhibition in An . Aquasalis and An . darlingi after ingestion of whole blood spiked with IVM or whole blood equivalent from volunteers administered ivermectin were analyzed using GraphPad Prism v7 . 02 . Normalized oocyst inhibition versus human volunteer plasma concentrations of IVM was analyzed using a nonlinear dose-response analysis with a variable slope . The maximum inhibition was fixed to 100% , and the minimum inhibition ( zero drug concentration ) and the drug concentration producing 50% of maximum effect ( IC50 ) was estimated . The P . vivax asexual blood-stage inhibitory concentrations ( i . e . IC50 , IC90 and IC99 ) for pIVM and pCQ were determined using the free software ICE estimator available online at http://www . antimalarial-icestimator . net/ .
Median peak plasma concentrations ( i . e . , plasma samples collected 4 hours post-dose ) was 79 . 8 ( 50 . 6–112 ) ng/mL after a single oral dose of 200 μg/kg of IVM . Only two out of five volunteers had detectable IVM plasma concentrations 10 days post-dose ( i . e . , 2 . 64 and 2 . 32 ng/mL ) , while only one volunteer had a detectable IVM plasma concentration on day 14 after dose ( i . e . , 1 . 87 ng/mL ) ( S1 Table ) . Different concentrations of pIVM were evaluated ( 5 , 10 , 20 and 40 ng/mL ) on the P . vivax infection prevalence , oocyst intensity , and An . aquasalis mortality ( Fig 2A–2C , S1 Dataset ) . There were significant differences among pIVM treatment groups for infection prevalence [F ( 3 . 23 ) = 11 . 58 , p = 0 . 001] compared to the control ( 0ng/ml ) group . P . vivax infection prevalence was significantly reduced in mosquitoes that ingested pIVM at 10 ng/ml by 33 . 2% [32 . 03% ( SD = 5 . 83% ) , p = 0 . 0051 , reps = 7 , n = 171] , 20 ng/ml by 33 . 7% [31 . 81% ( SD = 5 . 74% ) , p = 0 . 019 , reps = 7 , n = 68] and 40 ng/ml by 61 . 3% [18 . 60% ( SD = 5 . 81% ) , p<0 . 0001 , reps = 7 , n = 178 ) concentrations but not the 5ng/ml by 12 . 6% [41 . 93% ( SD = 7 . 67% ) , p = 0 . 141 , reps = 7 , n = 160 ) concentration ( Fig 2A ) . Also , the infection intensity ( i . e . number of oocysts per mosquito ) was reduced in the groups of mosquitoes that fed on infective blood meals containing 10ng/mL by 62 . 6% [5 . 89 ( SD = 5 . 87 ) , p = 0 . 0079 , reps = 7 , n = 171 ) , 20ng/mL by 72 . 6% [4 . 32 ( SD = 2 . 71 ) , p = 0 . 0298 , reps = 7 , n = 68] and 40ng/mL by 86 . 1% [2 . 20 ( SD = 0 . 97 ) , p<0 . 0001 , reps = 7 , n = 178 ) but not the 5ng/ml by 56 . 6% [6 . 84 ( SD = 3 . 20 ) , p = 0 . 99 , reps = 7 , n = 160 ) concentration of pIVM in comparison to the control group ( Fig 2B ) [15 . 73 ( SD = 12 . 97 ) , reps = 7 , n = 134] . There were significant differences among pIVM treatment groups for mosquito mortality at 7-day [F ( 3 . 23 ) = 2 . 99 , p = 0 . 052] compared to the control ( 0ng/ml ) group but the mortality rate was higher only in the 40 ng/ml group [78 . 11% ( SD = 7 . 17% ) , p = 0 . 0007] but not 20 ng/ml group [66 . 34% ( SD = 9 . 72% ) , p = 0 . 056] , 10 ng/ml group [49 . 09% ( SD = 20 . 03% ) , p = 0 . 646] or 5ng/ml group [63 . 38% ( SD = 13 . 26% ) , p = 0 . 080] compared to the control group [43 . 93% ( SD = 14 . 46% ) ] ( Fig 2C ) . The ingestion of mIVM presented in the infective blood meals affected the infection rate and the infection intensity and the mortality of the vectors An . aquasalis and An . darlingi as seen in the Fig 3A–3F . There were significant differences for among mIVM treatment groups for infection prevalence [F ( 3 . 31 ) = 119 . 02 , p<0 . 001] in An . aquasalis and [F ( 2 . 9 ) = 9 . 39 , p = 0 . 0063] in An . darlingi compared to the control ( 0 hour ) group . The oocyst infection rates were reduced in the groups of An . aquasalis fed volunteer mIVM plasma collected at 4 hours ( 48 . 0 ng/ml ) by 89 . 2% [7 . 82% ( SD = 7 . 04% ) , p<0 . 0001 , reps = 10 , n = 232 ) , 1 day ( 5 . 55 ng/ml ) by 51 . 5% [35 . 11% ( SD = 8 . 96% ) , p = 0 . 0038 , reps = 6 , n = 102] , and 5 days ( 1 . 58 ng/ml ) by 24 . 2% [54 . 88% ( SD = 8 . 57% ) , p = 0 . 0009 , reps = 9 , n = 300 ) but not 10 days by 14 . 3% [62 . 10% ( SD = 9 . 46% ) , p = 0 . 063 , reps = 9 , n = 258 ) or 14 days by 2 . 3% [70 . 75% ( SD = 7 . 57% ) , p = 0 . 073 , reps = 9 , n = 214 ) ( Fig 3A ) compared to the mIVM control [72 . 38% ( SD = 6 . 05% ) , reps = 10 , n = 448] . The oocyst infection rates were reduced in the groups of An . darlingi fed mIVM plasma collected at 4 hours ( 43 . 24 ng/ml ) by 91 . 1% [6 . 25% ( SD = 10 . 8% ) , p = 0 . 019 , reps = 4 , n = 8] and 1 day ( 5 . 69 ng/ml ) by 73 . 9% [18 . 33% ( SD = 18 . 48% ) , p = 0 . 014 , reps = 4 , n = 11] , but increased slightly but not significantly on 5 days ( 1 . 35 ng/ml ) by 1 . 95% [71 . 47% ( SD = 22 . 88% ) , p = 0 . 156 , reps = 4 , n = 67] , 10 days ( 2 . 48 ng/mL ) by 0 . 48% [70 . 44% ( SD = 28 . 17% ) , p = 0 . 098 , reps = 4 , n = 97] , or 14 days ( 1 . 87 ng/mL ) by 14 . 89% [80 . 54% ( SD = 7 . 5% ) , p = 0 . 350 , reps = 4 , n = 51] compared with the control [70 . 1% ( SD = 25 . 4% ) , reps = 4 , n = 97] ( Fig 3B ) . The mIVM concentrations imbibed by the mosquitoes were calculated by taking pharmacokinetic values and multiplying by 60% to account for the 40% hematocrit in each blood meal , several volunteers had no detectable ivermectin at days 10 and 14 so no concentrations ingested by mosquitoes could be estimated . We did not find significant differences for infection prevalence among both species in each treatment group: Control ( p = 0 . 9178 ) , 4 horas ( p = 0 . 527 ) , 1 day ( 0 . 287 ) , 5 days ( 0 . 2883 ) , 10 days ( p = 0 . 734 ) and 14 days ( p = 0 . 513 ) . However , the mean oocyst intensity for An . darlingi ( Fig 3D ) is substantially higher than An . aquasalis ( Fig 3C ) . On the other hand , the oocyst infection intensities were reduced in all the groups: plasma 4 hours ( 43 . 24 ng/ml ) by 96 . 1% [0 . 78 ( SD = 2 . 03 ) , p<0 . 0001 , reps = 6 , n = 232] , 1 day ( 5 . 69 ng/ml ) by 56 . 8% [8 . 52 ( SD = 10 . 09 ) , p<0 . 0001 , reps = 6 , n = 102] , 5 days ( 1 . 35 ng/ml ) by 29 . 4% [13 . 91 ( SD = 11 . 39 ) , p<0 . 0001 , reps = 6 , n = 300] , 10 days by 41 . 3% [11 . 57 ( SD = 5 . 50 ) , p = 0 . 008 , reps = 6 , n = 258] and 14 days by 58 . 2% [8 . 24 ( SD = 2 . 55 ) , p = 0 . 007 , reps = 6 , n = 214] after the volunteer pIVM intakes compared with the control [19 . 70 ( SD = 13 . 09 ) , reps = 6 , n = 448] in An . aquasalis ( Fig 3C ) and in plasma 4 hours ( 43 . 24 ng/ml ) by 99 . 9% [0 . 06 ( SD = 0 . 10 ) , p = 0 . 003 , reps = 4 , n = 8] , 1 day ( 5 . 69 ng/ml ) by 97 . 57% [2 . 31 ( SD = 3 . 77 ) , p = 0 . 008 , reps = 4 , n = 11] and 5 days ( 1 . 35 ng/ml ) by 56 . 8% [76 . 49 ( SD = 80 . 96 ) , p = 0 . 011 , reps = 6 , n = 67] , but not 10 days by 56 . 8% [99 . 77 ( SD = 125 . 9 ) , p = 0 . 593 , reps = 6 , n = 97] and 14 days by 56 . 8% [37 . 80 ( SD = 5 . 1 ) , p = 0 . 141 , reps = 6 , n = 51] after the volunteer pIVM intakes compared with the control [94 . 8 , ( SD = 90 . 18 ) , reps = 4 , n = 97] in An . darlingi ( Fig 3D ) . It is important to highlight that An . darlingi showed much less oocyst number on 4 hours and 1 day than An . aquasalis . At day 7 post blood meal , there were significant differences in the mortality rate among mIVM treatment groups in An . aquasalis [F ( 5 . 46 ) = 16 . 45 , p<0 . 0001] and mIVM treatment groups in An . darlingi [F ( 2 . 9 ) = 16 . 92 , p<0 . 0001] compared to the control groups ( plasma 0 hours ) . The mosquito mortality was significant higher 4 hours [78 . 33% ( SD = 17 . 73% ) , p = 0 . 0006] , 1 day [60 . 15% ( SD = 6 . 95% ) , p = 0 . 0017] , 5 days [36 . 84% ( SD = 11 . 99% ) , p = 0 . 046] , 10 days [50 . 01% ( SD = 11 . 44% ) , p = 0 . 014] and 14 days [39 . 83% ( SD = 17 . 58% ) , p = 0 . 043] ( Fig 3E ) than the control group [21 . 29% ( SD = 15 . 80% ) ] in An . aquasalis . However , at day 7 post blood meal only the An . darlingi groups feds on infective blood meals containing plasma collected 4 hours [97 . 43% ( SD = 2 . 56% ) , p = 0 . 027] and 1 day [97 . 07% ( SD = 3 . 23% ) , p = 0 . 025] had significantly increased mortality rates but not when fed plasma from 5 days [71 . 23% ( SD = 14 . 91% ) , p = 0 . 144] , 10 days [40 . 25% ( SD = 7 . 47% ) , p = 0 . 855] or 14 days [31 . 4% ( SD = 1 . 9% ) , p = 0 . 873] compared with the control [43% ( SD = 22 . 45%] ( Fig 3F ) . At 14 days post blood meal , after mIVM plasma from 4 hours , 1 and 5 days intake all the An . darlingi had died , therefore sporozoite prevalence at these time points could not be analyzed . Also , there were no significant differences on the mortality rate with respect to control [36 . 83% ( SD = 20 . 06% ) ] in the groups that were blood fed mIVM of 10 days [57 . 66% ( SD = 15 . 21% ) , p = 0 . 265] and 14 days [48 . 65% ( SD = 21 . 67% ) , p = 0 . 200] . An . darlingi sporozoite prevalence at day 14 post blood meal was not reduced with respect to control [90 . 27% ( SD = 13 . 39% ) ] in the groups that were blood fed mIVM from 10 days [90 . 38% ( SD = 10 . 88% ) , p = 0 . 992] or 14 days [82 . 14% ( SD = 1 . 68% ) , p = 0 . 824] . Interestingly , the mIVM ( IC50 = 5 . 68 [3 . 79–7 . 56] ng/ml [95%CI] ) of mosquito-stage P . vivax in An . aquasalis appears to be much lower compared to pIVM ( IC50 = 28 . 15 [16 . 36–39 . 94] ng/ml ) ( Fig 4A and 4B ) . This demonstrates that mIVM has a more potent sporontocidal effect against P . vivax compared to ivermectin compound . Anopheles darlingi and An . aquasalis had significantly reduced oocyst infection prevalence and intensity when ingested a second mIVM ( 4 hours ) non-infective bloodmeal 4 days post P . vivax infection compared to the control . An . aquasalis oocyst prevalence was reduced by 84 . 5% [10 . 72% ( SD = 6 . 79% ) , p = 0 . 047 , reps = 4 , n = 50] compared to the control [68 . 83% ( SD = 10 . 69% ) , reps = 4 , n = 94] ( Fig 5A ) and An . darlingi by 60 . 3% [33 . 98% ( SD = 17 . 8% ) , p<0 . 0001 , reps = 3 , n = 31] compared to the control [85 . 49% ( SD = 13 . 2% ) , reps = 3 , n = 73] ( Fig 5B ) . Oocyst intensity in An . aquasalis was reduced by 93 . 6% [0 . 94 ( SD = 0 . 67 ) , p<0 . 0001 , reps = 4 , n = 50] compared to the control [14 . 65 ( SD = 7 . 71 ) , reps = 4 , n = 94] ( Fig 5C ) and in An . darlingi by 97% [0 . 73 ( SD = 0 . 2 ) , p<0 . 0001 , reps = 3 , n = 31] compared to the control [24 . 0 ( SD = 53 . 3 ) , reps = 3 , n = 73] ( Fig 5D ) . There were significant reductions in oocyst infection prevalence and intensity compared to control when An . aquasalis ingested the drug-treated infected blood meals with unprocessed [F ( 3 . 17 ) = 159 . 90 , p<0 . 0001] and reconstituted [F ( 3 . 17 ) = 164 . 82 , p<0 . 0001] blood . For unprocessed blood , IVM+CQ by 63 . 7% [29 . 44% ( SD = 4 . 19% ) , p = 0 . 029 , reps = 3 , n = 64]; PQ+CQ by 71 . 1% [23 . 44% ( SD = 7 . 95% ) , p = 0 . 02 , reps = 3 , n = 161] and IVM+PQ+CQ by 66 . 5% [27 . 15% ( SD = 2 . 20% ) , p = 0 . 046 , reps = 3 , n = 99] but not CQ alone by 26 . 5% [59 . 49% ( SD = 10 . 9% ) , p = 0 . 072 , reps = 3 , n = 58] compared with the control [80 . 91% ( SD = 4 . 12% ) , reps = 3 , n = 615] , ( Fig 6A ) . Similar to observed in mosquitos fed with unprocessed blood , the infection rate was significantly reduced on mosquitos fed with P . vivax blood meal reconstituted with plasma from patients that undertook IVM+CQ by 65 . 6% [27 . 86% ( SD = 4 . 44% ) , p = 0 . 006 , reps = 3 , n = 205] , PQ+CQ 56 . 55% [35 . 21% ( SD = 4 . 17% ) , p = 0 . 004 , reps = 3 , n = 180] and IVM+PQ+CQ 62 . 9% [30 . 02% ( SD = 1 . 00% ) , p = 0 . 042 , reps = 3 , n = 140] but not CQ alone by 19 . 2% [65 . 40% ( SD = 7 . 63% ) , p = 0 . 077 , reps = 3 , n = 91] in comparison to control [80 . 91% ( SD = 4 . 12% ) , reps = 3 , n = 615] ( Fig 6A ) . There were no significant differences in oocyst prevalence between the unprocessed and reconstituted treatment regimens [IVM+CQ p = 0 . 246] , [CQ p = 0 . 784] , [PQ+CQ p = 0 . 120] and [IVM+PQ+CQ p = 0 . 128] . The infection intensity ( oocysts number ) was significantly reduced in mosquitos fed with unprocessed: IVM+CQ by 90 . 1% [1 . 99 ( SD = 2 . 01 ) , p<0 . 001 , reps = 3 , n = 64]; PQ+CQ by 90 . 8% [1 . 86 ( SD = 1 . 77 ) , p<0 . 001 , reps = 3 , n = 161] , IVM+PQ+CQ by 89 . 1% [2 . 19 ( SD = 0 . 96 ) , p<0 . 001 , reps = 3 , n = 99] and CQ alone by 56 . 5% [8 . 71 ( SD = 8 . 92 ) , p<0 . 001 , reps = 3 , n = 58] compared with the control [20 . 02 ( SD = 4 . 05 ) , reps = 3 , n = 615] ( Fig 6B ) and reconstituted with plasma from patients that undertook IVM+CQ by 84 . 1% [3 . 19 ( SD = 1 . 43 ) , p<0 . 001 , reps = 3 , n = 205] , PQ+CQ 71 . 6% [5 . 69 ( SD = 5 . 14 ) , p<0 . 001 , reps = 3 , n = 180] , IVM+PQ+CQ by 87 . 6% [2 . 50 ( SD = 0 . 64 ) , p<0 . 001 , reps = 3 , n = 140] and [20 . 02 ( SD = 4 . 05 ) , reps = 3 , n = 615] . There were no significant differences in oocyst prevalence between the unprocessed and reconstituted treatment regimens [IVM+CQ p = 0 . 246] , [CQ p = 0 . 784] , [PQ+CQ p = 0 . 120] and [IVM+PQ+CQ p = 0 . 128] . A total of 5 malaria vivax patients were recruited with high parasitemia ( 138–245 parasites/200 leucocytes ) , with at least 80% of parasites in the ring stage . We tested five isolates of P . vivax for pIVM drug sensitivity . The pIVM had some activity against two of five P . vivax isolates tested ( Table 1 ) . As expected in our site , chloroquine was fully effective as the blood schizonticidal treatment with pCQ IC50 ranging from 1 . 96 ng/mL to 7 . 53ng/mL . Interestingly , when mIVM was added to P . vivax culture in different dilutions , a significant reduction in parasite maturation was observed compared to drug free control 1:2 ( 34 . 24 ng/ml ) [52 . 31% ( SD = 18 . 06% ) , p = 0 , 011 , reps = 5]; 1:4 ( 17 . 12 ng/ml ) [52 . 34% ( SD = 11 . 49% ) , p = 0 . 0002 , reps = 5]; 1:8 ( 8 . 56 ng/ml ) [54 . 93% ( SD = 11 . 78% ) , p = 0 . 0013 , reps = 5] and 1:16 ( 4 . 28 ng/ml ) [51 . 77% ( SD = 9 . 92% ) , p = 0 . 0001 , reps = 5] ( Fig 7 ) .
Effective malaria transmission-blocking tools are an integral element in malaria eradication campaigns . MDA of IVM disrupted malaria parasite transmission in West Africa [9 , 49] by killing the vector An . gambiae [9 , 50] which shifts the population age structure , thereby reducing the sporozoite rate . Additional effects of ivermectin which likely further reduce transmission include inhibiting sporogony in the vector as demonstrated with P . falciparum in An . gambiae [45] and P . vivax in An . dirus , An . minimus [37] , and An . darlingi [43] . Previous findings from our group also showed effects of pIVM and mIVM on survivorship , fecundity and even the locomotor activity of An . aquasalis [39 , 48] . Therefore , IVM MDA has strong potential to be a novel tool for malaria transmission control . To our knowledge , this is the first study to evaluate mIVM effects on the P . vivax oocyst infection and intensity in Anopheles . This is also the first study to asses IVM effects against P . vivax asexual blood-stage development . Herein , it was demonstrated that IVM reduces the oocyst infection and intensity of P . vivax . When An . aquasalis were fed with different concentrations of pIVM , a reduction of oocyst infection rate and infection intensity at the 10 , 20 and 40 ng/ml concentrations ( Fig 2 ) . These ivermectin concentrations were selected based on the human pharmacokinetic curve for IVM which corresponds to approximately the concentration found in human blood between 4 to 60 hours post ingestion of IVM at 200 μg/kg [33 , 55] . A recent report demonstrates that An . darlingi had modest sporontocidal pIVM results wherein oocyst prevalence was significantly reduced by 22 . 6% at the lethal concentration that kills 50% ( LC50 ) ( 43 . 2 ng/ml ) and 17 . 1% at the LC25 ( 27 . 8 ng/ml ) but not significantly by 11 . 3% at the LC5 ( 14 . 8 ng/ml ) . Furthermore , there were no reductions in oocyst intensity in An . darlingi at any pIVM concentration tested [43] . Other reports also show the effect of pIVM on P . vivax oocyst prevalence and intensity in Asian vectors Anopheles dirus and Anopheles minimus wherein sporontocidal effect was far more impactful [38] . Interestingly , when An . aquasalis was fed P . vivax with mIVM , a reduction on infection rate and intensity was observed with plasma collected at 4 hours , 1 and 5 days after drug intake by the healthy volunteers . A far more potent sporontocidal effect was observed in An . aquasalis ingesting mIVM concentrations from day 1 ( 5 . 55 ng/ml ) reduced oocyst prevalence by 51 . 5% compared to pIVM at 5 ng/ml by 12 . 6% . A similar effect was observed in An . darlingi ingesting mIVM concentrations from day 1 ( 5 . 69 ng/ml ) reduced oocyst prevalence by 73 . 9% compared to pIVM at 14 . 8 ng/ml reduced oocyst prevalence by 11 . 3% [43] . Here we demonstrate that metabolized ivermectin has a more potent sporontocidal effect compared to ivermectin compound ( Fig 4 ) , which suggests that ivermectin metabolites may enhance the sporontocidal effect . Little is known about ivermectin metabolite production in humans , one small trial ( n = 4 ) indicated that mean peak plasma concentration of metabolites was 2 . 5 fold greater than that of the parent compound and the effective half-lives of the metabolites were approximately 2 . 9 days while the parent compound half-life was 11 . 8 hours [59] . Future studies should be designed to elucidate ivermectin metabolite production in orally-treated volunteers and their impact on mosquito survivorship and Plasmodium sporogony . Interestingly , oocyst prevalence and intensity was more intensely impacted with mIVM plasma from 1 day in An . darlingi compared to An . aquasalis ( Fig 3 ) but no significant reduction occurred when plasma from day 5 was fed to An . darlingi which suggests a shorter duration of but stronger sporontocidal effect compared to An . aquasalis . Mosquito mortality effect was observed at day 7 post-blood feed with mIVM plasma from 4 hours and 1 day in the two-species studied . This IVM effect on mosquito survival is similar to our previous findings in non-infected An . aquasalis which also used metabolized ivermectin showing the higher impact on the survival and reproductive fitness [39] and with other studies in different anopheline species infected and uninfected with Plasmodium [9 , 50 , 60 , 61] . In a single dose of 200 mcg/kg showed an increase in mosquito mortality in An . aquasalis when fed on mIVM at 1 day ( 5 . 55 ng/ml ) to 5 days ( 1 . 58 ng/ml ) the drug intake , but there was no longer effect when fed plasma collected from days 10 or 14 . The reduction in An . aquasalis and An . darlingi oocyst infection and intensity when mosquitoes were blood fed mIVM 4 days after infection expands the window that a blood meal containing IVM has an effect on P . vivax mosquito infection . Moreover , IVM was able to impair parasite development even when it was given to the mosquitoes after the midgut epithelium invasion by parasite ookinete suggesting that IVM has direct effects on already established and developing oocysts . Interestingly , these results were different from Kobylinski et al . [45] , wherein no sporontocidal effect was observed on early-stage oocyst development when pIVM was fed 3 days after P . falciparum infection in An . gambiae . This study also evaluates the in vivo exposure of P . vivax to IVM , CQ , and PQ in the human and its subsequent development in An . aquasalis . Mosquito oocyst infection and intensity were significantly reduced when mosquitoes were fed blood from patients treated with IVM+CQ , PQ+CQ or IVM+PQ+CQ but not CQ alone ( Fig 6 ) . Importantly , this is the first study to show that primaquine has a sporontocidal effect on P . vivax infection in the mosquito . However , the reduction on infection rate and intensity were not augmented on IVM +PQ+ CQ treated blood in relation to IVM+CQ only , suggesting that PQ does not have an additive or synergistic effect beyond IVM . The mosquitoes were exposed in parallel to P . vivax with unprocessed and reconstituted blood samples . Unprocessed blood samples allowed for in vivo exposure of P . vivax to the drugs in the human compared to reconstituted blood samples which investigated the impact of metabolized drugs on P . vivax . The reduction in oocyst infection and intensity found in IVM+CQ and IVM+PQ+CQ treatment groups was similar in mosquitoes fed with the unprocessed and reconstituted blood ( Fig 6 ) . These data indicate that the ivermectin transmission blocking effect occurs in the mosquito midgut and not in human blood . Similar results were observed in the PQ+CQ and IVM+PQ+CQ groups , suggesting that primaquine effect on mosquito infection also occurs in the midgut . Since these assays were performed at only one-time point after drug intake ( 4 hours ) , we could not discard a possible delayed in vivo effect of primaquine or ivermectin on P . vivax asexual stages and gametocytes in the patient . It is important to note that all patients who received the different regimens treatment were also supplied with CQ at the same time , following the guidelines of the Brazilian Health Ministry , which recommend all the patients ethically have to receive the CQ treatment at the same time that they are diagnosed . As expected , the mosquitoes fed a blood meal containing only CQ did not show a decrease in the oocyst infection prevalence of An . aquasalis with P . vivax , which is in accordance with other studies , where CQ did not affect the oocyst prevalence of Plasmodium berghei in An . gambiae [62] . However , this is the first report to demonstrate that oocyst intensity was reduced in mosquitoes fed a blood meal containing CQ . A reduction in oocyst intensity by CQ could be due to some direct action on P . vivax , or its immunosuppressive potential [63] may interfere with successful parasite midgut invasion leading to fewer oocysts . Our results showed the highest mortality rate reduction of infection rate and intensity of An . aquasalis and An . darlingi , two important vectors of South American on plasma 4 hours and considering that in the human pharmacokinetic curve of the IVM the mean peak plasma concentrations is ( 46 . 6 ± 21 . 9ng/ml ) at approximately 4 hours after dosing , with a IVM half-life from about 12 to 56 hours [33] . Similar peaks have been found in Primaquine and Chloroquine [64] . We can suggest the use of the IVM as a potent way to administration in combination with the other antimalaria drugs and mainly during the first hours after being detected the infection in the patient , which would have a higher impact in the Malaria elimination and eradication programs , specially , in endemics areas like Amazonas Region , which , have high incidence of Malaria cases by P . vivax . This is the first study to assess the effect of ivermectin against asexual P . vivax . Two previous studies demonstrated an inhibition of pIVM on P . falciparum asexual stage development but with IC50s in the 1–10 μg/ml range [51 , 65] . No effect of pIVM ( 3 . 9–1000 ng/ml ) was observed in the current study against asexual blood-stage P . vivax , but this may have been due to using too low concentration ( Fig 7 ) . On the other hand , when asexual P . vivax was incubated with 4 different dilutions ( 4 . 28 , 8 . 56 , 17 . 12 and 34 . 24 ng/ml ) of plasma obtained from healthy volunteers 4 hours after IVM administration , there was a significant decrease in P . vivax maturation in relation to the drug free control and incubations with plasma from healthy volunteers collected before IVM administration . It is important to highlight in the present study that when the asexual P . vivax was incubated with the pIVM ( 3 . 9 – 1000ng/ml ) the development was not affected , however , a considerable reduction in blood-stage development was observed when the asexual stages were incubated with mIVM ( 4 . 28–34 . 24 ng/ml ) . This discrepancy might be a result of IVM metabolites conferring the parasite maturation inhibition effect . It is important to note that mIVM concentrations that showed blood-stage inhibition were achieved following oral administration with a standard dose of ivermectin ( 200 μg/kg ) . Unfortunately , data collected in this study could not be used to elucidate the ivermectin mechanism of action against asexual P . vivax . Further studies are warranted to evaluate the safety and efficacy of ivermectin as an adjunct during P . vivax antimalarial therapy . We also have assayed the P . vivax sensitivity to pCQ on the same isolates used to examine the asexual maturation inhibition with IVM . These results showed pCQ IC50 values ranging from 1 . 96 ng/mL to 7 . 53ng/mL , similar to other reports [66–69] , which demonstrate the chloroquine effect on P . vivax . Our findings also confirm the effect of chloroquine in terms of its pharmacodynamics against P . vivax . In conclusion , our study shows for the first time the effect of mIVM on the oocyst infection and intensity of P . vivax in the South American malaria vectors An . aquasalis and An . darlingi . In both vectors it appears that mIVM has a stronger sporontocidal effect compared to pIVM , this suggests that ivermectin metabolites have sporontocidal effect . We report for the first time , the effect of IVM on ex vivo cultures of P . vivax and demonstrate that mIVM can inhibit P . vivax development . Moreover , it provides evidence that IVM may affect several parameters of Ross-MacDonald model [70] , including parasite life cycle stages , placing it as a strong candidate for malaria transmission reduction .
|
Malaria is one of the most important infectious diseases in the world with hundreds of millions of new cases every year . The disease is caused by parasites of the genus Plasmodium where Plasmodium vivax represent most of the cases in the Americas . Current strategies to combat malaria transmission are being implemented; however , widespread insecticide resistance in vectors threatens the effectiveness of vector control programs . Ivermectin ( IVM ) has arisen as a new potential tool to be added to these programs as it has mosquito-lethal and sporontocidal properties making it a promising transmission reduction drug . Plasmodium vivax was drawn from patients , mixed with powdered IVM and metabolized IVM in plasma collected from healthy volunteers receiving IVM , and fed to mosquitoes via membrane feeding . Powdered and metabolized IVM interrupt P . vivax transmission , reducing oocyst infection and intensity rate of two South American malaria vectors An . aquasalis and An . darlingi . We also demonstrate the effect of IVM on asexual stages development of P . vivax , providing evidence that IVM may affect different parasite life cycle stages . Our findings place IVM as a strong candidate for malaria transmission reducing interventions .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"death",
"rates",
"invertebrates",
"parasite",
"groups",
"medicine",
"and",
"health",
"sciences",
"body",
"fluids",
"oocysts",
"plasmodium",
"tropical",
"diseases",
"parasitic",
"diseases",
"animals",
"parasitology",
"apicomplexa",
"pharmacology",
"population",
"biology",
"insect",
"vectors",
"drug",
"metabolism",
"infectious",
"diseases",
"disease",
"vectors",
"insects",
"blood",
"plasma",
"pharmacokinetics",
"arthropoda",
"population",
"metrics",
"mosquitoes",
"eukaryota",
"blood",
"anatomy",
"physiology",
"biology",
"and",
"life",
"sciences",
"species",
"interactions",
"malaria",
"organisms"
] |
2018
|
Promising approach to reducing Malaria transmission by ivermectin: Sporontocidal effect against Plasmodium vivax in the South American vectors Anopheles aquasalis and Anopheles darlingi
|
The sphingolipid ceramide elicits several stress responses , however , organisms survive despite increased ceramide but how they do so is poorly understood . We demonstrate here that the AKT/FOXO pathway regulates survival in increased ceramide environment by metabolic adaptation involving changes in glycolysis and lipolysis through novel downstream targets . We show that ceramide kinase mutants accumulate ceramide and this leads to reduction in energy levels due to compromised oxidative phosphorylation . Mutants show increased activation of Akt and a consequent decrease in FOXO levels . These changes lead to enhanced glycolysis by upregulating the activity of phosphoglyceromutase , enolase , pyruvate kinase , and lactate dehydrogenase to provide energy . A second major consequence of AKT/FOXO reprogramming in the mutants is the increased mobilization of lipid from the gut through novel lipase targets , CG8093 and CG6277 for energy contribution . Ubiquitous reduction of these targets by knockdown experiments results in semi or total lethality of the mutants , demonstrating the importance of activating them . The efficiency of these adaptive mechanisms decreases with age and leads to reduction in adult life span of the mutants . In particular , mutants develop cardiac dysfunction with age , likely reflecting the high energy requirement of a well-functioning heart . The lipases also regulate physiological triacylglycerol homeostasis and are important for energy metabolism since midgut specific reduction of them in wild type flies results in increased sensitivity to starvation and accumulation of triglycerides leading to cardiac defects . The central findings of increased AKT activation , decreased FOXO level and activation of phosphoglyceromutase and pyruvate kinase are also observed in mice heterozygous for ceramide transfer protein suggesting a conserved role of this pathway in mammals . These data reveal novel glycolytic and non-autonomous lipolytic pathways in response to increased ceramide for sustenance of high energy demanding organ functions like the heart .
Energy balance is regulated by complex homeostatic control of lipid and glucose metabolism involving signaling pathways in multiple tissues and organs . Aberrant glucose and lipid metabolism arising from defects in these pathways accompanies a number of diseases including cardiovascular disorders , type II diabetes and metabolic syndrome associated with obesity and insulin resistance . Understanding regulatory mechanisms governing glucose and lipid metabolism is fundamental for identifying potential therapeutic targets for treatment of these metabolic diseases . Altered lipid metabolism is one of the important contributors to obesity and insulin resistance . Dyslipidemia in obesity is characterized by elevated free fatty acids , triglycerides ( TAG ) , VLDL and decreased HDL cholesterol . Increased levels of free fatty acids and lipids such as diacylglycerol and ceramide mediate insulin resistance by redistribution of fat metabolites to tissues not suited for lipid storage [1] , [2] . The sphingolipid , ceramide , is an integral component of cell membranes and also a bioactive lipid [3] . Ceramide elicits many cell-stress responses including apoptosis , senescence , inflammation , mitochondrial dysfunction and recent studies show that it can affect cellular metabolism . It can impair insulin stimulated glucose uptake , it serves as an intermediate linking glucocorticoids and saturated fatty acids to insulin resistance and has been implicated in the pathogenesis of lipotoxic cardiomyopathy [4]–[6] . Strategies that pharmacologically or genetically decrease ceramide have beneficial effects in reversing insulin resistance , preventing apoptosis of pancreatic β-cells and cardiomyocytes [4] , [7] . Despite the involvement of ceramide in many stress responses , organisms develop and survive with increased ceramide levels in certain circumstances . For example , Drosophila mutants of ceramide kinase and brainwashing ( mammalian homolog of alkaline ceramidase ) are viable while mutants of ceramidase are lethal despite similar increase in ceramide levels [8]–[11] . Mice and Arabidopsis deficient in ceramide kinase and mice deficient in sphingomyelin synthase 1 and 2 are viable despite accumulation of ceramide [12]–[16] . A likely explanation is that these organisms implement adaptive responses that allow them to survive and maintain equilibrium in the face of stress due to increased ceramide . These adaptive mechanisms are of critical importance in understanding metabolic homeostasis , but have received limited attention especially in multi-cellular organisms . In this study , we address this issue using Drosophila loss of function mutation in ceramide kinase ( CERK ) . Our earlier studies established CERK as an important regulator of phospholipase C ( PLC ) signaling and photoreceptor homeostasis [10] . A mutation in dCERK led to proteolysis of PLC , consequent loss of PLC activity and failure in light signal transduction . These defects were due to increased ceramide levels and not ceramide 1-phosphate ( C1P ) since levels of C1P did not change significantly in photoreceptors of dcerk1 mutants . In recent years , Drosophila has emerged as a useful organism to study metabolism [17] . The fly has organ systems such as the fat body ( adipose tissue ) , oenocytes ( hepatocyte-like cells ) , gut ( gastrointestinal tract ) and malphigian tubules ( kidneys ) that parallel those in mammals . Several studies highlight the conserved mechanisms of carbohydrate , lipid and energy homeostasis in flies and genes discovered in Drosophila have provided information relevant to human physiology including that of the heart [18]–[27] . We set out to globally discover what genes and metabolites change in ceramide kinase mutant flies ( dcerk1 ) by gene expression and metabolic profiling . Integrating information from these analyses with genetic and biochemical experiments , we show that dcerk1 mutants survive due to increased activation of AKT and decreased FOXO level . In the dcerk1 mutants , increase in ceramide leads to decreased ATP level due to compromised mitochondrial oxidative phosphorylation . Metabolic reprogramming to ceramide involves AKT induced glycolytic utilization of glucose through activation of phosphoglycerate mutase and also utilization of triglycerides by activation of two intestine specific lipases , CG8093 and CG6277 in dcerk1 . These are novel downstream targets of AKT/FOXO in glycolysis and lipolysis . With age , these compensatory mechanisms fail leading to abnormal cardiac function and decreased life span of the mutants . Gut specific reduction in these lipases in both wild type and mutant flies results in TAG accumulation , increased sensitivity to starvation and cardiac dysfunction . Using mice heterozygous for ceramide transfer protein ( CERT ) as a second model of ceramide increase , we substantiate AKT activation , decrease in FOXO level and activation of glycolytic genes , phosphoglycerate mutase and pyruvate kinase in a mammalian system [28] . We have thus discovered a novel connection in the PI3K-AKT-FOXO pathway for survival and metabolic adaptation to a ceramide environment that participates in energy response involving phosphoglycerate mutase and gut specific lipases , CG8093 and CG6277 as physiological targets .
Ceramide kinase mutants show an increase in steady state concentration of ceramides without significant decrease in the levels of ceramide 1-phosphate in whole fly extracts ( Figure S1 ) . The increase in C24:1 ceramide is most dramatic ( about 280% ) while other ceramides show between 30–50% increase in the dcerk1 mutants over control flies . Thus dcerk1 mutants will have to survive in an increased ceramide environment . An organism's attempt to adapt to stress or altered environment involves changes in gene expression and metabolite levels that aid in reestablishing homeostasis . Therefore , we reasoned elucidating both transcriptional and metabolite responses that change in mutants relative to control could help us understand possible adaptive mechanisms that operate in dcerk1 . We carried out a transcriptome-wide analysis in w1118 and dcerk1 flies . 310 transcripts were changed significantly in dcerk1 compared to control , with 152 genes being upregulated ( > = 2 . 0 fold ) and 158 genes being downregulated ( < = 0 . 5 fold , Table S1 ) . To identify pathways and predicted gene functions corresponding to the altered transcripts , we employed the DAVID gene ontology ( GO ) annotation and functional classification tools [29] . The top GO class for increased genes included serine type endopeptidases and hydrolases ( Figure 1A ) . Most of the serine type endopeptidases are mainly expressed in the larval/adult mid gut ( based on FlyAtlas expression patterns ) suggesting they are likely involved in the mobilization of dietary protein . Among the increased hydrolases were glycosyl hydrolases and those involved in metabolism of starch , sucrose and galactose and these are also expressed mainly in the adult gut . Thus , there is a significant upregulation of midgut genes involved in breakdown of dietary proteins and sugars in dcerk1 . Other classes of upregulated genes in dcerk1 were transporters , oxidoreductases and genes that constitute the peritrophic envelope , a lining composed of chitin and glycoproteins , separating the food from the midgut epithelium . The top GO category for downregulated genes were antimicrobial peptides ( AMPs ) involved in the innate immune response . Important metabolic genes that change in dcerk1 flies are phosphoenolpyruvate carboxykinase ( PEPCK , downregulated ) involved in gluconeogenesis , fatty acid synthesis ( downregulated ) , carnitine palmitoyltransferase ( β-oxidation , upregulated ) and pyruvate kinase ( upregulated ) involved in glycolysis . Since gene expression changes in known targets of FOXO such as PEPCK and AMPs were observed , we analyzed if other differentially expressed genes in our microarray dataset could be potential targets of FOXO . Recent ChIP-chip analysis revealed about 700 direct FOXO targets in adult Drosophila [30] . Using this information coupled with ChIP peak Annopackage we identified that 40 genes from our microarray data contained FOXO binding sites and could be potential targets of FOXO ( Table S2 ) [31] . In order to understand how changes in gene expression influence metabolism , we compared the metabolic profiles of w1118 and dcerk1 by mass spectrometry . About 175 metabolites were identified and classified into different functional pathways ( Table S3 ) . The fold changes ( dcerk1/w1118 ) for various amino acids , fatty acids , sugars and intermediates in glycolysis , TCA cycle , are shown in Figure 1B . Two important metabolic pathways that were different between w1118 and mutants were glycolysis and TCA cycle . dcerk1 exhibited increase in lactate and decrease in pyruvate , both glycolytic intermediates . While oxaloacetate and α-ketoglutarate were decreased , other TCA cycle intermediates , citrate , malate , fumarate and succinate were increased . The altered steady state concentrations of the metabolites indicate change in the flux of these pathways . Glycolysis , TCA cycle and mitochondrial oxidative phophorylation share control of energy metabolism and they interact to match cellular ATP demand with ATP production . To assess mitochondrial oxidative phosphorylation , we measured the enzyme activities of the five complexes of the electron transport chain in mitochondrial-enriched fractions from control and mutant flies . As seen in Figure 2A , the activities of complexes II , III , IV and V are decreased while I is increased in the mutant compared to control . To test if compromised oxidative phosphorylation led to alteration in energy level , we measured ATP level in mitochondria isolated from w1118 and mutant flies . Indeed , dcerk1 showed a 40% decrease in ATP level relative to w1118 ( Figure 2B ) . Since metabolic profiling emphasized changes in glycolysis , we first examined if there are changes in gene expression of enzymes of the glycolytic pathway in addition to pyruvate kinase , which was shown to be upregulated by microarray analysis . Quantitative PCR analysis of the different glycolytic genes showed a 15-fold increase in phosphoglycerate mutase ( Pglym ) , 3–4 fold increase in both enolase ( Eno ) and pyruvate kinase ( Pyk ) expression , enzymes that catalyze later reactions in the glycolytic pathway ( Figure 2C ) . Interestingly , traditional regulators of glycolysis such as hexokinase and phosphofructokinase are either unchanged or decreased . To test if increase in transcript levels translated to increased activity , we measured activity of the various enzymes . Pglym , Pyk , Eno and Lactate dehydrogenase ( Ldh ) showed increase in enzyme activity ( Figure 2D ) . Metabolically these changes in transcript and enzyme activity result in a 1 . 7 fold increase in lactate in dcerk1 ( Figure 1B ) . Thus a significant portion of glucose metabolized by the glycolytic pathway was converted to lactate by the action of lactate dehydrogenase . The above results suggest that the increased reliance on glycolysis by dcerk1 compensates for their decreased ability to produce ATP through oxidative phosphorylation . Since dcerk1 mutants show specific changes in glycolysis and differentially expressed genes in our transcriptional profiling data contained FOXO binding sites , we decided to test if AKT , a central modulator of metabolism and upstream regulator of FOXO was altered in dcerk1 . AKT is a master regulator of survival , proliferation and cellular metabolism [32] , [33] . AKT activation requires TOR dependent phosphorylation of Ser-473 ( Ser-505 in Drosophila ) at the C-terminus [34] , [35] . We measured AKT activation in dcerk1 flies by measuring the level of Ser-505 phosphorylated AKT protein by Western analysis . Phosphorylated AKT ( seen as two major and one minor band ) was increased while total AKT level was not different from w1118 ( Figure 3A ) . The blot also shows reduced levels of total and phospho AKT in akt4226 , a hypomorphic allele of AKT used in this study [36] . Densitometric scanning of Western blots shows about 50% increase in pAKT level in the dcerk1 mutant ( Figure 3B ) . To test if activity of the upstream activator PI3K is increased , we used an in vivo reporter , which contains the PH domain of general receptor for phosphoinositide fused to GFP ( tGPH ) [37] . tGPH is increased in the cell membrane when PI3K activation increases PIP3 level . We observed an increased membrane association of tGPH in dcerk1 compared to control in the adult mid gut ( Figure 3C , top panel ) and larval fat body ( Figure 3C , bottom panel ) . These results suggest increased PI3K activation leads to increased AKT activation in dcerk1 . A likely reason for the increased activation of PI3K could be the increased availability of PIP2 in the dcerk1 mutant [10] . AKT mediates many of the alterations in metabolism through regulation of FOXO and in Drosophila there is a single homolog of the FOXO transcription factor family [38]–[41] . To test the status of FOXO , we measured FOXO transcript level in mutant flies and found that it was significantly decreased ( Figure 3D ) . AKT activation in mammalian cells has been shown to lead to low FOXO level due to ubiquitination and proteosomal degradation of FOXO [42] . To address this possibility in Drosophila , we reduced ubiquitin mediated proteosomal degradation in dcerk1 using a dominant temperature sensitive ( DTS ) mutation , DTS5 that affects the β6 proteosomal subunit [43] . Since ubiquitous expression of DTS5 resulted in lethality , we overexpressed DTS5 in photoreceptors using the GMR GAL4 driver and measured FOXO transcript level in heads . If FOXO is targeted for degradation in dcerk1 , then expressing DTS5 should result in restoration of FOXO level and this is indeed the case ( Figure 3D ) . Thus increased activation of AKT leads to downregulation of FOXO transcription factor in dcerk1 mutant flies . To evaluate the importance of increased AKT activity in dcerk1 homozygous mutants , we generated akt4226 , dcerk1 double homozygotes . While dcerk1 or akt4226 homozygous flies alone are viable , double mutants are early larval lethal . Hence , compromised AKT function was incompatible with survival of dcerk1 flies . To test if AKT may be required for the increased glycolytic flux in mutants , we generated dcerk1 flies with reduced AKT function wherein one wild type copy of AKT was replaced with the hypomorphic allele of AKT . dcerk1 flies heterozygous for akt 4226 ( referred to as dcerk1 . akt/dcerk1 ) survive to adulthood and we have used them in subsequent experiments for evaluating the effects of reducing AKT function in dcerk1 . The data from our studies with Pglym are shown here since it showed significant increase in both transcript and enzyme activity in mutant flies ( Figures 3E , 3F ) . dcerk1 . akt/dcerk1 flies showed reduction in Pglym transcript and enzymatic activity . akt4226/+ flies were also tested in the above experiments and they did not differ significantly from w1118 control suggesting that heterozygous state of AKT alone was not responsible for the changes ( Figures 3E , 3F ) . Densitometric scanning of Western blots reflects a 50% decrease in total AKT and 30% decrease in pAKT levels in the akt4226/+ ( Figures S2A , S2B ) . As mentioned earlier , AKT alters metabolic balance through downregulation of FOXO . We tested its involvement in regulating Pglym , by overexpressing FOXO™ , a constitutively active form of the protein that localizes to the nucleus [44] . We expressed FOXO™ using a muscle driver , myosin heavy chain ( MHC ) -GAL4 and analyzed whole body Pglym transcript level [45] . Indeed , FOXO™ overexpression in the muscles resulted in significant decrease in Pglym transcript ( Figure 3G ) , suggesting Pglym regulation is FOXO dependent . The measurement of steady state levels of glucose ( mammals ) and trehalose ( Drosophila ) is an indicator of glycolytic utilization of sugars for energy metabolism . Therefore , we tested if AKT is required for maintaining normal trehalose levels in dcerk1 mutants . Trehalose levels in fly extract of dcerk1 were 90% of the control flies , whereas a reduction of akt dosage in dcerk1 mutants resulted in an increase in trehalose ( Figure 3H ) . Circulating trehalose levels also followed a similar trend as the whole body trehalose measurements in these flies ( data not shown ) . We then globally reduced Pglym expression by ubiquitous RNAi mediated knockdown using the actin-Gal4 driver ( Figure S3 ) . Notably , this reduction resulted in semi lethality with 40% of the expected number of flies reaching adulthood in dcerk1 . Trehalose levels increased by about 20% in w1118 and by 50% in dcerk1 when Pglym levels were reduced ( Figure 3I ) . Hence , our results suggest increased glycolytic flux through Pglym is not only important for survival but also for maintaining near physiological levels of trehalose in dcerk1 mutants . Our experiments thus far suggest that dcerk1 flies attempt to adapt to reduced energy availability . Triacylglycerol ( TAG ) is one of the major energy reserves , which is stored as cytoplasmic lipid droplets primarily in the adipose tissue [46] . In Drosophila , in addition to fat body cells , other tissues , most notably , the gut , oocytes , larval imaginal discs and oenocytes also accumulate TAG as intracellular lipid droplets [47] . During times of energy need such as starvation , stored TAG is hydrolyzed from these depots by activation of lipolytic mechanisms [48] . Fatty acids generated via lipolysis by lipases can be used for energy production through β-oxidation . Stored lipids are transported through the hemolymph as lipoprotein particles called lipophorins . So , we decided to test if energy rich TAG stores are altered in dcerk1 and if so , does the AKT/FOXO pathway play a role in this process . We measured whole animal TAG levels in ad libitum fed and 20 h starved w1118 and dcerk1 flies . The mutant flies showed 20% less TAG than w1118 in the fed state and 50% less TAG in the starved state ( Figure 4A ) . The decrease in TAG in the starved state is accompanied by increased free fatty acid and glycerol in whole fly extracts ( Figures 4B , 4C ) . Since TAG levels were significantly altered in the starved state , we assessed the ability of dcerk1 to survive starvation stress . As shown in Figure 4D , the susceptibility of mutant flies to starvation was similar to w1118 . These results suggest that dcerk1 mutant likely rely on increased TAG hydrolysis to survive starvation stress . To test if AKT mediated compensation allowed dcerk1 to survive starvation , we subjected dcerk1 . akt/dcerk1 flies to starvation stress . These flies died at a faster rate than dcerk1 showing reduction of AKT function in dcerk1 led to starvation sensitivity ( Figure 4D ) . TAG level in these flies was significantly higher than in dcerk1 , both in the fed and starved states ( Figure 4A ) . The starvation sensitivity of these flies despite increased TAG level suggested that they were unable to properly hydrolyze their TAG stores via lipolysis when AKT activity is reduced . Thus AKT dependent lipases may provide a critical function in dcerk1 flies . In order to identify lipases that are activated in dcerk1 , we first tested transcript levels of known fat body lipases in Drosophila , which include Brummer ( ATGL homolog ) , CG11055 ( putative homolog of hormone sensitive lipase ) and CG8552 ( homolog of major triglyceride lipase in a related insect ) [19] , [48] . The transcript level of Brummer was decreased in dcerk1 ( 0 . 5 fold compared to control ) while that of the other two lipases did not show changes ( data not shown ) . Thus traditional lipases were not upregulated for increased TAG utilization in dcerk1 . We then pursued potential candidates from microarray studies . Our microarray analysis identified that amongst the 60 lipases that have been identified in Drosophila , expression levels of two genes , CG8093 and CG6277 , that encode predicted TAG lipases were increased in dcerk1 . CG8093 is an acid lipase of the α/β hydrolase family related to human lysosomal acid lipase while CG6277 is a neutral lipase showing homology to lipoprotein lipase and to pancreatic triacylglycerol lipase family . Both CG8093 and CG6277 contain the catalytic triad of Ser , His , Asp residues important for lipase activity . Earlier studies involving microarray analysis in Drosophila demonstrated that these lipases are repressed by high sugar and in situ hybridization studies showed they were expressed in the midgut [49] . Our QPCR results also showed that these genes were highly expressed in the midgut ( Figure S4 ) . To test if these lipases could facilitate TAG utilization in dcerk1 , we first examined their transcript levels in fed and starved states ( Figure 4E ) . Both genes were upregulated in dcerk1 . This is an AKT mediated adaptive response since the transcript levels were downregulated when AKT is reduced in dcerk1 . To evaluate the involvement of FOXO , we overexpressed FOXO™ using gut ( esg-GAL4 ) and fat body ( C564-GAL4 ) specific GAL4 drivers [50] , [51] . Gut specific overexpression of FOXO™ decreased whole animal lipase transcripts while fat body GAL4 did not alter expression significantly in dcerk1 ( Figure 4F ) . Overexpression of FOXO™ in the gut also resulted in higher TAG level in dcerk1 compared to control ( Figure 4G ) . The expression of these lipases could be regulated directly or indirectly by FOXO . Our earlier analysis identified FOXO binding sites upstream of CG8093 ( Table S2 ) and thus could be a potential target of FOXO . To test if negative regulation of CG8093 by FOXO is mediated through these sites , we performed luciferase reporter assays . A 2 . 4 kb fragment containing two FOXO motifs from the regulatory region of CG8093 ( pGL4CG8093 2 . 4 kb ) and a fragment in which the motifs were deleted ( pGL4CG8093 1 . 3kbdel ) were cloned in a reporter vector containing the luciferase gene ( pGL4 ) . Drosophila S2 cells were cotransfected with constitutively active FOXO along with each of these constructs and luciferase activity was measured after 30 h [40] . Constitutively active FOXO suppresses the luciferase activity of pGL4CG8093 2 . 4 kb and this is considerably relieved in pGL4CG8093 1 . 3kbdel , which is deficient of the FOXO binding sites ( Figure 4H ) . This suggests FOXO could act via these sites to regulate CG8093 transcription . Our search for FOXO binding sites was limited to 10 kb upstream of the microarray target genes and CG6277 was not identified in this analysis . The promoter region of CG6277 does not contain obvious FOXO binding sites suggesting transcriptional regulation either takes place through the interaction with other transcription factors or is indirectly mediated through FOXO dependent induction of a transcriptional repressor protein . A possible candidate could be sugarbabe , which encodes a zinc finger protein . An earlier study that analyzed sugar dependent gene expression profiles in Drosophila larvae identified CG6277 could be negatively regulated by sugarbabe [49] . Our analysis of sugarbabe transcript in dcerk1 mutants suggests its level is low , which would be consistent with the observed upregulation of CG6277 transcript level . Also , analysis of FOXO binding sites identifies two potential regions within 10 kb upstream of sugarbabe ( data not shown ) . Thus a possible mode of regulation could be FOXO activates sugarbabe , which in turn negatively regulates CG6277 . Thus downregulation of FOXO in dcerk1 could lead to activation of CG6277 . These results suggest that increase in CG8093 and CG6277 through AKT/FOXO could lead to increased lipolysis and decreased TAG level in dcerk1 . To examine whether increased hydrolysis of fat from the gut did occur as a result of increased lipase activity , guts isolated from w1118 , dcerk1 and dcerk1 . akt/dcerk1 were stained with the neutral lipid dye Oil Red O under starved conditions ( Figure 4I ) . Oil Red O staining was evident in w1118 gut while dcerk1 showed little staining . Midgut isolated from dcerk1 . akt/dcerk1 also stained well with Oil Red O ( Figure 4I ) . Oil Red O puncta representing lipid droplets were clearly visible in w1118 and dcerk1 with reduced AKT but were substantially decreased in dcerk1 midgut ( Figure 4I , bottom panel ) . To ensure that changes in gene expression and increased utilization of lipid in the gut are not to due to compromised gut integrity or defective epithelial renewal , we tested if dcerk1 gut were morphologically different from w1118 gut . We observed that gut length , width and ability to ingest food were not different between control and mutant flies ( data not shown ) . Epithelial integrity and epithelial renewal were also assessed as described in Figure S5 and these data suggest that gut integrity or epithelial renewal in the gut is largely uncompromised in dcerk1 . Our data suggest that increased expression of CG8093 and CG6277 leads to increased lipolysis and a concomitant decrease in TAG level in dcerk1 . To ascertain this observation , we asked if impairing the function of these lipases in dcerk1 would lead to accumulation of TAG in the mutant . We attempted to reduce their levels by ubiquitous expression of UAS-CG8093 RNAi and UAS-CG6277 RNAi constructs in dcerk1 . However these conditions resulted in lethality of both wild type and dcerk1 flies . Since these lipases are expressed predominantly in the gut , we next drove the RNAi constructs using the midgut specific driver esg-GAL4 and were able to obtain adult flies . RNAi resulted in 80% reduction in the transcript levels of the lipases in control and dcerk1 ( Figure S6 ) . Midgut specific knockdown of each of these lipases resulted in significant increase in whole animal TAG levels in dcerk1 ( Figures 5A , 5B ) . Reduction of these lipases also resulted in increased TAG in control flies ( Figures 5A , 5B ) . Gut specific reduction of the lipases also resulted in increased sensitivity to starvation in both wild type and mutant background compared to GAL4 driver and RNAi control flies ( Figures 5C , 5D ) . Staining of guts isolated from knockdown flies with Oil Red O showed increase in the number of Oil Red O puncta indicative of defective mobilization of lipid stores ( Figure S7 ) . These results collectively suggest that both lipases are important for utilization of TAG and maintenance of TAG homeostasis not only in the mutant but also in wild type flies . Additionally , the life span of wild type and dcerk1 flies is significantly reduced upon gut specific knockdown of CG6277 ( data not shown ) . The heart has a high demand for energy and relies on fatty acids derived from lipolysis of lipoprotein bound TAG by lipases as energy substrates [52] . Cardiac dysfunction is linked to altered energy metabolism in many instances [53] . To test if the two lipases played a role in this process , we tested if their reduction in wild type flies compromised cardiac function using a variety of heart performance assays that have been developed in Drosophila [21] . Heart function was compromised in one-week old gut specific knockdown of CG6277 as manifested by decreased heart rate or increased heart period primarily due to an increased diastolic interval ( Figure 5E ) . Knockdown of CG6277 in dcerk1 mutant flies also resulted in increased heart period and a trend towards decreased heart rate ( Figure 5E ) . One-week old dcerk1 alone did not show significant changes in heart parameters , and only a slight trend towards a slower heart rate and dilation ( increased systolic and diastolic diameters ) . While gut specific knockdown of CG8093 in wild type and in dcerk1 flies did not show significant changes in most of the parameters tested for heart function , heart period is affected in dcerk1 due to expansion of diastolic interval , pointing to the importance of this lipase for heart function in dcerk1 flies ( Figure S8 ) . Since the knockdown of lipases resulted in increased sensitivity to starvation and cardiac defects , we also tested the effects of downregulating the glycolytic gene , Pglym in sensitivity to starvation and cardiac functions in wild type and mutant backgrounds . Ubiquitious knockdown of Pglym results in a significant increase in sensitivity to starvation stress in wild type and dcerk1 ( Figure S9 ) . Evaluation of various heart function parameters with Pglym knockdown flies in both wild type and dcerk1 mutant backgrounds showed a decrease in diastolic and systolic diameters resulting in a somewhat constricted phenotype without significant changes in other parameters ( Figure S9 ) . This is reminiscent of cardiac constricted phenotype of eas mutants , which show aberrations in phosphatidylethanolamine synthesis and elevated SREBP activity [26] . To test if the adaptive mechanisms we observe in young dcerk1 flies ( less than one week ) is sustained or fails with age , we tested for AKT activation and measured FOXO transcript level in aged dcerk1 ( 6 weeks ) . We observe a decrease in AKT activation and 4-fold increase in FOXO transcript level in older flies ( data not shown ) . Consequently , Pglym activity decreases significantly in old dcerk1 flies ( Figure 6A ) leading to increased trehalose level ( data not shown ) . Similarly , transcript levels of both lipases decrease with age ( Figure 6B ) , leading to 50% increase in TAG level in old dcerk1 ( data not shown ) . These results suggest the efficiency of the compensatory mechanisms decrease with age in the mutant flies . The decompensation leads to deterioration in many indices of heart performance between young and three-week old dcerk1 flies ( Figure 6C ) , a time point at which we begin to observe noticeable decrease in transcript levels of the lipases in the mutant . These changes follow a similar trend as those observed when CG6277 lipase is knocked down in healthy three-week old wild type flies ( Figure 6D ) . In both these situations in addition to heart rate and heart period changes , we also observe dilated heart chamber , perhaps due to increased dependence of heart function on this lipase with aging . We chose an interim period of three weeks because CG6277 lipase knockdown flies do not survive for six weeks . Thus , a decrease in CG6277 compromises cardiac function likely due to altered energy balance resulting from decreased TAG hydrolysis as observed in wild type flies when the transcript levels are reduced by RNAi or in aging dcerk1 mutants when AKT mediated compensatory changes begin to fail . We then tested whether impaired mitochondrial function and age related pathology such as triglyceride accumulation and decline in cardiac function impacted the life span of dcerk1 . While 90% of the mutants die by day 60 , only 20% of the controls are dead by this time , suggesting the adult life span of the mutant flies is shortened ( Figure 6E ) . We finally asked whether the adaptive mechanisms to increased ceramide identified here using Drosophila dcerk mutants are likely important in other scenarios where ceramide is increased . One example is the mouse ceramide transfer protein mutant , which accumulates ceramide and dies around embryonic day 11 . 5 [28] . These embryos show severe mitochondrial degeneration , decreased phospho AKT level and cardiac defects suggesting a likely failure of the adaptive mechanism . CERT heterozygous mice are viable , show mitochondrial damage and stress but not as severe as the homozygotes . We tested if our central results of AKT activation and decrease in FOXO level are observed in CERT heterozygous mice . Indeed , we see increased phosphoAKT level and decreased levels of different FOXO transcripts ( Figures 7A , 7B ) . While results from liver extracts are shown here , down regulation of FOXO is also observed in other tissues such as intestine and skeletal muscle . We also observe an increase in transcripts of phosphoglycerate mutase and pyruvate kinase , suggesting a likely increase in glycolysis in these animals ( Figure 7C ) . We then tested some of the mammalian lipase family members that showed homology to CG8093 ( LIPA , LIPF , LIPM ) and CG6277 ( pancreatic lipoprotein 1 , pancreatic lipoprotein related protein , lipoprotein lipase , endothelial lipase ) but could not detect changes in transcript levels . The lack of validation of targets in triglyceride utilization in the mice suggests that CERT heterozygotes may not be the ideal model for lipolytic mechanisms and also reflects the complexities involved in lipid sensing in the gut and coordination with other organs in the mammalian systems .
In this study , we have attempted to understand how dcerk1 flies survive and adapt to increased ceramide level , which disrupts several of the physiological metabolic pathways . Our results collectively suggest that in the presence of increased ceramide , the most important coping strategies for organismal survival are geared towards energy production through metabolic remodeling involving upregulation of genes that function in breakdown of dietary nutrients , increased glycolysis and utilization of stored fat in the gut . This is due to increased activation of PI3K/AKT/FOXO pathway and subsequent activation of novel downstream effectors ( summarized in a model , Figure 7D ) . Failure of this compensatory mechanism ( due to aging or genetic reduction ) leads to TAG accumulation , increased starvation sensitivity and cardiac abnormalities . While dcerk1 flies are viable , genetic reduction of AKT results in lethality showing survival of dcerk1 is dependent on increased activation of AKT . AKT activation leads to increased glycolytic flux in the mutants . Increase in lactate level in dcerk1 ( in the presence of sufficient oxygen ) is reminiscent of aerobic glycolysis observed in normal proliferating cells as well as in cancer cells . In normal proliferating cells , increasing glycolytic intermediates has been proposed to support macromolecular synthesis while a similar phenomenon called Warburg effect is observed in cancer cells where glycolysis is increased to provide energy to cells defective in mitochondrial respiration [54] , [55] . AKT does not activate its traditional targets of glycolysis such as hexokinase and phosphofructokinase in dcerk1 but rather Pglym and downstream glycolytic enzymes . Historically Pglym has not been viewed as a rate-limiting step in glycolysis although studies have suggested its importance in heart , phagocytic cells and cancer cells [56] , [57] . Thus , within the glycolytic pathway , control of metabolic flux can be achieved by exerting control on different enzymes depending on cellular conditions and Pglym could be an important node in normal proliferating as well as in stressed cells . Addition of ceramide analogs has been reported to inhibit AKT activation without affecting upstream signaling events [58] , [59] . Possible mechanisms include activation of protein phosphatase 2A , which could dephosphorylate AKT and interference with membrane recruitment of AKT by atypical protein kinase C [59] , [60] . These effects could in part be due to use of pharmacological amounts of ceramide analogs , which likely do not mimic endogenous ceramide . Nevertheless , these effects could be negated by overexpression of constitutively active AKT [59] . How is AKT activation achieved in dcerk1 ? While several mechanisms could contribute to AKT activation , an idea we favor is that more PIP2 is available for PI3K since utilization of PIP2 by PLC is severely decreased due to its degradation in dcerk1 [10] . While we see an increase in tGPH suggesting an increase in PIP3 , we have not been able to quantify PIP3 level by mass spectrometry . However , PIP2 level shows 60% increase in dcerk1 compared to control [10] . Studies in Drosophila have highlighted the role of Brummer lipase ( ATGL homolog ) in fat storage and mobilization from the fat body and in mice have shown ATGL mediated lipolysis is important for PPARα-PGC1 complex activity and heart function [19] , [61] . Magro , a target of the nuclear receptor DHR96 , hydrolyzes dietary lipids in the gut as well as stored cholesterol esters highlighting the importance of the Drosophila gut in maintaining TAG and cholesterol homeostasis [62] , [63] . In mammalian systems , understanding of lipid sensing by the intestine and its coordination with other organs such as the liver , brain and the signaling pathways in these organs that alter glucose and energy homeostasis are just beginning to be explored [64] . Our results suggest utilization of stored TAG from the gut is an important mechanism in times of energy need such as starvation stress and CG8093 and CG6277 are critical targets of AKT/FOXO in this process . In a Drosophila larval model of type 2 diabetes that is accompanied by high sugar and TAG levels , gene expression changes reveal that CG8093 and CG6277 are both significantly downregulated ( 15 and 5 fold respectively ) suggesting these lipases could have a role in hypertriglyceridemia associated with diabetes [65] . Metabolic control of lipolysis has been primarily studied at the level of nutrient-sensing signal transduction cascades rather than transcriptional regulation and hence not much is known about the regulation of lipases by FOXO . In Drosophila , dFOXO has been shown to directly regulate an acid lipase , dlip4 [66] . Recent studies show a hormone dependent module consisting of salt induced kinase SIK3 , the histone deacetylase HDAC4 , which regulate FOXO activity in lipolysis [67] . In C . elegans , fat mobilization through induction of a triglyceride lipase K04A8 . 5 , a target of DAF16 increases longevity [68] . In mammalian systems , FOXO has been shown to regulate lipolysis by either directly controlling expression of ATGL or through PPARγ [69] . A forkhead factor , FKHR has been shown to upregulate lipoprotein lipase expression in the skeletal muscle [70] . While FOXO binding sites have been identified in the promoter region of lipoprotein lipase , that it can modulate its expression directly has yet to be demonstrated [71] . Further understanding of the role of FOXOs in different organs will allow the exploration of the therapeutic value of targeting FOXO in metabolic diseases . Many recent studies show sphingolipids could be relevant to changes in carbohydrate and fat levels . In Drosophila , overexpression of glucosylceramide synthase that catalyzes synthesis of glucosylceramide increases TAG and carbohydrate levels while a reduction causes decrease in fat storage [72] . Drosophila ceramide synthase mutants also show reduction in fat storage [73] . In mammalian systems , mice lacking sphingomyelin synthase and ceramide kinase ( both of which show increase ceramide levels ) are resistant to high fat diet induced obesity [74] , [75] . It would be interesting to test if these animals show increased lipolytic mechanisms based on our observations in Drosophila . Our results with the dcerk mutant presented here as well as other mutants of Drosophila sphingolipid enzymes ( our unpublished observations ) suggest that an increase in ceramide in one cellular compartment eventually leads to an increase in ceramide in the mitochondria suggesting that a central effect of ceramide accumulation in an organism is impairment of mitochondrial function . This impact on mitochondria is also being validated in recent mice knock out studies such as the sphingomyelin synthase 2 and CERT mutants , which show mitochondrial dysfunction [16] , [28] . The mechanisms outlined here could not only mediate adaptation to ceramide , but could be important responses to reduced energy availability . The ‘protective effects’ of the adaptive mechanisms revealed here suggest that they have potential therapeutic implications in ceramide mediated stress involving decreased energy leading to high sugar and TAG levels , ultimately resulting in cardiac dysfunction . Also , importantly , the identification of novel downstream targets in the AKT/FOXO pathway opens the possibility of new therapeutic targets in treatment of cardiac dysfunction , hyperglycemia and hypertriglyceridemia .
Drosophila stocks were raised on standard corn meal agar and maintained at 25°C . 2–7 day old flies were used in all experiments unless otherwise indicated . Akt4226 , actin-GAL4 driver , MHC-GAL4 driver and tGPH flies were obtained from Bloomington Stock Center ( Indiana University ) . UAS-FOXO was a gift from Marc Tatar ( Brown University , Providence ) , esg-Gal4 from Norbert Perrimon ( Harvard Medical School , Boston ) C564-GAL4 flies from Neal Silverman ( UMass , Worcester ) and DTS flies were from Eric Baerahcke ( UMass , Worcester ) . UAS RNAi lines for CG8093 , CG6277 and Pglym were obtained from VDRC ( Vienna ) . Akt4226 was recombined to dcerk1 to obtain dcerk1 . akt/dcerk1 flies . UASRNAi lines for CG8093 , CG6277 and Pglym on the third chromosome were recombined to dcerk1 to obtain dcerk1 . UASCG8093RNAi , dcerk1 . UASCG6277RNAi and dcerk1 . UASPglymRNAi lines . UAS-FOXO , esg-GAL4 , actin-GAL4 , C564-GAL4 and MHC-GAL4 drivers on second chromosome were used . dcerk1 and w1118 flies were collected , and RNA was extracted using TRIzol ( Invitrogen ) and purified on RNeasy columns ( QIAGEN ) . All samples were prepared in triplicate to facilitate subsequent statistical analysis . Purified RNA was sent to Expression Analysis ( Durham , NC ) , where it was labeled and hybridized to Affymetrix Drosophila 2 . 0 microarrays . The details of the analysis are described in Supplemental Materials and Methods ( Text S1 ) . dcerk1 and w1118 ( 100 flies each , in triplicate ) were collected and frozen . The samples were prepared and analyzed by LC/MS , LC/MS/MS and GC/MS platforms by Metabolon ( Durham , NC ) . TAG measurement was carried out using the Serum triglyceride determination kit ( TR0100; Sigma , St . Louis , MO ) . Total body and circulating trehalose were measured using a glucose reagent ( GAGO-20 , Sigma ) followed by porcine kidney trehalase ( Sigma , T8778 ) . ATP measurements were carried out using an ATP assay kit ( Calbiochem ) . Glycerol and free fatty acid were determined using coupled enzyme assay kits ( Zenbio , NC ) . Detailed methodology is provided in Supplemental Materials and Methods ( Text S1 ) . In each experiment , 10 batches of 20 flies for each genotype were transferred to vials containing a small piece of sponge soaked with water . Dead flies were counted at 6 h intervals . Newly eclosed adult w1118 and dcerk1 were collected over a 24 hr period and divided into batches of 20 flies per vial . Flies were transferred to fresh food vials and scored for survival every 48 h . Semi-intact fly heart preparations were used for heart parameter measurements according to published protocols [20] . High speed 1 min movies were taken at a rate of >100 frames per second using a Hamamatsu CCD camera on a Leica DM LFSA microscope with a 10× dipping immersion lens . The images were processed using Simple PCI imaging software ( Hamamatsu Inc . ) . Cardiac parameters including heart rate , heart period , diastolic and systolic intervals , diameters , fractional shortening and arrhythmic index were generated using a MatLab-based image analysis program . For Oil red O staining , midguts from 2–7 days old adult flies were dissected and fixed in 4% paraformaldehyde/PBS for 30 min at room temperature . The midguts were washed four times in PBS and incubated for 20 to 30 min in 0 . 1% Oil Red O stain ( prepared fresh in isopropanol∶water and passed through a 0 . 45 µm syringe filter ) . Midguts were washed three times with PBS , and mounted in 20% glycerol/PBS . The guts were imaged immediately using Differential Interference Contrast microscopy in a Nikon Eclipse E-600 microscope . For visualization of tGPH , adult midgut or fat body from early third instar larvae were dissected in PBS and visualized immediately using Nikon Eclipse E-600 microscope and NIS-Elements Imaging Software . Total RNA was extracted using TRIzol reagent and cDNA was synthesized using the SuperScript III first–Strand Synthesis kit ( invitrogen ) . Real time PCR was performed in the ABI PRISM 7000 sequence Detection System ( Applied Biosystems ) using SYBR Green Supermix ( Invitrogen ) . Reactions were normalized to RP49 levels . Detailed methodology and primers used are provided in Supplemental Materials and Methods ( Text S1 ) . Flies were crushed in SDS-sample buffer; the extracts were centrifuged and separated by SDS-PAGE followed by Western blotting using antibodies to AKT ( 1∶1000 , Cell Signaling ) and phospho Drosophila AKT Ser505 ( 1∶1000 , Cell Signaling ) . Soluble extracts were prepared from approximately 1000 flies per batch using different buffers for different enzymes and spectrophotometric assays were carried out as detailed in Supplemental Materials and Methods ( Text S1 ) . Mitochondria were isolated from approximately 1000 flies per batch . The activity of each of the five complexes was determined spectrophometrically as detailed in Supplemental Materials and Methods ( Text S1 ) .
|
Ceramide belongs to an important class of lipids called sphingolipids . An increase in ceramide levels elicits several stress responses . We have used Drosophila mutated in ceramide kinase , an enzyme involved in metabolizing ceramide to understand how an organism adapts to stress imposed by ceramide . We find that in the mutants , increased ceramide decreases energy production via mitochondrial oxidative phosphorylation . The mutants compensate for decreased energy by upregulating glycolysis , a central pathway of intermediary metabolism . Another important mechanism that is activated in these ceramide kinase mutants is lipolysis , which is involved in breakdown of triglycerides for energy contribution . The mutants show increased hydrolysis of triglycerides in the gut , probably to provide sufficient energy for optimal performance of the heart , a high energy demanding organ . The upregulation of these pathways in the mutant result from increased activation of AKT and decreased FOXO . The modulation of glycolysis and lipolysis by AKT/FOXO is mediated by novel targets identified in this study . Failure of these adaptive mechanisms in aging flies lead to triglyceride accumulation , cardiac abnormalities , and ultimately a decrease in the lifespan of the mutant animals . The mechanisms identified here could have clinical implications in diseases associated with ceramide such as diabetes and obesity .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"biology"
] |
2013
|
Survival Response to Increased Ceramide Involves Metabolic Adaptation through Novel Regulators of Glycolysis and Lipolysis
|
The genomes of all organisms throughout the tree of life are compacted and organized in chromatin by association of chromatin proteins . Eukaryotic genomes encode histones , which are assembled on the genome into octamers , yielding nucleosomes . Post-translational modifications of the histones , which occur mostly on their N-terminal tails , define the functional state of chromatin . Like eukaryotes , most archaeal genomes encode histones , which are believed to be involved in the compaction and organization of their genomes . Instead of discrete multimers , in vivo data suggest assembly of “nucleosomes” of variable size , consisting of multiples of dimers , which are able to induce repression of transcription . Based on these data and a model derived from X-ray crystallography , it was recently proposed that archaeal histones assemble on DNA into “endless” hypernucleosomes . In this review , we discuss the amino acid determinants of hypernucleosome formation and highlight differences with the canonical eukaryotic octamer . We identify archaeal histones differing from the consensus , which are expected to be unable to assemble into hypernucleosomes . Finally , we identify atypical archaeal histones with short N- or C-terminal extensions and C-terminal tails similar to the tails of eukaryotic histones , which are subject to post-translational modification . Based on the expected characteristics of these archaeal histones , we discuss possibilities of involvement of histones in archaeal transcription regulation .
In eukaryotes , octameric histone cores compact DNA by wrapping an approximately 150-bp unit twice around its surface , forming a nucleosome [12 , 13] . Nucleosomes interact with each other , yielding an additional level of DNA organization in the form of a fibre . Besides a role in compaction , histones also play roles in genome organization , replication , repair , and expression , which highlights the nucleosome as a very important complex affecting a vast array of cellular processes . Characteristic of core histone proteins of all different origins is a common “histone fold”: two short and one long α-helix , separated by loops [14–18] . In eukaryotes , the histone core consists of two H2A-H2B dimers and a H3-H4 tetramer , around which approximately 146 bp of DNA is wrapped twice ( Fig 1A ) . It has been suggested that smaller histone assemblies , such as tetrasomes ( H3-H4 tetramers ) , hexasomes ( H3-H4 tetramers plus one H2A/H2B dimer ) , and hemisomes ( a H3-H4 dimer plus one H2A/H2B dimer ) , have functional roles as intermediate structures during , for example , transcription elongation [19–22] . The linker histone H1 ( which lacks the characteristic histone fold ) binds at the entry and exit points of the DNA wrapped around the octameric histone core [23 , 24] . The association of histone H1 constrains an additional 20 bp of DNA and allows for the formation of the 30-nm fibre , which results in tighter compaction [25 , 26] . Also , flexible N-terminal tails that protrude from eukaryotic histones contribute to tighter DNA packaging . These tails may interact with either the DNA or the histone surface on another nucleosome , which stabilizes the close association of nucleosomes [27–29] . Furthermore , post-translational modifications of amino acid residues in the N-terminal tails , such as acetylation , methylation , phosphorylation , ubiquitination , and biotinylation , are a key instrument for the cell to regulate gene expression , the DNA damage response , and many other processes [30–32] . For instance , while heterochromatin ( tightly packed DNA ) is typically devoid of acetylated lysines , euchromatic ( lightly packed ) regions typically contain histones with acetylated lysines . In general , euchromatin contains actively transcribed genes . Histone acetylation is believed to cause a locally less condensed chromatin structure in vivo , which is permissive to transcription . In particular the lysine-rich histone H4 tail seems to be crucial in the modulation of chromatin structure [27] . In vitro , H4 tails are required for higher order chromatin folding [33–35] , which can be disrupted by acetylation of K16 [27] . Nucleosome function and level of genome compaction can be altered in a multitude of ways , providing flexible and versatile mechanisms for tuning the cell’s dynamic chromatin structure and transcription regulation . Archaeal genomes also encode proteins that are involved in shaping DNA architecture . Genes coding for histones are found in many species throughout the domain ( Table 1 ) . In some species a homologue of the bacterial DNA bender HU was identified [36 , 37] . Nucleoid-associated proteins ( NAPs ) from the Alba family ( also known as the Sulfolobus solfataricus 10b ( Sso10b ) protein family ) are abundant and widely conserved in Archaea . Notably , Alba family proteins have also been identified in eukaryotes [38] . Characteristic of these proteins is the formation of protein–DNA filaments and bridges between DNA duplexes [39–42] . Two Alba family proteins with different functionalities have been studied in Archaea . Alba1 cooperatively forms filaments in a sequence-independent and concentration-dependent manner in Crenarchaeota , whereas Alba2 only occurs as heterodimer with Alba1 and does not form filaments [38 , 42] . Alba proteins have been shown to repress transcription in vitro [43] . In Euryarchaeota , some species express sequence-specific Alba proteins [44] , which , like Alba1 homodimers at low-protein concentrations and Alba1-Alba2 heterodimers , may form loops by bridging two DNA duplexes [45] . Other proteins affecting DNA conformation are Sso10a family proteins , which are able to bend and bridge DNA as well as form filaments on DNA [46 , 47] and the monomeric DNA benders Cren7 and Sul7 [48 , 49] . Cren7 and Sul7 have exclusively been identified in members of the Crenarchaeota phylum , whereas Sso10a has been found in some Crenarchaeota and Euryarchaeota . Other less widespread NAPs include transcription regulator of the maltose system-like 2 ( TrmBL2 ) , methanogen chromosomal protein 1 ( MC1 ) , Methanopyrus kandleri 7 kDa protein ( 7kMk ) , Sulfolobus solfataricus protein 7c ( Sso7c ) , and crenarchaeal chromatin protein 1 ( CC1 ) [50–55] . The histones found in Archaea are widespread throughout the domain but are absent in most Crenarchaeota . They have the same histone fold as eukaryotic histones , but N-terminal histone tails have not been identified ( Fig 1B ) . Linker histones , homologous to eukaryotic H1 , have not been found . Archaeal histones exist as dimers in solution , which have been shown to bend DNA [56 , 57] . These histone dimers can be homodimeric or heterodimeric [58] , as many archaeal species express , or at least encode , more than one histone variant . In Methanothermus fervidus ( class Methanobacteria ) , the two histone variants are expressed at different levels and ratios at different growth phases , suggesting a distinct function for both proteins [59] . In addition to binding as dimers , archaeal histones have been reported in vivo and in vitro to bind DNA as tetramers [60–62] , wrapping the DNA once . However , micrococcal nuclease ( MNase ) digestion patterns of Thermococcus kodakarensis ( class Thermococci ) chromatin suggest that histone–DNA complexes consist of discrete multiples of a dimeric histone subunit ( i . e . , not limited to dimers and tetramers ) in vivo without obvious dependence on the DNA sequence [63] . Based on the latter observations , it was proposed that histone dimers multimerize and wrap DNA into a filament of variable length [17 , 63] . The crystallography study of Luger and coworkers on histone HMfB from M . fervidus indicates that these histones assemble into an endless left-handed rod in vitro , which we propose to call a “hypernucleosome” ( Fig 2 ) . Note that these complexes were assembled on SELEX-optimized DNA previously shown to favor tetrameric nucleosome assembly [64] . The number of wraps in the hypernucleosome , which is the DNA bending 360° around the histone multimer , scales linearly with the number of histone subunits , resulting in a tight packaging of DNA . The authors also provide evidence that mutation-directed perturbation of hypernucleosome function in vivo alters response to nutrient change in T . kodakarensis , suggesting a role in transcription . Both eukaryotes and Archaea encode histone proteins , which seem to be involved in response to environmental cues by their involvement in transcription regulation . It has been suggested that eukaryotic histones evolved from archaeal histones [65] . This hypothesis is supported by the high similarity at the amino acid sequence level and in secondary structure [66 , 67] . Suggestive of an archaeal origin of eukaryotic histones is also the dimeric nature of archaeal histones; archaeal histone complexes are built from dimers , but members of the archaeal class Halobacteria express a “tandem histone . ” In these tandem histones , the histone folds are linked end-to-end [68–70] . This implies that the histone folds always occupy the same position and role in the naturally linked dimer . This leads to the relaxation of evolutionary constraints in parts of the histone , an example of subfunctionalization [71 , 72] . According to this hypothesis , the histone folds further evolved in a divergent way , leading to an asymmetric dimer . This may have been an ancestor of H3-H4 , which later separated to become two individual proteins and corresponding genes [66] . The eukaryotic H3-H4 tetramer resembles the tetramer found in Archaea , and it has been suggested that H2A and H2B have arisen from H3 and H4 later on in histone evolution [66] . Indeed , H3 and H4 are more similar to archaeal histones than H2A and H2B , supporting this hypothesis . From this point , eukaryotic histones have further evolved into histone variants , highly homologous substitutes of canonical eukaryotic histones , which often play a specialist role in a wide variety of cellular processes [73] . Unlike canonical histones , which are mainly expressed during DNA replication , histone variants are expressed in a replication-independent manner [74 , 75] . Histone variants of H2A and H3 are widely known and studied , whereas only a few examples have been found of diversified H2B and H4 [76] . The evolutionary pressure for the evolution of dimer-based histones to octameric histones and their subsequent variants was long believed to be DNA compaction [66] . The fact that eukaryotic cells undergo mitosis , in which chromosomes are highly compacted , together with the abundance of gene-poor regions may have favored a histone conformation that wraps DNA twice ( eukaryotic octamer ) instead of once ( archaeal tetramer ) and that via its N-terminal tails has the ability to compact DNA at a higher order . Open questions that remain are how histone evolution was driven and what the roles of archaeal histones and their variants are in genome packaging and regulation . Here , we discuss the amino acid residues that are responsible for the formation of the hypernucleosome based on a sequence analysis of a subset of archaeal histones that includes histones from all phyla that contain genes coding for histones ( Fig 3 ) . Also , we analyze the ability of histones to form a hypernucleosome and the effects of N- or C-termini longer or shorter than the consensus on histone multimerization and transcription regulation . We emphasize the histones in species from recently discovered phyla , which are believed to be an evolutionary link to eukaryotes [11 , 77] . Based on elements that archaeal histones have in common and elements that differ from that consensus , we discuss some of the open questions regarding gene regulation by archaeal histones .
A striking finding based on the amino acid sequence comparison reveals that two histones from Candidatus Heimdallarchaeota archaeon LC_3 ( only Histone A [HA] shown in Fig 3 ) , one from Candidatus Huberarchaea archaeon CG_4_9_14_3_um_filter_31_125 and one from Candidatus Bathyarchaeota archaeon B23 , contain an N-terminal tail , which was previously thought to exist only in eukaryotic histones and only recently reported for Heimdallarchaeota [64] . In eukaryotes , these tails stabilize a higher order of compaction by interacting with either the DNA or another nucleosome . The tails of the two histones from Heimdallarchaeota and Huberarchaea are of roughly the same length and sequence composition as eukaryotic H4 tails ( see Fig 3 ) . Prompted by the importance of the eukaryotic histone tails in modulating chromatin structure and function [27 , 32] , we constructed a molecular model of a hypernucleosome formed by Histone A ( HA ) from Heimdallarchaeota LC_3 to investigate its potential function ( see Methods section ) . The model illustrates how three subsequent arginines ( R17–R19 ) could facilitate passing of the tails through the DNA gyres ( Fig 4 ) . The tails exit the hypernucleosome through DNA minor grooves , similar to eukaryotic histone tails , and might position their lysine side chains to bind to the hypernucleosomal DNA or to other DNA close by , facilitating ( long-range ) genomic interactions in trans . Like the H4 tail that is subject to acetylation of lysines K5 , K8 , K12 , and K16 [91] , lysines in the Heimdallarchaeal histone tail may well be subject to acetylation . Archaeal genomes are known to have several candidate lysine acetyltransferase and deacetylase enzymes , including proteins belonging to the ELP3 superfamily , to which transcription elongation factor and histone acetyltransferase ELP3 belongs [92–94] . Searches using the ProSite database ( http://prosite . expasy . org , [95] ) and Protein Information Resource ( http://pir . georgetown . edu , [96] ) further reveal that the Heimdallarchaeota LC_3 genome contains multiple gene products containing the Gcn5-related N-acetyltransferase domain , which is present in many histone acetyltransferases [97] . Interestingly , a potential “reader” protein that binds modified lysines can also be identified . This protein , HeimC3_47440 , contains a YEATS-domain , which has recently been shown to bind histone tails that carry acetylated or crotonylated lysines [98–101] . Comparison with the closest homolog of known 3D structure , YEATS2 ( 35% identity , PDB-id 5IQL , [102] ) , shows that the binding site for the modified lysine side chain is strictly conserved in the archaeal protein . Notably , only Candidatus Bathyarchaeota , which also features tailed histones , contains a detectable homolog of HeimC3_47440 . The presence of lysine-containing N-terminal tails in combination with histone modification writers and readers suggests that Archaea use post-translational modifications in a similar way to Eukaryotes as modulators of genome compaction and gene activity . The tail of the Huberarchaea histone also contains lysine residues that are found at the same position as some of the lysines of the H4 tail . However , no proteins involved in post-translational modification of histone tails have been identified in this phylum . Other histones , for example from Candidatus Lokiarchaeota CR_4 , Candidatus Odinarchaeota LBC_4 , Nanoarchaeum equitans , and Thermofilum pendens , contain a short N-terminal tail of 5–10 residues . Also , histones with a C-terminal tail have been found . The histone from the euryarchaeal species Methanocaldococcus jannaschii ( class Methanococci ) has a 28-residue tail , which seems to be unique among archaeal histones . Other C-terminal tails are up to 11 residues long ( as compared to Methanothermus fervidus HMfB ) and appear in Caldiarchaeum subterraneum , Candidatus Bathyarchaeota SMTZ-80 , Candidatus Heimdallarchaeota LC_3 , Candidatus Lokiarchaeota CR_4 , and all histones found in Crenarchaeota . These short C-terminal tails are similar in length to the H4 C-terminal tail , that is reported to play a role in the promotion of histone octamer formation in eukaryotes [103] . The genomes of some archaeal species contain genes for histone truncates . The histone from Haloredivivus sp . G17 , member of the candidate phylum Nanohaloarchaeota , and the histone from Candidatus Bathyarchaeota archaeon B24 both lack part of the N-terminal α-helix ( α1 ) , and one histone from Candidatus Lokiarchaeota GC14-75 is reduced in length at the C-terminus . The remainder of the C-terminal amino acids likely does not form a C-terminal helix ( α3 ) in this histone from Candidatus Lokiarchaeota . Although histones of reduced length or containing tails lack part of the histone fold , they likely still possess DNA-binding properties . Therefore , they possibly have functional roles in the regulation of genes . Both eukaryotic histones and HMfB form dimers , a process that is driven by a hydrophobic core ( involving residues A24 , L28 , L32 , I39 , and A43 in HMfB ) as well as a crucial salt bridge for a stable histone fold ( R52-D59 in HMfB ) [14] . These hydrophobic residues and the salt bridge are conserved among Archaea . This indicates that archaeal histones have very similar tertiary structures [14 , 104] . Also , residues that play an important role in DNA binding are present in all examined histones , including the arginines that anchor archaeal histone dimers to the DNA minor grooves ( R10 and R19 in HMfB ) [14] . Both eukaryotic H3-H4-dimers and HMfB dimers can form tetramers by hydrogen bonding of H49 and D59 ( HMfB ) and additional hydrophobic interactions in the interface ( L46 and L62 in HMfB ) [105] , pairs of residues that , too , are generally conserved among archaeal histones ( Fig 3 ) . The HMfB–DNA cocrystal structure reveals left-handed wrapping of DNA around a histone-multimer core [64] ( Fig 2 ) . This structure supports the model in which HMfB dimers multimerize along DNA into an “infinite” hypernucleosome , thereby linearly compacting the DNA approximately ten-fold . It is likely that hypernucleosomes grow or shrink by association or dissociation of dimers at both ends . The resolution of the crystal structure allowed us to identify several interacting residues between layers of dimers that may be important for stabilizing the complex ( Fig 5 ) . Based on this structural information , the propensity of different archaeal histones to multimerize can be predicted . In Table 2 , we set out three criteria for hypernucleosome formation by archaeal histones . Firstly , conservation of residues in the dimer–dimer interface ( L46 , H49 , D59 , and L62 in HMfB ) is required , as forming a tetramer is the first step in multimerization . Secondly , residue G16 , which is positioned at the stacking interface of the hypernucleosome ( Fig 5 ) , is crucial in permitting formation of the hypernucleosome [64] . Bulkier residues at this position interfere with multimerization [64] . Lastly , favorable interactions between histone dimers i and i+2 and i+3 , here termed stacking interactions , will contribute to stability of the compacted hypernucleosome . The HMfB hypernucleosome crystal structure shows three stacking interactions , hydrogen bonds from K30 to E61 , E34 to R65 , and R48 to D14 ( Figs 3 and 5 ) . Scrutiny of histone sequences reveals that most archaeal histones meet these criteria and are thus likely to form hypernucleosomes ( Table 2 , marked + ) . We identified two to seven potential stacking interactions for this group of histones , which may affect hypernucleosome stability and compactness . Fewer interactions may allow for more “breathing” of the hypernucleosome structure , yielding hypernucleosomes that are more flexible or “floppy . ” We predict such structures to be formed also by a number of archaeal histones that do not fully meet our criteria ( Table 2 , marked ± ) . For example , Candidatus Heimdallarchaeota LC_3 HA and Candidatus Lokiarchaeota GC14_75 HLkE have H49N and D59S substitutions , respectively , which likely weakens the crucial hydrogen-bonding interaction at the dimer–dimer interface [105] . Similarly , substitution of the hydrophobic residues 46 and 62 for more hydrophilic or bulkier ones would lead to a less stable dimer–dimer interface , as for Candidatus Heimdallarchaeota LC_3 HC and Candidatus Bathyarchaeota B23 . In the presence of the canonical dimer–dimer interface , bulky substitutions at position 16 likely also result in a more open hypernucleosome structure , as for Candidatus Odinarchaeota LCB_4 . Three archaeal histone species fail multiple criteria in our analysis , indicating that these cannot form hypernucleosomes . These histone species are Haloredivivus sp G17 , Nanosalina J07AB43 HB , and Euryarchaeal Methanococcoides methylutens ( class Methanomicrobia ) that all combine defects in the dimer interface with a bulky substitution at position 16 and few potential stacking interactions ( Table 2 , marked– ) . In particular , Nanosalina J07AB43 Histone B ( HB ) shows a H49D substitution and a glutamic acid at position 62 , making the dimer surface highly negatively charged and thus very unlikely to interact with another dimer . It is remarkable that most of the histones having N- or C-terminal tails or N- or C-terminal truncations additionally have substitutions in the dimer–dimer and/or stacking interface that will affect hypernucleosome formation . Histones with reduced ability to form compact hypernucleosomes are expected to exhibit different roles in shaping the genome , like simple DNA bending or site-specific interference with histone multimerization . Interestingly , the genomes of several organisms encode histones that we predict are able to multimerize as well as histones that probably do not multimerize . This suggests that they may , in addition to directly binding to promoters , also be able to affect gene regulation by multimerization . MNase-seq experiments have shown that histones position upstream and downstream of a promoter region [106] . This , in combination with knock-out studies showing both up- and down-regulation of transcription levels , leads to the hypothesis that histones are important for transcription regulation in the relatively well-studied phylum Euryarchaeota [45 , 69 , 107 , 108] and may play a similar role in other histone-coding phyla . The exact mechanisms by which histones act in regulation are at this moment largely unknown . What is the mechanistic role of histones in the regulation of gene expression ? Is the hypernucleosome , with a mechanism analogous to that in bacterial gene repression , able to block promoter regions and other regulatory elements , thereby making them inaccessible to the transcription machinery [109–112] ? In Bacteria , such a mechanism exists for H-NS and partition protein B ( ParB ) proteins , in which filaments laterally spread from a nucleation site , often a high-affinity DNA sequence [113–116] . Specific high-affinity sites have been identified both in vivo and in vitro in Archaea [61 , 106 , 117 , 118] . The role of such high-affinity sites may be to position the hypernucleosome on the genome and could be a key feature in archaeal genome regulation . In Archaea , cooperative lateral spreading of filaments has been reported for Alba proteins [40 , 42 , 119 , 120] . Also , promoter occlusion mechanisms and competitive binding of archaeal NAPs and transcription factors have been reported [45 , 121 , 122] . In addition , how dynamic are hypernucleosomes , and how does the cell control the size of the hypernucleosome in order for it to be functional ? Is up- and down-regulation of histone expression important in fine tuning this process ? Another option for control of hypernucleosome size is heteromerization of histone variants with different stacking propensity . Heteromerization of such histone variants , for instance HA and HB from Nanosalina J07AB43 ( Table 2 ) , could restrict hypernucleosome size to fewer subunits . Distinct expression patterns of histone variants at different growth phases or as a result of environmental cues such as osmolarity [59 , 107] , may alter the composition and size of the hypernucleosome . However , so far , histone variants have been poorly studied in Archaea . The results of our predictions on hypernucleosome formation clearly point out the need for in vitro and in vivo studies explicitly addressing all of these questions .
Histones from Archaea and eukaryotes are similar in tertiary but not in quaternary structure when bound to DNA . While eukaryotic histones form octamers on the DNA , archaeal histones form filaments of variable size: hypernucleosomes . Important residues responsible for DNA binding , dimer–dimer interactions , and stacking interactions are mostly conserved among Archaea , including Asgard Archaea , Bathyarchaeota , and other newly discovered Archaea . In these recently discovered Archaeal phyla , histone tails and truncated histone variants were also found . In terms of evolution , it appears that , based on fragmentary data derived from extant lineages , the hypernucleosome has progressively become more flexible as histones with N-terminal and C-terminal tails and additional terminal helices ( like in H2A and H2B in the nucleosome ) developed . Furthermore , the appearance of additional DNA-binding residues and positively charged N-terminal tails may have increased the affinity of histones for DNA [123] . These changes in dimer structure and DNA affinity may have stabilized octameric nucleosomes and disfavored multimerization . Specifically , the emergence of the eukaryotic H2A-H2B heterodimer blocked hypernucleosome formation since H2A lacks the dimer–dimer interface , and H2B contains an additional helix at its C-terminus that blocks the stacking interface . The histone tails from Candidatus Heimdallarchaeota are likely to function in similar ways as those of eukaryotic histones . They are lysine rich and potentially subject to post-translational modification , thereby possibly affecting the histone’s interactions with other actors . Alternatively , they may provide stabilization of the hypernucleosome via interactions with DNA in cis or in trans . Since it is believed that eukaryotes share their latest common ancestor with Candidatus Heimdallarchaeota , eukaryotic histones may have evolved from the predecessors of the tail-containing Heimdallarchaeal histones . As some histone proteins that have an N-terminal tail ( Candidatus Heimdallarchaeota LC_3 HA and Bathyarchaeota archaeon B23 ) seem to form less stable hypernucleosomes , these histones may represent an evolutionary transition towards a different mechanism of gene regulation , switching from regulation by multimerization and compaction toward regulation by histone tail modifications . Although the hypernucleosome structure is suggestive of stacking interactions between dimers in adjacent turns , experimental evidence for such interactions is lacking . Also , the functional role of tails , as well as truncates , has yet to be proven experimentally . In vitro hypernucleosome reconstitution experiments and in vivo foot-printing assays of species expressing nonstandard histones combined with mutation of the residues proposed to be involved in stacking interactions could answer these questions . Lastly , the existence of post-translational modifications of residues in archaeal histone tails , as well as their effect on transcription regulation , remains to be discovered and would give an important insight into the evolution of transcription regulation and genome folding from Archaea to eukaryotes .
We have included histones from every histone-encoding ( candidate ) phylum within the archaeal domain in our analysis . We show different histones from the same organism if the predicted stacking properties are very dissimilar . Sequences were aligned with Clustal Omega [124] using default parameters , removing gaps . Structural analysis of the selected archaeal histones and assessment of potential hypernucleosome formation was done by inspecting the conservation of residues that are important for multimerization in the published HMfB hypernucleosome structure [64] . Comparative multichain modeling was performed in MODELLER [125] using default parameters to construct dimer models of the archaeal histones . These models were superimposed onto HMfB dimers in the hypernucleosome crystal structure to assess whether alternative or additional interactions were possible in the different archaeal histone complexes . The molecular model of the histone HA dimer from the Heimdallarchaeota LC_3 genome was constructed by multitemplate modeling in MODELLER [125] using otherwise default parameters . The HMfB dimer in the hypernucleosome [64] was used as a structural template for the histone fold and eukaryotic histone H3 and H4 as structural templates for the N-terminal tails . An initial model for the Heimdall HA hypernucleosome was obtained by superimposing the HA dimer model onto HMfB in the hypernucleosome crystal structure , with either an H3-like or an H4-like tail conformation . To optimize the path of the tails through the DNA gyres and remove major steric clashes , the HA dimer model and surrounding DNA was excised from the initial model and water refined separately using High-Ambiguity Driven Docking ( HADDOCK ) [126] , imposing ambiguous interaction restraints between HA residues 14–19 and the surrounding 3-bp section of DNA , using otherwise default parameters .
|
Both Archaea and eukaryotes express histones , but whereas the tertiary structure of histones is conserved , the quaternary structure of histone–DNA complexes is very different . In a recent study , the crystal structure of the archaeal hypernucleosome was revealed to be an “endless” core of interacting histones that wraps the DNA around it in a left-handed manner . The ability to form a hypernucleosome is likely determined by dimer–dimer interactions as well as stacking interactions between individual layers of the hypernucleosome . We analyzed a wide variety of archaeal histones and found that most but not all histones possess residues able to facilitate hypernucleosome formation . Among these are histones with truncated termini or extended histone tails . Based on our analysis , we propose several possibilities of archaeal histone involvement in transcription regulation .
|
[
"Abstract",
"Introduction",
"Histones",
"are",
"found",
"in",
"some",
"newly",
"discovered",
"Archaea",
"Conclusion",
"Methods"
] |
[
"chemical",
"compounds",
"dna-binding",
"proteins",
"microbiology",
"organic",
"compounds",
"review",
"archaea",
"basic",
"amino",
"acids",
"amino",
"acids",
"epigenetics",
"chromatin",
"chromosome",
"biology",
"proteins",
"gene",
"expression",
"chemistry",
"histones",
"nucleosomes",
"archaean",
"biology",
"biochemistry",
"eukaryota",
"cell",
"biology",
"organic",
"chemistry",
"genetics",
"biology",
"and",
"life",
"sciences",
"lysine",
"physical",
"sciences",
"organisms"
] |
2018
|
Structure and function of archaeal histones
|
Infection by Listeria monocytogenes ( Lm ) causes serious sepsis and meningitis leading to mortality in neonates . This work explored the ability of CD11chigh lineage DCs to induce CD8+ T-cell immune protection against Lm in mice before 7 days of life , a period symbolized by the absence of murine IL-12p70-producing CD11chighCD8α+ dendritic cells ( DCs ) . We characterized a dominant functional Batf3-dependent precursor of CD11chigh DCs that is Clec9A+CD205+CD24+ but CD8α- at 3 days of life . After Lm-OVA infection , these pre-DCs that cross-present Ag display the unique ability to produce high levels of IL-12p40 ( not IL-12p70 nor IL-23 ) , which enhances OVA-specific CD8+ T cell response , and regulatory IL-10 that limits OVA-specific CD8+ T cell response . Targeting these neonatal pre-DCs for the first time with a single treatment of anti-Clec9A-OVA antibody in combination with a DC activating agent such as poly ( I:C ) increased the protection against later exposure to the Lm-OVA strain . Poly ( I:C ) was shown to induce IL-12p40 production , but not IL-10 by neonatal pre-DCs . In conclusion , we identified a new biologically active precursor of Clec9A+ CD8α- DCs , endowed with regulatory properties in early life that represents a valuable target to augment memory responses to vaccines .
Early life is a period of immune maturation characterized by a high susceptibility to infectious diseases . The underdeveloped immune system gives a Th2-biased response and has an impaired ability to develop long-lasting protective CD8+ T cell immunity [1 , 2] . We are particularly interested in immune resistance to infections by Listeria monocytogenes ( Lm ) . Lm is a gram-positive opportunistic food-borne bacteria with a facultative intracellular life cycle that commonly causes sepsis and/or meningitis , leading to mortality in neonates but is asymptomatic in immunocompetent Lm-infected adults [3] . DCs are the key components of the immune system , determining susceptibility to infections . The primary function of DCs is the detection of pathogens and the initiation of the adaptive immune response . Such a response requires the DCs to present an antigen ( Ag ) from a specific pathogen , as well as an innate signal from microbes or damaged cells allowing DCs to orchestrate the adaptive immune response . Conventional DCs ( cDCs ) in mice can be divided into two distinct populations , one with high expression of CD8α ( CD8α+ cDCs ) and the other with no expression of CD8α ( CD8α- cDCs ) . These CD8α+ cDCs selectively express the C-type lectin receptor DNGR1 , also called Clec9A [4] . The development of CD8α + cDCs depends on a common set of transcription factors including Irf8 [5] , Batf3 [6] , Id2 [7] and Nfil3 [8] . CD8α+ cDCs are particularly efficient at internalizing and cross-presenting exogenous Ag on MHC class I molecules , especially from dead or dying cells [9–14] . cDCs represent a key subset which initiates cell-mediated immunity against tumors , viruses and bacteria [15 , 16] . Upon Lm infection , adult CD8α + DCs phagocytize the bacteria in the marginal zone of the spleen , and migrate to the T-cell zone in order to present the bacterial antigens to CD8+ T cells [17] . The resultant response involves the up-regulation of co-stimulatory molecules , the production of cytokines like IFN-γ and the generation of cytotoxic T-cell immunity . Finally , CD8α+ cDCs have been identified as professional IL-12p70 producers priming the adaptive immune cells towards Th1 differentiation [18–21] . In murine neonates , CD8α+ cDCs have been shown to be defective in the first 6 days of life . Beyond this time , the CD8α+ cDCs producing IL-12p70 induces the downregulation of the IL-4Rα/IL-13Rα1 on T cells , favoring a Th1 response [2] . Since the study by Lee H . et al . [2] , the immune neonatal period has been redefined . As a result , some of the previous reports on the quantitative and qualitative shortcoming of neonatal DCs have to be revisited . For example , it was demonstrated that at 7 days of life the Flt3 ligand-treated “neonatal” mice showed an increase in DCs lineage development and an increased in IL-12-dependent innate resistance against Lm [22] . Another study reported that one-day-old DCs were able to produce adult level of IL-12p70 , but only after IL-4 , a maturating cytokine , was added to GM-CSF and CpG in the culture [23] . Neonatal induction of Th1/Tc1 memory is still controversial . Neonates have shown to be more susceptible to intracellular pathogens due to a suboptimal capacity to mount an efficient cell-mediated immunity , particularly the memory CD8+ T cells . For instance , qualitative defect in neonatal Batf3-dependent CD103+ lung DCs were recently reported to influence the CD8+ T cell response , following respiratory syncytial virus ( RSV ) infection [24] . However , other studies have demonstrated that neonates could mount an adult-like CD8+ T cell immune response against human CMV or Trypanosoma cruzi [25 , 26] . Concerning Lm infection in early life , a previous study demonstrated that 5- to 7-days old neonates are able to develop robust primary and secondary CD4+ and CD8+ Th1-type responses against Lm without characterizing the antigen presenting cells that were involved [27] . The objectives of this study were to describe the phenotype of Batf3-dependent CD11chigh DC subset and to explore their abilities to induce a CD8+ T cell immune protection against Lm at 3 days of life . First , we characterized the splenic DC subset bearing DNGR1/Clec9A but not CD8α , a precursor of CD8α + DCs . This DNGR1/Clec9A bearing DC is the predominant lineage before 6 days of life . Next , we demonstrated the role of these early DCs in taking up and presenting exogenous Lm Ag to prime a CD8+ T-cell response . Additionally , we defined the role of IL-12p40 and IL-10 uniquely produced by these neonatal pre-DCs in the establishment of an adaptive response . Finally , we assessed vaccination strategies , directly targeting neonatal DCs using OVA coupled to anti-Clec9A in the presence of poly ( I:C ) . This study clarifies the function of pre-CD8α + DCs in early life and highlights the advantages for human neonatal vaccination strategies .
To determine the type of DC involved in the adaptive immune response against Lm at 3 days of life , we employed Batf3-/- mice known to lack the conventional CD8α+ type of DC [6] . We compared the OVA-specific primary immune response to the attenuated strain Lm actA-/--OVA in 3-day-old and adult C57BL/6 and Batf3-/- mice . As seen in Fig 1A , the production of IFN-γ , following restimulation with MHC I-restricted OVA peptides , was drastically reduced in splenic cultures from Batf3-/- mice compared with C57BL/6 mice at both ages . This suggested that Batf3-dependent DCs are required to trigger an IFN-γ T cell response against Lm infection at 3 days of life as well as at the adult stage . To characterize the DCs likely to generate an adaptive CD8+ T-cell response against Lm in neonates , we focused on the Batf3-dependent CD8α+ DC lineage . In naïve adult C57BL/6 spleen , the majority ( 69% ) of CD11b-CD11chigh cells were CD8α+ ( Fig 1B ) . In contrast , in neonates , the majority ( 85% ) of splenic CD11b-CD11chigh cells lacked CD8α expression but were positive for CD205 and DNGR1/Clec9A , which are features of CD8α+ DC family [4] , as confirmed in adult splenic CD8α+ DCs ( Fig 1B ) . We further observed that CD11chigh CD11b-CD205+DNGR1+ DCs in mesenteric lymph nodes collected from 5 day-old C57BL/6 mice ( S1 Fig ) expressed the same intermediate level of CD8α as their splenic counterpart . The neonatal splenic CD11chigh CD11blow CD205+CD8α- DCs also expressed CD24 and cKIT at similar levels but expressed MHCII , CD80 and CD86 molecules at a lower level , compared to adult CD8α + DCs ( S2A Fig ) . Interestingly , neonatal CD11chighCD11blowCD205+CD8α + or CD8α - displayed similar levels of aforementioned molecules suggesting that they were closely related phenotypically ( S2A Fig ) . No expression of CD207 , CD4 or B220 was detected in neonatal CD8α-CD11b-CD205+ DCs nor in adult CD8α+CD11b-CD205+DCs ( S2A Fig ) and the high expression of DNGR1/Clec9A was restricted to neonatal CD8α -CD11b-CD205+ DCs and not to neonatal pDCs or cDC2 ( S2B Fig ) . These results demonstrated that neonatal CD8α -CD11b- DCs display a surface phenotype closely similar to adult CD8α + DCs except for the expression of CD8α , what suggests that they are earlier forms of the same DC lineage . This was supported by analysis of Batf3-/- neonates in which no CD11c+CD11b-CD205+DNGR1+ CD8α - DCs could be detected ( Fig 1C and S3A Fig ) . Next we examined the developmental relationship between neonatal CD11b-CD8α - DCs and adult CD8α + DCs . A time course of the frequency of CD11c+CD11b-CD205+CD8α - DCs versus CD11c+CD11b-CD205+CD8α + DCs showed that the latter were dominant for the first 5 days of life . After day 7 , the proportions were totally reversed as CD11c+CD11b-CD205+ DCs expressing CD8α became dominant ( Fig 1D ) . We thus confirmed the appearance and accumulation of the CD8α + DCs at day 6 as previously reported [2] . Finally , to test if neonatal spleen contained precursors of the CD8α + DC type found in adults , we transferred , into adult C57BL/6 CD45 . 1+ mice , neonatal C57BL/6 CD45 . 2+ spleen cells that were depleted of CD3+ , CD19+ and Gr1+ cells and that contained mostly CD11c+CD11b-CD24+ CD205+ CD8α - cells . The number of DCs of donor origin expressing CD8α among CD45 . 2+CD11c+ CD11b- CD24+ CD205+ spleen cells showed a steady increase to adult levels by day-6 post transfer ( Fig 1E ) . Taken together , these results suggested that earlier forms of the CD8α + DCs , that we will call preCD8α Clec9A+ DCs , are predominant among splenic and mesenteric lymph nodes CD11chigh DCs from birth to 5 days of life . We further determined that C57BL/6 neonates that were submitted to a 3 day treatment with the dendritic cells growth factor , Flt3L [28] ( instead of 7 days as previously reported [22] ) , expanded preferentially the absolute number of preCD8α Clec9A+ DCs ( Fig 1F ) without affecting the number of neonatal plasmacytoid DCs or conventional DC2 ( S3C Fig ) and significantly enhanced their defense against Lm ( Fig 1G ) as previously described [22] . As presented in Fig 2A , we first demonstrated that neonatal preCD8α Clec9A+ DCs were able to phagocytize Lm-GFP ( 62%±2 . 6 ) as did adult CD8α + DCs ( 84 . 3%±3 ) . To assess capacity to cross-present cell-associated Ag during Lm infection , we purified splenic C57BL/6 neonatal preCD8α Clec9A+ DCs or adult CD8α + DCs , incubated them with OVA257-264 peptides or Lm-OVA and co-cultured them with OVA-specific OT-I T cells , with or without GM-CSF known to enhance cross-presentation capacity of newly formed DC [29]; and monitored the IFN-γ production ( Fig 2B ) . Neonatal preCD8α Clec9A+ DCs were able to cross-present OVA as efficiently as adult CD8α + DCs in the presence of GM-CSF . We evaluated at the RNA level the cDNA relative expression of a large number of genes involved in the cross-presentation machinery ( such as Rac2 , Ergic 1 , 2 and 3 , Rab14 , Erap1 , Sec22b , Syntaxin 4 , TAP1 and 2 ) and concluded that they were similar in sorted preCD8α DCs and adult CD8α DCs except for a slight difference with H2-K1 , β2m and Rab27a ( S4 Fig ) . This cross-presentation ability was confirmed ex vivo with neonates that were injected with Lm actA-/--OVA , from which preCD8α Clec9A+ DCs were sorted and co-cultured with OT-I T cells , with or without GM-CSF ( Fig 2C ) . We then assessed the capacity of neonatal preCD8α Clec9A+ to produce cytokines in several settings such as Lm exposure or TLR ligand encounter and compared it to fully competent adult CD8α + DCs . As an endosomial TLR receptor , TLR3 binds to ligand such as dsRNA , namely polyinosinic-polycytidylic acid ( Poly ( I:C ) ) which mimic the replication intermediate of virus . TLR3 agonist was shown to promote the cross-presentation of antigens that transit in the endosome as it increases MHC class I and costimulatory molecules of DCs and stimulate IL-12 secretion . It is considered as an efficient manner to optimally activate DCs to promote CD8+ T cell activation . Moreover , TLR3 is only expressed in CD8α + DC subset . Poly ( I:C ) was first injected into C57BL/6 adults and neonates and intracellular staining for IL-12p40 was performed on CD11c+CD11b-CD205+CD8α + and CD11c+CD11b-CD205+CD8α - at different time points ( Fig 3A and S5 Fig ) . As expected , the overall number and relative frequency of IL-12p40 producing CD8α + DCs was higher in adults than in neonates . Interestingly , 2h after poly ( I:C ) treatment , a high proportion of CD8α - DCs were producing IL-12p40 in neonates but not in adults ( Fig 3A ) . To assess cytokine responses by the DC to Lm infection , cytokine secretion of cultured cells and mRNA synthesis was utilized . For mRNA measurement , we sorted DC subsets from neonates infected with 5 x 105 Lm actA-/- or adults infected with 5 x 105 or 5 x 106 CFU Lm actA-/- . IL-12p35 mRNA synthesis was induced in adult CD8α + DCs but not in CD8α - DCs at both ages ( Fig 3B ) . A significant induction of IL-12p40 mRNA synthesis was observed in neonatal and adult CD8α- DCs and in adult CD8α + DCs upon Lm actA-/- infection ( Fig 3B ) . In addition , sorted neonatal preCD8α Clec9A+ DCs secreted quite similar protein levels of IL-12p40 as adult CD8α + DCs after in vitro stimulation with poly ( I:C ) or Lm ( Fig 3C ) . However , neither IL-12p70 ( Fig 3D ) nor IL-23 ( S6 Fig ) were significantly produced by these neonatal preCD8α Clec9A+ DCs after such stimulation , even in the presence of GM-CSF , IL-4 and IFN-γ; by contrast , CpG did induce IL-12p70 secretion in such maturating conditions ( Fig 3D ) as previously described [23] . IL-23p19 mRNA synthesis was not induced in all the tested DC subsets in response to Lm infection ( Fig 3B ) . Of particular interest , IL-10 transcripts were strongly induced only in neonatal preCD8α Clec9A+ DCs after Lm actA-/- infection whereas adult preCD8α Clec9A+ DC and adult CD8α + DCs did not ( Fig 3E ) . The exclusive production of IL-10 by neonatal preCD8α Clec9A+ DCs after Lm actA-/-stimulation was confirmed in vitro at the protein level ( Fig 3F ) . Finally , the TLR expression was compared between preCD8α DCs and CD8α + DCs to potentially explain their distinct functionality . As presented in S7 Fig , TLR9 RNA expressions were similar and TLR3 RNA was expressed only 1 . 14 times more in adult CD8α + DCs . Altogether , these results indicate that , upon Lm infection , splenic neonatal preCD8α Clec9A+ DCs have the unique ability to produce IL-12p40 and IL-10 but no IL-12p70 and IL-23 . We investigated the impact of the IL-12p40 subunit and of IL-10 in the protective immune response against Lm infection . Neutralizing anti-IL-12p40 mAb was administered to neonates exposed to Lm . The survival rate of neonates was significantly reduced compared to control isotype-treated infected mice ( Fig 4A ) . The impact of IL-12p40 on the primary CD8+ T-cell response during Lm infection was then assessed . IL-12p40 neutralization during Lm actA-/--OVA infection strongly inhibited the frequency of IFN-γ-producing CD8+ T cells ( Fig 4B ) as well as the secretion of IFN-γ ( Fig 4C ) in response to OVA257-264 peptides . Furthermore , when anti-IL-12p40 mAb was added during co-culture of sorted neonatal preCD8α Clec9A+DCs with Lm-OVA ( MOI 1 ) in the presence of GM-CSF and OT-I T cells , an inhibition of IFN-γ production was obtained ( Fig 4D ) . We further demonstrated that the in vivo proliferative response of transferred OT-I T cells in C57BL/6 neonates induced by Lm actA-/--OVA infection was significantly inhibited in IL-12p40 -/- neonates but not in IL-12p35-/- or IL-23p19 -/- neonates ( Fig 4E ) . Finally , the role played by IL-10 in CD8+ T-cell activation induced by Lm was assessed by blocking the IL-10 receptor . The IFN-γ production of OT-I T cells in response to Lm-OVA presentation in vitro by sorted neonatal preCD8α Clec9A+ DCs was enhanced in the presence of anti-IL-10R mAb ( Fig 4F ) . These results demonstrated that the IL-12p40 subunit produced by neonatal preCD8α Clec9A+ DCs is functional in inducing an efficient primary CD8+ T cell response whereas the IL-10 secreted by these precursor DCs moderates the CD8+ T cell activation . We then investigated whether targeting antigens to Clec9A on the neonatal preCD8α Clec9A+ DCs could be an effective strategy for immunization against Lm infection . We first tested whether the neonatal preCD8α Clec9A+ DCs are able to cross-present to CD8+ T cells in vivo through Clec9A targeting . C57BL/6 neonates or Batf3-/- neonates were i . v . injected with either the construct of OVA protein linked to anti-Clec9A antibody ( anti-Clec9A-OVA ) or with Lm actA-/--OVA , along with CFSE-labelled OT-I T cells that were monitored for proliferation . Constructs of OVA linked to an isotype-matched control Ab ( GL117-OVA ) was used as a control . As seen in Fig 5 , anti-Clec9A-OVA administration induced OT-I T cell proliferation in vivo , similar to that obtained with Lm actA-/--OVA ( Fig 5A ) . Furthermore , the proliferation induced by the Clec9A targeting was exclusively Batf3-dependent , since no proliferation was observed in the Baft3-/- mice compared to GL117-OVA treated mice ( Fig 5B ) whereas the Lm actA-/--OVA-induced OT-I cell proliferation was only partially dependent on Batf3 ( Fig 5B ) . We therefore conclude that neonatal preCD8α Clec9A+ DCs serve as effective targets for cross-presentation through Clec9A and for CD8+ T cell activation , such as described for MHC-I restricted CD8+ T-cell responses in adult mice [30–32] . In a next step we proceeded to determine whether targeting antigen to preCD8α Clec9A+ DCs in neonates by anti-Clec9A-OVA constructs leads to enhanced protection against later Lm infection , and whether this requires DC activating agents . Neonates were vaccinated with anti-Clec9A-OVA or GL117-OVA constructs , with or without poly ( I:C ) as adjuvant . The mice were challenged by infection with Lm-OVA 60 days later . A group of unimmunized control mice was infected at day 60 to provide a primary response comparison ( Fig 6A ) . Mice vaccinated at the neonatal stage with anti-Clec9A-OVA and poly ( I:C ) were partially protected , 55% surviving compared to 90–100% death in control groups ( Fig 6B ) . Immunization with OVA linked to a non-targeting isotype control antibody , or with the targeted construct alone without adjuvant , did not allow survival . The survival of mice that were vaccinated at 3 days of life and challenged 60 days later with Lm-OVA was correlated with the decrease of bacterial burden . Neonatal immunization with anti-Clec9A-OVA and poly ( I:C ) reduced the amount of Lm in the adult spleen compared to GL117-OVA and GL117-OVA + poly ( I:C ) ( Fig 6C ) . However , it was surprising to observe a significant early reduction of bacterial load in mice immunized with anti-Clec9A-OVA alone . We next assessed the efficiency of the secondary T-cell response after vaccination . Five days after infection with Lm-OVA , the frequency of splenic IFN-γ producing CD8+ T cells in Clec9A-OVA and poly ( I:C ) immunized group after infection was strikingly higher than in all the other groups ( Fig 6D ) . The use of poly ( I:C ) during immunization did increase the production of IFN-γ by the T cells after infection . These results with the OVA model antigen system raise the possibility of selectively delivering Lm antigens to neonatal preCD8α Clec9A+ DCs to trigger a protective immunity against Lm infections .
Quantitative and qualitative deficits in the neonatal innate immune response were proposed as causal factors to account for their inability to mount a protective IFN-γ-dependent CD8+ T-cell response against various viral pathogens such as respiratory syncytial virus , influenza virus , hepatitis B virus , herpes simplex virus as well as intracellular bacterial pathogens such as Lm . Indeed , enhanced neonatal protection against Lm infection can be restored through administering recombinant IFN-γ [33] , through Flt3L treatment [22] or through administering CpG oligonucleotides [34] , both last treatments inducing an IL-12-dependent innate resistance . However , the cellular elements of the neonatal innate immune system able to mount protective type 1 T-cell activation against Lm were not clearly identified , as the IL-12p70-producing CD8α + DC subset was reported to be defective in murine spleen before 7 days of age [2] . It has been well established that CD8α + DCs play a critical role in mounting an effective cytotoxic CD8+ T cell response to Lm infection in adult mice [15 , 35] . CD8α+ DCs are crucial for both efficient bacterial entry into the spleen and induction of the immune response [17 , 36] . In this work , we first demonstrated that a murine splenic neonatal CD8α-type DC precursor subset is predominant in the Batf3-dependent DC lineage before 6 days of life . These preCD8α Clec9A+ DCs express CD11chigh , DNGR1/Clec9A , CD24 , CD205 , and MHCII , without expression of CD8α , B220 , CD4 or CD207 . They are Batf3-dependent and expand preferentially after a limited Flt3L-treatment ( 3 days instead of 7 days ) , compared to pDC or cDC2 , which are respectively the transcription factor and the growth factor known to be involved in CD8α+ DC subset development [6 , 28] . They share the lineage features of CD8α+ DCs [37–39] and are most likely converted in vivo into CD8α + DC . These splenic DC precursors are still present in adults but constitute a minor population of the Batf3 lineage as reported by Bedoui et al . who described a similar immediate splenic CD11chigh , CD24+ , MHCII+ , CD205+ , CD8α- precursor subset capable of cross presenting Ag but poorly addressed their ability to produce cytokines leading to Th1/Tc1 response [40] . We further characterized these adult splenic DC precursors by demonstrating that they were high producers of IL-12p40 exclusively . We also demonstrated that in addition to phagocytosing Lm , neonatal preCD8α Clec9A+ DCs possess the ability to efficiently cross-present Ag , both in vitro and in vivo , an ability well described to be restricted to CD8α+ DCs [6 , 11 , 12 , 14] . Furthermore , GM-CSF enhances and allows full expression of their capacity to cross-present Ag , similar to the newly formed CD8α lineage DCs in Flt3 ligand stimulated bone-marrow cultures that require a maturation step promoted by GM-CSF to acquire the capacity to cross-present Ag [29] . We therefore conclude that the only CD11chigh DCs subset potentially able to cross-present Ag in neonatal spleen are the preCD8α Clec9A+ DCs , in contrast to adult spleen in which both the few CD24+CD8α- DCs and the predominant CD24+CD8α+ DCs have this capacity [40] . We clearly demonstrated that neonatal preCD8α Clec9A+ DCs are able to produce optimal levels of IL12p40 in response to poly ( I:C ) and Lm without IL-12p70 and IL-23 secretion . This is in accordance with the study which previously described the IL-12p35 gene expression defect of newborn monocyte-derived DC [41] . However , the secretion of IL-12p70 during early life has also been shown to be environment dependent . Indeed , one-day-old purified CD11c+ DCs were reported to be able to produce IL-12p70 in response to CpG if a cocktail of maturation agents like GM-CSF and IL-4 were added [23] . Here , we determined that the preCD8α Clec9A+ DCs are the subset able to produce IL-12p70 in response to CpG in combination with GM-CSF , IL-4 and IFN-γ . The fact that neonatal preCD8α Clec9A+ DCs did not produce IL-12p70 upon Lm and poly ( I:C ) stimulation , even in the presence of GM-CSF , IL-4 and IFN-γ , indicates that they are refractory to maturation through these pathways . Interestingly , we have shown that IL-12p40 secreted by neonatal preCD8α Clec9A+ DCs plays a role in neonatal T cell immunity . The inhibition of IL-12p40 during Lm infection increased the neonatal mortality and reduced significantly the Ag-specific CD8+ T-cell expansion and activation . Furthermore , the ability of neonatal preCD8α Clec9A+ DCs to cross-present OVA Ag in vitro or ex vivo after Lm-actA-/-OVA incubation was significantly inhibited when IL-12p40 was neutralized but not IL-12p35 or IL-23p19 . Some studies support an independent role for IL-12p40 and more precisely for the IL-12p ( 40 ) 2 homodimeric form . It has been shown that IL-12p40 could act negatively by competitively binding to the IL-12 receptor in an IL-12 mediated shock [42] or by inhibiting IL-23 functions [43] . In contrast , it has also been suggested that IL-12p40 could have a positive role in inducing immune responses [44] . It has been shown that IL-12p40 promotes macrophage inflammation , DC migration and has a protective function in Mycobacterial infection [45 , 46] . In line with our findings , it was also demonstrated that IL-12p ( 40 ) 2 was involved in activation of naive T cells and in the induction of IFN-γ production by CD8+ T cells [47 , 48] . Therefore , IL-12p70 but also IL-12p40 may act as a feed-back loop on costimulatory molecules and MHC molecule expression on dendritic cells to increase naive T cell activation and IFN-γ production [49] . We may therefore conclude that the CD8+ T cell immunity induced in early life against Lm is IL-12p40 dependent and IL-12p70 or IL-23 independent . These data are in contrast to previous studies showing the requirement of IL-23 in the protection against Lm in adults although this role was mostly associated to the activation of IL-17A/IL-17F producing γδ T cells [50] without identifying the cellular source of IL-23 . Concerning the role of IL-23 in the CD8+ T cell response to Lm , it seems to be minor in adult mice [51] . A surprising finding was that , in addition to secreting IL-12p40 , neonatal preCD8α Clec9A+ DCs produce IL-10 . It was already known that neonatal mice display an increased production of IL-10 early in the course of infection with Lm and after CpG stimulation , but the source of IL-10 in these neonatal studies had been shown to be macrophages and CD5+ B cells [52 , 53] . This is the first demonstration that Batf3-dependent DC precursors produce IL-10 , suggesting a new mechanism responsible for suboptimal activation of neonatal CD8+ T cells . Indeed , blocking IL-10R during cross-presentation after Lm stimulation enhanced the production of IFN-γ by Ag-specific CD8+ T cells . Numerous publications indicate a regulatory role of IL-10 in DC activation and in the Th1/Th2 polarization in both adults and neonates [52–54] . The direct impact of IL-10 secreted by neonatal preCD8α Clec9A+ DCs in T-cell polarization will be analyzed in future investigations . We demonstrated that these neonatal preCD8α Clec9A+ DCs could be used as a cellular target for direct delivery of Lm Ags in order to induce efficient immunization when poly ( I:C ) was co-administrated , allowing a later effective secondary immune response against Lm infection . This is in line with previous studies showing that delivering Ags into the cytoplasm of APCs , with for instance synthetic microspheres , was the key for a better induction of neonatal CD8+ T cell response [55–57] . Previous studies have shown that DNGR1/Clec9A excels as a target for enhancing CD8+ T-cell response and to generating follicular helper T cells in the presence of poly ( I:C ) , in part due to its restricted expression , predominantly in the CD8α+ DC lineage and at a lower level in PDC [30 , 58 , 59] . We observed that injection of anti-Clec9A-OVA construct with poly ( I:C ) in 3-day-old neonates enhances the frequency of OVA-specific IFN-γ producing CD8+ T cells . Furthermore , the absence of CD8+ T-cell proliferation in Batf3-/- mice following anti-Clec9A-OVA construct injection confirms a specific involvement of neonatal CD8α- DCs in this process , and argues against a potential role for pDCs . Specifically , we showed here for the first time that a single treatment with anti-Clec9A-OVA construct and poly ( I:C ) at 3 days of life is enough to significantly enhance the protection of mice against later exposure to the Lm-OVA strain . This protective secondary response was associated with a control of the bacterial burden and a memory CD8+ T cell response involving Ag-specific IFN-γ producing CD8+ T cells . The poly ( I:C ) treatment was shown to induce in vitro the IL-12p40 but not the IL-10 secretion by isolated preCD8α Clec9A+ DCs , inhibiting their regulatory properties . In summary , we have characterized a preCD8α Clec9A+ DC subset that is predominant in mouse spleen during the neonatal period . Compared with their adult counterpart or to the adult CD8α+ DCs , this neonatal Batf3-dependent DC precursors , that cross-present Ag , display the unique abilities to be high producers of IL-12p40 but also of IL-10 . Upon infection with Lm , we demonstrated that these preCD8α Clec9A+ DCs are endowed with regulatory properties that control the CD8+ T cell response through IL-10 . The capacity of these neonatal preCD8α Clec9A+DCs to induce a protective type 1 T-cell immune response against intracellular pathogens was allowed with anti-Clec9A construct and poly ( I:C ) treatment through a mechanism involving only the IL-12p40 subunit production with no IL-10 . This discovery opens new strategies for future human vaccine development . It requires the investigation of the ontogeny of the human equivalent of these neonatal preCD8α Clec9A+ DCs .
C57BL/6 CD45 . 2 , C57BL/6 CD45 . 1 , C57BL/6 IL-12p40-/- , C57BL/6 IL-12p35-/- , Batf3-/- [6] and OT-I TCR transgenic ( OT-I ) mice were purchased from Jackson Laboratory ( Bar Harbor , USA ) . C57BL/ 6 IL-23p19-/- mice , with EGFP reporter gene , were kindly provided by E . Muraille ( Université Libre de Bruxelles , Belgium ) . Mice were bred and housed in our specific pathogen-free animal facility . For all experiments , neonatal mice are defined as 3-day-old and adults as sex-matched 8-to 12-week-old mice . They were kept in sterile confinement in a P2 animal unit during infections . All animal studies were approved by the institutional Animal Care and Local Use committee . The animal handling and procedures of this study were in accordance with the current European legislation ( directive 86/609/EEC ) and in agreement with the corresponding Belgian law “Arrêté royal relatif à la protection des animaux d'expérience du 6 avril 2010 publié le 14 mai 2010” . The complete protocol was reviewed and approved by the Animal Welfare Committee of the Institute of Biology and molecular medicine ( IBMM ) from the Université Libre de Bruxelles ( ULB , Belgium ) ( Permit Number: 2014–43 ) . Lm-EGD strain ( Lm ) , Lm-EGD strain deficient for actA ( ( Lm actA-/- ) and Lm-GFP strain were kindly provided by Prof . P . Cossart ( Pasteur Institute , Paris , France ) . Lm-OVA and Lm actA-/--OVA were purchased from DMX incorporated ( Philadelphia , PA ) [60] . Bacteria were cultured in BBL Brain Heart Infusion ( BHI ) medium ( BD Diagnostics , USA ) and stored at -80°C in 10% DMSO . For survival studies , mice were injected i . p . for 7-day-old mice and i . v . for 3-day-old neonates or adults with different doses of Lm diluted in sterile-PBS . To determine the median lethal dose ( LD50 ) of neonates , 3 or 7-day-old and comparative adult C57BL/6 mice were injected with 4 doses ( 50 , 100 , 1000 and 10000 CFU ) of Lm WT and survival rates were observed for 15 days ( S8 Fig ) . All 3-day-old mice died 2 days following 10000 CFU and 5 days following 1000 CFU of Lm inoculation; only 20% survived after 100 CFU infections and 60% after 50 CFU . In contrast , adults survived every Lm dose administered . The 7 day-old mice showed intermediate sensitivity , dying 8 days after the highest dose of Lm , 50% surviving 1000 CFU and all surviving 100 CFU and 50 CFU . To quantify Lm burden , spleens were harvested 3 days after Lm infection and homogenized in LPS-free PBS using gentleMACS Dissociator ( Miltenyi Biotec , Leiden The Netherlands ) . Serial dilutions of homogenates were plated on BHI-Agar for 24h at 37°C and bacterial CFUs were assessed . For primary responses and/or vaccination , neonates mice were i . p . injected with 5 x 105 CFU of Lm actA-/- or Lm actA-/--OVA and adult mice were injected with 5 x 105 or 5 x 106 CFU Lm actA-/- . Neonates were vaccinated i . v . with 0 , 1μg of anti-Clec9A/OVA mAb or GL117/OVA control mAb . For secondary responses , 60 days after the first immunization , mice were i . v . challenged with 5 x 105 CFU of Lm-OVA . When indicated , neonates were injected with neutralizing purified NA/LE Rat anti-Mouse IL-12p40/p70 ( clone C17 . 8; 25μg/neonate 6 hours before , and 1 and 3 days after Lm injection ) ( BD Biosciences ) or isotype-matched Ab ( BD Biosciences ) and with poly ( I:C ) ( 1 mg/kg; Sigma-Aldrich ) . Mice were eventually s . c . injected with 50μl of saline buffer ( NaCl 0 , 9% ) or Flt3L ( 20μg/ml ) at day 0 , 1 and 2 of life ( Celldex , Phillipsburg , New Jersey ) . Seven days after primary response or 5 days after Lm-OVA challenge , spleen cells were harvested and cultured with OVA257-264 peptide ( 1μg/ml , Polypetides Laboratories , Strasbourg , France ) in complete culture RPMI 1640 medium ( Lonza Research Products , Switzerland ) as described ( 58 ) . When indicated , IL-2 was added to the culture ( 10ng/ml , R&D Systems , Minneapolis , USA ) . Production of IFN-γ by CD8+ T-cells was measured by cytometry and ELISA . OT-I cells were isolated from lymph nodes of OT-I TCR transgenic mice using the Dynabead untouched mouse CD8 cell protocols ( Invitrogen , Life Technologies Europe B . V , Ghent , Belgium ) . CFSE-labeling ( CellTrace CFSE Cell Proliferation Kit , Invitrogen , Life Technologies Europe B . V , Ghent , Belgium ) or VPD450-labeling ( Violet Proliferation Dye 450; BD Biosciences ) were done following the manufacturer’s protocol . C57BL/6 and Batf3-/- neonates were injected i . v . with 3x105 unlabeled or CFSE-/VPD450-labeled OT-I cells and with anti-Clec9A/OVA or GL117/OVA construct ( 0 , 1μg/mouse ) or Lm actA-/--OVA ( 5x105 CFU/mouse ) . Spleen cells were harvested 60h later . Proliferation of OT-I T cells was assessed on a Cyan ADP cytometer ( Dako Cytomation , Everlee , Belgium ) by the dilution of CFSE staining . All the following fluorochrome-conjugated mAbs ( B220 ( RA3-6B2 ) , CD3ε ( 500A2 ) , CD4 ( GK1 . 5 ) , CD8 ( 53–6 . 7 ) , CD11b ( M1/70 ) , CD11c ( HL3 ) , CD19 ( 1D3 ) , CD24 ( M1/69 ) , CD44 ( IM7 ) , CD62L ( MEL-14 ) , CD80 ( 16-10A1 ) , CD86 ( GL-1 ) , CD117 ( 2B8 ) , MHCII ( M5/144 . 15 . 2 ) and Sirpα ( P84 ) were purchased from BD Biosciences , fluorochrome-conjugated anti-CD205 mAb ( NLDC-45 ) was purchased from BioLegend and anti-PDCA1 ( eBio129c ) and CD207 mAb ( eBioL31 ) from ebiosciences ( San Diego , CA , USA ) . Anti-DNGR1 mAb was a kind gift from Dr . Caetano Reis e Sousa ( Immunobiology Laboratory , Cancer Research UK’s London Research Institute ) . For DC characterization , neonatal spleens and lymph nodes from C57BL/6 or Batf3-/- mice were harvested and disrupted using a Pyrex Potter tissue homogenizer ( VWR ) . Red blood cells were lysed by Ammonium-Chloride-Potassium ( ACK ) Lysing Buffer . Cells ( 2-5x106 ) were stained in FACS buffer ( PBS/0 , 5% BSA/2mM EDTA ) at 4°C in the dark for 20 min . After fixation in 1% paraformaldehyde ( Sigma-Aldrich BVBA , Diegem , Belgium ) , analysis was performed on a Cyan ADP ( Dako Cytomation , Everlee , Belgium ) . For DC and CD8+ T-cell intracellular staining , splenocytes were incubated at 37°C for 4h in complete culture medium with Golgiplug ( 1μl/ml; BD Biosciences ) . Cells were then harvested and stained with extracellular mAbs . Intracellular staining ( IFN-γ , clone XMG1 . 2 and IL-12p40 , C15 . 6 ) was then done following the manufacturer’s protocol ( Cytofix/Cytoperm; BD Biosciences ) . For DC sorting and adoptive transfer , spleen cells were first labeled with anti-CD3 , -CD19 , -B220 and -Gr1 biotinylated mAbs for negative selection using BD IMag Streptavidin Particles Plus DM ( BD Biosciences ) , following the manufacturer’s protocol . Enriched splenocytes were then injected or stained to sort CD11chighCD11b-CD205+CD24+ CD8α- DCs ( termed CD8α- DCs ) or CD8α+ DCs ( termed CD8α+ DCs ) on a BD FACSAria II Cell sorter ( BD Biosciences ) . For phagocytosis assays , spleen cells from neonates were cultured with Lm-GFP ( MOI 1:5 ) for 2 hours and stained with CD11c , CD11b , CD205 and CD8α specific mAbs for FACS analysis . 30x106 pre-purified C57BL/6 CD45 . 2+ neonatal splenocytes , collected from 20 to 30 neonates by negative selection as described above were i . v . injected in C57BL/6 CD45 . 1+ adults . Recipient spleen cells were harvested at different time points and were stained to follow CD8α expression on CD45 . 2+ CD11c+CD11b-CD205+CD24+ cells by flow cytometry . 2x104/well of indicated sorted DCs ( collected from 40 to 60 neonates and 5 to 7 adults ) were cultured for 24h . with CpG ( 2μg/mL ) , poly ( I:C ) ( 10μg/mL ) and Lm ( MOI 1:1 ) with or without GM-CSF ( 20 ng/mL , R&D ) , IL-4 ( 20 ng/mL , R&D Systems ) and IFN-γ ( 20 ng/mL , R&D ) . Supernatants were harvested for cytokine measurements by ELISA . For antigen presentation assay , sorted CD8α-DCs were cultured at 5000 cells/well in complete RPMI-1640 medium at 37°C in the presence of OVA257-264-peptide ( 1μg/mL , Polypeptide Laboratories , Strasbourg , France ) or Lm-OVA ( MOI 1:5 ) with or without GM-CSF ( 20ng/ml; R&D Systems ) , and purified NA/LE Rat Anti-Mouse CD210 ( 20ng/mL ) ( IL10R , clone 1B1 . 3a ) or Anti-Mouse IL-12p40/p70 ( 600ng/mL ) ( clone C17 . 8 ) or isotype matched rat Ig ( BD Biosciences ) . After 4h , isolated OT-I T cells were added to the cultures at a ratio of 1:5 ( sorted DC/T ) . After 48h , IFN-γ was measured by ELISA . For ex-vivo cross-presentation assay , neonates were injected i . p . with 106 CFU of Lm-actA-/- OVA . Neonatal CD8α- DCs were sorted 24h later and co-cultured at 2x104 cells/well with OT-I T cells at a ratio 1:5 , with or without GM-CSF ( 20ng/ml ) . IFN-γ was measured by ELISA after 48h . For total RNA Quantitative real time PCR and microArrays , neonates and adults were injected or not with 5 x 105 or 5 x 106 CFU of Lm-actA-/- for 24h . Total RNA from 40 , 000–60 , 000 sorted neonatal or adult CD8α- DCs or adult CD8α+ DCs ( collected from 40 to 60 neonates and 5 to 7 adults ) was extracted with phenol/chloroform and purified with the RNeasy microkit ( Qiagen ) according to manufacturer’s instructions . For quantification of transcripts , reverse transcription and quantitative real-time PCR were performed in a single step using the TaqMan RNA Amplification ( Roche Diagnostics ) on a Lightcycler 480 apparatus ( Roche Diagnostics ) . For individual samples , mRNA levels were normalized to those of β-actin . Sequence of primers and probes are available on request . For microArrays , total RNA was amplified using the Ovation PicoSL WTA System V2 ( NuGen ) , labeled with biotin using the Encore BiotinIL Module ( NuGen ) , and applied on Illumina HT12 bead arrays at the GIGA-GenoTranscriptomics platform ( Liège , Belgium ) . Microarray data ( derived from Affymetrix GeneChip arrays HG-U133 plus 2 . 0 ) from CD8α- neonatal and CD8+ adult DCs samples were obtained from the National Center for Biotechnology Information Gene Expression Omnibus . IL-12p40 , IL-10 and IFN-γ Duoset ELISA kits and Mouse IL-23 Quantikine ELISA Kit ( R&D , Minneapolis , USA ) were measured in culture supernatant according to the manufacturer’s instructions . For IL-12p70 ELISA assays , culture supernatant were measured as previously described ( 23 ) . Data are expressed as mean ± SEM . Statistical comparison between experimental groups was analyzed using a two-tailed nonparametric Mann-Whitney test for the CFUs , absolute number/% of cells and cytokine levels or with the logrank test for survival curves ( GraphPad Prism , GraphPad Software , Inc . ) . p values less than or equal to 0 . 05 were considered significant . * = p<0 , 05 , ** = p<0 , 01 , *** = p<0 , 001 .
|
Lm is a gram-positive food-borne pathogen that is the ethiological agent of listeriosis , a worldwide disease reported most frequently in developed countries . It can cause spontaneous septic abortions , fatal meningitis or encephalitis in immunocompromised and pregnant individuals . The murine model of systemic Lm infection has been demonstrated as a useful model to understand host resistance to intracellular pathogens . Neonates are highly susceptible to infections such as Lm , and display low responses to vaccines requiring IFN-γ producing T cells . In the present study , we characterized in murine neonates a precursor of conventional dendritic cells that is able to produce IL-12p40 and IL-10 cytokines and to modulate the development of the adaptive immune response , more particularly the CD8+ T cell response upon exposure to Lm . By targeting Lm-associated antigens to these conventional dendritic cell precursors in neonates , we succeeded to confer a partial protection to a lethal dose of Lm at the adult stage . Our study provides new insights into our understanding of the innate immune response to infections in early life and will help to design new vaccine strategies in newborns .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"blood",
"cells",
"flow",
"cytometry",
"innate",
"immune",
"system",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"immune",
"physiology",
"cytokines",
"spleen",
"immunology",
"developmental",
"biology",
"molecular",
"development",
"cytotoxic",
"t",
"cells",
"research",
"and",
"analysis",
"methods",
"specimen",
"preparation",
"and",
"treatment",
"staining",
"white",
"blood",
"cells",
"animal",
"cells",
"t",
"cells",
"immune",
"response",
"spectrophotometry",
"immune",
"system",
"cytophotometry",
"cell",
"staining",
"cell",
"biology",
"physiology",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"neonates",
"spectrum",
"analysis",
"techniques"
] |
2016
|
IL-12p40/IL-10 Producing preCD8α/Clec9A+ Dendritic Cells Are Induced in Neonates upon Listeria monocytogenes Infection
|
MicroRNAs ( miRNAs ) suppress the transcriptional and post-transcriptional expression of genes in plants . Several miRNA families target genes encoding nucleotide-binding site–leucine-rich repeat ( NB-LRR ) plant innate immune receptors . The fungus Fusarium oxysporum f . sp . lycopersici causes vascular wilt disease in tomato . We explored a role for miRNAs in tomato defense against F . oxysporum using comparative miRNA profiling of susceptible ( Moneymaker ) and resistant ( Motelle ) tomato cultivars . slmiR482f and slmiR5300 were repressed during infection of Motelle with F . oxysporum . Two predicted mRNA targets each of slmiR482f and slmiR5300 exhibited increased expression in Motelle and the ability of these four targets to be regulated by the miRNAs was confirmed by co-expression in Nicotiana benthamiana . Silencing of the targets in the resistant Motelle cultivar revealed a role in fungal resistance for all four genes . All four targets encode proteins with full or partial nucleotide-binding ( NB ) domains . One slmiR5300 target corresponds to tm-2 , a susceptible allele of the Tomato Mosaic Virus resistance gene , supporting functions in immunity to a fungal pathogen . The observation that none of the targets correspond to I-2 , the only known resistance ( R ) gene for F . oxysporum in tomato , supports roles for additional R genes in the immune response . Taken together , our findings suggest that Moneymaker is highly susceptible because its potential resistance is insufficiently expressed due to the action of miRNAs .
MicroRNAs ( miRNAs ) are single-stranded RNA molecules of approximately 20–24 nucleotides in length that are endogenously transcribed from single-stranded non-coding RNA species [1] , [2] . Plant miRNAs were first identified in 2002 [1] , [3] and have been shown to play vital roles in multiple biological processes , including leaf morphogenesis and polarity , floral organ identity , hormone signaling and stress responses [4] , [5] , [6] , [7] , [8] , [9] , [10] , [11] . miRNAs primarily act on their target mRNAs by influencing mRNA degradation or translational inhibition . In contrast to animals , plant mRNAs are not deadenylated prior to miRNA-guided transcript cleavage and degradation . Although there are several examples of translational inhibition of mRNAs by miRNAs in animals [12] , [13] , this phenomenon has only recently been reported in plants [14] . Expression of miRNA genes is regulated by external stimuli , including abiotic ( e . g . , drought , temperature , salinity ) and biotic ( e . g . , pathogens such as viruses , bacteria and fungi ) stresses . During pathogen attack , recognition of microbe-associated molecular patterns ( MAMPs ) by plant pattern-recognition receptors leads to pattern-triggered immunity ( PTI ) resulting in changes in gene expression that result in altered hormone and metabolite levels [15] . Pathogens have evolved effectors to sabotage PTI . In return , plants acquired disease resistance ( R ) genes , to recognize the presence or action of specific effectors , directly or indirectly , and to activate effector-triggered immunity ( ETI ) , a fast and strong form of immunity [16] . A role for miRNAs in regulating genes important for plant defense has been demonstrated for the response to several pathogens [17] . In tomato ( Solanum lycopersicum ) , the levels of miR319/miR159 and miR172 are induced during Tomato leaf curl New Delhi virus ( ToLCNDV ) disease progression [18] . miR393 , miR160 and miR167 are up-regulated in leaves challenged with the virulent bacterial pathogen Pseudomonas syringae pv . tomato ( Pst ) DC3000 [19] . Similarly , miR393 , miR319 , miR158 , miR160 , miR167 , miR165/166 and miR159 are induced , while miR390 , miR408 and miR398 are repressed , in Arabidopsis thaliana ( Arabidopsis ) leaves infected with Pst DC3000 [20] . Treating Arabidopsis with the MAMP flagellin-derived peptide , flg-22 , induces expression of miR393 , a negative regulator of mRNAs for the F-box auxin receptors TIR1 , AFB2 , and AFB3 [21] . miR482a , a member of the miR482/2118 superfamily , targets mRNAs for R proteins , with nucleotide-binding site ( NB ) and leucine-rich repeat ( LRR ) motifs , for degradation both directly and through generation of secondary small interfering RNAs ( siRNAs ) in Nicotiana benthamiana infected with Pst DC3000 [22] , [23] . miR5300 was first identified as a novel tomato miRNA [24] and later classified as a member of the miR482/2118 superfamily [23] . However , regulation of predicted target genes by miR5300 has not yet been reported [25] . Strains of the ascomycete fungus Fusarium oxysporum are ubiquitous soil inhabitants [26] , [27] . Accumulating data indicate that F . oxysporum is a large species complex , with more than 120 formae speciales causing disease in vegetables , fruit trees , wheat , corn , cotton and ornamental crops [26] , [27] . F . oxysporum infects vascular bundles in the plant host , leading to wilt symptoms . Germination of dormant spores in soil results in adherence and invasion of plant roots by fungal hyphae . The hyphae then move from the root cortex to the xylem where production and dissemination of microconidia spores is critical for disease progression [26] . Previous work has demonstrated that the I-2 gene of tomato confers resistance to race 2 strains of F . oxysporum f . sp . lycopersici ( hereafter referred to as F . oxysporum; [28] ) . The I-2 locus encodes a coiled-coil ( CC ) NB-LRR protein that recognizes the avr2 gene product from F . oxysporum [29] . The near-isogenic tomato cultivars Moneymaker and Motelle are susceptible ( i-2/i-2 ) and resistant ( I-2/I-2 ) genotypes , respectively , for I-2 and the response to F . oxysporum infection [30] , [31] , [32] . In this study , we explored a possible role for tomato miRNAs in the differential resistance of Moneymaker and Motelle to F . oxysporum . Our results indicate that two different miRNAs contribute to plant immunity in tomato by influencing mRNA stability or translation of at least three NB domain-containing proteins distinct from I-2 .
We investigated microRNA ( miRNA ) production in roots of tomato during infection with the wilt fungus F . oxysporum through construction of small RNA libraries and deep sequencing . We took advantage of two near-isogenic cultivars that show differential interaction with F . oxysporum – Moneymaker ( susceptible ) and Motelle ( resistant ) [30] , [31] , [32] . We generated a total of four libraries , including: Moneymaker treated with water ( MM_H2O ) , Moneymaker treated with F . oxysporum ( MM_Foxy ) , Motelle treated with water ( Mot_H2O ) and Motelle treated with F . oxysporum ( Mot_Foxy ) . Our goal was to identify miRNAs that were either upregulated in Moneymaker or down-regulated in Motelle after infection with F . oxysporum . Such a pattern of expression would presumably lead to upregulation of potential target mRNAs required for plant defense in Motelle , but not Moneymaker , after infection . Using Illumina sequencing , we obtained a total of more than 27 million high quality small RNA sequences from the four libraries that could be mapped to the tomato genome . Of these , 5 , 743 , 067 were from MM_H2O , 5 , 492 , 955 from MM_Foxy , 4 , 392 , 583 from Mot_H2O and 5 , 497 , 730 from Mot_Foxy ( Table S1 ) . Among all size classes , 24- , 21- and 22-nt small RNA species were the three most abundant ( Fig . 1A ) . These sizes are similar to those previously identified in tomato [22] , [23] . Within the miRNA population of sequences , more than 98% of the reads began with a uracil . It has been demonstrated that Argonuate proteins recruit small RNAs based on the 5′ terminal nucleotide: AGO2 and AGO4 recruit small RNAs with 5′ terminal adenosine , whereas AGO1 and AGO5 recruit small RNAs with a 5′ terminal uracil and cytosine , respectively [33] , [34] , [35] . We identified 82 predicted miRNAs with at least one raw sequence read in one of the four libraries ( Table S2 ) . miRNAs were considered for further analysis if there were at least 12 raw sequence reads and at least a two-fold change between the F . oxysporum-infected and control plant libraries . Based on these criteria , we identified 18 unique miRNA sequences corresponding to plant disease resistance , stress responses , transcription factors , and others ( Fig . 1B ) . Notably , among all of the regulated miRNAs identified , miR403 and miR398 are associated with disease resistance in other plant species [36] , [37] . miR398 is also implicated in the regulatory network for additional abiotic stresses , including salinity , water deficit , oxidative stress , high levels of abscisic acid , ultraviolet light , copper and phosphate deficiency and high sucrose [11] , [38] , [39] , [40] , [41] , [42] , [43] , [44] . In contrast to the other regulated miRNAs , functions for miR5300 have not been previously reported in any plant species , including tomato . As stated above , our objective in this study was to identify miRNAs that were present at increased levels in Moneymaker or decreased levels in Motelle after infection with F . oxysporum . Our results showed that the majority of miRNAs ( 15 ) were present at increased levels in Motelle plants after infection with F . oxysporum and we did not identify any miRNAs that were increased ( or decreased ) at least two-fold in Moneymaker after infection ( Fig . 1B ) . In contrast , slmiR398 , slmiR5300 and slmiR482f were all suppressed at least two-fold in Motelle plants after F . oxysporum treatment ( Fig . 1B ) , consistent with our original hypothesis . Northern blot analysis was performed to analyze expression of the three miRNAs that were demonstrated to be down-regulated in Motelle by deep sequencing . We analyzed 14 additional miRNAs ( 17 total ) in order to avoid excluding other possible candidates due to issues with sequencing data . The 13 miRNAs that could be detected using northern analysis are presented in Fig . 2 . A subset of the small RNA northern blot results was consistent with the deep sequencing data . A caveat to this analysis is that subfamily members ( e . g . , slmiR482a-f; [23] , [45] , [46] , [47] that share significant homology will cross-hybridize during this analysis . Of interest , both slmiR482f and slmiR5300 were decreased in Motelle plants treated with F . oxysporum as detected by both methods ( Fig . 1B and Fig . 2 ) . The reduction observed during northern analysis ( Fig . 2; 53% for slmiR482f and 58% for slmiR5300 ) was similar to that obtained during deep sequencing ( Table S2; 72% for slmiR482f and 61% for slmiR5300 ) . Deep sequencing data showed that slmiR398 was induced by 1 . 89-fold in Moneymaker by F . oxysporum infection , but suppressed by 71% in Motelle ( Table S2 ) . However , these expression trends were essentially reversed in the small RNA northern blot analysis . slmiR398 levels were similar in Moneymaker and Motelle controls , elevated in Motelle treated with F . oxysporum and barely detectable in Moneymaker under the same conditions ( Fig . 2 ) . Thus , the northern results for slmiR398 were reversed relative to those from deep sequencing for infected Moneymaker and Motelle plants . Deep sequencing data indicated that expression of slmiR403 was reduced by 26% in Moneymaker plants , but induced four-fold in Motelle , after treatment with F . oxysporum ( Fig . 1B ) . Although the small RNA northern results also detected slight reduction of slmiR403 in Moneymaker , slmiR403 levels were slightly reduced in Motelle treated with F . oxysporum ( Fig . 2 ) . Based on our original hypothesis , the results from both deep-sequencing and small RNA northern blot analysis pointed to slmiR482f and slmiR5300 as potential regulators of plant defense genes in tomato . Levels of both miRNAs decreased in the resistant Motelle plants after infection with F . oxysporum . These two miRNAs belong to the miR482/2118 Superfamily in tomato [23] , [48] , [49] . We utilized the psRNATarget algorithm [50] that predicts targets of plant miRNAs to identify mRNAs with binding sites for slmiR482f and slmiR5300 ( See Methods ) . For each miRNA , we found several potential targets in the tomato genome ( Fig . S1 ) . Interestingly , all top putative targets for either miRNA encode proteins with full or partial NB domains ( Fig . 3B ) . The binding site for both slmiR482f and slmiR5300 miRNAs is in the P-loop region of the NB domain in each target ( indicated by red arrow in Fig . 3B ) . For slmiR482f , the top two putative targets were Solyc08g075630 ( NB and CC domains ) and Solyc08g076000 ( NB and three LRR domains ) ( Fig . 3A , B ) . Solyc08g075630 has an atypical arrangement , with the CC domain following the NB domain ( Fig . 3B ) . For slmiR5300 , the top two putative targets were Solyc05g008650 and Solyc09g018220 ( Fig . 3A , B ) . Solyc05g008650 contains a truncated NB domain and overlapping DUF3542 and CC motifs ( Fig . 3B ) . We analyzed available RNAseq data , as well as all three reading frames of genomic sequence at the Sol Genomics database ( http://solgenomics . net/organism/Solanum_lycopersicum/genome ) downstream from this gene , but could not find sequence corresponding to the rest of the NB domain . DUF3542 is a domain of unknown function found in eukaryotes and viruses [51] . Interestingly , the CC-NB-LRR domain protein-encoding gene Solyc09g018220 is tm-2 [52] , [53] . Tomato cultivars Motelle and Moneymaker contain tm-2 ( http://tgc . ifas . ufl . edu/vol43/p79 . html ) , the susceptible allele of the Tm-22 locus [53] . Tm-22 is required for durable resistance of tomato to Tomato mosaic virus ( ToMV ) [52] . We next tested the possibility that the presence of slmiR482f or slmiR5300 would suppress expression of the target genes , leading to reduced levels of the encoded mRNAs and proteins . Before quantitating expression of putative target mRNAs , we first determined the expression levels of several control genes in our four RNA preparations using qRT-PCR ( Fig . 4A ) . These included I-2 , required for F . oxysporum resistance [29] , several I-2-homologous genes identified in the Sol Genomics database , and Mi-1 , required for resistance to nematodes and other pests , but not F . oxysporum . Mi-1 was chosen because Motelle and Moneymaker are also near-isogenic for this gene [30] . The results for I-2 and Mi-1 were in agreement with previous findings and the genotypes of the two cultivars [30] . I-2 was not detectable in Moneymaker , but levels increased more than 3-fold in Motelle after infection with F . oxysporum . Mi-1 levels were similar in both cultivars and did not change significantly after F . oxysporum treatment . Expression of the four I-2-homologous genes varied , but none exhibited a significant difference between water control and F . oxysporum exposure . We checked the mRNA levels of the putative targets under water or F . oxysporum treatment conditions in both tomato cultivars using qRT-PCR ( Fig . 4B ) and northern blot analysis ( Fig . S2 ) . The results of qRT-PCR showed that putative slmiR482f target Solyc08g075630 was induced by almost two-fold in Motelle , but unchanged in Moneymaker , after treatment with F . oxysporum ( Fig . 4B ) . The results from northern analysis of Solyc08g075630 closely mirrored those from qRT-PCR ( Fig . S2 ) . Both qRT-PCR and northern analysis demonstrated that Solyc08g076000 mRNA levels were not significantly changed by F . oxysporum treatment in either cultivar , although levels of Solyc08g076000 were elevated in Motelle relative to Moneymaker ( Fig . 4B , Fig . S2 ) . qRT-PCR and northern analysis revealed significant upregulation of slmiR5300 putative target Solyc05g008650 in Motelle ( 3–4 fold ) , but not Moneymaker , after infection with F . oxysporum ( Fig . 4B , Fig . S2 ) . Likewise , slmiR5300 target Solyc09g018220 exhibited 3 . 4- ( Fig . 4B ) or 5 . 9-fold ( Fig . S2 ) up-regulation by F . oxysporum infection in Motelle compared to water , while levels in Moneymaker were unchanged . Taken together , these results support regulation of at least three of the four predicted target genes at the mRNA abundance level by their respective miRNAs . The psRNATarget algorithm results predicted that both slmiR482f predicted targets and one slimiR5300 target ( Solyc09g018220 ) were regulated at the translational level , while the second slmiR5300 target ( Solyc05g008650 ) is regulated at the mRNA cleavage step . The results from our mRNA analysis were consistent with pre- or post-transcriptional regulation of certain target genes , as three targets exhibited elevated transcript levels in Motelle , but not Moneymaker , with fungal infection , while one target ( Solyc08g076000 ) was relatively unchanged . In order to further probe the possible mechanism for regulation of targets by the four miRNAs , as well as determine specificity of the miRNA/target interaction , we conducted Agrobacterium-mediated transient co-expression experiments in N . benthamiana . We used a binary construct to co-express the FLAG-tagged putative target protein gene and the respective miRNA gene . The presence of the FLAG tag would allow us to detect differences in protein levels and thus , possible translational or post-translational regulation of the target by the miRNA . Vectors with no insert , only a target gene , or containing the miRNA gene slmiR166 that does not recognize our predicted targets , were used as negative controls . We first performed qRT-PCR to check the mRNA levels of targets during co-expression ( Fig . 5A ) . In the presence of slmiR482f , expression of its both putative target genes Solyc08g075630 and Solyc08g076000 were not significantly decreased . In contrast , slmiR5300 targets Solyc05g008650 and Solyc09g018220 were greatly suppressed by the presence of the miRNA; in the case of Solyc09g018220 transcript levels were reduced by almost 90% ( Fig . 5A ) . We investigated possible translational control of target gene expression by checking levels of the target proteins , using western blot analysis with antibody against the FLAG-tag that was placed at the N-terminus of each target in our constructs . Our results showed that protein levels of all targets were down-regulated by the presence of the corresponding miRNA ( Fig . 5B ) . Proteins corresponding to slmiR482f target gene Solyc08g076000 and slmiR5300 target gene Solyc05g008650 were difficult to detect ( Fig . 5B ) . These results strongly suggest that slmiR482f and slmiR5300 are responsible for the down-regulation of their respective protein targets . The observation that levels of Solyc08g075630 and Solyc08g076000 proteins were greatly reduced , while transcript amount was only slightly affected , suggests that slmiR482f silences these two targets mainly via translational inhibition . These results are consistent with the predictions from the psRNATarget algorithm described above . On the other hand , both mRNA and protein levels of Solyc05g008650 and Solyc09g018220 were reduced by their corresponding miRNA , suggesting that slmiR5300 mainly acts at the transcriptional level . The results for Solyc09018220 contrast with the prediction of translational inhibition by psRNATarget . Taken together , these results revealed that all in silico predicted target genes are influenced at the mRNA abundance and/or protein level by co-expression of slmiR482f or slmiR5300 . The four miRNAs target cleavage sites were then validated by RNA Ligase-Mediated 5′ Rapid Amplification of cDNA Ends ( RACE ) [54] , [55] using total RNA isolated from the N . benthamiana leaves used for the co-expression studies described above . The 5′ end of the 3′ derived cleavage product without enzymatic pretreatment can be ligated directly to an RNA adaptor with T4 RNA ligase . Gene-specific 5′ RACE primers were designed to yield predicted products of between 300–400 bp if miRNA-guided cleavage occurred in vivo . In this way , miRNA-guided cleavage can be detected by sequence analysis of the cloned PCR products . With each primer , a major PCR product of the size predicted to be generated from a template resulting from a miRNA-guided cleavage event was detected . In all cases , at least 80% of the 5′ ends of inserts terminated at a position corresponding to the miRNA ( Fig . 5C ) . With the exception of Solyc05g008650 , all of the predicted miRNA-mRNA interactions contain mismatched positions . These findings show that perfect base pairing during the miRNA-mRNA interaction is not a strict requirement to guide cleavage of target RNAs , a result which has been reported by several groups [54] , [56] . We investigated a possible role for the four target proteins in resistance to F . oxysporum using a TRV-based virus-induced gene silencing ( VIGS ) system to down-regulate expression of each gene in the resistant tomato cultivar Motelle . For these studies , Phytoene Desaturase ( PDS ) TRV-silenced plants ( TRV-PDS ) were used as a positive control for silencing . The photobleached phenotype was consistently observed on the third and fourth leaves above the inoculated leaves 3–4 weeks after TRV infiltration [57] . Therefore , treatment with F . oxysporum was carried out four weeks after TRV infection . Transcript levels of genes were checked using qRT-PCR prior to F . oxysporum infection . All VIGS plants were tested for expression levels of all four miRNA target genes , as well as Mi-1 and I-2 , in order to detect possible off-target effects of the VIGS constructs ( Fig . 6 ) . The results demonstrated down-regulation of the corresponding mRNA for all four VIGS-target genes , with reductions ranging from ∼60–95% compared to control Motelle plants not treated with a TRV vector ( Fig . 6 ) . The VIGS was specific for the silenced genes , with the exception of one VIGS plants ( #2 ) for the Solyc08g075630 gene , which had significantly lower levels of I-2 expression than the water-treated control . However , the observation that the other three plants were not significantly different is consistent with a specific effect on the Solyc08g075630 gene . The VIGS constructs for all four target genes corresponded to the extreme 3′ end of the ORF and a portion of the 3′ untranslated region . The VIGS constructs for three out of four target genes did not display nucleotide identity with any other genes in the tomato genome using BLAST . However , the Solyc08g076000 VIGS construct exhibited significant nucleotide identity with a region of tomato gene Solyc02g014230 ( 38 identical nucleotides in the longest stretch ) . In order to determine whether the Solyc08g076000 VIGS construct down-regulated expression of Solyc02g014230 , we performed qRT-PCR on the same RNA samples used in Fig . 6 . The results revealed that Solyc02g014230 mRNA levels were not decreased in the Solyc08g076000 VIGS plants ( Fig . S3 ) . Having demonstrated that the VIGS constructs reduced expression of the appropriate genes in tomato , we assessed disease phenotypes for control and VIGS plants . Scoring was performed four weeks after F . oxysporum infection . Control Motelle plants treated with water and VIGS control Motelle plants carrying an empty TRV vector did not exhibit disease symptoms after infection with F . oxysporum ( Fig . 7A ) . As expected , Moneymaker plants infected with F . oxysporum displayed severe wilting symptoms ( Fig . 7A ) . Disease symptoms , including leaf wilting and discoloration , were observed in all VIGS plants inoculated with F . oxysporum but not in water-treated controls ( Fig . 7C; water-treated controls are plant or leaf #1 in each panel ) . All F . oxysporum-infected plants carrying a target VIGS construct grew more slowly than control plants treated with water ( Fig . 7C ) and exhibited wilting at the top leaves . In particular , line 2 of TRV-Solyc05g008650 and line 4 of TRV-Solyc09g018220 exhibited especially severe disease symptoms ( Fig . 7C ) . We quantified the degree of F . oxysporum infection in tomato leaves by amplifying the rRNA Intergenic Spacer Region ( IGS ) from genomic DNA isolated from leaves using qPCR [58] . In the control plants , levels of F . oxysporum were significantly elevated in Moneymaker after infection ( Fig . 7B , left panel ) , while they were relatively unchanged in Motelle with or without TRV vector ( Fig . 7B , center and right panels ) . The extremely high fungal load in Moneymaker after infection with F . oxysporum is in agreement with the severe disease symptoms of these plants ( Fig . 7A ) . For plants carrying a VIGS construct , our data indicate that F . oxysporum levels were elevated in inoculated plants for each of the four target genes ( Fig . 7D ) . The greatest levels were observed in line 2 of TRV- Solyc05g008650 and line 4 of TRV- Solyc09g018220 ( Fig . 7D ) , consistent with disease severity symptoms . However , the phenotypes observed in VIGS plants were not as severe as those of the control Moneymaker plants after infection ( Fig . 7A ) .
In this study , we exploited the availability of near-isogenic susceptible and resistant cultivars of tomato towards F . oxysporum to identify miRNAs important for plant defense . The results with these two cultivars guided our experiments , allowing us to focus on miRNAs that were down-regulated in the resistant Motelle cultivar during infection . We were able to quickly narrow down to a small group of miRNAs and identify two ( slmiR482f and slmiR5300 ) that correlated with disease . Knock-down of the target genes ( Solyc08g075630 and Solyc08g076000 for slmiR482f and Solyc05g008650 and Solyc09g018220/tm-2 for slmiR5300 ) caused the resistant Motelle cultivar to become susceptible to F . oxysporum . Our study provides a platform for differentially expressed miRNAs in tomato after F . oxysporum infection and demonstrates that plant miRNAs are involved in defense against F . oxysporum . Due to extensive DNA sequence homology with another gene ( Solyc02g014230 ) , we were not able to produce a VIGS construct that was specific for the slmiR482f target Solyc08g076000 . However , this construct resulted in susceptibility of the Motelle cultivar to F . oxysporum infection , and accumulation of fungal biomass comparable to that observed during knockdown of the other three miRNA target genes . Furthermore , qRT-PCR results demonstrated that Solyc08g076000 mRNA levels were greatly reduced , while Solyc02g014230 levels were relatively unchanged in the VIGS plants . Although we cannot rule out an effect on expression of Solyc02g014230 at an earlier time point that might affect plant defense , this finding supports an active role for Solyc08g076000 , but not Solyc02g014230 , in resistance to F . oxysporum in tomato . Tomato is one of the most economically important crops and a model system for fruit development . Although whole-genome sequencing of domesticated tomato has made it possible to characterize the entire family of miRNAs , only a small number of miRNAs have been implicated in tomato-specific processes , such as fruit development and ripening [59] , [60] , [61] . Our study showed that some disease resistance or abiotic stress-associated miRNAs , such as slmiR482f and slmiR398 , are suppressed in the resistant tomato cultivar Motelle after F . oxysporum treatment . In addition we determined that slmiR5300 , for which no functions had been previously ascribed [24] , [25] , was also suppressed in Motelle during F . oxysporum infection . SlmiR482 and slmiR5300 are members of the miR482/2118 superfamily and members of this family have been shown to target the p-loop motif in the mRNA of the NB-LRR encoding R genes [23] . The miR482 family has six members , including miR482a-f [23] , [45] , [46] , [47] . Plant defense responses can be activated very rapidly by pathogen infection . Studies in both plant and animal systems have demonstrated that some small RNAs are induced quickly and specifically by various pathogens and diseases [62] , [63] , [64] . Interestingly , employing the psRNATarget algorithm , the top targets predicted for slmiR482f ( Solyc08g075630 and Solyc08g076000 ) and slmiR5300 ( Solyc05g008650 and Solyc09g018220 ) encode proteins with partial or full NB-domains . Surprisingly , Solyc09g018220 is allelic with the susceptible allele of the ToMV R gene tm-2 . This finding implicates tm-2 in resistance to fungal attack in tomato and suggests that susceptible disease resistance alleles could have roles in immunity . It is intriguing to speculate that plants are able to use susceptible disease resistance alleles to broaden the pathogen recognition spectrum . Since slmiR482f should also target both the functional Tm-22 and broken Tm-2 alleles , it is likely that these genes can also perform the role demonstrated here for tm-2 in F . oxysporum resistance . The discovery of the need for additional full-length or truncated genes for a presumed single gene resistance has broad implication in breeding for resistance and transfer of these traits to plants from different families . It remains to be seen whether the interfamily barrier seen in the transfer of R genes is due to the absence of these additional genes [65] . None of the four miRNAs target I-2 and the three non-tm-2 targets are not homologs of known R proteins in tomato or other plant species ( Fig . S4 ) . Surprisingly , silencing each of their respective genes resulted in susceptibility of Motelle tomato to F . oxysporum . Although silencing of Solyc09g018220/tm-2 resulted in the most severe symptoms , the phenotypes of any single gene VIGS Motelle plant were not as drastic as those observed in the susceptible tomato cultivar Moneymaker lacking I-2 . This could be because down-regulation of one NB-domain containing protein is not sufficient to completely abolish effective disease resistance in tomato . Alternatively , the residual transcript levels of target genes in VIGS plants may have produced this outcome . Taken together , these results support the requirement for multiple proteins carrying the NB domain , including tm-2 , in resistance of tomato to F . oxysporum . There are other studies where inadequate expression of resistance to a eukaryotic plant pathogen occurs because resistance genes are suppressed or underexpressed in solanaceous plant species . For example , Tai et al . demonstrated that defense gene expression is relatively reduced in potato cultivars that are tolerant vs . resistant to the fungal pathogen Verticillium dahliae [66] . In another study , resistance to the oomycete pathogen Phytophthora infestans was analyzed by performing global transcriptional profiling of a susceptible and resistant tomato line using microarrays [67] . The resistance is encoded by a yet unidentified QTL locus . The results showed that VIGS of an R gene that was upregulated in the resistant ( but not susceptible ) line after pathogen infection led to reduced resistance in the normally resistant line . The mode of regulation of the transcripts in these studies was not reported , but the possibility remains that some of the affected mRNAs are down-regulated by miRNAs . Recently , several examples have been described for the requirement of a pair of NB-LRR proteins for the recognition of a specific Avr and disease resistance in a number of plant species , including Arabidopsis , N . benthamiana , rice and wheat [68] , [69] , [70] , [71] , [72] , [73] . In great majority of these cases , their corresponding R genes are located next to each other in tight physical linkage . However , in spite this physical linkage , not all these R gene pairs are homologous [69] , [73] . All four NB-domain containing sequences identified in our study are distinct from I-2 and each other . Moreover , all four are located on different chromosomes than I-2 , suggesting a distinct evolutionary mechanism for the I-2 resistance . Similarly , the three non-tm-2 targets do not appear to have a close evolutionary relationship to tm-2 ( Fig . S4 ) . It remains to be determined whether the various targets interact directly with I-2 , participate in the I-2 signaling complex or activate parallel signaling pathways [72] . Our results suggest that slmiR5300 catalyzes cleavage of both target mRNAs , while the two slmi482f targets are regulated at the translational level . There is a precedence for regulation of mRNA cleavage by miR482 family members in tomato [23] . We do not know of any examples where a miR482/2118 miRNA superfamily member regulates targets at the translational level . However , Arabidopsis miRNA172 regulates cell-fate specification as a translational repressor of APETALA2 [74] and miRNA156/157 inhibits translation of the SBP box gene , SPL3 [75] . It is worth noting that miRNAs have been demonstrated to generate secondary siRNAs from the 3′-UTR side of the target RNA sequence , and these secondary siRNAs can regulate gene expression in plants [22] , [45] , [76] , [77] , [78] . In particular , slmiR482a , a miR482/2118 superfamily member , targets the LRR1 mRNA as a siRNA-mediated secondary target [23] . To conclude , our results support the notion that the miR482/2118 superfamily-mediated reduction of gene expression involves multiple NB-domain-encoding genes , including tm-2 , and occurs via mRNA cleavage and/or translational control mechanisms in tomato . It remains to be determined whether introduction of artificial miRNAs that silence mature and or precursor forms of slmiR482f and slmiR5300 could up-regulate target gene expression in the susceptible Moneymaker plants . In this scenario , we would expect that silencing of miRNAs will enhance resistance to F . oxysporum and would therefore be a useful molecular tool to uncover functional roles for the increasing number of discovered miRNAs in tomato .
Two tomato near-isogenic cultivars ( cv . ) Motelle ( I-2/I-2 ) and Moneymaker ( i-2/i-2 ) that exhibit different susceptibilities to the root pathogen F . oxysporum were used for plant infection and library construction . The wild-type Fusarium oxysporum f . sp lycopersici strain used for all experiments is FGSC 9935 ( also referred to as FOL 4287 or NRRL 34936 ) . Profiling experiments were performed on two-week-old tomato seedlings grown at 23°C with a 16/8-h light/dark cycle . Plants were removed from soil and roots incubated in a solution of F . oxysporum conidia at a concentration of 1×108/ml for 30 min . Control tomato plants were treated with water . Plants were then replanted in soil and maintained in a growth chamber at 25°C for 24 h with constant light . Plants were removed from soil , and roots rinsed and excised using a razor blade . Roots were immediately frozen in liquid nitrogen and stored at −80°C . Total RNA was isolated from roots using either a method involving hot phenol extraction [79] or Trizol ( #15596-018 , Qiagen , Grand Island , NY ) according to the manufacturer's recommendations . Small RNA libraries for deep sequencing were constructed as described [80] and sequenced using an Illumina GSII sequencer at Los Alamos National Laboratory ( Los Alamos , NM ) . For small RNA northern blot analysis , 40 µg total tomato root RNA was resolved on 7 M urea/15% denaturing polyacrylamide gels in 1× Tris/Boric Acid/EDTA ( TBE ) . miRNA-specific oligonucleotide probes ( Table S3 ) were end-labeled using γ-32P-ATP ( #M0201 , New England Biolabs , Ipswich , MA; oligonucleotide probes were labeled according to the manufacturer's recommendations ) . Blots were stripped and reprobed using a U6 RNA oligonucleotide probe to provide a loading control . All blots were imaged using a PhosphorImager ( Molecular Dynamics/GE Life Sciences , Pittsburgh , PA ) and band intensities quantified using Imagequant software ( GE Life Sciences ) . Expression of target or control mRNAs was determined using northern blot analysis or quantitative reverse transcriptase PCR ( qRT-PCR ) . For northern analysis , 20 µg of total RNA was resolved on 1 . 2% agarose gels and processed as described [81] . Probe templates were prepared by amplification of genomic DNA using specific primers in PCRs ( Table S3 ) . This was facilitated by the availability of sequence in the 3′ UTR ( which exhibits the greatest diversity between genes , even close homologs ) for most genes . In the case of I-2 , the published sequence was not present in the tomato genome sequence at the Sol Genome database , presumably because the sequenced cultivar lacks I-2 . Therefore , I-2 primers had to be designed from a region of the ORF . In order to ensure specificity of amplification , we were able to identify two primer sequences with at least six mismatches with the other genes ( Fig . S5 ) . Probes were labeled using the random priming method according to the manufacturer's protocol ( #U1100 , Promega , San Luis Obispo , CA ) . Blots were stripped and reprobed using an 18S rRNA probe as a loading control . Blots were imaged and band intensities quantitated as described above for the small RNA northerns . For qRT-PCR analysis , one µg of total RNA was used for cDNA synthesis ( #4368813 , Life Technologies , Grand Island , NY ) according to the manufacturer's recommendations . Amplification of S . lycopersicum miRNA ( slmiRNA ) targets was carried out using qRT-PCR ( iQ5 , Bio-Rad , Philadelphia , PA ) . The sequences of primers used in qRT-PCR are listed in Table S3 . Illumina sequence reads were processed to remove adaptor sequences and quality filtered with FASTX toolkit ( Table S1; http://hannonlab . cshl . edu/fastx_toolkit/ ) . Annotation of miRNA gene expression was determined using CLC Bio Genomics Workbench ( http://www . clcbio . com/ ) miRNA mapping pipeline mapping against miRBase 17 ( http://www . ncbi . nlm . nih . gov/pubmed/21037258 ) ( Table S2 ) . Further validation of expression for closely related copies was applied using Bowtie alignments of short reads against the miRNA sequences of the S . lycopersicum miRNA 482a-f family ( Table S4 ) . To identify differentially expressed miRNAs across F . oxysporum infected and H2O treated tomato plants , we first mapped the pre-processed sequencing reads to the genomic loci of the known miRNAs ( using Bowtie ) based on miRBase annotation ( version 15 , http://www . mirbase . org ) . In order to deal with noise in sequencing , we only considered miRNAs with at least 12 raw sequencing reads mapped in a given library . In the mapping , we allowed up to 3-nt shifts upstream and downstream from the annotated starting locus of a miRNA to compensate for possible variation in Dicer activities ( Table S2 ) . To predict miRNA target genes , we followed the rules for target prediction as described by Allen et al . [76] . The psRNATarget algorithm that predicts targets of plant miRNAs [50] was used in this study ( http://solgenomics . net ) . There were several modifications , including no allowance for gaps in the miRNA . A maximum of three continuous mismatches was allowed if the mismatch region contained at least one G∶U pair . The penalty score of the region was multiplied by 1 . 5 . miRNAs and their corresponding target genes were inserted into vector GATEPEG100 . All constructs were transformed into Agrobacterium tumefaciens strain GV3101 . N . benthamiana plants , seeded directly in pots , were maintained in an incubator at 24°C with 12 h light/12 h dark cycle . A . tumefaciens cultures were grown in liquid LB medium with selection [82] . After 40 h , leaves were harvested , and protein extraction was performed [83] . Proteins were separated on 10% SDS–PAGE gels and transferred onto nitrocellulose membranes ( Millipore , Billerica , MA ) . Membranes were blocked using 5% milk in 1×TBST and then incubated with Anti-FLAG ( DYKDDDDK ) Antibody ( #635691 , Clontech , Mountain View , CA ) followed by a secondary horseradish peroxidase ( HRP ) -conjugated goat anti-Rabbit polyclonal antibody ( #A0504 , Sigma , St . Louis , MO ) according to the manufacturer's recommendations . Reactive species were visualized using SuperSignal West Pico Chemiluminescent Substrate ( #34087 , Pierce , Rockford , IL ) and imaging using a Biochemi system ( UVP , Upland , CA ) . Target validation was done using a RNA ligase-mediated rapid amplification of cDNA ends ( 5′RACE ) assay as described [54] , [55] , with slight modification using the FirstChoice RLM-RACE Kit ( #AM1700 , Invitrogen , CA ) . Total RNA was isolated from N . benthamiana leaves used for co-expression of miRNAs with predicted mRNA target genes . Poly ( A+ ) mRNA was prepared by two rounds of purification with an Oligotex mRNA Midi Kit ( #70042 Qiagen ) and directly ligated to the FirstChoice RLM-RACE Kit RNA Oligo adaptor without further modifications . Gene-specific primers were designed approximately 400 nucleotides to the 3′ side of predicted target sites ( Table S3 ) . The conditions used for this amplification step were those for gene-specific RACE recommended by the manufacturer . VIGS was used to suppress expression of the predicted mRNA targets using TRV-based vectors ( pTRV1 and pTRV2 ) [84] . Gene-specific VIGS constructs could only be developed for three of the four target genes in tomato , due to high nucleotide identity between Solyc08g076000 and Solyc02g014230 ( see Results ) . slmiR482f target genes Solyc08g075630 and Solyc08g076000 and slmiR5300 target genes Solyc05g008650 and Solyc09g018220 were amplified using gene-specific primers ( Table S3 ) and cloned into the pTRV2 vector . A vector carrying a fragment of the Phytoene Desaturase ( PDS ) gene was used as a positive control for silencing [57] . All TRV-VIGS constructs were transformed into A . tumefaciens strain GV3101 . Bacterial cultures were grown as described above . Equal volumes ( OD600 = 1 ) of A . tumefaciens carrying pTRV1 and suspensions containing pTRV2-derived constructs or pTRV2 empty vector were mixed prior to infiltration into leaves of 2 to 3-week-old tomato plants . pTRV2 empty vector was used as the negative control in this study [57] , [84] , [85] , [86] . Plants were maintained at 20°C for four weeks , until photobleaching symptoms were observed in the leaves of PDS TRV-silenced plants . At this time , leaflets were harvested from several plants for isolation of RNA and qRT-PCR analysis of the target genes to assess the degree of silencing . The same plants were then treated with F . oxysporum or water as described above for the small RNA library construction . After four more weeks , plants were scored for disease symptoms . Genomic DNA was isolated from leaves [87] , [88] and relative levels of F . oxysporum determined using qPCR of the rRNA intergenic spacer ( IGS ) sequence of F . oxysporum [58] using specific primer sequences ( Table S3 ) .
|
Fusarium oxysporum is a fungal pathogen that represents a species complex , with members that infect numerous crops . In spite of its importance to agriculture , very little is known about roles of small RNAs in plant immunity against F . oxysporum . In this study , we set up a screen for tomato microRNAs ( miRNAs ) that correlate with resistance to F . oxysporum f . sp . lycopersici by performing deep sequencing of small RNAs from a resistant and susceptible tomato cultivar . We focused on two miRNAs that are uniquely down-regulated in the resistant cultivar during fungal infection . All predicted targets of these miRNAs encode proteins with NB domains , a motif associated with pathogen resistance in plants . Using a heterologous system , we validated that the miRNAs could regulate expression of four targets . Silencing of the target genes in tomato resulted in decreased immunity to F . oxysporum in the normally resistant cultivar . The finding that none of our targets correspond to I-2 , the only known resistance ( R ) gene for F . oxysporum in tomato , supports roles for additional R genes in the immune response . Our results suggest that the potential resistance of the susceptible cultivar is insufficiently expressed due to the action of miRNAs .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"biochemistry",
"rna",
"micrornas",
"plant",
"science",
"plant",
"disease",
"resistance",
"gene",
"expression",
"genetics",
"gene",
"regulation",
"biology",
"and",
"life",
"sciences",
"plant",
"pathology",
"plant",
"pathogens"
] |
2014
|
MicroRNAs Suppress NB Domain Genes in Tomato That Confer Resistance to Fusarium oxysporum
|
Family studies of individual tissues have shown that gene expression traits are genetically heritable . Here , we investigate cis and trans components of heritability both within and across tissues by applying variance-components methods to 722 Icelanders from family cohorts , using identity-by-descent ( IBD ) estimates from long-range phased genome-wide SNP data and gene expression measurements for ∼19 , 000 genes in blood and adipose tissue . We estimate the proportion of gene expression heritability attributable to cis regulation as 37% in blood and 24% in adipose tissue . Our results indicate that the correlation in gene expression measurements across these tissues is primarily due to heritability at cis loci , whereas there is little sharing of trans regulation across tissues . One implication of this finding is that heritability in tissues composed of heterogeneous cell types is expected to be more dominated by cis regulation than in tissues composed of more homogeneous cell types , consistent with our blood versus adipose results as well as results of previous studies in lymphoblastoid cell lines . Finally , we obtained similar estimates of the cis components of heritability using IBD between unrelated individuals , indicating that transgenerational epigenetic inheritance does not contribute substantially to the “missing heritability” of gene expression in these tissue types .
The genome contains a complex set of instructions for the assembly and maintenance of an organism . A fundamental goal in biology is to understand the relationship between genotype and phenotype . This goal can be achieved in part by studying the genetic basis of gene expression , as many genotype-phenotype correlations are a consequence of genetically driven variation in gene expression [1] . A number of studies have mapped individual cis and trans regulatory variants in humans , and recent work has suggested that the majority of regulators act in trans [2]-[5]; regulation of gene expression has also been widely studied in animal models [6]-[9] . However , the bulk of variability in gene expression remains unexplained . Heritability analyses can shed light on the genetic basis of gene expression . Several previous studies have demonstrated substantial overall heritability of gene expression in family data sets , and heritability approaches have also been broadly applied to other phenotypes [10]-[14] . In this study , we used gene expression measurements [11] and genome-wide single nucleotide polymorphism ( SNP ) data [15] from 722 Icelanders from family cohorts to examine the heritability of gene expression in blood and adipose tissue . By studying more than one tissue type , we were able to analyze the regulation of gene expression both within and across tissues . Our goal was to answer three key questions about gene expression heritability . First , can heritability be partitioned into cis and trans components using local and genome-wide IBD between pairs of individuals ? Second , to what extent are heritable components of variance shared across tissues ? Third , to what extent does heritability extend to distantly related individuals inheriting IBD segments from distant ancestors ? We sought to partition the heritability of gene expression into cis versus trans components by comparing the effects of IBD at the genome-wide level ( trans ) to those of IBD at the local level ( cis ) , defined as the number of chromosomes ( 0 , 1 or 2 ) shared IBD at the genomic location containing the expressed gene . Our results show a substantially higher proportion of heritability due to cis regulation , 37% in blood and 24% in adipose tissue , than the 12% reported in a previous ancestry-based study of lymphoblastoid cell lines ( LCL ) in African Americans [16] . One possible explanation for this discrepancy is transgenerational epigenetic inheritance , which is one of the explanations proposed to account for the “missing heritability” in genetic studies of human traits [17]-[23] . Epigenetic inheritance would regulate gene expression at the cis locus , and would be expected to contribute to cis heritability in family-based analyses but not in ancestry-based analyses , given that this mode of inheritance persists over a relatively short time scale . However , by using IBD in distantly related individuals to produce similar estimates of cis heritability , we were able to rule out this hypothesis . Instead , our analyses indicate that the proportion of heritability attributable to cis regulation is tissue-specific , and that similarities in gene expression across tissues are primarily due to heritable cis effects . Thus , the proportion of gene expression heritability attributable to cis regulation is expected to increase as a function of the number of different cell types present in the tissue being assayed , consistent with results obtained from blood , adipose tissue and LCL .
This research was approved by the Data Protection Commission of Iceland and the National Bioethics Committee of Iceland . The appropriate informed consent was obtained for all sample donors . Relative abundances of 23 , 720 transcripts were obtained for blood samples from each of 1 , 001 individuals from the IFB cohort , as described previously [11] ( see Web Resources ) . Values were adjusted for sex and age . We removed 4 , 985 transcripts that either had >5% missing data , did not map to an autosomal chromosome , or mapped to more than one genomic location . We removed 16 individuals with >5% missing data and 269 individuals for which long-range phased SNP data were not available . This left 18 , 735 transcripts and 716 individuals . Most of out analyses focused on 2 , 233 related pairs ( individuals from the same family pedigree with genome-wide IBD >0 . 05 ) spanning a subset of 687 individuals . Relative abundances of 23 , 720 transcripts were obtained for adipose tissue samples from each of 673 individuals from the IFA cohort , as described previously [11] ( see Web Resources ) . Values were adjusted for sex , age and body mass index ( BMI ) , restricting to 638 individuals with BMI data . We removed 4 , 621 transcripts that either had >5% missing data , did not map to an autosomal chromosome , or mapped to more than one genomic location . We removed 2 individuals with >5% missing data and 67 individuals for which long-range phased SNP data were not available . This left 19 , 099 transcripts and 569 individuals . Most of our analyses focused on 1 , 700 related pairs ( individuals from the same family pedigree with genome-wide IBD >0 . 05 ) spanning a subset of 531 individuals . Individuals were genotyped using the Illumina 300K chip . Owing to the sensitive nature of genotype data , access to these data can only be granted at the headquarters of deCODE Genetics in Iceland . Given long-range phased Illumina 300K data [15] for a pair of individuals , we partitioned the genome into 2cM blocks and for each block performed 2×2 = 4 comparisons between haplotypes from the two individuals . We declared two haplotypes to be IBD if they matched at >95% of alleles in the block , non-IBD if they matched at <85% of alleles , and unknown-IBD otherwise . We excluded SNPs with missing data in one or both individuals , so that lack of a match implies a mismatch , and set IBD status to unknown for pairs of haplotypes with >5% of SNPs excluded . We defined local IBD as the total number of comparisons producing a match . We verified that this approach infers 0∶1∶2 copies IBD between parent-child pairs with probabilities 0 . 2%∶99 . 3%∶0 . 4% and 0∶1∶2: copies IBD between sibling pairs with probabilities 24 . 9%∶50 . 1%∶24 . 9% , excluding from this computation the 7% of pairs and blocks for which inferred IBD was unknown . These numbers are a function of the thresholds we used to define IBD and non-IBD; the thresholds were largely chosen for specificity rather than sensitivity since for our application it does not matter that inferred IBD is sometimes unknown . The numbers are very close to the expected theoretical probabilities ( for parent-child pairs , 2 copies IBD is expected to occasionally occur due to IBD in “unrelated” parents ) . This validates our use of long-range phased SNP genotypes to compute local IBD estimates . We computed genome-wide IBD estimates as the average of local IBD estimates across all 2cM blocks . We applied variance-components methods to estimate narrow-sense heritability [14] , [24] . The source code used in our heritability analyses is available for download ( see Web Resources ) . Let egs denote the gene expression for gene g and individual s , normalized to have mean 0 and variance 1 across individuals . Let θst denote the genome-wide IBD between individuals s and t ( 0≤θst≤1 ) and be the N×N matrix of genome-wide IBD , where N is the number of individuals . Let Vg denote the covariance matrix of normalized gene expression for gene g . We consider the model and fit hg2 , the heritability of gene g , to the observed normalized gene expression values egs by maximizing the likelihood , where . Values of egs , hg and Vg vary with tissue type , but we view tissue type as an implicit index rather than an explicit index for simplicity of notation . For both blood and adipose tissue , the estimated values of hg2 were ∼80% correlated to values that were computed previously using similar methods [11] , despite the fact that the current analysis was restricted to a subset of individuals for which long-range phased SNP data was available for local IBD inference . We declare hg2>0 to be nominally significant ( P<0 . 05 ) if hg2 is larger than each value analogously estimated from 19 data sets with sample labels randomly permuted ( thus , 5% of genes will be nominally significant even in the absence of a true effect ) . We estimated average h2 as the average of hg2 across genes g . We computed standard errors on h2 by performing independent runs with sample labels randomly permuted; we obtained identical standard errors using either five permutations or 20 permutations . Similar procedures were used in the cis vs . trans and cross-tissue analyses described below . We extended the variance-components approach to cis and trans heritability via the model , where is the N×N matrix of local ( cis ) IBD between individuals s and t at the genomic location proximal to gene g . We used the midpoint of the gene expression probe to define genomic location , but the value of γgst is not sensitive to this choice as local IBD segments between related individuals span many megabases . We scale γgst to have value 0 . 0 , 0 . 5 , or 1 . 0 ( for 0 , 1 , or 2 copies shared ) . We fit the cis heritability hg , cis2 and trans heritability hg , trans2 by maximizing the usual likelihood , fitting hg2 = hg , cis2+hg , trans2 and hg , cis2 in turn . We average across genes g to estimate hcis2 and htrans2 . We define the proportion of heritable gene expression variation that is due to cis regulation as πcis = hcis2/ ( hcis2+htrans2 ) . As above , all values vary with tissue type , which we view as an implicit index . The cross-tissue correlation ρ was computed as the correlation between normalized expression levels in blood and adipose tissue across genes and individuals . Due to the normalization , this is equal to the average of gene-specific correlations ρg . We computed standard errors of both gene-specific and average cross-tissue correlations via jackknife , repeating the computation with each individual removed in turn and estimating the standard error as times the standard deviation of the N estimates . We now describe our estimation of cross-tissue heritability . Let and denote normalized expression levels for gene g and individual s in blood and adipose tissue , respectively . Let Wg denote the covariance matrix of the vector ( ebg , eag ) of length 2N . Here the relevant equations are where ξ2 denotes cross-tissue heritability , ρ denotes cross-tissue correlation , and denotes the tensor product of a 2×2 matrix with an N×N matrix to form a 2N×2N matrix . For example , the first term of Wg has entries hbg , cis2γgst in the upper left N×N block , ξg , cis2γgst in the upper right N×N block , and so on . This generalization of the variance-components approach to cross-phenotype analyses has been previously described ( for the case of genome-wide IBD ) in an analysis of two height phenotypes , self-reported height and clinically measured height [25] . The likelihood is defined in the usual way , replacing Vg with Wg and eg with ( ebg , eag ) . We fit hbg2 = hbg , cis2+hbg , trans2 , hbg , cis2 , hag2 = hag , cis2+hag , trans2 , hag , cis2 , ρg , ξg2 = ξg , cis2+ξg , trans2 and ξg , cis2 in turn . For each of the parameters estimated , we compute average values by averaging across genes g .
For the analysis of gene expression in blood , we analyzed normalized intensity values for 18 , 735 mRNA transcripts . Analysis was restricted to 687 individuals from the IFB cohort for whom long-range phased SNP data were available ( see Methods ) . For each pair of individuals , we used the long-range phased SNP data to compute the number of chromosomes shared IBD at each location in the genome , and computed the genome-wide IBD as an average of these values ( Figure 1; see Methods ) . Our initial analyses focused on 2 , 233 related pairs with genome-wide IBD >0 . 05 . For the analysis of gene expression in adipose tissue , we similarly analyzed 19 , 099 mRNA transcripts of 531 individuals from the IFA cohort , focusing on 1 , 700 related pairs with genome-wide IBD >0 . 05 ( see Methods ) . The IFA cohort largely overlaps the IFB cohort , with 496 of the 722 individuals analyzed appearing in both cohorts . We estimated the overall heritability hg2 for each gene g using variance-component methods [14] ( see Methods ) . Although estimates for each gene g are statistically noisy at these sample sizes , histograms show a clear positive bias for both IFB and IFA cohorts ( Figure S1 and Table S1 ) , and hg2>0 was nominally significant ( P = 0 . 05; see Methods ) for an excess of genes: 42% for IFB and 63% for IFA . We computed the average h2 as the average of hg2 across genes g . A relevant question is whether or not to allow negative values of hg2 when computing this average [26] . Such values have no biological interpretation ( except in the case of negative correlation among siblings in traits that depend on birth order ) . However , because values close to zero may be either increased or decreased by statistical noise—leading to negative estimates of hg2 for 3 , 031 of 18 , 735 genes for IFB and 1 , 038 of 19 , 099 genes for IFA—we elected to allow negative values in our main computations so as to produce an unbiased estimate of average h2 . We obtained estimates of h2 = 0 . 150 for blood and h2 = 0 . 234 for adipose tissue . We obtained similar results when using a regression-based approach to estimate average h2 ( Text S1 ) , which more readily lends itself to visualization ( Figure 2A and 2B ) . ( When clipping negative hg2 values to zero , we obtained h2 = 0 . 159 for blood and h2 = 0 . 237 for adipose tissue . ) Our results are consistent with previous analyses reporting that expression levels of a substantial fraction of genes are significantly heritable at the level of h2 = 0 . 3 or higher [10]-[13] , [26] . While estimates of overall heritability are based on genome-wide IBD , it is possible to estimate cis versus trans heritability by extending variance components to consider both local ( cis ) IBD at the genomic location close to the expressed gene , and genome-wide ( trans ) IBD ( see Methods ) . As before , analyses were restricted to 2 , 233 and 1 , 700 related pairs from the IFB and IFA cohorts , respectively . Histograms of hg , cis2 and hg , trans2 estimates for each gene g show a clear positive bias for both IFB and IFA cohorts ( Figure S2 and Table S1 ) , with an excess of nominally significant ( P<0 . 05 ) genes for IFB ( hg , cis2>0: 16%; hg , trans2>0: 19% ) and IFA ( hg , cis2>0: 16%; hg , trans2>0: 30% ) . For IFB , we obtained average cis and trans heritability estimates of hcis2 = 0 . 055 and htrans2 = 0 . 095 , respectively , which sum to h2 = 0 . 150 . This leads to the conclusion that the proportion of heritability of expression due to cis variants in blood is πcis = 37% . For IFA , we obtained estimates of hcis2 = 0 . 057 and htrans2 = 0 . 177 , which sum to h2 = 0 . 234 . This yields an estimate of πcis = 24% in adipose tissue . The values of h2 and htrans2 in adipose tissue are significantly higher than for blood , but hcis2 is similar , leading to a lower value of πcis . We obtained similar results when using a regression-based approach to estimate average hcis2 and htrans2 ( Text S1; Figure 2C and 2D ) . We note that there is considerably less statistical uncertainty in estimates of hcis2 ( Figure 2C and 2D ) than in estimates of h2 ( Figure 2A and 2B ) . Indeed , we obtained standard errors of h2 = 0 . 150±0 . 011 , hcis2 = 0 . 055±0 . 001 and htrans2 = 0 . 095±0 . 010 for blood and h2 = 0 . 234±0 . 011 , hcis2 = 0 . 057±0 . 002 and htrans2 = 0 . 177±0 . 010 for adipose tissue ( see Methods ) . These standard errors are 7-100 times lower than standard errors for single-gene heritability estimates , which are inadequate for estimating πcis ( see Text S1 ) . The much lower standard errors for hcis2 are a consequence of variation in cis IBD across the genome that decouples the estimation of this parameter from the systematic noise covariance structure across all pairs of individuals ( see Text S1 ) . Based on these standard errors for hcis2 and htrans2 , πcis has little statistical uncertainty , although results may be affected by modeling uncertainty . Our heritability model does not account for the possibility of phenotypic similarity in related individuals due to shared environment , which can confound estimates of heritability [14] . We note that such effects would inflate estimates of h2 and htrans2 , but have a negligible impact on hcis2 , since the extent of shared environment would be related to genome-wide ( trans ) rather than local ( cis ) IBD . To investigate the possibility of confounding due to shared environment , we computed the average correlation in gene expression between spouses , who are genetically unrelated but have a shared environment . We observed average correlations of 0 . 074±0 . 042 in 33 IFB spouse pairs and 0 . 076±0 . 035 in 28 IFA spouse pairs , which are similar in magnitude to correlations between sib-sib or parent-child pairs that correspond to the average heritabilities reported above ( see Text S1 and Table S2 ) . Thus , there is strong evidence that shared environment can lead to similarity in gene expression phenotypes . We further investigated whether the gene by gene signature of correlations in spouse pairs matches the signature of correlations in sib-sib or parent-child pairs or estimates of hg2 , but found that it does not ( see Text S1 and Table S3 ) . Thus , we hypothesize that the correlations in spouse pairs are due to very recent shared environment ( e . g . diet ) arising from sharing the same household , whereas the correlations in sib-sib and parent-child pairs in this study ( who are unlikely to share the same household , since only adult individuals were sampled ) are due to genetic heritability . However , we cannot rule out a small amount of inflation in h2 and htrans2 estimates due to shared environment in related individuals . Our family-based estimates of πcis in blood and adipose tissue are considerably greater than a previous estimate of 12±3% obtained using lymphoblastoid cell lines ( LCL ) from African-Americans , in which local versus genome-wide European ancestry was used to infer the relative contribution of cis versus trans heritability [16] . An analogous ancestry-based analysis of LCL gene expression data [27] from admixed HapMap 3 Mexican-Americans [28] has produced a similarly low value of πcis = 13±9% . One possible explanation for the lower values as compared to family-based estimates could be the epigenetic inheritance of cis-acting factors other than DNA sequence that are transmitted from parent to offspring . Given the relatively short time scale of epigenetic inheritance , this would be expected to have a much greater impact on family-based estimates of πcis than those based on ancestry [22]-[23] . To further explore the epigenetic hypothesis , we repeated the cis versus trans analysis using subsets of unrelated or distantly related individuals ( genome-wide IBD <0 . 01 ) from the IFB and IFA cohorts . The mean genome-wide IBD for all such pairs of individuals was 0 . 0044 , with a standard deviation of 0 . 0018 , consistent with the known properties of distant relatedness between “unrelated” individuals from Iceland as well as other world populations [29]-[31] . We independently generated five random subsets of IFB individuals ( 85 , 87 , 92 , 93 , 91 individuals ) and five random subsets of IFA individuals ( 127 , 85 , 92 , 95 , 89 individuals ) with genome-wide IBD <0 . 01 between each pair of individuals in each subset , such that each subset was maximal subject to this constraint . The resulting estimates of hcis2 were 0 . 057±0 . 008 for blood and 0 . 067±0 . 005 for adipose tissue ( mean ± standard deviation across five subsets ) . These estimates of hcis2 were close to our previous estimates based on closely related pairs , thereby ruling out a substantial contribution of epigenetic inheritance to cis heritability ( see Discussion ) . However , we did not obtain meaningful estimates of htrans2 using distantly related individuals , due to the systematic noise covariance structure ( see Text S1 ) , and therefore πcis could not be estimated . We note that similar results for distantly related pairs were obtained using different IBD estimation algorithms ( see Text S1 ) . We conducted a cross-tissue analysis of expression heritability in blood and adipose tissue in 496 individuals who overlapped between the IFB and IFA cohorts . We determined that an individual’s blood expression for a particular gene is slightly but significantly correlated to the same individual’s adipose expression for the same gene , with an average correlation of ρ = 0 . 041±0 . 005 ( mean ± standard error ) ( see Methods ) . Although estimates for each gene g are statistically noisy at these sample sizes , histograms show a clear positive bias in ρg ( Figure S3 ) , and ρg>0 was nominally significant ( P = 0 . 05 ) for 20% of genes , a significant excess . We next investigated the relationship between an individual’s blood expression and a related individual’s adipose expression , using variance-components methods ( see Methods ) . This revealed that cross-tissue similarity varies with the level of family relatedness , with an average cross-tissue heritability estimate of ξ2 = 0 . 030±0 . 006 . Analogous to the analyses for single tissues , we partitioned the cross-tissue heritability into cis and trans components , yielding values of ξcis2 = 0 . 031±0 . 001 and ξtrans2 = -0 . 001±0 . 006 . We obtained similar results using regression-based approaches ( Text S1; Figure 3A and 3B ) . Histograms of cross-heritability estimates for each gene g show a positive bias for ξg2 and ξg , cis2 , but not ξg , trans2 , for which the histogram is symmetric about zero ( Figure S4 ) . While our estimate of ξtrans2 is not significantly different from zero , ξcis2 is highly significant and explains the bulk of our estimate of ρ . This implies that the extent to which gene expression in blood and adipose tissue is similar across genes and individuals is dominated by heritable effects at the cis locus . Our finding that cross-tissue similarities are dominated by heritable cis effects leads to the mathematical result that πcis is expected to increase with tissue heterogeneity: as the number of cell types represented in a tissue increases , the strongly correlated cis effects will add linearly but the uncorrelated trans effects will be diluted . In detail , let x and y denote cells types and suppose that Cov ( exgs , exgt ) = Cov ( eygs , eygt ) = hcis2γgst+htrans2θst for all genes g and individuals s≠t , and that all cis effects ( but no trans or non-genetic effects ) are shared across cell types . Thus , Cov ( exgs , eygt ) = hcis2γgst . Now consider a tissue z containing cell types x and y . Up to a normalization constant , Cov ( ezgs , ezgt ) = Cov ( 0 . 5 ( exgs+eygs ) , 0 . 5 ( exgt+eygt ) ) = hcis2γgst+0 . 5htrans2θst , so that πcis , z = hcis2/ ( hcis2+0 . 5htrans2 ) is larger than πcis , x = πcis , y = hcis2/ ( hcis2+htrans2 ) . We verified this theoretical result empirically by defining ezgs = ebgs + eags as the average of normalized gene expression in blood and adipose tissue , normalized to mean 0 and variance 1 . For synthetic tissue z , we obtained the value πcis = 0 . 41 , which is larger than the value of πcis for either blood or adipose tissue , and similar to the predicted value of 0 . 055/ ( 0 . 055+0 . 25 ( 0 . 095+0 . 177 ) ) = 0 . 45 based on hcis2 and htrans2 ( πcis < 0 . 45 is actually expected since not all cis effects are shared ) . Thus , the variability in πcis across tissue types ( 0 . 12 for LCL , 0 . 24 for adipose , 0 . 37 for blood ) is consistent with the fact that LCL represent a single cell type , whereas adipose tissue and blood contain many cell types: adipose tissue contains smooth muscle cells , fibroblasts , adipocytes , mast-cells and endothelial cells , while blood contains erythrocytes , thrombocytes , neutrophils , lymphocytes , monocytes , eosinophils and basophils in proportions that vary across individuals [32]-[34] . This also explains why studies of individual cell types have been more successful in identifying trans eQTLs than studies of whole tissues , and why most replications across tissue types occur at cis eQTLs [11] , [34]-[37] .
In this study , we observed a greater contribution of cis regulation in blood and adipose tissue than in a previous ancestry-based analysis of LCL in African-Americans [16] . This result is not sensitive to sample size , because although estimates for individual genes are statistically noisy , we considered averages across genes . We also observed that cross-tissue similarity between blood and adipose expression is genetically heritable and dominated by cis effects . These two results are highly concordant . Due to the dilution of trans effects that are not shared across cell types , cis regulation is expected to explain a greater proportion of heritability in tissue types that are heterogeneous in their cell composition , such as blood and adipose tissue—particularly blood , in which cell type proportions may vary among individuals . This highlights the importance of considering different tissue types [16] . However , other explanations for the higher contribution of cis regulation in this study than in the ancestry-based analysis are also possible . For example , epistasis between two neighboring cis variants would be included in cis heritabilities estimated via IBD , but not in the ancestry-based analysis in which ancestry is a partial proxy for SNP genotype but a very poor proxy for both genotypes of two interacting SNPs . In addition , epistatic interactions involving multiple loci may potentially be important , and may confound estimates of narrow-sense heritability , but are outside the scope of this study . A further possibility is that trans effects in LCL could be overstated due to genetically heritable variation of in vitro factors such as the response to EBV virus , which would mimic trans regulation in heritability analyses but does not reflect true biological trans regulation [38] . Distinguishing between these possibilities is an important direction of future work . Efforts to understand cis regulation are likely to benefit from combining information from many cell or tissue types , since underlying mechanisms can be either shared or cell-type specific . Indeed , our finding that on average roughly half single-tissue cis heritability ( hcis2 ) is shared across tissues ( ξcis2 ) is consistent with a recent study focusing on cis eQTLs , which reported that 54% , 50% and 54% of cis eQTLs in fibroblasts , LCLs and T cells , respectively , are cell-type specific [36] . Those percentages would be expected to be higher when considering only two cell types , but lower at larger sample sizes . On the other hand , studies of trans regulation should focus on a single cell type to avoid diluting trans effects that are not shared across cell types . New technologies to assay cell type-specific gene expression in complex tissues may also prove valuable [39] . Future experiments will shed light on whether similarity between tissues other than blood and adipose is also predominantly explained by heritable cis effects . Results may vary by organism as well as tissue type . Recent studies of fat , kidney , adrenal and heart tissues in rat recombinant inbred strains also observed reduced trans effects in more heterogeneous tissues , but reported some evidence of cross-tissue regulation in trans as well as in cis [8]-[9] . The similarity of cis heritability results using IBD in closely related versus distantly related individuals has significant implications . It has been suggested that epigenetic inheritance , defined as the transmission across generations of epigenetic changes not due to variation in DNA sequence , is a potential source of the “missing heritability” in genetic association studies [17]-[21] . Epigenetic inheritance would be expected to influence expression at the cis locus , and would be expected to contribute to cis heritability between closely related individuals but not between distantly related individuals , given that this mode of inheritance persists over a relatively short time scale [22]-[23] . Our failure to observe any such discordance suggests that transgenerational epigenetic inheritance is unlikely to play a major role in the missing heritability of gene expression and other traits , although it does not rule out a very small aggregate effect across all genes or large effects at certain metastable epialleles [40]-[41] , nor does it shed light on the importance of mitotically conserved epigenetic effects that are not transmitted from parent to offspring . Our results highlight the utility of using IBD in distantly related individuals to make inferences about heritability . This approach will be particularly valuable as sample sizes increase , since the number of pairs of individuals increases quadratically with sample size . Indeed , IBD in distantly related individuals has already proven useful for mapping specific loci [42] , and heritability-related analyses using identity-by-state ( IBS ) instead of IBD have also yielded important insights [43]-[45] . By using IBD segments shorter than those analyzed here to consider IBD sharing at different distances from genes , it may even be possible draw conclusions about the distribution of genomic distances at which cis regulation contributes to heritability .
|
An important goal in biology is to understand how genotype affects gene expression . Because gene expression varies across tissues , the relationship between genotype and gene expression may be tissue-specific . In this study , we used heritability approaches to study the regulation of gene expression in two tissue types , blood and adipose tissue , as well as the regulation of gene expression that is shared across these tissues . Heritability can be partitioned into cis and trans effects by assessing identity-by-descent ( IBD ) at the genomic location close to the expressed gene or genome-wide , respectively , and applying variance-components methods to partition the heritability of each gene . We estimated the proportion of gene expression heritability explained by cis regulation as 37% in blood and 24% in adipose tissue . Notably , the heritability shared across tissue types was primarily due to cis regulation . Thus , the relative contribution of cis versus trans regulation is expected to increase with the number of cell types present in the tissue being assayed , just as observed in our study and in a comparison to previous work on lymphoblastoid cell lines ( LCL ) . We specifically ruled out a substantial contribution of transgenerational epigenetic inheritance to heritability of gene expression in these cohorts by repeating our heritability analyses using segments shared IBD in distantly related Icelanders .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"genetics",
"and",
"genomics/gene",
"expression",
"genetics",
"and",
"genomics/complex",
"traits",
"genetics",
"and",
"genomics/genetics",
"of",
"disease",
"genetics",
"and",
"genomics/epigenetics",
"genetics",
"and",
"genomics",
"genetics",
"and",
"genomics/population",
"genetics"
] |
2011
|
Single-Tissue and Cross-Tissue Heritability of Gene Expression Via Identity-by-Descent in Related or Unrelated Individuals
|
In Europe , the Neolithic transition ( 8 , 000–4 , 000 b . c . ) from hunting and gathering to agricultural communities was one of the most important demographic events since the initial peopling of Europe by anatomically modern humans in the Upper Paleolithic ( 40 , 000 b . c . ) . However , the nature and speed of this transition is a matter of continuing scientific debate in archaeology , anthropology , and human population genetics . To date , inferences about the genetic make up of past populations have mostly been drawn from studies of modern-day Eurasian populations , but increasingly ancient DNA studies offer a direct view of the genetic past . We genetically characterized a population of the earliest farming culture in Central Europe , the Linear Pottery Culture ( LBK; 5 , 500–4 , 900 calibrated b . c . ) and used comprehensive phylogeographic and population genetic analyses to locate its origins within the broader Eurasian region , and to trace potential dispersal routes into Europe . We cloned and sequenced the mitochondrial hypervariable segment I and designed two powerful SNP multiplex PCR systems to generate new mitochondrial and Y-chromosomal data from 21 individuals from a complete LBK graveyard at Derenburg Meerenstieg II in Germany . These results considerably extend the available genetic dataset for the LBK ( n = 42 ) and permit the first detailed genetic analysis of the earliest Neolithic culture in Central Europe ( 5 , 500–4 , 900 calibrated b . c . ) . We characterized the Neolithic mitochondrial DNA sequence diversity and geographical affinities of the early farmers using a large database of extant Western Eurasian populations ( n = 23 , 394 ) and a wide range of population genetic analyses including shared haplotype analyses , principal component analyses , multidimensional scaling , geographic mapping of genetic distances , and Bayesian Serial Simcoal analyses . The results reveal that the LBK population shared an affinity with the modern-day Near East and Anatolia , supporting a major genetic input from this area during the advent of farming in Europe . However , the LBK population also showed unique genetic features including a clearly distinct distribution of mitochondrial haplogroup frequencies , confirming that major demographic events continued to take place in Europe after the early Neolithic .
The transition from a hunter–gatherer existence to a “Neolithic lifestyle , ” which was characterized by increasing sedentarism and the domestication of animals and plants , has profoundly altered human societies around the world [1] , [2] . In Europe , archaeological and population genetic views of the spread of this event from the Near East have traditionally been divided into two contrasting positions . Most researchers have interpreted the Neolithic transition as a period of substantial demographic flux ( demic diffusion ) potentially involving large-scale expansions of farming populations from the Near East , which are expected to have left a detectable genetic footprint [3] , [4] . The alternative view ( cultural diffusion model; e . g . , [5] ) suggests that indigenous Mesolithic hunter–gatherer groups instead adopted new subsistence strategies with relatively little , or no , genetic influence from groups originating in the Near East . Genetic studies using mitochondrial DNA ( mtDNA ) and Y-chromosomal data from modern populations have generated contradictory results , and as a consequence , the extent of the Neolithic contribution to the gene pool of modern-day Europeans is still actively debated [6]–[8] . Studies that suggest that the genetic variation in modern-day Europe largely reflects farming communities of the Early Neolithic period [9]–[11] contrast strongly with others that consider the input from the Near East an event of minor importance and ascribe the European genetic variation and its distribution patterns to the initial peopling of Europe by anatomically modern humans in the Upper Paleolithic [12]–[15] . These patterns are also likely to have been significantly impacted by the early Holocene re-expansions of populations out of southerly refugia formed during the Last Glacial Maximum ( ∼25 , 000 y ago ) and by the numerous demographic events that have taken place in post-Neolithic Europe . The genetics of prehistoric populations in Europe remain poorly understood , restricting real-time insights into the process of the Neolithic transition [16]–[21] . As a result , most attempts to reconstruct history have been limited to extrapolation from allele frequencies and/or coalescent ages of mitochondrial and Y chromosome haplogroups ( hgs ) in modern populations . Ancient DNA ( aDNA ) analyses now provide a powerful new means to directly investigate the genetic patterns of the early Neolithic period , although contamination of specimens with modern DNA remains a major methodical challenge [22] . A previous genetic study of 24 individuals from the early Neolithic Linear Pottery Culture ( LBK; 5 , 500–4 , 900 calibrated b . c . [cal b . c . ] ) in Central Europe detected a high frequency of the currently rare mtDNA hg N1a , and proposed this as a characteristic genetic signature of the Early Neolithic farming population [19] . This idea was recently supported by the absence of this particular lineage ( and other now more common European hgs ) among sequences retrieved from neighboring Mesolithic populations [20] , [21] . However , a study of 11 individuals from a Middle/Late Neolithic site on the Iberian Peninsula ( 3 , 500–3 , 000 cal b . c . ) did not find significant differences from modern populations , supporting a quite different population genetic model for the Neolithic transition in Iberia [18] . To gain direct insight into the genetic structure of a population at the advent of farming in Central Europe we analyzed a complete graveyard from the Early Neolithic LBK site at Derenburg Meerenstieg II ( Harzkreis , Saxony-Anhalt ) in Germany . The archaeological culture of the LBK had its roots in the Transdanubian part of the Carpathian Basin in modern-day Hungary approximately 7 , 500–8 , 000 y ago and spread during the subsequent five centuries across a vast area ranging from the Paris Basin to the Ukraine [23] , [24] . The graveyard samples provide a unique view of a local , closed population and permit comparisons with other specimens of the LBK archaeological culture ( the contemporaneous meta-population ) and with modern populations from the same geographical area ( covering the former range of the LBK ) , as well as groups across the wider context of Western Eurasia . Our primary aim was to genetically characterize the LBK early farming population: by applying comprehensive phylogeographic and population genetic analyses we were able to locate its origins within the broader Eurasian region , and to trace its potential dispersal routes into Europe .
All of the mtDNA SNP typing results were concordant with the hg assignments based on HVS-I sequence information ( Tables 1 and S1 ) and the known phylogenetic framework for the SNPs determined from modern populations [25] . The tight hierarchical structure of the latter provides a powerful internal control for contamination or erroneous results . Overall , both multiplex systems proved to be extremely time- and cost-efficient compared to the standard approach of numerous individual PCRs , and required 22–25 times less aDNA template while simultaneously reducing the chances of contamination dramatically . Also , both multiplex assays proved to be a powerful tool for analyzing highly degraded aDNA , and the GenoCoRe22 assay was able to unambiguously type four additional specimens that had failed to amplify more than 100 bp ( Table 1 ) from two independent extractions . However , for reasons of overall data comparability , we could not include these specimens in downstream population genetic analyses , which required HVS-I sequence data . The only artifacts detected were occasional peaks in the electropherograms of the SNaPshot reactions outside the bin range of expected signals . These were probably due to primers and were mainly present in reactions from extracts with very little or no DNA template molecules; they were not observed with better preserved samples or modern controls . In contrast , Y chromosome SNPs could be typed for only three out of the eight male individuals ( 37 . 5%; Table S2 ) identified through physical anthropological examination , reflecting the much lower copy number of nuclear loci [22] . After typing with the GenoY25 assay , individual deb34 was found to belong to hg G ( M201 ) , whereas individuals deb20 and deb38 both fall basally on the F branch ( derived for M89 but ancestral for markers M201 , M170 , M304 , and M9 ) , i . e . , they could be either F or H ( Table 1 ) . To further investigate the hg status beyond the standard GenoY25 assay , we amplified short fragments around SNP sites M285 , P287 , and S126 to further resolve deb34 into G1 , G2* , and G2a3 , and around SNP site M69 to distinguish between F and H [26] . deb34 proved to be ancestral for G1-M285 but derived for G2*-P287 and additional downstream SNP S126 ( L30 ) , placing it into G2a3 . deb20 and deb38 were shown to be ancestral at M69 and hence basal F ( M89 ) , and remained in this position because we did not carry out further internal subtyping within the F clade . The multiplexed single base extension ( SBE ) approach with its shortened flanking regions around targeted SNPs significantly increases the chance of successful Y-chromosomal amplifications , which have remained problematic for aDNA studies , as have nuclear loci in general , because of the much lower cellular copy number compared to mitochondrial loci . The multiplexed SBE approach promises to open the way to studying the paternal history of past populations , which is of paramount importance in determining how the social organization of prehistoric societies impacted the population dynamics of the past . Results of the qPCR revealed significantly ( p = 0 . 012 , Wilcoxon signed-ranks test ) more mtDNA copies per microliter of each extract for the shorter fragment ( 141 bp ) than for the longer ( 179 bp ) , with an average 3 . 7×104–fold increase ( detailed results are shown in Table S3 ) . This finding is consistent with previous observations demonstrating a biased size distribution for authentic aDNA molecules [22] , [27] , [28] and suggests that any contaminating molecules , which would also result in higher copy numbers in the larger size class , did not significantly contribute to our amplifications . To analyze the Neolithic mtDNA sequence diversity and characterize modern geographical affinities , we applied a range of population genetic analyses including shared haplotype analyses , principal component analyses ( PCAs ) , multidimensional scaling ( MDS ) , geographic mapping of genetic distances , and demographic modeling via Bayesian Serial Simcoal ( BayeSSC ) analyses ( Table 2 ) . We prepared standardized modern population datasets of equal size ( n = ∼500 ) from 36 geographical regions in Eurasia ( n = 18 , 039; Table S4 ) to search for identical matches with each LBK haplotype . Out of 25 different haplotypes present in 42 LBK samples , 11 are found at high frequency in nearly all present-day populations under study , a further ten have limited geographic distribution , and the remaining four haplotypes are unique to Neolithic LBK populations ( Table S4 ) . The 11 widespread haplotypes are mainly basal ( i . e . , constituting a basal node within the corresponding hg ) for Western Eurasian mitochondrial hgs H , HV , V , K , T , and W . While these haplotypes are relatively uninformative for identifying genetic affiliations to extant populations , this finding is consistent within an ancient population ( 5 , 500–4 , 900 cal b . c . , i . e . , prior to recent population expansions ) , in which basal haplotypes might be expected to be more frequent than derived haplotypes ( e . g . , end tips of branches within hgs ) . The next ten LBK haplotypes were unequally spread among present-day populations and for this reason potentially contain information about geographical affinities . We found nine modern-day population pools in which the percentage of these haplotypes is significantly higher than in other population pools ( p>0 . 01 , two-tailed z test; Figure 1; Table S4 ) : ( a ) North and Central English , ( b ) Croatians and Slovenians , ( c ) Czechs and Slovaks , ( d ) Hungarians and Romanians , ( e ) Turkish , Kurds , and Armenians , ( f ) Iraqis , Syrians , Palestinians , and Cypriotes , ( g ) Caucasus ( Ossetians and Georgians ) , ( h ) Southern Russians , and ( i ) Iranians . Three of these pools ( b–d ) originate near the proposed geographic center of the earliest LBK in Central Europe and presumably represent a genetic legacy from the Neolithic . However , the other matching population pools are from Near East regions ( except [a] and [h] ) , which is consistent with this area representing the origin of the European Neolithic , an idea that is further supported by Iranians sharing the highest number of informative haplotypes with the LBK ( 7 . 2%; Table S4 ) . The remaining pool ( a ) from North and Central England shares an elevated frequency of mtDNA T2 haplotypes with the LBK , but otherwise appears inconsistent with the proposed origin of the Neolithic in the Near East . It has been shown that certain alleles ( here hgs ) can accumulate in frequency while surfing on the wave of expansion , eventually resulting in higher frequencies relative to the proposed origin [29] , [30] . Several of the other population pools also show a low but nonsignificant level of matches , which may relate to pre-Neolithic distributions or subsequent demographic movements ( Figure 1 ) . Of the four unique mtDNA haplotypes , two were from an earlier study of the LBK ( 16286-16304 and 16319-16343; Table S5 and [19] ) . The haplotype 16286-16304 has many one- or two-step derivates in all parts of Europe and is therefore rather uninformative for inferring further geographical affinities . The only relatively close neighbor of haplotype 16319-16343 is found in Iraq ( 16129-16189-16319-16343 ) , in agreement with the Near Eastern affinities of the informative LBK haplotypes . The other two unique LBK haplotypes belong to N1a , the characteristic LBK hg . The frequency of N1a was 13 . 6% for Derenburg samples ( 3/22 ) and 14 . 3% for all LBK samples published to date ( 6/42 ) . Notably , N1a has not yet been observed in the neighboring hunter–gatherer populations of Central Europe before , during , or after the Early Neolithic [20] nor in the early Neolithic Cardial Ware Culture from Spain [18] . The Y chromosome hgs obtained from the three Derenburg early Neolithic individuals are generally concordant with the mtDNA data ( Table 1 ) . Interestingly , we do not find the most common Y chromosome hgs in modern Europe ( e . g . , R1b , R1a , I , and E1b1 ) , which parallels the low frequency of the very common modern European mtDNA hg H ( now at 20%–50% across Western Eurasia ) in the Neolithic samples . Also , while both Neolithic Y chromosome hgs G2a3 and F* are rather rare in modern-day Europe , they have slightly higher frequencies in populations of the Near East , and the highest frequency of hg G2a is seen in the Caucasus today [15] . The few published ancient Y chromosome results from Central Europe come from late Neolithic sites and were exclusively hg R1a [31] . While speculative , we suggest this supports the idea that R1a may have spread with late Neolithic cultures from the east [31] . Four Neolithic datasets were constructed ( Table 2 ) and compared with 55 present-day European and Near Eastern populations and one Mesolithic hunter–gatherer population [20] in a PCA ( Figure 2 ) . The PCA accounted for 39% of the total genetic variation , with the first principal component ( PC ) separating Near Eastern populations from Europeans ( 24 . 9% ) , and with LBK populations falling closer to Near Eastern ones . However , the second PC ( 17 . 4% ) clearly distinguished the four Neolithic datasets from both Near East and European populations . An MDS plot ( Figure S1 ) showed similar results , with the Near Eastern affinities of the LBK populations even more apparent . To better understand which particular hgs made the Neolithic populations appear either Near Eastern or ( West ) European , we compared average hg frequencies of the total LBK ( LBK42 ) and Derenburg ( DEB22 ) datasets to two geographically pooled meta-population sets from Europe and the Near East ( Tables 2 and S6; 41 and 14 populations , respectively ) . PC correlates and component loadings ( Figure 2 ) showed a pattern similar to average hg frequencies ( Table 2 ) in both large meta-population sets , with the LBK dataset grouping with Europeans because of a lack of mitochondrial African hgs ( L and M1 ) and preHV , and elevated frequencies of hg V . In contrast , low frequencies of hg H and higher frequencies for HV , J , and U3 promoted Near Eastern resemblances . Removal of individuals with shared haplotypes within the Derenburg dataset ( yielding dataset LBK34 ) did not noticeably decrease the elevated frequencies of J and especially HV in the Neolithic data . Most importantly , PC correlates of the second component showed that elevated or high frequencies of hgs T , N1a , K , and W were unique to LBK populations , making them appear different from both Europe and Near East . The considerable within-hg diversity of all four of these hgs ( especially T and N1a; Table 1 ) suggests that this observation is unlikely to be an artifact of random genetic drift leading to elevated frequencies in small , isolated populations . The pooled European and Near Eastern meta-populations are necessarily overgeneralizations , and there are likely to be subsets of Near Eastern populations that are more similar to the Neolithic population . Interestingly , both the PCA and the MDS plots identified Georgians , Ossetians , and Armenians as candidate populations ( Figures 2 and S1 ) . We generated genetic distance maps to visualize the similarity/distance of the LBK and Derenburg populations ( datasets LBK42 and DEB22 ) to all modern populations in the large Western Eurasian dataset ( Figure 3 ) . In agreement with the PCA and MDS analyses , populations from the area bounding modern-day Turkey , Armenia , Iraq , and Iran demonstrated a clear genetic similarity with the LBK population ( Figure 3A ) . This relationship was even stronger in a second map generated with just the Neolithic Derenburg individuals ( Figure 3B ) . Interestingly , the map of the combined LBK data also suggested a possible geographic route for the dispersal of Neolithic lineages into Central Europe: genetic distances gradually increase from eastern Anatolia westward across the Balkans , and then northwards into Central Europe . The area with lower genetic distances follows the course of the rivers Danube and Dniester , and this natural corridor has been widely accepted as the most likely inland route towards the Carpathian basin as well as the fertile Loess plains further northwest [23] , [32] , [33] . While an apparent affinity of Neolithic farmers to modern-day Near East populations is revealed by the shared haplotype analyses , PCA , MDS , and genetic distance maps , the population-specific pairwise FST values among ancient populations ( hunter–gatherers and LBK ) and the modern population pools ( Central Europe and Near East ) tested were all significant ( p>0 . 05; Table 3 ) , suggesting a degree of genetic discontinuity between ancient and modern-day populations . The early farmers were closer to the modern Near Eastern pool ( FST = 0 . 03019 ) than hunter–gatherers were ( FST = 0 . 04192 ) , while both ancient populations showed similar differences to modern Central Europe , with the hunter–gatherers slightly closer ( FST = 0 . 03445 ) than the early farmers ( FST = 0 . 03958 ) . The most striking difference was seen between Mesolithic hunter–gatherers and the LBK population itself ( FST = 0 . 09298 ) , as previously shown [20] . We used BayeSSC analyses to test whether the observed FST values can be explained by the effects of drift or migration under different demographic scenarios ( Figure S2 ) . This encompassed comparing FST values derived from coalescent simulations under a series of demographic models with the observed FST values in order to test which model was the most likely , given the data . By using an approximate Bayesian computation ( ABC ) framework we were able to explore priors for initial starting deme sizes and dependent growth rates to maximize the credibility of the final results . The Akaike information criterion ( AIC ) was used to evaluate a goodness-of-fit value of the range of models in the light of the observed FST values . In addition , a relative likelihood estimate for each of the six models given the data was calculated via Akaike weights ( ω ) . The highest AIC values , and therefore the poorest fit , were obtained for models representing population continuity in one large Eurasian meta-population through time ( Models H0a and H0b; Table 4 ) . Of note , the goodness of fit was better with a more recent population expansion ( modeled at the onset of the Neolithic in Central Europe ) and hence higher exponential growth rate ( H0a ) . The model of cultural transmission ( H1 ) , in which a Central European deme including Neolithic farmers and hunter–gatherers coalesced with a Near Eastern deme in the Early Upper Paleolithic ( 1 , 500 generations , or ∼37 , 500 y ago ) , resulted in intermediate goodness-of-fit values ( H1a and H1b; Table 4; Figure S2 ) . The best goodness-of-fit values were retrieved for models of demic diffusion ( model H2; Table 4 ) with differing proportions of migrants ( 25% , 50% , and 75% were tested ) from the Near Eastern deme into the Central European deme around the time of the LBK ( 290 generations , ∼7 , 250 y ago; Table 4 ) . Notably , the models testing 50% and 75% migrants returned the highest relative likelihood values ( 42% and 52% , respectively ) , and therefore warrant further investigation . However , while the demic diffusion model H2 produced values that approximated the observed FST between Neolithic farmers and the Near Eastern population pool , none of the models could account for the high FST between hunter–gatherers and early farmers or early farmers and modern-day Central Europeans . The models we tested represent major oversimplifications and it should be noted that modeling human demographic history is notoriously difficult , especially given the complex history of Europe and the Near East over this time scale . The fact that no model explained the observed FST between ancient and modern-day populations particularly well suggests that the correct scenario has not yet been identified , and that there is also an obvious need for sampling of material from younger epochs . Additionally , sampling bias remains an issue in aDNA studies , and this is particularly true for the chronologically and geographically diverse hunter–gatherer dataset . In the light of the models tested ( see also [19] , [20] ) , we would suggest that the basis of modern European mtDNA diversity was formed from the postglacial re-peopling of Europe ( represented here by the Mesolithic hunter–gatherers ) and the genetic input from the Near East during the Neolithic , but that demographic processes after the early Neolithic have contributed substantially to shaping Europe's contemporary genetic make up . The aDNA data from a range of Mesolithic hunter–gatherer samples from regions neighboring the LBK area have been shown to be surprisingly homogenous across space and time , with an mtDNA composition almost exclusively of hg U ( ∼80% ) , particularly hg U4 and U5 , which is clearly different from the LBK dataset as well as the modern European diversity ( Table 2 ) [20] . The observation that hgs U4 and U5 are virtually absent in the LBK population ( 1/42 samples ) is striking ( Table 2 ) . Given this clear difference in the mtDNA hg composition , it is not surprising that the pairwise FST between hunter–gatherers and the LBK population is the highest observed ( 0 . 09298 ) when we compared ancient populations with representative population pools from Central Europe and the Near East ( Table 3; see also [20] ) . If the Mesolithic data are a genuine proxy for populations in Central Europe at the onset of the LBK , it implies that the Mesolithic and LBK groups had clearly different origins , with the former potentially representing the pre-Neolithic indigenous groups who survived the Last Glacial Maximum in southern European refugia . In contrast , our population genetic analyses confirm that the LBK shares an affinity with modern-day Near East and Anatolia populations . Furthermore , the large number of basal lineages within the LBK , a reasonably high hg and haplotype diversity generated through one- or two-step derivative lineages , and the negative Tajima's D values ( Tables 1 and 2 ) indicate a recent expansion . These combined data are compatible with a model of Central Europe in the early Neolithic of indigenous populations plus significant inputs from expanding populations in the Near East [4] , [12] , [34] . Overall , the mtDNA hg composition of the LBK would suggest that the input of Neolithic farming cultures ( LBK ) to modern European genetic variation was much higher than that of Mesolithic populations , although it is important to note that the unique characteristics of the LBK sample imply that further significant genetic changes took place in Europe after the early Neolithic . aDNA data offers a powerful new means to test evolutionary models and assumptions . The European lineage with the oldest coalescent age , U5 , has indeed been found to prevail in the indigenous hunter–gatherers [12] , [35] . However , mtDNA hgs J2a1a and T1 , which because of their younger coalescence ages have been suggested to be Neolithic immigrant lineages [8] , [12] , are so far absent from the samples of early farmers in Central Europe . Similarly , older coalescence ages were used to support hgs K , T2 , H , and V as “postglacial/Mesolithic lineages , ” and yet these have been revealed to be common only in Neolithic samples . The recent use of whole mitochondrial genomes and the refinement of mutation rate estimates have resulted in a general reduction in coalescence ages [8] , which would lead to an improved fit with the aDNA data . However we advise caution in directly relating coalescence ages of specific hgs to evolutionary or prehistoric demographic events [36] . Significant temporal offsets can be caused by either observational bias ( the delay between the actual split of a lineage and the eventual fixation and dissemination of this lineage ) or calculation bias ( incorrect coalescent age estimation ) . aDNA has considerable value not only for directly analyzing the presence or absence of lineages at points in the past but also for refining mutation rate estimates by providing internal calibration points [37] . Archaeological and anthropological research has produced a variety of models for the dispersal of the Neolithic agricultural system ( “process of Neolithization” ) into and throughout Europe ( e . g . , [1] , [2] , [38] ) . Our findings are consistent with models that argue that the cultural connection of the LBK to its proposed origin in modern-day Hungary , and reaching beyond the Carpathian Basin [23] , [32] , [38] , [39] , should also be reflected in a genetic relationship ( e . g . , shared haplotype analyses; Table S4 ) . Therefore at a large scale , a demic diffusion model of genetic input from the Near East into Central Europe is the best match for our observations . It is notable that recent anthropological research has come to similar conclusions [40] , [41] . On a regional scale , “leap-frog” or “individual pioneer” colonization models , where early farmers initially target the economically favorable Loess plains in Central Europe [33] , [42] , would explain both the relative speed of the LBK expansion and the clear genetic Near Eastern connections still seen in these pioneer settlements , although the resolving power of the genetic data is currently unable to test the subtleties of these models . In conclusion , the new LBK dataset provides the most detailed and direct genetic portrait of the Neolithic transition in Central Europe; analysis of this dataset reveals a clear demonstration of Near Eastern and Anatolian affinities and argues for a much higher genetic input from these regions , while also identifying characteristic differences from all extant ( meta- ) populations studied . Ancient genetic data from adjacent geographic regions and time periods , and especially from the Near East and Anatolia , will be needed to more accurately describe the changing genetic landscape during and after the Neolithic , and the new multiplexed SBE assays offer a powerful means to access this information .
The archaeological site Derenburg Meerenstieg II ( Harzkreis , Saxony-Anhalt , Germany ) was excavated during three campaigns in 1997–1999 comprising an area of 3 ha . The archaeological context at this site shows a record of settlement activity ranging from the Early Neolithic ( LBK ) and Middle Neolithic ( Rössen and Ammensleben cultures ) to Bronze and Iron Age [43] . However , the main features of Derenburg are the LBK graveyard and its associated partial settlement approximately 70 m southwest . The archaeological data revealed that the larger part of the settlement has not yet been excavated and lies outside the area covered during these campaigns . In contrast , the graveyard was recorded in its entire dimension ( 25×30 m ) and encompassed a total of 41 graves . Two separate graves were found outside the graveyard ( 50 m WSW and 95 m SSE ) . Erosion and modern agricultural ploughing might have led to a loss of some graves at the plateau area . Here , the graves were shallow and in average state of preservation , whereas the graves embedded in deeper Loess layers showed an excellent state of preservation . In total , 32 single grave burials were found; there were also one double burial , one triple burial , two burials in settlement pits , two or three times additional singular bones in a grave , three burials with a secondary inhumation , and one empty grave . The majority of individuals ( 75% ) at Derenburg were buried in East–West orientation in a varying flexed position . The duration of usage of the graveyard spans over the entire time frame of the LBK and is reflected by the typology of the ceramics and associated grave goods ranging from older LBK pottery ( Flomborn style ) to youngest LBK pottery . Absolute radiocarbon dates confirm the usage over three centuries ( 5 , 200–4 , 900 cal b . c . ; see also Table 1 and [44] ) . From an initial 43 graves in the Derenburg graveyard , 31 indicated morphological preservation suitable for sampling and aDNA analyses . Five individuals had already been sampled in 2003 for our previous study and showed excellent preservation of aDNA , a negligible level of contamination , and an unusual mtDNA hg distribution , thereby justifying further investigation [19] . Hence , 26 additional individuals were processed in this study ( Table 1 ) . We amplified , cloned , and sequenced mitochondrial HVS-I ( nucleotide positions [np] 15997–16409; nucleotide position according to [45] ) as described previously [19] . mtDNA hg assignments were further supported by typing with a newly developed multiplex of 22 mtDNA coding region SNPs ( GenoCoRe22 ) . In addition , we typed 25 Y chromosome SNPs using a second novel multiplex assay ( GenoY25 ) . Final refinement of Y chromosome hg assignments was performed via singleplex PCRs . Lastly , the amount of starting DNA template molecules was monitored using qPCR on seven random samples ( Table S3 ) . aDNA work was performed in specialized aDNA facilities at the Johannes Gutenberg University of Mainz and the Australian Centre for Ancient DNA ( ACAD ) at the University of Adelaide according to appropriate criteria . All DNA extractions as well as amplification , cloning , and sequencing of the mitochondrial control region HVS-I were carried out in the Johannes Gutenberg University of Mainz facilities . Additional singleplex , all multiplex , and quantitative real-time amplifications , SNP typing , and direct sequencing of Y chromosome SNPs were carried at the ACAD as described below . The technique of SNP typing via SBE reactions ( also known as minisequencing ) has proven a reliable and robust method for high throughput analyses of polymorphisms , e . g . , human mitochondrial variation [46] , human X- and Y-chromosomal SNPs [47] , [48] , and human autosomal SNPs [49] . However , few SBE studies have addressed the special need for very short amplicon sizes to allow amplification from highly degraded DNA , as even forensic protocols have generally targeted relatively long amplicon sizes [50]–[54] . Our first multiplex ( GenoCoRe22 ) was designed to type a panel of 22 mitochondrial coding region SNPs that are routinely typed within the Genographic Project [25] , to allow for future maximum comparability with modern population data . A second multiplex ( GenoY25 ) targeted a basal , but global , coverage of 25 commonly typed Y chromosome SNPs , for maximum comparability of paternal lineages . The aim of the SNP assay design was to produce highly efficient and sensitive protocols , capable of working on highly degraded DNA , that also allow modern human DNA contamination to be detected at very low levels and monitored [51] . The GenoCoRe22 SNP panel was chosen to cover the basal branches of mitochondrial hgs across modern human mtDNA diversity [25] . The chosen SNP sites were identical to the initial set ( Figure 4 in [25] ) except for hg W ( SNP at np 8994 instead of np 1243 ) and hg R9 ( SNP at np 13928 instead of np 3970 ) , as a compromise arising from primer design within a multiplex assay . Selection of GenoY25 SNP panel for incorporation into the multiplex assay was performed using the highly resolved Y Chromosome Consortium tree and an extensive literature search for corresponding SNP allele frequencies in European populations [13] , [26] , [55] . Multiplex assays were set up , established , and performed at the ACAD facilities . Multiplex PCR using Amplitaq Gold ( Applied Biosystems ) was conducted in 25-µl volumes using 1× Buffer Gold , 6 mM ( GenoCoRe22 ) or 8 mM ( GenoY25 ) MgCl2 , 0 . 5 mM dNTPs ( Invitrogen ) , ≤0 . 2 µM of each primer , 1 mg/ml RSA ( Sigma ) , 2 U of Amplitaq Gold Polymerase , and 2 µl of DNA extract . Thermocycling conditions consisted of an initial enzyme activation at 95°C for 6 min , followed by 40–45 cycles of denaturation at 95°C for 30 s , annealing at 60°C ( GenoCoRe22 ) or 59°C ( GenoY25 ) for 30 s , and elongation at 65°C for 30 s , with a single final extension time at 65°C for 10 min . Each PCR included extraction blanks as well as a minimum of two PCR negatives at a ratio of 5∶1 . PCRs were visually checked by electrophoresis on 3 . 5% agarose TBE gels . PCR products were purified by mixing 5 µl of PCR product with 1 U of SAP and 0 . 8 U of ExoI and incubating at 37°C for 40 min , followed by heat inactivation at 80°C for 10 min . Because of the sensitivity of the multiplex PCR ( using fragment lengths of only 60–85 bp ) , and to be able to monitor potential human background contamination , usually all controls were included in downstream fragment analysis . Multiplex primer sequences and concentration are given in Table S7 . SBE reactions were carried out on the GenoCoRe22 and GenoY25 SNP multiplex assay using the ABI Prism SNaPshot multiplex reaction kit ( Applied Biosystems ) following the manufacturer's instructions , except that 10% 3 M ammonium sulfate was added to the extension primer mix to minimize artifacts [56] . SBE primers and concentrations are given in Table S7 . Cycling conditions consisted of 35 cycles of denaturation at 96°C for 10 s , annealing at 55°C for 5 s , and extension at 60°C for 30 s . SBE reactions were purified using 1 U of SAP , incubating at 37°C for 40 min , followed by heat inactivation at 80°C for 10 min . Prior to capillary electrophoresis , 2 µl of purified SNaPshot product was added to a mix of 11 . 5 µl of Hi-Di Formamide ( Applied Biosystems ) and 0 . 5 µl of Gene-Scan-120 LIZ size standard ( Applied Biosystems ) . Samples were run on an ABI PRISM 3130xl Genetic Analyzer ( Applied Biosystems ) after a denaturation carried out according to the manufacturer's instructions using POP-6 ( Applied Biosystems ) . Evaluation and analyses of SNaPshot typing profiles were performed using custom settings within the GeneMapper version 3 . 2 Software ( Applied Biosystems ) . Additional Y chromosome SNPs ( M285 , P287 S126 , and M69 ) were tested to determine specific downstream subclades based on the initial multiplex results in order to gain further resolution . We chose appropriate SNP loci by following general criteria , trying to keep the PCR amplicon size smaller than 90 bp in size and flanking DNA sequences free from interfering polymorphisms , such as nucleotide substitutions in potential primer binding sites . We selected PCR amplification primers that have a theoretical melting temperature of around 60°C in neutral buffered solutions ( pH 7–8 ) , with monovalent cation ( Na+ ) concentrations at 50 mM and divalent cation ( Mg++ ) concentrations at 8 mM . All primer candidates were analyzed for primer–dimer formation , hairpin structures , and complementarities to other primers in the multiplex using Primer 3 ( http://primer3 . sourceforge . net/ ) . Primer characteristics were chosen to ensure equal PCR amplification efficiency for all DNA fragments , as previously described [50] . The primers were HPLC-purified and checked for homogeneity by MALDI-TOF ( Thermo ) . Table S7 shows the sequences and the concentrations of the amplification primers in the final multiplex PCR . Additional Y chromosome SNP singleplex PCRs were carried out in the ACAD facilities . Standard PCRs using Amplitaq Gold ( Applied Biosystems ) were conducted in 25-µl volumes using 1× Buffer Gold , 2 . 5 mM MgCl2 , 0 . 25 mM of each dNTP ( Fermentas ) , 400 µM of each primer ( Table S7 ) , 1 mg/ml RSA ( Sigma-Aldrich ) , 2 U of Amplitaq Gold Polymerase , and 2 µl of DNA extract . Thermocycling conditions consisted of an initial enzyme activation at 95°C for 6 min , followed by 50 cycles of denaturation at 94°C for 30 s , annealing at 59°C for 30 s , and elongation at 72°C for 30 s , with a single final extension time at 60°C for 10 min . Each PCR reaction included extraction blanks as well as a minimum of two PCR negatives . PCR products were visualized and purified as described above and were directly sequenced in both directions using the Big Dye Terminator 3 . 1 Kit ( Applied Biosystems ) as per manufacturer's instructions . Sequencing products were purified using Cleanseq magnetic beads ( Agencourt , Beckman Coulter ) according to the manufacturer's protocol . Sequencing products were separated on a 3130xl Genetic Analyzer ( Applied Biosystems ) , and the resulting sequences were edited and aligned relative to the SNP reference sequence ( GenBank SNP accession numbers: M285 , rs13447378; P287 , rs4116820; S126 [also known as L30] , rs34134567; and M69 , rs2032673 ) using the software Sequencher 4 . 7 ( Genecodes ) . qPCR was used to determine the amount of DNA in the samples prior to amplification and to assess the authenticity based on the assumption that there is an inverse relationship between DNA quantity and fragment length for degraded aDNA [57] , [58] . Two different length fragments were amplified from the HVS-I: 141 bp ( L16117/H16218 ) and 179 bp ( L16209/H16348 ) [19] , [59] . All qPCR reactions were carried out in a 10-µl reaction volume containing 1× Express SYBR Green ER Supermix Universal ( Invitrogen ) , rabbit serum albumin ( 10 mg/ml ) , forward and reverse primers ( 10 µM ) , and 1 µl of DNA extract . Thermocycling conditions consisted of an initial enzyme activation at 95°C for 5 min , followed by 50 cycles of 94°C for 10 s , 58°C for 20 s , and 72°C for 15 s . The primer specificity was assessed using a post-PCR melt curve to visualize the dissociation kinetics . The primers were validated using modern DNA , and a single peak was observed for both fragments , indicating specific binding . The dissociation temperature ( TM ) was 80–80 . 3°C for the 141-bp fragment and 81 . 7–82 . 3°C for the 179-bp fragment . Both primer pairs showed an absence of primer dimers , indicated by the lack of a smaller peak on the melt curve ( ≈60°C ) and a single band on a 2% agarose gel . The starting quantity of DNA in the ancient samples was determined by comparison to a standard curve of a known amount of DNA . The standard curves for the two fragments were created from modern human DNA . The DNA was extracted from a buccal cheek swab of a single individual using DNeasy Blood and Tissue Kit ( Qiagen ) . mtDNA was amplified for the two fragments ( 141 bp and 179 bp ) using 1× Hotmaster Buffer ( Eppendorf ) , 0 . 5 U of Hotmaster Taq ( 5Prime ) , forward and reverse primers ( 10 µM ) , distilled water , and 2 µl of DNA extract . Thermocycling conditions consisted of an initial enzyme activation at 94°C for 2 min , followed by 30 cycles of 94°C for 20 s , 60°C for 10 s , and 65°C for 1 min . The PCR products were purified using Agencourt Ampure ( Beckman Coulter ) according to manufacturer's instructions . The DNA concentration for the 141-bp and 179-bp amplicons was measured twice at 1∶1 and 1∶10 dilutions with a Nanovue ( GE Healthcare ) . Ten-fold serial dilutions , from 1×106 to 10 copies/µl , of the purified fragments were used to make the standards . These were run with the qPCR conditions described above . For each standard , each 10-fold dilution was run in triplicate and the qPCR was repeated on a separate day . All the standards met the following criteria: ( 1 ) there was a linear regression relationship between DNA quantity and cycle threshold ( fluorescence above background ) , R2>0 . 95 , and ( 2 ) the reaction was efficient ( i . e . , a doubling of product per cycle in the exponential phase ) , between 90% and 110% . Ancient qPCRs were run in triplicate with extraction and PCR blanks , and PCR standards ( positive control ) run in duplicate . Amplifications were performed on Rotor-Gene 6000 and analyses on Rotor-Gene 6000 Series Software 1 . 7 ( Corbett ) . The difference in mtDNA quantity between fragment lengths ( 141 and 179 bp ) was assessed using a nonparametric version of a Student's t test , a Wilcoxon signed-ranks test . This test was selected because the data were not appropriate for a parametric test , displaying a mixture of normal ( 179 bp , p = 0 . 425 ) and non-normal ( 141 bp , p = 0 . 012 ) distributions , as determined from a Shapiro-Wilk W test , which is appropriate for testing the normality of groups with small sample sizes . In line with previous publications on aDNA and especially with criteria for working with human aDNA , it can be stated that a 100% authentication of ancient samples is virtually impossible [22] , [57] , [60] . However , we took all possible precautions to prevent modern contaminations , and we regard the results as authentically derived from endogenous DNA based on the following chain of evidence . ( 1 ) All samples were collected under DNA-free conditions after excavation . Samples were not washed , treated , or examined before taking DNA samples . ( 2 ) All preparation and analytical steps prior to DNA amplification were conducted in a clean room area solely dedicated to aDNA work located in a physically separated building without any modern DNA work ( pre-PCR area ) . Amplification , cloning , and sequencing were carried out in the post-PCR lab . ( 3 ) All steps were monitored by non-template controls and by using bovid samples in parallel . ( 4 ) All individuals were sampled twice from anatomically independent regions and treated independently . At least eight independent PCR reactions were carried out ( four overlapping fragments × two extractions ) per individual . In case of successful amplification of all eight fragments , these were cloned and an average of eight clones per amplicons was sequenced to detect heterogeneous sequences due to DNA degradation or contamination . All replicable polymorphic sites were consistent with existing mtDNA haplotypes , ruling out post mortem DNA damage as a potential source for erroneous sequences . ( 5 ) The new multiplexes not only clearly confirm hg assignment but also provide an ideal monitoring system for ancient human DNA samples , as they directly target SNPs defining all potential contaminating lineages . ( 6 ) qPCR was carried out on a selection of samples to ensure appropriate levels of DNA quantity and to assess DNA quality . ( 7 ) Samples were collected and processed by W . H . exclusively ( mtDNA hg H1 , np 15997–16409: 16189C 16311C , and Y chromosome hg E1b1b1a-M78 ) after excavation; no other staff were involved in any of the pre-PCR steps . Eventually , all listed criteria indicating authenticity or at least the plausibility of having retrieved endogenous DNA were evaluated , together with the sample's post-excavation history [60] . Four partly overlapping Neolithic datasets were analyzed: the 22 Derenburg individuals ( DEB22 ) ; 20 individuals from other LBK populations previously published ( LBK20; Table S5 and [19] ) ; the combined LBK dataset ( LBK42 ) ; and the combined LBK dataset excluding eight individuals of possible kinship ( LBK34 , see below ) to avoid overestimation of haplotype frequencies . These four Neolithic sets were analyzed against extant population data from the MURKA mitochondrial DNA database and integrated software , currently containing 97 , 523 HVS-I records from published sources , and maintained by coauthors V . Z . , E . B . , and O . B . of the Russian Academy of Medical Sciences . Analyses were restricted to 390 populations from Europe and the Near East ( 35 , 757 mtDNAs ) . For detailed analysis of shared haplotypes , we included only sequences spanning from np 16069 to np 16365 ( 34 , 258 samples , haplotype dataset ) . aDNA sequences were trimmed to the same length . For frequency-based analyses ( PCA , MDS , and genetic distance maps ) , we omitted mtDNAs whose hg affiliations were ambiguous ( absence of information on coding region SNPs ) , resulting in our final hg frequency dataset of 23 , 394 individuals from 228 population studies , which subsequently were pooled into 55 populations based on ethnicity , language , and/or geographical criteria as described in the original publications ( see Table S6 ) . The mtDNA and Y chromosome hg results were overlaid onto the map of the graveyard to elucidate the spatial relationships within the graveyard ( Figure S3 ) . Four haplotypes were shared by two individuals each , and two haplotypes by three individuals each , while the remaining eight individuals ( 36 . 4% ) showed unique haplotypes within the Derenburg graveyard . A number of shared haplotypes is not surprising in a medium sized , closed LBK graveyard where the influence of genetic drift and a certain level of biological kinship are likely . However , little positional structuring according to maternal lineages was observed . A clustering of mtDNA haplotypes H-rCRS ( deb9 and deb21 ) and HV ( deb4 , deb20 , and deb5 ) in the northwest corner of the cemetery is notable , whereas other shared haplotype “twins” or “trios” with a potential maternal relationship are spread across larger distances . However , it must be stated that there are many other factors influencing the layout of interments in a graveyard that cannot be unraveled by aDNA analyses . LBK burials commonly show a great variety of mortuary patterns or rites at the same site ( e . g . , burials within the settlement and burials in pits/middens ) , and it is therefore not clear whether individuals in the cemetery represent the norm or the exception , and how much of the initial genetic variation of the population is missing [44] . In any case , to avoid overestimation of haplotype frequencies in the LBK dataset , the eight duplicate haplotypes were excluded , and a reduced dataset ( LBK34 ) was used in population genetic analyses alongside the complete set to account for a potential kinship effect . Haplotype diversity ( h ) and Tajima's D were calculated using DnaSP version 5 [61] . In order to calculate the percentage of shared haplotypes between the LBK sample and modern-day populations , we chose modern populations of equal or larger sample sizes , resulting in 36 out of 55 pooled populations with sample size n = 500 or above . Pooling was based on geographic proximity and linguistic similarity . For population studies with n>500 , 500 samples were selected randomly . After pooling and random selection the dataset comprised 18 , 039 samples . A pivot table was created ( 4 , 140 haplotypes in rows and 36 populations in columns ) , and Neolithic LBK data were included . Similarity between LBK and other populations was described quantitatively in two ways: ( 1 ) indicating presence or absence ( 1/0 ) , i . e . , whether or not the particular Neolithic haplotype was found in a given modern population , and ( 2 ) indicating the number of hits , i . e . , how many times the particular haplotype was found in a given population . The 25 different LBK sequence haplotypes were sorted into clusters of noninformative ( 11 ) , informative ( 10 ) , and unique ( 4 ) haplotypes ( Table S4 ) . We then calculated the relative frequency of each of the shared informative vs . noninformative LBK sequence haplotypes in each of the 36 modern-day populations ( Table S4 ) . A two-tailed z test ( Excel version 12 . 1 , Microsoft Office ) was applied to determine which population pool showed a significantly higher or lower percentage of shared informative haplotypes ( Table S4 ) . Nonparametric bootstrapping of 100 replicates for each hg per population was used to generate the confidence intervals for the percentage of hgs that are shared between all matches , informative matches , and noninformative matches . Bootstrapping was performed in Excel . version 12 . 1 . Classical and categorical PCAs and MDS were performed using the hg frequencies dataset . To avoid overpopulating graphs with 228 populations , populations were pooled into 55 groups defined by ethnicity , language , and/or geography as described in the original publications ( see Table S6 ) . To minimize statistical noise caused by very rare hgs , we considered only the following 19 hgs with average frequency above 1% in Europe and Near East: preHV , H , HV , J , T , I , N1a , K , V , W , X , U2 , U3 , U4 , U5a , U5b , the group of African hgs ( L and M1 ) , the group of East Eurasian hgs ( A , B , C , D , F , G , and Z ) , and the group of all other ( rare ) hgs . PCAs and categorical PCAs ( used for the biplot graph in Figure 1 , with default settings to correspond to a classical PCA ) were performed and visualized using the software package SPSS Statistics 17 . 0 . Nei's genetic distances [62] were calculated using the software program DJ , written by Yuri Seryogin ( freely available at http://www . genofond . ru ) . The resulting distance matrix was visualized via MDS in SPSS Statistics 17 . 0 . The genetic distances from two Neolithic datasets ( DEB22 and LBK42 ) to populations in the hg frequencies dataset ( pooled into 120 populations with the average sample size n = 196 to gain a balanced geographical coverage ) were calculated using the software DJ . Distances were plotted on a geographic map of Europe using the software GeneGeo written by S . K . This software is the renewed GGMAG package previously used for gene geographical studies ( [63] and references therein ) . We calculated population-specific pairwise genetic distances ( FST ) in Arlequin version 3 . 5 [64] , using 377-bp HVS-I sequences ( np 16069–16365 ) assigned to one of four populations ( Table S6 ) : modern Central Europeans from the LBK core area ( n = 1 , 030 ) , modern Near Easterners ( n = 737 ) , LBK samples ( n = 42 ) , and hunter–gatherers ( n = 20 ) . FST values were estimated using the Kimura two-parameter model [65] using a gamma distribution with shape parameter of 0 . 205 [66] . To test whether drift can account for the high FST values between ancient and contemporary populations from Central Europe and the Near East we modeled three alternative population histories ( Figure S2 ) using simulated coalescent analyses in the program BayeSSC [67] , [68] . Under the null hypothesis ( H0 ) we considered one large continuous Eurasian population with an effective population size ranging from 100 , 000 to 30 million and an exponential growth starting from a small Palaeolithic deme of 5 , 000 females , 300 ( H0a ) or 1 , 500 ( H0b ) generations ago . Hypothesis 1 ( H1 ) assumes two exponentially growing populations , a Central European deme ( 100 , 000 to 12 million ) and a Near Eastern deme ( 100 , 000 to 12 million ) , which coalesce 1 , 500 generations ago ( 37 , 500 y ago , assuming 25 y per generation ) in an Early Upper Palaeolithic deme of 5 , 000 females and constant size . Here , ancient samples from hunter–gatherers and Neolithic farmers were included in the Central European deme; therefore , this model can be considered a test for genetic continuity of Central European lineages under a scenario of cultural diffusion/transmission . Alternatively , we modeled a contrasting ( “demic diffusion” ) scenario ( H2 ) , similar to H1 in structure but allowing for migration from the Near Eastern deme 290 generations ago . We tested a contribution of 25% , 50% , and 75% migrants from the Near Eastern to the Central European deme . Each model was simulated initially using BayeSSC for 100 , 000 genealogies and a fixed mutation rate of 7 . 5×10−6 per site per generation [66] . A uniform distribution was used for priors to estimate effective population sizes at time 0 for the Central European and Near Eastern demes ( Table 4 ) . To compare the simulated and observed data , five pairwise FST values were chosen that reflect population differentiation between each of the two ancient samples and modern populations ( Table 3 ) . The simulated and observed FST values were compared within an ABC framework [69] , in which the top 1% of simulations were retained . Posterior distributions for each of the parameters with a prior were assessed . ABC was performed in R version 2 . 11 . 0 using scripts freely available at http://www . stanford . edu/group/hadlylab/ssc/index . html . To compare the goodness of fit of each model using AIC [70] given the observed data , priors were removed from the model and replaced with absolute parameter values that gave the maximum likelihood . The model was rerun in BayeSSC for 1 , 000 genealogies . The AIC for each model was calculated in R , and Akaike weights ω to compare the relative likelihood of each model where calculated in Excel version 12 . 1 [71] , [72] .
|
The transition from a hunter–gatherer existence to a sedentary farming-based lifestyle has had key consequences for human groups around the world and has profoundly shaped human societies . Originating in the Near East around 11 , 000 y ago , an agricultural lifestyle subsequently spread across Europe during the New Stone Age ( Neolithic ) . Whether it was mediated by incoming farmers or driven by the transmission of innovative ideas and techniques remains a subject of continuing debate in archaeology , anthropology , and human population genetics . Ancient DNA from the earliest farmers can provide a direct view of the genetic diversity of these populations in the earliest Neolithic . Here , we compare Neolithic haplogroups and their diversity to a large database of extant European and Eurasian populations . We identified Neolithic haplotypes that left clear traces in modern populations , and the data suggest a route for the migrating farmers that extends from the Near East and Anatolia into Central Europe . When compared to indigenous hunter–gatherer populations , the unique and characteristic genetic signature of the early farmers suggests a significant demographic input from the Near East during the onset of farming in Europe .
|
[
"Abstract",
"Introduction",
"Results/Discussion",
"Materials",
"and",
"Methods"
] |
[
"evolutionary",
"biology",
"evolutionary",
"biology/human",
"evolution",
"genetics",
"and",
"genomics/population",
"genetics"
] |
2010
|
Ancient DNA from European Early Neolithic Farmers Reveals Their Near Eastern Affinities
|
Betanodaviruses cause massive mortality in marine fish species with viral nervous necrosis . The structure of a T = 3 Grouper nervous necrosis virus-like particle ( GNNV-LP ) is determined by the ab initio method with non-crystallographic symmetry averaging at 3 . 6 Å resolution . Each capsid protein ( CP ) shows three major domains: ( i ) the N-terminal arm , an inter-subunit extension at the inner surface; ( ii ) the shell domain ( S-domain ) , a jelly-roll structure; and ( iii ) the protrusion domain ( P-domain ) formed by three-fold trimeric protrusions . In addition , we have determined structures of the T = 1 subviral particles ( SVPs ) of ( i ) the delta-P-domain mutant ( residues 35−217 ) at 3 . 1 Å resolution; and ( ii ) the N-ARM deletion mutant ( residues 35−338 ) at 7 Å resolution; and ( iii ) the structure of the individual P-domain ( residues 214−338 ) at 1 . 2 Å resolution . The P-domain reveals a novel DxD motif asymmetrically coordinating two Ca2+ ions , and seems to play a prominent role in the calcium-mediated trimerization of the GNNV CPs during the initial capsid assembly process . The flexible N-ARM ( N-terminal arginine-rich motif ) appears to serve as a molecular switch for T = 1 or T = 3 assembly . Finally , we find that polyethylene glycol , which is incorporated into the P-domain during the crystallization process , enhances GNNV infection . The present structural studies together with the biological assays enhance our understanding of the role of the P-domain of GNNV in the capsid assembly and viral infection by this betanodavirus .
Nodaviridae is a family of positive-sense single-stranded RNA viruses with a non-enveloped T = 3 capsid . These viruses are characterized by a viral genome comprising two RNA molecules–RNA1 and RNA2 . RNA1 ( 3 . 1 kb ) encodes protein A , which is a RNA-dependent RNA polymerase ( RdRp ) responsible for viral RNA replication [1 , 2] . RNA2 ( 1 . 4 kb ) encodes the structural protein associated with assembly of the viral particle . The subgenomic RNA3 , located at the 3’-terminal region of RNA1 , encodes a non-structural B2 protein , which plays a role in inhibition of host RNA interference ( RNAi ) [3–6] . Alphanodaviruses and betanodaviruses are the major genera in the family Nodaviridae [7] . Alphanodaviruses infect primarily insects , and are related to the Nodamura virus ( NoV; PDB ID: 1NOV ) , Black beetle virus ( BBV; PDB ID: 2BBV ) , Pariacoto virus ( PaV; PDB ID: 1F8V ) and Flock house virus ( FHV; PDB ID: 4FSJ ) . Betanodaviruses are also called nervous necrosis viruses ( NNV ) because they cause an acute syndrome of viral nervous necrosis ( VNN ) [8] . VNN is a serious syndrome disease causing viral encephalopathy or retinopathy , and is responsible for the high mortality at the larval stage among a wide range of species ( warm- and cold-water fishes ) or even across species from marine to freshwater fishes in the aquaculture industry [9] . Betanodavirus strains are currently classified into four distinct genotypes based on the genes encoding the viral capsid protein ( CP ) . These include the Striped Jack nervous necrosis virus ( SJNNV ) , Tiger puffer nervous necrosis virus ( TPNNV ) , Red-spotted grouper nervous necrosis virus ( RGNNV ) and Barfin flounder nervous necrosis virus ( BFNNV ) [10] . Based on genome organization and on phylogenetic analysis of RNA1 or RNA2 , additional clusters of unclassified nodaviruses infecting nematodes , moths , butterflies and prawns have been identified recently [11] . One report identifies two unclassified nodaviruses ( shrimp nodavirus ) , Macrobrachium rosenbergii Nodavirus ( MrNV ) and Penaeus vannamei Nodavirus ( PvNV ) , which cause muscle necrosis in prawns [12] . These findings suggest that the family Nodaviridae includes not only the known types but also other members with a wide distribution . In the family Nodaviridae , an assemblage of 180 CPs form a T = 3 capsid of diameter ~29−35 nm . CP is typically composed of the core jelly-roll topology , forming a face-to-face β-sandwich with two pairs of anti-parallel β-sheets [13] . During assembly of the alphanodavirus particle , self-catalyzed cleavage of the precursor protein α generates proteins β and γ , which are required for structural maturation of the capsid [14] . Protein β forms the canonical eight anti-parallel β-strands with N- and C-termini located inside the virus particle . The highly basic N-terminus of protein β is required to neutralize the encapsidated RNA duplex [15 , 16]; it also acts as a molecular switch to control the heterogeneous size and shape of the particles [17] . The structural complementarities between the different strains of the genus alphanodavirus appear conserved , despite the existence of large evolutionary distances in phylogenetic relations [7] . However , there is no significant homology in the CP sequences between alphanodaviruses and betanodaviruses . Genotypes of the RGNNV-strain betanodavirus isolated from different grouper species , such as Orange-spotted grouper nervous necrosis virus ( OSGNNV ) , Dragon grouper nervous necrosis virus ( DGNNV ) and Malabaricus grouper nervous necrosis virus ( MGNNV ) , contain highly conserved genomes . Three uninterrupted major domains of MGNNV CP , including the N-terminal region , the β-sandwich surface domain and the trimeric protrusion domain , have been previously studied by cryo-electron microscopy ( cryo-EM ) imaging at 23 Å resolution and 3D-PSSM prediction [18] . However , there is currently no high-resolution structural information on the capsid-related organization of the genus betanodavirus . In this report , we describe the crystal structure of the grouper nervous necrosis virus ( GNNV ) of the genus betanodavirus in various forms: ( i ) a complete T = 3 GNNV-like particle ( GNNV-LP ) at 3 . 6 Å resolution; ( ii ) T = 1 subviral particles ( SVPs ) of the delta-P-domain mutant at 3 . 1 Å; ( iii ) the N-ARM deletion mutant at 7 . 0 Å; and ( iv ) the individual P-domain of GNNV CP at 1 . 2 Å . The crystal structure of GNNV-LP demonstrates several significant and distinct variations in capsid architecture and molecular mechanisms of capsid assembly compared to the genus alphanodavirus and other RNA viruses . In particular , we have identified the conserved structural characteristics of the shell domain on GNNV . Various forms of the T = 3 and T = 1 GNNV capsids show that the N-terminal arginine-rich motif ( N-ARM ) acts as a molecular switch . Second , the P-domain , with its DxD motif together with two bound Ca2+ ions , plays a pivotal role in the trimerization of the GNNV CP and the particle assembly . These high-resolution structural details contribute further to our in-depth understanding of the molecular mechanisms of viral assembly and infection , and should provide the structural basis for studying the evolution of the family Nodaviridae .
SUMO-GNNV CPs are overexpressed in Escherichia coli ( E . coli ) and the GNNV-LPs are self-assembled in vitro . Based on the EM images , the morphology of GNNV-LP shows a T = 3 capsid with a diameter of 30~35 nm ( Fig 1A and S2A Fig ) . We determine the crystal structure of the T = 3 GNNV-LP using the ab initio method with non-crystallographic symmetry ( NCS ) averaging and refine the structure to 3 . 6 Å ( S1 Fig ) . The electron density of the icosahedral asymmetric unit ( iASU ) of the T = 3 GNNV-LP allows modeling of residues 52−338 for subunits A and B , and residues 34−338 for subunit C . The rest of the N-terminal segment of each subunit , which contains N-ARM , the positively charged arginine-rich motif 23RRRANNRRRSN33 , is disordered . The overall topological structure of the GNNV CP consists of the N-terminal arm ( N-arm ) ( residues 34−51 ) , the shell domain ( S-domain ) ( residues 52−213 ) , the linker region ( residues 214−220 ) and the protrusion domain ( P-domain ) ( residues 221−338 ) ( Fig 1B ) . The ordered N-arm exists along the icosahedral two-fold ( I2 ) interface of the inner surface , and extends its N-terminus to the icosahedral three-fold ( I3 ) axis to form a β-annulus . The S-domain comprises an eight-stranded anti-parallel β-sandwich with three short α-helices , which is a canonical structural feature similar to other virus CPs [13] . The individual S- and P-domains of the GNNV CP , connected by the flexible linker region , do not interact with each other directly . The P-domain folds into an independent structure , including eight anti-parallel β-strands and a short α-helix connected with loops of various lengths ( Fig 1C ) . Sixty trimeric S-domains participate in inter-subunit contacts , forming a continuous thin shell of the capsid with an empty inner cavity . Three neighboring P-domains per iASU embrace one another at the quasi three-fold ( Q3 ) axes to form 60 protrusions on the particle surface ( Fig 1D ) . Three neighboring monomeric S-domains from subunits A , B and C are engaged in dimeric , trimeric and pentameric interactions along the I2 , I3 and icosahedral five-fold ( I5 ) axes ( Fig 1D ) . Although the GNNV CP ( 338 residues ) is shorter than the alphanodavirus CP ( 407 residues ) , the structural organization of the GNNV capsid with its 60 large protrusions reveals a T = 3 architecture with a particle size similar to the compact alphanodavirus structure , in which the N- and C-termini of the CP are both positioned within the capsid . Only the partial N-terminus of each subunit C is seen inside the capsid; the N-termini of subunits A and B are completely absent . The first 33 residues of the N-termini , namely the N-ARM , are disordered in all the subunits . This flexible structural feature of the basic N-ARM is thought to play an important role in the RNA encapsidation in the intact virus . Two ordered and extended N-arms from the subunit-C/C dimer , together with their corresponding N-ARMs , occupy the groove of the inner surface along the I2 interface ( Fig 2A ) . Residues 36−41 from subunits C1 , C10 and C12 are engaged through hydrogen bonding to form a β-annulus structure around the I3 axis ( Fig 2B ) . The β-annulus structure of GNNV is similar to that of the Rice yellow mottle virus ( RyMV ) [19] , but differs from that of the Sesbania mosaic virus ( SeMV ) , in which three N-arms from subunits C1 , C7 and C9 form a β-annulus structure around another I3 axis [20 , 21] . Notably , each genotype of the genus betanodavirus has a conserved residue , Pro38 , for stabilization of the β-annulus structure , and this proline residue corresponds to Pro35 in RyMV and Pro53 in SeMV ( Fig 2B ) [19–21] . The N-arm of subunit C1 in GNNV is oriented at the B1-C6 interface toward one I3 axis , similar to that in RyMV . In contrast , the N-arm of subunit C1 in T = 3 RNA plant viruses , such as SeMV , folds back to result in an anti-parallel topology facing the first β-strand B of the S-domain . This results in a hairpin conformation along the I2 interface . The San Miguel sea lion virus ( SMSV ) of the family Caliciviridae also contains three ordered N-arms from the C1 , C10 and C12 subunits located near the I3 axis similar to GNNV , but oriented toward another direction ( Fig 2C ) [22] . Thus , based on structural conformation , the N-arms of the viral CP can be classified into several categories . The S-domains of the CPs in the GNNV-LP form a conserved jelly-roll structure as in those of canonical viruses [13] . Within each CP , two four-stranded anti-parallel sheets ( β-strands BIDG and CHEF ) are connected with two α-helices between strands C and D and one α-helix between strands E and F , respectively . A search of structural homologs between the GNNV S-domain and the corresponding domain in the CPs of other viruses using the DALI program [23] shows the highest similarity with the Orsay virus ( Z-score 23 . 8 ) [24] and the Carnation mottle virus ( CMV ) ( Z-score 18 . 1 ) [25] . The CP subunits adapt to the quasi-equivalent interactions of the triangulated icosahedral lattices , suggesting that the N-terminus of the CP is a molecular switch to adjust the curvature of the subunit-A/B dimer along the quasi two-fold ( Q2 ) axis and the subunit-C/C dimer along the I2 axis during T = 3 particle assembly [26] . The bent conformation of the subunit-A/B dimer in GNNV-LP is similar to that observed for the CP in the alphanodavirus . The flat conformation of the subunit-C/C dimer is stabilized by two ordered N-arms alone , in contrast to alphanodavirus , where incorporation of the encapsidated RNA participates in the T = 3 quaternary organization [15 , 16] . The strand B and the D-E loop on subunit C interact with the N-arm from the neighboring subunit C6 through hydrogen bonds to stabilize the subunit-C/C dimer . Divalent metal ions , such as calcium , are typically associated with metal-coordinating residues for particle formation , stability and infectivity [27] . The GNNV-LP has three Ca2+ ions located at interfaces between pairs of subunits within each of the S-domains , which are coordinated with side chains of Asp130 and Asp133 to form the 130DxxDxD135 motif at the E-F loop , Gln100 at the C-D loop , Ser170 at the G-H loop and Glu213 near the linker region from the neighboring subunit ( S4A Fig ) . There are three S-domains per iASU , and they all share the same calcium-binding structures to facilitate subunit-subunit interactions , similar to those seen in the CP of some RNA plant viruses , such as tombusvirus ( DxDxxD ) [28–30] and SeMV ( DxxD ) [31] . In contrast , alphanodavirus utilizes Asp249 and Glu251 to form DxExxD motif and incorporate one or two Ca2+ along the Q3 axis in its CP [27 , 32] . The electrostatic potential surface in the region of the S-domain of the GNNV-LP shows distributions of positively- and negatively-charged regions that are more dispersed on the inner surface compared with T = 3 PaV ( S4B Fig ) . Earlier crystal and cryo-EM structures of the PaV have suggested that 30 copies of an ordered encapsidated RNA duplex formed a dodecahedral cage within the inner surface [15] These data indicate that the encapsidated RNA of GNNV may be involved in a non-specific interaction with the inner surface or a specific interaction with positively charged residues of the flexible N-ARM inside the GNNV capsid . In the cryo-EM structure of MGNNV , 60 large protrusions along the Q3 axes have been identified that are larger than the extended domain ( 34 residues ) of alphanodavirus [18] . Our crystal structure of the T = 3 GNNV-LP also shows 60 protrusions on the particle surface along the Q3 axes formed by three contiguous P-domains per iASU . Although the structure of the P-domain can be readily assigned , the protrusions of the GNNV-LP show too poor electron density after NCS-averaging to allow a complete characterization of the morphology of the P-domain , which might be caused by the high flexibility . To gain more complete and detailed structural information , we have determined the crystal structure of the truncated P-domain ( residues 214−338 ) at high resolution ( 1 . 2 Å; Fig 3A and S7A Fig ) . Since it lacks only the S-domain , crystal packing of the truncated P-domain remains a trimer in the ASU , with a similar orientation and formation as the P-domains in the GNNV-LP . The anti-parallel β-strands C’ and D’ are located at the interface of the Q3 axis , presenting an L-shaped geometry facing the other six-stranded β-sheets ( A’ , B’ , E’ , F’ , G’ and H’ ) . This conformational arrangement is different from the jelly-roll topology . The flexible C-terminus of the P-domain is located near the linker region at the interface space between the S- and P-domains in the GNNV-LP . Structural alignments of the GNNV P-domain with all structures in the PDB reveal low degrees of structural similarities with the P-domain of Orsay virus ( Z-score 5 . 1 ) [24] , the P1-domain of Hepatitis E virus ( HEV ) ( Z score 2 . 2 ) [33 , 34] and the P1-domain of Calicivirus ( Z-score 2 . 1 ) ( S5 Fig ) [22] . The high-resolution structure of the truncated P-domain has allowed us to clearly locate two Ca2+ ions near the non-crystallographic three-fold axis , which are coordinated with the C’-D’ loop to stabilize the trimeric structural fold ( Fig 3A and 3B and S7A Fig ) . The 273DxD275 motif on the C’-D’ loop from each neighboring subunit interacts with two Ca2+ ions and two water molecules through electrostatic and hydrogen-bonding interactions . This calcium-binding site is buried in the cavity of the protrusion at a distance of ~37 Å from the S-domain ( Fig 3A and 3B ) . The distances between two Ca2+ ions and the side chains of Asp273 and Asp275 from each subunit are ca . 2 . 4~2 . 5 Å . Notably , only two of the three Asp275 residues are asymmetrically coordinated to the two Ca2+ ions , and the other Asp275 coordinates with one water molecule . A similar asymmetrical binding of two Ca2+ ions and two water molecules with three Asp273 is observed ( Fig 3C ) . Analysis of the elution profiles of the P-domain after size-exclusion chromatography ( SEC ) showed a possible role of Ca2+ in the trimerization of P-domains , suggesting that formation of the trimeric structure of the P-domains might be initiated and completed in the absence and presence of Ca2+ , respectively ( S6B Fig ) . Water molecules have been observed at the inter-subunit interfaces within the complete viral capsid; they must be important in stabilizing association of the subunits [35] . From the high-resolution structure , we have delineated the distribution of water molecules in the P-domains . As mentioned above , there are two water molecules at the calcium-binding site providing the trimeric contacts and stabilizing the protrusion ( Fig 3C ) . At the interface between the D’ and E’ strands with the F’-G’ loop from neighboring subunits , we also find two invariant water molecules associated with 278VYWH281 , Gly299 , Gln322 and Ile323 through hydrogen bonds , which are also essential to maintain the conformation and stability of each of the trimeric P-domains ( S6A Fig ) . A multiple amino-acid sequence alignment of P-domains from different genotypes of the genus betanodavirus reveals that several regions , including residues 223−227 , 233−237 , 253−259 and 285−291 , are divergent . Notably , all these residue variations are located on the surface of the protrusion in the structure of the truncated P-domain ( Fig 3D ) . An inspection of the structure of the truncated P-domains reveals additional electron densities in several pockets on the surface . We consistently find three glycerol ( GOL ) molecules located between the B’-C’ and F’-G’ loops of the three P-domains , and one polyethylene glycol ( PEG ) molecule at the interface between the two F’-G’ loops from neighboring subunits ( S7B Fig ) . The B-factor values of GOL and PEG molecules are 21 . 6 and 26 . 5 Å2 , respectively . A previous study has showed that PEG could increase the ability of Hepatitis B virus ( HBV ) to bind to the cell surface and to enhance virus infection [36] . On the basis of this lead , we examine the grouper fin cell line ( GF-1 ) infected with GNNV in the presence ( 4% ) of PEG3000 or PEG8000 , respectively . Compared to the untreated group , the viral copy number was significantly higher in the presence of PEG , especially PEG8000 ( ~30 folds ) , within 24 hours ( Fig 3E ) . These data suggest that the infectivity of GNNV for GF-1 cells could be enhanced with PEG8000 ( 4% ) during infection . Based on the PEG-binding ability of the P-domain , we surmise that the presence of PEG might participate in the early step ( s ) of GNNV infection . The symmetry of the icosahedral particles can be related to a regulation process that dictates the choices of inter-subunit arrangements or protein-nucleotide interactions to guide the capsid assembly [37] . For T = 3 GNNV-LP , the N-terminus of the CP contains the disordered N-ARM for putative RNA interactions at the inner cavity of the particles . The next N-arm is ordered along the I2 interface only on subunit C . The P-domains of GNNV-LP show an independent trimeric organization , which is different from that of the S-domains . We therefore speculate that the N-ARM or P-domain of GNNV-LP might act as a major molecular switch in regulating T = 3 or T = 1 assembly . To address this issue , we have constructed two sub-clones , including ( i ) the delta-P-domain mutant ( residues 35−217 ) and ( ii ) the N-ARM deletion mutant ( residues 35−338 ) , and have determined their structures ( Table 1 ) . In the delta-P-domain mutant , sixty copies of the S-domain assemble with interactions of I2 , I3 and I5 symmetries into a T = 1 SVP with a diameter ~190 Å ( Fig 4A and S2C Fig ) . Only residues 52−214 of each subunit are observed at a resolution of 3 . 1 Å . As expected , the delta-P-domain mutant comprises a canonical eight-stranded anti-parallel β-sandwich with three short α-helices similar to the S-domain of T = 3 GNNV-LP . In the N-ARM deletion mutant , the crystals diffract to only 7 Å resolution . However , analyses of self-rotation functions and molecular replacement indicate that the N-ARM deletion mutant could form T = 1 capsid of diameter ~240 Å , which is consistent with the EM images ( Fig 4A and S2B and S3 Figs ) . Although the organization of the equivalent subunits around the I3 axes of the T = 1 delta-P-domain mutant is notably similar to the arrangement of the iASU subunits of the T = 3 GNNV-LP , the organization of the trimeric subunits is flatter than that of the T = 3 GNNV-LP . In the T = 1 delta-P-domain mutant , there is no Ca2+ observed at corresponding calcium-binding sites as seen in the S-domain of T = 3 GNNV-LP . The hollow or empty binding site exhibits an expanded geometry with maximum movement of ~2 . 6 Å of the main chains ( Fig 4B ) . We compare the quaternary organizations of the T = 1 delta-P-domain mutant and the T = 3 GNNV-LP by superimposing dimeric , trimeric and pentameric partners in order to evaluate the rotation and translation of selected subunit-pairs . The differences in rotational angles and translations at the interfaces of several subunits are identified ( Fig 4C ) . We find that , without the Ca2+-mediated interactions at the subunit interfaces , the weaker contacts cause changes in the inter-subunit organization , with expanded assembly of the T = 1 delta-P-domain mutant , similar to the structure in the Asp mutants of T = 1 SeMV [38] .
Studies investigating the formation of the β-annulus structure with three conserved proline residues around the I3 axis suggest that the N-arm of GNNV , containing only 18 residues , is too short to cooperate with the first β-strand B of the S-domain to form a hairpin structure as in SeMV [20 , 21] . Instead , the β-annulus in GNNV is formed by three N-arms from the C1 , C10 and C12 subunits at the different I3 axis , with a symmetric geometry similar to RyMV ( Fig 2C ) [19] . The residues Asp36–Lys41 of the GNNV N-arm with the conserved proline residue ( Pro38 ) contribute to the formation of the β-annulus around the I3 axis via hydrogen bonding , similar to those of other RNA plant viruses , such as RyMV [19] , SeMV [20 , 21] , CMV [28] , Tomato bushy stunt virus ( TBSV ) [29 , 30] and Southern bean mosaic virus ( SBMV ) [39 , 40] . Through this comparison , we might infer that the three ordered N-arms contributed by the C-subunits are involved in the trimeric β-annulus structure regardless of the sequence variations of CPs with large evolutionary distances or different fold-classifications of the following N-arms between GNNV and other RNA plant viruses . Furthermore , the flat contacts of the subunit-C/C dimer seem able to create a spacious locus to accommodate two ordered N-arms , which are stabilized by hydrogen bonds , in the T = 3 GNNV-LP structure without RNA encapsidation ( S4B Fig ) . This structure feature is different from that in alphanodaviruses such as PaV , where a piece of genomic RNA is incorporated with the ordered arm of the subunit A . This subunit A-RNA interaction has been proposed to be necessary to promote the flat conformation of the subunit-C/C dimer [16] . Taken together , it appears that the formation of the β-annulus with the three conserved Pro38 around the I3 axis , the specific length of N-arm along the I2 interface , and the cavity space created by flat contacts of the subunit-C/C dimer might be essential for the morphology and the order of the N-terminus of CPs during the T = 3 GNNV assembly . Despite significant variations in amino-acid sequences , the structure of the GNNV S-domain exhibits a jelly-roll topology , similar to other structural viral CPs ( Fig 2C ) . Divalent metal ions , such as Ca2+ or Zn2+ , have been shown to play a crucial role in subunit interactions , particle stability , virion infection and environmental resistance in polyomavirus [41] , rotavirus [42] , tombusvirus [28–30] , sobemovirus [19 , 31 , 40] and nodavirus [27 , 32] . In GNNV , three Ca2+ ions per iASU are incorporated into the 130DxxDxD135 motif at the interfaces of the S-domains , as found in similar structural regions of T = 3 RNA plant viruses [19 , 28–31 , 40 , 43] . Therefore , the characteristic folds of viral CPs might be most likely a consequence of the geometric requirements of the building block , which is favorable as the jelly-roll β-barrel fold with conserved sequence patterns , including a calcium-binding site for the distinctive viral shell architecture [13 , 44] . Investigation of the Asp mutations on DGNNV [43] and our GNNV-LP structure shows that Asp130 and Asp133 , but not Asp135 , coordinate with Ca2+ for particle formation and stabilization . There are four cysteine residues on the GNNV CP ( Cys115 , Cys187 , Cys201 and Cys331 ) . Based on the failure of VLP formation in the presence of single mutations of either C115A or C201A , the existence of a disulfide-bond linkage between Cys115 and Cys201 was previously postulated [45] . However , a structural inspection of the S-domain of GNNV shows that the distribution of Cys115 , Cys187 and Cys201 is too remote to establish intra- or inter-subunit disulfide-bond linkages , implying that the disulfide bond is not required for the proper assembly of the GNNV capsid . The locations of Cys115 and Cys201 in GNNV are similar to those of Cys131 and Cys252 of SeMV [46] but different from those of the Cys105−Cys197 disulfide found in Orsay virus [24] ( S4C Fig ) . Viral CP is generally divided into several categories according to the number of short connecting linkers between the P- and S-domains . We find that the number of linkers , one or two , might correspond to localizations of the N- and C-termini on the opposite or the same side , respectively . The structure of GNNV provides an example of a unique topology with only one linker connecting the P- and S-domains of T = 3 GNNV and trimeric P-domains with Ca2+ for 60 protrusions along the Q3 axes ( Fig 5A ) . In contrast , the dimeric P-domains for 30 protrusions along the I2 axes and 60 protrusions along the Q2 axes appear with one flexible hinge between P- and S-domains in several T = 3 viral capsids , such as the families of Caliciviridae and Tombusviridae [22 , 28] . Two anti-parallel linkers with the trimeric P-domains along the Q3 axes have been reported in the infectious bursal disease virus ( IBDV ) of the family Birnaviridae , which is similar to alphanodavirus [47] . There are two independent linkers from the P1-domain around the three-fold axes connecting the S-domain and the P2-domain for 30 protrusions along the I2 axes and 60 protrusions along the Q2 axes , respectively , on T = 3 HEV of the family Hepeviridae , similar to Caliciviridae [33 , 34] . We propose that the organization of betanodavirus in the family Nodaviridae is intermediate between the families of Tombusviridae , Caliciviridae and Birnaviridae through evolutionary lineage . Previous studies have indicated that the surface protrusions on a viral capsid play a crucial role in antigenicity and endocytosis as a result of receptor interactions during virus infection [22 , 34] . Our high-resolution structure of the truncated P-domain provides not only a structural framework to investigate the particle formation , but also the aetiological basis of host fish-species specificity . We show that the individual P-domain contains a significant 273DxD275 motif for calcium binding ( Fig 3A and 3C ) . One Zn2+ ion involved in the trimeric organization of VP6 on the rotavirus was previously found at the bottom of the protrusion and near the S-domain along the Q3 axes [42] . In contrast to the rotavirus , this asymmetrical arrangement of two Ca2+ ions and two water molecules coordinating with three sets of the 273DxD275 motif in the truncated P-domain structure might exist on 60 protrusions of the native T = 3 GNNV . We demonstrate that Ca2+ plays a significant role in the trimerization of P-domains ( S6B Fig ) . Two metal-binding regions – 130DxxDxD135 of the S-domain and 273DxD275 of the P-domain–might be essential for the organization and stabilization of T = 3 GNNV . In addition to Ca2+ , conserved water molecules are consistent and integral components of the interfaces between neighboring subunits . These water molecules constitute the primary components of the GNNV protrusion for stabilization through a network of hydrogen bonds ( Fig 3C and S6A Fig ) . In addition , a structural comparison reveals that the truncated P-domain contains the rigid P-domain with the disordered linker region , and this linker region of T = 3 GNNV-LP also exhibits large B-factor values ( S7C Fig ) . This analysis may provide insights into why the flexible linker region allows the entire solid P-domain to be malleable , resulting in the broken electron density of the P-domain with large B-factor values for the T = 3 GNNV-LP . Oligomerization of CPs is the first intermediate step in capsid assembly . Based on SEC analysis , the trimeric truncated P-domains appear in the presence of Ca2+ ( S6B Fig ) . Interestingly , we observe trimerization of the full-length GNNV CP ( 112 kDa ) and the N-ARM deletion mutant ( 100 kDa ) as well as dimerization of the delta-P-domain mutant ( 40 kDa ) using SDS-PAGE ( S6C Fig ) . Compared with the full-length GNNV CP and the N-ARM deletion mutant , only the T = 1 delta-P-domain mutant might exhibit the dimeric capsomer formation in the assembly process , which is similar to that in some RNA plant viruses as well as the Orsay virus , which exhibits the trimeric protrusion in solution ( Fig 5B ) [19 , 20 , 24] . These results suggest that the P-domain may play a major role in promoting trimerization of the GNNV CPs in the initial assembly processes of the T = 3 GNNV and the T = 1 N-ARM deletion mutant under a Ca2+ environment ( Fig 5B and S6B and S6C Fig ) . The genus betanodavirus is generally classified into four genotypes: SJNNV , BFNNV , TPNNV and RGNNV . A comparison of the genetic heterogeneity of each genotype indicates that the P-domain is a major distinct region [10] . Several hypervariable regions on the P-domain coincide with the protrusion surface associated with the functionalities of the receptor binding and host-cell specificity ( Fig 3D ) [48] . This observation suggests an evolutionary divergence , resulting in distinct phenotypes of betanodavirus with various fish-host specificities . Heparan sulphate proteoglycans ( HSPs ) are negatively charged components of the cell surface and play a role in virion attachment to host cells and binding to secondary host receptors during viral infection . For instance , the surface L-protein of HBV is reported to bind to glycosaminoglycans ( GAGs ) on the host-cell surface; its GAG-dependent binding is enhanced by PEG to facilitate viral infection [36] . Our study identifies a PEG-binding site on the P-domain of GNNV CP and confirms the enhancement of GNNV infection in presence of PEG ( Fig 3E ) . The heparin-binding ability of GNNV CP has also been demonstrated using immobilized heparin-affinity chromatography [49] . Interestingly , our qPCR analysis of virus copies in GF-1 cells with heparin-containing medium detected no significant signal , suggesting that the presence of heparin suppressed GNNV infection as well . Taken together , GNNV infection might be similar to HBV infection in that they both require an initial attachment to the carbohydrate side-chains of HSPs . Furthermore , the hydrophobic moiety of PEG incorporated on the GNNV P-domain might improve the penetration of non-enveloped viruses across the cell membrane [50] . The organization of the N-terminus and encapsidated RNA have been implicated in providing a dynamic equilibrium of the dimeric subunits between “bent” and “flat” conformations during viral assembly [26] . The N-terminus of GNNV CP is composed of the disordered N-ARM and the ordered N-arm comprising the β-annulus , similar to SeMV [20 , 21 , 38] , and plays a role in regulating T = 3 capsid assembly . However , the bent conformation of the subunit dimer leads to the disordered N-arm lying at the inner cavity of the T = 1 delta-P-domain mutant without β-annulus formation . This observation indicates that the N-ARM of GNNV CP makes an essential contribution to the organization of the β-annulus along the I3 axes . We propose that the β-annulus on T = 3 GNNV might be an outcome of T = 3 capsid assembly rather than a profound effect on switching structural symmetries . Both crystal structures of the N-ARM deletion mutant and the delta-P-domain mutant without the N-ARM of GNNV reveal exclusively T = 1 architecture . Comparatively , in alphanodavirus , a complete N-ARM deletion ( delta residues 1−54 ) leads to the inhibition of particle assembly . Conversely , the partial N-ARM deletion ( delta residues 1−31 ) was shown to cause the formation of highly heterogeneous particles , including small bacilliform-like and irregular structures [17] . Furthermore , particle polymorphism of cowpea chlorotic mottle virus ( CCMV ) was previously described [51] , and its N-ARM ( residues 1−25 ) was invisible in the crystal structure [52] . A N-terminal domain deletion mutant ( delta residues 1−34 ) of the CCMV CP resulted in three categories of particles: T = 3 VLPs and two SVPs of T = 2 and T = 1 architectures in vitro [53] . The disordered N-ARM might be a critical structural feature of a molecular switch for controlling particle assembly , but this phenomenon was not found in the case of the Orsay virus [24] . A sequence comparison of CPs in the family Nodaviridae shows that the Orsay virus CP contains a basic-charged N-terminus but lacks the N-ARM ( S8C Fig ) . The N-terminal deletion mutant of Orsay virus forms a T = 3 architecture similar to full-length CP . It is therefore reasonable to assume that the cumulative number of Arg residues on CP might form a proper N-ARM and lead to the spontaneous self-assembly of particle polymorphism , or even the failure of particle assembly . The T = 1 delta-P-domain mutant without Ca2+ incorporated shows that the region containing the residues that potentially coordinate Ca2+ exhibits an expanded geometry , decreasing subunit contacts along the two- , three- and five-fold axes , and a flatter dimeric contact of curvature ~155° , compared with T = 3 GNNV-LP ( Fig 4B and 4C ) . A mutational analysis of Asp residues on SeMV has previously showed that Ca2+ coordination is unnecessary for capsid assembly but essential for capsid stability [31] . This is consistent with the observations that Ca2+ ions participate in the stability of the GNNV capsid but are not critical for the formation of T = 3 or T = 1 particles . Taken together , the N-ARM precedes other primary structural components , such as the β-annulus , P-domain and Ca2+ , to be a molecular switch to ensure the error-free T = 3 GNNV assembly . In summary , this work provides several important structural insights into the genus betanodavirus GNNV . Despite conservation of a viral genome encoding three major proteins and a compatible geometry of the T = 3 architecture in the family Nodaviridae , the structure of the GNNV-LP obtained here allows us to delineate the key structural components that trigger the oligomerization and stabilize the capsid assembly . Although the jelly-roll fold of the S-domain and the structure of the β-annulus of GNNV capsid are similar to those of known T = 3 RNA plant viruses , GNNV exhibits different fold-classifications of the N-arm and the calcium-incorporating trimeric P-domains with a specific DxD motif for trimerization of CPs . The GNNV structure also shows that the hypervariable surface regions of the P-domain contribute to host binding and specificity . The molecular organizations and assembly mechanisms of GNNV reveal that the genus betanodavirus in the family Nodaviridae may belong to a significant genus under the viral evolutional pathway among the Tombusviridae , Caliciviridae and Birnaviridae families . Structural mapping of the GNNV P-domain might be useful for the development of vaccine strategies in the fish aquaculture industry .
All animal experiments were performed in strict accordance with the recommendations in the guide for the Institutional Animal Care and Use Committee , National Cheng Kung University . The protocol was approved under the Institutional Animal Care and Use Committee ( IACUC ) of National Cheng Kung University ( IACUC #100065 ) . A consensus CP DNA sequence from the orange-spotted grouper nervous necrosis virus ( OSGNNV ) RNA2 ( GenBank accession no KT071606 ) was amplified by PCR and cloned into a modified pET32-Xa/LIC vector carrying 6×histidine residues and yeast SUMO ( SMT3 ) as the N-terminal fusion tag [54] . This construct was expressed in E . coli BL21-CodonPlus ( DE3 ) -RIL ( Stratagene ) , and the cells were cultured in Luria Bertani ( LB ) broth ( Merck ) containing chloramphenicol ( 34 μg/ml ) and ampicillin ( 100 μg/ml ) until the OD reached 0 . 6–0 . 7 at 600 nm at 37°C . IPTG ( isopropyl β-D-thiogalactopyranoside ) ( Bioshop ) was added to a final concentration of 0 . 5 mM and cultures were incubated overnight at 18°C . The cells were harvested and disrupted by sonication in lysis buffer ( 50 mM Tris HCl ( pH 8 . 0 ) , 0 . 25 M NaCl , 20 mM imidazole , 5 mM β-mercaptoethanol and 1 mM EGTA ) . CP was purified through a Ni-NTA column ( GE Healthcare ) . The SUMO-tag was cleaved using SUMO protease that was later removed with a Ni-NTA column . The purified GNNV CP was diluted to a concentration of 0 . 3 mg/ml and dialyzed overnight at 4°C against lysis buffer without EGTA or β-mercaptoethanol at a ratio of 1:150 . ( NH4 ) 2SO4 ( 750 mM ) was added to the dialysis , and GNNV CP was finally dialyzed against the GNNV-LP formation buffer ( 20 mM Tris HCl ( pH 8 . 0 ) , 0 . 2 M NaCl , 1% ( v/v ) glycerol and 2 mM CaCl2 ) . The size of GNNV-LP was measured by size-exclusion chromatography on a Superose 6 10/300 GL column ( GE Healthcare ) . The purified GNNV-LP was concentrated to 30 mg/ml and stored at 4°C . The truncated P-domain ( residues 214−338 ) , delta-P-domain mutant ( residues 35−217 ) and N-ARM deletion mutant ( residues 35−338 ) proteins were prepared using the same methods described above for GNNV CP . The truncated P-domain ( 20 mg/ml ) and S-domain ( 30 mg/ml ) proteins were stored in a buffer containing 300 mM NaCl and 50 mM Tris HCl ( pH 7 . 5 ) , whereas the N-ARM deletion mutant ( 30 mg/ml ) was stored in the GNNV-LP formation buffer at 4°C . The purified GNNV-LP , the N-ARM deletion mutant and the delta-P-domain mutant were all diluted to a final concentration of 50 μg/ml and blotted on freshly glow-discharged , carbon-coated 200 mesh copper grids ( NISSHIN EM Co , Ltd . , Tokyo , Japan ) . Grids were negatively stained with 5 μl of 2% ( w/v ) uranyl acetate solution and screened using the H-7650 transmission electron microscope ( Hitachi High-Technologies Co . ) operated at 80 kV . All images were acquired using a 1024 x 1024 pixels CCD camera ( TVIPS , Gauting , Germany ) and recorded at a magnification of 100 , 000 × . The initial GNNV-LP crystallization experiment was performed at 18°C with the hanging-drop vapor-diffusion method . A Mosquito liquid-handling robot ( TTP Labtech ) was used for high-throughput crystallization condition screening . The initial condition of 0 . 2 M sodium formate ( pH 7 . 2 ) and 20% ( w/v ) PEG3350 was obtained from the PEG/Ion Screen I kit ( Hampton Research ) . This condition was further optimized to improve the diffraction quality and resolution of the crystals . Crystals appeared within 1−2 weeks . All crystals were cryoprotected with 25~30% ( w/v ) PEG3350 and frozen in liquid nitrogen before data collection . X-ray diffraction data were collected on BL44XU at SPring-8 ( Harima , Japan ) with a CCD detector ( MX225-HE , Rayonix ) using X-ray wavelength of 0 . 9 Å . All images were collected with an oscillation angle of 0 . 3° per frame with an exposure time of 3 s and a crystal-to-detector distance of 600 mm . A total of 600 frames were recorded on different positions from one crystal ( 0 . 3 x 0 . 1 x 0 . 1 mm3 ) . All diffraction data were processed with HKL2000 [55] . The GNNV-LP crystals belong to a monoclinic C2 space group with unit-cell dimensions of a = 477 Å , b = 422 Å , c = 337 Å , and β = 134° . The diffraction data of GNNV-LP crystals contained 499 , 184 reflections and was 98% complete at a resolution range from 50 to 3 . 6 Å . To help initial phase determination by ab initio phasing [56] , the very low-resolution data of the GNNV-LP crystals up to 266 Å were measured , and only a few reflections were not measured in the region of very low resolution ( > 100 Å ) . The initial crystallizations of the truncated P-domain , delta-P-domain mutant and N-ARM deletion mutant proteins were performed with similar approaches as for the GNNV-LP . The initial crystallization conditions of the truncated P-domain , the delta-P-domain mutant and the N-ARM deletion mutant were 0 . 2 M Ca acetate , 0 . 1 M MES ( pH 6 . 5 ) , 10% ( w/v ) PEG8000; 0 . 2 M MgCl2 , 0 . 1 M HEPES−Na ( pH 7 . 5 ) , 30% ( w/v ) PEG400; and 0 . 1 M NaCl , 0 . 1M lithium sulfate , 0 . 1 M MES ( pH 6 . 5 ) , 30% ( w/v ) PEG400 , respectively . All crystals appeared within one week . X-ray diffraction data of the truncated P-domain and delta-P-domain mutants were collected on BL15A1 with a CCD detector ( MX300-HE , Rayonix ) of NSRRC in Taiwan at a wavelength of 1 . 0 Å . The N-ARM deletion mutant was collected on BL44XU at SPring-8 ( Harima , Japan ) with a CCD detector ( MX300-HE , Rayonix ) at a wavelength of 0 . 9 Å . The diffraction data were processed with HKL2000 [55] . All data processing statistics are shown in Table 1 . The initial phases of the T = 3 GNNV-LP were determined by the ab initio method using icosahedral non-crystallographic symmetry ( NCS ) averaging [56] . Self-rotation functions of κ = 72° , 120° and 180° hemispheres were analyzed with Molrep [57] to confirm the icosahedral symmetries of GNNV-LP crystals and to determine the orientation of the icosahedral symmetry . There were two T = 3 GNNV-LP particles in the monoclinic unit cell with one two-fold NCS axis of the virus particle coinciding with the crystallographic two-fold axis . The asymmetric unit contained half of the particle or 30 copies of the icosahedrally-related trimeric CPs . The spherical-shell model with uniform density was used as the starting model . The inner and outer radii of 119 and 159 Å , respectively , were chosen as initial parameters of the model [56] . For the ab initio method , which used NCS-averaging ( NCSA ) with phase extension , a proper mask was necessary for dividing two regions: the protein region to be NCS-averaged and the solvent region to be flattened . The initial mask for NCSA and solvent flattening was created from the atomic structure of T = 3 FHV ( PDB ID: 4FSJ ) with a large mask-radius of 11~13 Å around each atoms ( S1A Fig ) . The initial NCS operators for averaging were derived from the self-rotation function . In a basic NCSA cycle between dual spaces , 30-fold NCSA and solvent flattening were applied in real space followed by phase combination with the Rayment weighting [58] in reciprocal space . In most of the procedure , programs from RAVE [59] and CCP4 [60] were used . After more than one hundred cycles of iteration at 25 Å resolution , the phase extension was performed from 25 Å to 3 . 7 Å with 50 iterations in one reciprocal lattice step ( ≈ 1/a ) ( This process is referred as “procedure” hereafter ) . During cycles of iterations , the R factor and correlation coefficient comparing Fobs and Fcalc were monitored . The interpretable electron density map was successfully obtained ( S1A , S1B and S1C Fig ) by this procedure . To improve the electron density , the mask was updated based on the resultant map . The NCS operators were refined from the orientation of the icosahedral symmetry to give the highest correlation coefficient . The procedures were started from the spherical-shell uniform density model with the updated mask and NCS operators . The best values of the R factor and the correlation coefficient appeared to be 0 . 20 and 0 . 92 , respectively , at ~6 Å resolution . The overall values of these calculations are given in Table 1 and the progress of the phase extension is shown in S1D Fig The enantiomorph of the phase set was checked by the electron density of the helical structural elements ( S1C Fig ) . In the last cycle of phase improvement , DM [61] was used for NCSA with refinement of NCS operators , and resolution was extended to 3 . 1 Å . Although diffraction data higher than 3 . 6 Å resolution was of poor quality , the electron density map calculated with the phases extended to 3 . 1 Å resolution gave the result better than the map calculated with the data extended to 3 . 6 Å . Quality of the final electron density maps was good enough for an atomic model building except the P-domain region of T = 3 GNNV-LP . We suspected that the P-domain of T = 3 GNNV-LP did not follow the strict icosahedral symmetry . However , utilizing of DM with various trials , including the individual mask and NCS operator around the P-domain , did not significantly improve the density map around the P-domain of T = 3 GNNV-LP . The initial model building of GNNV-LP was performed by Cα-tracing with Coot [62] from the DM maps ( 3 . 1 Å ) . The complete models of the S-domain and the linker region were subsequently built up based on the amino-acid sequence of the GNNV CP manually ( S1C Fig ) . Structure refinement of the T = 3 GNNV-LP was performed using REFMAC5 [63] with icosahedral NCS restraints . During the refinement , the additional restraint was required for the coordinates of the P-domain to avoid the divergence due to the poor electron density . The PROSMART [64] with the high-resolution truncated P-domain model , which was subsequently determined , was used as the initial model and restraint reference . The coordinates were refined to a crystallographic Rcryst of 0 . 257 and Rfree of 0 . 295 at 3 . 6 Å resolution . Analysis of the Ramachandran plot showed that 97% of the main-chain dihedral angles were in preferred regions; 3% was in the allowed regions; and none were in the outlier regions using MolProbity [65] . The results of the GNNV-LP structure determination are summarized in Table 1 . For structure determination of the T = 1 delta-P-domain mutant , the coordinate of the S-domain from the T = 3 GNNV-LP structure was used as the molecular-replacement initial model . The icosahedral 20-fold NCSA phase extension by DM [59] was used for phase improvement . The structural model of the T = 1 delta-P-domain mutant was refined with the NCS restraints using REFMAC5 [63] and manual revision using Coot [62] to fit the DM map . The electron density of the P-domain , which was cut out from the density map of the T = 3 GNNV-LP , was used as the search model for molecular replacement of the truncated P-domain crystal . After phase improvement by the multiple crystal averaging with self-made programs together with MAPROT [66] , the model building was automatically performed with ARP/wARP [67] . Structure refinement of the truncated P-domain was performed with PHENIX [68] . Resolution of the data for the N-ARM deletion mutant was rather modest at 7 Å . The rough structure of the N-ARM deletion mutant was obtained as a reasonable MR solution using PHASER [69] . The crystal packing and self-rotation function analyses are shown in S3A and S3B Fig . In the figures , one MR solution is shown , in which two particles are located at ( 1/3 , 2/3 , 1/4 ) and ( 2/3 , 1/3 , 3/4 ) in one unit cell . All graphics for the molecular structure were produced with the PyMOL ( http://www . pymol . org/ ) . The GF-1 grouper cell line [70] was cultured in antibiotic-free Leibovitz’s L-15 medium ( Gibco ) supplemented with 5% ( v/v ) fetal bovine serum ( FBS ) at 28°C . GNNV was isolated from naturally infected groupers ( Epinephelus coioides ) collected in Taiwan . The isolated virus was propagated in GF-1 cells and collected when 90% of the cells displayed a cytopathic effect ( CPE ) . GF-1 cells were seeded in 12-well plates at a density of 1 x 105 per well in 2 ml L-15 medium supplemented with 5% ( v/v ) FBS , and cultured to 80−90% confluence . For infection , GF-1 cells were washed with PBS three times and subsequently infected with GNNV at a titer of 104 TCID50/ml in serum-free medium , 4% ( w/v ) PEG3000- and 4% ( w/v ) PEG8000-containing serum-free medium , respectively . After incubation with the virus for 30 min , the cells were washed thrice with PBS and L-15 medium ( 2 ml ) supplemented with 1% ( v/v ) FBS in the presence or absence of PEG3000 and PEG8000 was added . After infection , GF-1 cells were washed with PBS , and RNA was extracted with the TRIzol reagent ( Invitrogen ) . Reverse transcription and real-time quantitative PCR were performed as previously described [71] . All data analyses were shown as mean ± SD of three independent experiments . Statistical analyses were assessed by one-way ANOVA with SPSS statistical software version 17 . 0 ( SPSS Inc . ) . P values < 0 . 05 , were considered statistically significant . Nucleotide sequences of PCR-amplified fragments of OSGNNV RNA2 from have been deposited in the GenBank nucleotide database under the accession code KT071606 . Atomic coordinates and diffraction data of the T = 3 GNNV-LP , the truncated P-domain ( 214−338 ) and the T = 1 delta-P-domain mutant ( 35−217 ) have been deposited at the Protein Data Bank ( PDB ) with accession codes 4WIZ , 4RFU and 4RFT , respectively .
|
Betanodaviruses belong to the family Nodaviridae and cause the mortality of numerous larval-stage fish species . Here we report protein crystal structures of a piscine betanodavirus , the Grouper nervous necrosis virus ( GNNV ) , in four different forms . Highlights are two structural features that contribute to the viral molecular mechanisms of the T = 3 and T = 1 capsid assembly: a calcium-associated protrusion domain and a functional arginine-rich motif . These results also shed insights into the structural basis for evolutionary lineage of the family Nodaviridae .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[] |
2015
|
Crystal Structures of a Piscine Betanodavirus: Mechanisms of Capsid Assembly and Viral Infection
|
Clathrin-mediated endocytosis ( CME ) and activity-dependent bulk endocytosis ( ADBE ) are two predominant forms of synaptic vesicle ( SV ) endocytosis , elicited by moderate and strong stimuli , respectively . They are tightly coupled with exocytosis for sustained neurotransmission . However , the underlying mechanisms are ill defined . We previously reported that the Flower ( Fwe ) Ca2+ channel present in SVs is incorporated into the periactive zone upon SV fusion , where it triggers CME , thus coupling exocytosis to CME . Here , we show that Fwe also promotes ADBE . Intriguingly , the effects of Fwe on CME and ADBE depend on the strength of the stimulus . Upon mild stimulation , Fwe controls CME independently of Ca2+ channeling . However , upon strong stimulation , Fwe triggers a Ca2+ influx that initiates ADBE . Moreover , knockout of rodent fwe in cultured rat hippocampal neurons impairs but does not completely abolish CME , similar to the loss of Drosophila fwe at the neuromuscular junction , suggesting that Fwe plays a regulatory role in regulating CME across species . In addition , the function of Fwe in ADBE is conserved at mammalian central synapses . Hence , Fwe exerts different effects in response to different stimulus strengths to control two major modes of endocytosis .
In the presynaptic terminal , continuous release of synaptic vesicles ( SVs ) results in vesicle pool depletion , plasma membrane expansion , and SV protein overloading at the release site [1] . Endocytosis is therefore tightly coupled to exocytosis [2] . Among the different modes of SV endocytosis , Clathrin-mediated endocytosis ( CME ) and activity-dependent bulk endocytosis ( ADBE ) are well characterized [3 , 4] . In mild stimulation paradigms , CME is the prevalent mode of retrieving exocytic SVs in the form of a single SV [5 , 6] . In intense stimulation paradigms , however , ADBE promotes the uptake of large pieces of fused membranes in bulk endosomes or cisternae [7 , 8] . Small SVs are then formed from these membranous structures . SV exocytosis is a prerequisite for CME and ADBE initiation [2] , indicating that specific SV cargoes , an exocytic protein complex , or both are needed to trigger both modes of endocytosis . Indeed , Synaptotagmin , the SV Ca2+ sensor for exocytosis [9] , and components of the soluble NSF attachment protein receptor ( SNARE ) complex play crucial roles in CME [10–14] . Moreover , recent studies have identified vesicle-associated membrane protein 4 ( VAMP4 ) , a vesicle-associated SNARE ( v-SNARE ) protein , as a selective SV cargo for ADBE . Interestingly , VAMP4 is responsible for the formation of ADBE as well [15] . Thus , SV proteins encode components that retrieve SV membrane in newly formed vesicles as well as coordinate the nature of the formation of SV endocytosis . An increased local Ca2+ concentration in the presynaptic terminal is necessary not only for exocytosis but also for endocytosis [16–20] . Moreover , the Calmodulin/Calcineurin complex was proposed to function as a Ca2+ sensor for CME and ADBE [19 , 21 , 22] . Therefore , Ca2+/Calcineurin likely acts as a universal signal that elicits most forms of the SV retrieval [2] . At the rat Calyx of Held , in addition to a role in triggering exocytosis , a high , transient Ca2+ influx via a voltage-gated Ca2+ channel ( VGCC ) triggers CME [18 , 23] . However , the Ca2+ channel for triggering ADBE is unknown . Our previous genetic screen in flies identified Flower ( Fwe ) , an evolutionarily conserved transmembrane protein [24] . Fwe forms a Ca2+-permeable channel when expressed in heterologous cells or when incorporated into proteoliposomes . This protein localizes to SVs , and , upon SV release , Fwe is transferred to the periactive zone , where it triggers CME , thereby coupling exocytosis to CME . In the present study , we show that Fwe initiates ADBE as well . Intriguingly , the effects of Fwe on CME and ADBE depend on the strength of the stimulus . We found that the function of Fwe for regulating CME does not involve Ca2+ channeling . Instead , upon intense stimulation , Fwe triggers a Ca2+ influx that elicits ADBE . Lastly , when we removed ratFwe in cultured rat hippocampal neurons through clustered regularly interspaced short palindromic repeats ( CRISPR ) / CRISPR-associated protein 9 ( Cas9 ) technology , CME is impaired but not completely blocked , similar to the defect caused by the Drosophila fwe mutation . Furthermore , our data reveal that RatFwe is also involved in the induction of ADBE at mammalian central synapses . In summary , the Fwe channel exerts two different functions in response to two different stimuli that govern distinct modes of SV retrieval , thereby coupling exocytosis to endocytosis .
We previously reported that the Fwe Ca2+ channel promotes CME in the synaptic boutons of Drosophila neuromuscular junctions ( NMJs ) [24] . To investigate whether the Ca2+ influx via Fwe plays a direct role in CME , we utilized the FweE79Q mutant whose channel activity is severely impaired [24] and assessed the ability to rescue CME defects associated with fwe mutants . We expressed UAS-flag-fwe-HA and UAS-flag-fweE79Q-HA with nSyb-GAL4 , a pan-neuron GAL4 driver , in a strong loss of fwe background ( fweDB25/fweDB56 ) ( S1A and S1B Fig ) [24] . α-HA antibody staining was used to examine the distribution and expression of Fwe proteins in boutons . α-horse radish peroxidase ( HRP ) antibody staining labels the insect neuronal membranes , thereby outlining presynaptic compartments [25] . Both the wild-type Fwe and FweE79Q proteins are evenly distributed in boutons ( S1A and S1B Fig ) . We then introduced a genomic HA-tagged fwe transgene to estimate the relative expression levels of UAS transgenes versus endogenous Fwe protein ( S1C–S1F Fig ) . The proteins are expressed at ~50% of the endogenous Fwe protein level in type Ib NMJ boutons ( S1L–S1N Fig ) , whereas their expressions in type Is NMJ boutons correspond to ~80% of the endogenous Fwe protein . To estimate the efficacy of CME , we performed the FM1-43 dye uptake assay with moderate stimuli , i . e . , 1-min 90 mM K+/0 . 5 mM Ca2+ and 10-min 60 mM K+/1 mM Ca2+ stimulations . The experimental paradigm is shown in S2A Fig . Both stimulation paradigms significantly elicit dye uptake in wild-type control larvae when compared to a resting paradigm ( 10-min incubation in 5 mM K+/0 mM Ca2+ solution ) ( S2B–S2E Fig ) . We then performed a transmission electron microscopy ( TEM ) assay to assess the formation of bulk cisternae , a hallmark of ADBE [2 , 8] . No bulk cisternae were induced under these conditions ( S2F–S2I Fig ) , showing that the strength of these stimuli is mild , which predominantly promotes CME . Upon 1-min 90 mM K+/0 . 5 mM Ca2+ stimulation , loss of fwe impairs FM1-43 dye uptake ( Fig 1A , 1B and 1H ) . It is possible that either a defect in SV endocytosis or exocytosis would reduce FM1-43 dye uptake in this case . To test the role of Fwe in SV exocytosis , we performed the FM1-43 dye loading/unloading assay . The experimental paradigm is indicated in S3A Fig . Both the control and fwe mutant boutons were subjected to 5-min 90 mM K+/0 . 5 mM Ca2+ stimulation , which labels the SV pool with FM1-43 dye ( S3B and S3D Fig ) . Subsequently , 1-min stimulation with the same solution releases the SVs and unloads the dye from SVs ( S3C and S3E Fig ) . The strength of SV exocytosis is correlated with the FM1-43 dye unloading efficiency ( [ ( Fload—Funload ) / Fload] ) . While the dye loading is significantly reduced when fwe is lost ( S3B , S3D and S3F Fig ) , the dye unloading efficiencies of controls and fwe mutants are comparable ( S3G Fig ) , indicating that Fwe plays a marginal or no role in SV exocytosis . Hence , the FM1-43 dye uptake deficit associated with fwe mutants mainly results from a defect in CME . Reduced FM1-43 dye uptake in fwe mutants is completely rescued by the reintroduction of 50% Fwe protein ( Fig 1C and 1H ) . However , a similar rescue was also observed when 50% FweE79Q is present ( Fig 1D and 1H ) . Although the channel function of FweE79Q is mostly absent when analyzed in fly salivary glands [24] , the remaining channel activity in this mutant might be sufficient to promote CME if enough proteins are present in the boutons . We therefore determined the minimal Fwe level required for CME to verify the channel function of Fwe . After surveying numerous GAL4 lines , elav-GAL4 and nSyb ( w ) -GAL4 were found to drive the expression of the transgenes at approximately 0% and 4% of the endogenous Fwe level , respectively ( S1G–S1J , S1M and S1N Fig ) . As shown in Fig 1E , 1F and 1H , boutons expressing 10% or 4% Fwe can take up FM1-43 dye efficiently . We noted that the dye uptake with 4% Fwe expression is marginally reduced when compared to other Fwe-expressing larvae , indicating that this level of Fwe expression is near the minimal level for efficient CME . Next , we assessed CME in 4% FweE79Q-expressing boutons ( S1J , S1K , S1M and S1N Fig ) . Intriguingly , the efficiency of the dye uptake is not different between 4% Fwe and 4% Fwe E79Q-expressing boutons ( Fig 1F–1H ) , suggesting that CME can occur despite the lack of a significant Ca2+ influx via Fwe . We previously showed that loss of fwe results in a reduced SV number and enlarged SVs [24] . To examine the changes in SV ultrastructure , we performed TEM . The wild-type control bouton under the resting condition contains numerous SVs ( Fig 1I ) , whereas the number of SVs is decreased upon loss of fwe ( Fig 1J and 1Q ) . This low SV number worsens following 10-min 60 mM K+/1 mM Ca2+ stimulation ( Fig 1M , 1N and 1Q ) . Either 4% Fwe or 4% FweE79Q expression rescues this SV loss ( Fig 1K , 1L and 1O–1Q ) . In addition , enlarged SV sizes associated with fwe mutants are normalized under both expression conditions ( Fig 1R ) . Hence , these data indicate that Fwe triggers CME independent of Ca2+ channeling . Upon mild stimulation , CME retrieves the membrane that corresponds in size to an SV . In contrast , in response to intense stimulation , ADBE takes up large quantities of fused SVs from the plasma membrane to form bulk cisternae . It has been shown that both SV exocytosis and intracellular Ca2+ elevation are essential for ADBE to proceed around the periactive zones [7 , 18 , 26] . This raises the possibility that Fwe may play a role in ADBE . High K+ and Ca2+-containing solutions have been widely used to elicit ADBE at several different synapses , including fly NMJ boutons [8 , 27–30] . To assess the role of Fwe in ADBE , we applied 10-min 90 mM K+/2 mM Ca2+ stimulation to induce ADBE and examine the formation of bulk cisternae using a TEM assay . The TEM image of control boutons reveals numerous cisternae ( >80 nm in diameter , red arrows ) elicited by this stimulation paradigm ( Fig 2A , 2B and 2G ) . These processes , however , are dramatically suppressed by loss of fwe ( Fig 2C , 2D and 2G ) . In unstimulated conditions , bulk cisternae are also less abundant in fwe mutants than in controls ( Fig 2G ) . This ADBE defect indeed results from the fwe mutation , as 50% Fwe expression rescues this ADBE phenotype ( Fig 2E–2G ) . Interestingly , the average size of the few bulk cisternae observed in fwe mutants is comparable to that observed in control boutons ( Fig 2H ) , suggesting that Fwe acts at the initiation step of ADBE rather than during a late membrane invagination process . Furthermore , following high K+ stimulation , the accumulation of early endocytic intermediates was observed around the periactive zone in fwe mutant boutons ( Fig 2D–2D1 and 2I , yellow arrows ) when compared to wild-type controls and 50% Fwe-rescued larvae ( Fig 2B–2B1 , 2F–2F1 and 2I ) . Since optimal SV exocytosis is shown as a prerequisite for triggering ADBE [7 , 26] , we therefore estimated the total SV area per bouton area under the resting condition . No difference between controls and fwe mutants was found ( Fig 2J ) , showing that the ADBE defect associated with fwe mutants is not due to insufficient supply of exocytic SV membrane upon stimulation . Moreover , following 90 mM K+/2 mM Ca2+ stimulation , the strength of SV exocytosis determined by the FM1-43 dye loading/unloading assay is comparable between controls and fwe mutants ( S3H–S3N Fig ) . Collectively , these results reveal that Fwe is responsible for initiating ADBE during intense activity stimulation . Acute inactivation of the components involved in CME , such as Clathrin , AP180 , and Dynamin , elicits bulk membrane invaginations [31–34] , suggesting that CME suppresses ADBE or that ADBE is the result of membrane expansions . To assess the role of Fwe in this process , we treated larvae with 200 μM chlorpromazine to inhibit Clathrin coat assembly [31] , followed by 10-min 90 mM K+/ 2 mM Ca2+ stimulation in the presence of FM1-43 dye . As shown in Fig 2K–2N , large membranous invaginations enriched with FM1-43 dye were detected in the controls . In contrast , these structures decrease upon loss of fwe . In summary , these results reinforce the functional importance of Fwe in ADBE . To assess whether Fwe mediates an intracellular Ca2+ increase to initiate ADBE upon intense activity stimulation , we expressed the lexAop2 transgene of the fast-decay version of the genetically encoded Ca2+ indicator GCaMP6 , GCaMP6f [35] , in the presynaptic terminals with vglut-lexA , a glutamatergic neuron driver . We have shown previously that fwe mutant boutons display low resting Ca2+ levels [24] . Similarly , decreased GCaMP6f fluorescence was observed upon loss of fwe ( Fig 3A , 3B and 3E , white arrows ) . This indicates a reduction in the resting Ca2+ levels , as the expression level of GCaMP6f in boutons is higher in fwe mutants than in controls ( S4A , S4B and S4E Fig ) . Next , we stimulated boutons with 90 mM K+/2 mM Ca2+ solution for 10 min , which elicits ADBE ( Fig 2A and 2B ) , and measured GCaMP6f fluorescence in the 6th and 10th min . In controls , the intracellular Ca2+ concentrations in response to stimuli are substantially increased ( Fig 3A–3A3 and 3F ) , whereas loss of fwe significantly impedes these Ca2+ elevations ( Fig 3B–3B3 and 3F ) . Hence , Fwe sustains presynaptic Ca2+ levels upon strong stimulation . To assess the role of Fwe-derived Ca2+ influx in regulating presynaptic Ca2+ level , we traced GCaMP6f fluorescence in 4% Fwe- and 4% FweE79Q-expressing boutons under the resting and high K+ stimulation conditions . In both conditions , the levels of GCaMP6f are expressed similarly to that in controls ( S4C–S4E Fig ) . We found that 4% Fwe but not 4% FweE79Q expression restores normal resting Ca2+ levels in fwe mutants ( Fig 3C–3E ) . Furthermore , a defect in high K+-induced Ca2+ elevation associated with fwe mutants is partially reversed by 4% Fwe expression ( Fig 3C–3C3 and 3F ) . In contrast , 4% FweE79Q expression fails to rescue this Ca2+ defect ( Fig 3D–3D3 and 3F ) . However , when we examined the presynaptic Ca2+ changes following 1-min 90 mM K+/0 . 5 mM Ca2+ stimulation , which prevalently elicits CME ( S2 Fig ) , the Ca2+ increases among all genotypes are quite similar ( Fig 3G ) . Hence , these results indicate that Fwe triggers a Ca2+ influx specifically in response to strong stimuli . To verify this stimulus-dependent Ca2+ channeling of Fwe , we measured the Ca2+ increase evoked by electric stimuli in wild-type and fwe mutant boutons . To this end , we expressed UAS-GCaMP6f with nSyb-GAL4 . The levels of GCaMP6f in control and fwe mutant boutons are comparable ( S4F , S4G and S4K Fig ) . The stimulation paradigms are shown in the top panel of Fig 3H . Upon 10–40 Hz train stimuli , intracellular Ca2+ increase in controls correlates with the stimulus strength ( S5A–S5A4 Fig and Fig 3H , white circle ) . In fwe mutants ( S5B–S5B4 Fig and Fig 3H , red circle ) , the Ca2+ increase at 10 Hz is slightly higher than that in controls , and the Ca2+ increase at 20 Hz is similar to that in controls . In contrast , at 40 Hz , loss of fwe significantly impairs the evoked Ca2+ increase ( Fig 3H and 3I ) . Moreover , 50% Fwe expression normalizes this deficit ( S4H , S4K and S5C–S5C4 Figs and Fig 3H and 3I , black circle ) , whereas a partial restoration by 50% FweE79Q expression was observed ( S4I , S4K and S5D–S5D4 Figs and Fig 3H and 3I , orange circle ) . Furthermore , we observed similar effects on rescuing low resting Ca2+ levels associated with fwe mutants ( S5F Fig ) . These results support the finding that Fwe does not mediate a Ca2+ influx under a moderate stimulation condition that predominantly induces CME . Instead , it conducts a Ca2+ influx when neurons undergo intense stimulation . To rule out that the defect in evoked Ca2+ increase may be attributed to slow CME associated with fwe mutants , we applied the same stimulation protocol to dap160 mutants , which exhibit a similar CME defect [36–38] . As shown in S5E–S5E4 Fig and Fig 3H and 3I ( green circle ) , in dap160 mutant boutons , the Ca2+ concentrations at 40 Hz are elevated to wild-type levels , although the Ca2+ increase at 10 and 20 Hz is higher than that observed in controls and fwe mutants . In addition , the fluorescence level but not the expression level of GCaMP6f under the resting condition is reduced upon loss of dap160 ( S4J , S4K , S5E and S5F Figs ) , suggesting that dap160 mutants display low resting Ca2+ levels . Therefore , our results argue that a defective CME does not account for presynaptic Ca2+ dysregulation in fwe mutants . In addition , we found no evidence for the changes in the distribution and expression of Cacophony , the major VGCC located at the active zone ( S6 Fig ) . The above-mentioned results prompted investigations into the role of Fwe-driven Ca2+ influx in ADBE . We showed previously that 4% Fwe is sufficient for CME . We therefore addressed if this level of Fwe is sufficient to promote ADBE . When boutons are expressed with 10% or 4% Fwe , ADBE elicited by 10-min 90 mM K+/2 mM Ca2+ stimulation efficiently produces bulk cisternae ( Fig 4A , 4B , 4D , 4E and 4K ) . Thus , a partial Ca2+ influx by 4% Fwe ( Fig 3F ) is sufficient for initiating ADBE . Consistently , 50% FweE79Q , which induces a fractional Ca2+ influx ( Fig 3H and 3I ) , robustly triggers ADBE after high K+ stimulation ( S7 Fig ) . However , high K+-induced bulk cisternae are significantly reduced in 4% FweE79Q-rescued larvae ( Fig 4G , 4H and 4K ) . Notably , the number of high K+-induced bulk cisternae between 4% FweE79Q-rescued and fwe mutant larvae is comparable ( fwe mutant larvae , 1 . 06 ± 0 . 4 , n = 22 , versus 4% FweE79Q-rescued larvae , 1 . 66 ± 0 . 42 , n = 26 , [Student’s t test , p = 0 . 31] ) . Similar to loss of fwe , there is an increase in the level of endocytic intermediates formed around the periactive zone in 4% FweE79Q-rescued boutons after high K+ stimulation when compared to 4% Fwe-rescued boutons ( Fig 4L ) , thus supporting an important role of Ca2+ influx via Fwe in ADBE . At rest , the total SV membrane area per bouton area is also comparable between 4% Fwe- and 4% FweE79Q-rescued boutons ( Fig 4M ) . Therefore , both expression conditions yield equal SV membranes available for SV exocytosis . These data suggest that Fwe triggers ADBE mainly through fluxing Ca2+ . Furthermore , after chlorpromazine treatment , bulk membrane invaginations are less abundant in 4% FweE79Q-rescued boutons when compared to 4% Fwe-rescued boutons , also documenting a role of Fwe-derived Ca2+ influx in chlorpromazine-induced bulk membrane invagination ( Fig 4N–4P ) . If a suboptimal Ca2+ level in the presynaptic terminals results in impaired ADBE phenotype in 4% FweE79Q-rescued larvae , then we expected that increasing overall intracellular Ca2+ concentrations either via the remaining channel activity of FweE79Q or the other Ca2+ channels during stimulation might compensate for this low intracellular Ca2+ and rescue the defective ADBE . We therefore raised the Ca2+ concentration from 2 mM to 5 mM in our 90 mM K+ stimulation solution . When we applied a 10-min 90 mM K+/5 mM Ca2+ stimulation to 4% FweE79Q-rescued boutons , the bulk cisternae number is significantly rescued ( Fig 4I and 4K ) . Similarly , 5mM Ca2+ also rescues the ADBE deficit associated with fwe mutants ( Fig 4J and 4K ) . In contrast , this treatment does not increase the number of bulk cisternae further in 4% or 10% Fwe-rescued larvae when compared to the 10-min 90 mM K+/2 mM Ca2+ stimulation condition ( Fig 4C , 4F and 4K ) . Hence , this rescue effect might be due to increased Ca2+ levels rather than enhanced ADBE in the presynaptic compartments . These data further support the role of Ca2+ influx via Fwe in triggering ADBE during intense activity stimulation . If Ca2+ influx via Fwe triggers ADBE , one would anticipate that reducing channel activity will abolish ADBE . La3+ is a potent blocker of some Ca2+-permeable channels [39 , 40] . It may therefore inhibit the Fwe channel activity . We previously showed that heterologous expression of Fwe results in Ca2+ uptake by Drosophila salivary gland cells [24] . To determine the effect of La3+ on the Ca2+ conductance of Fwe , we applied 100 μM La3+ to the glands that carry a GAL4 driver only or overexpress Fwe and performed Ca2+ imaging . The cells were loaded with Fluo-4 AM Ca2+ indicator and bathed in 100 μM extracellular Ca2+ solution . Fwe-overexpressing cells display a slow but significant Ca2+ uptake over a 1-h period when compared to controls ( Fig 5A , 5B and 5E ) , consistent with our previous observations [24] . However , application of 100 μM La3+ solution nearly abolishes the Ca2+ influx mediated by Fwe ( Fig 5D–5E ) , although a mild suppression was observed in control cells as well ( Fig 5C and 5E ) . These results indicate that , similar to other Ca2+-permeable channels , the channel pore region of Fwe has a high affinity for La3+ and is blocked by La3+ . Next , we assessed the impact of 100 μM La3+ on ADBE . Since La3+ impedes Ca2+ permeability of VGCCs [39 , 40] , and VGCC-triggered exocytosis is essential for ADBE initiation [7 , 26] , we first tested whether treatment with100 μM La3+ solution affects SV exocytosis . At 0 . 2 Hz in 1 mM Ca2+ , 100 μM La3+ reduces the amplitude of excitatory junction potentials ( EJPs ) by ~60%–80% when compared to untreated ones , indicating that La3+ blocks a Ca2+ influx mediated by VGCCs . In contrast , 10 μM La3+ does not significantly influence 0 . 2 Hz-elicited EJPs in controls ( Fig 5F , white column ) . As this low La3+ concentration might not block the Fwe-derived Ca2+ influx effectively , we used 10% Fwe-rescued larvae , which are more sensitive to 10 μM La3+ ( Fig 5B ) . Under this expression condition , the application of 10 μM La3+ does not affect the EJP responses at 0 . 2 Hz ( Fig 5F , green column ) but largely reduces ADBE upon high K+ stimulation ( Fig 5G and 5I ) . Notably , the number of bulk cisternae induced under La3+ treatment is almost identical to that observed in fwe mutants after high K+-stimulation ( Fig 5I ) , suggesting that La3+ suppresses ADBE by selectively inhibiting the channel activity of Fwe . In support of this , in fwe mutants , 10 μM La3+ does not alter the 0 . 2 Hz-evoked EJP amplitude ( Fig 5F , red column ) or the level of high K+-induced bulk cisternae ( Fig 5H and 5I ) . Overall , these data indicate a role of Fwe in Ca2+-mediated ADBE . In summary , Fwe governs two major modes of SV endocytosis to permit consecutive rounds of exocytosis of neurotransmitters following distinct activity stimuli . Fwe homologs are found in most eukaryotes [24] , but their role in SV endocytosis in vertebrates has not been established . The mouse Fwe ( mFwe ) gene can generate at least six alternative mRNA splicing isoforms [41] , producing five different mFwe isoforms ( S8 Fig ) . The mFwe isoform 2 ( mFwe2 ) is the most similar to Drosophila Fwe . Moreover , the mFwe2 and rat Fwe isoform 2 ( ratFwe2 ) share ~99% amino acid identity ( 170/172 ) . In adult rat brain , ratFwe2 mRNA is widely expressed ( Fig 6A ) . In the lysates of mouse neuroblastoma Neuro 2a ( n2a ) cells , our antisera against the C-termini of both mFwe2 and ratFwe2 ( α-m/ratFwe2 ) recognize a ~18 kDa protein band , corresponding to the predicted molecular weight of mFwe2 ( Fig 6B ) . This signal is significantly decreased when mFwe2 is knocked down by mFwe-microRNAi ( miRNAi ) ( Fig 6B ) , showing antibody specificity . mFwe2 is expressed in postnatal as well as adult mouse brains ( Fig 6B ) . Similarly , ratFwe2 was detected in rat brain and cultured rat hippocampal neurons ( Fig 6C ) . To determine if ratFwe2 is enriched in SVs , we purified SVs from adult rat brain using a series of centrifugations [42] . As shown in Fig 6D , ratFwe2 was specifically detected in the SV ( Lysate pellet 2 [LP2] ) fraction , marked by the presence of Synaptophysin ( Syp ) , an abundant SV protein . ratFwe2 is also present in the SV fractions of adult rat brain separated with sucrose gradients ( Fig 6E ) . To assess the subcellular localization of ratFwe2 , we performed immunostaining in cultured neurons . Although staining with our antisera can visualize the expression of ratFwe2 in the cell bodies ( S9B Fig ) , we failed to obtain specific staining in the presynaptic terminals . We therefore expressed HA-tagged mFwe2 in ratFwe knockout neurons ( see below for details ) and determined the SV localization of mFwe2-HA using α-HA staining . As shown in Fig 6F and 6G , mFwe2-HA protein is enriched in the presynaptic terminals and largely colocalized with Syp . The biochemical data combined with the in vivo localization data provide compelling evidence that ratFwe2 is associated with SV proteins , similar to Drosophila Fwe . To determine whether rodent Fwe2 functions equivalently to Drosophila Fwe , we expressed UAS-flag-mFwe2-HA transgene in fwe mutants using nSyb ( w ) -GAL4 . Overexpressed mFwe2 is localized to SVs in the boutons ( Fig 6H–6H2 ) and rescues the FM1-43 dye uptake defect ( Fig 6I–6K ) , as well as the reduced number of SVs ( Fig 6L ) in fwe mutants , showing that mFwe2 can promote CME in flies . Furthermore , the expression of mFwe2 corrects the ADBE deficit caused by loss of fwe ( Fig 6M and 6N ) . Hence , mFwe2 promotes ADBE as well . We also observed that the early lethality of fwe mutant larvae is rescued by mFwe2 expression . Our results therefore suggest a conserved role of Fwe in SV endocytosis in mammals . To verify the role of ratFwe2 in SV endocytosis , we knocked out ratFwe in cultured rat hippocampal neurons using CRISPR/Cas9 technology [43] . We designed a specific guide RNA ( gRNA; m/ratFwe-gRNA ) that targets the first intron/second exon junction of both mFwe and ratFwe genes . To estimate the knockout efficiency , we transfected the gRNA construct into mouse neuroblastoma n2a cells and established a mFwe knockout n2a cell line . While mFwe2 was detected in normal n2a cells , it is lost in mFwe knockout n2a cells ( S9A Fig ) . At 14 days in vitro ( DIV ) , ratFwe2 is present in the Golgi apparatus of the cultured rat hippocampal neurons ( S9B–S9B3 Fig ) , consistent with the fact that ratFwe2 is an SV protein sorted from the Golgi . In neurons expressing Cas9 and m/ratFwe-gRNA , the expression of ratFwe2 in the Golgi apparatus is significantly diminished ( S9C and S9D Fig ) . Thus , this gRNA can also efficiently remove ratFwe2 in cultured neurons when Cas9 is present . To assess the efficacy of CME , we elicited exocytosis of Synaptophysin-pHluorin ( SypHy ) [5] by delivering 200 action potentials at 20 Hz and monitored its retrieval via SV endocytosis . It has been documented that this mild stimulation paradigm prevalently induces CME [12 , 15 , 44] . In control neurons ( Fig 7A , black line ) bathed at room temperature , repeated exocytosis in response to 20-Hz stimuli increases SypHy fluorescence , followed by a gradual fluorescence decay caused by the reacidification of SVs formed via CME . However , in ratFwe knockout neurons ( Fig 7A , red line ) , the decay rate of SypHy fluorescence is much slower ( Fig 7A and 7C ) . To verify whether this defect is specific to loss of ratFwe2 , we expressed mFwe2-HA in ratfwe knockout neurons . mFwe2-HA properly localizes in the Golgi ( S9E Fig ) as well as in SVs ( Fig 6F and 6G ) . This protein further normalizes the slow SypHy fluorescence decay ( Fig 7A and 7C , blue line ) . Recent studies have revealed distinct properties of SV endocytosis under physiological conditions [15 , 45 , 46] . We found similar results when these recordings were performed at physiological temperatures ( Fig 7B and 7D ) . A slow decay of SypHy fluorescence is possibly due to either impaired CME or inefficient SV reacidification or both . To distinguish these hypotheses , we performed an acidic quenching assay [47 , 48] . As shown in S10 Fig , upon perfusion of an acidic buffer , the newly recycled SVs in both control and ratFwe knockout neurons are efficiently acidified . Hence , our data suggest that ratFwe2 promotes CME at mammalian central synapses . To assess the role of ratFwe2 in ADBE , we performed a dextran dye uptake assay in the control and ratFwe knockout neurons . We triggered ADBE with a stimulation of 1 , 600 action potentials delivered at 80 Hz in the presence of 40 kDa tetramethylrhodamine ( TMR ) -dextran [15 , 49] . As shown in Fig 7E–7H , in the control axons marked with green fluorescent protein ( GFP ) expression , the presynaptic terminals filled with dextran dye ( red puncta ) were observed frequently . In contrast , removal of ratFwe significantly diminishes dextran dye uptake . This phenotype is specific to loss of ratFwe2 , as the reintroduction of mFwe2-HA corrects this dye uptake defect . Hence , ratFwe2 is indispensable for ADBE . In summary , Fwe promotes CME and ADBE in mammalian neurons , thereby coupling exocytosis to two major modes of endocytosis .
A tight coupling of exocytosis and endocytosis is critical for supporting continuous exocytosis of neurotransmitters . CME and ADBE are well-characterized forms of SV endocytosis triggered by moderate and strong nerve stimuli , respectively . However , how they are coupled with exocytosis under distinct stimulation paradigms remains less explored . Based on the present data , we propose a model as shown in Fig 7I . When presynaptic terminals are mildly stimulated , SV release leads to neurotransmitter release and the transfer of Fwe channel from SVs to the periactive zone where CME and ADBE occur actively [7 , 26 , 38] . Our data suggest that this channel does not supply Ca2+ for CME to proceed . However , intense activity promotes Fwe to elevate presynaptic Ca2+ levels near endocytic zones where ADBE is subsequently triggered . Thus , Fwe exerts different activities and properties in response to different stimuli to couple exocytosis to different modes of endocytosis . We previously concluded that Fwe-dependent Ca2+ influx triggers CME [24] . However , the current results suggest alternative explanations . First , the presynaptic Ca2+ concentrations elicited by moderate activity conditions , i . e . , 1-min 90 mM K+/0 . 5 mM Ca2+ or 20-s 10–20 Hz electric stimulation , are not dependent on Fwe ( Fig 3G and 3H ) . Second , expression of 4% FweE79Q , a condition that abolishes Ca2+ influx via Fwe ( Fig 3F ) , rescues the CME defects associated with fwe mutants , including decreased FM1-43 dye uptake , a reduced number of SVs , and enlarged SVs ( Fig 1 ) . Third , raising the presynaptic Ca2+ level has no beneficial impact on the reduced number of SVs observed in fwe mutants ( Fig 4J ) . These data are consistent with the observations that a Ca2+ influx dependent on VGCCs triggers CME at a mammalian synapse [18 , 23] . Hence , Fwe acts in parallel with or downstream to VGCC-mediated Ca2+ influx during CME . ADBE is triggered by intracellular Ca2+ elevation , which has been assumed to be driven by VGCCs that are located at the active zones [18 , 26] . However , our data strongly support a role for Fwe as an important Ca2+ channel for ADBE . First , following exocytosis , Fwe is enriched at the periactive zone where ADBE predominates [7 , 24 , 26] . Second , Fwe selectively supplies Ca2+ to the presynaptic compartment during intense activity stimulation ( Fig 3 ) , which is highly correlated with the rapid formation of ADBE upon stimulation [8 , 50] . Third , 4% FweE79Q expression , which induces very subtle or no Ca2+ upon strong stimulation , fails to rescue the ADBE defect associated with loss of fwe ( Figs 3 and 4 ) . Fourth , treatment with a low concentration of La3+ solution that specifically blocks the Ca2+ conductance of Fwe significantly abolishes ADBE ( Fig 5 ) . Lastly , the role of Fwe-derived Ca2+ influx in the initiation of ADBE mimics the effect of Ca2+ on ADBE at the rat Calyx of Held [7] . As loss of fwe does not completely eliminate ADBE , our results do not exclude the possibility that VGCC may function in parallel with Fwe to promote ADBE following intense stimulation . Interestingly , Ca2+ influx via Fwe does not control SV exocytosis during mild and intense stimulations ( S3 Fig ) . How do VGCC and Fwe selectively regulate SV exocytosis and ADBE , respectively ? One potential mechanism is that VGCC triggers a high , transient Ca2+ influx around the active zone that elicits SV exocytosis . In contrast , Fwe is activated at the periactive zone to create a spatially and temporally distinct Ca2+ microdomain . A selective failure to increase the presynaptic Ca2+ level during strong stimulation is evident upon loss of fwe . This pinpoints to an activity-dependent gating of the Fwe channel . Consistent with this finding , an increase in the level of Fwe in the plasma membrane does not lead to presynaptic Ca2+ elevation at the Calyx of Held when the presynaptic terminals are at rest or subject to mild stimulation [23] . However , we previously showed that , in shits terminals , blocking CME results in the accumulation of the Fwe channel in the plasma membrane , elevating Ca2+ levels [24] . It is possible that Dynamin is also involved in regulating the channel activity of Fwe or that the effects other than Fwe accumulation associated with shits mutants may affect intracellular Ca2+ handling [51 , 52] . Further investigation of how neuronal activity gates the channel function of Fwe should advance our knowledge on the activity-dependent exo–endo coupling . Although a proteomic analysis did not identify ratFwe2 in SVs purified from rat brain [53] , our biochemical analyses show that ratFwe2 is indeed associated with the membrane of SVs . Our data show that 4% of the total endogenous Fwe channels efficiently promotes CME and ADBE at the Drosophila NMJ . If a single SV needs at least one functional Fwe channel complex during exo–endo coupling , and one functional Fwe complex comprises at least four monomers , similar to VGCCs , transient receptor potential cation channel subfamily V members ( TRPV ) 5 and 6 , and calcium release-activated channel ( CRAC ) /Orai1 [24 , 40 , 54 , 55] , then we anticipate that each SV contains ~100 Fwe proteins ( 4 monomers × 25 ) . This suggests that Fwe is highly abundant on the SVs . It is unlikely that many SVs do not have the Fwe , as a 25-fold reduction of the protein is enough to ensure functional integrity during repetitive neurotransmission . Finally , our results for the SypHy and dextran uptake assays at mammalian central synapses indicate the functional conservation of the Fwe channel in promoting different modes of SV retrieval . In summary , the Fwe-mediated exo–endo coupling seems to be of broad importance for sustained synaptic transmission across species .
Most of the experiments used y w; FRT80B isogenized fly , which was used for the generation of the fweDB25 and fweDB56 mutations [56] as the controls . Larvae were reared in standard fly food or on grape juice agar covered with yeast paste at 22°C . The genotypes of flies used in the experiments are described below . For GCaMP6f imaging and immunostaining , the genotypes that carry vglut-lexA and lexAop2-GCaMP6f are as follows: The genotypes that carry nSyb-GAL4 and UAS-GCaMP6f are as follows: The following UAS transgenes were made in this study: For electrophysiology , the following were used: For Cac-EGFP expression , the following were used: Those used for Fluo-4 AM experiments in salivary glands are as follows: Those used for TEM analysis , FM1-43 dye uptake assay , immunostaining of HA , and Fwe are as follows: The pCasper4-genomic HA-fwe construct was constructed by inserting a HA sequence to the site after the translational start codon of fwe-RB in the context of the pCasper4-genomic fwe construct [24] . To obtain the pUAST-flag-mFwe2-HA construct , the fwe-RB fragment of the pUAST-flag-fwe-RB-HA construct [24] was replaced with the mFwe2 coding region , which was amplified from total mRNA of the adult mouse brain . P-element-mediated transgenesis was achieved by the standard procedure . The introduction of these genomic fwe transgenes to the fwe mutant background rescues the early lethality associated with fwe mutants , demonstrating that fused tags do not affect normal functions of Fwe . To generate the mFwe-miRNAi constructs , the sequences of miRNAs were designed according to Invitrogen’s RNAi Designer . mFwe-miRNAi-1 targets nucleotides 183–203 of the mFwe2 coding sequence ( forward oligomer: TGCTGAAGG CGTTCATGATCATCCACGTTTTGGCCACTGACTGACGTGGATGAATGAACGCCTT; reverse oligomer: CCTGAAGGCGTTCATTCATCCACGTCAGTCAGTGGCCAAAACGTGG ATGATCATGAACGCCTTC ) . mFwe-miRNAi-2 targets nucleotides 236–256 of the mFwe2 coding sequence ( forward oligomer: TGCTGTTG CAAACTCCACAAACTGGCGTTTTGGC CACTGACTGACGCCAGTTTGGAGTTTGCAA; reverse oligomer: CCTGTTGCAAACTCC AAACTGGCGTCAGTCAGTGGCCAAAACGCCAGTTTGTGGAGTTTGCAAC ) . Synthetic oligomers were annealed and subcloned to pcDNA 6 . 2-GW/EmGFPmiR vector ( Invitrogen ) . To generate the pSpCas9 ( BB ) -based plasmids , pSpCas9 ( BB ) -2A-tagRFP was constructed by replacing the GFP region of the pSpCas9 ( BB ) -2A-GFP plasmid ( addgene#48138 ) [43] with a TagRFP coding sequence . m/ratFwe-gRNA was designed to target the first intron/second exon junction of the mFwe and ratFwe ( forward oligomer: CACCGTTTGAAGCCTGTGCCATCTC; reverse oligomer: TTTGCTCTACCGTGTCCGAAG TTTG ) . Annealed synthetic oligomers were placed into the BbsI site of pSpCas9 ( BB ) -2A-GFP and pSpCas9 ( BB ) -2A-tagRFP to obtain pSpCas9 ( BB ) -m/ratFwe-gRNA-2A-GFP and pSpCas9 ( BB ) -m/ratFwe-gRNA-2A-tagRFP constructs . pSpCas9 ( BB ) -m/ratFwe-gRNA-2A-GFP-2A-mFwe2-HA and pSpCas9 ( BB ) -m/ratFwe-gRNA-2A-tagRFP-2A-mFwe2-HA were generated by inserting the DNA fragment of 2A-mFwe2-HA , which was amplified by PCR with the primers ( forward primer: AAAAGCTTGGCAGTGGAGAGGGCAGAGGAAGTCTGCTAACATGCGGTGACGTCG AGGAGAATCCTGGCCCAAGCGGCTCGGGCGCC; reverse primer: CCCTCGAGTTA CGCGTAGTCGGGGAC ) , after GFP or tagRFP . Total RNA of different regions of the adult rat brain were extracted with TRIZOL reagent according to the manufacturer's instructions . Five μg of total RNA was mixed with oligo-dT primer in 20 μl of reverse transcription reaction solution . One μl of this mixture was used to amplify the cDNA of ratFwe2 mRNA with specific primers ( forward primer: GAAGATCTATGAGCGGCTCGGTCGCC; reverse primer: CGGAATTCTCACAGTTCCCCCTCGAATG ) . Twenty-five PCR cycles were used to allow exponential PCR amplification . The PCR products were sequenced to validate the identity of ratFwe2 mRNA . GST-fused polypeptides comprising seven tandem repeats of the C-terminus of Drosophila Fwe-PB isoform [24] were injected in guinea pigs to obtain GP100Y antisera . To generate α-m/ratFwe2 antisera ( GP67 ) , the DNA fragment encoding seven tandem repeats of ratFwe2 C-terminus ( a . a . 140–172 ) was subcloned to pET28a plasmid . His-fused polypeptides were purified and then injected into guinea pigs . Antibody generation was assisted by LTK BioLaboratories ( Taiwan ) . Specific antibodies were further purified by antigen-conjugated affinity columns . For immunostaining in fly NMJ boutons , larval fillets were dissected in ice-cold 1X PBS and fixed in 4% paraformaldehyde for 20 min at room temperature . The samples were permeabilized in 0 . 1% Triton X-100-containing 1X PBS solution for all staining , except staining with anti-Fwe ( GP100Y ) and anti-HA antibodies , which used 0 . 1% Tween-20-containing 1X PBS solution to prevent Fwe dissociation from the SVs . Primary antibody dilutions used mouse anti-Dlg ( mAb 4F3 ) [57] , 1:100 ( Hybridoma bank ) [58]; mouse anti-Brp ( nc82 ) , 1:100 ( Hybridoma bank ) [59]; rabbit anti-GFP , 1:500 ( Invitrogen ) ; mouse monoclonal anti-HA , 1:200 ( Sigma ) ; rabbit anti-HA , 1:200 ( Sigma ) ; rabbit Cy3 conjugated anti-HRP , 1:500 ( Jackson ImmunoResearch ) ; and guinea pig anti-Fwe-PB ( GP100Y ) , 1:100 . Secondary antibodies were diluted to 1:500 ( Jackson ImmunoResearch and Invitrogen ) . For immunostaining of cultured rat hippocampal neurons , DIV14 neurons were fixed in 4% paraformaldehyde/4% sucrose for 10 min at room temperature and permeabilized and washed with 0 . 1% Tween-20-containing 1X PBS solution . To reduce nonspecific staining , GP67 antibodies were absorbed with paraformaldehyde-fixed n2a cells before staining . Primary antibody dilutions used rabbit anti-GFP , 1:500 ( Invitrogen ) ; guinea pig ant-m/ratFwe2 ( GP67 ) , 1:100; rabbit anti-GM130 , 1:500 ( Adcam ) ; mouse anti-HA , 1:200 ( Sigma ) ; and mouse anti-SVP38 , 1:200 ( Sigma ) . Secondary antibodies were diluted to 1:500 ( Jackson ImmunoResearch and Invitrogen ) . DAPI was used in the 1:2 , 000 dilution ( Invitrogen ) . To compare the staining intensity of boutons among different genotypes , larval fillets used in the same graph were stained in the same Eppendorf tube . The images were captured using a Zeiss 780 confocal microscope , and the scan setup was fixed for the same experimental set . For data quantifications , single-plane confocal images were projected . The final staining intensity in boutons was calculated by subtracting the background fluorescence intensity in the surrounding muscles from the staining intensity in boutons . The staining intensities of all type Ib boutons from the same muscles 6 and 7 in one image were averaged to obtain each data value . Image processing was achieved using LSM Zen and Image J . Mouse neuroblastoma n2a cells were transfected with pSpCas9 ( BB ) -m/ratFwe-gRNA-2A-GFP plasmid . GFP-positive cells were sorted out using flow cytometry , and the cells were plated in a 96-well plate in which each well included approximately one cell . After 3-wk culture , single cell-driven colonies were subjected to immunostaining and immunoblotting for mFwe2 to verify the knockout of mFwe2 . One of the confirmed mFwe knockout neuroblastoma n2a cell lines was used in S9A Fig . For western blotting of n2a cell lysates , the cells lysed with RIPA buffer ( 50 mM Tris-HCl [pH 8] , 150 mM NaCl , 1% NP-40 , 0 . 5% sodium deoxycholate , and 0 . 1% SDS ) were boiled in 1X SDS sample buffer for 10 min . To prepare subcellular fractions of adult rat brain , one adult brain ( ~1 g ) was homogenized in 5 ml 0 . 32 M sucrose buffer ( 320 mM sucrose , 1 . 5 mM MgCl2 , 1 mM EGTA , and 10 mM HEPES [pH 7 . 5] ) using a Teflon glass homogenizer . The homogenates ( H ) were centrifuged at 800 × g for 15 min at 4°C to yield pellets ( P1 ) and supernatants ( S1 ) . S1 supernatants were centrifuged at 9 , 200 × g for 15 min at 4°C to obtain pellets ( P2 ) and supernatants ( S2 ) , which were centrifuged at 100 , 000 × g for 2 h at 4°C to obtain fractions of cytosol ( S3 ) and light membrane ( P3 ) . The pellets ( P2 ) were then lysed in ice-cold Mini-Q water , followed by equilibration with 4 mM HEPES . After 30-min mixing at 4°C , the lysates were centrifuged at 25 , 000 × g for 20 min at 4°C to yield the crude synaptic vesicle fraction ( LS1 ) and lysed synaptosomal membrane fraction ( LP1 ) . The LS1 fraction was further centrifuged at 100 , 000 × g to obtain crude synaptic vesicles ( LP2 ) and the synaptosomal cytosol fraction ( LS2 ) . A discontinuous sucrose gradient from 0 . 3–0 . 99 M was prepared by gradually layering the different concentrations of sucrose . The S1 supernatants were loaded on sucrose gradient solution and centrifuged at 33 , 000 × g for 3 h at 4°C . Fractions were collected from low- to high-density sucrose . These fractions were boiled in 1X SDS sample buffer and subjected to SDS-PAGE and western blotting . The primary antibody dilutions used were as follows: guinea pig anti-m/ratFwe2 ( GP67 ) , 1:500; rabbit anti-SVP38 , 1:1 , 000 ( Sigma ) ; rabbit anti-GM130 , 1:5 , 000 ( Abcam ) ; mouse anti-Tubulin , 1: 10 , 000 ( Sigma ) ; and mouse anti-α-Actin , 1:10 , 000 ( Sigma ) . Secondary HRP-conjugated antibodies were diluted to 1:5 , 000 ( Jackson ImmunoResearch ) . To induce CME , the third instar larvae were dissected in 0 mM Ca2+ hemolymph-like ( HL ) -3 solution at room temperature ( 70 mM NaCl , 5 mM KCl , 10 mM MgCl2 , 10 mM NaHCO3 , 5 mM trehalose , 5 mM HEPES [pH 7 . 2] , and 115 mM sucrose ) [60] and subjected to 1-min 90 mM K+/0 . 5 mM Ca2+ stimulation ( 25 mM NaCl , 90 mM KCl , 10 mM MgCl2 , 10 mM NaHCO3 , 5 mM trehalose , 5 mM HEPES [pH 7 . 2] , 30 mM sucrose , and 0 . 5 mM CaCl2 ) or 10-min 60 mM K+/1 mM Ca2+ stimulation ( 55 mM NaCl , 60 mM KCl , 10 mM MgCl2 , 10 mM NaHCO3 , 5 mM trehalose , 5 mM HEPES [pH 7 . 2] , 30 mM sucrose , and 1 mM CaCl2 ) in the presence of 4 μM fixable FM1-43 ( Invitrogen ) . Excess dye was extensively washed with 0 mM Ca2+ HL-3 solution for 10 min . Larval fillets were fixed in 4% paraformaldehyde for 10 min , washed , mounted , and imaged on a Zeiss 780 confocal microscope . The scan setup was fixed for all the sets of the experiments . For data quantifications , single-plane confocal images were projected , and the final FM1-43 dye intensity in the boutons was calculated by subtracting the dye fluorescence intensity in the surrounding muscles from the dye fluorescence intensity within the boutons . The dye fluorescence intensities of all type Ib boutons from the same muscles 6 and 7 were averaged to obtain each data value . For chlorpromazine treatment experiment , dissected larvae were incubated with 200 μM chlorpromazine ( Sigma ) in Schneider medium for 30 min . They were then stimulated with a solution of 90 mM K+/2 mM Ca2+/200 μM chlorpromazine ( 25 mM NaCl , 90 mM KCl , 10 mM MgCl2 , 10 mM NaHCO3 , 5 mM trehalose , 5 mM HEPES [pH 7 . 2] , 30 mM sucrose , 2 mM CaCl2 , and 200 μM chlorpromazine ) in the presence of 4 μM fixable FM1-43 for 10 min . Bulk membranous invaginations were defined as the internalized structures labeled with high levels of FM1-43 dye . The areas of individual type Ib boutons and bulk membranous invaginations were measured using Image J . For FM1-43 dye loading/unloading assays , larvae were dissected in 0 mM Ca2+ HL-3 solution at room temperature and subjected to a stimulation of 90 mM K+/0 . 5 ( or 2 ) mM Ca2+ HL-3 solution for 5 min to load boutons with the FM1-43 dye . Excess dye was removed by extensive washing with 0 mM Ca2+ HL-3 solution for 10 min . FM1-43 dye uptake by boutons was imaged to indicate “loading . ” Subsequently , the dye loaded in SVs was unloaded by stimulation using 90 mM K+/0 . 5 ( or 2 ) mM Ca2+ solution for 1 min . Released dye was removed by several washes with a 0 mM Ca2+ HL-3 solution . The remaining dye in boutons was imaged to indicate “unloading . ” The final FM1-43 dye intensity in the boutons was calculated by subtracting the dye fluorescence intensity in the surrounding muscles from the dye fluorescence intensity within the boutons . The dye fluorescence intensities of at least ten type Ib boutons from the same muscles 6 and 7 were averaged to obtain each data value . The dye unloading efficiency was indicated as ( Fload-Funload ) /Fload . Images processing was achieved using Image J and LSM Zen . The third instar larvae were dissected in 0 mM Ca2+ HL-3 at room temperature and then bathed in 1 mM Ca2+ HL-3 solution for 5–10 min before the recording . The mean value of the resistance of the recording electrode was ~40 MΩ when the electrode was filled with a 3M KCl solution . All recordings were obtained from muscle 6 of abdominal segment 3 . Each larva was only used for one recording . Recordings from the muscles that hold resting membrane potentials at less than −60 mV were used for further data quantifications . EJPs were evoked by stimulating the axonal bundle via a glass capillary electrode with an internal diameter of ~10–15 μm ( Harvard Apparatus Glass Capillaries GC120F-15 ) at 0 . 2 Hz . Stimulus pulses were fixed at 0 . 5 ms duration ( pClamp 10 . 6 software , Axon Instruments ) . To obtain maximal EJP amplitude , 3–5 mV electric stimuli were applied . EJPs were amplified with an Axoclamp 900A amplifier ( Axon Instruments , Foster City , California ) under bridge mode and filtered at 10 kHz . EJPs were analyzed by pClamp 10 . 6 software ( Axon Instruments ) . For the EJP amplitude at 0 . 2 Hz , the mean of the EJP amplitude was averaged from the amplitudes of 80 EJPs in one consecutive recording . Larval fillets were dissected in 0 mM Ca2+ HL-3 medium at room temperature . To trigger CME , samples were stimulated with 90 mM K+/0 . 5 mM Ca2+ HL-3 solution for 1 min or 60 mM K+/1 mM Ca2+ HL-3 solution for 10 min . To induce ADBE , larval fillets were subjected to stimulation of a 90 mM K+/2 mM or 5 mM Ca2+ HL-3 solution in the presence or lack of 10 μM La3+ for 10 min . Subsequently , the samples were fixed for 12 h in 4% paraformaldehyde/1% glutaraldehyde/0 . 1 M cacodylic acid ( pH 7 . 2 ) solution and then rinsed with 0 . 1 M cacodylic acid ( pH 7 . 2 ) solution . They were subsequently fixed in 1% OsO4/0 . 1 M cacodylic acid solution at room temperature for 3 h . The samples were subjected to a series of dehydration from 30% to 100% ethanol . After 100% ethanol dehydration , the samples were incubated in propylene , a mixture of propylene and resin , and pure resin . Lastly , they were embedded in 100% resin . The images of type Ib boutons were captured using Tecnai G2 Spirit TWIN ( FEI Company ) and a Gatan CCD Camera ( 794 . 10 . BP2 MultiScan ) at ≥4 , 400 × magnifications . The size of the SVs and the bulk cisternae and the area of type Ib boutons were measured using Image J . We identified type Ib boutons by multiple layers of subsynaptic reticulum . The radius of the bulk cisternae was calculated from A ( area ) = πr2 . Isolated membranous structures larger than 80 nm in diameter were defined as bulk cisternae . For SypHy imaging , DIV7 cultured rat hippocampal neurons were transfected with pSpCas9 ( BB ) and pCMV-SyphyA4 ( addgene#24478 ) plasmids [5] . DIV13–15 neurons were bathed in the imaging buffer in a chamber ( Warner instruments RC-25F ) with two parallel platinum wires separated by 5 mm . The imaging buffer consisted of 136 mM NaCl , 2 . 5 mM KCl , 2 mM CaCl2 , 1 . 3 mM MgCl2 , 10 mM glucose , 10 mM HEPES ( pH 7 . 4 ) , 10 μM CNQX , and 50 μM AP-5 . SV exocytosis was elicited with a train of 200 action potentials delivered with a 20-Hz electric field stimulation ( 50 mA , 1-ms pulse width ) . Single images were captured every 1 s using MetaMorph software and an ANDOR iXon 897 camera . The experiments were performed at either room or physiological temperatures controlled by a Warner temperature controller ( TC-344B ) . For SV reacidification experiments , the imaging chamber was perfused with the imaging buffer ( 136 mM NaCl , 2 . 5 mM KCl , 2 mM CaCl2 , 1 . 3 mM MgCl2 , 10 mM glucose , and 10 mM HEPES [pH 7 . 4] , 10 μM CNQX , and 50 μM AP-5 ) before perfusion with an acidic buffer ( 136 mM NaCl , 2 . 5 mM KCl , 2 mM CaCl2 , 1 . 3 mM MgCl2 , 10 mM glucose , and 10 mM 2-[N-morpholino] ethane sulphonic acid [pH 5 . 5] , 10 μM CNQX , and 50 μM AP-5 ) , which was prepared by replacing HEPES in the imaging buffer with 2-[N-morpholino] ethane sulphonic acid [47 , 48] . Next , the imaging buffer was perfused to allow surface SypHy to be fluorescent . Experimental temperatures were maintained at physiological temperatures . The final SypHy fluorescence intensities in the presynaptic terminals were calculated by subtracting the background fluorescence intensity on the surrounding coverslip from the SypHy fluorescence intensity within presynaptic terminals . Each data value was obtained from a single terminal . For 40 kDa TMR-dextran uptake assays , DIV13–15 neurons transfected with pSpCas9 ( BB ) plasmids were stimulated by a train of 1 , 600 action potentials delivered with an 80 Hz electric field stimulation ( 50 mA , 1-ms pulse width ) in the imaging solution ( 144 mM NaCl , 2 . 5 mM KCl , 2 . 5 mM CaCl2 , 2 . 5 mM MgCl2 , 10 mM HEPES [pH 7 . 5] , 10 μM CNQX , and 50 μM AP-5 ) [49] in the presence of 50 μM 40 kDa TMR-dextran ( Invitrogen ) . Subsequently , neurons were perfused with the same buffer for 5 min to remove excess dextran dye . Experiments were performed at room temperature . Imaging was achieved through MetaMorph software and an ANDOR iXon 897 camera . Image processing was achieved using Image J and LSM Zen . Paired and multiple data sets were compared by Student’s t test and one-way ANOVA statistical analyses , respectively . All data analyses were achieved using GraphPad Prism 7 . 0 . The numerical data used in all figures are included in S1 Data .
|
The arrival of an action potential at the nerve end induces synaptic vesicle ( SV ) exocytosis to allow the release of chemical neurotransmitters and the rapid transmission of signals . SV endocytosis is in turn elicited in order to rapidly replenish the vesicle pool in neurons . Therefore , tight coupling between exocytosis and endocytosis within these cells maintains constant synaptic transmission . Exocytosis and intracellular Ca2+ elevation are known to be key prerequisites for the two main modes of SV endocytosis , Clathrin-mediated endocytosis ( CME ) and activity-dependent bulk endocytosis ( ADBE ) , which are primarily triggered by moderate and strong nerve stimuli , respectively . However , how these two events cooperate to trigger endocytosis upon exocytosis remains unclear . In this study , we show that Flower ( Fwe ) , an SV-associated Ca2+ channel , plays a significant role in these processes in Drosophila . Upon mild stimulation , we find that Fwe is transferred from the SV to endocytic zones to promote CME independently of Ca2+ channeling . In contrast , upon intense stimulation Fwe triggers Ca2+ influxes that elicit ADBE . In addition , we find that Fwe promotes CME and ADBE also in mammalian central synapses , revealing a conserved role for Fwe in endocytosis . We conclude that the Fwe channel exerts two different functions in response to two different stimuli to govern distinct modes of synaptic vesicle retrieval , thereby coupling exocytosis to endocytosis .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] |
[
"medicine",
"and",
"health",
"sciences",
"neuromuscular",
"junctions",
"nervous",
"system",
"cell",
"processes",
"electrophysiology",
"pigments",
"neuroscience",
"surgical",
"and",
"invasive",
"medical",
"procedures",
"developmental",
"biology",
"functional",
"electrical",
"stimulation",
"materials",
"science",
"nerve",
"fibers",
"dyes",
"animal",
"cells",
"life",
"cycles",
"materials",
"by",
"attribute",
"presynaptic",
"terminals",
"exocytosis",
"fluorescent",
"dyes",
"endocytosis",
"cellular",
"neuroscience",
"cell",
"biology",
"secretory",
"pathway",
"anatomy",
"synapses",
"neurophysiology",
"physiology",
"neurons",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"physical",
"sciences",
"larvae"
] |
2017
|
A Ca2+ channel differentially regulates Clathrin-mediated and activity-dependent bulk endocytosis
|
The lack of a mutant phenotype in homozygous mutant individuals’ due to compensatory gene expression triggered upstream of protein function has been identified as genetic compensation . Whilst this intriguing process has been recognized in zebrafish , the presence of homozygous loss of function mutations in healthy human individuals suggests that compensation may not be restricted to this model . Loss of skeletal α-actin results in nemaline myopathy and we have previously shown that the pathological symptoms of the disease and reduction in muscle performance are recapitulated in a zebrafish antisense morpholino knockdown model . Here we reveal that a genetic actc1b mutant exhibits mild muscle defects and is unaffected by injection of the actc1b targeting morpholino . We further show that the milder phenotype results from a compensatory transcriptional upregulation of an actin paralogue providing a novel approach to be explored for the treatment of actin myopathy . Our findings provide further evidence that genetic compensation may influence the penetrance of disease-causing mutations .
Genetic compensation exists as a mechanism to buffer the organism against gene loss that would otherwise be deleterious to survival . Whilst this term has been used to describe dosage compensation , evolution resulting in reversion to ancestral phenotypes , and gene duplication compensating for mutation , in the present study we refer to genetic compensation as altered gene expression resulting in a normal phenotype in a homozygous mutant individual . Studies in plants [1] , worms [2] , and yeast [3 , 4] have demonstrated that genetic robustness can be achieved by the presence of duplicate gene copies , which have retained a similar biological function , or through the presence of redundant pathways . A single case of genetic compensation in zebrafish has been shown to result in the activation of a network of gene expression resulting from the presence of a mutation in egf17 , but not from gene knockdown [5] . Intriguingly , data from widespread application of whole genome sequencing demonstrates that homozygous mutations predicted to cause a loss of function , that would normally cause disease , may be present in healthy individuals in the human population [6] , suggesting that compensation may be more widespread than previously thought . However , the lack of functionally characterized examples raise the question of whether this is an isolated case , or if genetic compensation may be a common mechanism contributing to genetic robustness in vertebrates . In the skeletal muscle , expression of cardiac muscle α-actin ( ACTC1 ) can partially compensate for loss of ACTA1 to ameliorate the loss of function phenotype [7] . α-actin is an essential component of the thin filament , with its mutation resulting in a range of skeletal muscle disorders , including nemaline myopathy [8] . ACTA1 differs from ACTC1 by only four amino acids and both are co-expressed in the skeletal muscle and heart during development [9–11] . In vertebrates , ACTC1 is the predominant actin isoform in fetal skeletal muscle , however , after birth ACTC1 is down regulated [11] and by adulthood it comprises <5% of adult skeletal muscle [12] , with ACTA1 becoming the predominant form . Most ACTA1 mutations are de novo dominant mutations with only 10% of patients’ carrying recessively inherited loss of function mutations [8] . Patients displaying a complete absence of ACTA1 typically show retention of ACTC1 in their skeletal muscle , with the level of retention determining the level of clinical severity [7] . Transgenic overexpression of ACTC1 was also shown to rescue the early lethality observed in recessive ACTA1-/- knockout mice strains [13] , further demonstrating the potential for the levels of ACTA1 paralogues to influence disease severity . Here we uncover a novel case of genetic compensation in zebrafish within the highly conserved actin gene family . We [14] and others [15 , 16] have shown that loss or knockdown of skeletal α-actin is catastrophic for muscle structure and function . Antisense Morpholino ( MO ) knockdown of Actc1b results in the formation of nemaline bodies and reduced skeletal muscle performance [14] . In contrast , we observe very mild defects in a genetic actc1b mutant . We determine that this is due to a compensatory transcriptional upregulation of an α-actin paralogue in the skeletal muscle buffering the loss of Actc1b function . Our study not only supports the existence of genetic compensation as a phenomenon affecting phenotypic diversity in vertebrates , but also identifies a genetic process that may have therapeutic implications for nemaline myopathy and other actin-related diseases .
To investigate α-actin function in zebrafish we initially examined the expression of the α-actin paralogues . Four α-actin genes ( acta1a , acta1b , actc1a , and actc1b ) have been identified in the zebrafish [14] , and all are expressed in both skeletal and cardiac muscle during early zebrafish development ( S1 Fig ) . qRT-PCR analyses showed that , whilst all genes are expressed during early embryogenesis , actc1b is expressed at much higher levels than the other paralogues and that by 180 dpf ( days post-fertilization ) actc1a and actc1b are the predominant α-actin isoforms in the heart and skeletal muscle respectively ( Fig 1 , S1 Table , and S2 Table ) . We were surprised to find that the two cardiac α-actin genes were the most highly expressed in the muscle and therefore analyzed the sequence similarity between the α-actin isoforms . An analysis of the surrounding genomic regions shows conserved synteny between human ACTA1 and zebrafish acta1a and acta1b genes , and similarly between human ACTC1 and zebrafish actc1a and actc1b genes ( S2 Fig ) suggesting that indeed the zebrafish genes are orthologous to their respective human genes . Interestingly , a maximum likelihood phylogenetic analysis predicts that , following an initial duplication , the ACTA1 and ACTC1 copies separate into distinct clades except in zebrafish where actc1a , acta1b , actc1a , and actc1b all form a clade together with ACTA1 ( S3 Fig ) . Although this suggests that the zebrafish actc1 genes may have evolved to become more similar to ACTA1 than to their respective orthologue , the low bootstrapping values reflect a very high level of nucleotide sequence conservation between all genes ranging from 85–90% identity . To analyze Actc1b function in zebrafish we examined the phenotype of actc1b mutants ( actc1bsa12367 , referred to hereafter as actc1b-/- ) . This mutant was generated using ENU mutagenesis and the presence of a nonsense mutation at amino acid 5 was verified by competitive allele specific PCR ( KASP ) genotyping [17] . We verified the mutation in the actc1b-/- mutant strain and confirmed that the MO binding sites were intact ( S4 Fig ) . Surprisingly , we found no difference in the appearance of the muscle fibers between actc1b-/- embryos and their wildtype siblings ( Fig 2A ) and only a small , but significant , reduction in swimming capabilities of actc1b-/- embryos compared to wildtype siblings ( actc1b+/+ mean 1 . 0±0 . 1SEM and actc1b-/- mean 0 . 77±0 . 12SEM; p 0 . 0128; Fig 2B ) . Whilst actc1b-/- fish show only a 23% reduction in distance swum , in line with our previous work , Actc1b ex2 morphants show a dramatic reduction in swimming performance compared to Standard Control morphants , UTR MO morphants , and uninjected controls ( median values Actc1b ex2 MO 0 . 23 , UTR MO 0 . 89 , Standard Control MO 0 . 98 , and uninjected 1 . 03; p<0 . 0001 for ex2 MO against all other conditions; Fig 2C ) . To determine whether the phenotypic differences observed were due to MO off-target effects we injected the Actc1b ex2 , UTR MO , or Standard Control MO into an incross of actc1b+/- zebrafish and assessed phenotypic severity in their offspring in three independent experiments . The phenotypes were classed as either wildtype , mild ( slight disruption to the muscle fibers ) , or severe ( large disruption to the muscle fibers and Actinin2 aggregates at the myosepta ) ( Fig 3A ) . Remarkably , we found that actc1b-/- mutants never displayed a severe phenotype when injected with either the Actc1b ex2 MO ( 21 wildtype , 25 mild , 0 severe ) or Actc1b UTR MO ( 14 wildtype , 13 mild , 0 severe ) similar to the Standard Control MO ( 16 wildtype , 9 mild , 0 severe ) . However , severe phenotypes were observed in actc1b+/+ and actc1b+/- siblings injected with either Actc1b MO ( actc1b+/+ injected with the Actc1b ex2 MO: 0 wildtype , 2 mild , 39 severe or Actc1b UTR MO: 0 wildtype , 59 mild , 17 severe and for actc1b+/- injected with the Actc1b ex2 MO: 0 wildtype , 66 mild , 29 severe or Actc1b UTR MO: 25 wildtype , 55 mild , 1 severe ) which were not observed following injection with the Standard Control MO ( for actc1b+/+: 29 wildtype , 0 mild , 0 severe and for actc1b+/-: 35 wildtype , 0 mild , 0 severe ) ( Fig 3B ) . The change in proportions of phenotype classes was significant for both morpholinos into actc1b+/+ and actc1b+/- compared to control MO ( p<0 . 0001 , Chi-square test ) . The insensitivity of the actc1b-/- mutants to Actc1b MO knockdown demonstrates that the severe phenotypes are not due to off-target effects . In addition , we measured the locomotion of actc1b-/- mutants and their wildtype siblings injected with either the Actc1b ex2 , Actc1b UTR , or Standard Control MO at 6 dpf . We observed a significant reduction in distance travelled by actc1b+/+ and actc1b+/- siblings injected with an Actc1b ex2 MO ( median 0 . 14 for actc1b+/+ , p<0 . 0001; and 0 . 19 for actc1b+/- , p<0 . 0001 ) and for actc1b+/+ siblings injected with an Actc1b UTR MO ( median 0 . 83 , p0 . 046 ) compared to the Standard Control MO ( 1 . 0 for actc1b+/+ and 0 . 95 for actc1b+/- ) ( Fig 3C ) . In contrast , actc1b-/- mutants injected with either an Actc1b ex2 ( median 0 . 90 ) or an Actc1b UTR MO ( median 0 . 95 ) show comparable locomotion to those injected with a Standard Control MO ( median 0 . 83 ) confirming that actc1b-/- mutants are indeed unaffected by Actc1b knockdown ( Fig 3C ) . To determine why there was not a loss of function phenotype in actc1b-/- mutants we first measured α-actin levels in the skeletal muscle of actc1b-/- mutants , wildtype siblings , and Actc1b morphants by western blot analysis . As previously observed ( Sztal et al , 2015 ) , Actc1b morphants display decreased total actin levels compared to siblings injected with a control morpholino ( Fig 4A and S5 Fig ) . Quantification and analysis of western blots by two-way ANOVA shows that actc1b+/+ and actc1b+/- siblings injected with either an Actc1b UTR MO ( mean 0 . 94 SD 0 . 19 for actc1b+/+ ( p0 . 0048 ) and mean 1 . 15 ±0 . 18SD for actc1b+/- ( p0 . 0077 ) ) or Actc1b ex2 MO ( mean 0 . 51 ±0 . 03SD for actc1b+/+ ( p<0 . 0001 ) and mean 0 . 72 ±0 . 23SD for actc1b+/- ( p<0 . 0001 ) ) show significantly reduced α-actin compared to Control MO injected actc1b+/+ and actc1b+/- siblings ( mean 1 . 51 ±0 . 27SD for actc1b+/+and mean 1 . 63 ±0 . 09SD for actc1b+/- ) ( Fig 4B and S5 Fig ) . However , actc1b-/- mutants injected with either an Actc1b UTR MO ( mean 1 . 23 ±0 . 25SD ) or Actc1b ex2 MO ( mean 0 . 77 ±0 . 09SD ) , show no decrease in α-actin levels compared to or Standard Control MO ( mean 1 . 08 ±0 . 11SD; p0 . 5779 and p0 . 0955 respectively ) explaining the lack of phenotype compared to actc1b+/+ and actc1b+/- morphants . The analysis also identified a significant interaction between genotype and MO treatment ( p0 . 0088 ) confirming a genotype specific effect of the morpholinos . Given the normal levels of actin in actc1b-/- fish we hypothesized that other α-actin paralogues were compensating for the loss of Actc1b . To determine the expression of the four α-actin paralogues we assayed the skeletal muscle of actc1b+/+ , actc1b+/- , and actc1b-/- embryos using qRT-PCR . We found a significant decrease in RNA levels of actc1b in Actc1b morphants ( UTR MO mean 1457 ±70SD , ex2 MO mean 1160 ±237SD , compared to Standard Control MO mean 5418 ±505SD , p<0 . 0001 for both ) as previously shown in Sztal et al ( 2015 ) as well as in actc1b-/- mutants ( mean 563 ±142SD compared to actc1b+/+ mean 2243 ±848SD , p0 . 0172 ) , suggesting the actc1b gene product is degraded by nonsense mediated decay ( Fig 4C & 4D ) . Interestingly , actc1a expression was significantly increased in actc1b-/- mutants compared to actc1b+/+ siblings ( mean 15334 ±742SD and mean 374 ±80SD respectively , p0 . 0346 , Fig 4C ) . Conversely , Actc1b morphants did not display significant changes in the expression of any of the other α-actin isoforms compared to siblings injected with a Standard Control MO ( Fig 4D ) . To confirm that the compensatory upregulation of actc1a was responsible for the milder phenotype in actc1b-/- mutants , we reasoned that if Actc1a was reduced in actc1b-/- mutants , they would display a more severe phenotype , comparable to Actc1b morphants . We used a MO targeting the splice donor site of exon 2 ( Actc1a ex2 MO ) to knockdown Actc1a . To determine a subphenotypic dose for the Actc1a ex2 MO we injected 0 . 5 , 1 . 0 , or 2 . 0ng of MO into wildtype embryos and performed both RT-PCR and western analyses to determine MO efficiency . We were able to detect a small decrease in α-actin by western blot and RT-PCR analyses revealed two amplicons , a smaller product observed in all of the samples including controls which was reduced in the MO injected embryos , and a larger band , only observed in 1 . 0 and 2 . 0ng injected samples ( S6A Fig ) . We sequenced the larger band and confirmed that it corresponds to the inclusion of intron 2 , resulting from mis-splicing of exon 2 and 3 , causing the addition of three amino acids and a stop codon which would undoubtedly disrupt Actc1a function . Mutations in actc1a have been shown to cause heart defects resulting in decreased cardiac contractility and altered blood flow [18] . Although the skeletal muscle appeared unaffected by MO injections ( S6E Fig ) , we observed a slightly dilated heart in embryos injected with a 1ng MO concentration which became more severe as the MO dose increased ( S6D Fig ) . Based on these observations , we selected a 1ng MO dose to use in further experiments . We then injected the Actc1a ex2 MO ( or corresponding dose of a Standard Control MO ) into an incross of actc1b+/- zebrafish and assessed phenotypic severity in their offspring in three independent experiments . The phenotypes were classed as either wildtype , mild ( slight disruption to the muscle fibers ) , or severe ( large disruption to the muscle fibers and Actinin2 aggregates at the myosepta ) ( Fig 5A ) . In actc1b+/+ and actc1b+/- siblings injected with either the Standard Control ( actc1b+/+: 17 wildtype , 0 mild , 0 severe and actc1b+/-: 55 wildtype , 2 mild , 0 severe ) or Actc1a MO ( actc1b+/+: 28 wildtype , 0 mild , 0 severe and actc1b+/-: 59 wildtype , 0 mild , 0 severe ) we only observed a wildtype or mild phenotype . However , when we injected the Actc1a MO into actc1b-/- mutants we observed a severe phenotype in approximately 40% of fish ( 3 wildtype , 15 mild , 11 severe ) which was not observed in Standard Control MO injected actc1b-/- mutants ( 4 wildtype , 14 mild , 0 severe , change in phenotype proportions p0 . 0106 Chi-square test ) . Taken together these results demonstrate that upregulation of the actc1a paralogue is protective in actc1b-/- mutants .
We have identified compensation triggered by a mutation in actc1b but not following morpholino mediated knockdown of Actc1b . There has been considerable debate recently as a result of phenotypic differences between mutant lines and morpholino-mediated knockdown [19–22] . The study by Rossi et al ( 2015 ) provided an example where this difference in phenotype was due to compensation , rather than morpholino off target effects as previously suggested . The data presented in the current study identifies another example of the phenomenon of genetic compensation , and , rather than compensation in egf17 mutants being an isolated case , suggests that this process may be more widespread . It also suggests that in some cases , rather than the phenotypic differences between mutant and knockdown animals being due to off-target or non-specific effects , genetic compensation may influence the phenotypic penetrance of deleterious mutations . Morpholino knockdown may therefore potentially reveal the phenotype resulting from reduction of the protein , without any compensatory transcription upregulation of paralogues or alternative pathways [23] . In zebrafish actc1a is expressed in the skeletal muscle during early embryogenesis , but is downregulated in the skeletal muscle as development proceeds . However , upregulation of actc1a is sufficient to compensate for the loss of Actc1b mimicking the upregulation of ACTC1 in patients suffering from recessive nemaline myopathy caused by mutations in ACTA1 [7] . In this situation , patients have a complete absence of skeletal α-actin but instead cardiac α-actin is upregulated leading to a milder disease phenotype than patients with dominant mutations in ACTA1 [7 , 15] . The levels of cardiac α-actin in these patients determines the clinical severity of the disease [7] . In contrast to the compensation we have identified , the specific response in the skeletal muscle in individuals with recessive ACTA1 mutations is not sufficient to prevent disease and the majority of individuals die within 6 months of birth [7] . Transgenic expression of ACTC1 in the skeletal muscle is , however , able to rescue both recessive ACTA1-/- and dominant ACTA1D286G mutations in mice [13 , 16] , consistent with our findings that upregulation of actin paralogues can prevent a disease phenotype . The absence of a similar compensatory mechanism resulting from reduction of Actc1b following morpholino antisense mediated knockdown suggests that compensation is not induced by the reduction in Actc1b , but at a step prior to protein formation . While the mechanisms of genetic compensation remain unclear , two different models have been recently proposed suggesting that compensation may be activated through either a DNA damage response or by the degradation of mutant RNA and subsequent activation of common microRNAs or ribosomal binding proteins to stabilize compensatory interactions [24] . In the case of actc1b , we have shown that there are no alternative transcripts produced in actc1b-/- mutants and that the resulting mRNA is likely to be non-functional and degraded by nonsense mediated decay , which may activate compensation . However , we cannot rule out the possibility that it is either the DNA lesion itself , or the presence of the mutant mRNA that may trigger compensation . Recent studies have shown that a missense mutation in actc1a ( actc1as434 ) causes severe defects in cardiac contractility and altered blood flow resulting from the loss of polymerized cardiac actin [18] . Injection of an Actc1a MO mimicked the heart edema and lack of endocardial cushion formation observed in actc1a mutants [18] suggesting that compensation does not play a role in actc1as434 mutants . However , it may be that if an alternative mutation , such as a nonsense mutation , is introduced or nonsense mediated pathways are activated that compensatory α-actin paralogues responses would be induced . Nevertheless , identifying the specific factors that trigger the compensatory upregulation of cardiac α-actin in the skeletal muscle tissues is the next challenge and could have therapeutic applications to ameliorate ACTA1 skeletal muscle diseases . Genetic robustness against null mutations appears to be a universal phenomenon in all organisms , however , the mechanisms determining compensation may differ . The existence of duplicate gene copies to compensate for the loss of an essential gene has been previously observed in yeast [4] and worms [2] with genome-wide deletion experiments revealing a significantly lower percentage of duplicates compared to singletons are essential for viability and fertility . Inherited mutations have also been shown to have variable consequences in different individuals [25] , which may be due to a plasticity of genetic compensatory responses masking the phenotypic effect of deleterious alleles . Our study verifies the existence of compensatory mechanisms , leading to a milder phenotype in ACTA1 recessive myopathy . More importantly , we suggest that similar compensatory responses may underline phenotypic differences in disease penetrance in the human condition .
Fish maintenance and handling was carried out as per the standard operating procedures approved by the Monash Animal Services Animal Ethics Committee under breeding colony license MARP/2015/004/BC . Fish were anaesthetized using Tricaine methanesulfonate . Zebrafish were maintained according to standard protocols [26] . The Actc1b ex2 ( 5’ TGCAGTGTTTTTTTCACCTGGTGAC 3’ ) Actc1b UTR ( 5’ GGTCAAGTTGTTATCACAAGACTGA 3’ ) , Actc1a ex2 ( 5’ TACATGCTTTAGAAGCCCACCTGGT 3’ ) and Standard Control ( 5’ CCTCTTACCTCAGTTACAATTTATA 3’ ) MOs ( GeneTools ) were diluted in distilled water and co-injected with Cascade Blue labeled dextran ( Molecular Probes ) into one- to two-cell embryos MO concentrations were calibrated according to [27] at the indicated amounts ( 2 . 0 ng for the Actc1b ex2 and UTR MOs corresponding to a concentration of 0 . 5mM; 0 . 5 , 1 . 0 , or 2 . 0ng for the Actc1a ex2 MO corresponding to concentrations of 0 . 125mM , 0 . 25mM , and 0 . 5mM; and 1 . 0 or 2 . 0ng for the Standard Control MO corresponding to concentrations of 0 . 25mM and 0 . 5mM ) . At 1 dpf , the embryos were sorted for Cascade Blue labeling . The actc1b mutant line ( sa12367 ) was obtained from the Zebrafish International Resource Centre [17] . Allele specific PCR KASP technology ( Geneworks ) was used for genotyping . Whole-mount in situ hybridization was carried out as described previously [28] . Probes were constructed using specific gene primers ( acta1a F: CAACATCCTATCATTGCCTCCT and R: CATGTTCAGTTTTATTTGTCTGTTGA; acta1b F: ATTCATCGGCTGCATCTGTC and R: TTAACACATATGCGTCACAAAAA; actc1a F: CCAGCACAATGAAGATCAAG and R: CCAGCACAATGAAGATCAAG; actc1b F: TGACCGTATGCAGAAGGAGAT and R: TCTTATCACTTATCTGTTT ) . Imaging was performed with an Olympus SZX16 stereomicroscope . Total RNA was extracted using TRIzol reagent ( Invitrogen Life Technologies ) . RNA samples were treated with RQ1 RNase-free DNase ( Promega ) . cDNA was synthesized from 1μg of each RNA sample in a 20ml reaction using Protoscript first strand cDNA synthesis kit ( New England Biosciences ) and oligo ( dT ) 20 primer following the supplier’s instructions . Quantitative PCR was performed on a Roche Lightcycler instrument and normalized against β-actin and RPS18 [29] as reference genes . Primers for quantitative PCR are listed in S3 Table . The human actin protein sequences were used as a query against representative databases from mouse , chicken and zebrafish genomes using a BLASTp search . Corresponding orthologues to all six actin isoforms were identified and aligned using ClustalX [30] . After using the MEGA 5 . 05 [31] program to determine the best–fit model for the analysis , a neighbor-joining tree ( JTT , bootstrapping = 1000 ) was compiled using MEGA 5 . 05 [31] . A yeast ACT1 ( Genbank ID: 850504 ) protein sequence was used as an outgroup . To analyze ACTA1 and ACTC1 synteny , orthologous genes were identified in Ensembl ( http://asia . ensembl . org/index . html ) and the flanking genomic regions were annotated . Locomotion assays were performed on 6 dpf zebrafish as per [32] . An inactivity threshold of 6 mm/s , detection threshold of 25 mm/s and maximum burst threshold of 30 mm/s were used . The total distance swum above the inactivity threshold and below maximum burst threshold in a 10-min period were extracted using the ZebraLab software ( ViewPoint Life Sciences ) . Blinding of treatments groups was used in combination with randomization of experimental replicates to remove any bias . Once the testing and genotyping was completed the treatments groups were uncovered . Immunofluorescence was performed on 2 dpf zebrafish as per [14] using an anti-Actinin2 antibody ( Sigma clone A7811 , 1:200 ) , AlexaFluorTM-labelled-596 secondary antibody ( Molecular Probes , 1:200 ) and rhodamine-tagged phalloidin ( Molecular Probes , 1:200 ) . For phenotypic experiments , samples were blinded during analyses and genotypes were revealed one all samples were scored . For western blot assays , the head and tails were separated from 2 dpf zebrafish from each condition for three independent biological replicates . The heads were used for genotyping and 20 tails per condition were used for protein lysates as per [33] and quantified using the Qubit fluorometric quantification ( Thermo Fisher Scientific ) . 10–20μg of each sample , along with reducing agent ( Life Technologies ) and protein loading dye ( Life Technologies ) , was heated at 70°C for 10 min and separated by SDS-PAGE on NuPAGE 4–12% Bis-Tris gels . The protein was transferred onto PVDF membrane ( Millipore ) , following which , the membrane was blocked with 5% skimmed milk in PBST and subsequently probed with anti-Actin ( Sigma , A2066 , 1/1000 ) , washed and incubated with HRP-conjugated secondary antibody ( Southern Biotech , 1:10 000 ) . Immunoblots were developed using ECL prime ( GE healthcare ) and imaged using a chemiluminescence detector ( Vilber Lourmat ) . The membrane was subsequently stripped , reprobed with anti-β-tubulin antibody ( Abcam , ab6046 , 1/5000 ) , incubated in HRP-conjugated secondary antibody and developed as above . The membrane was subsequently stripped and stained with Direct Blue 71 ( Sigma ) to identify total protein . Fiji was used to quantify protein intensity . For swimming analyses , all values were normalized to the average actc1b+/+ siblings injected with a control MO , or actc1b+/+ siblings . Normality of data was determined using a D’Agostino and Pearson test for normality . Normal data ( Fig 2B ) was analysed by one-way ANOVA using Dunnett’s correction for multiple comparisons . For data failing the normality test ( Fig 2C and Fig 3C ) , the test was repeated after the outliers were removed by the ROUT method ( Q = 1% ) or the data was logtransformed . In neither case did this result in a normal distribution of data . Therefore , in these cases the data from the three replicates was pooled and a Kruskal-Wallis test was performed and correction for multiple comparisons conducted using Dunn’s test . For phenotypic analyses ( Fig 3B and Fig 5 ) , the results of the three replicates were used to determine the mean percentage of each phenotype and to plot the graphs . The proportion of the phenotypes was determined by pooling the data from all three replicates and conducting a Chi-square test for each treatment against its respective control . For qRT-PCR data ( Fig 4C and 4D ) a one-way ANOVA was conducted for each gene comparing actc1b+/- and actc1b-/- to actc1b+/+ or UTR MO and ex2 MO to Standard Control MO using Dunnett’s test for multiple comparisons . All statistical analyses were conducted using GraphPad Prism 7 .
|
Many healthy individuals carry loss of function mutations in essential genes that would normally be deleterious for survival . Intriguingly , it may be the presence of the genomic lesion itself in these individuals that triggers the compensatory pathways . It is not known how widespread this phenomenon is in vertebrate populations and how genetic compensation is activated . We have shown that knockdown of actin causes nemaline myopathy as indicated by the formation of nemaline bodies within the skeletal muscle and reduced muscle function which , remarkably , we did not observe in an actin genetic mutant . We have identified that protection from the disease phenotype results from transcriptional upregulation of an actin paralogue restoring actin protein in the skeletal muscle . This study demonstrates that genetic compensation may be more prevalent than previously anticipated and highlights phenotypic differences resulting from genetic mutations versus antisense knockdown approaches . Furthermore , we suggest that activating compensatory pathways may be exploited as a potential novel therapeutic approach for human disorders caused by loss of function mutations .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] |
[
"fish",
"medicine",
"and",
"health",
"sciences",
"morpholino",
"rna",
"interference",
"skeletal",
"muscles",
"messenger",
"rna",
"vertebrates",
"nucleotides",
"animals",
"animal",
"models",
"osteichthyes",
"developmental",
"biology",
"model",
"organisms",
"untranslated",
"regions",
"experimental",
"organism",
"systems",
"epigenetics",
"embryos",
"research",
"and",
"analysis",
"methods",
"embryology",
"musculoskeletal",
"system",
"genetic",
"interference",
"gene",
"expression",
"muscles",
"antisense",
"oligonucleotides",
"biochemistry",
"zebrafish",
"rna",
"eukaryota",
"anatomy",
"nucleic",
"acids",
"phenotypes",
"oligonucleotides",
"genetics",
"biology",
"and",
"life",
"sciences",
"genetics",
"of",
"disease",
"organisms"
] |
2018
|
Genetic compensation triggered by actin mutation prevents the muscle damage caused by loss of actin protein
|
Currently , our knowledge of how pathogenic fungi grow in mammalian host environments is limited . Using a chemotherapeutic murine model of invasive pulmonary aspergillosis ( IPA ) and 1H-NMR metabolomics , we detected ethanol in the lungs of mice infected with Aspergillus fumigatus . This result suggests that A . fumigatus is exposed to oxygen depleted microenvironments during infection . To test this hypothesis , we utilized a chemical hypoxia detection agent , pimonidazole hydrochloride , in three immunologically distinct murine models of IPA ( chemotherapeutic , X-CGD , and corticosteroid ) . In all three IPA murine models , hypoxia was observed during the course of infection . We next tested the hypothesis that production of ethanol in vivo by the fungus is involved in hypoxia adaptation and fungal pathogenesis . Ethanol deficient A . fumigatus strains showed no growth defects in hypoxia and were able to cause wild type levels of mortality in all 3 murine models . However , lung immunohistopathology and flow cytometry analyses revealed an increase in the inflammatory response in mice infected with an alcohol dehydrogenase null mutant strain that corresponded with a reduction in fungal burden . Consequently , in this study we present the first in vivo observations that hypoxic microenvironments occur during a pulmonary invasive fungal infection and observe that a fungal alcohol dehydrogenase influences fungal pathogenesis in the lung . Thus , environmental conditions encountered by invading pathogenic fungi may result in substantial fungal metabolism changes that influence subsequent host immune responses .
The incidence of life-threatening human fungal infections has increased during the last three decades as medical therapies , organ transplantations , an increasing geriatric population , and HIV infections have generated a significant rise in the number of susceptible patients [1] , [2] , [3] . Aspergillus fumigatus , a commonly encountered mold found in soil and organic debris [4] , [5] , is responsible for a number of clinically relevant diseases in immunocompromised and immunocompetent individuals . Among these , invasive pulmonary aspergillosis ( IPA ) is the most lethal with mortality rates ranging from 30–90% depending on the patient population [6] , [7] , [8] , [9] , [10] , [11] . To cause lethal disease , A . fumigatus must face and overcome a number of in vivo microenvironment challenges once it is inhaled into the lower respiratory tract . However , our understanding of the dynamic microenvironments encountered by the fungus in the mammalian lung , and the mechanisms by which it grows in these microenvironments , are poorly understood . Arguably , understanding the mechanisms by which A . fumigatus is able to grow in the mammalian host environment will lead to either improvement of existing therapeutic options or development of novel treatments through the identification of novel drug targets . Some of the previously studied environmental factors encountered by A . fumigatus during in vivo growth include: high temperature , changes in pH , oxidative stress , and a restricted nutrient supply . In all probability , these stresses are similar to those that the mold has to overcome to be a highly competitive member of the compost microflora , and subsequently it has evolved multi-faceted and robust mechanisms to overcome these challenges [12] , [13] , [14] , [15] , [16] . An important characteristic of A . fumigatus's saprophytic lifestyle that has largely been overlooked is its ability to adapt to a wide range of oxygen levels . Aspergillus species are generally considered obligate aerobes , but A . fumigatus has been observed to tolerate oxygen levels as low as 0 . 1% [17] , [18] . In compost piles , oxygen concentrations range from atmospheric ( 21% ) to hypoxic ( 1 . 5% and lower ) and rapidly change with the metabolic activity of the compost microflora [19] . Thus , organisms such as A . fumigatus that thrive in such environments likely have evolved mechanisms to tolerate hypoxia . With regard to fungal-human interactions , oxygen availability in mammalian tissues is also substantially below atmospheric levels [20] , [21] , [22] . Even in the alveoli of healthy lungs , the most aerated organ and primary site of Aspergillus deposition , the oxygen level is around 14% . By the time oxygen reaches the capillaries and diffuses into surrounding tissues its availability is much lower with levels of 2–4% reported [23] , [24] . Thus , microorganisms that colonize , inhabit and infect mammalian hosts are subject to dynamic ranges of oxygen availability depending on their location in the mammalian body . Moreover , the collateral effects of microbial infections , inflammation , thrombosis , and necrosis , are often thought to decrease available oxygen concentrations even further [25] , [26] , [27] , [28] . However , the occurrence and effects of hypoxia on the outcome of human fungal infections , especially those that primarily occur in the lung , are poorly understood [29] , [30] . Recent evidence supporting the hypothesis that hypoxia is a significant component of fungal pathogenesis comes from studies on the sterol-regulatory element binding protein ( SREBP ) transcription factor in Cryptococcus neoformans and A . fumigatus . Null mutants of the respective SREBP in each fungal pathogen were incapable of growth in hypoxic conditions and subsequently were also avirulent in murine models of cryptococcosis and IPA [31] , [32] , [33] . Though the virulence defect in these SREBP mutants may be caused by other pleiotropic factors , their ability to grow in normoxic but not hypoxic conditions strongly suggests that adaptation and growth in hypoxia are contributing factors to the avirulence of these strains . Yet , as mentioned , whether hypoxia occurs during an invasive fungal infection in commonly used murine models of fungal disease is unknown . In this study , we observe for the first time that hypoxic microenvironments do occur in three immunologically distinct murine models of IPA . We also observe that a key gene , which encodes an enzyme required for the last step of ethanol fermentation in response to hypoxia , influences IPA pathogenesis through modulation of the inflammatory response . Thus , we conclude that in vivo hypoxic microenvironments do occur during IPA and that fungal responses to these conditions can influence fungal pathogenesis . These results lay the foundation for further studies to identify how human pathogenic fungi adapt to hypoxia and how these adaptation mechanisms ultimately influence the outcome of fungal pathogenesis in mammals .
In order to gain an understanding of the important metabolic pathways utilized by Aspergillus fumigatus during growth in the mammalian lung , we examined qualitative production of metabolites in a chemotherapeutic murine model of invasive pulmonary aspergillosis ( IPA ) utilizing broncheoalveolar lavages ( BAL ) and 1H-NMR . Our chemotherapeutic model of IPA is characterized by the use of cyclophosphamide and the corticosteroid triamcinolone to induce immunosuppression . Visual inspection of the BAL sample spectra taken from uninfected control mice and mice inoculated with A . fumigatus on day +3 post-inoculation revealed a relatively small number of identifiable metabolites and few differences . Identifiable metabolites in all samples included taurine , choline , creatine , acetate , and lactate ( Figure S1 ) . Given the low complexity of BAL samples that are predominately 0 . 7% saline , this result is not surprising . Surprisingly , however , ethanol was detected in 4 of the 10 mice infected with A . fumigatus , but in none of the uninfected controls ( Figure S1 ) . Attempts to detect ethanol at later time points during infection using BALs or lung homogenates were not successful . Thus , the extent of ethanol production in vivo during IPA remains to be determined . However , to support the hypothesis that the ethanol production was fungal in origin , we next tested whether A . fumigatus was capable of fermenting glucose to ethanol in vitro in glucose minimal media ( 1% glucose ) under normoxic or hypoxic conditions . While no ethanol was detectable after 24 h in either condition , we could detect ethanol in the culture supernatants after 48 , 72 , and 96 h of growth in hypoxia ( 1% oxygen , Figure 1 ) . The glucose levels expectedly dropped during the time course of the shake flask cultures , from 55 mM at time zero to less than 4 mM at 96 h , and this corresponded with a decrease in detectable ethanol ( Figure 1 ) . Fermentation is associated with hypoxic or anoxic environments in many organisms including plant pathogenic fungi , Crabtree negative yeast , pathogenic bacteria , and even plants [34] , [35] , [36] , [37] , [38] , [39] , [40] . We thus hypothesized that A . fumigatus encounters hypoxic or anoxic microenvironments during IPA . To test the hypothesis that A . fumigatus encounters hypoxia during IPA , we used a hypoxia marker , pimonidazole hydrochloride , a nitroheterocyclic drug whose hypoxia-dependent activation by cellular mammalian nitroreductases ( severe hypoxia: 10 mmHg partial oxygen pressure , ≤1% oxygen ) leads to the formation of covalent intracellular adducts with thiol groups on proteins , peptides , amino acids , and the drug itself [41] , [42] , [43] , [44] . The resulting protein adducts are effective immunogens and can be used to “visualize” hypoxia in vivo with immunofluorescence . We tested for the development of hypoxia in vivo in three immunologically distinct murine models of IPA ( chemotherapeutic , corticosteroid , and X-CGD ) ( Figure 2 ) . Each model represents a different clinically relevant mechanism of immunosuppression . As mentioned , the chemotherapeutic model attempts to reproduce the immunological state of severely immunosuppressed patients who have often undergone a bone marrow transplant . This model is characterized by a severe depletion of neutrophils and other important immune effector cells needed to prevent and control invasive fungal infections . Another patient population highly susceptible to invasive fungal infections is those patients on high doses of corticosteroids for treatment of graft versus host disease or other autoimmune type diseases . In our model of this patient population , we utilized a single high dose of the corticosteroid triamcinolone . Unlike the chemotherapeutic model , this model is not characterized by depletion of effector cells , but rather by a suppression of their antifungal activity that leads to altered inflammatory responses . Finally , we utilized transgenic mice that are deficient in the gp91 Phox subunit of the NADPH oxidase . These mice are a close model for the genetic disorder chronic granulomatous disease ( CGD ) , and are highly susceptible to Aspergillus infection . In the Triamcinolone model , histopathology of A . fumigatus inoculated mice show lesions with a strong influx of immune cells ( blue ) and strong growth of fungal hyphae ( green ) invading into the lung parenchyma from the airways ( Figure 2A , Figure S2 ) . By day 3 p . i . , hypoxia could readily be detected in the center of larger lesions ( red ) , and while the hypoxic areas of the lesions are comparable on day 3 . 5 p . i . , they are significantly expanded on day 4 p . i . . The isotype control staining of the same lesions in subsequent tissue sections , as well as complete staining of tissue sections from inoculated mice without hypoxyprobe injections , demonstrate the specificity of the hypoxyprobe and antibodies utilized ( Figure 2A ) . In contrast to the Triamcinolone model , lesions in the chemotherapeutic model are dominated by massive fungal growth causing significant tissue necrosis with minimal inflammation ( Figure 2B , Figure S2 ) . Given the severe neutropenia associated with this model , this result is expected . Despite the strong reduction in the inflammatory response and extensive fungal growth in this model , we were able to detect hypoxia in these lesions at similar time points to the Triamcinolone model ( Figure 2A and B ) . However , it is clear that the amount and extent of hypoxia is significantly reduced in this murine model . In mice that lack the gp91phox component of NADPH oxidase ( a model of X-CGD ) the lesion size gradually increased during the time course of infection from day 3 to day 5 , which was due to a strong increase in the inflammatory response of the host [45] . Fungal growth in this model was strongly reduced in comparison to the other two tested murine models ( Figure 2C , Figure S2 ) . On day 3 p . i . minimal amounts of hypoxia were detected , but by day 4 p . i . significant levels of hypoxia were observed in the center of the lesions . On day 5 p . i . hypoxia was abundantly present in the center of the lesions , and throughout the surrounding tissue indicating that significant parts of the lung experience hypoxia in this murine model of IPA ( Figure 2C ) . Indeed , in some animals in this model almost the entire lung seemed hypoxic at later time points just prior to mortality ( data not shown ) . Taken together , these data confirm that A . fumigatus encounters hypoxic microenvironments ( oxygen concentrations ≤1% ) and a dynamic range of oxygen availability during murine models of IPA . The extent of hypoxia , fungal growth , and host immune responses in the different models suggests that the host inflammatory response plays an important , but not exclusive , role in the generation of the hypoxic microenvironment . Given the evidence that A . fumigatus encounters hypoxia during IPA and that production of ethanol occurs in vivo during infection and is normally used by microbes to survive in low oxygen environments [34] , [35] , [36] , [37] , [38] , [39] , we next tested the hypothesis that ethanol fermentation was a key mechanism for hypoxia adaptation and fungal virulence . To determine the effects of ethanol fermentation on IPA pathogenesis , we searched the A . fumigatus genome sequence for genes encoding enzymes known to be involved in ethanol fermentation [46] , [47] . Using the A . nidulans pdcA ( pyruvate decarboxylase ) ( An_pdcA – AN4888 ) gene sequence for a BLASTX search of the A . fumigatus genome ( CADRE genome database ) we identified three potential candidates that may encode for pyruvate decarboxylases and named them pdcA ( AFUB_038070: 85% identity to An_pdcA ) , pdcB ( AFUB_096720: 39% identity to An_pdcA ) , and pdcC ( AFUB_062480: 33% identity to An_pdcA ) ( Table 1 ) . Protein sequence analysis ( InterProScan Sequence Search , http://www . ebi . ac . uk/Tools/InterProScan/ ) suggested all three A . fumigatus proteins were pyruvate decarboxylases ( Table 1 ) . In a similar manner , the known alcohol dehydrogenase ( Adh ) gene sequences of A . nidulans ( AdhI: An_alcA – AN8979 , AdhII: An_alcB – AN3741 , and AdhIII: An_alcC – AN2286 ) were used to identify the most likely genes encoding alcohol dehydrogenases in A . fumigatus . However , as the Aspergillus fumigatus genome contains several predicted alcohol dehydrogenases we restricted our search to proteins with high identity and similarity to the A . nidulans Adh proteins ( alcA – AFUB_087590: 87% identity and 94% similarity to An_alcA , alcB – AFUB_089920: 79% identity and 89% similarity to An_alcB , and alcC – AFUB_053780: 80% identity and 91% similarity to An_alcC ) ( Table 1 ) . Given the findings that A . fumigatus utilizes ethanol fermentation in in vitro and possibly in vivo hypoxic environments and the apparent gene redundancy in the predicted ethanol fermentation pathway , we next sought to determine which of the genes transcriptionally responds to hypoxia . Quantitative real-time PCR comparing the mRNA abundance of the ethanol fermentation genes under hypoxic and normoxic conditions revealed an immediate increase in mRNA abundance of all three pdc genes as well as the alcC gene to hypoxic growth conditions ( Figure 3 ) . While mRNA abundance levels of pdcB and pdcC show a ∼9-fold higher normalized fold expression after 24 h in hypoxia , pdcA mRNA showed a 64-fold increase compared to normoxic culture conditions . This data suggest that PdcA is the primary pyruvate decarboxylase that responds to hypoxia in A . fumigatus . With regard to the alcohol dehydrogenase encoding genes , the mRNA abundance of alcC significantly increased in response to hypoxia while mRNA abundance of the other two alcohol dehydrogenase encoding genes did not ( Figure 3 ) . These data suggest that alcC is the primary gene encoding an alcohol dehydrogenase that responds to hypoxia in A . fumigatus . To determine whether these genes are involved in ethanol fermentation , we generated null mutants of the genes encoding PdcA , PdcB , PdcC , and AlcC by replacement of the coding sequence in A . fumigatus strain CEA17 with the auxotrophic marker pyrG from A . parasiticus ( Figure 4 and data not shown ) . A pdcA/pdcB double mutant was also generated . Ectopic re-introduction of the wild type pdcA and alcC allele into ΔpdcA and ΔalcC ( resulting in strains pdcA recon and alcC recon ) allowed us to attribute all resulting phenotypes specifically to the absence of pdcA or alcC . The genotype of all strains was confirmed with PCR analyses ( data not shown ) and Southern blots ( Figure 4 and data not shown ) . Southern blot analysis of the alcC recon strain revealed a double insertion of the alcC encoding sequence in the genome and the alcC recon strain displayed a 10-fold higher mRNA abundance in response to hypoxia as the alcC allele in the wild type strain ( data not shown ) . However , the double insertion had no detectable phenotypic effect on the reconstituted strain , since the alcC recon strain showed the same phenotype as the wild type in all further experiments with only a slight but statistically insignificant increase in ethanol production ( Figure 5B ) . Next , we examined the ability of the generated null mutant strains to produce ethanol in response to in vitro hypoxic growth conditions . The loss of pdcA decreases Pdc enzyme activity in hypoxia by approximately 80% ( Figure 5A ) and the activity can be restored to wild type levels in the pdcA recon strain . The ΔpdcB and ΔpdcC , as well as the ΔalcC strain showed no significant decrease in Pdc activity ( data not shown ) , confirming the hypothesis that pdcA is the most important pdc gene in A . fumigatus for production of ethanol , at least in vitro . However there is still residual activity detectable in the ΔpdcA and the ΔpdcA/ΔpdcB strains ( 0 . 0047 ± 0 . 0041 in normoxia and 0 . 0039 ± 0 . 0064 in hypoxia; data not shown ) . A triple mutant of all three putative PDC encoding genes would need to be generated to definitively answer whether the observed residual activity from the cell free extracts is indeed real Pdc activity . Using the culture supernatants from the above experiments we examined the amount of ethanol produced by the respective fungal strains . The pdcB and pdcC null mutant strains show a slight but statistically insignificant decrease in ethanol production that is essentially similar to wild type levels ( CEA10: 0 . 071±0 . 035%; ΔpdcB: 0 . 047±0 . 002; ΔpdcC: 0 . 040±0 . 006; P>0 . 4 ) . Importantly , no ethanol could be detected in ΔpdcA and ΔalcC culture supernatants , while the wild type and respective reconstituted strains produced ethanol ( Figure 5B ) . These results support the gene expression and Pdc enzyme activity assays that suggest PdcA is the primary Pdc and that AlcC is the primary alcohol dehydrogenase required for in vitro ethanol production in A . fumigatus . The function of the remaining Pdc and Alc genes in A . fumigatus thus is not currently clear . To determine whether ethanol fermentation is important for growth under hypoxic conditions , we examined radial growth on solid media under normoxic and hypoxic conditions . As previously described , A . fumigatus grows well under hypoxic conditions on the fermentable carbon source glucose [18] , [31] ( Figure 5C ) . Surprisingly , the ethanol fermentation deficient mutants show no growth defect on glucose containing media under hypoxic ( 1% or 0 . 2% O2 ( data not shown ) ) conditions compared to the wild type and the reconstituted strains ( Figure 5C ) . In addition , the wild type and mutant strains are all also able to grow on the non-fermentable carbon sources ethanol , lactate and glycerol , although the growth rate is decreased compared to growth on glucose ( data not shown ) . Germination rates were the same for all strains utilized and no defects in conidia viability were observed ( data not shown ) . Liquid biomass quantification with the respective mutant strains also revealed no growth differences between wild type and ethanol deficient strains in hypoxia ( data not shown ) . Taken together , these results suggest that PdcA and AlcC are the primary enzymes involved in ethanol fermentation in A . fumigatus , but that other unidentified mechanisms are utilized to grow under hypoxic conditions on fermentable carbon sources when ethanol fermentation is not possible . Despite the general lack of a pathogenesis associated phenotype of the in vitro ethanol production deficient strains , ethanol itself has been observed to have significant immunomodulatory properties [48] , [49] , [50] , [51] , [52] . In addition , utilizing quantitative real-time PCR we found that alcC is expressed in vivo during fungal pathogenesis on day 3 and 4 post inoculation in the triamcinolone model ( Figure 6 ) suggesting that this gene and the enzyme it encodes may be important for in vivo growth . Therefore , we sought to determine the effects of loss of PdcA and AlcC on the pathogenesis of IPA . We first examined the virulence of the ΔpdcA and ΔalcC strains in the chemotherapeutic and X-linked chronic granulomatous disease ( X-CGD , gp91phox−/− mice ) murine models of IPA [45] , [53] . Irrespective of the fungal strain , A . fumigatus infected mice , in both models , displayed well described symptoms of A . fumigatus infection including hunched posture , ruffled fur , weight loss , and increased respiration as early as day +2 of inoculation . Subsequently , no difference in mortality was observed between the null mutant ( ΔalcC , ΔpdcA , and ΔpdcAΔpdcB ) and wild type strains ( Figure 7 and data not shown ) . To further examine the impact of the mutant strains on the pathogenesis of IPA in these murine models , lung histopathology was performed . Lungs from X-CGD mice displayed the expected histopathology for this model including a large inflammatory response with reduced fungal growth ( Figure S3 ) . In this model , histopathology of wild type and ΔalcC inoculated mice looked identical at all time points examined ( Figure S3 ) . In the chemotherapy model , pulmonary lesions of wild type infected animals show substantial fungal growth and invasion of the lung parenchyma with a minimal influx of immune cells but extensive tissue necrosis , hemorrhaging , edema , and tissue damage ( Figure 8A ) . Importantly , mice inoculated with ΔalcC show lesions with reduced fungal growth and more inflamed tissue compared to wild type inoculated mice ( despite the overall lesion sizes being comparable between the two inoculation groups ) ( Figure 8A ) . Taken together , this result suggests that AlcC plays a potential role in the pathogenesis of IPA . To explore this hypothesis further , we utilized the Triamcinolone ( corticosteroid ) model of IPA . IPA in mice treated with corticosteroids have previously been observed to induce hyper-inflammatory responses that are speculated to be the primary cause of mortality in this model [54] . Thus , we rationalized that any changes in the inflammatory response to A . fumigatus in the absence of AlcC would be potentiated in this model . As in the chemotherapeutic and X-CGD murine models , ΔalcC infected animals displayed wild type levels of mortality in the Triamcinolone model ( Figure 7C ) . A similar change in gross histopathology of the Triamcinolone compared to the chemotherapeutic model infected with the ΔalcC strain is also observed ( Figure 8B ) . Consistent with the observations in the chemotherapy model , ΔalcC inoculated animals show less fungal growth but increased levels of inflammation ( Figure 8B ) . Altogether these observations suggest that loss of AlcC results in an increased inflammatory response to A . fumigatus . To further characterize and quantify these histopathology observations , we analyzed the cellular infiltrates in BAL fluids of Triamcinolone treated mice from 2 different inoculation experiments using flow cytometry and differential cell counts . On day 3 p . i . , ΔalcC inoculated mice show an increased quantity of F4/80+/CD11c+ cells and a significant increase in GR-1+/CD11b+ cells in the BAL fluids compared to wild type infected animals ( Figure 9A and B , and Figure S4 and S5 ) . F4/80+/CD11c+ cells most likely represent macrophages while GR-1+/CD11b+ cells are most likely neutrophils . As expected , control mice BALs contained F4/80+/CD11c+ cells but nearly no GR-1+/CD11b+ cells . Furthermore , differential cell counts of BAL fluids revealed that macrophages , monocytes and particularly neutrophils were the dominant cell types found in the BAL samples of A . fumigatus inoculated mice . Consistent with the observed histopathology as well as the flow cytometry data , differential cell count numbers of neutrophils were significantly increased in ΔalcC inoculated mice ( p<0 . 05; Figure 9C and D ) . Taken together , these data support the histopathology observations that indicate an increased inflammatory response in mice infected with ΔalcC . To further quantify the differences in immune response to ΔalcC , we examined the production of cytokines normally associated with neutrophil recruitment in mice ( murine IL-8 homologs , KC and MIP2 ) . We observed that protein levels of the two murine neutrophil chemo-attractants KC and MIP-2 were significantly increased in BALs from ΔalcC inoculated animals compared with wild type ( Figure 10A and B; p<0 . 05 ) . IL-6 was also slightly elevated while TNF-α protein levels were reduced in comparison to BAL fluids of mice inoculated with the wild type ( Figure 10C and D; p>0 . 05 ) . Altogether , these results indicate that loss of AlcC modulates the immune response of the host to A . fumigatus causing increased recruitment of immune effector cells to the site of infection , particularly neutrophils , and associated altered cytokine responses . The comparison of histopathology between wild type and ΔalcC inoculated mice suggested reduced fungal growth by the ΔalcC strain in both the chemotherapy and the Triamcinolone murine models ( Figure 8 ) . In order to confirm this important observation , we quantified the pulmonary fungal burden on days 3 and 4 p . i . by quantitative RT-PCR . Consistent with the GMS histopathology , qRT-PCR confirmed a reduced pulmonary fungal burden in mice inoculated with ΔalcC compared to wild type ( Figure 11 ) . In addition , we examined LDH ( lactate dehydrogenase ) and Albumin release in BAL fluid to determine the degree of tissue damage caused by ΔalcC and the wild type strain . Intriguingly , both strains cause the same levels of LDH and Albumin release on day 3 and 4 post inoculation ( Figure 12 ) . Collectively , the lower fungal burden , the increased host inflammatory response , and the wild type level of tissue damage in response to ΔalcC strongly suggest the A . fumigatus alcohol dehydrogenase , AlcC , plays an important role in IPA . The observed altered host response and reduced fungal burden in animals infected with ΔalcC led us to question the mechanism behind these phenotypes . As inflammatory responses to fungal pathogens are often mediated by the fungal cell wall , we tested whether loss of AlcC resulted in unexpected changes to the cell wall of this strain that could account for the in vivo phenotypes . Conidia and ultraviolet ( UV ) irradiated germlings or hyphae from A . fumigatus wild type or ΔalcC were co-incubated with bone marrow-derived macrophages ( BMM∅ ) and inflammatory cytokine responses were measured ( Figure 13 ) . No differences in the secretion of TNF-α or MIP-2 by BMM∅ were observed to any of the tested A . fumigatus growth stages with regard to ΔalcC or wild type strains . Moreover , no difference in the response to chemical cell wall perturbing agents ( caspofungin and congo red ) was observed with ΔalcC ( data not shown ) . Thus , our data suggest that the increased inflammatory response observed in ΔalcC inoculated mice is not caused by changes in the fungal cell wall . Thus , the exact mechanism for the altered pathogenesis in mice inoculated with ΔalcC remains to be determined .
Metabolic adaptability and flexibility are important attributes for pathogens to successfully colonize , infect , and cause disease in a wide range of hosts . Importantly , these processes are dynamic , and pathogen and host metabolism are likely to change as the result of the host-pathogen interaction , which alters the localized microenvironment . In this study , we present new insights into the pathogenesis of IPA in commonly used experimental murine models . We present data that confirms previous circumstantial data that suggested that hypoxia may be part of the pathogenesis of IPA [55] , [56] . To our knowledge , this is the first confirmation of the occurrence of in vivo hypoxic microenvironments in an invasive fungal infection . Our results thereby add a “new” in vivo stress to which human fungal pathogens must adapt to cause lethal disease , and it will be intriguing to define the occurrence of hypoxia in other models of human fungal disease . To determine whether hypoxia actually occurs in the lung during IPA , we used the hypoxia marker , pimonidazole hydrochloride , which is an investigational oncology probe used as a hypoxia-imaging agent in clinical studies to detect reduced oxygen concentrations in animal and human tumors [43] , [44] , [57] , [58] , [59] . In our study , we observed that lesions in lungs of mice infected with A . fumigatus are hypoxic , as evidenced by the formation of a stable adduct between reduced pimonidazole and host proteins at sites of A . fumigatus infection . However , the extent and timing of hypoxia development differed between the immunologically distinct murine models of infection . While hypoxia did occur to some degree in all three models , the chemotherapy model exhibited the least amount of hypoxia in terms of size of the hypoxic area in the lung . This result suggests that the influx and activity of host cells is a strong contributor to the development of hypoxia . However , the persistence of hypoxia in this model , albeit not as extensive as in the other models , also suggests that fungal activities/components contribute to hypoxic lesion . For example , a recent study has observed that A . fumigatus can modulate host angiogenesis by secretion of secondary metabolites such as gliotoxin , which may further compromise tissue perfusion and ultimately contribute to coagulative necrosis , and thus limit oxygen delivery to sites of infection [60] . Importantly , though hypoxia was not detected on day +1 or day +2 of infection in any of our models we cannot rule out , and indeed would expect , that conidia and growing hyphal tips experience low oxygen tensions as they are engulfed by various host cells and ultimately penetrate the lung parenchyma and invade into the vasculature . Thus , we conclude that during colonization and subsequent infection , A . fumigatus is exposed to a dynamic range of oxygen levels in the lung . The significance of hypoxia , and the timing and extent to which it occurs during IPA , remain important areas of investigation . One important question that we have started to explore in this study is related to the potential clinical significance of fungal mechanisms of hypoxia adaptation . Previous studies in our and other laboratories with fungal SREBPs have suggested that fungal adaptation to hypoxia is critical for virulence . If true , these mechanisms would become an attractive therapeutic target . However , SREBPs are transcription factors that likely modulate numerous important mechanisms of fungal physiology , and thus it is not currently possibly to attribute the avirulence phenotype of these strains solely to their inability to grow in hypoxia . In general , mechanisms of hypoxia adaptation in human fungal pathogens are unexplored . Most eukaryotic cells , like A . fumigatus , obligatorily use oxygen to carry out many of their biochemical reactions . Oxygen is a key component of energy production where it functions as a terminal electron acceptor in the formation of ATP from glucose during aerobic respiration . When exposed to microenvironments with limited levels of oxygen , many microorganisms utilize fermentation as a potential metabolic mechanism for dealing with the lack of oxygen [34] , [35] , [36] , [37] , [38] , [39] , [40] . Fermentation allows the fungus to replenish sources of NAD+ and thus to generate ATP through continued use of glycolysis . Importantly , our interest in hypoxia and fungal pathogenesis began with the discovery of ethanol in BAL samples from A . fumigatus infected mice immunosuppressed with our chemotherapeutic model ( Figure S1 ) . Thus , in this study , we explored the potential role of ethanol fermentation in A . fumigatus hypoxia adaptation and fungal virulence by identifying and characterizing the ethanol fermentation pathway genes in this pathogen . Our in vitro molecular genetic analyses strongly suggest that the main route of ethanol fermentation in A . fumigatus is through the pyruvate decarboxylase , PdcA , and subsequent alcohol dehydrogenase , AlcC . Null mutants in both of these genes were unable to produce detectable ethanol in vitro under hypoxic conditions . These results are in agreement with observations in Aspergillus nidulans , where a pdcA deletion strain also fails to produce ethanol [61] and alcC activity is induced by hypoxic conditions [62] . However , our results also suggest that ethanol fermentation per se is not required for fungal growth in vitro as none of the ethanol fermentation deficient strains displayed any growth differences in in vitro normoxic or hypoxic growth conditions . We cannot definitively rule out that a small amount of undetectable ethanol fermentation still occurs in our mutant strains , however , we feel it is more likely that other fermentation pathways exist and/or that sufficient mitochondrial respiration still occurs under the conditions examined to support robust growth . Despite the persistent growth of the ethanol fermentation deficient stains under hypoxia , the ΔalcC displayed a very different phenotype in vivo in our murine models of IPA . In our three immunologically distinct models of IPA , no difference in mortality could be observed in mice infected with the wild type and ethanol fermentation mutant strains . However , histopathology of the chemotherapy and Triamcinolone models indicated an increased influx of immune cells and reduced fungal growth in ΔalcC inoculated mice . These observations were confirmed by flow cytometry and differential cell counts as well as quantitative fungal burden measurements by qRT-PCR . Although , we observed less fungal burden in ΔalcC inoculated mice , the overall lesion size was comparable to wild type caused lesions and both strains caused the same level of tissue damage as measured by LDH and Albumin assays . This result is probably due to the increased influx of neutrophils and macrophages to sites of ΔalcC infection . It is tempting to speculate then that ΔalcC inoculated mice might succumb to the infection because of the increased host inflammatory response rather than by the invasive growth of the mold . The decrease in fungal burden in mice infected with the ΔalcC strain might suggest that this response is partially effective at limiting fungal growth , but with collateral damage to the host that results in mortality . Of note , Mehrad et al . recently observed that an overproduction of KC results in a lower fungal burden and higher levels of neutrophil recruitment in a murine model of IPA , which leads to more resistance to A . fumigatus infections [63] . As we observed increases in KC and MIP2 levels in response to ΔalcC with a concomitant decrease in fungal burden , it may be possible that the increased inflammatory response to ΔalcC is at least partially antifungal . The in vivo phenotype of ΔalcC raises some intriguing questions about the mechanism behind the observed increase in host inflammatory response and subsequent reduction in fungal burden . Previous observations have indicated that ethanol is a potent immunosuppressive agent , and thus it seems reasonable to hypothesize that loss of ethanol production at the site of infection at least partially explains the observations with ΔalcC [48] , [49] , [52] , [64] , [65] , [66] , [67] , [68] . Acute and chronic ethanol exposures have been shown to alter the immune response to both bacterial and viral pathogens [69] , [70] , [71] . Ethanol decreases clearance of pneumococci and Klebsiella species from the lungs of ethanol-fed mice , which is mainly due to an impaired response of the phagocytic cells [72] . With regard to fungal pathogens , Zuiable et al . observed that human blood monocytes incubated with ethanol have impaired killing of Candida albicans [51] . Along these lines of thinking it may be possible that A . fumigatus is able to partially suppress localized host immune responses by utilizing ethanol fermentation in response to hypoxic microenvironments during IPA . However , to confirm this hypothesis , more sensitive detection methods for the localized and low levels of ethanol produced at the site of A . fumigatus infection are needed . Currently , ethanol detection in our murine models is inconsistent as exhibited by our initial experiment with BALs and 1H-NMR . BALs only sample the airway and do not sample localized infection sites located in the lung parenchyma so it is potentially not surprising that this method may not consistently detect a small molecule in the lung such as ethanol . To overcome this potential limitation , we attempted to utilize whole lung homogenates and two different ethanol detection methods including an enzymatic based approach and a GC-MS approach . Unfortunately , either the complexity of the samples , the chemical nature of ethanol itself , or the metabolism at the site of infection prevented reproducible detection of ethanol . Thus , currently , we cannot directly attribute the increased inflammatory response observed with ΔalcC to loss of ethanol production . However , development of more sensitive detection methods is underway in our laboratory . It is intriguing to note that the possibility of fermentation being important for hypoxic growth during fungal infections is further supported by the finding of ethanol in cerebral tissue of rats infected with C . neoformans [73] . Moreover , in support of the hypothesis that it is at least a secreted factor that is affecting the host response to ΔalcC , UV killed wild type and ΔalcC strains at 3 distinct growth phases do not exhibit a difference in pro-inflammatory responses ex vivo with bone marrow derived macrophages ( Figure 13 ) . Thus , the most likely culprit for an altered inflammatory response , the fungal cell wall , appears to not be altered in ΔalcC . Future studies will continue to probe the mechanism behind the reduced fungal growth and increased inflammatory response of ΔalcC . Altogether , in this study we present the first in vivo observations of hypoxic microenvironments during an invasive pulmonary fungal infection and shed light on how the mold A . fumigatus adapts to low oxygen environments to cause disease . These results , along with other published data from our laboratory , continue to support the hypothesis that hypoxia adaptation and growth is an important component of the pathogenesis of IPA [29] , [31] , [74] . Our results further emphasize the dynamic and complex interactions that occur between fungi and their hosts during an invasive pulmonary fungal infection . Future studies will continue to explore the effects of infection localized microenvironment stresses on invasive pulmonary aspergillosis pathogenesis . It will be intriguing to learn if other human fungal pathosystems also involve significant levels of hypoxia at sites of infection and whether ethanol fermentation pathway mutants also alter the host response .
A . fumigatus strain CEA17 ( gift from Dr . J . P . Latgé , Institut Pasteur , Paris , France ) was used to generate the pdcA ( AFUB_038070; ΔpdcA::A . parasiticus pyrG pyrG1 ) , pdcB ( AFUB_096720; ΔpdcB::A . parasiticus pyrG pyrG1 ) , pdcC ( AFUB_062480; ΔpdcC::A . parasiticus pyrG pyrG1 ) , and alcC ( AFUB_053780; ΔalcC::A . parasiticus pyrG pyrG1 ) null mutant strains . A . fumigatus strain CEA17 is a uracil-auxotrophic ( pyrG1 ) mutant of A . fumigatus strain CEA10 [75] , [76] . In this study , we used CEA10 ( CBS144 . 89 ) ( gift from Dr . J . P . Latgé , Institut Pasteur , Paris , France ) as the wild type strain in all experiments except the 1H-NMR metabolite profiling experiment which utilized strain AF293 , ΔpdcA , ΔpdcB , ΔpdcC , ΔalcC , and the ectopic complemented control strains , pdcA recon ( ΔpdcA::A . parasiticus pyrG + pdcA ) and alcC recon ( ΔalcC::A . parasiticus pyrG + alcC ) . All strains were routinely grown in glucose minimal medium ( GMM ) with appropriate supplements as previously described [77] at 37°C . To prepare solid media 1 . 5% agar was added before autoclaving . Generation of the pdc null mutants and the alcC null mutant in A . fumigatus strain CEA17 were accomplished by replacing the ORF of the target genes with A . parasiticus pyrG . The replacement construct was generated by cloning a sequence homologous to the gene locus into plasmid pJW24 ( donated by Dr . Nancy Keller , University of Wisconsin – Madison ) . Homologous sequences , each ∼1 kb in length and 5′ and 3′ of the gene coding sequence , were cloned to flank A . parasiticus pyrG in pJW24 . The resulting plasmids , pPDCAKO , pPDCBKO , pPDCCKO , and pALCCKO , were used as templates to amplify a disruption construct ( 3 . 6–4 . 7 kb ) for use in fungal transformation . To complement the ΔpdcA and ΔalcC strains the genes were amplified using genomic DNA of CEA10 as template and primers ∼1 kb 5′ and ∼500 bp 3′ of the gene coding sequence . The PCR products were cloned in front of the hygromycin B resistance gene into plasmid pBC-hygro ( Silar 1995 , obtained from the Fungal Genetics Stock Center , Dr . Kevin McCluskey ) using SpeI and NotI restriction sites [78] , [79] . The resulting plasmids , pBC-hyrgo-PDCA and pBC-hygro-ALCC , were used as template to amplify complementation constructs ( ∼7 . 4 kb ) , which were used in a fungal transformation and selection was for colonies able to grow on media containing 150 μg/ml hygromycin B . The primers utilized in vector construction are presented in Table S1 . Standard fungal protoplast transformation was used to generate mutant and reconstituted strains as previously described [31] . Transformants were initially screened by PCR to identify potential homologous recombination events at the gene locus using primers designed to amplify only the mutated gene locus ( Table S1 ) . Single conidia of each transformant were prepared and screened by PCR to eliminate the chance of heterokaryons . Homologous recombination was confirmed by Southern analysis with the digoxigenin labeling system ( Roche Molecular Biochemicals , Mannheim , Germany ) as previously described [80] . Strains were grown on or in GMM at 37°C . Normoxic conditions were considered general atmospheric levels within the lab ( ∼21% ) . For hypoxic conditions two different devices were used , a Biospherix C-Chamber with O2 levels controlled by a PRO-Ox controller and CO2 levels controlled with PRO-CO2 controller ( Biospherix , Lacona , NY , USA ) and an INVIVO2 400 Hypoxia Workstation ( Ruskinn Technology Limited , Bridgend , UK ) . For these experiments , the O2 set point was 1% and the CO2 set point was 5% . Oxygen levels were maintained with 94% N2 and a gas regulator . Colony growth was quantified as previously described [31] . For normoxic samples , strains were grown in GMM with 1x106 conidia/ml , 300 rpm , at 37°C for 16 hrs . 25 ml of the normoxic culture were mixed with 15 ml of fresh GMM and incubated for an additional 24 hrs under hypoxic conditions ( 130 rpm , 37°C ) . Mycelia of normoxic and hypoxic cultures were harvested , rinsed twice with distilled water , transferred to 2 ml screw cap tubes with 0 . 1 mm glass beats , immediately frozen in liquid nitrogen and weighed . After adding 1 ml of extraction buffer ( 100 mM KH2PO4 , 2 mM MgCl2 , and 1 mM DTT ) , the samples were twice placed in a mini beadbeater ( Biospec products , Bartlesville , OK , USA ) for 30 sec with 5 min on ice in between . After centrifugation ( 13 , 000 rpm , 20 min , 4°C ) the cell free extracts were transferred to new , cold reaction tubes and kept on ice until use . The protein concentration of the cell free extracts was defined by using the Coomassie Plus – The Better Bradford Assay Kit ( Pierce , Rockford , IL , USA ) following the method recommended by supplier . Enzyme activity was determined using a method adapted from Lockington et al . 1997 [61] . The assay volume was adjusted to 200 µl for use in 96-well plates . 25 µl cell free extract ( sample and control ) or 25 µl extraction buffer ( blank ) were added to wells in duplicates and then 175 µl of the sample mix ( 50 mM histidine-HCl , 0 . 35 mM MgCl2 , 0 . 35 mM TPP , 67 mM pyruvate , 6 U yeast ADH , 0 . 5 mM fresh NADH , added up with water to 175 µl ) were added to the samples and the blank , and 175 µl of the control mix ( sample mix without pyruvate ) were added to the cell free extract as controls . The rate of decrease in absorbance at 340 nm was followed in a Spectramax Plus ( Molecular Devices , Sunnyvale , CA , USA ) , measuring every 10 sec , at 37°C over 5 min after mixing for 1 sec . The pyruvate decarboxylase activity was calculated as described by http://cmbe . engr . uga . edu/assays/pyruvatedecarboxylase . pdf . The calculation had to be adjusted to the reduced length of the light path in the 96-well plate by multiplying the molar extinction coefficient for NADH ( 6 . 22 L/mmol for a path length of 1 cm ) with 0 . 788 . Experiments were done in three separate biological replicates and the mean and standard error calculated with Prism software version 5 . 0b ( GraphPad Software Inc . ) . To detect ethanol in the culture supernatant of in vitro fungal cultures , high performance liquid chromatography ( HPLC ) was performed using a Shimadzu system ( Kyoto , Japan ) , consisting of a solvent delivery module , a low pressure gradient pump unit , a degasser , an autoinjector , a column oven and a refractive index . The column used for the analytical separation was the Aminex Fermentation Monitoring column ( 150 mm×7 . 8 mm , BioRad , Hercules , CA ) . The mobile phase consisted of 0 . 001 M H2SO4 , the flow rate was 0 . 8 ml/min , the column temperature was 60°C , and the sample injection volume was 25 µl . As external standard ethanol solutions with known concentration ( 2 , 1 , 0 . 5 , 0 . 1 , 0 . 05 , 0 . 01 ( v/v ) % ) were used in every experiment and a standard curve was generated and used to determine concentration . Data was normalized to mycelial dry weight . In addition , a Shimadzu ( Kyoto , Japan ) QP2010 GC/MS with an electron ionization ( EI ) source was used for metabolite separation and identification . A 30 m 0 . 25 mm id 0 . 25 um film thickness , RTX-5MS ( 5% Diphenyl - 95% dimethyl polysiloxane ) fused silica capillary column from Restek ( Bellefonte , PA ) was used for all separations . The GC column was temperature programmed as follows: 5 min isothermal at 100°C , then raised at 20°C/minute to 120°C , and held for 30 seconds . Helium gas served as the carrier gas at a flow rate of 0 . 73 ml/min . Split injections were performed at a 1 to 20 ratio . The injection port was held constant at 200°C . The interface temperature was set at 200°C and the ion source at 200°C . EI fragmentation was performed scanning from 40 to 400 at 0 . 2 seconds/scan . The instrument was calibrated with Perfluorotributylamine ( PFTBA ) prior to analysis . Standards of ethanol were analyzed for rention time and a response curve using a 3 point serial dilution . These response curves were used to calculate detected compounds in each sample . One micro-liter of each sample and standard was used for analysis . Identification was matched to NIST 21 and NIST 107 libraries commercially purchased as well as secondary confirmation with standards , previously mention , purchased from Sigma ( Saint Louis , MO ) . For the culture samples , strains were grown as described above in the pyruvate decarboxylase assay description . After 24/48/72/96 hrs 2 ml of the culture supernatant were transferred into sterile reaction tubes on ice , and filtered through a Millipore membrane filter ( 0 . 45 µm , Millipore , Yonezawa , Japan ) into HPLC vials ( Sun Sri , Rockwood , TN ) . Experiments were done in three biological replications . Conidia from freshly harvested GMM plates were inoculated in 5 ml GMM in a 6-well plate to a concentration of 1×107 conidia/ml . Cultures were grown aerobically for 24 h . For normoxic growth , cultures were maintained in atmospheric conditions . For hypoxic growth , cultures were placed in the hypoxic chamber for 24 h . Fungal mats were flash frozen in liquid nitrogen and lyophilized prior to disruption using a bead beater . To assess fungal gene expression in vivo , the triamcinolone immunosuppression model was utilized as described below . Mice were sacrificed on day 3 and 4 post inoculation , and lungs were harvested and immediately frozen in liquid nitrogen . Samples were freeze-dried and homogenized with glass beads on a Mini-Beadbeater ( BioSpec Products , Inc . , Bartlesville , OK , USA ) . Total RNA was extracted using TRIsure Reagent ( Bioline , Tauton , MA , USA ) according to the manufacturer's instructions . After treatment with DNase I ( Ambion , Austin , TX , USA ) , 500 ng of total RNA were used to generate first-strand cDNA with the reverse transcriptase kit ( Qiagen , Hilden , Germany ) . Real-time PCR assays were performed with 20 µl reaction volumes that contained 1x iQ SYBR green master mix ( Biorad , Hercules , CA , USA ) , 0 . 2 µM of each primer ( Table S1 ) , and 2 µl of a 1∶5 dilution of the cDNA using a Bio-Rad MyiQ single Color real-time PCR detection System with iCycler . No-RT controls for each primer set were also assayed to confirm that no DNA contamination was present , respectively . Real-time PCRs were performed in triplicates , and the expression levels of all genes of interest were normalized to ß–tubulin levels or tefA ( translation elongation factor alpha subunit ) levels for in vivo experiments . The thermal cycling parameters consisted of a 3-min Taq polymerase hot start at 95°C , followed by template amplification of 40 cycles of 95°C for 10 sec , 58°C for 30 sec . Fluorescence was measured during the annealing/extension step ( 58°C ) and a disassociation analysis ( melting curve ) was performed to confirm that a single amplified product was present . Following amplification , data was analyzed with the Bio-Rad iQ5 2 . 0 Standard Edition Optical System Software . The ΔΔCt method of analysis was used to determine fold changes of gene expression in the mutants relative to the wild type CEA10 strain . The virulence of the A . fumigatus strains was tested in three immunologically distinct murine models of invasive pulmonary aspergillosis . All animals were housed five per cage in an environment with HEPA filtered air , autoclaved food at libitum , and prophylactic treatment with antibiotic water containing clindamycin ( 150 mg/ml ) , vancomycin ( 1 mg/L ) and gentamicin ( 100 mg/ml ) . CD1 male and female mice , 6–8 weeks old were used in all experiments for the triamcinolone and chemotherapeuatc murine models . Mice were obtained from Charles River Laboratories ( Raleigh , NC ) or from a breeding colony located in the Animal Resources Center ( ARC ) at Montana State University . For the Chronic Granulomatous Disease murine model , 6–8 week old mice with a null allele corresponding to the X-linked gp91phox component of NADPH oxidase ( B6 . 129S6-Cybbtm1Din ) were bred in the ARC at Montana State University [45] . For the triamcinolone ( corticosteroid ) model mice were immunosuppressed with a single dose of Kenalog ( Bristol-Myers Squibb Company , Princeton , NJ , USA ) injected subcutaneously ( s . c . ) at 40 mg/kg 1 day prior to inoculation . For the chemotherapy model mice were immunosuppressed with intraperitoneal ( i . p . ) injections of cyclophosphamide ( Baxter Healthcare Corporation , Deerfield , IL , USA ) at 175 mg/kg 2 days prior to inoculation and with Kenalog injected subcutaneously ( s . c . ) at 40 mg/kg 1 day prior to inoculation . On day 3 post-inoculation ( p . i . ) , repeat injections were given with cyclophosphamide ( 150 mg/kg i . p . ) and on day 6 p . i . with Kenalog ( 40 mg/kg s . c . ) . For the detection of hypoxia in vivo , 15 unanesthetized mice inhaled 40 ml of an aerosolized suspension of 1×109 conidia/ml of A . fumigatus strain CEA10 . 6 uninfected control mice inhaled 40 ml of aerosolized 0 . 01% Tween 80 in a Hinners inhalational chamber for 45 min as previously described [81] . One A . fumigatus infected mouse was sacrificed immediately after infection , lungs were removed , homogenized and the number of CFU was determined ( ∼1×104 conidia per mouse ) . After hypoxyprobe injection ( see below ) , mice were sacrificed at set time points after A . fumigatus challenge and lungs were processed for hypoxyprobe immunohistochemistry . For survival studies and histopathology , 10 mice per A . fumigatus strain ( CEA10 , ΔpdcA , pdcA recon . , ΔalcC ) either inhaled 40 ml of an aerosolized suspensions of 1x109 conidia/ml ( control mice inhaled 40 ml of aerosolized 0 . 01% Tween 80 ) or the animals were inoculated intranasally with 1×106 conidia in 25 µl and monitored twice a day . Infection inoculum was prepared by growing the A . fumigatus isolates on GMM agar plates at 37°C for 3 days . Conidia were harvested by washing the plate surface with sterile phosphate-buffered saline 0 . 01% Tween 80 . The resultant conidial suspension was adjusted to the desired concentration by hemacytometer count . Mice were observed for 14 days after A . fumigatus challenge . Any animals showing distress were immediately sacrificed and recorded as deaths within 24 hrs . No mock infected animals perished in either murine model in all experiments . Lungs from all mice sacrificed on different time points during the experiment were removed for fungal burden assessment , infiltrate and cytokine analysis as well as histopathology . Animal experiments were all repeated in duplicate . 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 animal experimental protocol was approved by the Institutional Animal Care and Use Committee ( IACUC ) at Montana State University ( Federal-Wide Assurance Number: A3637-01 ) . For histopathology , CD1 mice were inoculated as described above , and sacrificed at set time points after A . fumigatus challenge . When mice were sacrificed , lungs were removed on that day . Lung tissue was fixed in 10% phosphate-buffered formalin , embedded in paraffin , sectioned at 5 µm , and stained with hematoxylin and eosin ( H&E ) or Gomori methenamine silver ( GMS ) by using standard histological techniques . Microscopic examinations were performed on a Nikon Eclipse 80i microscope and imaging system ( Nikon Instruments Inc . , Melville , NY , USA ) . A total of 3 mice were examined at each respective time point . The chemotherapeutic murine model of IPA was utilized in these experiments . Mice were inoculated in the Hinner's chamber with either 0 . 01% Tween 80 or A . fumigatus wild type strain AF293 . A total of 10 mice were used in each treatment group . On day +3 post infection , BALs were performed with each BAL a final total volume of ∼1 . 5 ml . Deuterated water was added to 600 μL of each BAL to provide a field frequency lock and an internal standard of 0 . 03% 3- ( Trimethylsilyl ) -Propionic acid-D4 sodium salt ( TSP ) was added to each sample to provide a chemical shift reference at 0 ppm . For 1H-NMR , a one-pulse sequence was used with a 2-second pre-saturation pulse and 7-second repetition time . The resulting one-dimensional spectra were compared using MestReC NMR analysis software ( Mestrelab Research ) to monitor the presence and absence of identifiable metabolites . Mice were intravenously injected with hypoxyprobe at a dose of 60 mg/kg weight of the mouse ( Hypoxyprobe Inc . , Burlington , MA , USA ) . After 60 to 90 min , mice were sacrificed by pentobarbital anesthesia ( 100 µg/g body weight ) followed by exsanguination . The left lung of each mouse was filled with OCT ( frozen tissue matrix ) and after embedding in OCT immediately frozen in liquid nitrogen . The lungs were cryosectioned into 5 µm sections and stored at −80°C until stained . After thawing , the sections are fixed in cold acetone ( 4°C ) for 15 min , followed by washing the sections ( PBS , 2×5 min ) and blocking with normal serum block ( NSB: PBS +10% goat serum +1 . 25% mouse serum ) at RT . After 30 min , sections were washed and incubated overnight at 4°C with the mouse monoclonal antibody FITC-Mab1 ( Hypoxyprobe-1 Plus Kit , Hypoxyprobe Inc . , Burlington , MA , USA ) diluted 1∶400 in NSB and with a rabbit polyclonal antibody to Aspergillus ( Abcam Inc . , Cambridge , MA , USA ) . Aspergillus isotype control slides were incubated only with FITC-Mab1 and hypoxyprobe isotype control slides only with the Aspergillus antibody . Isotype controls are a measure of unspecific staining of the secondary antibody . After another wash , sections were incubated for 60 min at room temperature with DyLight 594-conjugated mouse Anti-FITC ( Jackson ImmunoResearch Laboratories , West Grove , PA ) and AlexaFluor488-conjugated goat Anti-rabbit ( Invitrogen , Carlsbad , CA , USA ) diluted 1∶400 . After another washing step , prolong Gold antifade reagent with DAPI ( Invitrogen , Carlsbad , CA , USA ) was added to each section . Microscopic examinations were performed on a Zeiss Axioscope 2-plus microscope and imaging system ( Zeiss , Jena , Germany ) . For each time point , a total of 2 to 4 mice were examined and experiments were repeated in triplicate . Broncheoalveolar lavages ( BALs ) were performed by intratracheal instillation and extraction of 3 ml 1x PBS . Total lung lavage cell numbers were determined by hemacytometer count , spun onto glass slides , and stained with Diff-Quick ( Fisher Scientific , Pittsburgh , PA , USA ) for differential counting . For flow cytometry , BAL cells were centrifuged and resuspended in phosphate-buffered saline with 2% calf serum and an anti-mouse Fc receptor antibody ( Trudeau Institute , Saranac Lake , NY , USA ) to a concentration of approximately 107 cells/ml . The cells were then stained with a mixture of fluorophore-conjugated antibodies against mouse GR-1 ( APC-Cy7 ) ( BD Pharmingen , San Diego , CA , USA ) , F4/80 ( PE-Cy7 ) ( eBioscience , San Diego , CA , USA ) , CD11b ( AlexaFluor700 ) ( BioLegend , San Diego , CA , USA ) , and CD11c ( APC ) ( purified from the hamster cell line N418 ( ATCC , Manassas , VA , USA ) and fluorophore-conjugated using the AlexaFluor633 protein labeling kit ( Invitrogen , Carlsbad , CA , USA ) ) and then examined on a BD LSR flow cytometer ( BD Biosciences , San Jose , CA , USA ) . Analysis of cytometry data was performed with FlowJo software Version 8 . 8 . 7 DMG and numbers of relevant cell types were determined by combining flow cytometry data ( percentage of a given cell type ) with BAL cell counts . Data presented are the mean and standard error of N = 5 mice at each time point . The BD Cytometric Bead Array Mouse Inflammation Kit ( BD Biosciences , San Jose , CA , USA ) was used according to the manufacturers instructions to quantitatively measure IL-6 , IL-10 , MCP-1 , IFN-γ , TNF-α , and IL-12p70 protein levels in mouse BAL fluids utilizing a FACSCalibur flow cytometer ( Becton Dickinson , Mountain View , CA , USA ) . IL-17 , MIP-2 , KC , and VEGF levels in mouse BAL samples were determined using the mouse cytokine/chemokine Milliplex Map Kit ( Millipore Corporation , Billerica , MA , USA ) according to the manufacturers instructions and then examined and analyzed on the BioPlex 200 System ( Bio-Rad , Hercules , CA , USA ) . Data presented are the mean and standard error of N = 5 mice at each time point . In vivo lung tissue damage was determined by measurement of LDH and Albumin levels in mouse BAL samples by using a LDH assay ( CytoTox 96 Non-Radioactive Cytotoxicity Assay , Promega , Madison , WI , USA ) and an albumin assay ( Albumin ( BCG ) Reagent Set , Eagle Diagnostics , Cedar Hill , TX , USA ) according to the manufacturers' instructions . To assess fungal burden in lungs , the triamcinolone immunosuppression model was utilized as described above . Mice were sacrificed on day 3 and 4 post inoculation , and lungs were harvested and immediately frozen in liquid nitrogen . Samples were freeze-dried , homogenized with glass beads on a Mini-Beadbeater ( BioSpec Products , Inc . , Bartlesville , OK , USA ) , and DNA extracted with the E . N . Z . A . fungal DNA kit ( Omega Bio-Tek , Norcross , GA , USA ) . Quantitative PCR was performed as described previously [82] . Bone marrow cells were eluted from the tibias and femurs of 8–12 week old C57BL/6 mice , lysed of red blood cells , and cultured in RP20 ( RPMI , 20% FCS , 5 mM HEPES buffer , 1 . 1 mM L-glutamine , 0 . 5 U/ml penicillin , and 50 mg/ml streptomycin ) supplemented with 30% ( v/v ) L929 cell supernatant ( source of M-CSF ) . Bone marrow cells were plated in a volume of 20 ml at a density of 2 . 5×106 cells/ml in 10 ml Petri dishes . The medium was exchanged on day 3 . Adherent bone marrow-derived macrophages ( BMMØs ) were harvested on day 6 . Cells were washed and plated in 0 . 2 ml RP10 at a density of 5×105 cells/ml in 96 well plates and stimulated for 18 hours with conidia ( 5×105 per well ) , UV irradiated germlings ( 105/well ) , or UV irradiated hyphae ( 2×104/well ) prepared as described before [83] . After 18 hours co-culture supernatants were collected for ELISA . Commercially available ELISA kits for TNF ( BD Biosciences , San Jose , CA , USA ) and MIP-2 ( R&D Systems , Minneapolis , MN , USA ) were used according to the manufacturers' instructions .
|
Metabolic flexibility is important for human pathogens like Aspergillus fumigatus as it allows adaptation to dynamic infection induced microenvironments . Consequently , identification of fungal metabolic pathways critical for in vivo growth may uncover novel virulence mechanisms and new therapeutic opportunities . To date , the mechanisms used by A . fumigatus to adapt to microenvironments in immunosuppressed mammalian hosts are poorly understood . In this study we discover that A . fumigatus is exposed to oxygen limiting microenvironments during invasive pulmonary aspergillosis ( IPA ) . Thus , this result builds on growing evidence that suggests hypoxia is a significant in vivo stress encountered by human fungal pathogens . We tested the hypothesis that genes encoding enzymes involved in ethanol fermentation are important for in vivo fungal responses to hypoxia . We consequently observed a significant increase in the inflammatory response that correlated with reduced fungal growth in the lungs of mice inoculated with an alcohol dehydrogenase null mutant . Altogether , our study suggests that fungal responses to in vivo hypoxic microenvironments can directly affect host immune responses to the invading fungal pathogen . A better understanding of these mechanisms will increase our understanding of IPA and other human diseases caused by fungi and could potentially lead to improved therapeutic options .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"mycology",
"biology",
"microbiology",
"host-pathogen",
"interaction",
"pathogenesis"
] |
2011
|
In vivo Hypoxia and a Fungal Alcohol Dehydrogenase Influence the Pathogenesis of Invasive Pulmonary Aspergillosis
|
It is generally accepted that neuroendocrine cells regulate dense core vesicle ( DCV ) biogenesis and cargo packaging in response to secretory demands , although the molecular mechanisms of this process are poorly understood . One factor that has previously been implicated in DCV regulation is IA-2 , a catalytically inactive protein phosphatase present in DCV membranes . Our ability to directly visualize a functional , GFP-tagged version of an IA-2 homolog in live Caenorhabditis elegans animals has allowed us to capitalize on the genetics of the system to screen for mutations that disrupt DCV regulation . We found that loss of activity in the transcription factor PAG-3/Gfi-1 , which functions as a repressor in many systems , results in a dramatic up-regulation of IDA-1/IA-2 and other DCV proteins . The up-regulation of DCV components was accompanied by an increase in presynaptic DCV numbers and resulted in phenotypes consistent with increased neuroendocrine secretion . Double mutant combinations revealed that these PAG-3 mutant phenotypes were dependent on wild type IDA-1 function . Our results support a model in which IDA-1/IA-2 is a critical element in DCV regulation and reveal a novel genetic link to PAG-3-mediated transcriptional regulation . To our knowledge , this is the first mutation identified that results in increased neurosecretion , a phenotype that has clinical implications for DCV-mediated secretory disorders .
Dense core vesicles ( DCVs ) mediate the storage and secretion of hormones ( e . g . insulin ) and neuropeptides ( e . g . vasoactive intestinal peptide , VIP ) from a variety of neuroendocrine cells ( e . g . pancreatic β-cells ) . The biogenesis of DCVs and cargo is coordinately regulated , allowing secretory cell types to adjust their output in response to nutrient ( e . g . glucose ) and endocrine signals ( e . g . glucagon ) ( reviewed in [1] , [2] ) . The molecular and cellular mechanisms by which neuroendocrine cells achieve this coordinated secretory regulation are poorly understood . Recent work suggests that the DCV-associated protein Islet Antigen-2 ( IA-2 , a . k . a ICA512 ) and its paralog , IA-2β ( a . k . a . Phogrin ) , might hold clues to these regulated secretory processes . IA-2 and IA-2β are members of a family of protein tyrosine phosphatases ( PTP ) that lack catalytic activity [3]–[8] . Both IA-2 and IA-2β are transmembrane proteins localized to DCVs and are present in most , if not all , neuroendocrine tissues [6] , [9] . Mouse knockouts of either IA-2 or IA-2β result in impaired glucose-induced insulin secretion and elevated responses to glucose tolerance tests [10] , [11] . The IA-2 and IA-2β double knockout mice have enhanced phenotypes , including glucose intolerance [12] and infertility in female mice due to reduced luteinizing hormone secretion [13] . Over expression of IA-2 in insulin-expressing MIN6 cells markedly increases glucose- or K+-induced insulin secretion , the number of secretory vesicles , the levels of vesicle-associated proteins ( such as synaptotagmin and VAMP2 ) , and the insulin content suggesting that IA-2 might be involved in the stabilization of DCVs [14] . Finally , recent work suggests that IA-2 can be cleaved in pancreatic β-cells upon glucose stimulation . This cleavage releases a cytoplasmic tail of IA-2 that modulates the activity of STAT5 , a transcription factor regulating the transcription of the insulin gene and other DCV components [15] . Together , these results implicate IA-2 and IA-2β as key players in DCV and associated cargo abundance in pancreatic β-cells . To further explore the role of IA-2/IA-2β in neuroendocrine secretion , we have studied its functions in the model organism C . elegans . C . elegans is a genetically tractable system with a simple nervous system ( 302 neurons ) that has made major contributions to our understanding of synaptic vesicle ( SV ) [16] , [17] and DCV trafficking and neuroendocrine secretion [18]–[21] . It has previously been shown that the gene ida-1 encodes the single C . elegans factor related to both mammalian IA-2 and IA-2β , and that IDA-1 is an evolutionarily conserved neuroendocrine protein that is involved in acetylcholine release and functions in the insulin-like signaling pathway [7] , [18] , [22] . Our DCV-specific IDA-1::GFP integrated transgenic strain ( KM246 ) has provided a unique marker for studying DCV function using internal reflection fluorescence microscopy and direct electrophysiological assays [21] . We report here on the characterization of a mutant isolated in a genetic screen for abnormal patterns of IDA-1::GFP reporter gene expression . The identified allele , gv560 , results in markedly increased levels of gene expression for multiple DCV components , increased DCV numbers , and exhibits several behavioral phenotypes linked to increased neuronal secretion . This mutation maps to the gene pag-3/Gfi-1 , which encodes a zinc-finger transcription factor that functions as a repressor in many biological systems ( reviewed in [23] ) . We further show that the effects of pag-3 mutations on DCV biosynthesis are IDA-1-dependent , suggesting that PAG-3-mediated repression is part of a regulatory mechanism governing DCV numbers and cargo release .
In order to investigate the molecular mechanisms regulating DCV biosynthesis and turnover in vivo , we screened a mutagenized population of C . elegans animals for altered patterns of IDA-1::GFP ( Figure 1A–E ) . The wild type IDA-1::GFP reporter strain we used ( KM246 ) harbors a full length translational fusion of IDA-1 to GFP driven by a portion of its own promoter ( Figure 1A ) that results in strong gfp expression in a subset of larval and adult neurons; the endogenous distribution of IDA-1 is more widespread than this particular reporter , and is perhaps ubiquitous within all neurons and neurosecretory cells of the animal [18] , [24] . Specifically , we observed GFP in the following neurons in strain KM246 at the frequencies indicated parenthetically: ALA ( a single neuron of unknown function that extends a processes the length of the body ( 100% ) ) , VCs ( hermaphrodite specific Ventral Cord motor neurons involved in egg laying ( 100% VC4&VC5; 31% VC6; 26% for VC1-3 ) , the HSNs ( Hermaphrodite Specific Neurons regulating egg laying ( 29% ) ) , and PHCs ( tail neurons , possibly mechanosensory ( 18% ) ) ( Figure 1B , C , E ) [18] . A screen of 4 , 400 haploid genomes revealed three mutations that significantly altered this reporter gene expression pattern; two had additional developmental defects and were not maintained . The remaining mutant allele , gv560 , was homozygous viable and was fully penetrant for multiple phenotypes , including significantly up-regulated levels of IDA-1::GFP levels in a subset of neurons where it is normally observed ( Figure 1B–E ) . Some gv560 mutant neurons consistently displayed increased levels of GFP ( eg . VC4 , VC5 , and ALA ) whereas other neurons ( eg . VC1∼3 , VC6 , HSNs , and PHCs ) had a more variable pattern . These differences roughly correlated with the degree of mosaic expression normally observed for this integrated transgene in a wild type background , suggesting these differences might simply reflect the functional efficiency of the promoter element used in this particular reporter . In addition , the gv560 allele resulted in extra presumptive neurons along the posterior ventral midline ( Figure 1D ) and an uncoordinated ( Unc ) phenotype , particularly when moving in the reverse direction . All gv560 phenotypes were recessive . The location and appearance of the extra neurons in gv560 animals suggested they might be surviving P cell lineage descendants that normally are eliminated during larval development by cell death [25] . A reporter gene driven by the lin-11 promoter ( Plin-11::GFP ) has previously been used to identify these neurons in cell death mutants [26] . We crossed gv560 mutant animals into a strain harboring Plin-11::GFP . While the Plin-11::GFP is only present in VC1∼VC6 neurons in wild type animals ( Figure 1F ) , we found that this reporter was also active in many of the extra gv560 mutant cells ( Figure 1G ) [26] . This result strongly suggested that most , if not all , extra neurons observed in homozygous gv560 animals were VC-like neurons derived from Pn . aap-like cell lineages that had failed to undergo normal programmed cell deaths . The failure of programmed cell deaths in ced-3 mutants also results in extra VC-like neurons [27] . Therefore , we crossed ced-3 ( n717 ) mutants with our IDA-1 reporter and assayed GFP levels . IDA-1::GFP was not increased in the VC or VC-like neurons of ced-3 mutants when compared to wild type animals ( Figure 1H ) . Quantitative analysis of IDA-1::GFP intensity in VC4 and VC5 neurons showed that the GFP signal in gv560 is ∼3 . 4-fold and ∼5 . 6-fold of wild type respectively , while the GFP signal in ced-3 mutant is similar to wild type ( Figure 1I ) . Using genetic and single nucleotide polymorphism ( SNP ) markers , we mapped the gv560 mutant to the right end of the X chromosome ( Figure 2A ) ( see Materials and Methods ) . There are three known mutations in this region that result in an Unc phenotype; unc-3 , pag-3 , and unc-7 . One of these , pag-3 , has previously been shown to result in extra VC-like neurons due to cell lineage alterations [26] . We found that gv560 failed to complement two different pag-3 alleles ( Figure 2B ) . Moreover , a 6 . 5 kb genomic fragment from the pag-3 genomic locus rescued the gv560 mutant phenotypes ( Figure 2C , D ) , indicating that gv560 is a mutant allele of pag-3 . The pag-3 gene encodes a zinc-finger ( Zn-finger ) transcription factor that was originally identified in C . elegans due to altered gene expression patterns in specific neurons ( pattern of gene expression abnormal , pag-3 ) [28] . More recently , pag-3 mutations have been shown to affect cell fate decisions within certain lineages resulting in the generation of additional posterior neurons [26] . Sequence analysis of the pag-3 gene in the C . elegans gv560 mutant animals revealed a C to T transition in exon 7 resulting in a S223F amino acid change within the fourth Zn-finger coding region ( Figure 2E ) . Multiple sequences alignment showed that S227 in PAG-3 is evolutionarily conserved from C . elegans to humans ( data not shown ) . Examination of two previously identified pag-3 alleles ( is20 and n3098 ) ( Figure 2E ) , demonstrated that each had the same phenotype as gv560 with respect to IDA-1::GFP levels , either alone or in trans-heterozygous combination with gv650 ( Figure 2B ) . Because pag-3 ( ls20 ) ( Figure 2E ) has previously been shown to behave genetically as a null allele [28] , we conclude that gv560 is also genetically null; we can not eliminate the possibility that it retains some molecular function . To determine if PAG-3 activity acted cell autonomously , we sought to rescue the GFP over expression phenotype in a subset of neurons expressing IDA-1::GFP in the pag-3 ( gv560 ) mutants . The cat-1 gene encodes a synaptic vesicular monoamine transporter [29] that has been reported to be expressed in several neurons ( Shawn Lockery , personal communication; http://chinook . uoregon . edu/promoters . html ) , including VC4 , VC5 and the HSNs where we see IDA-1::GFP phenotypes in the pag-3 ( gv560 ) mutant . We confirmed this expression pattern using a ∼3 kb region upstream of the cat-1 coding region to drive GFP ( data not shown ) ; this reporter was not expressed in ALA . The Pcat-1 promoter was placed upstream of a wild type , full-length cDNA encoding PAG-3 and used to make stable , extrachromosomal transgenic strains in the Pida-1::IDA-1::GFP; pag-3 ( gv560 ) mutant background . Interestingly , we found that this transgene was often toxic , resulting in several strains in which most transgenic animals arrested at the L1 stage of development and subsequently died . However , we were also able to generate multiple , independent stable strains in which transgenic animals grew normally , presumably reflecting expression of the transgene at a sub-toxic dose . We scored GFP levels in these healthy strains , focusing on the VC4/5 versus ALA over expression phenotypes because these cells always express IDA-1::GFP . A visual screen revealed that 28% ( n = 100 ) of Pida-1IDA-1::GFP; pag-3 ( gv560 ) mutants harboring the Pcat-1::PAG-3 transgene had a reduction in VC4/5 GFP to near wild type levels whereas GFP over expression in ALA remained unaffected ( Figure 2F , G ) . We did not expect full rescue as the extrachromosomal rescuing transgene is mitotically unstable . A quantification of GFP intensity in VC4 and VC5 for 15 selected animals of each strain similarly showed a reduction ( 31% ) in average intensity in the rescued strain ( Figure 2H ) . These results demonstrated that wild type PAG-3 activity in VC4 and VC5 was likely acting cell autonomously , with no change in the GFP over expression phenotypes seen in neurons that did not express the Pcat-1::PAG-3 transgene . To quantify the increased IDA-1::GFP levels in pag-3 ( gv560 ) mutants , equal amounts of protein extract from gv560 animals were compared to the parental reporter strain on Western blots probed with anti-GFP and anti-IDA-1 antibodies . We found that GFP levels in the gv560 mutant were approximately five times higher than that in the parental controls when internally compared to a tubulin protein control ( Figure 3A ) . Interestingly , the IDA-1 peptide containing protein levels in gv560 mutant , derived from both the endogenous gene and the reporter fusion transgene , were nearly eight-fold higher than in controls . This demonstrated that the changes in the reporter accurately reflected alterations in endogenous IDA-1 protein levels and not merely changes in the reporter gene expression . To probe whether the effect of pag-3 ( gv560 ) was limited to IDA-1 protein or more generally applicable , we assayed five additional neurosecretory components for which specific C . elegans antibodies were available: ( 1 ) synaptotagmin ( SNT-1 ) , an integral membrane protein of synaptic vesicles ( SVs ) [30] that has also been implicated in DCV cargo release [31]; ( 2 ) synaptobrevin ( SNB-1 ) , an integral membrane protein of SVs implicated in neurotransmitter release [32] and for which mammalian data demonstrates its presence in DCVs [33] . ( 3 ) RAB-3 , a member of the Ras GTPase superfamily that regulates the axonal distribution of SVs and DCVs and , consequently , neurosecretion [22] , [34]; ( 4 ) UNC-31 , a calcium-activated protein for secretion ( CAPS ) homolog required for DCV docking and cargo release ( Livingstone , 1991 , Ph . D . Thesis ) [21] , [35] , [36]; and ( 5 ) serotonin , a DCV cargo component present in HSN neurons and several neurons in the head region [37] . We found that the levels of SNT-1 and serotonin are substantially increased in gv560 mutants ( Figure 3A , B ) , whereas the SNB-1 level is more moderately increased . In contrast , the level of proteins lacking vesicle transmembrane motifs ( e . g . RAB-3 and UNC-31 ) and the control ( α-tubulin ) did not change in the gv560 mutant ( Figure 3A ) . It was possible that the increases in endogenous IDA-1 , SNT-1 , and serotonin observed in pag-3 ( gv560 ) mutants were due to the extra VC-like ( and perhaps other undetected ) neurons in these mutants . To address this , we checked protein levels in the mutant ced-3 that also results in extra VC-like neurons ( and other cell types ) due to a lack of programmed cell death [27] . There was no detectable increase in any assayed protein in ced-3 ( n717 ) mutants compared to wild type ( data not shown ) , suggesting that observed protein increases in the pag-3 ( gv560 ) mutant are due to up-regulation or increased stability of these factors rather than the increased neuronal cell number . The significantly increased IDA-1 protein levels in the pag-3 ( gv560 ) mutant prompted us to explore DCV components more thoroughly in ida-1 mutants alone . We assayed by Western blot the levels of DCV-related proteins ( IDA-1 , SNT-1 , SNB-1 , RAB-3 , and UNC-31 ) in wild type animals and an ida-1 ( ok409 ) mutant using tubulin as a control ( Figure 4A ) . IDA-1 was undetectable , as expected , in the predicted null allele of ida-1 [18] . Of the remaining proteins tested , SNT-1 and RAB3 showed a slight reduction ( ∼80% of the N2 control , p<0 . 05 ) in ida-1 ( ok409 ) mutants ( Figure 4A ) . We also assayed serotonin levels in ida-1 ( ok409 ) mutants as an example of a DCV cargo component and found a significant reduction ( ∼50% of the N2 controls , p<0 . 05 ) ( Figure 4B ) . Interestingly , the increased levels of SNT-1 and serotonin found in pag-3 ( gv560 ) animals are also eliminated when IDA-1 activity is removed ( Figure 4B–D ) . A comparison between ida-1 ( ok409 ) ;pag-3 ( gv560 ) double mutant protein levels to ida-1 ( ok409 ) mutants alone demonstrates that ida-1 is epistatic to pag-3 for the change in both SNT-1 and serotonin . We concluded that loss of PAG-3 activity results in an up-regulation of at least some vesicle-related components and that this up-regulation also requires functional IDA-1 . Previously described roles of PAG-3/Gfi-1 and related proteins in transcriptional repression ( see review in [23] ) , coupled with the increased levels of DCV component proteins in the gv560 mutant , led us to examine DCV and SV component gene expression in these mutant animals . We selected four genes whose products are DCV-associated proteins , four genes that encode DCV cargo proteins , four genes expected to function primarily in SV pathways , and three non-vesicle related genes as controls ( Table 1 ) . We used quantitative real-time polymerase chain reaction ( RT-PCR ) assays to measure the expression of these 15 genes in young adult wild type and pag-3 ( gv560 ) mutant animals relative to internal controls . We found that pag-3 ( gv560 ) mutants showed significantly increased expression of three genes ( Table 1 ) , including two that encode DCV-membrane associated proteins ( IDA-1 and SNT-1 ) and one encoding DCV cargo related protein ( INS-1 or insulin-like molecule 1 ) . There were no changes in the expression of SV genes and the three non-vesicle associated genes . Interestingly , the DCV membrane-associated gene ida-1 showed the greatest magnitude of change in pag-3 ( gv560 ) mutants ( Table 1 ) . Because the up-regulation of at least some of the vesicle-related proteins in pag-3 ( gv560 ) is dependent on the presence of wild type IDA-1 , we thought it possible that a feedback mechanism could link vesicle biogenesis , utilization , or stability to gene expression . Previous work has shown that a cleaved cytosolic tail of mammalian IA-2/IDA-1 interacts and modulates the transcription factor STAT-5 that , in turn , positively regulates insulin gene expression [15] . This link between IA-2 and the transcription of a DCV cargo-encoding gene in mammals prompted us to investigate in C . elegans the role of ida-1 and sta-1 ( encoding the lone STAT-related factor ) in DCV homeostasis . We first examined the mRNA levels of all 15 genes listed in Table 1 by quantitative real-time PCR in ida-1 mutants; we found no significant differences compared to wild type levels . We also crossed ida-1 ( ok409 ) into the pag-3 ( gv560 ) mutant background and repeated the gene expression analysis . We found that the mRNA levels in the pag-3 ( gv560 ) ;ida-1 ( ok409 ) double mutants remained the same as in the pag-3 ( gv560 ) single mutant ( except for ida-1 itself , data not shown ) , demonstrating that transcriptional alterations caused by loss of PAG-3 are not dependent on IDA-1 activity . Finally , we crossed the presumptive null deletion allele sta-1 ( ok587 ) into the IDA-1::GFP reporter strain , either alone or in combination with pag-3 ( gv560 ) , and assayed IDA-1 and IDA-1::GFP protein levels . The absence of STA-1 activity did not change any IDA-1 protein phenotypes . Our data strongly suggest that the presence or absence of IDA-1 alone has no effects on the transcriptional regulation of the DCV and SV genes we examined in C . elegans nor does the absence of STAT activity influence vesicle-associated protein levels . Because the up-regulation of DCV components observed in pag-3 ( gv560 ) is dependent on wild type IDA-1 activity , whereas changes in gene expression are not , we conclude that the role of IDA-1 in DCV regulation is post-transcriptional . To determine if the changes in gene expression and protein levels observed in pag-3 ( gv560 ) mutants had behavioral consequences , we tested two behaviors that reflect vesicle ( DCV and SV ) cargo release: egg-laying and aldicarb resistance . Egg ( embryo ) laying by gravid hermaphrodites is a complex neuroendocrine behavior involving serotonin and acetylcholine released by VC and HSN neurons acting on the vulval muscles [38] . The rate of egg laying is dependent on many variables , including animal age , population density , and food supply . To eliminate some of these variables , individual young gravid hermaphrodites were picked into single wells of a 96-well micro titer plate containing M9 buffer . Exogenous serotonin ( 12 . 9 mM ) was added to stimulate wild type egg laying to a rate ( ∼8 eggs/hr/animal ) that could be quantified easily within a 60 minute assay and various mutant animals were tested . We found that pag-3 ( gv560 ) mutants had a significant increase ( ∼34% ) in egg-laying compared to wild type controls ( Figure 5A ) . In comparison , ida-1 ( ok409 ) single mutants and ida-1 ( ok409 ) ;pag-3 ( gv560 ) double mutants had a ∼33% and ∼26% reduction , respectively , in egg laying compared to wild type controls . As observed for protein level changes , the egg-laying phenotype of gv560 was dependent on wild type IDA-1 activity . To control for possible effects of the extra VC-like neurons present in gv560 animals on egg-laying , we also tested laying rates in ced-3 ( n717 ) mutants that also have extra VC-like neurons; ced-3 mutant rates of egg laying were comparable to wild type animals ( data not shown ) . These results suggested that increased neurosecretion from the HSNs , and/or the VCs , and not the presence of extra VC-like neurons , likely underlies the enhanced egg-laying phenotype of gv560 mutants in response to serotonin . As a second test of neurosecretion , we used the acetylcholinesterase inhibitor aldicarb to assess acetylcholine ( Ach ) release and presynaptic function [39] , [40] . Ach release is traditionally considered an assay of SV function , not DCVs . However , it has been previously shown that disruption of DCV function impaired , directly or indirectly , the release of Ach [18] , [35] , [41] , [42] . We used a population growth rate assay in which we counted the number of progeny from parental animals after 96 hr of growth on plates containing various concentrations of aldicarb [18] , [39] . This assay provides a more sensitive and quantitative measure of aldicarb sensitivity compared to scoring paralysis alone . We found that pag-3 ( gv560 ) mutants showed greater sensitivity to aldicarb than wild type animals ( Figure 5B ) , suggesting that these mutants release more Ach than wild type controls . In contrast , the ida-1 ( ok409 ) mutant animals showed resistance to aldicarb as previously reported [18] due to disruption of DCV function and decreased neurosecretion . Again , the gv560 aldicarb phenotype was suppressed in double mutant combination with ida-1 ( ok409 ) . The increased rates of egg-laying and increased sensitivity to aldicarb observed for pag-3 ( gv560 ) are both consistent with enhanced neurosecretion in these mutant animals . We also investigated the role of pag-3 in dauer formation , an alternative part of the life cycle in C . elegans that is regulated ( in part ) by an insulin-like signaling pathway dependent on proper neuroendocrine secretion . We compared dauer formation in wild type , pag-3 ( gv650 ) and temperature-sensitive ( ts ) daf-28 ( sa191 ) mutants , each reared at an elevated temperature of 25°C . The wild type and pag-3 mutant animals showed no appreciable dauer formation at this temperature ( Figure 5C ) . In contrast , the daf-28 mutant harboring a ts allele of the insulin-like protein it encodes resulted in 40% dauers due to reduced signaling through the insulin-like signaling pathway at this temperature [43] . We found that the daf-28 ( sa191 ) ;pag-3 ( gv650 ) double mutant resulted in a ∼75% reduction in dauer formation at 25°C compared to daf-28 ( sa191 ) alone ( Figure 5C ) demonstrating enhanced signaling through the pathway . This result is consistent with our egg-laying and aldicarb assays and together suggested that loss of PAG-3 activity results in enhanced neuroendocrine secretion affecting many cell types . Previous cell culture work showed that DCV numbers are increased 2∼3-fold following over expression of mammalian IA-2 in mouse MIN6 cells , as assayed by vesicle counts from electron micrographs [14] . We have demonstrated here that IDA-1/IA-2 is up-regulated in pag-3 mutants , perhaps similarly resulting in increased DCV numbers and increased neurosecretion . Although technically challenging in C . elegans , we sought to determine by electron microscopy if the overexpression of ida-1 observed in pag-3 ( gv560 ) mutants was accompanied by increased numbers of DCVs per neuron . Wild type and mutant animals were fixed , sectioned and examined by electron microscopy ( see Materials and Methods ) . We focused on the ventral nerve cord region and identified presynaptic regions by serial sections; individual neurons in any section were not precisely identified . We found that the average density of presynaptic region DCVs in pag-3 ( gv560 ) was about twice the value of that in wild type controls ( Figure 6A , B , E ) . We also quantitated DCV numbers in ida-1 ( ok409 ) mutants and found a 50% reduction compared to wild type ( Figure 6C , E ) . Of particular interest , we found that the pag-3 ( gv560 ) ;ida-1 ( ok409 ) double mutants eliminated the increased DCV numbers observed in the pag-3 ( gv560 ) single mutant ( Figure 6D , E ) . Our results demonstrate that presynaptic DCV numbers correlate with neurosecretion-based phenotypes in the mutants studied . That is , pag-3 ( gv560 ) mutants have increased numbers of presynaptic DCVs in ventral cord neurons and display phenotypes of enhanced egg-laying , enhanced aldicarb sensitivity , and enhanced signaling through the insulin-like pathway . All of these enhanced phenotypes are dependent on wild type IDA-1 activity .
We have capitalized on the genetics of C . elegans to identify factors that contribute to the regulation of dense core vesicles ( DCVs ) and neuroendocrine secretion . Using a reporter gene to mark DCVs in live animals , we carried out a genetic screen and identified a mutation that results in both over and ectopic expression of the reporter . The mutation , that behaves genetically as a recessive null allele , results in an amino acid change ( S223F ) in the fourth zinc finger of the transcription factor PAG-3/Gfi-1 . This mutation ( gv560 ) results in increased gene expression and levels of several DCV-associated proteins and increased numbers of DCVs in presynaptic regions of ventral cord neurons . These increases correlate with enhanced neurosecretory phenotypes in pag-3 mutants and are dependent on wild type IDA-1 , the lone C . elegans PTP-like factor related to mammalian DCV membrane proteins IA-2 and IA-2β . Although mutations in many other genes , such as unc-4 and unc-37 [44] , [45] , unc-86 [46] , [47] and the EGL-46 mammalian homolog INSM1/IA-1 [48]–[50] , have been shown to decrease SV and DCV components and numbers in various neurosecretory systems , this is the first report we are aware of in which a mutation results in DCV component up-regulation . The results of this and previous work suggest that DCV homeostasis is regulated by both stimulatory and inhibitory mechanisms that correlate with levels of IDA-1-related proteins . Our genetic studies demonstrate that the up-regulation of DCV protein components in pag-3 mutants is dependent on functional IDA-1 activity . It is possible that this dependence is indirect , with IDA-1 and PAG-3 acting in separate pathways of DCV regulation . For example , the enhanced neurosecretory phenotypes of pag-3 ( gv560 ) could be suppressed by loss of IDA-1 activity due simply the averaged effects ( one positive and one negative ) of the two individual mutant phenotypes , each acting in different regulatory cascades . We do not favor this explanation for two reasons . First , both ida-1 transcription and IDA-1 levels are elevated in pag-3 mutants . Second , where quantifiable , the genetic interactions between ida-1 and pag-3 mutants we observe strongly suggest an epistatic relationship . That is , the ida-1 ( ok409 ) ;pag-3 ( gv560 ) double mutants more closely resemble ida-1 ( ok409 ) alone ( with the exception of gene expression ) rather than a quantitative average of ida-1 ( ok409 ) and pag-3 ( gv560 ) effects together . These results strongly suggest that both IDA-1 and PAG-3 act in a common pathway to regulate DCV homeostasis . A model consistent with our results places IDA-1 in an important position in the modulation of DCV numbers and secretion by directly influencing their biogenesis , stability and/or utilization ( Figure 7 ) . Actions that increase IDA-1 ( e . g . loss of PAG-3 ) result in increased presynaptic DCV numbers and increased neurosecretion whereas actions that decrease IDA-1 ( e . g . loss of IDA-1 itself ) result in decreased DCVs and secretion . Loss of both PAG-3 and IDA-1 in double mutant animals results in a DCV phenotype resembling loss of IDA-1 . This demonstrates that effects of increased transcription of some DCV component genes , due to loss of PAG-3 , are masked by the loss of the post-transcriptional role of IDA-1 in influencing the biogenesis , stability , or utilization of DCVs . The molecular mechanism whereby IDA-1 influences DCV steady state levels is currently unknown and represents an important area for future research . Although IDA-1 activity is an important barometer of DCV homeostasis , it is worth noting that it is not essential for DCV formation and organism viability . Instead , its loss or hyper production in C . elegans results in an approximately 2-fold swing , down or up respectively , in DCV numbers . Thus , IDA-1 is part of a larger and as yet undefined , multi-component regulatory system that ensures DCV numbers and cargo reflect the neurosecretory demands of the cell . Our results are consistent with recent results from vertebrate studies , providing an organismal context for understanding IDA-1/IA-2 function in DCV regulation . Over expression of mouse IA-2 in insulin-expressing MIN6 cells markedly increased induced insulin secretion and the number of DCVs [14] . Studies by Solimena and colleagues have provided evidence that after fusion of the DCV with the plasma membrane during insulin secretion , the cytosolic tail of IA-2 ( containing the catalytically inactive PTP-like domain ) is cleaved by a calcium-induced and calpain-dependent mechanism [51] , [52] . They have proposed that this cleavage event initiates a retrograde signal in which the IA-2 cytosolic domain enters the nucleus to complex with the transcription factors STAT3 and STAT5 , thereby blocking their inactivation by de-phosphorylation [15] , [52] . In this model , IA-2 is a critical element that links DCV exocytosis to feedback regulation of transcription such that genes encoding IA-2 and other DCV components and cargo are up-regulated upon induced insulin secretion . We find in C . elegans that all up-regulated changes in DCV numbers and secretory behaviors due to pag-3 mutations are ameliorated by loss of IDA-1/IA-2 . However , loss of IDA-1 has no effect on the steady state mRNA levels of any vesicle-related genes we assayed ( listed in Table 1 ) , nor do we ever detect our IDA-1::GFP translation fusion protein in the nucleus . Thus , functional IDA-1 appears to be necessary post-transcriptionally for the coordinated regulation of DCV biogenesis and utilization , although we have no evidence that reveals the molecular mechanisms by which this occurs . We were unable to link these effects genetically to the single C . elegans STAT encoding gene sta-1 , suggesting that any feedback mechanisms that function in C . elegans are not entirely dependent on STAT activity . Similarly , both of the pancreatic β-cell-specific STAT3 and STAT5 knockout mice showed normal insulin content and islet mass [53] , [54] , suggesting that STAT3 and STAT5 are functionally redundant or are not critical elements for insulin-containing DCV regulation . It is also possible that other transcription factors act redundantly with the STATs , thereby masking the effects of STAT loss of function in this process in both C . elegans and mice . Regardless of the role of STAT activity in DCV regulation , our results reveal an additional , post-transcriptional mechanism whereby IDA-1 influences DCV homeostasis . Our work reveals a novel transcriptional link between IDA-1/IA-2 and DCV regulation that is mediated by PAG-3 . PAG-3 is a five C2H2-type zinc finger transcription factor and it is orthologous to vertebrate Gfi-1 ( growth factor independence-1 ) and Drosophila Senseless [28] , [55] , [56] . In the C . elegans nervous system , PAG-3 appears to function primarily as a transcriptional repressor [28] , a theme common to the functions described to date for its vertebrate and Drosophila orthologs [23] , [57] . Interestingly , we find a similar mis-regulation of our IDA-1::GFP reporter gene in unc-3 mutant animals ( unpublished ) . UNC-3 was recently identified as PAG-3 interacting protein and is an ortholog of the Olf-1/Early B cell factor family of transcription factors involved in neuronal cell development [58] . It is possible that PAG-3 and UNC-3 act together as part of a common transcriptional complex that can influence , directly or indirectly , the expression of genes whose products are necessary for DCV homeostasis . However , the relationship between the activities of these two factors appears to be complex . UNC-3 has not been reported in the VC4 and VC5 neurons [58] that show mis-regulation of IDA-1::GFP in the pag-3 mutant background . In addition , our rescue experiments demonstrate that PAG-3 acts cell autonomously in VC4 and VC5 with respect to IDA-1 up-regulation . Finally , unc-3 and pag-3 mutants have different cell lineage transformations underlying their extra VC and VC-like neuron phenotypes [58] . Taken together , these results suggest that UNC-3 and PAG-3 may act at multiple times in the cell lineage giving rise to these cells , perhaps both together and independently . Characterizing the neurosecretory phenotypes of unc-3 mutants and exploring the molecular mechanisms that might link UNC-3 and PAG-3 are challenges for future studies . Given the dramatic up-regulation of ida-1 gene expression and other DCV-associated gene products in pag-3 mutants , it is tempting to speculate that PAG-3 directly represses ida-1 and other DCV component gene expression . We have used the known binding site of PAG-3 in vitro , as well as its related mammalian factors , to search bioinformatically for promoters containing matching sequences . Although several of the DCV genes we searched , including ida-1 , do have sequences within their promoters that resemble the consensus Gfi-1 binding site [59] , our attempts ( and that of others ) to demonstrate direct PAG-3 binding in vitro have not been successful . Therefore , the question of whether the effects of pag-3 mutations on DCV component gene expression are direct or indirect remains unanswered . Identification of PAG-3 as a factor involved in DCV regulation may have clinical relevance . Recent studies show that mammalian Gfi-1/PAG-3 is involved in the regulation of neuroendocrine cell development and controls neuroendocrine cancer growth [60] , [61] ( also see review by [62] ) . Gfi-1 co-expression with neuroendocrine markers , including the DCV-related protein chromogranin A and calcitonin peptide , in small cell lung carcinoma ( SCLC ) strongly suggests its involvement in the maintenance of the neoplastic phenotype of neuroendocrine lung tumors [60] . We have shown previously that IA-2 is strongly expressed in many small cell lung carcinoma ( SCLC ) cells with a neuroendocrine phenotype , but not in non-neuroendocrine carcinomas [63] . In fact , IA-2 is expressed in many other , if not all , human neuroendocrine cancer cells examined such as insulinoma [64] and serotonin-secreting tumors of the midgut [65] . Our current results suggest that deregulation of Gfi-1/PAG-3 in neuroendocrine neoplastic disease may severely alter IA-2-related homeostatic control . This dysregulation may contribute directly to the tumor hypersecretion phenotypes that are caused by more than twenty of the different neuropeptides and hormones abnormally released from DCVs such as gastrin-releasing peptide , chromogranin A , atrial natriuretic peptide ( ANP ) , and growth hormone-releasing hormone [66]–[69] . Further studies are required to look for mutations in neuroendocrine pathway genes and to examine directly the role of Gfi-1/PAG-3 and IA-2 in SCLC and related neuroendocrine cancers . Our current study adds to a growing body of evidence linking IA-2-related protein levels to neuroendocrine secretion and highlights the usefulness of model organisms in dissecting basic biological mechanisms . For human diseases that might benefit from increased neurosecretion , such as diabetes mellitus , targeted knockdown of Gfi-1 function represents a potential therapeutic target . Although there is substantial evidence linking IA-2 protein levels to DCV numbers and secretion , the molecular mechanisms underlying DCV homeostasis and its link to transcriptional control remain unclear . The identification of PAG-3/Gfi-1 as a component of this regulatory system provides another clue for increasing our understanding of these molecular mechanisms .
C . elegans were grown at 20°C on NGM plates seeded with OP50 bacteria [70] unless otherwise noted . Strains and alleles used were as follows: wild type ( N2 ) , CB4856 , ida-1 ( ok409 ) , KM246{gvIs246[Pida-1IA-2::gfp]} , pag-3 ( ls20 ) , pag-3 ( n3098 ) , pag-3 ( gv560 ) ;gvEx[pag-3 ( + ) ] , ced-3 ( n717 ) , unc-3 ( e151 ) , unc-7 ( e5 ) , daf-28 ( sa191 ) , unc-31 ( e928 ) , unc-64 ( e246 ) and sta-1 ( ok587 ) . A N2 strain harboring Plin-11::GFP was kindly provided by Michael R . Koelle . Double mutants were confirmed by phenotype , molecular genotyping of specific alleles , or both . Transgenic L4 larvae of the strain harboring the IDA-1::GFP translational fusion reporter ( KM246 ) [18] were treated with ethyl methanesulfonate ( EMS ) as described by [70] . Mutagenized worms were placed on seeded 10-cm NGM agar plates and allowed to lay eggs overnight . 2200 individual F1 mutagenized animals were separated onto 550 plates and visually screened three days later as young adults for alterations in IDA-1::GFP in VC and ALA neurons , using a dissecting fluorescence microscopy . Among the mutants examined , we recovered a single penetrant mutation that showed increased IDA-1::GFP expression in neuronal cells . This recessive mutation , gv560 , was out crossed four times and mapped to LGX . Worms were harvested , washed twice with M9 buffer , and frozen in liquid nitrogen . For RNA preparation , worms were thawed at 65°C for 10 min , and RNA was isolated using the TRIZOL LS Reagent ( Invitrogen , Carlsbad , CA ) . Isolated total RNA was subjected to DNAse treatment and further purified using RNAeasy ( Qiagen , Valencia , CA ) . cDNA was prepared from 5 µg of total RNA in a 100 µl reaction using the SuperScript First-Strand Synthesis System ( Invitrogen ) . The cDNA was used in quantitative real-time PCR ( RT-qPCR ) using standard conditions . Primer pairs and probes ( sequences available upon request ) were diluted into 96-well plates at a concentration of 3 µM . Real-time amplification of the cDNA was performed using the TaqMan Universal Master Mix ( Applied Biosystems , Foster City , CA ) . All RT-PCR reactions were carried out and analyzed on an ABI-Prism 9700TH Sequence Detection System ( Applied Biosystems ) , according to the manufacturer's directions . Data were collected using RNA from three independent C . elegans populations . To determine the relationship between mRNA abundance and PCR cycle number , all primer sets were calibrated using serial dilutions of cDNA preparations . Relative abundance is reported as the mRNA abundance of each experimental gene relative to the mRNA abundance of several control genes . Western blots were performed using Amersham ECL Plus™ Western Blotting Detection Reagents ( GE Healthcare Bio-Sciences Corp , Piscataway , NJ ) . Immunostaining was performed using a modified protocol [18] . Antisera against an IDA-1 peptide corresponding to amino acid residues 47–61 ( CYSSESGSPEPTVLD ) was produced in rabbits and purified with Immobilized Protein G Agarose ( Pierce Biotechnology , Rockford , IL ) . Antisera against synaptotagmin ( antibody 1095 , kindly provided by M . Nonet , Washington University School of Medicine , St . Louis , MO ) , IDA-1/IA-2 ( J . Hutton , University of Colorado at Denver , CO ) , UNC-31 ( K . Miller , Oklahoma Medical Research Foundation , Oklahoma City , OK ) , and RAB-3 ( K . Iwasaki , National Institute of Bioscience and Human Technology , Ibaraki , Japan ) were used as described [71] . Antisera against serotonin were from H . Steinbusch ( Maastricht University , Maastricht , Netherlands ) and detected using the method of Loer ( http://home . sandiego . edu/̃cloer/loerlab/anti5htshort . html ) , anti-synaptobrevin ( SNB-1 ) from Developmental Studies Hybridoma Bank ( University of Iowa , IA ) , and anti-GFP antibodies from Clontech ( Palo Alto , CA ) . Antibodies were detected using rhodamine- or FITC-conjugated goat anti-rabbit or anti-mouse secondary antibodies ( Jackson ImmunoResearch , West Grove , PA ) . Animals were grown at 20°C and different life stages were analyzed from a growing population . For image acquisition , animals were incubated with 10 mM NaN3 in M9 buffer for 1 h and mounted on agar pads . Stacks of confocal images with 0 . 3 to 0 . 4 µm vertical pitch were recorded with identical exposure times using a Leica TCS SP2 microscope . Maximum intensity projections of all images from a given animal were generated using the NIH ImageJ software package ( http://rsbweb . nih . gov/ij/download . html ) . Individual young hermaphrodites were placed in 96 wells of a microtiter plate containing 12 . 9 mM serotonin creatinine sulfate ( Sigma ) in 100 µl of M9 buffer [72] . After 60 min incubation at room temperature , the eggs laid by each animal were counted . All assays were performed in triplicate ( n = 10 ) . Chronic effects of aldicarb on mutants were quantified in a growth assay as previously described [18] , [39] , [73] . Single L1 larva were placed on individual culture plates containing 0 , 10 , 25 , 50 , 75 , 100 , 200 , 300 , or 400 µM aldicarb and grown at room temperature ( 22°C ) for 96 hr . Growth was then stopped by putting the plates at 4°C . For each aldicarb concentration , the number of progeny was counted as a percentage of the total number of progeny produced on the no-drug control . Triplicate assays were performed and an N2 control set was included with each set of strains to allow for comparisons between different sets of assays . Adults were allowed to lay eggs on bacterially seeded plates for 3 hr at room temperature and progeny were scored after 48 hr at 25°C as indicated in results [18] , [74] . Dauers were distinguished by the distinctive body shape , darkly pigmented intestine , and a constricted pharynx . Each set of assays included all of the relevant strains and the actual ambient incubator temperature surrounding the plates was monitored with a digital thermometer ( Barnant Co . , Barrington , IL ) . Genetic mapping was carried out using single nucleotide polymorphisms ( SNPs ) [75] . The gv560 mutant allele was crossed into the CB4856 wild type strain [76] . F2 Unc progeny with extra VC-like neurons ( P9–P12 ) were scored for the presence of CB4856 SNP markers . Thirteen SNP markers were selected between the interpolated genetic map position 2 . 86∼23 . 92 on the X chromosome; pkP6114 ( C05C9 ) , pkP6060 ( C35C5 ) , pkP6130 ( B0198 ) , pkP6125 ( C23H4 ) , pkP6131 ( K02A4 ) , pkP6115 ( C05E7 ) , pkP6161 ( F11C1 ) , pkP6132 ( F46G10 ) , pkP6133 ( C49F8 ) , pkP6164 ( R03E1 ) , pkP6116 ( F33C8 ) , pkP6087 ( C33A11 ) , pkP6093 ( F01G12 ) . Detailed location and primer sequence of each SNP are provided at the Genome Sequencing Center ( http://genome . wustl . edu/genome/celegans/celegans_snp . cgi ) . The gv560 allele was mapped between SNP marker pkP6116 on cosmid F33C8 and pkP6093 on cosmid F01G12 . To confirm the gene location of gv560 , small PCR fragments ( 1 kb in size ) covering the ∼2-kb promoter region , entire coding region , and intron-exon junction regions of pag-3 were generated for sequencing ( ABI 373 DNA sequencer ) . Primer sequences are available upon request . Further confirmation of the pag-3 ( gv560 ) mutation was provided by rescuing the mutant phenotypes with a 6 . 5-kb genomic fragment covering the entire coding region of pag-3 that was amplified by long-PCR ( TaKaRa LA Taq DNA Polymerase , Takara Bio USA Corporate , Madison , WI ) . Animals were injected with the purified PCR products at 1–10 ng/µl together with the co-injection marker pRF4 ( rol-6 ( su1006 ) ) at 100 ng/µl and analyzed for rescue of IDA-1::GFP intensity and distribution . To determine whether the effect of pag-3 mutants on IDA-1::GFP expression is a cell autonomous , the full-length cDNA of pag-3 was specifically expressed in a limited number of neurons ( including VC4 and VC5 ) driven by the 3-kb promoter of cat-1 to generate Pcat-1PAG-3 in the PCRII-TOPO vector that was confirmed by sequencing . Ten N2 , pag-3 ( gv560 ) , pag-3 ( gv560 ) ;ida-1 ( ok409 ) , and ida-1 ( ok409 ) young adult hermaphrodites were subjected to high-pressure freezing . Animals were placed in a 100-µm depth specimen chamber ( part no . LZ 02316VN , Technotrade International ) , frozen rapidly at −176°C under high pressure ( pressure >2 , 100 bar ) in a Bal-Tec HPM010 apparatus . Frozen animals were subjected to chemical fixation and dehydration in a Reichart AFS apparatus ( Leica ) : incubation with 0 . 1% tannic acid and 0 . 5% glutaraldehyde in anhydrous acetone for 72 h at −90°C followed by several washes in acetone over 6 h . Samples were moved to 2% osmium tetradioxide in anhydrous acetone for 4 h at −90°C before raising the temperature to −20°C at 5°C/h . Samples were then incubated for 16 h at −25°C before raising the temperature to 4°C over 3 h . Samples were then washed in anhydrous acetone for 3 times , each for 15 min and embedded in Epon-Araldite during a 48-h period at 60°C . Blocks were subsequently sectioned and analyzed . Serial sections ( 80 nm ) were counterstained and imaged on CM 120 transmission electron microscope at 120 kV with a Gatan digital camera . A synapse was defined in serial sections that contained the same presynaptic specialization in at least three consecutive sections . Fifteen total ventral cord synapses from two WT animals and seven total ventral cord synapses from two pag-3 ( gv560 ) animals and eight total ventral cord synapses from two pag-3 ( gv560 ) ;ida-1 ( ok409 ) or two ida-1 ( ok409 ) animals were analyzed .
|
Within secretory cells , hormones are packaged into vesicles ( called DCVs ) that are released upon stimulation . The number of DCVs is regulated to meet the secretory demands of the cell by a mechanism that is poorly understood , although a protein in the membrane of DCVs , called IA-2 , is thought to play a role . A genetic screen in the nematode C . elegans is used , here , to find mutations that mis-regulate the corresponding worm protein called IDA-1 . Capitalizing on the simple neuroanatomy of the nematode and its transparency , we visualize IDA-1 protein levels directly in the animal using a fluorescent tag . We find that mutations in the transcription factor PAG-3/Gfi-1 result in elevated levels of IDA-1 protein , increased numbers of presynaptic DCVs , and behaviors consistent with increased neurosecretion . Our results demonstrate that IDA-1/IA-2 protein levels correlate with the biogenesis , utilization , or stability of DCVs . We propose that PAG-3 normally down regulates the production of IDA-1 , thus serving as part of the mechanism underlying DCV regulation . This is the first reported mutation that increases DCV numbers and secretion , offering insight into DCV homeostasis and a potential therapeutic target for diseases that would benefit from a boost in neuroendocrine secretion .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"genetics",
"and",
"genomics",
"developmental",
"biology",
"diabetes",
"and",
"endocrinology",
"cell",
"biology/neuronal",
"signaling",
"mechanisms"
] |
2009
|
Loss of the Transcriptional Repressor PAG-3/Gfi-1 Results in Enhanced Neurosecretion that is Dependent on the Dense-Core Vesicle Membrane Protein IDA-1/IA-2
|
Japanese encephalitis ( JE ) is a global public health issue that has spread widely to more than 20 countries in Asia and has extended its geographic range to the south Pacific region including Australia . JE has become the most important cause of viral encephalitis in the world . Japanese encephalitis viruses ( JEV ) are divided into five genotypes , based on the nucleotide sequence of the envelope ( E ) gene . The Muar strain , isolated from patient in Malaya in 1952 , is the sole example of genotype V JEV . Here , the XZ0934 strain of JEV was isolated from Culex tritaeniorhynchus , collected in China . The complete nucleotide and amino acid sequence of XZ0934 strain have been determined . The nucleotide divergence ranged from 20 . 3% to 21 . 4% and amino acid divergence ranged from 8 . 4% to 10 . 0% when compared with the 62 known JEV isolates that belong to genotype I–IV . It reveals low similarity between XZ0934 and genotype I–IV JEVs . Phylogenetic analysis using both complete genome and structural gene nucleotide sequences demonstrates that XZ0934 belongs to genotype V . This , in turn , suggests that genotype V JEV is emerging in JEV endemic areas . Thus , increased surveillance and diagnosis of viral encephalitis caused by genotype V JEV is an issue of great concern to nations in which JEV is endemic .
Japanese encephalitis ( JE ) , which is caused by JE virus ( JEV ) , is one of the most important viral encephalitis in the world [1]–[4] . It is prevalent mostly in Asia including eastern Asia [5]–[7] , southern Asia [8] and southeast Asia [9] , [10] . JE has extended its geographic range to the south Pacific region , including Australia [11] , [12] . An estimated 3 billion persons live in countries where JE is endemic [1]–[3] . JEV has a zoonotic transmission cycle between mosquitoes ( principally of the genus Culex ) and vertebrate hosts such as bats , water birds and pigs [1] , [4] . Human beings contract JEV when bitten by infected mosquitoes . Around 35 , 000–50 , 000 JE cases are reported each year , of which 10 , 000–15 , 000 are fatal [1]–[3] , [13] . Approximately 50% of JE patients present severe neurological and mental sequelae such as motor deficits , and cognitive and language impairment [14]–[16] . JEV is a member of the genus Flavivirus , family Flaviviridae [1] . Like other flaviviruses , the JEV genome is a single-stranded positive-sense RNA of approximately 11 kb in length . It is capped at its 5′ end and has a single open reading frame ( ORF ) that encodes a polyprotein . The ORF is flanked by 5′ and 3′ untranslated regions ( UTRs ) . The viral structural proteins are encoded by the 5′ one-third of the ORF and consist of the capsid ( C ) , membrane ( M; formed by proteolytic cleavage of its precursor protein PrM ) and envelope ( E ) proteins . The remaining 3′ region encodes non-structural proteins ( NS1 to NS5 ) [1] , [17] . JEVs have been divided into five genotypes ( genotype I , II , III , IV , V ) , based on nucleotide sequence of E gene [18] . Genotypes I-IV have been isolated from many vectors [19]–[21] , bats [22] , and patients [7] , [8] , [21] , [23] , [24] in Asia ( including eastern , southern and southeast Asia ) and Australia . To date , the Muar strain , which was isolated from specimens of brain tissues of patients with viral encephalitis in Malaya in 1952 , is the only example of genotype V JEV [18] , [25] . Since that time , no genotype V JEV has been detected . In this study , genotype V JEV was isolated from Culex tritaeniorhynchus collected in China in 2009 . This suggests that genotype V JEV is re-emerging in Asian country after a 57 year hiatus .
C6/36 ( Aedes albopictus ) cell line was grown in minimal essential medium ( HyClone ) with Hanks' salt solution supplemented with 10% fetal bovine serum ( FBS , HyClone ) , 2 mM glutamine , 0 . 12% NaHCO3 , and 100 U ml−1 penicillin and streptomycin . Cells were propagated and maintained at 28°C [21] . BHK-21 cells were grown in minimal essential medium ( HyClone ) with Earl's balanced salt solution supplemented with 10% FBS , 2 mM glutamine , 0 . 12% NaHCO3 , and 100 U ml−1 penicillin and streptomycin . BHK-21 cells were propagated and maintained at 37°C under a 5% CO2 atmosphere [21] . An arbovirus survey was conducted in Tibet in the summer of 2009 . Mosquitoes were collected in Mainling County ( altitude 2900 m ) and Medog County ( altitude 1000 m ) in the Nyingchi area of Tibet . Mosquito samples were collected using mosquito-trapping lamps ( Wuhan Lucky Star Environmental Protection Tech Co . , Ltd . , Hubei , China ) in the evening . Collection locations were proximal to sites of frequent human activity . Collection nets containing mosquitoes were frozen for 30 min at −20°C and transferred onto an ice plate for determination of mosquito species ( blood-fed and male mosquitoes were discarded ) . Female mosquitoes were identified to species level by morphologic characteristics and sorted into pools of 100 specimens according to species . The pools were put into collection tubes individually and stored in liquid nitrogen [26] , [27] . Mosquito pools were added to 1 . 5 ml minimal essential medium ( HyClone ) , supplemented with 2 mM glutamine , 0 . 12% NaHCO3 , and 100 U ml−1 penicillin and streptomycin , followed by grinding in a pre-cooled sterile plastic grinding tube using a TissueLyser ( GIAGEN , Germany ) . Homogenized samples were centrifuged at 17 , 000× g in a microcentrifuge for 20 min at 4°C , and the clarified supernatants were used to inoculate monolayers of BHK-21 and C6/36 cells and incubated at 37°C and 28°C , respectively . The cells were observed daily to check for development of cytopathic effects ( CPE ) . A sample was regarded as virus-positive if it caused CPE in successive cell passages [26] , [27] . Viral supernatants were applied to six-well plates ( Corning , USA ) of confluent BHK-21 cells and incubated for one hour . Plates were first overlaid with medium containing 75% agarose and then with medium containing neutral red vital stain after three days incubation at 37°C in a 5% CO2 incubator . Plaques of different sizes and shape were shattered in 500 ul MEM medium after being picked out using a sterile pipette tip . As described previously [28] , this process was repeated until a single plaque-shaped virus was obtained . Viral RNA was extracted from 140 ul supernatant from virus-infected BHK-21 cell cultures using a Viral RNA Mini Kit ( QIAGEN , Germany ) according to the manufacturer's instructions . cDNA was synthesized using a Ready-to-Go You-Prime First-Strand Beads Kit ( GE healthcare , UK ) and random hexanucleotide primers . PCR amplification using universal primers specific for flaviviruses , alphaviruses and bunyaviruses was conducted for identification of virus isolates [29] . Primers ( Table 1 ) were designed for full-length genome amplification and sequencing of JEV using the PREMIER Primer 5 software package . Thermal cycling parameters were as follows: one cycle of denaturation ( 94°C , 5 min ) followed by 35 cycles of 94°C denaturation for 30 s , 55°C annealing for 30 s , and 72°C extension for 1 min . The programme ended with an extension step at 72°C for 10 min . Amplified products were examined by agarose gel electrophoresis ( 1% ) , purified using a QIAquick Gel Extraction kit ( QIAGEN , Germany ) , and then sequenced directly . Sequencing of the 5′ UTR and 3′ UTR were determined by using 5′ RACE and 3′ RACE system for Rapid Amplification of cDNA Ends ( Invitrogen ) respectively . 5′ RACE was performed according to standard protocols ( Invitrogen 5′ RACE kit ) . 3′ RACE was performed by first adding a polyA tail using PolyA polymerase ( New England Biolabs ) and then conduct RT-PCR with gene specific primers and an oligo-dT-adapter primer . The full-length genome of the XZ0934 strain was compiled using SeqMan in the Lasergene software package ( DNASTAR ) . Nucleotide and amino acid sequence alignments were generated by ClustalX version 2 . 0 . 9 [22] , [30] . Analysis of nucleotide and deduced amino acid sequence identities was performed using GeneDoc and MegAlign in the Lasergene software package ( DNASTAR ) . Full-length nucleotide sequences of 32 selected JEV strains of varying genotype isolated from different locations and sources , and across a number of years , were downloaded from GenBank ( Table 2 ) . The reported structural gene sequence of genotype V JEV ( Muar strain ) [25] was used to build phylogenetic trees . Neighbor-joining phylogenetic trees based on nucleotide sequences were constructed using MEGA version 4 . 0 . 2 [18] , [31] . The robustness of phylogenetic constructions was evaluated by bootstrapping using 1000 replicates . To better understand the phylogenetic relationship between genotype V JEV and other flaviviruses , full-length nucleotide sequences of previously published JEV strains and other flaviviruses were downloaded from GenBank , including sequences from Murray Valley encephalitis virus ( MVEV ) , West Nile virus ( WNV ) , Kunjin virus ( KUNV ) , St . Louis encephalitis virus ( SLEV ) , Dengue virus 1 ( DENV1 , Dengue virus 2 ( DENV2 ) , Dengue virus 3 ( DENV3 ) , Dengue virus 4 ( DENV4 ) , Yellow fever virus ( YFV ) , Powassan virus ( POWV ) , Langat virus ( LANV ) , Louping ill virus ( LIV ) , Tick-borne encephalitis virus ( TBEV ) and Culex flavivirus ( Table 3 ) .
After homogenized supernatants were inoculated onto monolayers of BHK-21 and C6/36 cells , a single pool containing 100 specimens of Culex tritaeniorhynchus yielded a virus isolate designated XZ0934 . The supernatant of pool XZ0934 caused cytopathic effects ( CPE ) in BHK-21 and C6/36 cells in successive cell passages . The C6/36 cells became aggregate , and showed fusion and shedding while the BHK-21 cells became aggregate and began shedding by 72 h post-infection . All plaques in BHK-21 cell monolayers were of identical size ( mean 1 . 5 mm , n = 10 ) . Two plaques were picked from them and subjected to a second round of plaque purification . The resultant data were consistent with the former . Viral RNA was extracted and amplified by PCR using primers specific for flaviviruses , alphaviruses and bunyaviruses . XZ0934 was positive when primers specific for flaviviruses ( FU1/cFD2 ) [29] were used , and nucleotide sequencing confirmed that XZ0934 was a JEV . To ensure the consistency of different viral plaques , six purified plaques were picked and amplified using flavivirus-specific primers ( FU1/cFD2 ) . The nucleotide and amino acid sequence identities of the six purified plaques were 100% . This indicates that each of the six purified plaques was generated by an identical JEV strain . Therefore , one plaque was selected for full-length genome amplification and sequencing . Recent reports have suggested that JEVs currently circulating in China belong to genotypes I and III [5]–[7] , [21]–[23] . Thus , 32 primers were designed using the complete sequences of genotype I JEV Ishikawa ( GenBank accession number AB051292 ) and 48 from the sequence of genotype III JEV Beijing-1 ( GenBank accession number L48961 ) . These were used for amplification of the entire XZ0934 genome . PCRs were positive with 4 genotype I and 10 genotype III primers . Based on obtained nucleotide sequences , primers were designed to close the majority of gaps between assembled contigs by PCR amplification in order to determine the whole genome of XZ0934 . A further 24 primers ( Table 1 ) were designed and used to verify the accuracy of sequencing . The complete genome ( 10 , 983 nt ) of XZ0934 was sequenced ( GenBank accession no . JF915894 ) and found to possess one open reading frame ( ORF ) . When the complete genome sequence of isolate XZ0934 was compared with those of 62 known JEV isolates ( genotypes I–IV ) in Genbank , sequence identities varied from 78 . 6% ( KV1899 , K94P05 ) to 79 . 7% ( CC27-L1 ) and amino acid sequence identity from 90 . 0% ( KV1899 , K94P05 ) to 91 . 6% ( K87P39 ) . Thus , these data reveal low similarity between XZ0934 and genotype I–IV JEVs . Because the structural gene sequence of genotype V ( Muar ) has been reported [25] , an identity analysis of JEV structural genes ( C , PrM , M , E ) of XZ0934 , Muar and other selected genotype I–IV JEV strains was conducted ( Table 4 ) . C gene sequence homology varied from 78 . 2% ( G IV , JKT 6468 ) to 88 . 5% ( G V , Muar ) for nucleotides and 72 . 4% ( G IV , JKT 6468 ) to 85 . 8% ( G V , Muar ) for amino acids . That of the PrM gene varied from 71 . 7% ( G IV , JKT 6468 ) to 84 . 1% ( G V , Muar ) for nucleotides and 81 . 5% ( G IV , JKT 6468 ) to 90 . 2% ( G V , Muar ) for amino acids . M gene sequence homology varied from 80 . 0% ( G IV , JKT 6468 ) to 95 . 6% ( G V , Muar ) for nucleotides and 85 . 3% ( G IV , JKT 6468 ) to 100 . 0% ( G V , Muar ) for amino acids . E gene sequence homology varied from 77 . 0% ( G I , Ishikawa ) to 86 . 0% ( G V , Muar ) for nucleotides and 89 . 4% ( G I , Ishikawa ) to 93 . 2% ( G V , Muar ) for amino acids . These data demonstrate that the structural gene sequence homology of XZ0934 was higher with genotype V JEV ( Muar ) than with other genotype I–IV JEV strains . To establish the phylogenetic relationship between XZ0934 and other JEV strains , a phylogenetic tree was constructed using the complete genome sequences of XZ0934 and 32 selected JEV strains ( genotypes I–IV ) . Murray Valley encephalitis virus ( MVEV ) was used as an outgroup . Five distinct phylogenetic groups were identified . The XZ0934 strain , which was isolated from China , formed a branch divergent from other genotype I–IV JEV strains ( Figure 1A ) . Therefore , XZ0934 should be regarded as a novel , non-genotype I–IV , JEV isolate . To study their phylogenetic relationship , a phylogenetic tree was constructed using the reported structural gene nucleotide sequences of Muar [25] , XZ0934 , and other JEV strains ( genotype I–IV ) . No matter which structural gene was used to construct the phylogenetic tree , the topology was similar . Five distinct phylogenetic groups were evident in each tree . XZ0934 and Muar fell into the same group when the tree was constructed using the C ( Figure 1B ) , PrM ( Figure 1C ) , M ( Figure 1D ) or E ( Figure 1E ) genes . This result suggested that XZ0934 was a novel genotype V JEV isolate . A phylogenetic tree was also constructed using genomic nucleotide sequences in order to understand the phylogenetic relationship between XZ0934 and other flaviviruses . Data indicated that XZ0934 was indeed a JEV rather than any of the other 14 flaviviruses ( Figure 2 ) .
In recent years , the sequence of the JEV viral envelope ( E ) gene has been used by various authors to perform phylogenetic analyses [18] , [21] , [32] , [33] . Based on the resultant data , JEV strains have been divided into five genotypes ( genotypes I-V ) [18] . Genotypes I and III are distributed widely in Asia , including Japan , Korea , China , India , Vietnam and Philippines . Genotype II includes isolates from southern Thailand , Malaysia , Indonesia , and northern Australia . Genotype IV has been isolated only in Indonesia [18] . The Muar strain , isolated in Malaya in 1952 , is regarded as the only genotype V JEV isolate [18] , [25] , [33] . In this study , phylogenetic analysis of structural genes and whole genome sequences also suggested the existence of five JEV genotypes . Thus , the E gene is confirmed to be a useful phylogenetic marker for JEV . Primers designed for JEV genotypes I and III were used for full-length amplification of XZ0934 . Of these , only a few ( 4/32 genotype I and 10/48 genotype III ) resulted in successful amplification . This suggests a low whole genome sequence homology between XZ0934 and genotype I and I JEV isolates . In order to further understand the differences between XZ0934 and other JEV strains ( genotype I–IV ) , an identity analysis was conducted using the full-length nucleotide sequences of XZ0934 and 62 known JEV isolates ( genotypes I–IV ) in Genbank . Data suggested that XZ0934 and the genotype I–IV JEV strains were dissimilar . The nucleotide sequence identity varied from 78 . 6% to 79 . 7% and amino acid sequence identity from 90 . 0% to 91 . 6% . Indeed , the sequence divergence ranged from 20 . 3% to 21 . 4% ( nt ) and 8 . 4%–10 . 0% ( aa ) . It has been suggested that the nucleotide sequence divergence between different JEV genotypes is ∼10% [13] . The sequence divergence ( 20 . 3%–21 . 4% ) between XZ0934 and the genotype I–IV JEVs was greater than 10% , suggesting that XZ0934 is not a member of JEV genotypes I–IV . To confirm that XZ0934 was a JEV and not some other flavivirus , 14 flavivirus strains , including mosquito-borne and tick-borne flaviviruses , were used to build a phylogenetic tree . The data indicated that XZ0934 was indeed a JEV ( Fig . 2 ) . Four viral encephalitis cases were reported in Malaya ( n = 1 ) and Singapore ( n = 3 ) in the summer of 1952 . All patients exhibited high fever , vomiting , headache , disturbance of consciousness , stiff neck and deep coma with rapid progression to death by respiratory failure . Four virus strains were isolated from brain tissue specimens and identified as JEV by neutralization test using the Japanese Nakayama JEV strain [24] . Of these , the Muar strain , isolated from a 19-year-old male patient in Malaya in 1952 , has been assigned to genotype V based on the E gene sequence [18] , [25] , [32] , [33] . During the following 57 years ( 1952–2009 ) , no genotype V JEV has been reported . In this study , XZ0934 , isolated from Culex tritaeniorhynchus collected in China , has been identified as a genotype V JEV , based on phylogenetic analysis using both full-length genome and structural gene nucleotide sequences . This represents only the second instance of isolation of genotype V JEV worldwide since 1952 . Thus , genotype V JEV is not limited to southeast Asia and has begun to be emerge in the world . Many factors may contribute to spread of JEV [1] , [2] , such as changed agricultural practices ( which provide new breeding sites for mosquitoes ) , animal husbandry ( which provides host animals for transmission ) [34] , migrating birds and even wind-blown mosquitoes [11] , [35] . Each of the five known JEV genotypes originated in the Indonesia-Malaysia region [18] , so why has genotype V JEV not been detected for 57 years ? How did it spread to China from southeast Asia , a distance of thousands of kilometers ? Does this virus exist somewhere along the path from Malaysia to China ? All these issues are worthy of further study . Moreover , genotype V JEV was first isolated from human specimens , suggesting a high pathogenicity and the possibility of viral encephalitis . Therefore , increased surveillance and more effective diagnosis of viral encephalitis caused by genotype V JEV is an issue of great concern to nations in which JEV is endemic .
|
Japanese encephalitis virus ( JEV ) is a mosquito-borne virus that causes Japanese encephalitis ( JE ) with significant morbidity and mortality . Five genotypes ( genotype I–V ) have been identified based on the nucleotide sequence of viral envelope ( E ) gene of JEV . To date , the only known strain of genotype V is Muar strain , isolated from patient in Malaya in 1952 . Since then , no genotype V JEV has been detected in the world . In this study , the JEV strain , XZ0934 , was isolated from mosquito samples collected in China in 2009 . The full-length genome sequences of the XZ0934 strain was determined and founded to be the second strain of genotype V JEV based on the phylogenetic analysis using the complete genome and structural gene sequences . This suggests that genotype V JEV is re-emerging after 57 years ( 1952–2009 ) . Therefore , increased surveillance and more effective diagnosis for cases of JE caused by genotype V JEV are needed .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] |
[
"virology",
"emerging",
"viral",
"diseases",
"biology",
"microbiology"
] |
2011
|
Genotype V Japanese Encephalitis Virus Is Emerging
|
The interplay between excitatory and inhibitory neurons imparts rich functions of the brain . To understand the synaptic mechanisms underlying neuronal computations , a fundamental approach is to study the dynamics of excitatory and inhibitory synaptic inputs of each neuron . The traditional method of determining input conductance , which has been applied for decades , employs the synaptic current-voltage ( I-V ) relation obtained via voltage clamp . Due to the space clamp effect , the measured conductance is different from the local conductance on the dendrites . Therefore , the interpretation of the measured conductance remains to be clarified . Using theoretical analysis , electrophysiological experiments , and realistic neuron simulations , here we demonstrate that there does not exist a transform between the local conductance and the conductance measured by the traditional method , due to the neglect of a nonlinear interaction between the clamp current and the synaptic current in the traditional method . Consequently , the conductance determined by the traditional method may not correlate with the local conductance on the dendrites , and its value could be unphysically negative as observed in experiment . To circumvent the challenge of the space clamp effect and elucidate synaptic impact on neuronal information processing , we propose the concept of effective conductance which is proportional to the local conductance on the dendrite and reflects directly the functional influence of synaptic inputs on somatic membrane potential dynamics , and we further develop a framework to determine the effective conductance accurately . Our work suggests re-examination of previous studies involving conductance measurement and provides a reliable approach to assess synaptic influence on neuronal computation .
Neurons receive myriad excitatory ( E ) and inhibitory ( I ) synaptic inputs at dendrites . The spatiotemporal interaction between these E and I inputs are crucial for neuronal computation [1–3] , for instance , to shape neural activity [4 , 5] , to enhance feature selectivity [6 , 7] , to modulate neural oscillations [8] , and to balance network dynamics [9 , 10] . To understand synaptic mechanisms underlying neuronal computation , it is important to investigate the dynamics of the pure E and I inputs to a neuron via electrophysiological recording techniques . Somatic voltage clamp has become a popular approach to achieve this both in vitro and in vivo studies over the last thirty years [11] . For instance , voltage clamp has been extensively applied to areas including visual [12–14] , auditory [4 , 15–17] , and prefrontal cortex [18 , 19] . To reveal the quantitative information of E and I inputs , data collected in voltage clamp mode needs to be further processed to determine the input conductance values . In the traditional method , the dynamics of the neuronal voltage is described as [20] c d V d t = - g L ( V - ε L ) - g E ( V - ε E ) - g I ( V - ε I ) + I i n j , ( 1 ) where c is the membrane capacitance , V is the membrane potential , gL , gE and gI are the leak , E , and I conductances , respectively , εL , εE and εI are the corresponding reversal potentials , respectively , and Iinj is the externally injected current . Here all potentials are relative to the resting potential . Using the voltage clamp to hold the somatic voltage V at different levels , i . e . , c d V d t = 0 , one can obtain the corresponding synaptic currents I s y n i n j = g E ( ε E - V ) + g I ( ε I - V ) ( the superscript “inj” emphasizes that the synaptic current is measured in the presence of injected current given by the voltage clamp ) and linearly fit an I-V relation at each time point . By casting I s y n i n j as I s y n i n j = - k V + b , the slope k = g E + g I ( 2 ) is the total conductance ( the linear summation of the E and I conductances ) and the intercept b = g E ε E + g I ε I ( 3 ) is the reversal current ( the weighted summation of the E and I conductances ) . Therefore , by measuring the slope and the intercept of the I-V relation , one can solve Eqs 2 and 3 to obtain the values of gE and gI . Despite the extensive application of voltage clamp to determine E and I conductances for decades , it has yet to address various important issues related to the validity of the above approach . First , in Eq 1 , is the assumption of the linear summation of the synaptic current from the dendrites and the injected clamp current from the soma valid in a neuron with spatial dendrites ? Second , in the presence of the space clamp effect [21–27] , the membrane potential at distal synapses can deviate greatly from the holding potential and the E and I conductances obtained from Eqs 2 and 3 in the traditional method can be distorted significantly from the synaptic conductances on the dendrites . How to interpret the value of the conductance measured using the traditional method ? Whether there is a direct relation between the measured conductance and the local conductance that allows one to assess synaptic influence on neuronal computation ? Third , if the measured conductance does not correlate with the local conductance , then how to characterize synaptic impact on neuronal computation under the constraint of the space clamp effect ? Using theoretical analysis , electrophysiological experiments , and realistic neuron simulations , here we demonstrate that there does not exist a transform between the local conductance and the conductance measured by the traditional method because of the neglect of a nonlinear interaction between clamp current at the soma and synaptic currents from the dendrites in the traditional method . Consequently , the conductance determined by the traditional method may not correlate with the local conductance , and it could give unphysically negative value as observed in experiments . Under the constraint of the space clamp effect , we propose the concept of effective conductance , which reflects directly the functional impact of synaptic inputs on action potential initiation and thereby neuronal information processing . We then devise a framework for determining the effective conductance and accordingly verify it in both electrophysiological experiments and realistic neuron simulations , thereby establishing a biologically plausible metric for elucidating synaptic impact on neuronal computation . We discuss the scientific advance of our study in contrast to existing studies addressing the space-clamp effect and issues relevant to the application of our method in the section of Discussion .
The preparation of acute hippocampal slices ( 350 μm thick ) from Sprague Dawley rats of postnatal days 15-20 followed a method described in our previous study [28] . The animal experimental protocol was approved by the Animal Use and Care Committee of State Key Laboratory of Cognitive Neuroscience & Learning at Beijing Normal University ( IACUC-BNU-NKLCNL-2016-02 ) . In brief , rats were deeply anesthetized by i . p . injection of pentobarbital ( 30 mg/kg weight ) , and the brain was quickly dissected and then incubated in the ice-cold artificial cerebrospinal fluid ( aCSF ) , which was oxygenated with 95% O2 / 5% CO2 . Coronal hippocampal slices were sectioned with vibratome ( VT1200 , Leica ) and incubated in oxygenated aCSF at 34 °C for 30 min , followed by an incubation at 20-22 °C till the use for the electrophysiological recording . The aCSF contained ( in mM ) 125 NaCl , 3 KCl , 2 CaCl2 , 2 MgSO4 , 1 . 25 NaH2PO4 , 1 . 3 sodium ascorbate , 0 . 6 sodium pyruvate , 26 NaHCO3 , and 11 D-glucose ( pH 7 . 4 bubbled with 95% O2 / 5% CO2 ) . Whole-cell recording was made on the hippocampal CA1 pyramidal cell ( PC ) in slices in a chamber perfused with the same aCSF solution as that used for the brain slicing ( 2 ml/min; 30-32°C ) , under an Olympus upright microscope ( BX51WI ) that was equipped with the differential interference contrast ( DIC ) and fluorescence optics as well as an infrared camera ( IR-1000E , DAGE-MTI ) . The borosilicate-glass micropipettes were pulled by a Sutter puller ( P-1000 ) and filled by an internal solution containing ( in mM ) 145 K-gluconate , 5 KCl , 10 HEPES , 10 disodium phosphocreatine , 4 Mg-ATP , 0 . 3 Na-GTP and 0 . 2 EGTA ( pH 7 . 3 , 295 mOsm ) . Simultaneous recordings from the cell body and dendrite of a PC followed a procedure reported previously [29] , in which whole cell recording on the soma was first made using a micropipette ( 3-5 MΩ; with 20 μM Alexa Fluor 488 , InvitroGene ) , followed by another recording on Alexa Flour 488 ( green ) -labeled apical dendritic arbor at position ∼100 μm away from the soma with a micropipette ( 10-15 MΩ , filled with the internal solution without Alexa Fluor 488 ) . The serial resistance was compensated by >90% using the built-in function of the amplifier MultiClamp 700B ( Molecular Devices ) . Holding potentials of recorded cells were corrected for a calculated liquid junction potential [30] of ∼15 mV . In the dynamic clamp recording experiments , either AMPA type glutamate receptor-mediated excitatory conductance or GABAA receptor-mediated inhibitory conductance was intracellularly injected to the recorded PCs through the whole-cell recording pipette , using the built-in dynamic-clamp function of a 1401 Power3 digitizer ( CED ) and the Spike2 software ( v5 . 08; CED ) . Kinetics of AMPA or GABAA receptor conductance were in the form of two exponential functions with different rise/decay time constants: 5/7 . 8 ms for AMPA conductance; 6/18 ms for GABAA conductance . Their respective reversal potentials , EAMPA and E G A B A A , were set as 0 mV and −70 mV . Membrane voltage or current signals were amplified with a MultiClamp 700B amplifier ( Molecular Devices ) , filtered at10 KHz ( low-pass ) , digitalized by an analog-digital converter ( 1401 Power3 , CED ) at 50 KHz , and then acquired by the Spike2 software into a computer for further analysis . The experimental data of a sample neuron is available at https://github . com/songting858/Intercept-Method-code . The laboratory protocols used in this study has been uploaded to protocols . io , http://dx . doi . org/10 . 17504/protocols . io . wm4fc8w . We adapted the multi-compartment neuron model used in our previous studies [28 , 31 , 32] for our realistic pyramidal neuron simulation . The morphology of the reconstructed pyramidal neuron , which includes 200 compartments , was obtained from the Duke-Southampton Archive of Neuronal Morphology [33] . The passive cable properties and the densities of active conductances in the neuron model were based on published experimental data obtained from the hippocampal and cortical pyramidal neurons [34–46] , and the passive cable properties were slightly tuned to capture the distance-dependent space clamp effect measured in an experiment [26] ( S1 Fig ) . In particular , the multi-compartment neuron model included the voltage-gated sodium channel , the delayed rectifier potassium channel , two variants of A-type potassium channel , and the hyperpolarization activated channel . The synaptic inputs were given through AMPA receptor with rise/decay time constants: 5/7 . 8 ms and GABAA receptor with rise/decay time constants: 6/18 ms . The resting potential was set to Vr = −70 mV , and the E and I reversal potentials were set to EAMPA = 0 mV , E G A B A A = - 80mV . We used the NEURON software Version 7 . 4 [47] to simulate the model with time step of 0 . 1 ms . The detailed model description and the simulation code are available at https://github . com/songting858/Intercept-Method-code . We generalize the static two-port analysis [20 , 28] to study the property of effective conductance , to determine theoretically the effective conductance , and to illustrate the deficiency of the traditional method in the determination of conductance due to the neglect of a nonlinear interaction between clamp current at the soma and synaptic currents from the dendrites . Note that the purpose of this analysis is to provide insights into the issue of conductance measurement . Therefore , for the sake of simplicity , we focus on the case of time-independent synaptic inputs in the analysis . The case of time-dependent synaptic inputs will be demonstrated in both electrophysiological experiments of rat CA1 pyramidal neurons and simulations of the realistic pyramidal neuron model in the section of Results .
As introduced in the section of Materials and Methods , the effective conductance is defined by the ratio of the synaptic current I s y n e f f arriving at the soma to the driving force ( difference between the reversal potential ε and the somatic membrane potential V ) in the presence of either E or I input , i . e . , g e f f = I s y n e f f ε - V . ( 37 ) It should be stressed that , in order to distinguish from the synaptic current measured using voltage clamp in the traditional method , I s y n e f f is the synaptic current in the absence of any externally injected current . By performing a static transfer resistance analysis ( see Materials and methods for details ) , one can show that the effective conductance at the soma depends linearly on the local conductance on the dendrite , g E e f f = K E S K S S g E and g I e f f = K I S K S S g I ( 38 ) to the first order accuracy of gE and gI , where g E e f f and g I e f f are the effective E and I conductances respectively , and gE and gI are the corresponding local ones . Here the transfer resistance KAB is defined as the ratio of the voltage change in location B to the magnitude of the injected current in location A . The validity of the static transfer resistance analysis and the derived linear relationship between the effective and local conductances ( Eq 38 ) relies on the assumption that transfer resistance is a well-defined property of a neuron independent of input strength , which has been verified in the simulation of our realistic pyramidal neuron model ( Fig 1a ) . The effective conductance is an important concept not only because it is a proportional indicator of the local conductance on the dendrite ( Eq 38 ) , but also it reflects directly the functional impact of synaptic inputs on the spike trigger mechanism and thereby neuronal information processing at the soma . For example , a strong synaptic input at distal dendrite and a weak synaptic input at proximal dendrite can give rise to a similar magnitude of effective conductance when they arrive at the soma after dendritic filtering and integration , thus inducing a similar somatic response to initiate action potentials and propagate signals . Therefore , measuring the effective conductance is valuable for understanding the influence of synaptic activities on somatic membrane potential change and information coding . On account that the clamp can control the voltage sufficiently well only at the soma and its nearby regions , by applying the somatic voltage clamp , it is difficult to measure the local conductance accurately . However , it remains possible to measure the effective conductance which by definition only requires the local control of the voltage at the soma . As will be demonstrated below , the conductance determined by the traditional method ( using Eqs 2 and 3 ) is close to neither the local conductance on the dendrite nor the effective conductance at the soma . Here we perform the static transfer resistance analysis to illustrate the deficiency of the traditional method . When a neuron receives both E and I synaptic inputs on the dendrite with its somatic membrane potential clamped at various levels , our analysis yields a linear relation between synaptic current and voltage in voltage clamp mode ( see Materials and methods for details ) , i . e . , I s y n i n j = - k V + b , where k = ( K E S K S S ) 2 g E + ( K I S K S S ) 2 g I ( 39 ) and b = K E S K S S g E ε E + K I S K S S g I ε I ( 40 ) to the first order accuracy . Alternatively , if we cast Eqs 39 and 40 in terms of effective conductance using Eq 38 , we can obtain k = K E S K S S g E e f f + K I S K S S g I e f f ( 41 ) and b = g E e f f ε E + g I e f f ε I ( 42 ) to the first order accuracy . Therefore , to determine the local conductance , one needs to solve gE and gI from Eqs 39 and 40; while , to determine the effective conductance , one needs to solve g E e f f and g I e f f from Eqs 41 and 42 . Clearly , in contrast to Eq 2 in the traditional method , the slope k of the I-V relation in Eqs 39 or 41 is neither the total effective conductance nor the total local conductance . In this sense , the conductance determined by the traditional method using Eqs 2 and 3 is neither the local conductance nor the effective conductance . We can further show that there does not exist a transform between the local conductance and the conductance determined by the traditional method ( see Materials and methods ) , indicating that the measured conductance may not correlate with the local conductance thus its biological interpretation is unclear . The error of the traditional method is caused by the prefactors K E S K S S and K I S K S S in Eqs 39–41 , which arise from the nonlinear interaction between the injected current at the soma and the synaptic current from the dendrite ( see Materials and methods for details ) . Only when the E and I inputs are given at the soma will the prefactors vanish . In this particular limit , the local and effective conductances become identical ( Eq 38 ) . In addition , Eqs 39 , 40 , 41 and 42 further reduce to Eqs 2 and 3 , which enables one to use the traditional method to determine the local or effective conductance accurately . However , in general , these prefactors cannot be naively assumed to be unity ( i . e . , no nonlinear interaction ) since they can distort significantly the determination of conductance , as will be demonstrated below . As it is challenging in experiment to elicit inputs which are spatially broadly distributed on the distal dendrite , we resort to numerical simulations using the realistic pyramidal neuron model to investigate the spatial dependence of the prefactors K E S K S S and K I S K S S . Our numerical result shows that the prefactors across the entire dendritic tree decay from unity to zero rapidly with the increase of the distance between the synaptic input sites and the soma ( Fig 1b ) . Therefore , if one attempts to determine the effective conductance using the traditional method based on Eqs 2 and 3 rather than Eqs 41 and 42 , the errors can become prominent . For instance , when the E and I inputs are given at the distal dendrite where the prefactors are small , the value of conductance could vanish when determined by the traditional method . In addition , in our analysis , when the prefactors become sufficiently small , a negative conductance can arise via the traditional method ( Fig 1c and 1d ) . This possibly explains why a negative conductance was observed in early experiments [26] . Our theoretical prediction is particularly of note that the measurement of I conductance is distorted more significantly than E conductance by the traditional method—as a consequence of the ratio of measurement error between the E and I conductances being proportional to the ratio between the I reversal potential εI ( e . g . , −10 mV relative to the resting potential ) and the E reversal potential εE ( e . g . , 70 mV relative to the resting potential ) , i . e . , ΔgE: ΔgI = −εI: εE ( see Fig 1c and Materials and methods ) . From our theoretical analysis above , the deficiency of the traditional method in measuring the effective conductance results from the neglect of the nonlinear interaction between synaptic current and injected clamp current . To confirm our theoretical results , we perform electrophysiological experiment to demonstrate the existence of the interaction between synaptic current and injected clamp current . In the experiment ( see Materials and methods for details ) , we record 7 rat hippocampal CA1 pyramidal neurons using somatic voltage clamp . The resting potential of each neuron ranges from −57 mV to −68 mV . E and I synaptic inputs are given via a dynamic clamp at the location on the dendrite about 100 μm away from the soma . The absolute E and I reversal potentials are set as EAMPA = 0 mV and E G A B A A = - 70mV , and the relative reversal potentials εE and εI can be determined by subtracting the resting potential from EAMPA and E G A B A A respectively . The local input synaptic conductances through the dynamic clamp take the form of a difference of two exponential functions whose time constants were derived from voltage traces in experiment [28] , with rise time constant 5 ms ( 6 ms ) and decay time constant 7 . 8 ms ( 18ms ) for E ( I ) conductance . The peak amplitude of E and I synaptic conductances ranges from 2 nS to 5 nS and 3 nS to 6 nS , respectively . For each pyramidal neuron , we clamp the voltage at the soma with five levels from −50 mV to −90 mV with an increment of 10 mV ( Fig 2a ) . For an individual E input given on the dendrite via dynamic clamp ( Fig 2b ) , we can then record five excitatory postsynaptic current ( EPSC ) traces I s y n i n j at the soma corresponding to the five holding voltage levels , and determine the corresponding E conductance traces using g E i n j = I s y n i n j / ( ε E - V ) . Here εE and V are reversal and holding potentials relative to the resting potential , and I s y n i n j is the synaptic current which is the current increment from the baseline injected current that is used to clamp the neuron to a steady state of voltage ( see Materials and methods ) . Again , the superscript “inj” in the notations g E i n j and I s y n i n j emphasizes the fact that they are determined in the presence of the injected clamp current . We obtain the final profile of the resulting conductance as a function of a difference of two exponentials using least square fitting . Fig 2c shows that five E conductances g E i n j thus obtained are not identical with disparity between these conductances well beyond recording statistical fluctuations . The dependence of the conductance value of g E i n j on the clamp voltage as shown in Fig 2c contradicts the assumption of the traditional method that the synaptic conductance is independent of injected current as described in Eq 1 . It confirms our result that the synaptic current from the dendrite and the injected current on the soma cannot be linearly summed as a consequence of their interaction with each other . The difference between the I conductances g I i n j estimated from different voltage clamp levels is more prominent than the E case ( Fig 2d and 2e ) . The voltage dependence of the I conductance g I i n j can be highlighted by the following limiting case . When the soma is clamped at the I reversal potential E G A B A A = - 70mV , the value of the I conductance g I i n j would become unphysically infinity because the denominator in the expression g I i n j = I s y n i n j / ( ε I - V ) vanishes ( such a case is not displayed in Fig 2e ) . There is further evidence demonstrating the interaction between synaptic current and injected clamp current as observed in the experiment . Were the synaptic and injected currents linearly summable ( Eq 1 ) , then the conductances obtained from the slope and those from the intercept of the I-V relation would be identical through the traditional method . However , this turns out not to be the case in our experimental observation . To be specific , given an individual E or I input on the dendrite , we can obtain a linear I-V relationship between the voltage and the synaptic current I s y n i n j at each time point after the onset of the stimulus . An example of the I-V relationship at the time 10 ms after the stimulus onset is shown in Fig 2f . Upon casting g ( ε − V ) as I s y n i n j and following Eqs 2 and 3 for the case of only a purely E or I input , we obtain the ratio of the conductance value estimated from the slope to that from the intercept . This ratio deviates greatly from unity—the expected result obtained by the traditional method . It is nearly a constant and is independent of conductance amplitude ( Fig 2g ) . The value of ratio for E inputs is nearly identical to that for I inputs when the E and I inputs are given at the same dendritic location ( Fig 2g ) . This observation is consistent with our theoretical prediction from Eqs 41 and 42 for the case of a purely E or I input , for which the above ratio is the same as the ratio of the effective conductance g E e f f ( g I e f f ) at the soma to the local conductance gE ( gI ) on the dendrite ( Eq 38 ) . In principle , the deficiency of the traditional method could be eliminated by measuring the value of the prefactors associated with input locations based on Eqs 41 and 42 . However , for a neuron receiving a large number of spatially broadly distributed synaptic inputs , it remains difficult to have all a priori information about the transfer resistances between the synaptic input sites and the soma , thus hampering the recovery of the E and I conductances from Eqs 41 and 42 directly . From our theoretical analysis , we note that the intercept b in Eq 42 possesses the form of effective reversal current without the explicit information of transfer resistances ( see Materials and methods ) . Therefore , we propose to recover the effective E and I conductances only from the intercept value . In principle , the effective E and I conductances can be recovered from multiple I-V relations by varying the E or I reversal potential at various levels , or by pharmacologically blocking the E or I synaptic receptor . As a proof of concept , below we examplify the recovery of the effective conductances from the intercept information via the change of synaptic reversal potential . To be specific , we can vary the I reversal potential from εI to ε I ′ to obtain a second intercept equation b ′ = g E e f f ε E + g I e f f ε I ′ , ( 43 ) and then the effective E and I conductances can be obtained from Eqs 42 and 43 . In physiological experiment , to change reversal potential , one may need to effect a change of the intracellular fluid environment as further discussed in the section of Discussion . From now on , we refer to this method based on Eqs 42 and 43 as the intercept method ( IM ) , and the traditional method based on Eqs 2 and 3 as the slope-and-intercept method ( SIM ) . The key difference between SIM and IM lies in the transfer resistance . Based on our static transfer resistance analysis , the slope of the linear I-V relation follows Eq 41 , and the intercept of the linear I-V relation follows Eq 42 , in which the pre-factors KES/KSS and KIS/KSS are determined by the synaptic input locations , and their values are in general difficult to measure . Conceptually , IM does not ignore these pre-factors but SIM does . In IM , by accounting for the fact that these pre-factors are generally unknown , one can only use the intercept information to recover g E e f f and g I e f f . And because there are two unknown variables g E e f f and g I e f f to solve , IM suggests to provide at least two equations of I-V relations . For example , if the reversal potential is changed , one can obtain a second intercept equation ( Eq 43 ) . Subsequently , in IM , the effective E and I conductances g E e f f and g I e f f are determined by solving Eqs 42 and 43 . In contrast , in SIM , by naively assuming the location-dependent pre-factors in Eq 41 to be unity , the effective E and I conductances are determined by solving Eqs 41 and 42 ( with the assumption that KES/KSS = 1 and KIS/KSS = 1 in Eq 41 ) . However , based on the static transfer resistance analysis , the existence of the pre-factors results from the nonlinear interaction between the clamp current at the soma and the synaptic current from the dendrites , and the pre-factors in general deviate from unity as shown in Fig 1b . Therefore , the conductances measured by the traditional method is expected to deviate from the true effective conductances , as shown in both electrophysiological experiments and realistic neuron simulations below . Technically , when applying the intercept method to measure the effective E and I synaptic conductances of a neuron , the first step is to clamp the somatic voltage at various levels and measure the corresponding synaptic currents arriving at the soma . In this step , one shall exert a well control of the experimental condition such that the E and I synaptic inputs received by the neuron under different holding voltage are approximately the same . This step is identical to that in the traditional slope-and-intercept method . The second step is to fit a linear relation between the holding voltage and the synaptic current and read out the intercept value from the I-V relation at each time point , which contains information of the effective E and I conductances g E e f f and g I e f f described by Eq 42 . The third step is to vary the reversal potential to a different value and repeat the first two steps to obtain its corresponding intercept value at each time point , which also contains information of g E e f f and g I e f f described by Eq 43 . The final step is to recover g E e f f and g I e f f by solving Eqs 42 and 43 . We next perform experiment to demonstrate the validity of IM by contrasting its error with that of SIM . Because we need to vary the reversal potential to a different value in IM , we have first verified that the value of the effective conductance is nearly independent of the reversal potential value ( Fig 2h ) . Next , we proceed to determine the reference effective E and I conductances . Based on our analysis ( Eq 42 ) , the conductance obtained from the intercept of the I-V relation is the true effective conductance , i . e . , the conductance at the soma induced by a synaptic input on the dendrite in the absence of the injected current . Therefore , we choose the values of the E and I conductances estimated from the intercepts for the case of only pure E or I inputs as the reference conductances to evaluate IM and SIM . We note in passing that the effective reference conductances determined in this way are more accurate than those determined directly from Eq 1 in the absence of voltage clamp for we can avoid taking time derivative of noisy experimental voltage data—a procedure that would introduce large numerical errors . In our realistic neuron simulation to corroborate our experimental results below , however , we can use Eq 1 in the absence of voltage clamp to determine the reference conductance since the numerical simulation is sufficiently accurate for obtaining the time derivative of voltage . Given an individual E pulse input at a dendritic location about 100 μm away from the soma and placing the voltage clamp at the soma , we can record a set of synaptic current I s y n i n j under five holding voltages from −50 mV to −90 mV . We then determine the effective E conductance from the intercept of the I-V relation at each time point . A similar procedure is carried out separately for the effective I conductance . A pair of measured effective E and I conductances determined in this way is displayed in Fig 3c as the reference values ( solid curves ) , against which we evaluate the performance of IM and SIM . Next , with a voltage clamp placed at the soma , we simultaneously elicit the E and I pulse inputs same as for the reference ones , i . e . , input at the same dendritic location with the same strength ( Fig 3a ) . Five total synaptic currents I s y n i n j at the soma are obtained under five holding voltages from −50 mV to −90 mV . We then also observe a linear relation between the synaptic current and the membrane potential at each time point . An example of the I-V relation at the time 10 ms after the onset of the stimulus is shown in Fig 3b . Finally , we determine a pair of values of the E and I conductance pulses from the linear I-V relation by using SIM . Meanwhile , by changing the I reversal potential E G A B A A from −70 mV to −80 mV and repeating the above procedure ( Fig 3b ) , a pair of alternative values of E and I conductances can be obtained using IM . By comparing the values of the conductance pulses measured by the two methods with those of the reference conductance pulses in Fig 3c , we observe that the conductance estimated by IM nearly overlaps with the true conductance , whereas the conductance estimated by the traditional SIM deviates greatly from the true conductance . As shown in Fig 3d , across 7 pyramidal neurons , the effective conductance measured by IM has a relatively small error on average , with a relative error of peak amplitude ranging from 0 . 6% to 15 . 6% for E conductance and from 3 . 0% to 14 . 3% for I conductance . In contrast , the conductance measured by SIM yields a large relative error of peak amplitude as great as from 13 . 1% to 40 . 0% for E conductance and from 10 . 1% to 44 . 9% for I conductance . According to our theoretical analysis , the error in SIM is caused by failing in taking into account the nonlinear interaction between the synaptic current from the dendrite and the injected current at the soma—thus missing the prefactors KES/KSS and KIS/KSS in Eq 41 , whose strength has a sensitive dependence on the dendritic location ( Fig 1b ) . As observed in a previous experiment [26] , for synaptic inputs received at the proximal dendrite 100 μm away from the soma , the recovered synaptic current at the soma by using the somatic voltage clamp is already below 60% , and the escape voltage level is above 50% , indicating that KES/KSS and KIS/KSS in Eq 41 are unneglectable in this case . In our experiment , we show that when the inputs are given at the proximal dendrite about 100 μm away from soma , the error of SIM has already reached ∼ 40% ( Fig 3d ) . For a synaptic input location further towards the distal dendrite , the error is expected to become substantially larger . As it is a rather challenging experimental task to elicit inputs on multiple dendritic locations in general , and on the distal dendrite in particular , we turn to realistic neuron simulations to demonstrate the validity of IM by contrasting its error with that of SIM for distal inputs or spatiotemporally broadly distributed multiple inputs on the dendrite . In our realistic pyramidal neuron model ( see Materials and methods for details ) , the resting potential is set to Vr = −70 mV , and the absolute E and I reversal potentials are initially set to EAMPA = 0 mV , E G A B A A = - 80mV . The relative reversal potentials are then determined as εE = 70 mV and εI = −10 mV . We first demonstrate the validity of our realistic neuron model as a good model of a biological neuron by examining whether the performance of SIM and IM for the model neuron is similar to their performance for the pyramidal neurons recorded in the experiment , when the E and I synaptic inputs are given at the dendritic trunk of the model neuron about 100 μm away from the soma—the same input condition as that in the experiment . In the model , we measure the distance by taking into account the zig-zag geometry of the dendrites . As above , we first determine the reference E and I conductances by giving the neuron an individual E and I input separately . For an individual E pulse input at a dendritic location about 100 μm away from the soma but without the injected clamp current at the soma , we can numerically record the corresponding EPSP at the soma and invoke Eq 37 to determine the value of the effective E conductance pulse from the point-neuron model ( for which we set Iinj = 0 , gI = 0 in Eq 1 ) . A similar procedure can be carried out for the effective I conductance pulse in response to an individual I pulse input at a dendritic location about 100 μm away from the soma ( again , in the absence of injected current at the soma ) . In the simulation , the experimental result is also confirmed that the value of the effective E or I conductance is nearly identical under different synaptic reversal potentials ( Fig 4a ) . When the E and I inputs are given simultaneously , the application of voltage clamp gives rise to an I-V relation at each time point as in experiment . By altering the I reversal potential E G A B A A from −80 mV to −90 mV , an additional I-V relation at the same time point can be obtained . As shown in S3 Fig , the effective E and I conductances measured by IM has a relative error of peak amplitude about 13 . 5% for E conductance and 9 . 6% for I conductance . In contrast , the conductance measured by SIM yields a large relative error of peak amplitude as great as 18 . 8% for E conductance and 37 . 4% for I conductance . This result falls within the range of the error measured in the experiment as shown in Fig 3d , and is similar to the error measured in the example neuron as shown in Fig 3c . We next investigate the performance of IM when a pair of synaptic inputs are given on the distal dendrite of the model neuron . For simultaneous E and I inputs given at the dendritic trunk about 350 μm and 300 μm away from the soma respectively , the application of voltage clamp gives rise to an I-V relation at each time point as in experiment . An additional I-V relation at the same time point after the onset of the stimulus results from altering the I reversal potential E G A B A A from −80 mV to −90 mV ( Fig 4b ) . As shown in Fig 4c , the conductances measured using IM have a small relative error compared with the corresponding reference values , with a maximum error of 14 . 5% for E conductance and 11 . 1% for I conductance in the peak amplitude . In contrast , those determined using SIM yield an error as large as 36 . 8% for E conductance and 98 . 1% for I conductance . To model the situation in vivo , we distribute 15 E inputs and 5 I inputs across the entire dendritic tree of the pyramidal neuron ( Fig 4d ) . At each synaptic location , the arrival time of each input is randomly selected between 0 ms and 1000 ms with input rate of 100 Hz . We use the identical input for both the measurement of time evolution of the effective conductance as reference without the voltage clamp ( using Eq 1 ) as that of the conductances with the voltage clamp . Comparison of the values of conductance measured by the IM and SIM methods with the reference conductance in Fig 4e demonstrates that the effective E or I conductance estimated by our IM is in good agreement with the true effective conductance . Meanwhile , the conductance estimated by SIM deviates greatly from the true one in general , and particularly substantial for the inhibitory case . In the subthreshold regime , the conductances measured by IM incur a relatively small error with time averaged relative error of 16 . 3% for E conductance and 6 . 3% for I conductance , whereas those determined using SIM yield a time averaged relative error as large as 42 . 7% for E conductance and 102 . 6% for I conductance . In this simulation , while the true I conductance is substantially larger than the true E conductance , the I conductance estimated by SIM turns out to be substantially smaller than the E conductance . It is important to stress that the value of the I conductance could even become negative , thus demonstrating the severe deficiency of SIM . We note that in Fig 4c the tails of the reference conductances also become slightly negative , which arises from the repolarization of the membrane potential before relaxing to its resting state . This phenomenon has also been observed in experiments [28] and is attributed to the activity of voltage-gated ion channels , which have not been taken into account in the simple point neuron model ( Eq 1 ) . Different from the case in Fig 4c , the negative conductance determined by SIM in Fig 4e originates from the deficiency of the method itself instead of active channels . In fact , the negative value of conductance in Fig 4e is only observed for the inhibitory one , for which its reference conductance always stays at a positive level . We now address the question of how the error of the two methods depends on the input locations on the dendrite . For a pair of synaptic inputs at various locations on the dendrite , our simulation shows that IM can control the error about 10% even for the distal inputs ( Fig 4f and 4h ) , whereas the error of SIM increases rapidly to 100% as the input location moves away from the soma to the distal dendrite ( Fig 4g and 4i ) . In some remote distal dendritic sites—greater than 400 μm away from the soma , the estimated I conductance can also become negative ( Fig 4i ) , accentuating the SIM deficiency . To investigate the performance of SIM and IM when the inputs are given on a dendritic branch , we have simulated the following three cases , i . e . , when both the E and I inputs are located at the dendritic trunk 200 μm away from the soma , when the E input is located at a dendritic branch 200 μm away from the soma and the I input is located at the dendritic trunk also 200 μm away from the soma , and when both the E and I inputs are located at a dendritic branch 200 μm away from the soma . As shown in S4 Fig , the performance of IM is almost the same for the three cases , while the performance of SIM improves a little when both the E and I inputs are located at the dendritic branch . As discussed previously , the error of SIM results from the incorrect assumption KES/KSS = 1 and KIS/KSS = 1 in Eq 41 . Therefore , the slightly improved performance of SIM for inputs received on secondary dendrites can be explained by the fact that the prefactors KES/KSS and KIS/KSS are closer to unity when the inputs are on a branch compared to the case when the inputs are on the dendritic trunk , as shown in Fig 1b and Eqs 29 and 30 . A further validation of IM is shown in Fig 5 . For the same pair of transient E and I inputs in Fig 4c , on the one hand , we can measure the slope and the intercept of the I-V relation using voltage clamp when the E and I inputs are given simultaneously; on the other hand , we can determine the effective E and I reference conductances g E e f f and g I e f f when the E and I inputs are given separately and then reconstruct the total conductance as g E e f f + g I e f f and the effective reversal current as g E e f f ε E + g I e f f ε I . Fig 5 shows that the effective reversal current overlaps well with the intercept of the I-V relation , while the violation of SIM is instantiated by a rather substantial difference between the total effective conductance and the slope of the I-V relation .
The space clamp effect has been well known for decades [21–27] which limits the control of the voltage clamp on the membrane potential across the entire dendritic arbor , thus potentially impeding the quantitative understanding of synaptic physiology , especially the interaction between excitation and inhibition . The error of conductance measurement induced by the space clamp effect has been quantified in experiment [26] . In addition , early experiments show that the leak conductance of a neuron can be blocked pharmacologically to increase membrane resistance [38] , and the time course of synaptic currents can be slowed down in room temperature to increase neuronal input resistance [54] . Based on these experiments , to alleviate the space clamp issue , several approaches have been proposed including pharmacologically reducing the leak conductances of the neuron or slowing the rate of synaptic conductance changes by cooling the neuron [26 , 55] . However , it has been shown in experiment that these approaches are unable to resolve the space clamp problem in real neurons [26 , 55] . In particular , it has been reported in [26] that the intracellular diffusion of cesium from somatic and dendritic patch pipettes for the pharmacological reduction of resting and leak conductance improves little in the measurement of synaptic currents by the somatic voltage clamp . Alternative approaches including the voltage jump method [56 , 57] , dendritic recording [58] , and multi-compartment neuron modeling [59] have been applied to investigate synaptic physiology at dendrites . But these approaches are unable to separate the E and I inputs information received by a real neuron , hampering the understanding of the interaction between E and I inputs . Different from the previous works , our work is not a re-examination of the space clamp effect . We aim to develop a method to measure a biologically interpretable quantity associated with E and I synaptic inputs under the constraint of the space clamp effect . We have first addressed the relation between the measured conductance by the traditional method SIM and the local conductance in the presence of the space clamp effect . Although it has been demonstrated that the measured conductance substantially deviates from the local conductance , it remains unclear whether the conductance measured using the traditional method can encode the information of the local conductance . Our work has demonstrated that there does not exist a transform between the measured conductance and the local conductance , i . e . , the measured conductance may not correlate with the local conductance . Consequently , severe problems have been revealed by our realistic neuron simulations that , by using the traditional method SIM , the excitation could be misinterpreted as being substantially larger than the inhibition , which completely contradicts to the fact that inhibition is dominant to excitation ( Fig 4e ) . And the measured inhibitory conductance could be unphysically negative ( Fig 4e ) . In addition , we have pointed out that the major deficiency of the traditional method SIM arises from the slope equation ( Eq 2 ) but not the intercept equation ( Eq 3 ) , which has not been noticed by previous works and our work potentially provides some guidance for future voltage clamp data analysis . Furthermore , previous works have only investigated the error induced by the space clamp effect per se without providing a solution to extract E and I inputs information under the constraint of the space clamp effect . In contrast , here we have developed a method named IM to measure the effective conductance based on the intercept equation ( Eq 3 ) . The notion of the effective conductance obviates the issue of the space clamp effect since one only needs to control the clamped voltage well at the soma in our method . In addition to the space clamp effect , dendritic spines also present great challenge for the estimation of local conductance because of the high spine neck resistance [60] . However , this issue is circumvented as well if one concentrates on the influence of synaptic inputs on the soma after dendritic filtering . Similar to several other multi-trial methods of conductance measurement [11 , 13 , 61 , 62] , a limitation of our method is the requirement of repeatable network behavior from trial to trial . To overcome this limitation , several methods have been proposed to be performed on a single voltage trace [63–67] . However , these methods are limited to special cases so far . For example , some of the single-trace methods require assumptions of the form of conductance dynamics [63] or membrane potential dynamics [65 , 67] . In these cases , our method will be more efficient than them when the experiment is relatively well controlled and repeatable with a good precision . In practice , our method and the single-trace methods may complement each other depending on experimental conditions . It is worthwhile to comment that so far our method is developed to measure the effective conductance at the soma rather than the local conductance on dendrites . In order to study the integration of synaptic inputs on local dendrites and dendritic phenomena such as dendritic spikes [68 , 69] , the IM method shall be further improved to infer the local conductance from the effective conductance based on our derived proportional relation between the local and effective conductance . In addition , our analysis is accurate only to the first order approximation of the effective conductance , while higher order corrections may also contribute to the conductance value; And our analysis is based on a simple point model of the soma . Yet the dendritic integration of synaptic inputs can potentially lead to a more complicated form of a point-neuron model of the soma [70]; Further , the change of reversal potential in our method in principle requires a change of the intracellular fluid environment , which could be experimentally challenging especially in vivo . Therefore , our method is a proof of concept at the current stage , but the concept of IM already contrasts the severe deficiency of the traditional method . In addition , to provide more potential solutions , we have proposed a new alternative method to obtain a second intercept equation , i . e . , intracellular blockade of GABA receptors using drugs . For example , it has been shown in experiment that fluoride ions were effective for intracellular blockade of IPSCs [71] , which may make IM effective potentially . However , this approach so far has additional cellular effects as reducing a neuron’s selectivity and utility [71] . Therefore , despite that a direct validation of IM is difficult to achieve at present , we believe that our method will become useful with the further development of pharmacological tools . To move further steps , it is important to address these issues in future studies for a quantitative understanding of the synaptic dynamics of neurons with higher accuracy .
|
To understand synaptic mechanisms underlying neuronal computations , a fundamental approach is to use voltage clamp to measure the dynamics of excitatory and inhibitory input conductances . Due to the space clamp effect , the measured conductance in general deviates from the local input conductance on the dendrites , hence its biological interpretation is questionable , as we demonstrate in this work . We further propose the concept of effective conductance that is proportional to the local input conductance on the dendrites and reflects directly the synaptic impact on spike generation , and develop a framework to determine the effective conductance reliably . Our work provides a biologically plausible metric for elucidating synaptic influence on neuronal computation under the constraint of the space clamp effect .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion"
] |
[
"medicine",
"and",
"health",
"sciences",
"nervous",
"system",
"membrane",
"potential",
"electrophysiology",
"neuroscience",
"simulation",
"and",
"modeling",
"bioassays",
"and",
"physiological",
"analysis",
"neuronal",
"dendrites",
"research",
"and",
"analysis",
"methods",
"excitatory",
"postsynaptic",
"potentials",
"animal",
"cells",
"electrophysiological",
"techniques",
"cellular",
"neuroscience",
"cell",
"biology",
"anatomy",
"synapses",
"physiology",
"neurons",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"neurophysiology"
] |
2019
|
Determination of effective synaptic conductances using somatic voltage clamp
|
Blood flukes of the genus Schistosoma cause schistosomiasis—a neglected tropical disease ( NTD ) that affects more than 200 million people worldwide . Studies of schistosome genomes have improved our understanding of the molecular biology of flatworms , but most of them have focused largely on protein-coding genes . Small non-coding RNAs ( sncRNAs ) have been explored in selected schistosome species and are suggested to play essential roles in the post-transcriptional regulation of genes , and in modulating flatworm-host interactions . However , genome-wide small RNA data are currently lacking for key schistosomes including Schistosoma haematobium—the causative agent of urogenital schistosomiasis of humans . MicroRNAs ( miRNAs ) and other sncRNAs of male and female adults of S . haematobium and small RNA transcription levels were explored by deep sequencing , genome mapping and detailed bioinformatic analyses . In total , 89 transcribed miRNAs were identified in S . haematobium—a similar complement to those reported for the congeners S . mansoni and S . japonicum . Of these miRNAs , 34 were novel , with no homologs in other schistosomes . Most miRNAs ( n = 64 ) exhibited sex-biased transcription , suggestive of roles in sexual differentiation , pairing of adult worms and reproductive processes . Of the sncRNAs that were not miRNAs , some related to the spliceosome ( n = 21 ) , biogenesis of other RNAs ( n = 3 ) or ribozyme functions ( n = 16 ) , whereas most others ( n = 3798 ) were novel ( ‘orphans’ ) with unknown functions . This study provides the first genome-wide sncRNA resource for S . haematobium , extending earlier studies of schistosomes . The present work should facilitate the future curation and experimental validation of sncRNA functions in schistosomes to enhance our understanding of post-transcriptional gene regulation and of the roles that sncRNAs play in schistosome reproduction , development and parasite-host cross-talk .
Human schistosomiasis is a chronic , snail-borne , neglected tropical disease ( NTD ) caused predominantly by infections with the blood flukes Schistosoma haematobium , S . mansoni and S . japonicum [1] . This disease is highly prevalent in sub-Saharan Africa , where infections with S . haematobium and S . mansoni affect ~ 200 million people [2] . The decoding of schistosome draft genomes [3–5] has , to some extent , led to an improved understanding of the molecular biology of these parasites and associated disease , but investigations have mostly focused on the protein-coding gene complement . Only recently , non-protein-coding regions in flatworm genomes have come under the spotlight . For instance , small non-coding RNAs ( sncRNAs ) present in exosomes have been proposed to play a role in modulating parasite-host interplay [6 , 7] , and the identification of other sncRNA elements suggests similar functionalities in flatworms [6–10] . Clearly , a comprehensive characterisation of sncRNA elements in S . haematobium and other flatworm taxa , for which data are currently lacking , will likely underpin investigations to understand mechanisms governing parasite-host interactions and disease . Deep sequencing of the RNA populations of schistosomes has revealed a diversity of non-protein-coding RNA families , including degradation-like small RNA fragments originating from microRNAs ( miRNAs ) [10–13] , small interfering RNAs ( siRNAs ) and transfer RNAs ( tRNAs ) [13 , 14] . In these studies , miRNAs have been a focus , mostly due to the recognised structures of miRNA precursors and their abundance in the small RNA ( sRNA ) libraries sequenced to date . Comparative studies have identified that the complements of miRNAs in parasitic flatworms display substantial differences in size and sequence when compared with those of other eukaryotes studied to date [8] , but such studies require further support by the sequencing of transcribed sRNA from additional species . To date , most studies have focused on either in silico predictions of miRNA precursors from genomic data [11] or on the detection of mature miRNA elements within sequenced sRNAs and the subsequent identification of precursors within genomic data [10] . Parasitic flatworms lack the canonical Piwi pathway [15 , 16] , suggesting that , whilst many non-coding elements are relatively conserved among eukaryotic species , little is known about the precise composition of non-coding RNA families and their functions in platyhelminths . Due to an absence of sRNA datasets for most lophotrochozoans and the evolutionary distance between lophotrochozoans and model eukaryotes ( e . g . , Caenorhabditis and Drosophila ) with well-characterised sRNA complements [17–20] , efforts to characterise sRNA fragments other than miRNAs in flatworms have been restricted largely to S . japonicum [13 , 14 , 21] , and are still far from comprehensive . In the present study , we explored S . haematobium sRNAs at the transcriptome level . As a foundation for annotating these sRNAs , we improved the annotation of genes , including 5’- and 3’-untranslated regions ( UTRs ) and tRNA genes . We produced sRNA fragment libraries from adult males and females of S . haematobium , in order to define conserved miRNAs among schistosomes species for which miRNA data are available , and characterised novel RNA fragments to provide additional evidence for conserved roles of coding and non-coding RNAs in schistosome development and reproduction .
The sRNA libraries were constructed from total RNA extracted from the same pools of adult male ( n = 50 ) and adult female ( n = 50 ) S . haematobium worms used previously for mRNA sequencing [5] . Briefly , this S . haematobium strain originates from Egypt and was maintained in the Biomedical Research Institute , Rockville , Maryland [22] in Bulinus truncatus ( snail intermediate host ) and Mesocricetus auratus ( hamster; mammalian definitive host ) . Helminth-free hamsters were each infected with 1000 cercariae . After 90 days , paired male and female adults of S . haematobium were collected from M . auratus , following the perfusion of the mesenteric and intestinal vessels using physiological saline ( 37°C ) [5] . As 5’- and 3’-UTRs were not defined in the S . haematobium genome [5] , they were predicted here by de novo assembly of published RNAseq data for adult and egg stages of S . haematobium ( NCBI SRA accession numbers: SRR6655497 , SRR6655495 and SRR6655493 ) using Trinity ( release 10 Nov . 2013 [23] ) . First , transcripts ( BioProject: PRJNA431931 ) were aligned to published genome scaffolds ( BioProject: PRJNA78265 ) using BLAT v . 34x12 [24] , with the highest-scoring transcript alignments recorded in GTF format using a Perl script ( available from https://github . com/vikas0633/perl/blob/master/blat2gff . pl ) . Second , combined RNAseq datasets ( accession numbers: SRR6655497 , SRR6655495 and SRR6655493 ) were mapped to the genome using Tophat v . 2 . 1 . 1 and Cufflinks v . 2 . 2 . 1 [25] , providing published gene predictions ( -G option ) [5] . Third , the published gene set ( GFF format ) , gene models of curated gene families ( GFF format ) [26 , 27] , and aligned and mapped transcripts ( GTF format ) were merged using Cuffmerge [25] . Fourth , predicted genes nested within or partially overlapping with another gene were identified using gffread ( -E option; https://github . com/gpertea/gffread ) and manually curated or removed ( redundant gene loci ) . Finally , the merged gene annotation file was processed using GAG v . 2 . 0 . 1 ( https://genomeannotation . github . io/GAG/ ) , Sequin ( https://www . ncbi . nlm . nih . gov/Sequin/ ) and tbl2asn ( https://www . ncbi . nlm . nih . gov/genbank/tbl2asn2/ ) to confirm high-quality gene models and to remove overlapping , low complexity genes . The gene set , including annotations of the mRNAs , coding domains ( open reading frames , ORFs ) and UTRs , was submitted to NCBI ( BioProject: PRJNA78265 ) and used in the sRNA analyses described herein . In addition to coding domains , tRNA genes encoded in the genome were predicted using tRNA-scan v . 1 . 4 [28] and manually curated . The identification , classification and positions of repeats in the S . haematobium genome had been established previously [5] . Methods used for the isolation and quality assessment of total RNA were described previously [5] . The sRNA libraries representing male or female adults of S . haematobium ( accession numbers: SRR6655496 and SRR6655494 ) were constructed ( TruSeq Small RNA Library , Illumina ) and sequenced ( HiSeq 2500 sequencing platform , Illumina ) according to manufacturer’s instructions . Reads were filtered for low quality ( Phred score: < 35 ) and adapters removed using Trimmomatic [29] , retaining only reads with the 3’-adapter plus ≥ 8 bases . For the prediction of miRNAs and the full sncRNA complement , miRDeep2 v . 2 . 0 . 0 . 5 [30] and ShortStack v . 1 . 2 . 4 [31] were used , respectively , employing combined ( i . e . representing adult male and female S . haematobium ) sRNA sequence libraries . For miRDeep2 , known miRNAs of S . mansoni [10–12 , 32] and of S . japonicum [13 , 14 , 33 , 34] , and the Rfam microRNA database ( version 11 [35] ) , which represents members of Lophotrochozoa , were employed . For ShortStack , the minimum read depth was set at 10 , and the minimum and maximum Dicer processing sizes at 18 and 30 , respectively . Only gene regions containing sRNA read clusters ( i . e . precursor elements ) predicted to be miRNAs ( using miRDeep2 ) or containing hairpin ( HP ) structures ( using ShortStack ) were retained for further analysis . Candidate miRNAs were filtered based on miRDeep2 scores of ≥ 10 , or 100% sequence identity to known miRNAs ( across the seed region ) from other lophotrochozoans represented in miRBase [36] or Rfam [35] , or to miRNAs of S . mansoni and/or S . japonicum [10–14 , 32–34] . Stable ( i . e . non-randomly degraded ) , bona fide sRNAs were identified by selecting locus-specific ShortStack sRNA clusters that were less than 20 nucleotides in length and were represented by 100 or more mapped sRNA reads . In addition , ShortStack precursors exhibiting a Watson-Crick strand bias [37] between 0 . 2 and 0 . 8 were excluded . Since sRNA sequencing libraries were constructed from 50 male and 50 female worms , the most prevalent sRNA read was selected from each ShortStack precursor to represent the non-redundant consensus sRNA within the precursor ( = representative sRNA ) . Following definition of the non-redundant sRNA sequences , each ShortStack precursor representative sRNA was classified using Infernal v . 1 . 1 . 1 [38] and assessed for homologs in the Rfam database . ShortStack precursors predicted by Infernal to be tRNA or rRNA were removed from this sRNA set and recorded separately . Subsequently , sRNA clusters that were located entirely within a repeat region were identified . A cluster was defined as repeat-derived or repeat-complementary , depending on whether it was predicted on the same or the opposite strand as the repeat . The stability of selected sRNAs was assessed using the RNAfold software in the ViennaRNA v . 2 . 1 . 8 package [39] . Transcription levels of miRNAs and all sncRNAs were inferred and normalised as counts per million reads mapped ( CPM ) using miRDeep2 and ShortStack , respectively . Each 3'-UTR was screened for miRNA binding sites using the programs miRanda v . 3 . 3a [40] and PITA v . 6 [41] . A binding site was considered as valid if it had a score of > 300 ( miRanda ) and < -10 ( PITA ) .
The improved gene set for S . haematobium inferred here contains 10 , 837 genes coding for 11 , 140 proteins , compared with 13 , 073 originally reported , for which no transcript isoforms had been predicted [5] . Of the 10 , 837 improved gene models predicted here , 4100 ( 38% ) included annotations for both 5’- and 3’-UTRs , 1244 ( 11% ) for 5’-UTRs , and 2596 ( 24% ) for 3’-UTRs . In total , ~ 49 . 6 million and ~ 41 . 3 million sRNA reads were sequenced for male and female adults of S . haematobium , respectively ( Table 1 ) . Following filtering , ~ 42 . 2 million ( male ) and ~ 33 . 9 million ( female ) sRNA reads that had a 3’-adapter , lacked 5’-contaminants and were > 18 nucleotides in length were used for analyses ( Table 1 ) . Of these , 3 , 078 , 078 ( 7 . 3%; male ) and 3 , 983 , 507 ( 11 . 8%; female ) were distinct ( non-redundant ) reads ( Table 1 ) , with lengths usually ranging between 18 and 28 nucleotides ( median of 20 to 21; Fig 1 ) . Most quality-filtered reads from female ( 68 . 1% ) and male worms ( 80% ) mapped to the S . haematobium genome and were then used to identify and classify sncRNA elements transcribed in adult S . haematobium . A total of 89 transcribed miRNAs were identified in S . haematobium , which compares with 112 in S . mansoni [10] and 78 in S . japonicum [13] . Of these 89 miRNAs , 27 , 16 and 12 miRNAs were inferred to have an homologous miRNA seed in S . mansoni , S . japonicum and among the three schistosome species , respectively ( Fig 2; S1 Table ) . Five miRNA seeds were homologous to miRNAs in the Rfam [35] or miRBase [36] databases , namely: mir-398 ( Rfam ID: RF00695 ) , mir-450 ( RF00708 ) , mir-598 ( RF01059 ) , mir-785 ( RF02244 ) and cte-mir-981 ( miRBase ID: MI0010092 ) . Reads representing miRNAs accounted for 16% and 9% of all sRNA reads sequenced from male and female adults of S . haematobium , respectively . The most highly transcribed miRNAs in male worms were sha-mir-1 ( 5 . 9% of mapped sRNA reads ) , sha-mir-71a ( 5 . 8% ) , sha-mir-125b ( 1 . 6% ) , sha-mir-7a ( 1 . 0% ) and sha-let-7 ( 0 . 6% ) . In female worms , the most highly transcribed miRNAs were sha-mir-71a ( 3 . 6% of mapped sRNA reads ) , sha-mir-1 ( 2 . 0% ) , sha-mir-71b ( 0 . 7% ) , sha-mir-125b ( 0 . 7% ) and sha-bantam ( 0 . 3% ) ( S1 Table ) . In total , 59 of 89 miRNAs ( 66% ) exhibited substantial female-biased transcription , with CPMs of more than twice that of respective miRNAs in the male library . Of these 59 miRNAs , 44 ( 75% ) were novel ( or ‘orphan’ ) miRNAs without known homologs in other schistosome species studied to date or in miRNA databases . In contrast , five of the 89 miRNAs ( 6% ) had male-biased transcription ( more than two-fold higher CPM values ) , with four of them being orphan miRNAs ( Fig 2; S1 Table ) . A total of 798 potential binding interactions were identified between S . haematobium miRNAs and predicted 3’-UTRs , of which 519 represented unique interactions between 46 miRNAs and 332 genes ( excluding isoforms; S2 Table ) . Genes with more than five miRNAs predicted to bind a 3’-UTR included a “methyltransferase-like protein 14” ( MS3_05841; eight miRNAs ) , a “serine/threonine-protein phosphatase 2B” ( MS3_00821; seven miRNAs ) , a “putative member of the sno and ski oncogene family” ( MS3_02059; six miRNAs ) and an uncharacterised protein ( MS3_10507; six miRNAs ) ( S2 Table ) . The majority ( n = 38; 82 . 6% ) of miRNAs predicted to bind 3’-UTR elements were associated with more than one gene; these miRNAs included sha-miR-450_RF00708 ( 104 genes ) , sha-miR-71a ( 53 genes ) , sha-miR-71b ( 43 genes ) , sha-miR-new_38 ( 39 genes ) and sha-miR-new_36 ( 37 genes ) . A large number of sRNA reads mapped to rRNA ( 22 , 697 , 561 of 56 , 793 , 790 total reads mapped; 40%; 747 clusters ) and tRNA ( 3 , 386 , 888 of 56 , 793 , 790 total reads mapped; 6%; 782 clusters ) regions ( S3 and S4 Tables ) . Of the remaining reads , which formed 3875 sncRNA clusters , 49 ShortStack precursors had RNA products of a consistent size and characteristic , containing a representative RNA sequence with conservation to sRNA in the Rfam database ( S5 and S6 Tables ) . Among these precursors were clusters encoding most known spliceosomal small nuclear RNA ( snRNA ) classes , including ten U6 ( RF00026 ) , three U5 ( RF00020 ) , three U1 ( RF00003 ) , two U6atac ( RF00619 ) , two U2 ( RF00004 ) and one U4atac ( RF00618 ) sequences . Of these snRNAs , U1 and U2 were the most highly transcribed ( CPM for U1 in library of male worms: 91 . 9; CPM for U2 in library of female worms: 74 . 4 ) . Three other clusters encoded the small nucleolar RNAs ( snoRNAs ) SNORD15 ( RF00067; two clusters; CPM range in library of male worms: 66 . 7–67 . 8; library of female worms: 56 . 3–58 . 8 ) and SNORA19 ( RF00413; one cluster; CPM range: 0 . 7–0 . 8 ) ( S5 Table ) . Additionally , 16 elements had similarity to “hammerhead 1 ribozyme-like RNA” ( RF00163; HHR ) within sRNA clusters that were between 55 and 404 nucleotides in length , of which seven and two overlapped completely with DNA predicted to encode SINE ( short interspersed nuclear element ) /tRNA and LINE ( long interspersed nuclear element ) /RTE-BovB elements , respectively ( S5 Table ) . None of these clusters overlapped with a region in the genome predicted to encode tRNAs ( to which repetitive elements coding for HHRs can be similar in sequence [42] ) , suggesting that they are bona fide SINEs/LINEs . Of all 16 clusters , nine were predicted ( using ShortStack ) to be ‘hairpin-derived’ . Eleven sRNA cluster sequences shared 95–100% sequence identity with predicted S . haematobium HHRs in the Rfam database . In addition , one cluster ( Cluster_22636 ) matched ( 91% nucleotide identity across 56nt using blastn [43] ) to a previously characterised S . haematobium HHR ( accession number: AF036390 . 1 [42] ) . Matching regions of the consensus sequences of these clusters were extracted and , using RNAfold , folds resembling that of an HHR were predicted for eight sequences . Three of the 11 putative HHRs derived from sRNA clusters ( Cluster_12997 , Cluster_13427 and Cluster_13865 ) contained the CUGANGA motif , which is a conserved part of the HHR catalytic core [42] . All HHR-like clusters were transcribed at low levels , except for Cluster_16211 and Cluster_86205 which were predominantly transcribed at higher levels in libraries representing male ( CPM: 13 . 2 ) and female ( CPM: 16 . 5 ) adults of S . haematobium , respectively . In addition to snRNAs and hammerhead-like RNAs , 30 potential miRNA precursors were identified , including mir-80 ( RF00817 ) and lin-4 ( RF00052 ) , which lacked the required characteristics of miRNAs , including not being identified by miRDeep2 and not being predicted to have the characteristic miRNA hairpin secondary structure ( S5 Table ) . In addition to miRNAs and other known sncRNAs , 3798 sRNA clusters in the genome of S . haematobium were predicted to share characteristics of non-randomly degraded RNA but could not be annotated based on known sRNA elements in other organisms . To explore these elements in more detail , we assessed whether these sRNA clusters were represented in coding or non-coding regions of the genome and whether they overlapped with a repeat region ( S6 Table and Fig 3A ) . In total , 120 clusters were within exons ( from 102 genes ) , of which 25 were within the translation start site ( TSSa ) of the gene . Of these 25 , two clusters had high transcription levels in the library derived from male worms ( CPM range 12 . 1–43 . 1 ) , whereas the two most highly transcribed sRNAs in the library derived from female worms exhibited moderate transcription levels ( CPM range: 6 . 6–8 . 6 ) . In coding regions , eight clusters of sRNAs were on the strand complementary to an exon and were thus predicted to play a role in ( siRNA-mediated ) gene silencing ( Fig 3A ) . Of these clusters , Cluster_26190 , Cluster_8504 and Cluster_84791 were the most highly transcribed in one or both sexes ( CPM range: 1 . 2–24 . 1 ) . On the sense strand , 483 clusters were found within introns , of which 270 and 213 clusters mapped to regions containing repeat elements on the sense and antisense strand , respectively ( Fig 3A and 3B; S6 Table ) . Approximately 1% of the reads representing uncharacterised sRNA clusters mapped to transposable elements ( TEs; such as DNA/En-Spm , DNA/Merlin , LINEs , long terminal repeats ( LTRs ) and SINEs/tRNAs; Fig 3A , 3B and 3C ) with a slight antisense bias ( > 51% of reads ) , consistent with Piwi-interacting RNA-like elements ( piRNAs ) [44] . Although schistosomes lack a canonical Piwi pathway [15 , 16] , three sRNA precursors complementary to repeat elements were identified in the genome . However , these elements were not highly transcribed in libraries derived from male or female worms ( CPM range: 0 . 4–2 . 3; S6 Table ) .
Employing a substantially improved gene set for S . haematobium with enhanced gene annotation ( relating to 3’- and 5’-UTRs , longer coding sequences and reduced redundancy ) and sRNA-Seq libraries for male and female adults of S . haematobium , we defined here the transcribed complements of miRNAs and other sncRNAs in this species . Although the number of miRNAs transcribed in S . haematobium adults was comparable with numbers reported for other schistosome species studied thus far [10 , 11 , 13 , 21] , there was little conservation of miRNA homologs among schistosome species , and many miRNAs in S . haematobium appeared to be species-specific . These findings are consistent with reports of a substantial loss of ‘conserved’ miRNAs from flatworms and a gain of novel miRNA families [8 , 10] . For example , the 12 schistosome-specific miRNAs conserved between S . mansoni and S . japonicum [10] were not found in S . haematobium , and of the miRNAs reported as S . mansoni-specific [10] , only one was conserved and shared by S . haematobium . In this context , it is noteworthy that only miRNAs transcribed in adult S . haematobium worms were identified in this study . Thus , the apparent lack of some miRNA homologs might be explained by their specific transcription in one or more developmental stages for which data exist for S . mansoni and/or S . japonicum , but not yet for S . haematobium . For three other miRNAs—mir-190 , mir-281 and mir-8451—reported to be conserved among the Bilateria , Protostomia and Platyhelminthes , respectively [10] , no representatives were detected in S . haematobium . For mir-8451 , this finding might be explained by this miRNA not being transcribed in the adult stages of schistosomes , consistent with earlier findings [10 , 13 , 21] . However , transcription in adult stages has been reported for mir-190 in both S . japonicum [13 , 21] and S . mansoni [10] , and for mir-281 in S . mansoni [10] , suggesting that homologs of these miRNAs were not identified here in S . haematobium due to the stringency with which miRDeep2 defines high-confidence homologs , requiring an exact nucleotide match for positions 2–8 ( ‘seed’ ) of the mature sequence . To test this hypothesis , we assessed whether mature and precursor sequences of sma-mir-190 , sma-mir-281 and sma-mir-8451 were homologous to any S . haematobium miRNAs identified using a less stringent ( blastn -task “blastn-short” ) approach [43] . This analysis revealed exact matches of 7–9 nucleotides to several S . haematobium miRNAs ( including novel or orphan sequences ) for all three S . mansoni sequences . However , when directly comparing positions 2–8 of the homologous mature sequences of S . haematobium and S . mansoni , exact matches could not be inferred for any of them . This finding suggests that the lack of homologs for some S . haematobium miRNAs is due to different approaches used to infer consensus pre-miRNA sequences and mature sequences from mapped sRNA reads . Thus , sequences were not annotated as homologs to a known sequence , unless homology could be inferred using the stringent , high-confidence approach employed in miRDeep2 . Future studies should focus on defining miRNA complements of additional developmental stages of S . haematobium to allow for a better comparison to miRNAs reported in different developmental stages of other schistosome species [13 , 21] , and to assess potential losses/gains of particular miRNAs in S . haematobium and other species of schistosomes . The availability and quality of genome assemblies can also have a marked impact on miRNA detection , as exemplified in a recent publication [7] , which demonstrated significantly improved inference of miRNA employing a draft genome for Fasciola hepatica over findings without a genome [45] . Thus , additional miRNAs are likely to be detected using enhanced genome assemblies for S . haematobium in the future . The present finding that many miRNAs of S . haematobium are primarily transcribed in female adult worms is largely concordant with those reported previously for S . mansoni and/or S . japonicum [10 , 21] . Specifically , female-biased transcription of six and two miRNAs , whose homologs also have female-biased transcription in S . mansoni [10] and S . japonicum [21] , respectively , and male-biased transcription for mir-1b ( reported for S . mansoni [10] ) were established . Notably , among the female-biased transcripts were the three miRNAs , sha-mir-71b , sha-mir-2b and sha-mir-2c ( encoded in a cluster on scaffold KL251164 . 1 ) , whereas the related miRNAs sha-mir-71a and sha-mir-2a ( encoded in a cluster on scaffold KL250488 . 1 ) did not show sex-biased transcription . These results are consistent with reports of a mir-71/mir-2 cluster duplication in S . mansoni [10 , 11] and S . japonicum [14 , 34] on the female-sex ( W ) chromosome and on one autosome ( chromosome 5 ) , and further support roles in sex-specific traits , sexual differentiation , pairing of adult worms and reproductive processes in schistosomes . Similar to the finding of conserved miRNAs among schistosomes , female-biased transcription for the majority of novel , S . haematobium-specific miRNAs , supports proposed roles in reproductive biology and/or pairing of adult worms . Moreover , the four novel miRNAs transcribed principally in male adults , including one miRNA ( sha-miR-new_46 ) encoded on the same scaffold as the mir-71/mir-2 cluster ( KL250488 . 1 ) with a CPM more than 20-times higher than that in the library derived from adult female worms , suggest male-specific roles for these miRNAs in gene regulation . In addition to the analysis of transcriptional profiles , an improved gene annotation and inference of 3’-UTRs enabled a homology-based prediction of miRNA binding sites in the genome . Predictions of targets of S . haematobium miRNAs ( S2 Table ) did not agree with those made for the respective S . mansoni homologs [11] . The number of targets for mir-71a and mir-71b identified here was similar to that of a previous study of S . mansoni miRNAs [11] , but overall , the number and identity of targets were discordant with results from the present study . Reasons for this discrepancy might be: ( i ) that the previous study of S . mansoni employed miRanda ( minimum score threshold: 120 ) , whereas here , a combination of miRanda ( minimum score threshold: 300 ) and PITA was used to predict targets; or ( ii ) due to difference in the number of annotated 3’-UTRs in the gene sets between the two species . The current S . mansoni gene set ( PRJEA36577; WormBase ParaSite v . 9 ) has annotations for 3’-UTRs representing 4 , 534 transcripts , in contrast to 6 , 696 transcripts with 3’-UTRs identified for S . haematobium . A comparison of 3’-UTRs of 10 , 015 pairwise ( amino acid ) orthologs between S . haematobium and S . mansoni showed that only 21% ( n = 2064 ) of orthologs had a 3’-UTR annotation in both species . Clearly , future work ( including additional RNA-Seq libraries and/or HITS-CLIP [46] ) to unify and improve the annotation of UTRs in schistosome species is warranted to gain a better understanding of miRNA-targeting in schistosomes . In this context , it would also be valuable to experimentally identify and validate miRNA targets , as has been recently reported for S . japonicum [47] . Other sRNAs identified included snRNAs , snoRNAs , HHR-like RNAs and uncharacterised sncRNAs , some of which were predicted to be TE-derived . For snRNAs , most classes known to constitute the spliceosome and the minor spliceosome [48] were identified , but representatives of U4 , U11 or U12 were not . It is likely that S . haematobium sequences representing these three classes are more similar in sequence to the Rfam models for U4atac , U1 and U2 , respectively , and were thus assigned to the latter ( or other ) classes . This prediction is supported in that two to ten loci were identified to encode five different classes of snRNAs . Similarly , only three representatives of snoRNAs which are predicted to function in the biogenesis of other rRNAs , tRNAs and snRNAs [49] were detected . Other snoRNAs in S . haematobium might not have been classified due to their sequence divergences from Rfam models and might thus be amongst the 3798 unclassified sRNAs . Of the unclassified sRNAs , sRNA products of a consistent size mapped to the antisense strand of predicted TEs . Although S . haematobium and other schistosomes lack a canonical Piwi pathway [15 , 16] , this information suggests that schistosomes can suppress TEs via TE-derived sncRNAs , thus contributing to genomic stability , as proposed for sncRNAs in S . japonicum [13] . Schistosomes lack Piwi and Vasa proteins and other argonaute proteins of the Piwi family [15] , which have been recognised as being relatively conserved among members of the Deuterostomia and Ecdysozoa [15] . Thus , it is plausible that S . haematobium and other schistosomes have evolved alternative sRNA-mediated RNA degradation pathways to control ‘jumping’ genes [50] , similar to those in nematodes [51] . It is possible that the degradation of TEs is regulated by some of the 16 HHR-like sRNAs identified here , with varying levels of evidence and confidence ( i . e . having a hairpin-fold , conserved motifs and/or sequence homology; S5 Table ) . HHRs are RNA enzymes found within repetitive TEs , such as SINEs [52] . They can self-cleave the SINE in which they are encoded , thus controlling the propagation of TEs in the genome . Additionally , some HHRs can cleave other RNA targets [42] , thus regulating their transcription , and might also have roles in the processing of tRNAs , sRNAs , and RNAi inhibition [52] . Although the present results do not allow for the prediction of a function for HHRs identified in S . haematobium , they provide experimental evidence ( at the RNA level ) that they are encoded in the genome . Combined with earlier reports of both HHRs [52] and TEs [53] in several trematodes , the present data provide a foundation for future investigations of these intriguing RNA molecules and their roles in the molecular biology of schistosomes .
The present work represents the first genome-wide miRNA and sncRNA resource for S . haematobium , extending previous work on schistosomes , and providing additional evidence with regard to the conservation of miRNAs across flatworms [8] . The outcomes from this work should facilitate future research of the post-transcriptional regulation of genes in schistosomes , and the roles that sncRNAs play in development and parasite-host interactions . Given the proposed involvement of these RNAs in the infection process , parasite-host cross-talk and development of drug resistance , and their potential relevance as drug targets [46 , 54] , an improved understanding of sncRNAs transcribed in developmental stages within the definitive host , are much needed . The findings from the present study provide a foundation for such future endeavours .
|
Human schistosomiasis is a chronic , neglected tropical disease ( NTD ) that is predominantly caused by the blood flukes Schistosoma haematobium , S . mansoni and S . japonicum . Infections by S . haematobium and/or S . mansoni are highly prevalent in Africa , affecting ~ 200 million people . The decoding of schistosome draft genomes has , to some extent , improved our understanding of the molecular biology of these parasites and now allows for non-protein-coding regions in these genomes to be characterised . Here , we explored small RNAs in adult S . haematobium by deep sequencing , reference genome mapping and detailed bioinformatic analyses . This study provides the first genome-wide miRNA and sncRNA resource for S . haematobium , extending earlier work on schistosomes and facilitating future curation efforts and functional investigations of schistosome sncRNAs . These efforts should enable a better understanding of post-transcriptional RNA modifications , gene regulation and novel aspects of parasite development , parasite-host cross-talk and disease at the molecular level .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion",
"Conclusions"
] |
[
"sequencing",
"techniques",
"schistosoma",
"invertebrates",
"schistosoma",
"mansoni",
"helminths",
"gene",
"regulation",
"animals",
"invertebrate",
"genomics",
"micrornas",
"genome",
"analysis",
"molecular",
"biology",
"techniques",
"rna",
"sequencing",
"research",
"and",
"analysis",
"methods",
"schistosoma",
"japonicum",
"genomic",
"libraries",
"schistosoma",
"haematobium",
"gene",
"expression",
"molecular",
"biology",
"animal",
"genomics",
"biochemistry",
"rna",
"eukaryota",
"nucleic",
"acids",
"genetics",
"biology",
"and",
"life",
"sciences",
"genomics",
"non-coding",
"rna",
"computational",
"biology",
"organisms"
] |
2018
|
The small RNA complement of adult Schistosoma haematobium
|
During cell division , segregation of sister chromatids to daughter cells is achieved by the poleward pulling force of microtubules , which attach to the chromatids by means of a multiprotein complex , the kinetochore . Kinetochores assemble at the centromeric DNA organized by specialized centromeric nucleosomes . In contrast to other eukaryotes , which typically have large repetitive centromeric regions , budding yeast CEN DNA is defined by a 125 bp sequence and assembles a single centromeric nucleosome . In budding yeast , as well as in other eukaryotes , the Cse4 histone variant ( known in vertebrates as CENP-A ) is believed to substitute for histone H3 at the centromeric nucleosome . However , the exact composition of the CEN nucleosome remains a subject of debate . We report the use of a novel ChIP approach to reveal the composition of the centromeric nucleosome and its localization on CEN DNA in budding yeast . Surprisingly , we observed a strong interaction of H3 , as well as Cse4 , H4 , H2A , and H2B , but not histone chaperone Scm3 ( HJURP in human ) with the centromeric DNA . H3 localizes to centromeric DNA at all stages of the cell cycle . Using a sequential ChIP approach , we could demonstrate the co-occupancy of H3 and Cse4 at the CEN DNA . Our results favor a H3-Cse4 heterotypic octamer at the budding yeast centromere . Whether or not our model is correct , any future model will have to account for the stable association of histone H3 with the centromeric DNA .
During eukaryotic cell division sister chromatids , containing identical copies of genetic information , are pulled apart and driven towards opposite spindle poles by the microtubules of the mitotic spindle , which attach to the centromeric DNA sequences of the sisters via kinetochore protein complexes . It is imperative for proper chromosomal segregation that each chromosome assembles the kinetochore only at one site . The sites of kinetochore assembly are marked by specialized nucleosomes . Budding yeast represents the simplest case in which a single microtubule attaches to the so-called “point” kinetochore assembled around a single centromeric nucleosome . More complicated “regional” centromeres of most other eukaryotes are composed of arrays of specialized centromeric nucleosomes interspersed with conventional nucleosomes [1] and support the assembly of several microtubule attachment sites . Centromeric nucleosomes were reported to have histone H3 substituted by a histone variant , CENP-A , called Cse4 in budding yeast [2] . It displays more than 60% similarity with the conventional histone H3 within the histone fold domain and has an additional N-terminal extension [3] . CENP-A has been demonstrated to co-purify with a subset of kinetochore proteins and is likely to provide interaction surfaces for kinetochore assembly [4] , [5] . Recruitment of CENP-A to centromeric DNA requires the CENP-A targeting domain ( CATD ) , comprised of loop1 and the α2-helix [6] , [7] , and is regulated by a number of other proteins [8] . One example is the non histone protein Scm3 ( HJURP in human [9] ) , which is believed to be a histone chaperone required for recruitment of CENP-A to centromeres [10]–[18] . CENP-A overexpression in metazoans [19] and budding yeast [20] leads to its mislocalization . In budding yeast mislocalized Cse4 is very unstable [21] . Although budding yeast [22] and fission yeast [14] , [23] , [24] appear to be an exception , in several organisms CENP-A is loaded on the DNA outside of S phase , in anaphase of mitosis or the following G1 [25] , [26] , when it is proposed to replace histone H3 . Despite a significant progress in the field , the exact function of CENP-A at the centromere remains a mystery . CENP-A and H4 were reported to form a more compact and conformationally more rigid heterotetramer compared to the heterotetramer of histones H3 and H4 [6] , [27] . However , the significance of the structural differences between H3 and CENP-A to their function is unknown . Even the question of the exact composition and localization of centromeric nucleosomes has not been resolved to date and remains the subject of controversy [28] . Besides an octamer composed of two molecules each of CENP-A , H2A , H2B and H4 , a hexamer model in which Scm3 replaces H2A and H2B [11] , [17] and a hemisome model which proposes a tetramer consisting of one copy each of Cse4 , H4 , H2A and H2B [29]–[32] were also proposed . Regional centromeres of higher eukaryotes can accommodate different versions of CENP-A-containing nucleosomes . While budding yeast with their point centromeres is an appealing model system to study the centromeric nucleosome , it is possible that the yeast centromeric nucleosome might also possess unique features . Here we report the results of our analysis of the yeast centromeric nucleosome using a novel chromatin immunoprecipitation technique and discuss them in the context of the previously proposed models of the CENP-A containing nucleosome .
The composition of the centromeric nucleosome was previously analyzed by means of chromatin immunoprecipitation ( ChIP ) [11] , [12] in yeast . In a conventional ChIP approach proteins are chemically cross-linked to DNA , the chromatin is fragmented by sonication to about 500 bp size , and immunoprecipitated fragments are identified in PCR or microarray hybridization assays . This approach suffers certain drawbacks when applied to the centromere . The DNA fragment size is much larger than the region accommodated by a conventional nucleosome ( 146 bp ) , which limits the resolution . This problem can in principle be overcome by the treatment of chromatin with micrococcal nuclease , which specifically digests the internucleosomal linker DNA . However the size of kinetochore footprint is highly variable depending on the digest conditions [33] , [34] and apparently poses an accessibility problem for antibodies since the efficiency of the co-immunoprecipitation of the CEN DNA with canonical histones is very low compared to pericentric regions [11] , [12] , [35] . In addition , PCR with a specific pair of primers or microarray hybridization detect larger DNA fragments without identifying them by size , which imposes further limits on resolution . We developed new versions of ChIP to reveal the composition of the centromeric nucleosome in budding yeast . There are three main differences from conventional ChIP . First , we performed our experiments with and without the chromatin cross-linking . We reasoned that omitting cross-linking improves the accessibility of the centromeric nucleosome to antibodies and prevents potential artifacts due to the cross-linking of loosely associated proteins . However , because cross-linking prevents local re-arrangements due to nucleosomal sliding along the DNA , we also included cross-linked samples in our analysis . Second , we flanked CEN DNA by restriction sites and excised it by a specific endonuclease similar to earlier studies by [36] . Finally , analysis of the immunoprecipitated DNA was performed using methods that identify the isolated fragments by size , initially by a Southern blot with specific probes hybridizing to the excised CEN fragment . In experiments where qPCR with a specific pair of primers was used , the immunoprecipitated DNA was size-fractionated prior to PCR to preclude the detection of uncut DNA . The Biggins's laboratory recently employed a similar approach [37] . In this study , micrococcal-nuclease digested chromatin was immunoprecipitated with an anti-Cse4 antibody and analyzed by Southern blot . The results demonstrated a single Cse4 nucleosome positioned at the budding yeast centromere but did not address its composition further . In our initial experiments we used a small minichromosome that contained the CEN region of chromosome IV ( Figure S1A ) . We utilized strains with HA-tagged versions of H3 and Cse4 and found that the minichromosome can be specifically co-immunoprecipitated with an anti-HA antibody even in the absence of cross-linking ( Figure 1A ) . This result demonstrates that the minichromosome assembles conventional nucleosomes as well as a centromeric nucleosome . Next , we tested whether it is possible to digest the minichromosome in yeast cell lysate and subsequently immunoprecipitate the fragments . We constructed minichromosomes with BglII sites at different positions with respect to CEN . The digest efficiency was highly variable depending on the position of the BglII site ( Figure S1B ) . It was previously reported that the centromeric DNA is inaccessible for the nuclease digest [33] , [34] . However , under our conditions it was possible to excise CEN DNA and even to cut it between CDEII and CDEIII in agreement with the previous results by [38] , [39] . In subsequent ChIP experiments we used a minichromosome with BglII restriction sites 50 bp upstream and downstream of CEN4 boundaries flanking a 214 bp CEN fragment . The chromatin was digested with the endonuclease BglII and immunoprecipitated with an anti-HA antibody ( Figure 1B ) . A probe hybridizing to the TRP1 gene located on the minichromosome outside of CEN was used for the Southern blot . Due to an incomplete chromatin digest , a linearized full-length minichromosome and a CEN-less fragment could be detected . Only the full-length linearized minichromosome co-immunoprecipitated with Cse4-HA6 while both the full-length linearized minichromosome and the CEN-less fragment were recovered with HA-tagged histones H4 , H2A , H2B and H3 ( Figure 1C ) . Therefore , although the minichromosomes assemble conventional nucleosomes along their entire length , only CEN DNA is associated with Cse4 , which is in agreement with [37] . Since it was proposed recently that the Scm3 histone chaperone might replace H2A/H2B dimers in the centromeric nucleosome [11] , [17] we performed the minichromosome ChIP with the Scm3-HA6 strain . We could not co-immunoprecipitate the minichromosome with HA-tagged Scm3 under our conditions indicating that Scm3 is unlikely to be a part of the centromeric nucleosome ( Figure 1C ) . The observation that no CEN-less fragment was recovered in the Cse4-HA6 immunoprecipitation rules out lateral sliding of Cse4 nucleosome during the course of the immunoprecipitation as well as tethering of DNA fragments via protein-protein interactions , e . g . , between centromeric and conventional nucleosomes in our assay . The efficiency of immunoprecipitation of the minichromosome fragments of approximately 1000 bp and longer was exceptionally high and close to 100% . When a 930 bp fragment from ARS1 until position +50 downstream of CDEIII was excised , it could be depleted from yeast cell lysate with anti-HA antibodies recognizing Cse4-HA6 while virtually none of the remaining CEN-less fragment of the minichromosome could be detected on the beads ( Figure S2 ) . Considering the immediate proximity of the +50 cutting site to the centromere it is highly unlikely that there was a significant local rearrangement of nucleosomes and/or tethering of the CEN fragment to the rest of the minichromosome under our experimental conditions . The detection of the small 214 bp CEN fragment was very inefficient using the 32P-labelled probe . Therefore we employed a digoxygenin ( DIG ) -labeled locked nucleic acid ( LNA ) oligonucleotide ( Figure 1D ) with improved hybridization properties [40] . Using the LNA probe it was possible to detect the 214 bp fragment released from 6 pg of the minichromosome which corresponds to about 0 . 1% efficiency of immunoprecipitation starting with 150 ml of yeast culture in the early log phase ( Figure S3 ) . We could detect the 214 bp CEN fragment in the immunoprecipitates with Cse4 , H4 , H2A and H2B . Surprisingly , we reproducibly observed an interaction of H3 with the 214 bp CEN fragment using this method ( Figure 1D ) . This was in contrast with previous studies proposing that H3 is replaced by Cse4 at the centromere [2] . We next tested whether the interaction of H3 with CEN is dependent on the cell cycle stage as it is possible that Cse4 replaces H3 at a specific point in the cell cycle . The notion that the composition of the centromeric nucleosome might vary through the cell cycle was proposed earlier [17] , [28] . Yeast cultures were arrested in G1-phase with alpha-factor and in G2-phase with nocodazole/benomyl ( Figure S4B ) , and chromatin was digested with BglII to release the 214 bp CEN fragment prior to immunoprecipitation . Both H3 and Cse4 , as well as H2B , were found to be associated with CEN in G1-phase and in G2-phase ( Figure 2A ) . Although nearly a 100% efficiency of co-immunoprecipitation of the minichromosomes with Cse4-HA6 ( Figure 1A ) indicated that it is unlikely to be the case , it is possible that a fraction of minichromosomes assemble a conventional nucleosome at the centromere and this would explain the association of H3 with CEN DNA in the above experiments . To address this possibility we adapted our ChIP approach to the native centromeres on the chromosomes and introduced BglII restriction sites 50 bp upstream and downstream of CEN on chromosome IV . The excised “native” 214 bp CEN4 fragment could be efficiently co-immunoprecipitated with H3-HA3 and Cse4-HA6 ( Figure 2B ) . We conclude that both histones H3 and Cse4 localize to centromeric DNA in budding yeast . In order to rule out the possibility that Cse4 is replaced by H3 during our immunoprecipitation procedure , we mixed yeast cell lysate of an H3-HA3 strain that does not carry minichromosomes with lysate of an untagged H3 strain carrying the minichromosomes . We could not observe any immunoprecipitation of the minichromosome with anti-HA antibody from those mixed lysates ( Figure 2C ) . Thus there is little or no turnover of minichromosome-associated H3 in our cell lysates . However , this experiment could not rule out local rearrangement of nucleosomes such as lateral sliding in the course of our experimental procedure , which included long incubations . Therefore we cross-linked proteins to DNA with formaldehyde prior to immunoprecipitation . Adding formaldehyde to the spheroplasts dramatically reduced the efficiency of centromeric DNA co-immunoprecipitation with either Cse4 or H3 . This was partially due to the low yield of the minichromosome in the cleared lysate after centrifugation presumably because the minichromosomes were cross-linked to larger structures . However , when formaldehyde was added directly to yeast lysate the immunoprecipitation was not impeded . In order to minimize the potential rearrangement of nucleosomes after cell lysis , the duration of the restriction digest of the minichromosomes was limited to 5 minutes followed by formaldehyde addition and immunoprecipitation . We were able to efficiently co-immunoprecipitate the 214 bp CEN fragment with both Cse4 and H3 after cross-linking ( Figure S5A ) . Therefore , it is unlikely that the detection of H3 at the CEN DNA is due to nucleosomal sliding during our experimental procedure . A qPCR-based approach was employed to compare the efficiencies of co-immunoprecipitation of the CEN DNA with H3-HA3 and Cse4-HA6 . After excision of the 214 bp CEN fragment CEN DNA was co-immunoprecipitated with Cse4-HA or H3-HA using anti-HA antibodies , eluted off the beads using SDS , size-fractionated via agarose gel-electrophoresis to separate it from full-length minichromosome and quantified using a quantitative PCR reaction . Using this procedure , we ensured that the 214 bp CEN fragment was exclusively detected since no PCR product was obtained when the restriction digest step was omitted ( Figure 3A ) . We did not observe any significant differences in ChIP efficiencies with H3 and Cse4 when the same anti-HA antibody was used . Similar IP/input ratios were observed with and without crosslink ( Figure 3B ) with the CEN DNA located on a minichromosome and on the native chromosome IV flanked by restriction sites ( Figure 3C ) . Thus we have no indication that only some centromeres are associated with H3 . The association of H3 and Cse4 with yeast centromeres can be mutually exclusive , i . e . , a fraction of the centromeres are occupied by the Cse4 nucleosome while a different fraction assembles a conventional nucleosome containing H3 . Alternatively , H3 and Cse4 are co-occupying the centromeric DNA at the same time . In order to distinguish between these two possibilities we performed a sequential ChIP experiment . After excision of the 214 bp CEN fragment and formaldehyde cross-linking CEN DNA was co-immunoprecipitated with Cse4-Myc using anti-Myc antibodies covalently coupled to the beads ( Figure S5B and S5C ) , eluted off the beads using SDS , and re-immunoprecipitated with anti-HA antibodies recognizing H3-HA . The CEN DNA fragment eluted off the beads was decross-linked , size-fractionated via agarose gel-electrophoresis to separate it from uncut DNA , and quantified using a quantitative PCR reaction ( Figure 3D–3F ) . The efficiency of the second immunoprecipitation step in this experiment was approximately 100 fold higher than the “mock” immunoprecipitation from a strain in which only Cse4 was tagged and was comparable to that of H3-HA re-immunoprecipitation in the experiment where both the first and the second steps were performed with anti-HA antibodies . Similar results were obtained when CEN DNA was excised from the minichromosome ( Figure 3E ) or native chromosome ( Figure 3F ) . We conclude that H3 and Cse4 co-exist at least at some centromeres . Unfortunately , we could not perform the reverse experiment , i . e . , to immunoprecipitate the CEN DNA via HA-tagged histone H3 and then re-precipitate via Myc-tagged Cse4 , since we could not re-precipitate CEN DNA from Cse4-Myc strain with anti-Myc antibody in 0 . 1% SDS . Switching the tags was also unsuccessful since the H3-Myc6 strain was not viable . Because the length of our excised centromeric fragment ( 214 bp ) is much shorter than would be necessary to accommodate two conventional nucleosomes ( 292 bp assuming no linker DNA in-between ) or a conventional nucleosome and a Cse4 nucleosome ( 268 bp if the Cse4 nucleosome organizes only 121 bp of DNA [41] ) , it is plausible that the centromeric nucleosome is a heterotypic octamer with one molecule of H3 and one molecule of Cse4 . If the structure of this hypothetical heterotypic nucleosome is similar to the structure of the conventional nucleosome and the CENP-A containing nucleosome [41] , [42] , histones H3 and Cse4 are expected to form a four-helix bundle with parts of their α2 and α3 helices . In vertebrates and many other organisms the α2 helix of H3 contains a cysteine residue , C110 . These cysteine residues from two histones H3 within the same nucleosome are within 6 . 2 Å from each other [42] and were reported to form a disulfide bond under oxidizing conditions in vitro [43] . In human CENP-A the corresponding residue is a leucine , L112 , although CENP-A proteins from some other mammals , such as platypus , as well as birds and amphibians have a cysteine in this position . In the recently reported crystal structures of human CENP-A nucleosome the two leucines 112 are 4 . 8–5 . 7 Å apart [27] , [41] , which should allow cross-linking if they are mutated to cysteines . ( Figure S6A ) . In order to test whether a cross-link between two Cse4 molecules or between Cse4 and H3 is at all possible we co-expressed the histone fold domain of Cse4-Cys and the full-length H3-Cys in bacteria . We could observe the formation of spontaneous covalently cross-linked H3 homodimers , Cse4 homodimers and some H3/Cse4 heterodimers . The dimers were detected after denaturing SDS-electrophoresis and could be resolved by β-mercaptoethanol treatment indicating that they indeed resulted from the formation of the disulfide bond between the cysteine residues ( Figure S6B ) . We reasoned that disulfide bond formation between the two α2 helix cysteines would only be possible if the two histones form a four helix bundle and the ability to cross-link Cse4 and H3 would be a test of a heterotypic octamer model . Since in budding yeast neither H3 nor Cse4 contain cysteine residues , we mutated the corresponding alanine 111 and leucine 204 to cysteines . We were able to cross-link homodimers of H3-Cys in crude lysates and on isolated chromatin in the presence of 5 , 5′-dithiobis- ( 2-nitrobenzoic acid ) ( DTNB , Ellman's reagent ) , which has been reported to facilitate intermolecular disulfide bond formation between H3 histones in chicken nucleosomes [44] ( Figure S7A ) . We could also cross-link H3-Cys histones using cysteine-specific cross-linkers , bBBr and BMOE . However , we did not observe a reproducible cross-link either between two Cse4-Cys molecules or between Cse4-Cys and H3-Cys ( Figure S7B ) in crude yeast lysate or isolated chromatin . Thus we currently have no direct evidence for the presence of the heterotypic octamer at budding yeast centromeres . It is possible that the heterotypic nucleosome has a very unusual structure compared to the conventional H3-H3 nucleosome [42] or the human CENP-A-CENP-A octamer that were recently reported [27] , [41] and that this structure does not allow for the cysteine cross-link . It remains to be confirmed whether the cysteines can be cross-linked in the context of the fully assembled octamers . An alternative to the octamer is the hemisome model , which proposes a tetramer consisting of Cse4 , H4 , H2A and H2B histones [30] , [31] . Our refinement of this model will imply that in budding yeast in the immediate vicinity of the Cse4 hemisome there is either a conventional nucleosome or , possibly , an H3-containing hemisome . According to the recently reported structure , the human CENP-A-containing octamer assembled in vitro organizes 121 bp of DNA [41] while a conventional nucleosome wraps 147 bp of DNA . Thus , a Cse4 hemisome and a conventional nucleosome without any linker in-between would require approximately 207 bp which would fit with the size of our excised centromeric fragment of 214 bp . An important and testable prediction of this model is that Cse4 and histone H3 are incorporated into distinct structures , which can be potentially mapped to different stretches of DNA . The budding yeast centromere is defined by a 125 bp sequence [45] consisting of three elements . CDEI is a non-essential 8 bp palindrome , CDEII is 78–86 bp long and is composed of 87–98% A/T , and CDEIII is a highly conserved 25 bp sequence which binds the CBF3 protein complex [46] . We conducted a series of experiments in which we tested whether Cse4 and histone H3 associate with distinct elements within CEN DNA . It was reported earlier that CSE4 genetically interacts with CDEI and CDEII but not with CDEIII [47] suggesting that the Cse4-containing nucleosome is localized upstream of the CDEII/CDEIII boundary . Since we were able to cut the minichromosome between CDEII and CDEIII we hoped to gain further insights in the exact localization of Cse4 with regard to CEN by using our ChIP approach . We created a minichromosome with a restriction site between CDEII and CDEIII and a restriction site outside of the CEN DNA , in ARS1 . Using our ChIP approach we were able to co-immunoprecipitate Cse4-HA6 with both the CDEI/CDEII and the CDEIII-containing fragments ( Figure 4A ) suggesting that the centromeric nucleosome straddles the boundary between CDEII and CDEIII . However , an interaction with the CDEIII fragment appeared less efficient , indicating that the Cse4-containing nucleosome interacts mostly with the CDEI/CDEII region of the CEN DNA . An important corollary from this observation is that in our assay the Cse4-containing nucleosome ( or hemisome ) is not displaced from the CEN DNA to the edge of the 214 bp fragment . To gain further insight into spatial distribution of H3 and Cse4-containing nucleosomes on CEN DNA we next excised a 139 bp fragment from position −50 upstream of CDEI until the CDEII/CDEIII boundary . When cross-linked , this fragment could be co-immunoprecipitated with both H3 and Cse4 ( Figure 4B ) . This result demonstrates that H3 is present at the CDEI/II region of the centromere and/or at the preceding 50 bp of the non-centromeric DNA . Since the detection of a fragment containing CDEIII and 50 bp of DNA downstream of the CEN DNA with the LNA probe was not possible , we followed the association of histone H3 and Cse4 with CDEI/II and CDEIII elements using qPCR . Both the fragment containing CDEI/II region with upstream 50 bp and the fragment containing CDEIII region with the downstream 50 bp could be co-immunoprecipitated with HA-tagged Cse4 and histone H3 with and without crosslinking ( Figure 4C ) . Therefore histone H3 and Cse4 appeared to be inseparable when associated with the CEN DNA implying that they are likely to be a part of one and the same structure . We would like to note that since Cse4 is capable of tethering CDEII and CDEIII fragments together ( Figure 4A ) , the co-immunoprecipitation of the small CDEI/II and CDEIII fragments with H3 might be due to the small CDE-containing fragments maintaining the association with the large CDE-less fragment of the minichromosome throughout co-immunoprecipitation . No such tethering was observed when the complete 214 bp CEN DNA containing fragment was excised from the minichromosome ( Figure 1C and Figure S2 ) .
Three models of the centromeric nucleosome are proposed in the literature . In the first model the centromeric nucleosome is an octamer , where Cse4/CENP-A replaces histone H3 . While octameric nucleosomes with two copies of budding yeast Cse4 [48] , [49] or human CENP-A [41] were assembled in vitro , whether only one or both copies of H3 are replaced in vivo is not known . There is evidence from different organisms for and against either of these possibilities . In HeLa cells CENP-A released from chromatin by micrococcal nuclease digestion is still associated with histone H3 even after 2M NaCl treatment resulting in dissociation of H2A and H2B , implying heterotypic tetramers with two histones H4 , one H3 and one CENP-A [4] . In contrast , in Drosophila S2 and Kc cells when chromatin is digested with micrococcal nuclease and CENP-A/CID is immunoprecipitated , no H3 co-purifies with CENP-A [1] . It was recently reported that Drosophila CENP-A/CID forms homodimers in vivo , which are unexpectedly very salt-sensitive but could be crosslinked via cysteines in the four-helix bundle after a prolonged incubation [50] . The authors did not exclude the formation of H3-CENP-A/CID heterodimers in addition to CENP-A/CID homodimers and it remains possible that different forms of CENP-A/CID nucleosomes are simultaneously present at the regional centromeres of Drosophila and possibly other higher eukaryotes . In this study we demonstrate that a budding yeast centromeric DNA fragment of only 214 bp is associated in vivo with both H3 and Cse4 . We can exclude a homotypic octamer with two copies of Cse4 . Our experiments suggest a very intimate spatial association between the conventional histone H3 and centromeric Cse4 . This association cannot be explained if the Cse4-containing centromeric nucleosome is separated from the neighboring conventional H3 nucleosomes by spacer DNA as was proposed recently [51] but rather suggests that H3 and Cse4 co-occupy the CEN DNA fragment of only 214 bp in length . We favor the Cse4-H3 heterotypic octamer model ( Figure 5 , model 1 ) . This octamer appears to be resistant to cysteine cross-linking , which might be due to the reduced stability of the four-helix bundle similar to the Drosophila CENP-A/CID [50] . The hexamer model postulates that in budding yeast the non-histone protein Scm3 replaces H2A and H2B and the nucleosome is composed of two copies each of Scm3 , CENP-A and H4 [11] , [17] . Although it was initially proposed that the Scm3 dimer constitutes an integral part of the centromeric hexasome [11] , the recent structures of budding yeast Scm3 associated with Cse4/H4 [16] , [18] and human HJURP in complex with CENP-A/H4 [52] , [53] revealed that binding of DNA as well as the ( Cse4/H4 ) 2 heterotetramer formation are incompatible with Scm3 binding . In the experiments in vitro it was demonstrated that Scm3 association with the reconstituted ( Cse4/H4 ) 2 nucleosome-like particles depends on a DNA binding domain within Scm3 [17] . Our results are compatible with the view that Scm3 does not form a part of the centromeric nucleosome . Under our experimental conditions we were able to co-immunoprecipitate minichromosomes with Cse4 , H4 , H2A , H2B and H3 but not with Scm3 , which most likely dissociated from the centromere in yeast lysate . Finally , the hemisome model proposes a tetramer consisting of Cse4 , H4 , H2A and H2B histones [30]–[32] . According to this model , the Cse4 hemisome is positioned mostly at CDEII [20] and is expected to occupy approximately 60 bp of DNA [41] . This scenario leaves approximately 77 bp on each side of our 214 bp fragment available to accommodate the H3-containing nucleosome ( s ) . We can speculate that a hemisome with Cse4 might , for example , be incorporated into a DNA loop between the two halves of an H3-containing octamer ( Figure 5 , model 3 ) . This model might explain the tripartite organization of the budding yeast centromere that was observed in the micrococcal nuclease protection pattern [20] . Although it is technically possible that 77 bp upstream and downstream of the hypothetical centromeric hemisome are wrapped around ½ of the flanking conventional nucleosomes ( Figure 5 , model 4 ) , this model will result in tethering of the excised 214 bp fragment to the rest of the minichromosome which we did not observe ( Figure 1C and Figure S2 ) and therefore can be excluded . More exotic models can be also considered . Two recent studies compared Cse4-GFP fluorescence in vivo to independent standards and found 3 . 5–6 . 0 [54] or even 7 . 6 [55] Cse4-GFP molecules per budding yeast centromere in anaphase . Even more surprisingly , in prolonged G1 arrest Cse4-GFP fluorescence was reduced more than two-fold [55] . These observations are inconsistent with the notion of a single Cse4 nucleosome at the budding yeast centromere [37] . It was proposed that the budding yeast centromere is in fact a regional centromere with additional Cse4s associated with the flanking DNA similar to the much larger centromeres of higher eukaryotes [54] . However , we could not observe any Cse4 associated with the non-centromeric part of the 2 . 4 kb minichromosome , which is expected to assemble 10 conventional nucleosomes . Therefore no additional Cse4 nucleosomes assemble , at least at these relatively short flanking sequences . Our results are consistent with those of [20] , [56] who did not detect additional Cse4 nucleosomes in centromere-flanking regions by high-resolution mapping of yeast genome . The additional Cse4 molecules at the centromere could result from Cse4 mis-incorporation which is observed in strains overexpressing Cse4 [20] and could potentially be caused by GFP-tagging . Alternatively , additional Cse4 molecules may not be incorporated into the centromeric nucleosome but are rather associated with it via protein-protein and/or protein-DNA interactions ( Figure 5 , model 2 ) . In this scenario the centromeric nucleosome can be a Cse4-H3 heterotypic octamer to which more Cse4 molecules are bound . Intriguingly , when ( Cse4/H4 ) 2 heterotetramers were reconstituted in the presence of Scm3 into nucleosome-like particles on a 207 bp-long high affinity nucleosome positioning DNA sequence in vitro , high molecular weight complexes possibly representing additional Cse4/H4 in loose association with the Cse4/H4/DNA complex were detected [49] . Similar complexes were reported to be assembled in vitro on a 148 bp CEN3 DNA [17] . It is more than a decade now since it was proposed that H3 is replaced by the histone variant Cse4 [2] . Our results appear to contradict this well-established dogma . If Cse4 and H3 indeed co-localize to the centromeric DNA why wasn't it noticed before ? We can offer the following explanation . We have noticed that in most publications reporting ChIP experiments at the budding yeast centromere , the absolute efficiency of ChIP of the CEN DNA with H3 and Cse4 is very similar and typically in the range of 1% [11] , [35] . The claim that only Cse4 is associated with the CEN DNA is then based on an observation that non-centromeric DNA is co-immunoprecipitated with H3 at about 5 to 10-fold higher rate than CEN DNA while almost no non-CEN DNA is found associated with Cse4 ( Figure S8 ) . We suggest that if CEN DNA were generally difficult to immunoprecipitate , for example due to cross-linking of the large number of kinetochore proteins during the in vivo cross-linking , this would explain the reduced efficiency of H3 ChIP at the centromere compared to the chromosomal arms . Our results appear to contradict those of [35] . This group could co-immunoprecipitate differentially tagged versions of Cse4 from budding yeast but did not observe co-immunoprecipitation of tagged Cse4 and H3 . However , one of the tagged Cse4s was expressed from a plasmid and Cse4 overexpression was reported to result in its ectopic incorporation genome-wide into octameric nucleosomes that were not observed in the wild type strain [20] . It remains possible that even in budding yeast there is a degree of heterogeneity in the composition of the centromeric nucleosomes among different chromosomes and that either a homotypic Cse4/Cse4 octamer or a heterotypic Cse4/H3 octamer can provide the essential function . At this time we can only speculate at the function of H3 at the budding yeast point centromere . It is possible that the presence of two different nucleosomes ( or hemisomes ) , one with Cse4 and one with H3 provides structural asymmetry which might form the basis for two separate surfaces , one facing the sister centromere and another providing the attachment site for the spindle microtubule .
Generation of the minichromosome containing a 850 bp long sequence from chromosome IV encompassing CEN4 was described earlier [57] , [58] . A version without Tet operators was used to introduce BglII restriction sites using QuikChange Site-Directed Mutagenesis Kit ( Stratagene ) . A SalI digest and religation was used to remove the pUC19 sequence from the final construct prior to transformation into yeast . To introduce BglII restriction sites flanking the CEN DNA into the native chromosome IV , the region of CEN4 +/− 200 bp was cloned into the PvuII site of pOM10 ( courtesy of Anne Spang ) and BglII sites were introduced by mutagenesis . A yeast strain was transformed with a PCR product containing CEN4 DNA with BglII sites , marker , and a CEN flanking sequence . The BglII flanked CEN4 DNA was recombined into the endogenous locus and the marker cassette was removed with Cre recombinase [59] leaving 85 bp of the pOM10/loxP sequence 200 bp downstream of CDEIII ( Figure 2B ) . The whole CEN4 region was sequenced . Cse4 was tagged with HA6 , Myc6 or Myc3 at an internal XbaI site as described in [2] . All other histones were tagged at the C-terminus and the second gene was either left untagged ( H4 ) or deleted ( H2A , H2B , H3 ) . The strains are described in Table S1 . Yeast strains transformed with the minichromosome were grown overnight in synthetic medium without tryptophan at 30°C , were inoculated into fresh medium to a final OD600 of 0 . 2 , and grown until the OD600 reached 1 . 6 . For G1 arrest , yeast culture was grown from an OD600 of 0 . 05 until an OD600 of 0 . 2 and then arrested with 2 µg/ml alpha factor for 1 hour . After 1 hour , additional 1 . 5 µg/ml alpha factor was added followed by an additional hour of incubation . For G2/M arrest , 15 µg/ml nocodazole and 10 µg/ml benomyl were added to a yeast culture at an OD600 of 0 . 65 in YEPD medium , and cells were incubated for 1 . 5 hours . Spheroplasting was carried out with lyticase ( Sigma , L2524 ) as described in [60] . Spheroplasts were lysed for 30 min on ice in 2 . 5 ml of lysis buffer ( 25 mM HEPES/KOH [pH 8 . 0] , 50 mM KCl , 10 mM MgSO4 , 10 mM Na citrate , 25 mM Na sulfite , 0 . 25% TritonX-100 , 1 mM PMSF , 3 mM DTT , 1× complete EDTA-free protease inhibitors ( Roche ) and 100 µg/ml RNase A ) . The lysate was cleared by centrifugation at 10 , 000 rpm for 5 min in an Eppendorf microcentrifuge . For DNA cleavage , lysate was incubated with 1 unit/µl of BglII ( NEB ) for 2 hours with rotation at 4°C before adding NaCl to a final concentration of 300 mM to stop the digest . For strains with BglII sites on chromosome IV the crude lysate was incubated with BglII and cleared after 2 hours of digestion . Pre-cleared lysate ( 2 ml ) was incubated with 25 µg of anti-HA ( 12CA5 ) antibody and 0 . 5 ml suspension of protein A Dynabeads ( Invitrogen ) overnight . Beads were washed 3 times with 1 . 5 ml of the lysis buffer with 300 mM NaCl . Isolated DNA was eluted off the beads two times with 250 µl of 50 mM Tris [pH 8 . 0] , 10 mM EDTA and 1% SDS at 65°C . For cross-linked chromatin the DNA digest with BglII was performed for 5 min at 37°C , the digest was stopped by adding 300 mM NaCl and chromatin was cross-linked by adding 0 . 1% formaldehyde for 30 min and 125 mM glycine for 15 min on ice . The cross-linked lysate was incubated with protein A Dynabeads covalently coupled to either anti-HA ( 12CA5 ) or anti-Myc ( 9E11 ) antibody with DMP ( dimethyl pimelimidate ) according to the manufacturer's guidelines . For the sequential immunoprecipitation the chromatin was eluted off the beads as described above , diluted to 0 . 1% SDS with lysis buffer with 300 mM NaCl and immunoprecipitated with protein A Dynabeads covalently coupled to anti-HA ( 12CA5 ) . The DNA was eluted off the beads as above . All the samples were adjusted to 1% SDS final concentration , extracted twice with phenol/chloroform/isoamyl alcohol ( 25∶24∶1 ) , ethanol precipitated in the presence of 20 µg glycogen ( Roche ) and samples were dissolved in 20–40 µl TE . For the Southern blots detected with a 32P-labelled probe specific for TRP1 or CEN4 , samples were separated on a 1% agarose gel with ethidium bromide and a capillary transfer to Hybond-N+ ( GE ) was carried out under neutral conditions . Blots were scanned on Personal Molecular Imager ( Bio-Rad ) and bands quantified with QuantityOne 4 . 6 . 7 . For Southern blots detected with double-DIG labeled LNA probe ( AAAGTTGATTATAAGCATGTGAC , Exiqon ) samples were separated on a denaturing 6% TBE polyacrylamide gel followed by an electrophoretic transfer to Hybond-N+ at 80 V for 1 hr in 1× TBE in the Trans-Blot System ( Biorad ) . Hybridization with DIG labeled LNA probe was performed according to instructions of DIG High Prime DNA Labeling and Detection Starter Kit II ( Roche ) . For qPCR the samples were size fractionated on a 2% agarose gel ( Certified Low Range Ultra Agarose , Bio-Rad ) , gel excised to separate from uncut and linear minichromosome and subjected to qPCR with the primers AGTAACTTTTGCCTAAATCAC and TAGGTAGTGCTTTTTTTCCA for the 214 bp CEN4 , TAGTAACTTTTGCCTAAATC and TAATAAATAAATTATTTCATTTATGTTT for the 139 bp CDEI/II fragment , and TGTTTATGATTACCGAAACA and TTAGGTAGTGCTTTTTTTCC for the 77 bp CDEIII fragment , qPCR analysis was performed using LightCycler 480 SYBR Green I Master ( Roche ) according to the manufacturer's manual . Spheroplasting was carried out using the same procedure as for ChIP . Spheroplasts were washed in 1 M sorbitol and lysed in cold reaction buffer ( 25 mM Sodium Phosphate [pH 7 . 0] , 100 mM KCl , 2 . 5 mM MgCl2 , 0 . 25% TritonX-100 ) for 15 min on ice . Chromatin was pelleted using a low-speed centrifugation ( 4 , 000 rpm , 1 min ) and the supernatant was discarded . The chromatin pellet was then resuspended in the reaction buffer with varying concentrations of the cross-linker . DTNB ( 5 , 5′-dithiobis- ( 2-nitrobenzoic acid ) , Sigma ) was prepared as a 50 mM stock in DMSO and diluted into the reaction mixture as appropriate . Cross-linking was allowed to proceed for 1 hour on ice . The chromatin was pelleted by centrifugation and resuspended in SDS-PAGE loading dye without DTT or β-mercaptoethanol . Codon optimized sequences of yeast histone H3-Cys , N-terminally tagged with Avitag ( Avidity ) , and the histone fold domain of Cse4-Cys ( D150-end ) , N-terminally tagged with 6xHis , were cloned either together into pRSFDuet1 ( Novagen ) or separately , Cse4 in pETDuet1 and H3 in pRSFDuet1 , transformed and expressed in BL21 ( DE3 ) according to the manufacturer's instructions . Aliquots of bacterial culture were harvested and resuspended in SDS-PAGE loading buffer with and without β-mercaptoethanol . Samples were separated on a 15% SDS-PAGE and Western blots were analyzed with Streptavidin-HRP ( Pierce ) for H3-Cys and with anti-Penta-His antibody ( Qiagen ) for Cse4-Cys .
|
During cell division , replicated DNA molecules are pulled to daughter cells by microtubules , which originate at the spindle poles and attach to a multiprotein complex , the kinetochore . The kinetochore assembles at a special region of the chromosome , termed the centromere . The kinetochore is comprised of more than 50 different proteins whose precise functions are far from being fully understood . The kinetochore assembles on the foundation of a specialized centromeric nucleosome . A nucleosome is a complex of eight subunits , termed histones , which compacts the DNA by wrapping it around itself in 1 . 7 turns of a superhelix . The centromeric nucleosome is very special , and its stoichiometry and structure are a subject of intense debate . It is believed that the centromeric nucleosome is devoid of histone H3 and instead contains its variant , termed CENP-A in vertebrates or Cse4 in budding yeast . Here we report that in budding yeast both CENP-A and histone H3 localize to a small centromeric DNA fragment that , due to its size , cannot accommodate more than a single nucleosome . Our results necessitate a revision of what is known about the structure of the inner kinetochore and the role of CENP-A in its assembly .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"cell",
"biology",
"chromosome",
"biology",
"centromeres",
"biology",
"molecular",
"cell",
"biology"
] |
2012
|
Histone H3 Localizes to the Centromeric DNA in Budding Yeast
|
In addition to the DNA contributed by sperm and oocytes , embryos receive parent-specific epigenetic information that can include histone variants , histone post-translational modifications ( PTMs ) , and DNA methylation . However , a global view of how such marks are erased or retained during gamete formation and reprogrammed after fertilization is lacking . To focus on features conveyed by histones , we conducted a large-scale proteomic identification of histone variants and PTMs in sperm and mixed-stage embryo chromatin from C . elegans , a species that lacks conserved DNA methylation pathways . The fate of these histone marks was then tracked using immunostaining . Proteomic analysis found that sperm harbor ∼2 . 4 fold lower levels of histone PTMs than embryos and revealed differences in classes of PTMs between sperm and embryos . Sperm chromatin repackaging involves the incorporation of the sperm-specific histone H2A variant HTAS-1 , a widespread erasure of histone acetylation , and the retention of histone methylation at sites that mark the transcriptional history of chromatin domains during spermatogenesis . After fertilization , we show HTAS-1 and 6 histone PTM marks distinguish sperm and oocyte chromatin in the new embryo and characterize distinct paternal and maternal histone remodeling events during the oocyte-to-embryo transition . These include the exchange of histone H2A that is marked by ubiquitination , retention of HTAS-1 , removal of the H2A variant HTZ-1 , and differential reprogramming of histone PTMs . This work identifies novel and conserved features of paternal chromatin that are specified during spermatogenesis and processed in the embryo . Furthermore , our results show that different species , even those with diverged DNA packaging and imprinting strategies , use conserved histone modification and removal mechanisms to reprogram epigenetic information .
The totipotency of a new embryo depends upon the reprogramming of epigenetic information carried over from the sperm and oocyte , each of which distinctly packages its DNA . For example , canonical histones H2A , H2B , H3 , H4 and the linker histone H1 within mammalian sperm are replaced by sperm nuclear basic proteins ( SNBPs ) including histone variants , transition proteins , and protamines [1]–[4] . This repackaging results in global transcriptional repression and tight compaction of sperm chromatin [4] . After fertilization , sperm and oocyte chromatin also undergo distinct processes [5] , [6] . Oocyte chromosomes complete meiotic divisions . Then , while maternal and paternal chromatin decondense , SNBPs are removed from sperm chromatin . Thus , distinct programs package the chromatin from each gamete type during their development and after fertilization . However , still largely unknown are the epigenetic features that are inherited , how they are reprogrammed in the new embryo , and the extent to which these reprogramming steps are conserved across species . During sperm formation , histone variants are incorporated to restructure chromatin . In human , mouse , and fly , testes-specific variants of H1 , H2A , H2B , and H3 displace canonical histones prior to incorporation of transition proteins and protamines [7]–[9] . In vitro studies demonstrate that some of these variants cause nucleosome destabilization [8] , [10]–[15] . Other histone variants mark specific chromatin domains . For example in mouse spermatids , H2AL1/2 are incorporated into heterochromatic pericentric chromatin [14] , inherited by the embryo , and removed upon paternal chromatin decondensation [16] . The fate of other paternally contributed histone variants remains unclear . Chromatin restructuring during spermatogenesis is also regulated by histone post-translational modifications ( PTMs ) . In Drosophila and mammals , global acetylation of histones precedes their replacement after meiosis [7] , [17]–[21] . In particular , H4K16 acetylation ( H4K16ac ) targets histones for polyubiquitin-independent degradation within sperm-specific proteasomes called spermatoproteasomes [18] . During spermatogenesis in Drosophila , mouse , and rat , histone H2A is mono-ubiquitinated ( H2AK119ub ) [21]–[24] . However , the extent to which histone removal requires H2A ubiquitination by the ubiquitin ligase RNF8 remains controversial [23] , [25] . Many other histone PTMs have been identified in sperm or testes but roles in the histone-to-protamine transition for most of these PTMs are unknown [26] , [27] . Other roles for histone PTMs are as paternal imprints . In human sperm , marks of active transcription , such as H3K4me2 and H3K4me3 , localize to paternally-expressed imprinted loci and to promoters of genes required for development , suggesting these paternal marks play key roles in developmental programs in the embryo [28] . Conversely , a mark of repressed chromatin , H3K27me3 , is enriched at paternal promoters that are repressed in gamete formation and early embryos [28] , [29] . In mouse , H4K8ac and H4K12ac , which mark heterochromatin regions in elongating spermatids , are also carried over by sperm [30] , [31] . In C . elegans , the exclusion of H3K4me2 at inactive chromatin domains during spermatogenesis is maintained at those domains within the embryonic genome [32] . Thus histone PTMs can serve as epigenetic marks to transmit the history of specific chromatin regions to the new embryo [5] , [33] . With the array of possible histone PTMs , there is significant potential for other paternal chromatin PTMs to play directive roles in the embryo . In this work , we define the unique complement of epigenetic information contributed by histones in sperm chromatin in C . elegans . Histones are hypothesized to be the major constituent of epigenetic information in C . elegans because this organism lacks DNA methylation , which is classically associated with genomic imprinting in some species [34] . Though three highly homologous putative protamines , called SPCH-1 , SPCH-2 , and SPCH-3 ( SPCH-1 , 2 , 3 ) , have been identified in C . elegans [35] , histones and histone PTMs are retained in sperm chromatin and delivered to the new embryo . For example , previous studies have detected canonical and variant histones [32] , [35] , GFP-labeled histone H3 . 3 , or specific post-translationally modified histones ( such as H3K4me2 ) in C . elegans sperm [32] , [36] . However , a global profile of histone variants and PTMs in paternal chromatin and an understanding of their fate during the reprogramming of the embryonic genome are lacking . To address this , we employ a combination of approaches . First , we use proteomic and western analysis to define the histone profile of isolated sperm and compare it to the profile of mixed-stage embryos . Immunostaining is then used to track the temporal and spatial dynamics of how histone marks are established during spermatogenesis and oogenesis and then processed in the early embryo . This combination of approaches identifies paternal-specific epigenetic features and shows how parent-specific information can be retained or erased before and after fertilization .
To assess the extent of histone retention in sperm , we compared histone levels in sperm ( a relatively uniform population of transcriptionally inactive , differentiated haploid cells [37] ) to that of mixed-stage embryos ( transcriptionally active and developing diploid cells ) . Western blot analysis shows canonical histone H3 , H4 , and the ubiquitous H2A variant HTZ-1 are present in both sperm and embryos , while the H2A variant HTAS-1 and the putative protamines SPCH-1 , 2 , 3 are found only in sperm ( Figure 1A–E ) . Quantification revealed that lysates from sperm possess roughly 37% of canonical histones H3 and H4 compared to those of embryos that contain equivalent amounts of DNA ( Figure 1A , B , Methods ) . HTZ-1 was found at 30% in sperm lysates compared to embryo lysates ( Figure 1C , Methods ) . Thus , C . elegans retain a considerable portion of histones in sperm compared to humans ( ∼4–15% ) , mice ( ∼1% ) , and Drosophila ( ∼0% ) [21] , [28] , [38] , [39] . To further explore the histone content of chromatin from sperm and embryos we employed Multidimensional Protein Identification Technology ( MudPIT ) , a sensitive approach that can detect low-abundance chromatin proteins and PTMs in complex mixtures of proteins [40]–[43] . We purified and acid solubilized chromatin from sperm and mixed-stage embryos [44] . SDS PAGE analysis revealed histones H2A , H2B , H3 , and H4 were enriched in both chromatin samples ( Figure S1A , B ) . These samples were digested with proteases and then analyzed by MudPIT . Spectra from each sample were matched by the ProLuCID algorithm to predicted peptides and assembled into corresponding C . elegans proteins by the DTASelect program [45]–[47] . Consistent with SDS PAGE and western analysis ( Figure 1A–D , Figure S1A , B ) , histones were among the most highly represented from the 1889 spermatogenic and 2421 embryonic proteins identified as evaluated by either total spectral counts , normalized spectral abundance factor ( NSAF ) values , or Exponentially Modified Protein Abundance Index ( EMPAI ) values ( Table S1 , Table S2 ) [41] , [48]–[50] . We looked for chromatin proteins enriched in embryo or sperm chromatin samples compared to one another utilizing the spectral count feature of MudPIT , which gives an idea of the relative levels of more abundant proteins [41] , [51] , [52] . Indeed , we found that the H2A variant HTAS-1 is highly enriched in sperm chromatin with 285 spectral IDs out of 3595 total H2A spectra IDs ( which include modified and unmodified canonical H2A , HTZ-1 , and HTAS-1 spectra ) compared to embryo chromatin ( 0/4758 total H2A spectra IDs ) ( Table S3 ) . Furthermore , MudPIT analysis was sensitive: we were able to detect 36 different histone PTMs including mono-methylated ( me1 ) , di-methylated ( me2 ) , tri-methylated ( me3 ) and acetylated ( ac ) residues ( Table 1 , Tables S3–S6 , Files S1–S4 ) . Few phosphorylated residues were found , likely because phosphorylation is inherently labile and phospho-enrichment methods were not used . The coverage of the N- and C-termini of histones was also low because they are rich in lysine and arginine , the recognition sites of the proteases used . Overall , proteomic analysis found 31 histone PTM marks in embryos compared to 22 marks in sperm ( Table 1 ) , some of which would not have been identified by immunostaining or western analysis because antibodies that recognize them are not available . Our analysis also found ∼2 . 4 fold fewer modified peptides in sperm samples ( 127 PTM histone spectra IDs/7190 total histone spectra IDs = 1 . 8% ) compared to embryo samples ( 253 PTM histone spectra IDs/5998 total histone spectra IDs = 4 . 2% ) ( Tables S3 , S4 , S5 , S6 , Methods ) . A comparison of histone PTMs between sperm and embryos revealed differences in the identification of some PTM classes in each histone subtype . In general , acetylation on histone H2A , H2B , and H4 were identified more frequently and on more sites in embryos compared to sperm , whereas methylation on histone H3 was detected at similar levels in both samples ( Table 1 ) . To further investigate the presence of histone PTMs identified by proteomic analysis , we applied western analysis on sperm and mixed-stage embryo chromatin . One drawback of conducting westerns is the availability , reliability , and expense of suitable antibodies [53] . Indeed , of 27 available antibodies to modified and unmodified histones we screened ( Table S7 ) , 12 recognized bands at the expected molecular weight ( Table S7 , Figure 1F ) . A band was detected in 14/16 instances for sites where proteomic analysis had also identified the histone PTM ( Figure 1F ) ; therefore , proteomic identification is suitable for detecting both high and low abundance histone PTMs . However , in 2/4 instances western analysis detected a band for sites where proteomic analysis did not identify a modification ( Figure 1F ) . Hence , the lack of detection by either proteomics or westerns is not in itself definitive evidence that the modification is absent . An antibody that recognizes ubiquitinated histone H2A ( H2Aub ) [ ( at lysine K119 in mammals , which corresponds to C . elegans K120 ( File S5 ) ] also detected H2Aub in embryos but not sperm ( Figure 1F ) [54] . Thus , both proteomic and western analysis support that the global histone variant and PTM profile of sperm is distinct from that of mixed-stage embryos . Because proteomic and western analysis on sperm and embryo chromatin indicated HTAS-1 is enriched in sperm ( Table S3 , Figure 1D ) , we verified it is sperm-specific . First , HTAS-1 is present in mutant strains that produce only sperm and not in those that produce only oocytes , similar to the sperm-specific Major Sperm Protein ( MSP ) ( Figure S1C ) . In contrast , another H2A variant , HTZ-1 , was detected in both ( Figure S1C ) . Furthermore , immunolocalization on dissected males and hermaphrodites found that HTAS-1 is expressed only during spermatogenesis in both sexes ( Figure 2A , Figure S2A ) [35] . HTAS-1 incorporation into chromosomes starting from the late pachytene stage precedes incorporation of the putative protamines SPCH-1 , 2 , 3 ( Figure 2D , F ) . HTAS-1 is strongly enriched on condensing DNA . It is also detectable on all post-meiotic spermatids when SPCH-1 , 2 , and 3 levels are high . In contrast , HTZ-1 and histone H1 are present at all stages ( Figure 2A , F , Figure S2B ) . Thus , from proteomic , western , and cytological analysis , we conclude that HTAS-1 is a sperm-specific chromatin protein . HTAS-1 incorporates into chromosomes as they condense and transcription becomes globally reduced ( Figure 2A ) [55] , [56] . We were thus interested in determining the localization of HTAS-1 relative to HTZ-1 , which incorporates at promoters and regulates transcription in embryos [57]–[59] . Though HTZ-1 is present in all cells , several chromosomal regions exhibit relatively low levels of HTZ-1 , particularly the silenced and condensed X chromosome , which has been shown to be marked by dimethylation of lysine 9 on histone H3 ( H3K9me2 ) in male germ lines ( Figure 2B , C ) [60]–[63] . Levels of HTZ-1 staining do not substantially diminish as HTAS-1 is incorporated , suggesting that the function of HTAS-1 is not solely to displace HTZ-1 . Instead , HTAS-1 is integrated into similar chromosomal regions as HTZ-1 ( Figure 2C and D ) and both are retained on post-meiotic spermatids ( Figure 2E ) . Staining of HTZ-1 and histone H1 , which was previously found to be retained in sperm [64] , decreases in later spermatids ( Figure S2B ) ; however , both are visible on paternal chromatin in newly fertilized embryos ( Figure S3A ) , supporting the idea that they are retained in sperm and carried over at fertilization [55] , [56] . In the new embryo , differences in histone variant dynamics were found by examining chromatin transitions unique to each gamete of origin ( Figure 3A ) . Prior to fertilization , oocytes adjacent to the spermatheca mature , a process that includes nuclear envelope breakdown , nuclear migration , and cortical rearrangements [65] . After sperm entry , the oocyte pronucleus completes meiotic divisions to produce a haploid complement of maternal DNA and two polar bodies [6] , [66] . Both the oocyte and sperm chromatin then decondense , become ensheathed within pronuclear envelopes , and undergo DNA replication [67] . The oocyte and sperm pronuclei migrate , transition to metaphase and undergo mitosis . Embryos in the 1- to 2-cell stage rely on maternal stores of proteins and mRNAs because embryonic transcription does not initiate until the 4-cell stage [68] , [69] . Embryos complete transfer from maternal to embryonic control of development around the 40-cell stage . Interestingly , histone H2A variants have different fates in the embryo . HTZ-1 levels are high on both paternal and maternal chromatin just after fertilization ( Figures 3B , Figure S3A ) , but levels drop on both after oocyte meiotic divisions ( Figure 3C , D ) . Thus HTZ-1 is either removed by maternally loaded factors or displaced during DNA replication . HTZ-1 levels remain low in 1- and 2-cell stage embryos ( Figures 3C–E and S3A ) then rise in the 4-cell stage on all nuclei ( Figure 3F ) . This includes the transcriptionally silenced germline precursor P2 cell [70] , [71] , suggesting embryonic transcription is not necessary for HTZ-1 incorporation . In contrast , HTAS-1 is retained on paternal chromatin and absent on maternal chromatin throughout the 1-cell stage ( Figure 3B–D , Figure S3B ) . HTAS-1 marks only the paternal half of the chromatin mass at metaphase , revealing the striking compartmentalization of chromosomes at this stage ( Figure 3D ) . After division , HTAS-1 is detectable on chromosomes and levels subsequently decline with each cell division , becoming undetectable after the 4-cell stage ( Figure 3E , F ) . Thus , HTZ-1 is removed after fertilization while HTAS-1 is retained . These results show that histone H2A variants are differentially processed in the early embryo and indicate distinct mechanisms recognize and either remove or retain each variant . The dynamics of HTZ-1 and HTAS-1 before and after fertilization suggests histone subunit exchange during these critical transitions . A clue about the mechanism of this process was revealed by western blot analysis , which detected H2Aub in chromatin from embryos but not sperm ( Figure 1F ) [54] . To investigate this difference further , we tested four antibodies via immunostaining in C . elegans that were previously shown to specifically recognize H2Aub in mammals ( Table S7 ) [54] , [72] , [73] . Of these , one antibody , the monoclonal E6C5 antibody detected H2Aub in both male and hermaphrodite germ lines but not in later stages of spermatogenesis ( Figure S4A ) . During the pachytene stage , H2Aub is found on all chromosomes ( Figure S4A ) , which is in contrast to the heterochromatic sex body localization observed at that stage in mice [22] , [74] . As H2Aub levels on chromosomes decrease in early pachytene , the E6C5 antibody also detected foci not associated with DNA , a pattern we subsequently refer to as “off-chromatin foci” ( Figure S4A ) that are distinct from germline P-granules [75] . The H2Aub off-chromatin foci do share partial overlap with poly-ubiquitin conjugates linked through either lysine 48 ( Ub-K48 ) , which target proteins to the proteasome , or lysine 63 ( Ub-K63 ) , which can act as a signal for autophagy ( Figure S4B–D ) [76] . The overlapping localization of Ub-K48 and the proteasome together with the presence of Ub-K63 foci suggest these pathways are active during late spermatogenesis , when chromatin is undergoing extensive remodeling . The lack of H2Aub on spermatid chromatin is consistent with western blot analysis and , together with the potential overlap of off-chromatin foci with ubiquitin conjugates , suggests H2A may be removed during late spermatogenesis . Because spermatids lack the H2Aub mark before fertilization , we were surprised to observe high levels of H2Aub on paternal chromatin and in adjacent off-chromatin foci in newly fertilized embryos using either the E6C5 monoclonal or the polyclonal ABE569 and #308 anti-H2Aub antibodies ( Figure S5 , Figure 4A , B ) . Similar structures around paternal DNA were previously found to harbor sperm membranous organelles ( MOs ) that become poly-ubiquitinated on lysines 63 and 48 to target them for destruction by autophagy and possibly the proteasome after fertilization ( Figure 4C–F ) [76] , [77] . H2Aub staining partially overlaps with Ub-K48 and Ub-K63 staining on paternal DNA and in some off-chromatin foci as oocyte chromosomes complete meiosis ( Figure 4C–E ) and is distinct from the localization of MOs both before and after fertilization ( Figure 4F ) . The partially overlapping regions of staining with Ub-K48 and Ub-K63 suggest that paternally-delivered H2A may , like MOs , be targeted for destruction after fertilization [76] . Maturing and meiotically-dividing oocyte chromosomes also exhibit low H2Aub levels , where off-chromatin foci were also visible ( Figure S5 ) . After oocyte meiosis , levels of chromatin-associated H2Aub fall ( Figure S5 ) . By the 16-cell stage , H2Aub chromosomal staining rebounds ( Figure S5 ) but Ub-K48 and Ub-K63 staining are absent [77] . Together , these results indicate that H2A may be removed and targeted for destruction by ubiquitination as paternal and maternal chromosomes decondense , suggesting that ubiquitination of H2A plays a role in H2A exchange in the newly fertilized embryo . The differential acetylation profile of sperm chromatin was next examined by testing commercially available antibodies to find those that recognized specific chromatin patterns in C . elegans ( Table S7 ) . We found that 8 acetylated histone marks [H4K16ac , H2BK12ac , H3K23ac , H3K27ac , H4K5ac , H4K8ac , H4K12ac , and H2A acetylated on sites K5 , K9 , K13 , K15 ( hereafter referred to as H2Apan-ac ) ] dramatically decrease during sperm chromosome condensation ( Figure S6 , Table S7 , File S5 ) . Six of the sites ( H3K27ac , H4K5ac , H4K12ac , H4K16ac , H2Apan-ac , and H2BK12ac ) showed no staining on nuclei undergoing the second meiotic division ( Figure S6A–F ) , while 2 sites - H4K8ac and H3K23ac - were detectable at low levels on metaphase 2 nuclei ( Figure S6G , H ) . Thus , overall acetylation levels are decreased on post-meiotic sperm , consistent with proteomic results ( Table 1 ) . Immediately after fertilization , the lack of staining for 5 sites ( H4K16ac , H3K27ac , H4K5ac , H4K12ac , H2Apan-ac ) on paternally-derived chromatin distinguishes it from the maternal chromatin , where high levels of staining were observed ( Figure 5A–E , Figure S3C–E ) . In the early embryo , acetylation dynamics reveal differences in the fate of different PTMs on each pronucleus and mechanisms of histone acetylation/deacetylation . In particular , H4K16ac staining is retained on the maternal chromosomes but excluded from the paternal chromosomes in 1-cell embryos ( Figure 5A , Figure S3C ) . Levels of H4K16ac remain low in 1- to 2-cell stage embryos but increase dramatically in 4-cell stage embryos in all cells , including the P2 cell , suggesting transcription is not necessary for H4K16 acetylation ( Figure S7A ) . Overall , the sperm-specific loss of H4K16ac results in a maternal-specific H4K16 acetylation mark in 1-cell embryos . Acetylation dynamics of H3K27 suggest that some histone acetylases are maternally loaded . H3K27ac levels are high on maternal chromatin after fertilization , but absent on paternal chromatin ( Figure 5B ) . However , H3K27ac levels on paternal chromatin gradually rise until staining is strong in 2-cell embryos . This gradual rise in 1-cell embryos suggests maternal loading of the acetylase for this site , resulting in an “equalization” of H3K27ac levels between paternal and maternal pronuclei in the 1-cell embryo . A general equalization of acetylation levels can occur via histone deposition during DNA replication . Histone H4 is marked by di-acetylation at K5 and K12 as it incorporates during DNA replication in Tetrahymena , Drosophila , and human cells [78] . Indeed in C . elegans , we also found H4K5ac and H4K12ac are present on the maternal pronucleus but absent on paternal chromatin after fertilization; however , levels of each are uniform on both pronuclei after DNA replication ( Figure 5C , D , Figure S3D ) . The timing of this equalization suggests these marks may be added to the paternal chromatin via histone deposition during DNA replication . An antibody that recognizes multi-acetylated H2A ( H2Apan-ac ) on K5 , K9 , K13 and K15 ( the only antibody we tested against H2A that produced positive immunostaining results , see Table S1 and File S5 ) exhibited similar dynamics ( Figure 5E , Figure S3E ) . H2AK5 acetylation is influenced in the germline by XND-1 , a protein involved in regulating meiotic crossovers in C . elegans [79]; however , little is known about acetylation of on these residues in any organism . Deacetylation in oocytes before fertilization can also contribute to equalizing acetylation levels . Oocytes closest to the spermatheca , which are maturing [65] , exhibit a dramatic drop in H2BK12ac levels on chromosomes during diakinesis ( Figure 5F , Table S7 , File S5 ) , suggesting that H2BK12 deacetylation could be triggered by oocyte maturation signals . However , we also observe deacetylation of H2BK12 also occurs at diakinesis in spermatocytes ( Figure S6F ) . H2BK12ac levels remain low on both sperm and oocyte chromatin until the 2–4 cell stage , when H2BK12ac then decorates all nuclei including the P2 cell ( Figure S7B ) , suggesting that embryonic transcription is not directly correlated with its presence [71] . Epigenetic acetylation marks present at different levels in both pronuclei at sperm entry can become equalized . H4K8ac and H3K23ac exhibit high levels of staining in oocytes and low levels in late stages of sperm meiosis ( Figure 5G , H , Figure S6G , H ) . Consistent with this , post-fertilization H4K8ac and H3K23ac staining is much weaker on paternal chromatin compared to maternal chromatin ( Figure 5G , H ) . H4K8ac levels on both pronuclei increase dramatically during oocyte meiotic divisions , while H3K23ac levels remain modest in 1-cell embryos but become strong in 2-cell embryos . The increase in acetylation of both sites indicates that the acetylases that recognize each are likely to be maternally loaded . Overall , we have identified features of paternal and maternal chromatin histone acetylation dynamics before and after fertilization in C . elegans that are both shared and distinct from that of other organisms . These observations show that 1 ) paternal chromatin undergoes erasure of many acetylation sites during sperm formation , 2 ) sperm and oocyte chromosomes carry parent-specific markings into the new embryo , 3 ) paternal and maternal pronuclei can undergo distinct histone remodeling events both before and after fertilization , 4 ) the processing of epigenetic information is dependent on maternal loading of histone modification enzymes . Parent-specific imprinting in C . elegans can be specified by histone methylation that marks the transcriptional activity of chromatin domains by two different mechanisms [32] , [61] , [80] . Meiotic silencing of unpaired chromatin ( MSUC ) is implemented in part by excluding H3K4me , a mark of germline transcriptional activity , from the single X chromosome in XO males [61] . The silenced male X chromosome instead harbors H3K9 methylation marks [61]–[63] , [81] . After fertilization , only the paternal X remains resistant to H3K4 methylation while autosomes and the maternal X exhibit this mark . In contrast , H3K36me2 is a different mark added by the MES-4 methyltransferase only on autosomes in hermaphrodites and males to transmit a memory of gene expression patterns from the parental germ line to germline precursors in offspring [80] , [82] . Thus H3K36me2 is absent on both the paternal and maternal X chromosomes before and after fertilization . Our proteomic analysis identified other histone methylation marks on sperm chromatin ( Table 1 ) . We used immunostaining to investigate the retention of three methylation marks ( H3K36me1 , H4K20me1 , and H3K79me2 ) with available commercial antibodies ( Table S7 ) . We found H3K36me1 is likely specified by mechanisms that implement MSUC and not by MES-4 . During spermatogenesis in males , H3K36me1 is absent from the single X chromosome but present on autosomes in nuclei at all stages , though low on post-meiotic sperm chromatin ( Figure S8A ) . During oogenesis in hermaphrodites , all chromosomes , including the maternal X , stain with H3K36me1 at the diplotene stage ( Figure 6A ) . In early embryos , only the paternal pronuclei lack one region of H3K36me1 staining , which , because it lacked staining before fertilization , is presumably the paternal X chromosome . This pattern is distinct from that of H3K36me2 [80] , [82] and instead mirrors the staining pattern observed after fertilization for H3K4me [61] , [62] . Thus , the dynamics of H3K36me1 we observe more closely correlates with the imprinting mechanisms of MSUC and not MES-4 [61] , [80] . H4K20me1 exhibits a distinctive localization pattern when carried over to the embryo . In C . elegans , though H4K20me1 is enriched on dosage-compensated X chromosomes in somatic cells , it is reduced on meiotically-silenced X chromosomes during the pachytene stage [83] . We observe that by diplotene/diakinesis , H4K20me1 becomes focused in one or two bright chromatin-associated spots ( Figure 6B ) . In male sperm-producing germ lines , levels of H4K20me1 are also low on the single X chromosome during pachytene ( Figure S8B ) then fall; however , one or two bright chromatin-associated foci appear . Costaining with an antibody that recognizes H3K9me2 , a marker for the X chromosome [61] , [62] , [81] , show that these foci are not on X ( Figure S8B ) . After fertilization , the paternal and maternal pronuclei are similarly decorated with small H4K20me1 foci . Levels then rise to cover all chromosomes in both pronuclei ( Figure 6B ) , suggesting the H4K20 methylase is maternally loaded . These results show that H4K20me1 is retained on sperm chromatin in a novel localization pattern and demonstrate that paternal and maternal nuclei are equivalent in H4K20me1 status in the new embryo . H3K79me2 is also carried over to the new embryo from both sperm and oocytes ( Figure S8C , Figure 6C ) . H3K79me2 is found at high levels on oocyte chromosomes before and after fertilization . However , though H3K79me2 status is equivalent in paternal and maternal pronuclei after sperm entry; levels drop in 1-cell embryos ( Figure 6C ) , suggesting a H3K79 demethylase is maternally loaded to reprogram paternally delivered marks in the new embryo . A similar demethylation of H3K79 also occurs in mouse [84] . H3K79me2 levels rise in 4-cell embryos , including the transcriptionally inactive P2 cell ( Figure S7C ) , indicating its presence is not dependent on transcription [71] . Not all histone methylation marks are retained in sperm chromatin . H3K79me3 was not identified in sperm chromatin by proteomic analysis . In the male sperm-producing germ lines , H3K79me3 levels fall before meiotic divisions ( Figure S8D ) . In contrast , H3K79me3 levels are high in oocyte-producing germ lines through diakinesis . Subsequently , lack of H3K79me3 distinguishes the paternal pronucleus from the maternal pronucleus after fertilization ( Figure 6D ) . Levels of H3K79me3 become low in both maternal and paternal pronuclei then rise after the 16-cell stage . This suggests that the enzyme responsible for H3K79me3 is likely not maternally supplied . The removal of maternal H3K79me3 in early embryos was similarly observed in mouse early embryos [84] .
In this study , we define the epigenetic profile of C . elegans sperm chromatin . This distinctive profile includes ( 1 ) sperm-specific incorporation of the histone H2A variant HTAS-1 , ( 2 ) a depletion of histone acetylation marks , and ( 3 ) the retention of specific histone marks , like H3K36me1 , that convey the transcriptional history of chromatin domains . Overall , we find that sperm generally harbor less histone PTMs on fewer sites compared to embryos and tracked 13 sites via immunostaining before and after fertilization to define distinctive features of paternal and maternal chromatin ( Tables 1 , 2 ) . An important product of these studies is a set of temporal markers of the dramatic chromatin restructuring and reprogramming events that occur during sperm formation and in early embryo development ( Table 2 , Figure 7 ) . The sperm epigenetic profile is established by four partially overlapping stages during sperm formation in C . elegans ( Figure 7A , B ) that are then processed in the embryo . The first involves H2A ubiquitination . H2A ubiquitination occurs in Drosophila , mouse , and rat spermatogenesis [21] , [23] , [24]; however , the role of this modification remains unclear because reports concerning how loss of the ubiquitin conjugating enzyme , RNF8 , affects histone replacement during spermiogenesis vary in mouse [23] , [25] . A recent report has instead shown histones may be targeted to sperm-specific proteasomes via H4K16 acetylation [18] . In our studies , chromatin-associated H4K16ac and H2Aub are removed as chromosomes condense for meiotic divisions . Additionally , we found evidence of H2Aub off-chromatin foci sharing partial overlap with Ub-K48 conjugates , which target proteins to proteasomes , and Ub-K63 conjugates , which are linked to the DNA damage response and autophagy ( Figure S4 ) , though only one ( E6C5 ) of four antibodies we tested showed a clear signal in the male germ line [77] . However , the dynamics of ubiquitin conjugates and the proteasome localization defined here for the first time in the C . elegans germ line suggests that Ub-K48 and Ub-K63 pathways are active during late spermatogenesis when chromatin is remodeled ( Figure S4 ) . Furthermore , after fertilization , the appearance of H2Aub detected by three antibodies ( E6C5 , ABE569 , and #308 ) as paternal chromatin decondenses suggests paternally contributed histones are targeted for destruction , just as other sperm-specific proteins , like membranous organelles , are targeted for autophagy [76] , [85] . Thus , ubiquitination may play a general role in the immediate clearance of sperm proteins in the embryo after fertilization . Incorporation of the sperm-specific H2A variant HTAS-1 is the next phase of chromatin remodeling ( Figure 7A , B ) . Incorporation of histone variants before completion of meiosis occurs in other organisms ( Figure 7C ) . For example , TH2B facilitates protamine incorporation in mouse and rat [9] , [86] . Other histone variants , such as H2AL1/2 in mouse , are retained at pericentric regions , then delivered to the embryo and removed upon sperm decondensation [14] , [16] . In contrast , HTAS-1 is carried over and not erased in the 1-cell embryo . This raises the intriguing possibility that HTAS-1 may be specifically retained at specific genomic sites for function in the embryo . In contrast , HTZ-1 is removed from paternal and maternal chromatin rapidly in the new embryo . Though it is not clear if HTZ-1 removal coincides with DNA replication , studies in mice show that maternally-contributed H2A . Z removal is DNA replication-independent [87] . Thus , it remains to be determined which structural differences between H2A variants may alter the stability of nucleosomes during DNA replication or are recognized by histone chaperones or chromatin modifiers [88] , [89] . The third stage of chromatin remodeling during sperm formation is the bulk loss of many histone PTMs that results in the lack of histone acetylation marks on paternal chromatin in the embryo ( Figure 5 , Table 2 ) . Levels of all histone PTMs fall as sperm chromosomes condense and divide . However , we found that removal prior to the second meiotic division correlates with erasure , indicating this cell cycle transition may be a critical juncture that coordinates histone PTM removal ( Figure 7A , Table 2 ) . After fertilization , many maternal and paternal chromatin histone PTM marks subsequently become equalized . Work from this study and others show this occurs via distinct mechanisms [5] , [33] , [84] . A notable mark that is not equalized is H4K16ac , which marks oocyte chromatin but remains absent from paternal chromatin . However , by the 4-cell stage , levels of all histone PTMs progressively rebound ( Figure 7A ) . While embryonic transcription may be a factor , several marks are established on chromatin of the transcriptionally-silenced germline precursor cell ( Figure S7 ) . Thus the regulatory factors that temporally regulate histone modifications during the oocyte-to-embryo transition remain to be elucidated . Histone PTMs that persist into the second meiotic division during spermatogenesis are retained ( Figure 7A , Table 2 ) . For example H4K20me1 forms small chromatin-associated foci detectable before and after fertilization . In later embryos , H4K20me1 levels rebound on all chromosomes then later become enriched on X chromosomes to function in dosage compensation [83] , [90] . Further investigation will reveal whether H4K20me1 dynamics play roles in sperm or early embryos , as H4K20 methylation has a broad range of functions , including DNA damage response , DNA replication , mitotic condensation , and transcription [91] . Overall , we have found that many of the PTM marks in C . elegans share similar fates as those in mouse ( Table S8 ) . Importantly , our proteomic analysis has also identified both histone methylation and acetylation marks retained in sperm chromatin that also have the potential to pass on specific paternal epigenetic information to the embryo ( Table 1 , Table 2 , Table S8 ) . The last stage of restructuring is the incorporation of sperm nuclear basic proteins ( SNBPs ) , which exhibits differences compared with mammals ( Figure 7B , C ) . Most strikingly , in C . elegans , incorporation of HTAS-1 and the putative protamines SPCH-1 , 2 , 3 overlaps with meiotic divisions ( Figure 2 , Figure 7A , B ) [35] , [55] , in contrast to mammalian transition proteins and protamines that are incorporated after meiosis ( Figure 7C ) [92] . Furthermore , C . elegans sperm retain more histones compared to mammals . This may be due to the rapid progression of spermatogenesis as well as a lack of identifiable C . elegans transition proteins . Future studies , including assessing spch gene loss-of-function , are necessary to support the possibility that SPCH-1 , 2 , and 3 function as C . elegans protamines and to determine whether each plays a distinct role in sperm formation . The overlap of events also results in a lack of a post-meiotic burst of transcription in C . elegans spermatogenesis , which prepares gametes for chromosome restructuring and spermiogenesis in mammals [4] ( Figure 7B , C ) . In C . elegans , transcription is globally repressed before meiotic divisions and not reactivated after meiosis [55] , [56] , [93] . Further , C . elegans does not exhibit post-meiotic histone hyperacetylation , which plays roles in transcriptional activation and histone displacement in mammals and Drosophila [18]–[20] , [94] . In C . elegans , histone acetylation levels remain high on nuclei until meiotic divisions , at which point levels rapidly drop ( Figure 7B and Figure S6 ) [56] . The overlap , omission , or abbreviation of these events likely contributes to the rapid progression of sperm formation observed in C . elegans ( ∼24 hours ) compared to mouse ( ∼30 days ) or humans ( ∼60 days ) [56] , [95]–[97] . Historically , identifying epigenetic marks in sperm and embryos solely using cytology or western analysis in any organism has been challenging . Antibody availability is limited and specificity in different applications is variable [53] . For example , in this work , we tested four distinct antibodies used in previous studies to detect H2A ubiquitination: the monoclonal E6C5 and the highly purified polyclonal ABE569 , #87 , and #308 antibodies [54] , [72] . Antibody #87 exhibited specificity in western analysis ( Figure 1F ) , while E6C5 , ABE569 and #308 showed distinct staining patterns in immunostaining experiments ( Figure 4 , Figure S4 , Figure S5 ) . While one possibility is that they may recognize off-target proteins or unmodified H2A in immunostaining experiments , it is also possible that they are specific for distinct epitopes that are accessible under specific conditions ( i . e . denatured or partially folded protein conformations ) because each was raised in individual animals against antigens with multiple possible epitopes . Another issue is whether lack of staining indicates the absence of a protein or the inaccessibility of antibodies to the tightly compacted paternal chromatin . Indeed , we observed the interior of the C . elegans spermatid nucleus is resistant to antibody staining regardless of the antibody or fixation condition used . Thus , the sole use of antibodies to identify histone PTMs should be used with caution [98] . Overall our study shows that despite the limitations of proteomics , western analysis and cytology , their combined use was an effective strategy to mine for modification sites and then define the spatial and temporal dynamics of their addition and removal . Further high-resolution studies of gametes and early embryos are necessary to distinguish specific genes associated with the epigenetic markers identified here . However , our studies provide key landmarks for understanding the dynamics of epigenetic erasure and reprogramming required for both sperm cell and early embryonic development .
C . elegans strains were maintained using standard conditions [99] . Strains used in this study are CB1489 him-8 ( e1489 ) , DR466 him-5 ( e1490 ) , TY0119 fem-1 ( hc17ts ) and JK0816 fem-3 ( q20ts ) . Strains were cultured at 20°C , except for TY0119 fem-1 ( hc17ts ) and JK0816 fem-3 ( q20ts ) , which were maintained at 15°C . Large scale culturing of C . elegans hermaphrodites was conducted as in [35] except that fem-3 ( q20ts ) animals were cultured at 15°C . Synchronous cultures of fem-3 ( q20ts ) hermaphrodites were treated with a 1 . 2% sodium hypochlorite , 0 . 5 N NaOH solution with vigorous vortexing to release embryos . After washing in M9 buffer to remove hypochlorite , the embryos were either quick-frozen in liquid nitrogen and stored at −80°C degrees for chromatin isolation or were hatched overnight in M9 solution . The resulting L1 larvae were plated onto egg plates and cultured for 4 days at 25°C . Sperm was isolated as in [35] and frozen at −80°C degrees . Approximately 600 µL of mixed-stage N2 embryos and 600 µL of fem-3 ( q20ts ) sperm were collected and lysed by 6 rounds of sonication ( 3 seconds at 10% amplitude ) and rest ( 1 minute on ice ) . Genomic DNA was isolated using phenol/chloroform ( Koelle lab ) from 200 µL of N2 embryo lysate or 200 µL of fem-3 ( q20ts ) sperm lysate , resuspended in 100 µL of dH2O , and quantified with the Qubit dsDNA BR Assay Kit on a Qubit 2 . 0 Fluorometer ( Life Technologies ) . Titrations of sperm and embryo lysates representing equivalent amounts of DNA were subjected to 4–20% SDS PAGE ( Bio-Rad ) or 14% SDS PAGE ( Fisher Scientific ) and western analysis . Primary antibodies ( and the dilutions ) used include: anti-HTAS-1 ( at 1∶1000 , polyclonal rabbit antibodies 3656 and 3657 generated against the last 15 amino acids of the C-terminal , LPKKKAKEDDKENNS ) , anti-SPCH ( at 1∶500 , rabbit 2391 [35] ) , anti-HTZ-1 ( a polyclonal rabbit antibody from the lab of Dr . Gyorgyi Csankovszki used at 5 ug/mL for the westerns ) and anti-histone PTMs shown in Table S7 . HRP-conjugated donkey anti-rabbit ( abcam ab6802 ) was used at a 1∶1000 dilution . HRP signal was detected using SuperSignal Chemiluminescent Substrate ( Pierce ) . Western blots were analyzed using a Kodak Imager . The Gel Analyzer tool from ImageJ was used to quantify bands from western blot analysis . For chromatin preparations , a minimum of 500 µL of frozen sperm or embryos was used per isolation . Cells were lysed as described above . Chromatin was isolated as in [35] except centrifugation through a sucrose gradient was conducted at 100 , 000×g for 1 hour at 4°C and 1 mM sodium orthovanadate ( Na3VO4 ) , 10 mM sodium butyrate ( Na ( C3H7COO ) ) , and 0 . 02% protease inhibitors ( Calbiochem ) were added to all buffers prior to use . Nuclear basic proteins from sperm and embryo chromatin were isolated by resuspending samples in 0 . 4 N H2SO4 acid for 1 hour at 4°C [44] . Samples were subjected to centrifugation at 16 , 000×g for 10 minutes at 4°C . The resulting supernatant was resuspended in 500 µl of Buffer A [35] . TCA precipitated sperm and embryo acid-solubilized chromatin proteins representing one-fifth of that analyzed by MudPIT were resuspended in 50 µL of 1×LSM . 5 µL were subjected to SDS PAGE and western blotting as described above . Primary antibodies and the dilutions used for westerns are described above ( HTAS-1 , HTZ-1 , SPCH ) or shown in Table S7 . Mouse anti-MSP ( from the lab of Dr . David Greenstein ) was used at 1∶1000 dilution [105] . Because sperm DNA is tightly compacted , we tested different fixation methods to achieve consistent and sensitive detection of proteins associated with spermatid chromatin . A methanol/acetone fixation ( as detailed below ) was the most effective for detecting integral chromatin proteins , such as histones , in comparison to methanol or ethanol/paraformaldehyde methods [35] , [56] , [106] . Using this method we detect staining on early spermatid chromatin for histone H1 ( Figure S2 ) , a protein previously shown to be present in sperm chromatin [64] . For most proteins tested , staining levels on sperm chromatin decrease gradually after completion of meiosis , suggesting that chromatin becomes progressively more compacted [56] . Each staining experiment was conducted a minimum of four times with at least three slides ( each with 30–50 animals ) for every experiment . For methanol/acetone fixation , 30–60 adult him-8 ( e1489 ) males or hermaphrodites were dissected into sperm salts ( 50 mM Pipes , pH 7 , 25 mM KCl , 1 mM MgSO4 , 45 mM NaCl , and 2 mM CaCl2 ) on a slide . The slide was then freeze-cracked . For antibodies that recognize histones , slides were placed into 100% methanol for 10 minutes , then 100% acetone for 5 minutes and briefly air-dried . Slides were then washed 3 times with PBST ( 1× PBS , 0 . 5% Triton X-100 , and 1 mM EDTA , pH 8 ) for 10 minutes each wash . For antibodies that recognize ubiquitin , MOs , and the proteasome , a methanol fixation procedure was also used , in which slides were placed in methanol for 30 minutes then processed as described . Primary and secondary antibodies were incubated overnight . Slides were washed with PBST as above then stained with DAPI ( 1 µg/ml ) and mounted with Vectashield . Confocal images were obtained using a Zeiss LSM710 confocal microscope . Primary antibodies against commercially available post-translationally modified histones used in this study are listed in Table S7 .
|
Successful reproduction depends upon the receipt and processing of distinct chromatin packages from sperm and oocytes . This includes not only a unique complement of DNA , but information in the form of proteins , such as histones , that differentially package the DNA in each cell type . For example , histone variants and post-translationally modified histones can carry epigenetic information across generations . Such information is then reprogrammed in the new embryo to ensure proper development . However , it is unclear how many of these marks are established during gamete formation and reprogrammed after fertilization . We define a signature histone variant and post-translational modification profile of sperm chromatin from C . elegans . This profile is established during sperm formation in part by exchanging canonical histones with sperm-specific histone proteins . Histone variants and modifications passed from sperm and oocytes are differentially removed or retained , suggesting that the embryo can reprogram epigenetic information from each parent distinctly . These C . elegans studies reveal that both conserved and novel histone modification and exchange mechanisms are used across diverse species to establish and reprogram epigenetic information .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"biochemistry",
"developmental",
"biology",
"model",
"organisms",
"cell",
"biology",
"genetics",
"biology",
"and",
"life",
"sciences",
"proteomics",
"molecular",
"cell",
"biology",
"research",
"and",
"analysis",
"methods"
] |
2014
|
The Specification and Global Reprogramming of Histone Epigenetic Marks during Gamete Formation and Early Embryo Development in C. elegans
|
Flexible or innovative behavior is advantageous , especially when animals are exposed to frequent and unpredictable environmental perturbations . Improved cognitive abilities can help animals to respond quickly and adequately to environmental dynamics , and therefore changing environments may select for higher cognitive abilities . Increased cognitive abilities can be attained , for instance , if environmental change during ontogeny triggers plastic adaptive responses improving the learning capacity of exposed individuals . We tested the learning abilities of fishes in response to experimental variation of environmental quality during ontogeny . Individuals of the cichlid fish Simochromis pleurospilus that experienced a change in food ration early in life outperformed fish kept on constant rations in a learning task later in life—irrespective of the direction of the implemented change and the mean rations received . This difference in learning abilities between individuals remained constant between juvenile and adult stages of the same fish tested 1 y apart . Neither environmental enrichment nor training through repeated neural stimulation can explain our findings , as the sensory environment was kept constant and resource availability was changed only once . Instead , our results indicate a pathway by which a single change in resource availability early in life permanently enhances the learning abilities of animals . Early perturbations of environmental quality may signal the developing individual that it lives in a changing world , requiring increased cognitive abilities to construct adequate behavioral responses .
The ability of adapting to changes in the environment is an important driving force of evolution , as recognized already by Darwin in his famous quote: “It is not the strongest of the species that survives…it is the one that is the most adaptable to change” [1] . Animals may adapt by altering their behavior , physiology , or morphology . The construction of behavioral responses is thought to be the fastest and most flexible way of adapting to new situations . Animals often have to deal with new situations for which they must devise novel or flexible solutions [2] . Field observations and laboratory studies showed that the advantages of novel or altered behaviors increase with the complexity of the environment ( reviewed in [3] , [4] ) . This suggests that frequent and unpredictable environmental changes may select for increased cognitive abilities allowing animals to meet these challenges by constructing adequate behavioral responses . In mockingbirds , for example , the complexity of songs is assumed to reflect their cognitive abilities , and species inhabiting areas with a low predictability of climatic patterns show more elaborate song displays than species in stable environments [5] . On the level of the individual , environmental instability can be encountered by plastic trajectories of the development of cognitive abilities . Environmental fluctuations early in life are known to enhance the behavioral flexibility of animals with regard to predator avoidance strategies [6] , [7] , feeding performance [7] , and social behavior [6] , [8] . A possible explanation for these behavioral effects is that variable environments evoke repeated neural stimulations resulting in faster and better learning [7] . Several studies showed that neural stimulation over longer periods by exposing animals to enriched environments ( e . g . , [9] , [10] ) can enhance brain development [5] , [11] , for example through an increased synaptic density [12] , and can lead to improved learning abilities and memory capacity [12] . A food manipulation experiment indicated that a single change of diet can constrain neural development if later cognitive abilities are traded against the benefits of a compensatory growth response [7] , [13] . On the contrary , an environmental change early in life should be expected to favor enhanced cognitive development , if this early perturbation signals the developing individual that it lives in a more variable environment . In response to this signal animals should develop increased cognitive abilities , which help them to construct adequate behavioral responses to the environmental challenges . An experimental evaluation of this hypothesis has been hitherto lacking . Individuals of the African cichlid Simochromis pleurospilus live in a stable environment , but parts of the population experience a habitat shift around maturation [14] . If increased cognitive performance confers a fitness advantage when shifting habitats , we should expect that improved cognitive abilities can be triggered in S . pleurospilus by experimentally varying their juvenile environmental quality . We tested this prediction by investigating how the performance in a learning task was influenced by different juvenile feeding regimes in S . pleurospilus . Fish were fed either on a stable high or a stable low food ration , or rations were switched from low to high or vice versa . We trained the fish to associate a visual cue with food and tested how often they selected the positive stimulus . We tested their cognitive performance twice , at the end of the juvenile period and 1 y later when the fish were adults . We adhere to the broad definition of “cognition” as comprising “all mechanisms that invertebrates and vertebrates have for taking in information through the senses , retaining it , and using it to adjust behavior to local conditions” [15] .
The tests of learning abilities yielded similar results in juveniles ( J ) and adults ( A ) . Neither the amount of food received before the switch ( J: p = 0 . 62 , A: p = 0 . 53 ) nor after the switch ( J: p = 0 . 21 , A: p = 0 . 12 ) influenced the number of correct choices significantly . The interaction between early and late food treatment was significant ( J: p = 0 . 029 , A: p = 0 . 005 , Table 1 ) however , demonstrating that fish that had experienced a switch in feeding regime outperformed those fed constant rations . This effect is independent of the direction of the diet change ( high-to-low or low-to-high; Figure 1 ) . Alternatively , learning ability might be affected by the average amount of food during the juvenile period . In that case we should expect the learning performance to increase linearly with food ration or , if an optimal food level exists , to follow a dome-shaped relationship . To test for this possibility , we determined the average amount of food each fish consumed relative to its own body mass . However , the number of correct choices was not related to mean relative food ration , neither with a linear ( GLzM: J: p = 0 . 59 , A: p = 0 . 86 ) nor with a quadratic predictor ( GzLM: J: p = 0 . 26 , A: p = 0 . 90 ) . During the test of juvenile cognitive abilities the animals of different treatment groups differed in body size ( overall difference: ANOVA , df = 3 , F = 103 . 88 , p<0 . 001; differences between individual groups: Tukey's hsd test , all p<0 . 001 , Figure 2A ) and in the latency times to approach the stimulus ( ANOVA , df = 3 , F = 7 . 81 , p<0 . 001; differences between individual groups: Tukey's hsd test , all p<0 . 01 , Figure 2B ) . These differences had disappeared by the time the fish were tested for a second time during adulthood ( size: ANOVA: df = 3 , F = 0 . 61 , p = 0 . 61 , Figure 2C; latency time: ANOVA: df = 3 , F = 0 . 38 , p = 0 . 77 , Figure 2D ) .
Juveniles differed in size across the treatment groups as they were subject to different feeding rations during the tests . NHH and SLH fish received near ad libitum food and were presumably satiated , whereas NLL and SHL fish experienced a food shortage most likely resulting in a stronger motivation of these two groups to approach the test apparatus . This is reflected in substantial differences in the times to leave the shelter and to approach the choice apparatus . A differential motivation cannot explain our results however , as in this case the learning performance should differ between NHH/SLH and NLL/SHL fish . Moreover , these differences in latency time had disappeared , when testing the fish the second time during adulthood . Potential motivational differences were now eliminated as the size differences had vanished and all fish received the same food rations . Poor early nutrition can adversely affect neural development [16]–[18] and it can have a negative impact on song learning in birds [19] , [20] and intelligence quotients in humans [21]–[23] . Also this factor cannot explain our findings as NHH fish did not perform better than NLL fish . We assume that the low-food ration was sufficiently high to sustain normal neural development , as in a previous study [14] , [24] S . pleurospilus raised on the NLL ration developed and reproduced normally . Body size is amongst the traits under strongest selection [25] . Juvenile fish should have high incentives to grow fast , since the number of potential gape-size limited predators decreases exponentially with increasing body size [26] . Fast growth can have negative effects , however ( reviewed in [27] ) , including a negative influence on cognitive performance . In zebra finches , birds that had the highest rates of compensatory growth after experiencing a period of a reduced food ration performed worst in a subsequent learning task [13] . This effect might result from a trade-off between investment in growth versus neural development [28] , [29] or from prolonged stress due to increased foraging activity leading to chronically elevated levels of corticosterone , which in turn can adversely affect neural development [30] . If compensatory growth had affected the learning performance of S . pleurospilus , NHH fish should have outperformed those groups that had not started on a high-food diet and that exhibited compensatory growth in our experiment ( all fish reached similar sizes at the time when adults were tested , but NLL fish took longer to do so than SLH fish ) . If the brain was especially vulnerable to negative effects of compensatory growth during the juvenile period , then the fish that experienced a switch from a low to a high ration ( SLH; highest degree of compensatory growth ) should perform worst . The opposite was the case , however , as SLH fish outperformed NLL fish . To ensure that the learning experience of juveniles did not influence the subsequent learning performance of adults [31] , we performed a test series , which confirmed that the fish did not remember the conditioned cue of the juvenile test series before starting the second series . Other fish species have been shown to forget learned foraging techniques already within 2 d [32] , whereas the tests of juveniles and adults in our experiment were more than 1 y apart . We are therefore confident that we tested independent learning abilities of the fish in both test series rather than memory effects . Twelve individuals , which replaced fish that had died until the onset of the second test series and which had not been tested as juveniles , slightly outperformed the previously tested fish in the learning tests ( Table 1; see Material and Methods for details ) . Possibly fish tested 1 y before may still have associated the test apparatus with a food reward , as apparently they were less afraid of the test apparatus ( shorter latency times until approach; see Material and Methods ) . They may therefore have been less attentive to the type of stimulus cue during the training phase than previously untested fish . These fish approached the apparatus more cautiously and hence may have had more time to associate the cue with food . The effect of previous learning experience was the same across all treatment groups . Environmental conditions during development may trigger changes in morphology , physiology , or behavior , which can confer an adaptive advantage later in life if an animal remains in these conditions [33] . The main mechanisms proposed to explain such plastic responses to environmental cues involve repeated stimulation , for example , through physical exercise facilitating muscle development or by early neural stimulation through environmentally enriched raising conditions , which enhances cognitive abilities later in life [7] , [12] , [34] . But neither environmental enrichment nor training through repeated neural stimulation can explain our findings , as the sensory environment was kept constant during ontogeny and resource availability was changed only once . Our results rather show that already a single event—a change of food ration—during early ontogeny triggers learning ability possibly indicating the existence of a novel pathway of plastic neural development . It has been hypothesized that changing environments improve learning abilities , which consequently may allow animals to behave more adequately and flexibly [7] . Our results support this hypothesis by showing that environmental change can indeed directly affect learning abilities , independently of motivational differences between individuals . Changing environments experienced early in ontogeny can greatly improve the flexibility of behavior [6]–[8] . If such effects result partially from better learning abilities induced by early environmental change , these studies elucidate the manifold possible consequences of improved learning abilities , which extend to a wide range of behavioral contexts . The life history and ecology of S . pleurospilus suggests that the improvement of cognitive abilities in response to environmental change is adaptive . S . pleurospilus are algae grazers and hence depend upon the primary production of turf algae , which is influenced mainly by light intensity and a suitable substrate such as rocks and stones [35] . Algae productivity decreases exponentially with depth [35] . While juvenile S . pleurospilus inhabit the shallow regions of the lake with the highest algae intensity and some fish stay there throughout adulthood , other fish start to settle in deeper water around maturation [14] . These fish should benefit from increased cognitive abilities , as they have to cope with entirely different nutritional conditions . Improved cognition may help them to find and remember patches of high-quality turf algae ( reviewed in [36] ) , while those fish remaining stationary in the natal habitat do not necessarily require a better cognitive performance . Hence our findings suggest that habitat shifts can make these animals smarter . More generally , animals forced to cope with environmental changes as caused , for example , by anthropogenic perturbations of their habitats may benefit from improved cognitive abilities induced by these perturbations when forced to adjust to the new conditions . In conclusion we show for the first time that a single change in food availability early in life can enhance life-long learning abilities . Hence our study provides experimental support for the hypothesis that selection favors higher cognitive abilities in unpredictable or changing environments [5] , [9] . It also suggests a mechanism of how animals can acquire better abilities to cope with such environments: an environmental switch early in ontogeny may enhance learning ability persistently .
Simochromis pleurospilus is a maternally mouthbrooding cichlid of the subfamily Tropheini endemic to Lake Tanganyika , East Africa . It lives along the rocky shores of the lake where it feeds on epilithic turf algae . S . pleurospilus reproduces all year round and adult males defend small , adjoining territories of 2–4 m2 where females visit them to spawn . Juveniles and females are non-territorial and use large home ranges . After spawning females leave the male territory immediately and care for the clutch on their own [24] . Approximately 28 d after spawning , the young are independent . Juveniles and adults live sympatrically , but juveniles are confined to the shallow areas between 0 . 5 and 2 m depth , whereas adults often disperse to feed in greater depth between 1 and 12 m ( [14] , A . Kotrschal & B . Taborsky submitted ) . We raised 130 fishes in separate 20-l Plexiglas tanks , each equipped with a layer of sand , a flower-pot half for shelter , and an internal biological filter ( see [24] for details on experimental set-up ) . The experimental fish were derived from seven clutches of different females , and siblings were proportionally distributed over all treatments . We exposed the fish to two different feeding conditions in early and late adolescence , respectively , using a full-factorial design . Fish either received ( 1 ) a high food ration both in early and late life ( abbreviated as NHH , where “N” stands for “Not switched”; n = 40 ) ; ( 2 ) a low food ration both in early and late life ( NLL , n = 40 ) ; ( 3 ) a high food ration in early life , switched to a low food ration in late life ( SHL , where “S” stands for “Switched”; n = 22 ) ; or ( 4 ) a low food ration in early life , switched to a high food ration in late life ( SLH , n = 22 ) . Diet switches were performed either at 77 d ( i . e . , after the first third of the juvenile period; SLH: n77d = 11; SHL: n77d = 11 ) or at 133 d of age ( i . e . , after the second third of the juvenile period; SLH: n133d = 11; SHL; n133d = 11 ) . We had switched diets at two different ontogenetic stages to enhance the chances to capture a potential sensitive period when a change in resource availability affects cognitive abilities . As in the learning trials fish switched at day 77 did not perform differently from fish switched at day 133 , neither as juveniles ( GzLM; SLH: 77 d versus 133 d: χ2 = 2 . 1 , p = 0 . 15; SHL: 77 d versus 133 d: χ2 = 0 . 3 , p = 0 . 57 ) nor as adults [SLH: 77 d versus 133 d: χ2 = 0 . 09 , p = 0 . 77; SHL: 77 d versus 133 d: χ2 = 2 . 3 , p = 0 . 13 ( details of model see section Statistical Analysis ) ] , we combined the data of early and late switched fish resulting in four treatment groups: NHH , NLL , SLH , and SHL . Fish were fed 6 d a week with standardized agarose cubes containing an amount of Tetramin flake food corresponding to 12% ( near ad lib ) or 4% of mean body weight plus 5% Spirulina algae . All fish of a treatment group received the same food ration , which was based on the mean body mass of fish within this group . We adjusted the food rations to increasing mean body weights every 14 d . We stopped adjusting the rations to body weight in NHH fish at 189 d , because they no longer depleted the food cubes . We continued to adjust the ration for the NLL , SLH , and SHL fish until day 259 when they reached the same body size as NHH fish . Thereafter all fish were kept on the same food ration . We measured lengths and weights of fish every 3 wk . Standard lengths were read from a measuring board with a 1 mm grid and were estimated to the nearest 0 . 5 mm by eye . Weights were read to the nearest 1 . 0 mg from an electronic balance . All measurements were taken before the daily feeding and done by the same person ( A . K . ) . We first trained the animals to associate a certain visual cue with food and thereafter determined the number of correct decisions made when presenting the cue . We did the first test series in the juvenile phase shortly before maturation ( J ) at an age of 172 d ( ±10 d ) when the fish still received different food rations and differed in body size between treatments . The second test series was done 1 y later in adults ( A ) at an age of 585 d ( ±10 d ) , when all fish were fed the same rations and were of similar size . Each fish was tested in its individual raising tank . Twelve juveniles were excluded , as they were used in a pilot study after which we adjusted the testing protocol . Furthermore 12 fish never left their shelter within 12 min , yielding a final sample of 66 fish for the juvenile test series ( NHH = 19 , NLL = 19 , SHL = 14 , SLH = 14 ) . One year later some fish that had been tested previously had died in the meantime . Therefore we added 12 previously untested individuals to increase sample sizes . We used all SHL , SLH , and NLL fish still alive and 30 NHH fish for the second test run . Five adults refused to take food from the test apparatus and eight adults never left their shelter within 12 min , resulting in a sample size of 77 fish for the adult test series ( NHH = 30 , NLL = 25 , SHL = 8 , SLH = 14 ) . Overall 46 fish participated in all 6 juvenile trials , and 69 fish participated in all 10 adult trials . The mean rate of participation was 5 . 3 ( ±1 . 4 SE ) times out of 6 in juveniles and 9 . 7 ( ±1 . 0 SE ) times out of 10 in adults . Although juvenile NLL and SHL fish participated more often than NHH and SLH fish ( ANOVA: F = 4 . 63 , p = 0 . 005 ) , this did not bias the results because the statistical model accounts for participation rate ( see Statistical Analysis ) . Adult fish of all treatment groups participated at a similar level ( ANOVA: F = 1 . 04 , p = 0 . 38 ) . Since not all fish participated in every trial we used binary probit-link generalized linear models ( GzLM ) to analyze the cognitive performance with the total number of correct choices as the dependent variable and the number of times the fish participated as the independent variable [37] . We included food ration in early adolescence ( “early food treatment” ) and in late adolescence ( “late food treatment” ) as fixed factors . Twelve adults that had not been tested as juveniles took on average 70 s longer to enter the choice area ( Mann-Whitney U: Z = −2 . 19 , p = 0 . 028 ) , but they outperformed those fish already tested as juveniles ( GzLM , χ2 = 6 . 11 , p = 0 . 013 ) . As the latter effect occurred across treatment groups ( indicated by an absence of a significant interaction between treatment group and previous test experience , GzLM: χ2 = 352 , p = 0 . 84 ) , we included previous test experience ( yes or no ) in the model of adult learning performance . To examine whether the amount of food per se influenced the likelihood of correct choices , we tested if a positive , a negative ( linear predictor ) , or a dome-shaped ( quadratic predictor ) relationship exists between these two variables . We determined the amount of food received by each individual by calculating the percentage of food mass contained in the food pellets relative to the body mass of individuals using data from our tri-weekly body mass measurements . We then took the mean of these values during the entire juvenile period ( i . e . , until week 30 ) as a measure of food consumed by individual fish . All analyses were done with SPSS 17 . 0 , SPSS Inc . , Chicago .
|
Animals with higher cognitive abilities should be better capable of producing new , modified , or innovative behaviors as this ability could allow them to cope better with unpredictable environmental changes . Changing environments may hence select for higher cognitive abilities . Similarly , changing conditions during ontogeny can cause plastic responses , helping individuals to adapt to their current environment . In this study , we have used the cichlid fish Simochromis pleurospilus to show experimentally that individuals subjected to a change in food ration early in life ( i . e . , low to high or vice versa ) outperform fish kept on constant rations in a learning task later in life . Remarkably , this result was independent of the direction of the implemented change or the average amount of food each fish received , and the results in the juvenile stage did not change in adulthood . Our results suggest that a single environmental change early in life might enhance cognitive abilities in animals .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"evolutionary",
"biology/animal",
"behavior"
] |
2010
|
Environmental Change Enhances Cognitive Abilities in Fish
|
Numerous biomolecular interactions involve unstructured protein regions , but how to exploit such interactions to enhance the affinity of a lead molecule in the context of rational drug design remains uncertain . Here clarification was sought for cases where interactions of different ligands with the same disordered protein region yield qualitatively different results . Specifically , conformational ensembles for the disordered lid region of the N-terminal domain of the oncoprotein MDM2 in the presence of different ligands were computed by means of a novel combination of accelerated molecular dynamics , umbrella sampling , and variational free energy profile methodologies . The resulting conformational ensembles for MDM2 , free and bound to p53 TAD ( 17-29 ) peptide identify lid states compatible with previous NMR measurements . Remarkably , the MDM2 lid region is shown to adopt distinct conformational states in the presence of different small-molecule ligands . Detailed analyses of small-molecule bound ensembles reveal that the ca . 25-fold affinity improvement of the piperidinone family of inhibitors for MDM2 constructs that include the full lid correlates with interactions between ligand hydrophobic groups and the C-terminal lid region that is already partially ordered in apo MDM2 . By contrast , Nutlin or benzodiazepinedione inhibitors , that bind with similar affinity to full lid and lid-truncated MDM2 constructs , interact additionally through their solubilizing groups with N-terminal lid residues that are more disordered in apo MDM2 .
A large fraction of proteins contain substantial regions that are unstructured in native conditions [1 , 2] . Protein disorder plays a key role in biomolecular function , enabling proteins to tune binding affinity and specificity to diverse partners [3] . In particular protein-complexes that involve interactions with disordered protein regions often involve disorder-to-order transitions ( and vice versa ) in one or both partner [4] . Nature is a rich source of inspiration in the search for new therapeutic-agents . Much successful medicinal chemistry has arisen from efforts to mimic biomolecular recognition mechanisms , prominent examples include GPCR- ( ant ) agonists or transition state analogue enzyme inhibitors . Likewise , there is evidence that small-molecules can productively target disordered protein regions [5] . For instance the Metallo lab has reported several small-molecule ligands that interact with disordered regions of the transcription factor c-Myc [6] , though concerns about binding specificity have been raised [7] . Herbert et al . discovered an allosteric inhibitor of FGFR that induces ordering of an unstructured segment into a helical region [8] . Similar mechanisms have been inferred for allosteric inhibitors of pyruvate kinase [9] . How to anticipate productive interactions in the context of rational drug design with experimental or computational methods remains however uncertain [10] , and detailed investigations are necessary to progress our understanding of this molecular recognition mechanism . This report focusses on the consequences of small-molecule interactions with disordered protein regions , and their computational treatment . A clear illustration of the opportunities is provided by the oncoprotein MDM2 . Disrupting the interaction of MDM2 with the tumor suppressor p53 is an attractive strategy in oncology [11–15] . The N-terminal domain of human MDM2 ( ca . 120 residues ) interacts with the transactivation domain ( TAD ) of p53 . This interaction is mediated by Phe19 , Trp23 and Leu26 of p53 that protrude into three hydrophobic pockets of MDM2 [16 , 17] . Additionally , the first 24 residues of the N-terminal domain of MDM2 form an unstructured flexible lid , that can adopt both “open” or “closed” states , the latter competing for the p53-binding site through a pseudo-substrate mechanism ( Fig 1A ) [1 , 2 , 18–20] . Ground breaking NMR studies from Showalter et al . indicated that the exchange between open and closed lid conformations occur on a >10 ms time-scale [3 , 18] . Potent p53/MDM2 inhibitors that bind to the p53-binding site of MDM2 have been developed , including Nutlins [4 , 21 , 22] , 1 , 4-benzodiazepine-2 , 5-diones [5 , 23] , and piperidinones [6 , 24] ( Fig 1B ) . Many other classes of inhibitors have been reported , and some have progressed to clinical trials [7 , 25 , 26] . Although it has been known for some time that the lid responds differentially to large peptide-like ligands and small-molecules [18] , this MDM2 region has not historically been fully considered in structure-based campaigns since the high lid flexibility hinders considerably biochemical studies and biophysical measurements [8 , 25] . Also , similar binding affinities to full-length MDM2 and shorter truncated lid constructs ( typically 17–125 ) were reported for earlier inhibitors , suggesting that interactions with the lid region are of little importance to optimize binding affinity and selectivity of p53/MDM2 antagonists . However in late 2012 , Michelsen et al . reported a remarkable disorder-to-order transition of the lid region upon binding of piperidinone-2 ( Pip2 ) ligands [24] , with the lid adopting a short α-helix ( residues 21–24 ) followed by a β-turn ( residues 17–20 ) and β-strand ( residues 14–16 ) . Significantly , unlike other ligands , the binding affinity of Pip2 towards lid-truncated MDM2 decreased by ca . 25 fold , leading Michelsen et al . to suggest that targeting this ordered MDM2 lid conformation may provide new opportunities for the design of potent and selective p53/MDM2 inhibitors . Given the surprising outcome and high-potential for drug design purposes , clarification into the details of small-molecule lid interactions was sought for this ligand-dependent disorder-order transition . To this purpose extensive computation of atomistic lid conformational ensembles in explicit solvent for apo MDM2 and for MDM2 in complex with the four ligands depicted in Fig 1B was pursued with the aid of an enhanced molecular simulation protocol . This featured accelerated molecular dynamics ( aMD ) [27–29] , umbrella sampling ( US ) [30] , and variational free energy profile ( vFEP ) methods [31 , 32] . Analysis of the resulting lid structural ensembles identified significant differences in lid recognition mechanisms for the different ligands , and suggested a rationale for the high affinity of Pip2 ligands for extended-lid MDM2 constructs .
As expected in light of the anticipated time scale for transitions between open and closed lid states , unbiased equilibrium MD simulations on the timescale of several hundreds of ns of apo MDM2 , initiated from a range of different initial lid conformations , failed to generate a single transition from open to closed states of the lid ( S1 Fig ) . By contrast the aMD simulation protocol sampled transitions between open and closed apo lid states in simulations of ca . 100 ns ( S1 Fig ) . In the simulations of small molecule bound complexes , complete lid opening was not observed with cMD or aMD protocols , but enhanced conformational fluctuations were observed with the latter protocol ( S1 Fig ) . Though some variability is apparent and no low-dimensionality projection is fully satisfactory for such complex system , the present protocol enabled the detection of significant differences between the different complexes , and a broad range of lid conformations were observed ( Fig 2 ) . Uncertainties in the computed free energy surfaces for each system were assessed by monitoring convergence over regular time-intervals ( S2–S3 Figs ) . Equilibrium properties of the ligand-bound complexes are reasonably reproducible , with greater uncertainties observed for the most flexible lid residues in apo MDM2 ( Figs 3–5 ) . Three major low free energy regions were identified in the FES of apo MDM2 ( Fig 2A ) . The lowest corresponds to a “closed” state ( CV1 = 31 Å; CV2 = 62° ) , with the lid adopting a semi-extended conformation in contact with α2 helix residues . While this conformation would hinder binding of the p53 TAD as a result of steric clashes with the lid , the Phe19-Trp23-Leu26 cleft was still accessible to small molecules such as Nutlins . The second region ( CV1 = 7 Å; CV2 = 119° ) corresponds to an “open” , compact state of the lid . In this state , the p53-binding site is fully accessible to large and small ligands . No open , extended lid conformations were observed , thus the lid in the open state adopts collapsed structures . A third additional region ( CV1 = 10 Å; CV2 = 64° ) corresponds to an intermediate “semi-open” state . In this conformation , the lid approaches the core of MDM2 . Although the hydrophobic pocket was still fully accessible to small ligands , binding of the larger p53 TAD would be hindered . On the basis of NMR spin relaxation measurements , Showalter et al . found that apo MDM2 favors a closed state over an open state by ca . 1 . 2 kcal/mol . While the definition of an open or closed state is somewhat arbitrary , comparison with the present results was pursued by classifying each bin in the computed two-dimensional free energy surface as “open” or “closed” . Defining a closed state for bins with a cutoff >23 Å for the lid extension and <80° for the lid-core angle collective variables , the closed state is favored over the open state by 0 . 3±0 . 1 kcal/mol . Varying this cutoff by ±3 Å and ±6° has little effect on the result ( ±0 . 1 kcal/mol ) . Thus the simulations predict a higher population of open states than the NMR data , though the agreement is still reasonable . This finding is consistent with reports from Best and co-workers that have found unfolded conformational ensembles to be too compact and stabilized by residual secondary structures with current classical force fields [33 , 34] . The p53 TAD-MDM2 simulations ( Fig 2B ) revealed a single broad free energy basin corresponding to a range of fully “open” and compact conformations ( CV1 = 12 Å; CV2 = 105° ) . Due to the presence of the p53 peptide in the hydrophobic pocket , the lid is displaced towards a broad range of open disordered states . Nevertheless , the most stable lid conformations detected for the p53 TAD-MDM2 complex were similar to the most open conformations shown in the apo-MDM2 system , both displaying a lid core angle of ca . 120° and a compact state of the lid ( CV1 ca . 10 Å ) . A low free energy basin ( Fig 2C ) corresponding to a “closed” , extended state was found for Nutlin-3a/MDM2 ( CV1 = 27 Å; CV2 = 41° ) . In comparison with apo MDM2 , the occurrence of open lid state is not only decreased , but the closed state differs in nature , adopting even lower lid-core angle values . This is because the position of the lid in the closed state shifts from contacting helix α2 to cover Nutlin-3a . Thus , Nutlin-3a does not compete with the lid for access to the p53-binding site , and stabilizes a different “closed” lid state than observed in apo MDM2 . An additional local minimum , corresponding to “semi-open” conformations , was also observed for the Nutlin-3a/MDM2 complex , with the lid region still interacting with the ligand but in a more compact conformation . In the presence of Bzd , a change in favored “closed” lid state is also observed ( Fig 2D ) . In this case the FES presented a single free energy basin ( minimum at CV1 = 38 Å; CV2 = 37° ) that corresponds to a “closed” and fully extended lid conformation over the hydrophobic cleft . The lid extension in this case was ca . 10 Å greater than that observed in low free energy conformations in the apo and Nutlin3a-MDM2 systems . The lid may readily adopt even lower CV2 angle ( <40° ) values than with Nutlin-3a , because the smaller size of the Bzd ligand enables the lid to close further the p53 binding cleft . The FES of MDM2 in complex with Pip2 displayed a major free energy basin ( CV1 = 16 Å; CV2 = 62° ) indicative of a relatively compact “Pip2” lid state ( Fig 2E ) that significantly differs from the previously observed states . Lid conformations in this region of the FES contain a short ordered α-helix at residues 21–24 , a sharp bend around residues 17–20 , and residues 10–16 adopt an extended conformation . This lid conformation is consistent with the X-ray and NMR data reported by Michelsen et al . for MDM2 ( 6–125 ) [24] . A secondary minimum is also apparent ( CV1 = 10 Å; CV2 = 85° ) , in this “open-Pip2” conformation the lid still contains a short α-helix at residues 21–24 , but segment 1–16 adopts a collapsed rather than extended conformation . Taken together , these results indicate that the conformational preferences of the MDM2 lid region are remarkably influenced by the chemical structure of the different classes of small-molecule MDM2 antagonists , and that multiple distinct closed states are readily achieved by the MDM2 lid region . Further insights into these intriguing results are gained by qualitative inspection of equilibrium lid intermolecular and intramolecular interaction patterns . Attention was focused on lid residues that exhibit an average number of hydrophobic contacts or hydrogen bonds that is significantly above typically observed values . Cutoff values of 4 hydrophobic contacts or 0 . 25 hydrogen bonds for intermolecular interactions , and 4 hydrophobic contacts or 1 . 0 hydrogen bond for intramolecular lid interactions were deemed sufficient to identify the most significant interactions . In addition , the observed contacts were mapped to specific MDM2 core residues and ligand functional groups involved in interactions with the lid . This was done by generating and visualizing representative structural ensembles of each complex by resampling of computed conformations according to their equilibrium probabilities .
Little is known about the conformational preferences of the MDM2 lid region owing to its considerable flexibility that hinders experimental studies , and a slow rate of exchange ( >10 ms ) between open and closed states that is inaccessible to conventional molecular simulations . Further , because the lid is highly flexible in its distinct closed or open states , an accurate characterization of lid interactions necessitates a description in terms of structural ensembles rather than a single representative structure . The present study addressed the technical challenge of computing atomically detailed lid structural ensembles with the aid of accelerated molecular dynamics , umbrella sampling and variational free energy profile methodologies . Altogether , over 10 microseconds of biased molecular dynamics simulations was used to generate free-energy landscapes for the MDM2 lid region in five different conditions . This likely represents the largest effort to date to resolve the conformational ensembles of the MDM2 lid region by means of explicit-solvent atomistic molecular dynamics simulations . The computed FES were deemed reasonably well converged ( S2–S3 Figs ) , and the predicted free energy difference between open and closed lid states in apo MDM2 was in reasonable agreement with experimental data ( within 1 kcal . mol-1 ) . The shift to an open lid state in the MDM2-p53 peptide complex was observed , in expectation with previous NMR studies . [18] These observations give confidence in the accuracy of the computed lid ensembles for the small-molecule bound simulations , for which relatively little was known prior to this study . A striking novel result from this study is that the MDM2 lid adopts different closed states when bound to different classes of small-molecules . In apo MDM2 the closed state of the lid lies mainly over helix α2 , but in Nutlin-3a or Bzd bound simulations , the lid moves away from helix α2 to cover the small-molecule ligands . In Pip2 bound simulations , the base of the lid orders into a α-helix/β-turn motif , with the rest of the lid showing considerable disorder . Thus it is inadequate to picture the MDM2 lid in equilibrium between a well-defined open and closed lid state . Instead the flexibility of the lid enables considerable adjustments in the closed state to best accommodate chemically distinct ligands . Comparison of the structures sampled in the FES depicted in Fig 2 shows that the conformations the lid adopts when p53 is bound are broadly present in the apo FES . However , conformations similar to the major conformations seen in the presence of the small-molecule ligands were not detected in the apo ensemble . This suggests that significant induced-fit is necessary to fine-tune interactions between the small-molecule ligands and the lid region . The Nutlin-3a and Bzd ligands were both found to establish significant interactions with the lid region ( Figs 6C , 6D , 8C2 and 8D2 ) . In the case of Nutlin-3a , significant hydrophobic contacts and hydrogen-bonding interactions involve lid residue Thr10 and the piperazinone carbonyl group . In the case of Bzd , the carboxylate moiety forms significant hydrogen-bonding interactions with lid residues Thr16 and Ser17 . By comparison with the structural ensembles of apo-MDM2 and Pip2-bound MDM2 , it can be seen that these additional interactions drag the lid further over these ligands . In the Pip2-MDM2 complex , the lid cannot easily access the Pip2 carboxylate moiety because it is partially covered by core residues His96 and Lys94 . Successful MDM2 antagonists generally position hydrophobic moieties in the p53/MDM2 binding cleft owing to the relatively apolar character of this binding site . Acceptable solubility of MDM2 small-molecule antagonists typically requires the introduction of an additional solubilizing group . On the basis of structural models where the MDM2 lid was absent , solubilizing groups in many classes of MDM2 antagonists have generally been positioned to lie over the surface or away from the protein towards region of space that have been assumed to be occupied solely by solvent molecules [37] . Although , it has been known for a long time that the precise chemical nature of the solubilizing group can substantially modulate the binding affinity of MDM2 ligands , an explanation for this observation has been lacking . For instance , removal of the solubilizing group in Nutlin derivative RG7112 decreases binding affinity by a factor of 100 , whereas in Bzd analogues , the length and acidic/basic nature of the solubilizing group can modulate binding affinity by a factor of 85 [37 , 38] . The present results suggest that MDM2 antagonist design programs should routinely consider the possibility that solubilizing groups may interact with the lid region and that this may significantly impact binding affinities . The Nutlin-3a and Bzd ligands are known to show little gains in potency through lid interactions , yet significant interactions with a large portion of the lid region were observed ( Fig 6C and 6D ) . Energetic analysis revealed that the favorable ligand-lid interactions are here offset by unfavorable lid-core and lid-lid interactions ( Fig 9 ) . By contrast , the potency of Pip2 ligands benefits significantly from the presence of the lid region , yet relatively fewer contacts between the ligand and the lid are observed ( Fig 6E ) . This is corroborated by the energetic analysis ( Fig 9 ) that revealed unfavorable lid-ligand and lid-lid interactions that are offset by favorable lid-core interactions . Additionally all Pip2-lid contacts occur at or after residue 14 in the lid primary sequence , whereas Nutlin-3a and Bzd ligands also engage with lid residues in segment 6–14 , notably with their solubilizing groups . Interactions with this segment are associated with a significant local decrease in lid flexibility ( Fig 3A , 3C and 3D ) . These observations suggest an entropy-enthalpy compensation mechanism is also at play; in other words Nutlin-3 and Bzd ligands do not benefit significantly from the additional contacts formed with the lid because those also involve residues that were significantly disordered in apo MDM2 . A significant difference in Pip2-bound simulations is that lid residue Glu23 forms a stable salt-bridge with Arg97 in the core of MDM2 , while Gln24 is hydrogen bonded to backbone atoms of Pro20 , Ile19 and Thr16 . This network of hydrogen-bonding interactions effectively locks the base of the lid into the observed α-helix/β-turn motif ( Fig 4E ) . However this pattern of interactions is not observed in the apo MDM2 simulations , suggesting that this conformation is only stable in the Pip2 complex because the meta-chlorophenyl ring of Pip2 also packs against lid residues 14–16 ( Fig 8E2 ) . Evidently Pip2 cannot engage directly with the lid region in short MDM2 constructs that are truncated at residue 17 . Therefore the origin of the high affinity of Pip2 for the extended lid MDM2 construct of Michelsen et al . is attributed to the indirect stabilization of the lid α-helix at residues 21–24 via hydrophobic contacts between the meta-chlorophenyl ring of Pip2 and Val14/Thr16 . Interestingly , Bista et al . have recently reported that chloroindole carboxylate derivatives are also able to interact with the lid through conformational adjustments that fit a para-chlorobenzyloxy benzyl moiety deep near the α-helix lid [39] . Though the binding affinity data for these ligands was not reported for the MDM2 constructs studied here , this suggests that different strategies may be available to stabilize the α-helix lid region . Intriguingly , although absent in the Nutlin-3a and Bzd-bound MDM2 simulations , the α-helix lid has been observed in some crystallographic complexes of these ligands with shorter MDM2 ( 17–125 ) constructs [21 , 23] . However in the present simulations , lid residues 1–16 engage in interactions with the ligands and the MDM2 core region that preclude formation of the α-helix lid in segment 19–24 . This is corroborated by additional ( ca . 100 ns timescale ) MD simulations of Bzd/Nutlin-3a bound to MDM2 1–119 and 17–125 that suggest the α-helix lid is only stable in the short construct ( S6–S7 Figs ) . Taken together , these observations suggest that a strategy to productively exploit ligand-lid interactions is to 1 ) stabilize with hydrophobic contacts a lid conformation where lid residues 14–16 packs against the α-helix lid , and 2 ) hinder undesirable lid interactions with residues 1–13 by positioning solubilizing groups to interact with ordered MDM2 core residues . In Pip2 this is achieved through hydrophobic contacts of the meta-chlorophenyl ring with Val14/Thr16 and hydrogen-bonding interactions of a carboxylate moiety with His96 and Lys94 , but other solutions may be possible . In summary , a detailed analysis of the interactions of the N-terminal domain of MDM2 with several ligands was undertaken to elucidate conformational preferences of the MDM2 lid region and to rationalize the origin of the ca . 25-fold activity improvement of the Pip2 ligand for constructs including an extended lid region . The simulations of apo-MDM2 indicate that the lid is disordered , adopting a mixture of open and closed lid states . Binding of p53 shifts the equilibrium towards an open disordered state , in agreement with reported NMR data . [18] A novel significant finding is that the MDM2 lid exhibits different conformational preferences and significant interactions with different classes of small-molecule p53/MDM2 antagonists . Structural and energetic analyses show that the enhanced affinity of Pip2 for MDM2 constructs that include the full lid is due to hydrophobic contacts that facilitate structuring of an α-helix/β-turn motif in lid residues 17–24 . Nutlin-3a or Bzd ligands that show similar affinity for short or long MDM2 lid constructs hinder formation of this motif because they also engage through their solubilizing groups segments of the lid earlier in the primary sequence that are more disordered in apo MDM2 . Taken together , these findings suggest that a strategy to productively exploit MDM2 lid-interactions for inhibitor design is to target the base of the lid with deep hydrophobic contacts , and to position solubilizing groups so as to minimize the likelihood of interactions with polar lid residues and lid residues distant from the lid base . These findings may be of significance to facilitate the development of novel potent and selective p53/MDM2 ligands as putative anti-cancer agents , and more generally to suggest new hypotheses for productively targeting disordered protein regions in structure-based drug design efforts .
The bulk of computational studies of MDM2 ligand interactions have neglected the lid , [47–56] Notable exceptions include work from Verkhivker and Dastidar et al . that have studied lid conformations over ca . 10 ns time scales [57 , 58] . However work from Showalter et al . suggested that the lid exchanges between closed and open states over a much slower ( >10 ms ) timescale [18] . To overcome this technical challenge , a protocol featuring use of aMD [27–29] , US [30] , and vFEP [31 , 32] was adopted . Three-dimensional structures and trajectories were visually inspected using the computer graphics programs PyMOL[77] and VMD [78] . Hydrogen bonds and hydrophobic contacts between the MDM2 lid and MDM2 core/ligand regions were monitored using ptraj and cpptraj modules in AmberTools 12 [40 , 79] . The formation of a hydrogen bond was considered when the distance between donor and acceptor was shorter or equal to 3 . 0 Å and the angle between the acceptor , hydrogen and donor atoms was equal or larger than 154° . A hydrophobic contact was defined when the distance between two carbon atoms was less than 5 Å . To avoid counting a large number of trivial contacts , for every lid residue i , intramolecular lid-lid hydrophobic contacts with immediate neighboring residues ( i+1 , i-1 ) were excluded from the analysis . Diverse ensemble-averages of lid-residue properties were computed , including the average number of hydrogen bonds with core/ligand atoms , average hydrophobic contacts with core/ligand atoms , average backbone heavy-atom RMSD to a representative conformation taken from the lowest free energy bin ( RMSDi ) and average helical ( Hi ) , sheet ( Si ) turn ( Ti ) or coil ( Ci ) propensity , the latter four computed according to the DSSP code [35 , 36] . Lid energetics were also characterized by component analysis of interaction energies using the mmpbsa . py software [80] . Specific quantities evaluated for each complex were: ligand-lid interaction energies ( Elig-lid ) , lid-core interaction energies ( Elid-core ) , and lid-lid intramolecular non-bonded energies ( Elid-lid ) . All observables and quantities were obtained by reweighting statistics from the US sampled snapshots according to eq 3 . 〈Ai〉=∑j=1N ( 1M∑k=1MAj , k ) ⋅e−βΔGj∑j=1Ne−βΔGj ( 3 ) where <Ai> is the ensemble average of the property of interest for lid residue i , N is the number of US bins , M is the number of snapshots in bin j , Aj , k is the value of property Ai for snapshot k in bin j , and ΔGj is the free energy of bin j obtained by vFEP reweighting . To estimate uncertainties in the computed properties , the simulation data was split in four consecutive blocks of 1 ns each and property values computed separately . Mean <Ai> values are reported along with one standard error . Lastly , graphical depictions of representative lid structural ensembles for each system were obtained by randomly sampling 20 snapshots from the pooled US snapshots according to their computed equilibrium properties .
|
Life as we know it depends on interactions between proteins . There is substantial evidence that many interactions between proteins involve very flexible protein regions . These disordered regions may undergo disorder/order transitions upon forming an interaction with another protein . Many successful approaches to medicinal chemistry are based on mimicking the interactions of biological molecules with man-made small molecules . However how drug-like small-molecules may modulate protein disorder is currently poorly understood , largely because it is difficult to measure in details this type of interaction with experimental methods . Here we have used computer simulations to resolve with great details the process by which different small-molecules modulate the flexibility of a disordered region of the protein MDM2 . This protein is overexpressed in many cancers and small-molecules that recognize MDM2 have been developed over the last decade as possible novel anti-cancer agents . We show that the flexible MDM2 “lid” region adopts different conformational states in the presence of different small-molecules . Our results suggest why some classes of small-molecules form favorable interactions with the lid region , whereas others do not . These findings may prove crucial to develop new and more effective MDM2 inhibitors , and more generally to help drug designers target disordered proteins regions with small-molecules .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[] |
2015
|
Elucidation of Ligand-Dependent Modulation of Disorder-Order Transitions in the Oncoprotein MDM2
|
Centromeres are specialized chromatin regions marked by the presence of nucleosomes containing the centromere-specific histone H3 variant CENP-A , which is essential for chromosome segregation . Assembly and disassembly of nucleosomes is intimately linked to DNA topology , and DNA topoisomerases have previously been implicated in the dynamics of canonical H3 nucleosomes . Here we show that Schizosaccharomyces pombe Top3 and its partner Rqh1 are involved in controlling the levels of CENP-ACnp1 at centromeres . Both top3 and rqh1 mutants display defects in chromosome segregation . Using chromatin immunoprecipitation and tiling microarrays , we show that Top3 , unlike Top1 and Top2 , is highly enriched at centromeric central domains , demonstrating that Top3 is the major topoisomerase in this region . Moreover , centromeric Top3 occupancy positively correlates with CENP-ACnp1 occupancy . Intriguingly , both top3 and rqh1 mutants display increased relative enrichment of CENP-ACnp1 at centromeric central domains . Thus , Top3 and Rqh1 normally limit the levels of CENP-ACnp1 in this region . This new role is independent of the established function of Top3 and Rqh1 in homologous recombination downstream of Rad51 . Therefore , we hypothesize that the Top3-Rqh1 complex has an important role in controlling centromere DNA topology , which in turn affects the dynamics of CENP-ACnp1 nucleosomes .
Centromeres are unique regions of eukaryotic chromosomes that are essential for chromosome segregation at mitosis and meiosis . The specialized centromeric chromatin directs assembly of kinetochores , which serve as points of attachment for the spindle apparatus . In all eukaryotes , centromeric chromatin is marked by the presence of nucleosomes containing the histone H3 variant Centromere Protein-A ( CENP-A ) , which is a key determinant for centromere identity and essential for centromere function . The specific incorporation and maintenance of CENP-A at centromeres is an epigenetic phenomenon that is not yet fully understood , and new factors involved in CENP-A dynamics are continuing to be discovered . During DNA replication in S-phase pre-existing CENP-A is equally partitioned to sister centromeres and after chromosome segregation newly synthesized CENP-A is incorporated specifically at pre-existing centromeres , possibly involving a feed-forward mechanism between pre-existing CENP-A chromatin and CENP-A assembly factors [1]–[2] . In agreement , it was recently shown that the constitutive centromere-associated network ( CCAN ) component Centromere Protein C ( CENP-C ) is required for recruitment of the Mis18 complex [3] . The Mis18 complex is in turn necessary for localization of CENP-A to centromeres and is hypothesized to have a role in centromere priming [4]–[7] . Furthermore , Mis18 is required for centromere targeting of the CENP-A-specific chaperone Holliday Junction Recognition Protein ( HJURP ) , which is both necessary and sufficient for stable recruitment of CENP-A [6] , [8]–[9] . It binds specifically to pre-nucleosomal CENP-A and histone H4 , and has been shown to facilitate assembly of nucleosomes containing CENP-A in vitro [8]–[14] . The Mis18 complex and/or the CENP-A pre-nucleosomal complex also associates with the chaperone RbAp48 , which can mediate assembly of nucleosomes containing CENP-A in vitro [4]–[5] , [10]–[11] , [15] . Despite continuous advances in the identification of pathways and factors controlling CENP-A nucleosome assembly , the molecular architecture of CENP-A nucleosomes remains unclear . CENP-A , H4 , H2A and H2B can be assembled into conventional octameric nucleosomes with left-handed negative wrapping of DNA in vitro [13] , [16]–[19] . However , they can also be assembled into tetrameric hemisomes containing only one copy of each histone with right-handed positive wrapping of DNA [20] . Different studies aiming at determining the composition of CENP-A nucleosomes in vivo have found contradicting evidence for both octamers and tetramers [17] , [20]–[24] . Interestingly , recent studies imply that there may be cell cycle-dependent transitions in the structure of CENP-A nucleosomes in vivo [25]–[26] . This would reconcile the contradictions between different studies on CENP-A nucleosome structure . A few other models for the structure of CENP-A nucleosomes have also been proposed , but are associated with less experimental evidence . DNA topoisomerases catalyze changes in DNA topology by cutting , shuffling and re-ligating DNA strands . Nucleosome dynamics are intimately linked to DNA topology [27] . Since DNA is wrapped around the histone core of nucleosomes in a left-handed negative direction , negative supercoiling of DNA favors nucleosome assembly while positive supercoiling of DNA favors nucleosome disassembly . In agreement , DNA topoisomerases have been implicated in the dynamics of canonical H3 nucleosomes in vitro and in vivo [28]–[31] . Eukaryotic topoisomerase III is a type 1A topoisomerase capable of relaxing negatively supercoiled DNA [32]–[33] . Topoisomerase III displays evolutionarily conserved genetic and physical interactions with RecQ helicases and RecQ-mediated genome instability ( Rmi ) proteins [34]–[39] . RecQ helicases and Rmi1 stimulate relaxation of negative supercoiling by topoisomerase III and together they also have the ability to fully de-catenate and catenate DNA molecules as well as to ‘dissolve’ double Holiday Junctions [40]–[43] . The single RTR complex in Schizosaccharomyces pombe consists of Rqh1 , Top3 and Rmi1 . These proteins are critical for genome stability and have so far been implicated in homologous recombination ( HR ) and DNA damage checkpoint activation in vivo [35] , [44]–[50] . Both top3 and rmi1 deletion mutants stop dividing after just a few generations with severe defects in nuclear morphology and chromosome segregation [37] , [51]–[52] . Thermo-sensitive top3 mutants display growth defects , sensitivity to DNA damaging agents , illegitimate recombination , altered nuclear morphology , and defects in chromosome segregation at restrictive temperatures [46] , [53] . The lethality of top3 and rmi1 deletion mutants can be rescued by mutations in rqh1 , likely because Rqh1 creates intermediate structures in HR that without Top3 remain unresolved and prevent chromosome segregation [51]–[52] . Mutations in rqh1 result in similar but less severe phenotypes compared to top3 and rmi1 mutations [49] , [54]–[55] . In this report , we investigated the genome-wide localization of S . pombe Top3 and discovered that it is preferentially found at intergenic regions ( IGRs ) , sub-telomeres and centromeres . Top3 occupancy at IGRs is similar to that of Top1 and Top2 . On the other hand , high relative enrichment of Top3 at centromeric central domains is unique , and is positively correlated with CENP-ACnp1 occupancy . Both top3 and rqh1 mutants display defects in chromosome segregation and increased relative enrichment of CENP-ACnp1 at centromeric central domains . Thus , the Top3-Rqh1 complex normally limits the levels of CENP-ACnp1 in this region . Altered levels of CENP-ACnp1 are accompanied by changes in the levels of the CENP-ACnp1-specific chaperone HJURPScm3 and are independent of HR downstream of Rad51 . We therefore suggest that the Top3-Rqh1 complex has an important role in controlling centromere DNA topology and thereby the dynamics CENP-ACnp1 nucleosomes . Specific removal of negative supercoiling by Top3 should inhibit assembly of and destabilize octameric CENP-ACnp1 nucleosomes with left-handed negative wrapping of DNA . In addition , removal of negative supercoiling may limit centromeric transcription , which is hypothesized to promote CENP-ACnp1 nucleosome assembly . In this model , impaired Top3 activity would facilitate assembly of CENP-ACnp1 nucleosomes . Alternatively , the activity of Top3 may create a unique topological state at centromeres that specifically favors formation of tetrameric CENP-ACnp1 hemisomes with right-handed positive wrapping of DNA over octameric CENP-ACnp1 nucleosomes . In this model , impaired Top3 activity would result in a shift from formation of CENP-ACnp1-containing hemisomes towards octameres at centromeres .
In this study we used the previously isolated but uncharacterized thermo-sensitive top3-105 mutant [56] . We sequenced the top3 open reading frame and identified a A762G base pair substitution that results in a Tyr209Cys amino acid change ( Figure 1A ) . This residue is found in a region involved in conformation changes upon DNA binding and is conserved in budding yeast Top3 as well as in metazoan Top3α [57] . Growth of the top3-105 mutant is similar to wild type at 25°C and 30°C , but severely impaired at the restrictive temperature of 36°C ( Figure 1B ) . When cultures grown at 25°C are shifted to 36°C , the top3-105 mutant initially proliferates with kinetics similar to wild type , but after approximately two generations proliferation is severely slowed , with essentially no further increase in cell numbers ( Figure 1C ) . Deletion of rqh1 rescues the lethality of a top3 deletion [51]–[52] . Growth of the top3Δ rqh1Δ double mutant is similar to the rqh1Δ mutant and somewhat slower compared to wild type at all temperatures ( Figure 1B and 1C ) . The lethality of a top3 deletion can also be suppressed by deleting rad51 , which is upstream of Top3 and Rqh1 in the ‘dissolution’ pathway in HR [35] . The rad51Δ single mutant displays slow growth , which is further enhanced at 36°C , but not to the same extent as for the top3-105 mutant ( Figure 1B and 1C ) [58] . The severe growth defect of the top3-105 mutant is not suppressed by deletion of rad51 as the top3-105 rad51Δ double mutant displays similarly impaired growth as the top3-105 mutant at 36°C . Moreover , the top3-105 rad51Δ double mutant displays a small synthetic growth defect at 25°C and 30°C . Growth of the rqh1Δ rad51Δ double mutant is largely similar to the rad51Δ single mutant at all temperatures [59] . Subsequently , we looked at nuclear morphology in the mutants using 4′ . 6-diamidino-2-phenylindole ( DAPI ) to stain DNA and anti-tubulin immunofluorescence to stain the mitotic spindle , respectively . After 8 hours at 36°C , there is a large increase in the fraction of abnormally long cells ( >15 µm ) for the top3-105 mutant ( 28% , n = 500 ) compared to wild type ( 0 . 4% , n = 500 ) , which is indicative of cell cycle delay ( Figure 2A ) . Both the rqh1Δ and top3Δ rqh1Δ mutants display less pronounced increases in the fractions of unusually elongated cells ( 14% and 17% , respectively , n = 500 ) . The rad51Δ , rad51Δ top3-105 and rad51Δ rqh1Δ mutants all display equally large fractions of elongated cells ( 53% , 52% and 53% , respectively , n = 500 ) [58] . Furthermore , the top3-105 mutant display various nuclear and mitotic defects , including amorphous and fragmented nuclei , unequal segregation of DNA , lagging chromosomes and a ‘cut’/‘torn’ phenotype , after 8 hours at 36°C ( Figure 2A–2B ) . The rqh1Δ and top3Δ rqh1Δ mutants display similar but less pronounced defects in chromosome segregation compared to the top3-105 mutant ( p = 0 . 024 and p = 0 . 00023 ) ( Figure 2A–2B and Table S1 ) . Deletion of rad51 rescues the severe mitotic defects of the top3-105 mutant ( p = 0 . 0046 ) . However , the top3-105 rad51Δ double mutant and the rad51Δ single mutant both still display moderate defects in chromosome segregation . Moreover , deletion of rad51 does not affect the chromosome segregation defects of the rqh1Δ mutant , as the rqh1Δ rad51Δ double mutant displays similar levels of mitotic defects as the rqh1Δ and rad51Δ single mutants ( Figure 2A–2B and Table S1 ) . Therefore , it seems that Top3 , Rqh1 and Rad51 are all important for normal mitotic chromosome segregation . Next , we investigated the genome-wide relative enrichment of Top3 by chromatin immunoprecipitation ( ChIP ) of c-Myc epitope tagged Top3 expressed from the endogenous locus and hybridization to high-resolution tiling microarrays ( ChIP-chip ) . Along the euchromatic chromosome arms high relative enrichment of Top3 is preferentially found at intergenic regions ( IGRs ) , while it is generally depleted from open reading frames ( ORFs ) ( Figure 3A ) . Top3 occupancy is also found at sub-telomeric regions , including the rDNA clusters at the left and right sub-telomeric regions of chromosome III ( tel3L and tel3R ) ( Figure 3A and Figure S1A–S1B ) . Telomeric repeats are not represented on the array and therefore telomere occupancy could not be investigated . Moreover , we observed a consistently high relative enrichment of Top3 at centromeres ( Figure 3A and Figure S1A–S1B ) . Next , we determined the genome-wide average levels of Top3 when genes are aligned at the transcription start site ( TSS ) and transcription termination site ( TTS ) , respectively . Top3 display high average relative enrichment at the TTS , but depletion from the TSS and the ORF ( Figure 3B ) . There is also a small average enrichment of Top3 just upstream of the TSS , where gene promoters are generally localized . Overall , this binding pattern is similar to that of Top1 and Top2 . Depletion of Top3 from the TSS and ORF as well as enrichment at the TTS is more pronounced for strongly transcribed genes , whereas Top3 occupancy at promoter regions is preferentially found at non-transcribed and weakly transcribed genes ( Figure S1C ) . Looking at individual genes , we identified 233 5′IGRs that display a >1 . 5-fold and 455 3′IGRs that display a >2-fold average relative enrichment of Top3 . These overlap significantly ( p<0 . 001 ) with 5′ and 3′IGRs that are enriched for Top1 and Top2 ( p<0 . 001 for all pair wise comparisons ) ( Figure 3C ) . Thus , similar to Top1 and Top2 , Top3 preferentially binds to intergenic regions along the euchromatic chromosome arms . In addition , Top3 is enriched at sub-telomeres and centromeres . S . pombe Top1 and Top2 have previously been implicated in various stages of transcription [30]–[31] . We investigated the genome-wide transcription levels in the top3-105 mutant by total RNA extraction , reverse transcription and hybridization to tiling microarrays . After 8 hours at 36°C , there is a small reduction in the genome-wide average RNA levels from ORFs in the top3-105 mutant compared to wild type ( Figure 4 ) . However , after filtering out non-transcribed genes ( AU<100 in both wild type and the top3-105 mutant ) we only identified 118 ( 106 protein coding ) genes that displayed a >1 . 5-fold decrease of RNA levels in the top3-105 mutant . Moreover , there are also 299 ( 211 protein coding ) genes that display a >1 . 5-fold increase of RNA levels in the top3-105 mutant . Thus , the top3-105 mutant displays minor changes in RNA levels , associated with both up- and down-regulation of gene transcription . Top3 occupancy at centromeres is most pronounced at the central domain , consisting of the central core ( cnt ) and the innermost repeat ( imr ) regions , where there is 4–8 fold enrichment of Top3 relative to the rest of the genome ( Figure 5A and Figure S2 ) . Interestingly , both Top1 and Top2 occupancies are relatively low at this region . The central domain , and particularly the cnt region , is also highly enriched for CENP-ACnp1 . Centromeric CENP-ACnp1 occupancy displays a strong positive correlation with Top3 occupancy , but no correlation with Top2 and a weaker positive correlation with Top1 occupancy ( Figure 5B ) . In fact , there is a subset of probes forming a separate cluster in the scatter plot for which there is a unique positive correlation with Top3 enrichment as opposed to Top1 and Top2 enrichment . This cluster corresponds to the majority of the cnt probes , for which the relative enrichment of CENP-ACnp1 is distinctively high . In S . pombe , it has previously been shown that CENP-ACnp1 also localizes to gene promoters [60] . 5′IGRs that display a >3-fold average relative enrichment of CENP-ACnp1 significantly overlap with those that display a >1 . 5-fold average relative enrichment of Top3 ( p<0 . 001 ) ( Figure 5C ) . However , as expected from the similar bindings of Top1 and Top2 to promoter regions , there are also significant overlaps with those that display a >1 . 5-fold average relative enrichment of Top2 and Top1 , respectively ( p<0 . 001 for pair-wise comparisons ) . Moreover , average CENP-ACnp1 occupancies at these 5′IGRs display a weaker positive correlation with Top3 occupancy , as well as with Top2 and Top1 occupancies ( Figure 5D ) . Thus , Top3 co-localizes with CENP-ACnp1 at both centromeres and 5′IGRs , but centromeres are distinctive in that they display a unique correlation between Top3 occupancy and high levels of CENP-ACnp1 occupancy , which is not seen for Top1 and Top2 . Next , we used ChIP-chip to investigate the genome-wide relative enrichment of CENP-ACnp1 in the top3-105 mutant after 8 hours at 36°C . There is a small increase in the genome-wide average relative enrichment of CENP-ACnp1 at promoter regions in the top3-105 mutant compared to wild type ( Figure 6A ) . However , there is no difference in the average relative enrichment of CENP-ACnp1 at the 198 5′IGRs that display a high ( >3-fold ) relative enrichment of CENP-ACnp1 in wild type ( Figure 6B ) . Among these sites , there are 13 that display a >1 . 5-fold increase and 54 that display a >1 . 5-fold decrease in the relative enrichment of CENP-ACnp1 in the top3-105 mutant . Moreover , we identified 85 5′IGR that are not enriched for CENP-ACnp1 in wild type , but display a >1 . 5-fold increase resulting in a rather high ( >1 . 5-fold ) relative enrichment of CENP-ACnp1 in top3-105 mutant . Thus , there is both an overall increase in the average levels and a redistribution of CENP-ACnp1 at promoter regions in the top3-105 mutant . At centromeres , where there is a unique overlap between Top3 and CENP-ACnp1 occupancy , there is a consistent increase in CENP-ACnp1 enrichment at the central domains in the top3-105 mutant compared to wild type after 8 hours at 36°C ( Figure 6C and Figure S3 ) . This does not seem to be due to indirect effects by changes in transcription , since no genes known to be involved in CENP-ACnp1 assembly or disassembly were found among those that displayed >1 . 5-fold increase or decrease in RNA levels in the top3-105 mutant ( Table S2 ) . We also investigated the total amounts of CENP-ACnp1 protein by acid extraction of histones from chromatin and immunoblotting using strains expressing FLAG epitope tagged CENP-ACnp1 from the endogenous locus . In agreement with the ChIP data , there is an increase in the total levels of CENP-ACnp1 protein associated with chromatin in the top3-105 mutant compared to wild type ( Figure S4 ) . The soluble fraction of CENP-ACnp1 is small in both wild type and mutants , but may be slightly higher in the mutant . Using ChIP and qPCR we confirmed that there is an increase in the relative enrichment of CENP-ACnp1 at the cnt region of chromosome I in the top3-105 mutant compared to wild type after 8 hours at 36°C ( p = 0 . 0030 ) ( Figure 6D ) . This phenotype is found also in the rqh1Δ mutant and the top3Δ rqh1Δ mutant ( p = 0 . 00026 and p = 0 . 045 , respectively ) . The rad51Δ single mutant on the other hand displays similar levels of CENP-ACnp1 compared to wild type in this region . Moreover , deletion of rad51 does not suppress the increase in CENP-ACnp1 enrichment in the top3-105 and rqh1Δ mutants , as the rad51Δ top3-105 and rad51Δ rqh1Δ double mutants still display increased relative enrichment of CENP-ACnp1 compared to wild type at the cnt region of chromosome I ( p = 0 . 00035 and p = 0 . 037 ) ( Figure 6D ) . Thus , Top3 and Rqh1 normally limit the levels of CENP-ACnp1 by a mechanism that is largely independent of their role in HR downstream of Rad51 . Next , we investigated whether the altered levels of CENP-ACnp1 is associated with changes in the relative enrichment of the CENP-ACnp1-specific chaperone HJURPScm3 by ChIP-qPCR using strains expressing Pk/V5 epitope tagged HJURPScm3 from the endogenous locus . Surprisingly , the enrichment of HJURPScm3 at the cnt region of chromosome I is reduced in the top3-105 , rqh1Δ and top3Δ rqh1Δ mutants compared to wild type after 8 hours at 36°C ( p = 0 . 028 , p = 0 . 24 and p = 0 . 011 , respectively ) ( Figure 6E ) . Thus , Top3 and Rqh1 have the opposite effect on the levels of HJURPScm3 compared to the levels of CENP-ACnp1 in this region . Last , we tested if the altered levels of CENP-ACnp1 at centromeres were associated with altered levels of histone H3 in this region . However , there was no significant difference in the relative enrichment of histone H3 at the cnt region of chromosome I in the top3-105 mutant . Thus , Top3 and Rqh1 affect centromeric chromatin in a way that specifically limits the levels of CENP-ACnp1 and promotes association of HJURPScm3 at central domains .
The top3-105 mutant carries an A762G single base pair substitution resulting in a Tyr209Cys amino acid change . This residue is conserved in S . cerevisiae Top3 as well as in metazoan Top3α , and resides in an important region involved in conformational changes upon DNA binding [57] . Similar to other thermo-sensitive top3 mutants , the top3-105 mutant displays impaired growth , altered nuclear morphology and various defects in chromosome segregation soon after a shift to the restrictive temperature [46] , [53] . Deletion of top3 is known to cause very severe nuclear and mitotic defects resulting in cell death after just a few generations [51]–[52] Since the lethality can be suppressed by deletion of rqh1 it is hypothesized that these extreme nuclear defects are caused by rapid accumulation of unresolved Rqh1-dependent HR intermediates that prevent chromosome segregation and ultimately causes lethality . However , both the rqh1Δ single mutant and the top3Δ rqh1Δ double mutant still display moderate mitotic defects , indicating that there may be additional roles for the Top3-Rqh1 complex in chromosome segregation . Rad51 has previously been shown to be important for normal mitotic chromosome segregation and the rad51Δ mutant displays moderate defects in chromosome segregation [58] . Deletion of rad51 also suppresses the severe chromosome segregation defects of the top3-105 mutant , but the top3-105 rad51Δ double mutant also still display moderate defects in chromosome segregation . Deletion of rad51 has no effect on the mitotic defects in the rqh1Δ mutant , as the rqh1Δ rad51Δ double mutant display similar levels of chromosome segregation defects as the rqh1Δ and rad51Δ single mutants . Thus , it seems that Top3 , Rqh1 and Rad51 are all important for normal mitotic chromosome segregation . The genome-wide localization of Top3 reveals high relative enrichment at IGRs , towards subtelomeric regions and at centromeres . The binding pattern for Top3 at IGRs is similar to those of Top1 and Top2 , possibly indicating that all tree topoisomerases have overlapping functions at promoters and TTSs , such as maintenance of 5′ and 3′ nucleosome depleted regions ( NDRs ) important for transcription [30]–[31] . However , we did not find any major effects on transcription for the top3-105 mutant , and Top3 association at promoters is mostly seen at non-transcribed genes . The relative enrichment of Top3 is also high towards subtelomeric regions , including the rDNA clusters at tel3L and tel3R . This is in agreement with the role of S . pombe Top3 and Rqh1 in replication recombination at rDNA and telomere repeats [46]–[47] , [61] . Surprisingly , Top3 is also enriched at centromeres and particularly at centromeric central domains , where there is no pronounced enrichment of Top1 and Top2 . Thus , Top3 is the major topoisomerase present at centromeric central domains and may have a unique function in this region . The chromatin structure found at centromeric central domains is unique in that it contains high levels of the histone H3 variant CENP-ACnp1 . Interestingly , Top3 occupancy displays a unique positive correlation with CENP-ACnp1 occupancy in these regions , and especially at the cnt regions where the enrichment of CENP-ACnp1 is particularly high . This led us to investigate if Top3 has an effect on CENP-ACnp1 nucleosomes in this region . Intriguingly , the relative enrichment of CENP-ACnp1 at central domains is increased both in the top3-105 mutant , the rqh1Δ mutant and the top3Δ rqh1Δ double mutant . This suggests that the activity of the Top3-Rqh1 complex normally limits the levels of CENP-ACnp1 in this region . Intriguingly , the altered structure and/or dynamics of centromeric CENP-ACnp1-containing chromatin may contribute to the chromosome segregation defects seen in the top3-105 , rqh1Δ and top3-105 rqh1Δ mutants . However , this new role is clearly not the cause of the extremely severe chromosome segregation defects and lethality in the top3Δ mutant , as the top3Δ rqh1Δ mutant is viable while still retaining this phenotype . As previously described , the lethality of the top3Δ mutant likely depends on accumulation of unresolved Rqh1-dependent recombination intermediates , which are probably independent of changes in centromeric chromatin . Interestingly , CENP-ACnp1 has also been shown to be associated with gene promoters [60] . Like for centromeric CENP-ACnp1 , there is an overall increase in the relative enrichment of CENP-ACnp1 at promoter regions in the top3-105 mutant . In addition , there is a partial redistribution of non-centromeric CENP-ACnp1 . Thus , it is clear that Top3 also affects the dynamics of CENP-ACnp1 outside of centromeres . However , since Top3 , Top2 and Top1 are all enriched at promoter regions , the dynamics of CENP-ACnp1 outside centromeres is likely to depend on all three topoisomerases , making the situation complex . The most established role of the Top3-Rqh1 complex is in HR , where they act downstream of Rad51 . One possibility is that the effects on CENP-ACnp1 enrichment relates to the role of Top3 and Rqh1 in this pathway . Rad51 is required for accurate chromosome segregation and has previously been shown to suppress chromosomal rearrangements at centromeres [58] , [62] . Moreover , it was recently hypothesized that HR could be involved in higher-order organization of S . pombe centromeres [63] . However , increased relative enrichment of CENP-ACnp1 is still seen in the top3-105 rad51Δ and rqh1Δ rad51Δ double mutants , but not in the rad51Δ single mutant . Therefore , the role of the Top3-Rqh1 complex in limiting the levels of CENP-ACnp1 seems largely independent of Rad51-dependent HR . Chromosome segregation defects are on the other hand also seen in the rad51Δ single mutant , indicating that HR may somehow be important for proper chromosome segregation . Thus , chromosome segregation defects in top3 and rqh1 mutants could originate both from defects in HR and from altered levels of CENP-ACnp1 at centromeres . Another possibility is that the Top3-Rqh1 complex has an indirect effect on CENP-ACnp1 dynamics due to altered transcription of genes involved in CENP-ACnp1 nucleosome dynamics . However , we did not find any significant change in transcription for any of the genes currently known to be involved in CENP-ACnp1 dynamics . A third possibility is that Top3 and Rqh1 affects the stability of the CENP-ACnp1 protein . The total amount of CENP-ACnp1 present in cells is increased in the top3-105 mutant compared to wild type . However , this seems to mostly relate to an increase in the total levels of CENP-ACnp1 associated with chromatin , while the amount of soluble CENP-ACnp1 is low in both wild type and mutant . In budding yeast , it has been shown that the amount of soluble CENP-A is tightly controlled by rapid proteolysis , while nucleosome assembly at centromeres stabilizes the protein [64] . In agreement , reduced levels of CENP-ACnp1 at centromeres have previously been associated with a reduction of the total CENP-ACnp1 protein levels found in cells [65] . Thus , increased protein levels of CENP-ACnp1 is in agreement with increased levels of CENP-ACnp1 nucleosomes at centromeres , and it seems less likely that impaired Top3 function would stabilize soluble CENP-ACnp1 . Instead , we suggest that Top3 together with Rqh1 affect the assembly and disassembly of CENP-ACnp1 nucleosomes by regulating centromere DNA topology . Nucleosome assembly and disassembly are intimately linked to DNA topology and DNA topoisomerases have previously been shown to affect the assembly of canonical H3 nucleosomes [27]–[31] . In S . pombe , Top1 and Top2 have been implicated in disassembly of H3 nucleosomes at 5′ and 3′ IGRs [30]–[31] . Here , removal of negative supercoiling by Top1 and Top2 is hypothesized to stimulate nucleosome disassembly mediated by the chromatin remodeler Hrp1 . Centromeres are likely to be topologically constrained regions where nucleosome dynamics are highly dependent on DNA topology and topoisomerases . Since Top3 is highly enriched at central domains as compared to the other topoisomerases , nucleosome dynamics in this region likely depends particularly on Top3 . Top3 is unique in that it preferentially removes negative supercoiling . Thus , the activity of Top3 should have a negative effect on assembly of CENP-ACnp1 nucleosomes with left-handed wrapping of DNA . This could be a way of controlling and fine-tuning the assembly of CENP-ACnp1 nucleosomes mediated by factors such as HJURPScm3 . In the top3-105 mutant , a shift towards a state of more negative supercoiling would result in increased stability and facilitated assembly of CENP-ACnp1 nucleosomes , altering the dynamics of CENP-ACnp1 nucleosomes . In agreement with increased relative enrichment of CENP-ACnp1 also in the rqh1Δ mutant , efficient relaxation of negative supercoils by Top3 has been shown to be dependent on RecQ helicases [41] , [66] . HJURPScm3 associates with centromeric chromatin during most of the cell cycle , independently of CENP-ACnp1 , but dissociates from centromeres right after assembly of newly synthesized CENP-ACnp1 in the G2 phase of the cell cycle [8]–[9] , [67] . Reduced levels of HJURPScm3 at centromeres in the top3-105 and rqh1Δ mutants may thus reflect facilitated loading of CENP-ACnp1 from the pre-nucleosomal complex onto centromeric DNA and a more rapid dissociation of HJURPScm3 . In addition to nucleosome dynamics , DNA topology is also known to be important for transcription . Recent studies have shown that transcription is permissive also at the CENP-ACnp1 containing centromeric central domains in fission yeast [60] . Although the exact function is unclear , carefully modulated centromeric transcription has been suggested to play a role in formation of kinetochores as well as assembly of CENP-ACnp1 nucleosomes [68] . In support , factors known to be important for transcription have been implicated in CENP-ACnp1 assembly [65] , [69]–[70] . Thus , it is possible that Top3 as the main DNA topoisomerase at centromeric central domains affects transcription-coupled CENP-ACnp1 assembly . However , this is not mutually exclusive with a direct effect on CENP-ACnp1 nucleosome assembly . The molecular architecture of CENP-A nucleosomes is a subject of debate . Some studies suggest that CENP-A , H4 , H2A and H2B form hemisomes with right-handed wrapping of DNA [20] , [22]–[24] . In this case , due to opposite wrapping of the DNA double helix , preferential relaxation of negative supercoils by Top3 should increase the stability of CENP-ACnp1 hemisomes . In this model , impaired Top3 activity in the top3 and rqh1 mutants may result in a shift from assembly of right-handed CENP-ACnp1 hemisomes toward assembly of left-handed octameric CENP-ACnp1 nucleosomes at centromeres , thus giving increased levels of CENP-ACnp1 . Such a structural transition has previously been observed upon ectopic incorporation of S . cerevisiae CENP-ACse4 at non-centromeric loci [24] . Recent studies have also suggested that CENP-A nucleosomes may cycle between octameres and tetramers during the cell cycle [25]–[26] . In this case , the Top3-Rqh1 complex may affect one or both species . Interestingly , it was also shown human HJURP and budding yeast Scm3 associates with centromeres specifically during formation of CENP-A hemisomes [25]–[26] . Thus , reduced levels of Scm3 at centromeres in top3 and rqh1 mutants can thus also be reconciled with a structural change in CENP-A chromatin during some part of the cell cycle . Moreover , the fact that CENP-ACnp1 levels are increased , while H3 levels remains the same in the Top3 mutant , would be consistent with a structural change specific for CENP-ACnp1 nucleosomes . In conclusion , we found that S . pombe Top3 displays a unique enrichment at centromeres where it affects centromeric chromatin in a way that limits the levels of CENP-ACnp1 . We suggest that the Top3-Rqh1 complex has an important role in regulating centromeric DNA topology , thereby affecting CENP-ACnp1 nucleosome dynamics and perhaps the structure CENP-ACnp1 nucleosomes . Thus , the Top3-Rqh1 complex may be part of the intricate network of factors and pathways that comes together to carefully regulate CENP-ACnp1 nucleosome dynamics . This function could contribute to the observed chromosome segregation defects in top3 and rqh1 mutants .
Standard procedures for genetic manipulation and growth of S . pombe were used [71] . The strain expressing Pk/V5 epitope tagged HJURPScm3 was a gift from Professor M . Yanagida , the thermo-sensitive top3-105 mutant and the rad51Δ mutant were acquired from the Yeast Genetic Resource Center ( YGRC ) . The rqh1Δ mutant was a gift from S-W . Wang . Strains used in this study are listed in Table S3 . Genomic DNA was isolated as previously described , except that cell walls were digested with 0 . 4 mg/ml zymolyase 100T ( USBiological ) and RNA was digested with 10 µg/ml RNase ( Roche ) for 60 minutes at 37°C [71] . The top3 open reading frame was amplified by PCR and the purified PCR product was sent to Eurofins MWG Operon for custom DNA sequencing . Primers used are listed in Table S4 . Protein sequence alignments were generated using ClustalW2 ( EMBL-EBI , http://www . ebi . ac . uk/Tools/msa/clustalw2/ ) . Cells grown to log-phase at 25°C were spotted in 5-fold serial dilutions and incubated at 25°C , 30°C or 36°C . Cells were grown to early log-phase at 25°C and then shifted to 36°C for 9 hours , while determining cell density hourly using a microscope counting chamber . Immunofluorescence microscopy was performed as previously described with the following details and alterations [72] . Cells were grown to mid-log phase first at 25°C and then at 36°C for 8 hours . Cells were fixed in 3 . 7% formaldehyde and 0 . 2% glutaraldehyde for 60 minutes before digestion with 0 . 5 mg/ml zymolyase 100T ( US Biological ) for 70 minutes at 37°C . Cells were incubated with 1∶80 dilution of TAT1 mouse anti-tubulin serum ( a gift from K . Gull ) over night and then with 1∶100 dilution of FITC-conjugated goat anti-mouse ( F-1010 , Sigma ) over night . Cells were stained with 0 . 2 µg/ml 4′ . 6-diamidino-2-phenylindole ( DAPI ) and mounted on poly-L-lysine-coated microscopy slides ( LabScientific ) using Vectashield mounting media ( Vector laboratories ) . Cells were examined by fluorescence microscopy using a Zeiss Axioplan 2 ( Carl Zeiss ) equipped with a Plan-Neofluar 63X/1 . 25 PH3 oil objective ( Carl Zeiss ) , an ORCA-100 CCD camera ( Hamamatsu Photonix ) and Openlab 5 . 0 . 2 software ( Improvision ) . P values for comparing chromosome segregation defects were generated using a two-tailed Fishers exact test . ChIP was performed as previously described with the following details and alterations [73] . Cells were grown to mid-log phase at 30°C or first at 25°C and then at 36°C for 8 hours . Cells were lysed using a FastPrep-24 homogenizer ( MP Biomedicals ) with seven 30 second pulses at 6 . 5 m/s . Chromatin was fragmented using a Vibra-cell VCX 130 sonicator ( Sonics ) equipped with a 2 mm stepped microtip , set to 40% amplitude with 10 second pulses and 15 second pauses for two minutes . Chromatin fragments were immunoprecipitated with 1 µg 9E10 mouse anti-Myc ( M4439 Sigma ) , 3 µg rabbit anti-H3 ( Ab1791 Abcam ) , 10 µl rabbit anti-Cnp1 antiserum ( a gift from R . Allshire ) or 3 µg mouse anti-V5 ( MCA1360 Serotec ) . DNA was recovered using QIAquick PCR Purification with Buffer PB ( Qiagen ) . For analysis on GeneChip S . pombe Tiling 1 . 0FR Arrays ( Affymetrix ) 5 mM dUTP was added to the second round of DNA amplification . Fragmentation , labelling and hybridization were performed by the Affymetrix core facility at Karolinska Institiutet ( BEA ) using standard protocols ( http://www . affymetrix . com ) . Cells were grown to mid-log phase first at 25°C and then at 36°C for 8 hours . Total RNA was extracted using hot acid-phenol and chloroform . Reverse transcription , labeling , fragmentation and hybridization to GeneChip S . pombe Tiling 1 . 0FR Arrays ( Affymetrix ) was performed by BEA . Raw data ( . CEL format ) was analyzed using Affymetrix Tiling Analysis Software ( TAS ) v1 . 1 . One sample analysis and linear scaling was used for RNA samples . Two-sample comparison of immunoprecipitated and input samples , with linear scaling and separate quantile normalization was used for ChIP samples . Probe signals were generated using a bandwidth of 100 and assigned to S . pombe genome coordinates ( Sanger 2007 and Sanger 2004 for centromeres ) . Browser images were generated using Integrated Genome Browser ( IGB ) ( Affymetrix ) and PodBat ( www . podbat . org ) . Podbat was used for averaging signals across ORFs and IGRs , for comparing gene lists and for moving averages ( bandwidth 100 probes , step size 20 probes ) after alignment of genes at the TSS and TTS based on previous annotations [74] . For statistical analysis , a t-test was performed for each bin under the null hypothesis that there is no difference between two sets . 5′IGRs and 3′IGRs were defined as 500 bp upstream or downstream of the ORF or up until the neighbouring gene if shorter . CENP-ACnp1 , Top1 and Top2 ChIP-chip and wild type transcription at 30°C raw data ( . CEL format ) are from previous studies [30] , [75] . The microarray data from this publication have been submitted to the GEO database [http://www . ncbi . nlm . nih . gov/geo/] and assigned the accession number GSE44206 . Quantitative real-time PCR ( qPCR ) was performed using the 7500 Fast Real-Time PCR System ( Applied Biosystems ) and the associated Sequence Detection Software v . 1 . 3 ( Applied Biosystems ) . ChIP relative enrichment was calculated using the ddCt method , normalizing to ChIP input ( dCt ) and to a control locus ( ddCt ) ( act1 ) . For each experiment all samples were also normalized to the average of the wild type . Standard deviations were calculated for the total of six samples from two independent experiments . P values were generated using an unpaired two-sample t-test . Primers used are listed in Table S4 .
|
Centromeres are unique regions on eukaryotic chromosomes that are essential for chromosome segregation at mitosis and meiosis . Centromere identity and function depends on the presence of specialized chromatin with nucleosomes containing the centromere-specific histone H3 variant CENP-A . Assembly and disassembly of nucleosomes have previously been shown to involve a family of enzymes known as DNA topoisomerases . We show that centromeres are unique in that they are associated with high levels of Top3 , but low levels of Top1 and Top2 , suggesting that Top3 is particularly important for centromeric DNA topology . Impaired function of Top3 or its partner Rqh1 results in chromosome segregation defects and increased levels of CENP-ACnp1 at centromeres . This role in limiting the levels of CENP-ACnp1 at centromeres is independent of the established role for the Top3-Rqh1 complex in homologous recombination . Therefore , we hypothesize that the Top3-Rqh1 complex exerts this effect by regulating centromere DNA topology , which in turn affects CENP-ACnp1 nucleosome dynamics . Specific removal of negative supercoiling by Top3 could directly have a negative effect on assembly of CENP-ACnp1 nucleosomes with left-handed negative wrapping of DNA and/or act indirectly by inhibiting transcription-coupled CENP-ACnp1 assembly . Alternatively , Top3 may be a factor that promotes formation of CENP-ACnp1 hemisomes with right-handed wrapping of DNA over conventional octamers . This suggests a new role for the Top3-Rqh1 complex at centromeres and may contribute to the understanding of the structural and functional specification of centromeres .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"genetics",
"epigenetics",
"biology",
"chromatin",
"genetics",
"and",
"genomics"
] |
2013
|
DNA Topoisomerase III Localizes to Centromeres and Affects Centromeric CENP-A Levels in Fission Yeast
|
Efficient infection control requires potent T-cell responses at sites of pathogen replication . However , the regulation of T-cell effector function in situ remains poorly understood . Here , we show key differences in the regulation of effector activity between CD4+ and CD8+ T-cells during skin infection with HSV-1 . IFN-γ-producing CD4+ T cells disseminated widely throughout the skin and draining lymph nodes ( LN ) , clearly exceeding the epithelial distribution of infectious virus . By contrast , IFN-γ-producing CD8+ T cells were only found within the infected epidermal layer of the skin and associated hair follicles . Mechanistically , while various subsets of lymphoid- and skin-derived dendritic cells ( DC ) elicited IFN-γ production by CD4+ T cells , CD8+ T cells responded exclusively to infected epidermal cells directly presenting viral antigen . Notably , uninfected cross-presenting DCs from both skin and LNs failed to trigger IFN-γ production by CD8+ T-cells . Thus , we describe a previously unappreciated complexity in the regulation of CD4+ and CD8+ T-cell effector activity that is subset-specific , microanatomically distinct and involves largely non-overlapping types of antigen-presenting cells ( APC ) .
Infection results in the priming of pathogen-specific T-cell responses in LNs draining the site of infection . Depending on the nature of the pathogen , this critical step in generating adaptive immunity involves the interaction of naive T cells with various types of migrating and LN-resident DCs [1] , [2] . During skin infection with herpes simplex virus ( HSV ) -1 , LN-resident CD8α+ DCs and skin-derived CD103+ DCs can activate naïve CD8+ T-cells , presumably through the cross-presentation pathway involving the acquisition of noninfectious antigen [1]–[4] . By contrast , all subsets of skin-derived migratory DCs , including epidermal Langerhans cells , dermal CD11b+ and dermal CD103+ DCs , in addition to LN-resident CD8α+ DCs acquire the ability to stimulate naive HSV-specific CD4+ T cells [1] , [2] , [4] . Following appropriate activation by DCs , T cells undergo a program of clonal expansion , which is accompanied by the acquisition of effector functions and the induction of migration molecules that facilitate their infiltration of infected tissues . While CD4+ helper T cells support the generation of antibody and CD8+ T-cell responses in lymphoid tissues , both CD4+ and CD8+ T-cells also contribute directly to pathogen control at sites of infection [5] , [6] . The latter is achieved through two principle effector functions: the contact-dependent elimination of infected tissue cells and the local production of inflammatory and antimicrobial cytokines [5] , [6] . The extent to which these T-cell activities contribute to immunity depends on the nature of the infection . For instance , control of non-cytopathic viruses , such as lymphocytic choriomeningitis virus , strictly requires cytolytic T-cell activity [7] . By contrast , immunity against cytolytic viruses , such as vaccinia and vesicular stomatitis virus , does not rely on target cell elimination by T cells [8] . Instead , under circumstances where infection will ultimately result in lytic cell death regardless of T-cell killing , pathogen containment and clearance is dependent on the production of cytokines by effector CD4+ and CD8+ T cells [9]–[11] . Together these diverse effector T-cell ( TEFF ) activities are essential for efficient immune protection , however , they may also cause the destruction of uninfected tissues , as seen in the context of immunopathology , autoimmunity or transplant rejection . Therefore , a detailed understanding of T-cell-mediated immunity in peripheral tissues forms an essential basis for therapeutic interventions to modulate T-cell responses against both harmful and innocuous antigens . Nevertheless , the cellular and molecular mechanisms controlling T-cell effector activities in nonlymphoid organs remain poorly defined [2] , [10] . At its simplest , T-cell effector functions are regulated by T-cell receptor ( TCR ) stimulation through peptide-MHC complexes on APCs . Importantly in this respect , disengagement of the TCR from antigen-MHC complexes results in the immediate cessation of T-cell cytokine production [10] , [12] . This “on-off cycling” of effector activity provides a sophisticated level of antigen specificity and places important temporal and spatial constraints on TEFF-cell responses [10] . As a consequence , effector T cells circulating through the blood or uninfected tissues are thought to shutdown cytokine production and to regain this effector function only upon reencounter with antigen in infected tissues [10] . In addition , noncognate signals delivered through inflammatory mediators and costimulatory molecules , such as interleukin ( IL ) -18 , IL-12 , type I IFNs or CD80 and CD86 , may also trigger or further modulate T-cell cytokine production and cytotoxic activity [13]–[16] . Thus , the presence of appropriate APCs providing antigen stimulation together with accessory signals is critical in regulating T-cell immunity in situ and targeting effector activities to pathogen-containing tissues [2] , [17] . Indeed , various types of professional and nonprofessional APCs , including monocyte-derived inflammatory DCs , B cells , neutrophils and parenchymal cells , have been suggested to elicit T-cell effector functions within nonlymphoid tissues [15] , [18]–[21] . Key aspects in this regulation , however , particularly those pertaining to the infection status of APCs and the role of distinct APC subtypes in driving CD4+ versus CD8+ TEFF-cell responses , remain poorly understood . Here , we define the cellular interactions that control TEFF-cell activity during the course of skin infection with HSV-1 . We focus our analysis on the production of IFN-γ , a central component of adaptive immune responses . IFN-γexerts proinflammatory and regulatory effects on a variety of target cells , including the stimulation of antimicrobial activity and the induction of MHC molecules and inflammatory chemokines [5] , [6] . Protection from HSV infection strictly requires TEFF-cell activities , with both CD4+ and CD8+ T cells contributing to virus control in skin , mucosa and sensory ganglia [22]–[26] . Moreover , efficient immunity against HSV infection requires IFN-γ [27] and , interestingly , it has been proposed that production of this cytokine rather than cytolytic activity is the major CD8+ T-cell mechanism for virus control in neuronal tissues [28] , [29] as well as during lytic infection in genital mucosa [26] . The shared role of IFN-γ as a key effector molecule produced by both CD4+ and CD8+ T cells allowed us to directly compare the regulation of these TEFF-cell subsets side by side . We further took advantage of the tropism of HSV-1 for epithelial tissues [30] to document a distinct anatomical distribution of IFN-producing TEFF-cell subsets in relation to the presence or absence of infectious virus in different microanatomical compartments . Importantly , this unexpected spatial segregation of TEFF-cell effector activity was a direct result of the involvement of largely non-overlapping subsets of professional and nonprofessional APCs in driving CD4+ and CD8+ TEFF-cell responses .
To determine population kinetics and cytokine production by TEFF cells in lymphoid and peripheral tissues , we utilized a skin infection with HSV-1 in combination with adoptive transfer of TCR-transgenic T cells specific for determinants derived from the HSV glycoproteins gB ( CD8+ gBT-I cells ) [31] and gD ( CD4+ gDT-II cells ) [4] , respectively . Consistent with the tropism of HSV for epithelial tissues , immunofluorescence microscopy ( IFM ) of skin revealed that infection was largely confined to the epidermal layer and hair follicle epithelium ( Fig . S1A ) . Separation of epidermal and dermal tissue ( Fig . S1B , C ) revealed that HSV-specific TEFF cells began to infiltrate infected skin around 5 days post-infection , albeit with fewer cells in the smaller epidermal compartment ( Fig . 1A ) . TEFF-cell numbers peaked around 8 days after inoculation and declined thereafter ( Fig . 1A ) . To analyze the production of IFN-γ by TEFF cells in situ , we adopted protocols that facilitate intracellular cytokine staining following exposure to the Golgi inhibitor brefeldin A ( BFA ) , either in vivo after intravenous injection [15] , [32] and/or ex vivo immediately after tissue harvest and during enzymatic digestion [21] . Of note , in order to focus our analysis on IFN-γ production in situ , neither of these approaches involved overt restimulation with high concentrations of peptide antigen ex vivo . A considerable portion of gBT-I and gDT-II TEFF cells in the epidermis and dermis of infected skin produced IFN-γ 5–6 days post-infection ( Fig . 1B–D ) . Concomitant with clearance of infectious virus from skin [25] , IFN-γ production by TEFF cells ceased around day 7 , with virtually no IFN-γ+ TEFF cells present 8 days post-infection ( Fig . 1D ) . Similar kinetics of IFN-γ production were also observed for endogenous CD8+ and CD4+ TEFF cells ( Fig . S2A , B ) . While roughly equal portions of gDT-II cells produced IFN-γ in the epidermal and dermal layers of skin 5 days after infection , the fraction of IFN-γ+ gBT-I cells was approximately 3-fold higher in the epidermis as compared to the dermis ( Fig . 1C , D ) . Note , that dermal preparations contained hair follicles of epithelial origin and therefore also harbored some replicating virus ( Fig . S1A–C ) . The broader distribution of IFN-γ+ gDT-II cells in the skin also extended to lymphoid tissues , with skin-draining axillary and brachial LNs , but not spleen , containing an appreciable fraction of IFN-γ-producing gDT-II cells ( Fig . 1E , F ) . By contrast , IFN-γ+ gBT-I cells were virtually absent from all lymphoid tissues . Together , these results suggested a distinct anatomical distribution of IFN-γ-producing CD4+ and CD8+ TEFF cells in both peripheral and lymphoid tissues . To gain further insight into the microanatomical localization of IFN-γ-producing TEFF-cell subsets , we obtained skin tissue for IFM analysis . Staining of skin sections with anti-IFN-γ antibodies confirmed the presence of IFN-γ-producing cells during the acute phase of infection ( Fig . 2A , B ) , with both endogenous CD4+ and CD8+ ( Fig . S2C , D ) , as well as transgenic gBT-I and gDT-II TEFF cells ( Fig . 2C , D ) contributing to this response . Interestingly , although gBT-I TEFF cells were broadly distributed throughout the skin , IFN-γ+ gBT-I cells were strictly confined to the epidermis and hair follicle epithelium ( Fig . 2C ) . By contrast , the majority of IFN-γ+ gDT-II cells localized to the dermal layer , where they were found either in association with hair follicles or in considerable distance to the epithelium ( Fig . 2D ) . Thus , in contrast to the strict confinement of IFN-γ+ CD8+ TEFF cells to the epithelium , the dermal layer was the predominant site of the CD4+ T-cell IFN-γ response . The kinetics of IFN-γ production by TEFF cells suggested that the presence of infectious virus was likely to play a role in the induction of cytokine production . This was indeed the case , as gBT-I and gDT-II TEFF cells primed by HSV infection did not produce IFN-γ in non-specifically inflamed skin after treatment with 1-fluoro-2 , 4-dinitrobenzene ( DNFB ) ( Figs . 3A and S3A ) . Likewise , in vitro activated gBT-I TEFF cells transferred into HSV-infected mice lacking H-2Kb molecules ( H-2Kb−/− ) did not produce significant amounts of IFN-γ in the skin ( Figs . 3B , C and S3B ) . Transfer of activated TEFF cells was necessary as H-2Kb−/− mice cannot prime naïve gBT-I cells due to lack of the relevant MHC-I restriction element . IFN-γ production by transferred gBT-I TEFF cells was completely restored in similar experiments following HSV-1 infection of bone marrow chimeric mice in which H-2Kb molecules were expressed exclusively in radioresistant cells , but were absent from the radiosensitive hematopoietic compartment ( Figs . 3D and S3C ) . By contrast , we observed a significant reduction in the frequency of in vivo primed IFN-γ+ gBT-I cells in chimeric mice in which H-2Kb molecules were selectively missing from radioresistant cells , when compared with fully MHC-I-sufficient control chimeras ( Figs . 3E and S3D ) . Interestingly , the overall frequencies of IFN-γ+ gBT-I TEFF cells appeared to be increased in this particular experimental set-up , possibly related to altered immune activation thresholds in previously irradiated recipient mice . Regardless , these results indicated that presentation of viral antigens by radioresistant epithelial cells , such as keratinocytes , Langerhans cells [33] and dendritic epidermal T cells ( DETC ) [34] , was necessary and sufficient for optimal IFN-γ responses by gBT-I TEFF cells . IFN-γ production by gDT-II cells was also a consequence of antigen recognition , as in vitro activated gDT-II TEFF cells transferred into infected MHC-II-deficient mice ( I-A/E−/− ) failed to produce IFN-γ ( Figs . 3F , G and S3E ) . Transfer of activated gDT-II cells was necessary to overcome the inability of I-A/E−/− mice to support CD4+ T-cell priming . By contrast , gDT-II TEFF cells primed in vivo produced IFN-γ in chimeric mice in which only radiosensitive , but not radioresistant cells expressed MHC-II molecules , although we observed a moderate , yet significant reduction in the frequency of IFN-γ+ gDT-II cells in this situation ( Figs . 3H and S3F ) . These results implied that bone marrow-derived MHC-II+ APCs were largely responsible for presentation of viral antigens and eliciting cytokine production by CD4+ TEFF cells . Consistent with an involvement of MHC-II-expressing APCs in driving CD4+ TEFF-cell responses , MHC-IIhi cells accumulated in the skin during the first week post-infection [35] ( Fig . S4A , B ) . The majority of these cells had a CD11cintCD11b+ phenotype and further expressed CD64 and MAR-1 , identifying them as monocyte-derived inflammatory DCs [36] , [37] ( Fig . S4C ) . IFM of infected skin revealed a broad distribution of MHC-II+ cells , with the vast majority localizing to the dermal layer , where they were found in close proximity to IFN-γ+ gDT-II cells ( Fig . 4A ) . Indeed , partial depletion of CD11c+ DCs upon diphtheria toxin ( DT ) treatment of CD11c . DTR mice resulted in a significant reduction in the frequency of IFN-γ+ gDT-II cells in infected skin ( Figs . 4B , C and S5A ) . Furthermore , treatment of mice with antibodies blocking the costimulatory molecules CD80 and CD86 , typically expressed by professional APCs such as DCs , also abrogated the CD4+ TEFF-cell IFN-γ in skin and draining LNs ( Figs . 4D and S5B ) . In stark contrast , costimulation blockade had no bearing on IFN-γ production by gBT-I TEFF cells ( Fig . 4D ) . These experiments suggested that MHC-II+ DCs were the main drivers of cytokine production by CD4+ TEFF cells . Nevertheless , in Ccr2−/− mice , an absence of monocyte-derived DCs [20] , [38] , which numerically dominated the cutaneous DC network during acute infection , rather increased than decreased the frequency of IFN-γ+ gDT-II and gBT-I cells ( Fig . S5C , D ) , potentially related to impaired virus control in these mice [20] . Furthermore , we observed normal IFN-γ production by gDT-II and gBT-I TEFF cells in the absence of Langerhans cells , CD103+ dermal DC and CD8+ LN-resident DCs upon DT treatment of Langerin . DTR mice [39] , [40] , and similarly , also in B-cell-deficient μMT mice [41] ( Fig . S5E–G ) . These results suggested a level of redundancy regarding the involvement of different types of professional APCs in regulating CD4+ TEFF activities in infected skin . To more directly establish a role for DCs in CD4+ T-cell responses , we utilized an ex vivo stimulation assay in which APCs purified from infected mice were cocultured with in vitro generated TEFF cells . Initial experiments revealed that maximal IFN-γ production occurred 18 hours after antigen-dependent restimulation for gDT-II and 5 hours for gBT-I TEFF cells ( Fig . S6A , B ) . We purified CD11chi DCs from HSV-infected skin and divided them into CD11bhi and CD11blo subsets ( Fig . 5A ) , with the former expected to contain monocyte-derived and dermal DCs and the latter expected to contain Langerhans cells and CD103+ dermal DCs [4] , [37] , [40] . Notably , both subsets induced robust IFN-γ production by gDT-II TEFF cells , whereas monocytes ( CD11b+CD11c−Ly6Chi ) and neutrophils ( CD11b+CD11c−Ly6Cint ) failed to do so ( Fig . 5B , C ) . Unexpectedly , none of these APCs induced IFN-γ production by gBT-I TEFF cells ( Fig . 5B , C ) . This was not related to potentially compromised expression of H-2Kb molecules after APC isolation , since all subtypes triggered IFN-γ production by gBT-I TEFF cells when pulsed with high doses of gB-peptide prior to cell sorting ( Fig . S6C , D ) . In separate experiments , we specifically sorted CD103+ dermal DCs , as this DC subset is capable of cross-presenting viral antigens to CD8+ T cells during skin infection [4] . Once again , while both CD103+ and CD103− CD11chi DCs triggered IFN-γ production by gDT-II TEFF cells , neither of the two subsets activated gBT-I TEFF cells ( Fig . 5D ) . Remarkably , this disparate response also extended to DCs isolated from LNs draining the site of infection with CD103+ , CD11b+ , CD8α+ and Langerhans cell-containing CD103−CD11b−CD8α− subsets inducing IFN-γ production by gDT-II , but not gBT-I TEFF cells ( Fig . 5E ) . Of note , gBT-I TEFF cell unresponsiveness towards DC stimulation was observed irrespective of the culture period for 5–18 hours ( not shown ) . Together , these results demonstrated that various skin-derived and LN-resident DC subsets acquired and presented viral antigen for the activation of CD4+ TEFF cells . By contrast , none of these DCs elicited IFN-γ-production by CD8+ TEFF cells , even though at least some of them , namely the CD8α+ and CD103+ subsets , have the ability to present viral antigens for the activation of naïve CD8+ T cells [3] , [4] . Given that IFN-γ+ gBT-I cells were found exclusively in skin epithelium ( Figs . 1B–D and 2B ) , we reasoned that this compartment contained APCs capable of stimulating CD8+ TEFF cells . Therefore , we purified CD45 . 2+ hematopoetic and CD45 . 2− parenchymal and stromal cells from epidermal sheets of infected mice by cell sorting and tested their ability to activate CD8+ TEFF cells . Note that here we used expression of CD45 . 2 to distinguish between hematopoeitc and non-hematopoeitc epidermal cells , whereas in other analyses we used this molecule as a marker for CD45 . 1+CD45 . 2+ gBT-I and gDT-II cells . Both fractions induced IFN-γ production by gBT-I TEFF cells , although the keratinocyte-containing CD45 . 2− subset appeared to be slightly more potent in this regard ( Fig . 6A , B ) . Induction of IFN-γ production was antigen-specific , since co-cultured OT-I TEFF cells of an irrelevant specificity did not respond to either of the APC subsets ( Fig . S7A , B ) . Furthermore , stimulation of IFN-γ production by CD45 . 2− epidermal APCs was specific for CD8+ TEFF cells since these APCs failed to activate gDT-II TEFF cells ( Figs . 6B and S7C ) . In line with this , the vast majority of CD45 . 2− epidermal cells from infected skin lacked expression of MHC-II molecules ( Fig . S7D ) . To better define the nature of epidermal APCs capable of stimulating CD8+ TEFF cells , we sorted these cells into keratinocytes ( CD45 . 2−EpCAM+ ) , DCs ( CD45 . 2+CD11chi ) , DETCs ( CD45 . 2+Vγ3+ ) , as well as residual CD45 . 2+CD11c− cells . As expected , keratinocytes induced moderate levels of IFN-γ production by gBT-I TEFF cells , and so did epidermal DCs ( Fig . 6C , D ) , in contrast to their counterparts isolated from total skin preparations and LNs ( Fig . 5C–E ) . While residual CD45 . 2+ epithelial cells had only a weak stimulatory capacity , remarkably , DETCs were by far the most potent APCs triggering IFN-γ production by gBT-I TEFF cells ( Fig . 6C , D ) . Given that the epidermis was the predominant site of viral replication in vivo , we hypothesized that the stimulatory capacity of epidermal APCs resulted from their direct infection . In agreement , intravital two-photon microscopy of skin infected with a cyan fluorescent protein ( CFP ) -expressing HSV-1 strain revealed that slow-moving gBT-I TEFF cells were swarming around virally infected epidermal cells ( Movie S1 ) . By contrast , gBT-I TEFF cells more distal to infection foci displayed significantly higher mean velocities ( Fig . 7A , B ) . Importantly , using IFM , we observed that IFN-γ+ gBT-I cells co-localized with HSV-infected cells in the epidermis and hair follicle epithelium ( Fig . 7C ) . To further identify infected cells , we inoculated mice with a recombinant strain of HSV-1 expressing green fluorescent protein ( HSV . GFP ) and analyzed epidermal cells 5 days post-infection using flow cytometry . We observed small numbers of GFP-expressing cells amongst various populations of epidermal cells , including keratinocytes ( CD45 . 2−EpCAM+ ) , DETCs ( CD45 . 2+Vγ3+ ) , Langerhans cells ( CD45 . 2+EpCAM+MHC-IIhi ) , other DCs ( CD45 . 2+EpCAM−MHC-IIhi ) , as well as undefined CD45 . 2+MHC-II− cells ( Fig . 7D ) . As expected , GFP+ cells were absent after infection with the wild-type HSV-1 KOS strain . Next , we sorted GFP+ and GFP− DETCs , MHC-IIhi DCs and keratinocytes from epidermal sheets ( Fig . 7E ) and tested their ability to stimulate TEFF cells ex vivo . Strikingly , all GFP+ , presumably infected , APCs were able to trigger IFN-γ production by gBT-I TEFF cells ( Fig . 7F , H ) . In contrast , IFN-γ production was not elicited by their GFP− counterparts . Finally , DCs , but not DETCs , elicited IFN-γ production by gDT-II TEFF cells , irrespective of their GFP-expression status ( Fig . 7G , H ) . Together , these results indicated that various types of epidermal APCs induced cytokine production by CD8+ TEFF cells . Importantly , direct viral infection was a strict requirement for their stimulatory capacity . Overall , our data highlight a previously unappreciated complexity in the regulation of T-cell effector activity that was subset-specific , microanatomically distinct and involved largely non-overlapping subsets of professional and nonprofessional APCs for CD4+ and CD8+ T-cell responses ( Fig . S8 ) .
Our results highlight a stringent and complex regulation of TEFF-cell responses that targets effector activities strictly to the site of infection and related lymphoid tissues . Thus , IFN-γ production was limited to the time of acute infection and occurred in an antigen-dependent fashion , requiring in situ restimulation via peptide-MHC complexes on bone marrow-derived professional APCs for CD4+ TEFF cells and on directly infected tissue cells for CD8+ TEFF cells . Consistent with a critical involvement of DCs in peripheral CD4+ TEFF-cell responses , we observed a pronounced accumulation of monocyte-derived inflammatory DCs in infected skin . Such DCs are thought to exert multiple functions , including the local production of inflammatory mediators [42] , trafficking of antigen to lymph nodes [43] and replenishment of peripheral DC populations following the resolution of infection [35] , [44] . In addition , our experiments employing genetic approaches , ex vivo stimulation assays and costimulation blockade provide compelling evidence that DCs in inflamed skin play a key role in stimulating CD4+ T-cell effector activity . These results reinforce the concept that DCs regulate various aspects of peripheral T-cell responses [2] , [15] , [20] , [21] , [45] , [46] . In fact , the presence of DCs appeared to be essential for CD4+ TEFF-cell responses as non-professional APCs , such as keratinocytes and DETCs , largely lacked MHC-II expression and failed to elicit IFN-γ production in ex vivo assays . It is possible that certain APC subsets may dominate the regulation of peripheral CD4+ TEFF-cell activities , as suggested for CD11chiCD11bhi dermal DCs after skin injection of model antigens [21] or CCR2-depedent monocyte-derived inflammatory DCs during mucosal HSV-2 infection [20] . Nevertheless , our study revealed a considerable degree of redundancy in this regard , with various DC populations from skin and LNs displaying strong stimulatory capacities for CD4+ TEFF cells . These results imply that all DC subsets , relative to their abundance in infected skin , contribute to CD4+ TEFF activation in vivo . Consistent with this , we observed normal IFN-γ+ production in mice deficient in specific APC subsets , such as monocyte-derived DCs , Langerhans cells , CD103+ dermal DCs or B cells . According to our analysis , monocyte-derived inflammatory DCs are by far the most abundant DC subtype in HSV-infected skin [35] and therefore , may be the major drivers of peripheral CD4+ TEFF-cell responses during HSV-1 skin infection . Nevertheless , our results demonstrate that they may not be essential in this regard as other DC subsets may compensate for their absence . The IFN-γ response by CD4+ T cells occurred in both draining LNs and the epithelial and dermal layers of infected skin , including regions a considerable distance away from infection foci in the epithelium . This remarkably broad distribution echoes the diverse functions of CD4+ TEFF cells in infection control , ranging from the initiation of antibody class-switching in LNs to the regulation of inflammatory cell infiltration and activity as well as direct antimicrobial effects within infected tissues [5] . Interestingly in this respect , IFN-γ can exert long-range effects on target cells located as far as 80 µm from CD4+ TEFF-cell-APC conjugates , as recently shown for skin infection with Leishmania major [47] . Thus , DC-mediated CD4+ TEFF-cell activation in the dermis may be an essential component of the host defense that restricts infection to the skin epithelium and limits its spread after virus reemergence in sensory nerve endings . The importance of the CD4+ TEFF-cell response is further illustrated by the lack of CD8+ TEFF-cell IFN-γ production in the dermis , as shown here . Supporting this notion , CD4+ TEFF-cell responses are thought to dominate the clearance of HSV-1 from the skin [22] , [23] , most likely via antibody-independent functions such as direct inflammatory and antiviral activities [5] , [48] , [49] . In striking contrast to the stimulation requirements for CD4+ TEFF cells , we identified nonprofessional APCs , such as keratinocytes and DETCs , as the main drivers of IFN-γ production by CD8+ TEFF cells . According to our histological analysis , keratinocytes are the most abundant cell type in infected epidermis suggesting that they may be largely responsible for activating CD8+ TEFF cells . In addition , various types of inflammatory cells that infiltrate the epithelial layer during infection may contribute to this response . Importantly , both keratinocytes and DETCs are highly susceptible to direct infection by HSV-1 in vivo [30] , [50] . Indeed , their stimulatory capacity , and surprisingly also that of epidermal DCs , was strictly dependent on direct infection . DETCs are invariant γδ-T cells that form a dense network in the epidermis of mice and have been implicated in both innate and adaptive immune responses [51] . Interestingly , it has been speculated that human γδ-T cells may act as professional APCs capable of stimulating naïve CD4+ T cells [52] , although we did not find evidence supporting a similar role for DETCs , as they failed to activate CD4+ TEFF cells . Nevertheless , their contribution to CD8+ T-cell responses may be particularly relevant at lesion borders where high numbers of infected DETCs [50] could elicit strong IFN-γ responses required to curb the lateral spread of infection . In addition to DETCs , other epidermal-infiltrating T cells may also act as potent APCs for local CD8+ TEFF cells upon infection with virus [53] . A role for infected T cells and DCs in triggering local CD8+ TEFF-cell activity is further supported by our observation that chimeric mice , in which only radiosensitive APCs could activate CD8+ T cells , had a residual IFN-γ response . Despite the fact that all DC subsets could activate CD4+ TEFF-cells , only HSV-infected epidermal DCs were capable of activating CD8+ TEFF cell to produce IFN-γ . Strikingly , uninfected DCs from the same location failed to stimulate CD8+ TEFF cells , highlighting the importance of direct infection in determining the outcome of CD8+ TEFF-cell-DC interactions . Previous studies have suggested that various types of professional and nonprofessional APCs , including inflammatory DCs and neutrophils , can trigger IFN-γ production by CD8+ TEFF cells during pulmonary infection with influenza virus [15] , [18] . Given the ability of influenza virus to infect a broad range of target cells in addition to its primary tropism for lung epithelial cells [17] , it is tempting to speculate that in this situation , direct infection may also be required for CD8+ T-cell activation . Supporting this notion , a large number of inflammatory DCs and neutrophils from influenza virus-infected lungs express viral antigens , most likely as a consequence of direct infection [15] , [18] , and infected neutrophils display a far superior ability to elicit CD8+ TEFF-cell cytokine production than their uninfected counterparts [18] . One of the more surprising findings from our study was that uninfected DCs failed to elicit IFN-γ production by CD8+ TEFF cells , even though they had access to viral antigen and efficiently activated CD4+ TEFF cells . We have previously shown that LN DCs are able to activate naïve CD8+ T cells at various stages during skin infection , unequivocally demonstrating that they present viral antigens in the context of MHC-I molecules [3] , [4] , [54] . The CD8α+ and CD103+ DC subsets are of particular interest in this regard , because those DCs cross-present viral antigens in draining LNs 5 days after HSV-1 infection [4] , corresponding to the time point analyzed here . Despite this , CD103+ DCs from skin and LNs failed to trigger IFN-γ production by CD8+ TEFF cells . This finding parallels observations with CD8+ T-cell responses to migratory DCs after influenza virus infection , where naïve but not memory T cells proliferate in response to antigen presented on migratory DCs [55] . Although not directly addressed in our study , it is tempting to speculate that the inability of uninfected DCs to activate CD8+ TEFF cells may be a means to prevent their elimination by triggering cytotoxic effector functions . While T-cell killing of DCs has been proposed to represent a negative feedback regulation on T-cell priming [56] , [57] , it should be noted that prolonged antigen presentation is a common feature in a variety of infections [58]–[61] . Therefore , CD8+ T-cell-mediated elimination of DCs in vivo may be inefficient at best , which is also consistent with our preliminary data demonstrating the failure of gBT-I TEFF cells to lyse CD103+ DCs from LN of HSV infected mice during short-term co-culture . The DC-dependent production of IFN-γ by CD4+ TEFF cells observed in our study further supports this assumption . Of note in this respect , minute amounts of surface antigen can trigger T-cell cytotoxicity [62] , [63] . Furthermore , cross-presenting nonprofessional APC , such as liver sinusoidal endothelial cells , can drive CD8+ T-cell TNFα production during viral hepatitis [64] . Thus , it appears unlikely that quantitative differences in antigen presentation between directly infected and cross-presenting DCs alone can explain the CD8+ TEFF-cell unresponsiveness described here . In the light of this , and given that infected epidermal DCs were indeed able to trigger CD8+ TEFF-cell IFN-γ production , we speculate that direct infection may alter the functional status of DCs , for instance through interference with putative inhibitory pathways [65] , to allow for the activation of CD8+ TEFF cells . Such modulation of DC stimulation thresholds may be particularly relevant when low levels of peptide-MHC-I complexes are presented to CD8+ TEFF cells . Further studies will be required to elucidate the precise molecular mechanisms operating in DCs and/or T cells to prevent the activation of CD8+ TEFF cells by uninfected cross-presenting DCs . Overall , the stringent temporal , cellular and molecular constraints on TEFF-cell responses identified in our study are likely in place to prevent collateral damage and autoimmune inflammation initiated by TEFF-cell activation in tissues not involved in infection . Our results are compatible with a scenario where CD8+ T-cell responses are strictly focused on infected tissue compartments , whereas CD4+ responses may induce a more regional state of antimicrobial protection in tissues surrounding infection foci . Having identified dramatically distinct requirements for CD4+ and CD8+ T-cell effector activity , our study has provided novel insights into the regulation of cellular immune responses in nonlymphoid tissues . Such knowledge has the potential to guide the development of T-cell subset-specific approaches for therapeutic and prophylactic intervention in antimicrobial immunity and autoimmunity .
All experiments were done according to Australian NHMRC guidelines contained within the Australian Code of Practice for the Care and Use of Animals for Scientific Purposes and under approvals ID1112038 and ID1112345 from the University of Melbourne Animal Ethics Committee . C57BL/6 , B6 . SJL-PtprcaPep3b/BoyJ ( B6 . CD45 . 1 ) , gBT×B6 . CD45 . 1 ( gBT-I . CD45 . 1 ) , gBT-I . EGFP , gBT-I . DsRed , OT-I×B6 . CD45 . 1 ( OT-I . CD45 . 1 ) , gDT-II×B6 . CD45 . 1 ( gDT-II . CD45 . 1 ) , gDT-II . EGFP , C57BL/6Ji-Kbtm1N12 ( H-2Kb−/− ) , B6 . 129S2-H2dlAb1-Ea/J ( MHC-II−/− ) , CD11c . DTR , B6129 . CCR2 ( Ccr2−/− ) , Langerin . DTR . EGFP ( Lg . DTR . EGFP ) and μMT mice were bred in the Department of Microbiology and Immunology . gBT-I . CD45 . 1 and gDT-II . CD45 . 1 are F1 generation offspring expressing both CD45 . 1 and CD45 . 2 . HSV-1 KOS , KOS pCMV/EGFP Cre ( HSV . GFP ) and 17 HSVgDUL47ΔYFP ( HSV . CFP ) were grown and titrated as previously described [46] . HSV . GFP expresses an EGFP/Cre fusion gene under the control of the CMV-IE promoter from the intergenic space between UL3 and UL4 resulting in EGFP expression by infected cells . HSV . CFP was derived from HSV-1 gDUL47 ( strain 17 expressing YFP from UL47 and CFP-tagged gD protein ) [66] , from which the YFP was removed and sequences from UL47 restored . HSV . GFP and HSV . CFP were made by homologous recombination between parent viral genomes with appropriate transfer plasmids in 293A cells after which green/yellow fluorescence were selected for and against , respectively . Final recombinants were verified by PCR and sequencing of relevant parts of the genome after at least three rounds of plaque purification . Mice were irradiated with two of doses 550 cGy 3 hours apart followed by reconstitution with 5×106 T-cell-depleted donor bone marrow cells and treatment with purified anti-Thy1 antibody ( clone T24/31 . 7 from hybridoma supernatants ) 1 day later . Chimeric mice were allowed to reconstitute for at least 8 weeks before experiments . WT→I-A/E−/− chimeras were not treated with anti-Thy1 antibody and received 5×106 enriched splenic CD4+ T cells from wild-type mice 1 day and 4 weeks after irradiation . Mice were infected on their flanks with 1×106 plaque-forming units of HSV-1 , as previously described [25] . For DNFB treatment , 15 µL of 0 . 5% ( w/v ) DNFB was applied to flank skin , as previously described [67] . For depletion of CD11c+ cells , CD11c . DTR mice were injected with 200 ng diphtheria toxin ( DT ) or PBS intraperitoneally and intradermally 4 d post-infection . Langerin . DTR mice were injected with 500 ng DT or PBS control intraperitoneally . For costimulation blockade , mice were injected with 0 . 25 mg anti-CD80 ( 16-10A1 ) and -CD86 ( GL1 ) blocking antibodies or rat IgG2 ( B81-3 ) and IgG2a ( R35-95 ) control antibodies ( BD Pharmingen ) intraperitoneally 4 d post-infection . gBT-I and gDT-II cells were isolated from lymphoid tissues and gDT-II cells were further enriched by positive and negative selection using magnetic beads , as described previously [4] . 5×104 gBT-I or 1×104 gDT-II cells were transferred into naïve mice intravenously via the tail vein , respectively . For transfer of in vitro activated T cells , 1 . 5×106 transgenic cells were injected . Splenic gBT-I or OT-I cells were activated with peptide-pulsed splenocytes , as previously described [25] . Purified gDT-II cells were activated by co-culture for 5 days with 4×107 irradiated wild-type splenocytes pulsed with 10 µM of gD315–327 in the presence of 2 µg LPS ( Sigma ) . Cells were cultured in 20 ml RPMI 1640 ( Department of Microbiology and Immunology ) supplemented with 10% FCS ( CSL ) , 5 mM HEPES ( Gibco ) , 2 mM glutamine ( Gibco ) , 5×10−5 M 2-β-mercaptoethanol ( Sigma ) , antibiotics ( Gibco , CSL ) ( RP-10 ) and 4 µg lipopolysaccharide . On days 2–4 , respectively , all cultures were diluted 1∶2 in fresh medium containing 20 U/mL recombinant human IL-2 ( PeproTech ) . As indicated , skin tissue was chopped and digested with 3 mg/ml collagenase type 3 ( Worthington Biochemicals , USA ) and 5 µg DNase ( Roche , Germany ) for 90 min at 37°C . Alternatively , skin was digested in 2 . 5 mg/mL dispase II ( Roche ) diluted in PBS for 90 min at 37°C . Then , the epidermis and dermis were separated mechanically and epidermal sheets were incubated in trypsin/EDTA ( 0 . 25%/0 . 1% ) ( SAFC Biosciences ) , while the dermis was chopped and incubated in collagenase type 3 and DNase , as previously described [49] . When analyzing ex vivo IFN-γ production , 10 µg/mL Brefeldin A ( Sigma ) was included during each enzymatic digestion step . For some experiments , mice were additionally injected with 0 . 25 mg BFA intravenously 6 hours prior to sacrifice . Axillary lymph nodes of HSV-1-infected mice were desiccated with a scalpel blade and digested with continual mixing in RPMI 1640 containing 1 mg/mL collagenase type 3 and 2 µg/mL DNase for 20 min , prior to the addition of 600 µL 0 . 1 M EDTA and continual mixing for 5 minutes further . DCs were subsequently enriched for by magnetic beads , as previously described [4] . APC subsets were stained with the appropriate monoclonal antibodies , purified by cell sorting using a FACSAria III ( BD Pharmingen ) and then washed and resuspended in RP-10 . Increasing concentrations of the APCs were cultured with 1 . 25×104 in vitro activated transgenic T cells in round bottom plates for 5 to 18 hours , as indicated , in the presence of 10 µg/mL BFA for the last 5 hours . Antibodies were from BD Pharmingen: anti-CD3 ( 145-2C11 ) , -CD4 ( RM4-5 ) , -CD8α ( 53-6 . 7 ) , -CD11b ( M1/70 ) , -CD19 ( ID3 ) , -CD45 . 1 ( A20 ) , -CD80 ( 16-10A1 ) , -CD86 ( GL1 ) , -IFN-γ ( XMG1 . 2 ) , -Ly6C ( AL21 ) , -NK1 . 1 ( PK136 ) , -Vα2 ( B20 . 1 ) and -Vβ8 ( MR5-2 ) ; from eBioscience: anti-CD45 . 2 ( 104 ) and -CD11c ( N418 ) ; or from BioLegend: anti-CD326 ( g8 . 8 ) and -Vα3 . 2 ( RR2-16 ) . For intracellular staining , cells were fixed with a Cytofix/Cytoperm kit ( BD Pharmingen ) . A FACSCanto II ( BD Pharmingen ) and FlowJo software ( TreeStar ) were used for analysis . Propidium iodide ( Sigma Aldrich ) and SPHERO calibration particles ( BD Pharmingen ) were added for identification of live cells and enumeration . Skin was fixed at room temperature for 30 min in PLP buffer ( 0 . 2 M NaH2PO4 , 0 . 2 M Na2HPO4 , 0 . 2 M L-lysine and 0 . 1 M sodium periodate with 2% paraformaldehyde ) , washed twice with PBS and incubated for 30 min in 20% sucrose , prior to being embedded , frozen , cut and stained as previously described [48] . IFN-γ staining ( AlexaFluor647 , BD Pharmingen ) was performed overnight at 4°C ( 1∶75 in PBS containing 2 . 5% [w/v] donkey serum ) . Anti-keratin-5 and -14 polyclonal antibodies were from Jomar Biosciences; anti-CD4 ( RM4-5 ) from BioLegend; anti-CD8 ( 53 . 67 ) from BD Pharmingen and polyclonal anti-HSV from Dako North America . Slides were mounted with ProLongGold antifade media ( Invitrogen ) , air-dried and viewed using a Zeiss LSM700 confocal microscope and Imaris 7 . 1 software ( Bitplane ) . Mice were anesthetized and HSV-1-infected flank skin was mounted on an imaging platform and acquired with an upright LSM710 NLO multiphoton microscope as described previously [48] . Imaging data was processed and automatic cell tracking aided by manual corrections was performed with Imaris 7 . 1 software . For movies , image sequences were composed in Adobe After Effects CS5 . Graphs were plotted using Prism 5 ( Graphpad ) and comparison of data sets was performed by one-way analysis of variance followed by Tukey post-test , or Mann-Whitney or student t tests , as indicated . All graphs depict means ± s . e . m . .
|
HSV-1 is a widely distributed pathogen causing a life-long latent infection associated with periodic bouts of reactivation and severe clinical complications . Adaptive immune responses encompassing CD4+ and CD8+ T-cell activities are key to both the clearance of infectious virus and the control of latent infection . However , precisely how such T-cell responses are regulated , particularly within acutely infected peripheral tissues , remains poorly understood . Using a mouse model of HSV-1 skin infection , we describe a complex regulation of T-cell responses at the site of acute infection . These responses were subset-specific and anatomically distinct , with CD4+ and CD8+ T-cell activities being directed to distinct anatomical compartments within the skin . While IFN-γ-producing CD4+ T cells were broadly distributed , including skin regions a considerable distance away from infected cells , CD8+ T-cell activity was strictly confined to directly infected epithelial compartments . This unexpected spatial segregation was a direct consequence of the involvement of largely non-overlapping types of antigen-presenting cells in driving CD4+ and CD8+ T-cell effector activity . Our results provide novel insights into the cellular regulation of T-cell immunity within peripheral tissues and have the potential to guide the development of T-cell subset-specific approaches for therapeutic and prophylactic intervention in antimicrobial immunity and autoimmunity .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"blood",
"cells",
"dermatology",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"herpes",
"simplex",
"immune",
"activation",
"antigen-presenting",
"cells",
"immunology",
"microbiology",
"infectious",
"disease",
"immunology",
"skin",
"infections",
"sexually",
"transmitted",
"diseases",
"animal",
"models",
"of",
"infection",
"infectious",
"diseases",
"white",
"blood",
"cells",
"animal",
"cells",
"t",
"cells",
"immune",
"system",
"cell",
"biology",
"clinical",
"immunology",
"immunity",
"virology",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"viral",
"diseases",
"acquired",
"immune",
"system"
] |
2014
|
Distinct APC Subtypes Drive Spatially Segregated CD4+ and CD8+ T-Cell Effector Activity during Skin Infection with HSV-1
|
Planar cell polarity ( PCP ) signaling controls the global orientation of surface structures , such as hairs and bristles , in both vertebrates and invertebrates . In Frizzled6-/- ( Fz6-/- ) mice , hair follicle orientations on the head and back are nearly random at birth , but reorient during early postnatal development to eventually generate a nearly parallel anterior-to-posterior array . We report the identification of a naturally occurring exon 5 deletion in Astrotactin2 ( Astn2 ) that acts as a recessive genetic modifier of the Fz6-/- hair patterning phenotype . A genetically engineered Astn2 exon 5 deletion recapitulates the modifier phenotype . In Fz6-/-;Astn2ex5del/del mice , hair orientation on the back is subtly biased from posterior-to-anterior , leading to a 180-degree orientation reversal in mature mice . These experiments suggest that Astn2 , an endosomal membrane protein , modulates PCP signaling .
In complex multi-cellular organisms , individual surface structures such as hairs , feathers , scales , and bristles typically exhibit a high degree of spatial order . In birds and mammals , the stereotyped orientations of feathers and hairs reflect the underlying orientations of follicles within the dermis . Hair follicle orientation is controlled by planar cell polarity ( PCP ) signaling , as determined by the changes in follicle orientation associated with mutations in the core PCP genes Frizzled6 ( Fz6 ) , Celsr1 , and Van Gogh-like2 ( Vangl2 ) in mice [1–5] . In the absence of Fz6 , the initial orientations of hair follicles on the head and back appear to be largely randomized , in contrast to the nearly parallel orientations of follicles on most of the body surface of wild type ( WT ) mice . During the first postnatal week , hair follicles in WT mice undergo a subtle reorientation , referred to as “refinement” , which minimizes angular differences among neighboring follicles . This process also leads to a more precise alignment of follicles with the body axes ( on the back ) or with local anatomic structures ( on the limbs ) [2 , 3] . In Fz6-/- mice , the refinement process is associated with far larger angular reorientations than in WT mice , presumably because Fz6-/- follicles exhibit a greater diversity of initial orientations [2 , 3] . In Fz6-/- back skin , this process leads initially to a series of large-scale patterns , such as whorls , most of which disappear by postnatal day ( P ) 10-P15 as the field of follicle vectors progressively aligns along an anterior-to-posterior direction . Current evidence suggests that PCP proteins are essential for cell-to-cell propagation and intracellular interpretation of polarity information , but the molecules and mechanisms responsible for setting up the initial asymmetry in spatial information remain unknown [6] . In the present study , we identify a genetic modifier of the PCP hair patterning phenotype that imposes a large-scale asymmetry on hair follicle orientation .
This work began with the chance discovery of an unusual and stereotyped hair pattern among siblings in a Fz6-/- intercross , referred to hereafter as the ridge phenotype . This phenotype is characterized by a transverse ridge across the back , which arises when hairs in the upper back that are oriented in an anterior-to-posterior direction encounter hairs on the lower back that are oriented in a posterior-to-anterior direction ( Figs 1 and S1 ) . The ridge pattern is not observed in typical Fz6-/- mice . As seen in Figs 1 and S1 , typical Fz6-/- back skins at P8 exhibit limited deviations from the strictly anterior-to-posterior follicle orientation of WT follicles . Additional crosses established that the ridge phenotype segregates as a recessive trait and is only observed in the absence of Fz6 . As the genetic background of our Fz6-/- line consisted of contributions from C57Bl6/J and SV129 , as well as an indeterminate contribution from a Flp-expressing line , we guessed that the Fz6-/- line might harbor sufficient genetic diversity that a genome-wide SNP screen could identify the locus responsible for the ridge phenotype . This strategy revealed a single linkage peak based on typing of 1 , 449 loci across the genome in 43 ridge+ and 39 ridge- progeny from a Fz6-/- intercross that was segregating the ridge phenotype . The peak resides on chromosome 4 and has a multipoint LOD score of 30 ( Fig 2A ) . To narrow the region within which the ridge locus resides , we scored hair patterns and polymorphic markers flanking the critical interval in >1 , 500 progeny of Fz6-/-;ridge/ridge x Fz6-/-;ridge/+ parents , and then fine-mapped the recombination breakpoints in the subset of progeny that exhibited a recombination event within the critical interval ( Fig 2B ) . This analysis narrowed the critical interval to a 2 . 3 Mb segment encompassing or adjacent to the genes for Toll-like receptor4 ( Tlr4 ) , Astrotactin2 ( Astn2 ) , and Trim32 , a gene embedded within intron 16 of the Astn2 gene ( Fig 2C ) . A 74 . 7 kb spontaneous deletion that overlaps the Tlr4 gene , and that eliminates Tlr4 mRNA and protein production [7 , 8] , was crossed into the Fz6-/- line and found to have no effect on hair patterning , indicating that the ridge phenotype does not arise from a loss of Tlr4 function . PCR amplification and sequencing of all of the exons of these three genes showed only one difference between control C57Bl6/J and ridge chromosomes: a consistent failure to amplify Astn2 exon 5 from ridge chromosomes ( Figs 2D and S2 ) . Tests with ten additional PCR primer pairs in the flanking introns revealed a ~30 kb deletion that encompasses Astn2 exon 5 and has endpoints within a pair of LINE elements ( Fig 2E ) . To search for the origin of the Astn2 exon 5 deletion , we analyzed the ES cells that were used to generate the targeted Fz6 null allele [1] . This ES cell line ( “R1” ) was derived by Nagy et al [9] from a cross between two 129/Sv lines [10–12] . PCR typing showed that the ~30 kb deletion is present in both R1 ES cells and in 129X1/SvJ mice , but not in the closely related 129S1/SvlmJ or 129S6/SvEvTac lines ( Fig 3A ) . NextGen sequencing of genomic DNA from the critical interval confirmed the presence of this deletion in 129X1/SvJ and Fz6-/-;ridge/ridge lines but not in the 129S1/SvlmJ or 129S6/SvEvTac lines ( S3 Fig ) . Importantly , when the Astn2 allele present in each of the three 129 lines was crossed into the Fz6-/- background and assessed in the homozygous state , only the 129X1/SvJ-derived Astn2 allele produced the ridge phenotype ( Fig 3B ) . The data presented thus far provide strong correlative evidence that the Astn2 exon 5 deletion causes the ridge phenotype . To definitively test this hypothesis , we used gene targeting in ES cells derived from 129S6/SvEvTac mice to generate a conditional allele in which Astn2 exon 5 is flanked by loxP sites ( S4 Fig ) . When this conditional allele is made homozygous in a Fz6-/- background ( Fz6-/-;Astn2ex5fl/fl ) the hair pattern is indistinguishable from that seen in Fz6-/- mice , i . e . it lacks a ridge . By contrast , Fz6-/-;Astn2ex5del/del mice , which lack Astn2 exon 5 ( following Cre-mediated germ-line recombination of the Astn2ex5fl allele ) , show a ridge phenotype indistinguishable from that seen in the original Fz6-/-;ridge/ridge mice ( Fig 4A ) . As with the naturally occurring ridge genotype , the Astn2ex5del/del genotype does not produce a hair patterning phenotype on a Fz6+/- background . These experiments demonstrate that a small deletion encompassing Astn2 exon 5 is responsible for the ridge phenotype , thereby identifying a mammalian modifier locus and revealing its origin as a recent spontaneous deletion . It is possible that the naturally occurring ~30 kb ridge deletion eliminates transcriptional regulatory sequences in addition to eliminating Astn2 exon 5 . Such a possibility is less likely for the engineered exon 5 deletion ( Astn2ex5del ) , which is only 1 . 07 kb in length . The transcription start sites of the Trim32 and Astn2 genes are at distances of 443 kb and 345 kb from Astn2 exon 5 ( Fig 2C ) . RT-PCR analysis of Trim32 transcripts in embryonic day ( E ) 15 . 5 Fz6-/- and Fz6-/-;ridge/ridge skin showed qualitatively similar expression levels ( S5 Fig ) . A similar RT-PCR analysis of Astn2 transcripts in E15 . 5 Fz6-/- , Fz6-/-;ridge/ridge , and WT skin also showed qualitatively similar expression levels ( S6A and S6B Fig ) . Although we cannot exclude the formal possibility that sequences in or immediately adjacent to Astn2 exon 5 regulate the expression of a more distant gene and that the ridge phenotype reflects perturbations in that regulation , the weight of the evidence supports the conclusion that the ridge phenotype reflects the absence of Astn2 exon 5 coding sequences . Astn2 and its close homologue Astn1 have been implicated in neuronal migration along glial scaffolds [13 , 14] . Astn1 and Astn2 are predicted to have a signal peptide , two transmembrane domains , and an unusual transmembrane topography in which both N- and C-termini reside on the extracellular face of the membrane ( S6C and S6D Fig ) . Both proteins localize to endosomes , and are expressed in multiple tissues during development [14] . By in situ hybridization , we observed Astn2 expression in hair follicles starting at the earliest stage of their development ( Fig 4B ) . Although the precise mechanism of action of the Astrotactins is still unclear , their endosomal localization suggests that they might be involved in recycling of plasma membrane proteins [14] . Interestingly , Devenport et al . [15] observed that PCP protein complexes in the developing epidermis are internalized into endosomes and then reassembled at the plasma membrane with every cell division . These observations suggest the possibility that an alteration in Astn2 might modify the Fz6-/- phenotype by affecting PCP protein trafficking . If correct , this hypothesis would also imply that Fz6-/- embryos retain some level of PCP signaling in the skin . Deletion of Astn2 exon 5 leads to an in-frame deletion of 36 amino acids in the predicted cytosolic domain , a region with no homology to any proteins other than Astn1 . Interestingly , constitutive alternative splicing leads to frequent skipping of Astn2 exon 4 , which leads to an in-frame deletion of 52 amino acids , also in the predicted cytosolic domain ( Figs S6 and S7 ) . Constitutive exon 4 skipping implies that large changes in the putative intracellular domain are compatible with protein stability and function , which suggests that deletion of exon 5 may alter but not abolish Astn2 function . How might deletion of Astn2 exon 5 influence hair follicle development to uniformly reverse the orientations of thousands of follicles in Fz6-/-;ridge/ridge mice ? The answer to this question could be related to the striking changes in orientation that occurs among Fz6-/- follicles during early postnatal development . As noted in the Introduction , at birth , Fz6-/- mice show hair follicle orientations on the back that appear to be approximately random , but over the first 1–2 postnatal weeks , these follicles reorient to generate a series of increasingly organized macroscopic patterns , eventually reorienting in an anterior-to-posterior direction . Our earlier work suggested that the reorientation process obeys a local consensus rule that minimizes angular differences among neighboring follicles [2 , 3] . Computer simulations demonstrated that this process efficiently enhances the amplitude of any global bias in initial orientation while simultaneously suppressing random orientation noise , with the result that a small initial orientation bias produces a uniform reorientation of all follicles along the direction defined by that bias [2] . An initial clue to the mechanism of follicle orientation reversal in Fz6-/-;ridge/ridge mice emerged when we quantified hair follicle orientations on the head and lower back of Fz6-/- , Fz6-/-;ridge/ridge , and Fz6-/-;Astn2ex5del/del mice at P3 ( Fig 4C ) . On the lower back , Fz6-/-;ridge/ridge and Fz6-/-;Astn2ex5del/del follicles show a subtle posterior-to-anterior bias , whereas Fz6-/- follicles show a subtle anterior-to-posterior bias ( compare panels q vs . r and t in Fig 4C ) . This trend was less apparent on the head ( compare panels g vs . h and j in Fig 4C ) . To extend this analysis , we quantified the orientations of >11 , 500 follicles from the lower backs of eight Fz6-/- , five Fz6-/-;ridge/ridge , and nine Fz6-/-;Astn2ex5del/del mice at P3 ( Fig 4D ) . The results confirm the directional bias noted above , with pairwise P-values of 1 . 7x10-5 for the Fz6-/- vs . Fz6-/-;ridge/ridge comparison and 9 . 7x10-7 for the Fz6-/- vs . Fz6-/-;Astn2ex5del/del comparison ( student’s t-test ) . The P-value for the comparison of Fz6-/- vs . the combination of Fz6-/-;ridge/ridge and Fz6-/-;Astn2ex5del/del is 3 . 1x10-9 . Interestingly , the follicle orientation histograms from all three genotypes exhibit minima at orientations perpendicular to the anterior-posterior axis ( Fig 4D ) , suggesting an additional bias favoring follicle orientations that are either parallel or anti-parallel to this axis . These quantitative analyses suggest that loss of Astn2 exon 5 either ( 1 ) continuously acts to reorient follicles in the lower back in a posterior-to-anterior direction , or ( 2 ) creates an initial posterior-to-anterior orientation bias , which is subsequently enhanced by the local refinement process . Although we cannot , at present , distinguish between these alternative models , the morphologic data imply that dramatically different follicle orientation patterns in mature skin can be consistently generated from subtly different patterns in immature skin . The performance of this system is all-the-more-remarkable because , in the Fz6-/- background , the field of immature follicle vectors has a very low signal-to-noise ratio . PCP signaling plays a central role in a wide variety of developmental processes . In addition to hair follicle orientation , these include neural tube closure , the orientation of motile cilia and of vestibular and auditory hair cells , and axon guidance [6] . The shared dependence of these processes on PCP signaling suggests that insights obtained from studying any one of them may shed light on the others . The relationship between hair follicle orientation and axon guidance is especially intriguing and is emphasized by the requirement for Celsr and Frizzled family members in both processes [5 , 16 , 17] and by the partial interchangeability of Fz6 and its close homologue Fz3 , which controls axon guidance [18] . The role of Astrotactins in both neuronal migration and hair follicle orientation suggests an even closer connection between patterning mechanisms in skin and brain .
This study was performed in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health . All of the animals were handled according to approved Institutional Animal Care and Use Committee ( IACUC ) protocol MO13M469 of the Johns Hopkins Medical Institutions . Fz6-/- mice are described in Guo et al . [1] . 129X1/SvJ , 129S1/SvlmJ and the Tlr4 deletion ( JAX #003752 ) lines were purchased from Jackson Laboratories . The 129S6/SvEvTac line was purchased from Taconic . The Astn2 floxed exon5 targeting construct was electroporated into MC1 ES cells ( from 129S6/SvEvTac-mice; a kind gift from Mitra Cowan ) and plated in G418 and ganciclovir for positive and negative selection . Colonies were screened by Southern blotting , and clones carrying the targeted allele were injected into C57BL/6 blastocysts . Chimeras were bred to C57BL/6 , and the FRT-flanked PGK-neo cassette was removed by crossing to germline Flp mice to generate the Astn2ex5fl allele . The Astn2ex5del allele was generated by crossing mice carrying Astn2ex5fl to mice carrying germline Sox2-Cre [19] ( Tg ( Sox2-Cre ) 1Amc/J; from Jackson Laboratories ) . For meiotic mapping , Fz6-/-;ridge/+ and Fz6-/-;ridge/ridge progeny of Fz6-/-;ridge/+ x Fz6-/-;ridge/ridge parents were phenotyped at ~P8-P10 by visual inspection of the hair pattern ( i . e . examined for the presence or absence of the transverse ridge ) , and genotyped by scoring microsatellite insertion/deletion variants with the PCR primers listed in S1 Table . High resolution mapping of the critical interval was performed by SNP genotyping with the PCR primers listed in S2 Table . PCR primers for amplifying the 23 Astn2 exons are listed in S3 Table . PCR primers for mapping the ~30 kb deletion encompassing Astn2 exon 5 ( Fig 2E ) are listed in S4 Table . RT-PCR primers for amplifying Dbc1 and Trim32 are listed in S5 Table . An Illumina mouse SNP array with 1 , 449 loci was used to type 43 ridge+ and 39 ridge- progeny from a Fz6-/- intercross that was segregating the ridge phenotype . The multipoint LOD score was calculated using R software with the quantitative trait locus ( QTL ) bioinformatics add-on package Version 1 . 21–2 ( release March 18 , 2011; http://www . rqtl . org ) . The calculation used a hidden Markov model with the Haley-Knott regression . The highest LOD score was 29 . 7 on Chromosome 4 , with the peak at position 63 . 65 . Genomic DNA from 129S1/SvlmJ , 129S6/SvEvTac , 129X1/SvJ , and Fz6-/-;ridge/ridge mice was purified from brain tissue by proteinase K digestion and CsCl centrifugation , fragmented to a mean size of ~350 bp , captured on a custom designed Agilent SureSelect oligonucleotide array that covered all non-repetitive sequences in the interval 6 . 58–6 . 72 Mb on chromosome 4 ( mouse genome build 38 ) , and subjected to 150 base paired-end sequencing on an Illumina MySeq to a mean coverage depth of ~50X . Skin flat mounts were prepared as described in Chang et al [20] . To visualize follicles using the endogenous melanin pigment , the dorsal back skin ( at P3 and P8 ) was dissected and flattened by pinning its edges to a flat Sylguard surface , fixed overnight in 4% paraformaldehyde in PBS , dehydrated through a graded alcohol series , and then clarified with benzyl benzoate:benzyl alcohol ( BBBA ) in a glass dish . Images were collected with a dissecting microscope . Hair follicle orientations were scored one at a time by placing the image of a freely rotatable vector over the skin flatmount image ( 3 . 2 mm x 2 . 5 mm ) , superimposing the vector on the follicle of interest , and assessing the best fitting vector orientation by visual inspection . Two sampling strategies were used . For low-density sampling ( Fig 4C ) , orientations were determined only for the 81 follicles closest to each point of intersection of the nine vertical and nine horizontal lines in a 2 . 88 mm x 2 . 25 mm grid overlaid on each image . For high-density sampling ( Fig 4D ) , all follicles within each image were scored . Vector orientations were measured in Photoshop and ImageJ . Statistical comparisons were performed in Microsoft Excel . To assess the reproducibility of the scoring method , images of two P3 back skin flat mounts ( one Fz6-/- and the other Fz6-/-;Astn2ex5del/del ) were rotated 180 degrees , and follicle orientations for all four images ( two original and two rotated; n is approximately 700 follicles per image ) were determined by an individual who was blinded to the genotypes and to the relatedness of the images . As shown in S8 Fig , when corrected for the 180 degree rotation , the distributions of follicle angles in the two rotated images were found to be nearly identical to the distributions in the original images , and each image reproduced the distinctive genotype-specific patterns shown in Fig 4D , which were based on quantification of 22 back skin images . In situ hybridization was performed as described [21] . Digoxigenin-labeled riboprobes were transcribed using T7 RNA polymerase from the Astn2 cDNA ( coding regions within exons 19–23 ) , which was cloned from WT mouse E15 . 5 skin by RT-PCR . Images were captured on an Imager Z1 microscope ( Zeiss ) using Openlab software .
|
Hair , feather , and scale patterns are a universal feature of vertebrate surface morphology . These patterns are under precise genetic control as seen by their species-specificity and by their alterations in different breeds of domesticated animals . The first clues to the mechanism of hair patterning in mammals came from genetic analyses of proteins that are homologous to a small set of Drosophila proteins that control patterning of bristles and hairs on the insect body surface and wings . The patterning process , referred to as planar cell polarity , involves a cell surface protein , Frizzled6 , which is produced in skin and hair follicles . Following a chance observation that some Frizzled6 mutant mice exhibit an unusual hair pattern in which all of the hair follicles on the posterior half of the back have reversed orientations , we have identified a single spontaneous mutation that accounts for this reversal . The mutation removes a single coding exon from the gene coding for the membrane protein Astrotactin2 . Interestingly , a closely related protein , Astrotactin1 , has been implicated in directed neuronal migration along a glial substrate , suggesting a mechanistic connection between patterning mechanisms in skin and brain .
|
[
"Abstract",
"Introduction",
"Results",
"and",
"Discussion",
"Materials",
"and",
"Methods"
] |
[] |
2015
|
Identification of Astrotactin2 as a Genetic Modifier That Regulates the Global Orientation of Mammalian Hair Follicles
|
We have developed a new approach to characterize allele-specific timing of DNA replication genome-wide in human primary basophilic erythroblasts . We show that the two chromosome homologs replicate at the same time in about 88% of the genome and that large structural variants are preferentially associated with asynchronous replication . We identified about 600 megabase-sized asynchronously replicated domains in two tested individuals . The longest asynchronously replicated domains are enriched in imprinted genes suggesting that structural variants and parental imprinting are two causes of replication asynchrony in the human genome . Biased chromosome X inactivation in one of the two individuals tested was another source of detectable replication asynchrony . Analysis of high-resolution TimEX profiles revealed small variations termed timing ripples , which were undetected in previous , lower resolution analyses . Timing ripples reflect highly reproducible , variations of the timing of replication in the 100 kb-range that exist within the well-characterized megabase-sized replication timing domains . These ripples correspond to clusters of origins of replication that we detected using novel nascent strands DNA profiling methods . Analysis of the distribution of replication origins revealed dramatic differences in initiation of replication frequencies during S phase and a strong association , in both synchronous and asynchronous regions , between origins of replication and three genomic features: G-quadruplexes , CpG Islands and transcription start sites . The frequency of initiation in asynchronous regions was similar in the two homologs . Asynchronous regions were richer in origins of replication than synchronous regions .
DNA replication in eukaryotes starts at DNA sequences termed origins of replication . The timing of DNA replication within the S-phase of the cell cycle depends on the distribution and the relative order of activation of those origins . Eukaryotic DNA replication is mediated by the sequential formation of a pre-Replication Complex ( preRC ) in the G1 phase of the cell cycle , starting with the binding of the origin recognition complex ( ORC ) to DNA and followed by the binding of additional licensing proteins that lead to activation of replication origins in S phase [1]–[3] . Although in some loci origins are constitutive and initiate replication each cell cycle , in most loci origin usage is very flexible and there are many more pre-RC complexes than origins that actually initiate replication [1] , [4] , [5] . The timing of initiation events and the precise chromosomal location of the origins of replication seem to be determined independently [6] . The timing of replication might be determined at the level of large chromatin domains early in G1 , while origin selection , a process that determines which potential replication origin will initiate replication in a particular cell cycle , would occur later in G1 . Several reports have shown that origin selection is also influenced by events taking place in mitosis [7]–[9] . The flexibility in origin usage results in the detection of replication initiation zones , which are DNA regions in which many origins can be mapped and therefore where initiation seems very frequent . Origins in initiation zones are used alternatively , apparently stochastically , in the different cells of a homogeneous cell population [10] . This flexibility might be necessary to allow multicellular organisms to replicate their genomes in cells with different chromatin structure or after DNA damage [1] . Several models have been proposed to explain the general organization of the replication program . These models differ in two basic assumptions . First , one set of models assumes that the frequency of initiation from each particular origin might be regulated independently of the time of replication during the cell cycle [11] , whereas another set of models assumes that the replication time and initiation frequency are co-regulated [12] , [13] Second , these models also differ in whether they postulate the existence of mechanisms that coordinate initiation from multiple origins of replication ( See [14] for a detailed discussion ) . Computer simulations and analyses at the single molecule level have led to the independent origin hypothesis , which proposes that origins initiate replication stochastically , independently from neighboring origins [11] . According to this model , the sole determinant of the timing of replication would be the efficiency of individual origins rather than a mechanism that directly regulates the time at which origins start replication . Simulations based on the independent origin hypothesis can model replication very successfully in budding yeast [11] and have been applied effectively to explain the regulation of the IgH locus in mammals [14] . In the case of mammalian cells , the next-in-line and the domino-cascade models [12] , [13] have been proposed to take into consideration the idea that neighboring origins can be co-regulated . On a larger scale , evidence has accumulated that there are chromatin mechanisms that coordinate initiation from multiple origins in regions that can be as large as several 100 kb . These suggestions are based on studies of the sub-nuclear localization of replicons during S phase , on observations ( using high-throughput methods ) of large contiguous regions of the genome that seem to exhibit coordinated replication , and on the findings that rif1 knock-out or knock-down and non-coding RNAs can affect the replication times of large chromosomal domains [15]–[17] . Based on these results , the Gilbert lab has proposed the replication domain model which postulates the existence of distinct epigenetic entities ( domains ) with physical boundaries independent of replicon distribution . According to this model , epigenetic mechanisms would be able to control replication timing by influencing initiation of DNA replication from all the origins within these domains [18] . No consensus sequences have been demonstrated for all origins of replication in mammalian cells [19] . No consensus sequences were found in studies of short newly replicated nascent strands ( NS ) isolated from 1% of the genome in transformed cells [20] and on 0 . 4% of the genome in mouse embryonic stem cells [21] . Pioneering studies on about 2% of the mouse genome in several cell lines demonstrated an association between origins of replication and G-rich regions [10] , [22] . In agreement , genome-wide studies by nascent strand DNA sequencing and bubble-seq demonstrated genome-wide association between CpG islands , transcription start sites , G4 quadruplexes and uncovered novel associations among origins , specific histone modifications and DNase I hypersensitive sites [23]–[25] . Although the primary sequence can clearly affect initiation capacity of individual replication origins [26]–[29] , these findings raised questions regarding the relative contributions of genetic and epigenetic regulatory mechanisms in determining the locations , initiation frequencies and timing of initiation from replication origins . The genome-wide timing profiles produced so far represented the average replication timing of the two alleles in a population of cells . In such profiles , regions that appear to replicate in mid-S phase can be interpreted either as regions in which the two alleles replicate at the same time in mid-S phase or as regions in which the two alleles replicate asynchronously or randomly . The size of asynchronously replicated domains and the mechanism of their establishment remain largely unknown . Three mechanisms are known to be associated with asynchronous replication of the two chromosome homologs: X chromosome inactivation , parental imprinting and mono-allelic expression [30] , [31] . A recent genome-wide allele-specific study in human lymphoblastoid cell lines has shown that the inactive X chromosome , replicated in a less structured manner than the active X chromosome and early replicating regions [32] . Whether other mechanisms can lead to replication asynchrony is not known . To characterize the extent of replication asynchrony in human cells and to begin to address the possible causes of asynchrony , we have developed an approach to measure allele-specific replication timing genome-wide . In addition , we have developed an approach to differentiate between genetic and epigenetic mechanisms based on the analysis of the phased genomes of a family quartet . We have used this method as well as two distinct origin mapping methods based on non-overlapping assumptions to provide a high resolution view of the regulation of replication in human primary cells .
The TimEX method compares the amount of DNA present in the S and G1 phases of the cell cycle as a surrogate for the timing of replication . The approach was first implemented at a low resolution using micro-arrays [33] and then extended genome-wide to high-throughput sequencing [34] ( Figure 1A & B ) . TimEX-seq measurements are very accurate because they do not require drug-induced cell synchronization , which can induce artifacts , or sorting of multiple S phase fractions , which can be difficult to calibrate . Genome-wide allele-specific analysis requires phased genome sequences which can be obtained by several methods including sequencing members of the same family [35] . To establish an experimental system that would allow allele-specific analysis in human primary cells , we recruited family FNY01 , a large family with no known genetic diseases , completely sequenced the genomes of two sisters and their biological parents ( Figure 1C ) and phased all genetic variants using a combination of transmission , physical and population-based phasing methods [36] . The TimEX-seq method was then applied to human primary basophilic erythroblasts that were produced by in vitro differentiation of circulating peripheral blood hematopoietic stem and progenitor cells from two members of the family [37] . All of the timing experiments were therefore performed twice on two different ( but related ) individuals . Allele-specificity was achieved by only considering reads containing phased heterozygous SNPs that could be assigned to either the maternal or paternal chromosome homologs . For each individual tested , we obtained between 43 and 55 million reads that contained informative SNPs . Sequencing depth per called SNPs averaged between 23 and 28 for the S and G1 profiles ( Table S1 ) . Allele-specific replication timing profiles were generated by calculating the number of reads from the maternal or paternal homologs obtained from cells in the S or G1 phase of the cell cycle in 500 bp genomic windows , smoothing the data with a Gaussian filter ( sigma equal to 100 kb ) and computing the S/G1 ratios of these smoothed allele-depth sums . Gaussian filters are similar to moving-average or Loess filters , but the averaged values are assigned weights that decrease following a Gaussian curve as a function of genomic distance . SNP density in family FNY01 varies between 1 SNP every 500 bp and one SNP every 5 kb because of the haplotype structure of the human genome [36] . The resolution of the replication timing information obtained therefore depended on the SNP density and regions of very low SNP density resulted in gaps . We estimate that we obtained accurate TimEX values for about 85% of the genome . The general shapes of the allele-specific profiles thus obtained were very similar to the shapes of the non-allele specific profiles that we and others previously generated ( Figure 1D ) [34] , [38] . About 96% of the TimEX values were between 0 . 9 and 2 . 1 in good agreement with the theoretically expected range of 1 to 2 . About 70% of the outliers were between 0 . 8 and 2 . 3 and fell mostly in regions of low SNP density . The Pearson's correlation coefficient between the timing profiles of the maternal and paternal homologs was very high ( r-squared = 0 . 95; Figure 1E ) . Histograms of the differences in TimEX values between the two alleles , in all 5 kb intervals genome-wide , had a mean equal to zero , but there was significant variation from normality at the extremes of the distribution reflecting the existence of regions in which the two alleles replicated asynchronously ( Figure 1F and S1 ) . To assess the average level of asynchrony , we first calculated the difference in the timing profiles for the two individuals tested . Assuming an 8 hour S phase , the differences in timing of replication between the two alleles was less than 10% of the length of S phase in 88% of the genome . This suggests that the two homologs replicated within less than 48 minutes of each other . Therefore , at the resolution of this study , the vast majority of the genome replicated synchronously . In the 12% of the genome that exhibited asynchronous replication , the timing differences did not generally exceed 30% of S phase length , corresponding to a maximum of 2–3 hours of replication delay between the two alleles . This was consistent with previous reports that showed that the time of replication of imprinted genes varies by only a few hours [39] . Larger differences in timing of replication likely exist between the active and inactive X chromosomes , but their detection would require the study of clonal cell populations . To characterize the mechanisms underlying replication asynchrony , we first attempted to determine if genetic differences between the two homologs could account for some of the replication asynchrony . There are four haploid genomes in a family quartet: two contributed by the mother and two by the father . The random assortment of chromosomes and the crossovers that occur during meiosis create recombined versions of these genomes in the children . Four inheritance states can be recognized [35] . When the two children inherit the same alleles from their mother and father , they are in the identical inheritance state; when they inherit different alleles , they are in the non-identical state; when they inherit the same allele from one parent but a different one for the other parent; they are in the haplo-identical maternal or paternal state . Each state represents about 25% of the genome of each child , the boundaries between the states being defined by the crossovers that occurred during meiosis in the parents . To determine if genetic differences can affect timing of replication , we compared the timing of replication in the identical and non-identical portions of the genome . The coefficients of correlation between the maternally inherited chromosomes of FNY01 3_2 and 3_3 were very similar in the identical and non-identical regions despite the presence of more than a million SNPs and indels between the maternally inherited homologs in the non-identical regions ( Figure 2A ) . Very similar results were found when the paternally inherited chromosomes were compared ( Figure 2A ) and when the calculated times of replication were compared in 500 kb windows rather than in 5 kb windows ( Figures S2A ) . Likewise , the distributions of the differences in timing between the maternal homologs or between the paternal homologs in the identical and non-identical regions were almost indistinguishable ( Figures 2B and S2B ) . We conclude from this comparison that the presence of SNPs and indels has no detectable on replication timing , at the resolution of this study . One likely explanation for this important result is that the timing of replication in any region is influenced by the combined initiations from many origins and therefore that differences in origin efficiency induced by genetic variations in the non-identical part of the genome cancel each other out and do not result in any detectable timing differences when large timing domains are considered . An alternative but less likely explanation is that there might be an evolutionary bias against mutations in origins of replication . Since the timing of replication can be determined by distal elements and epigenetic modifications , as well as by primary sequences [40] , redundancy of regulatory elements might also contribute to the robustness of the regulation of the timing of replication . Additional experiments will be necessary to determine if SNPs and indels can affect initiation from individual origins on a genome scale . Sequencing of family quartet FNY01 had revealed that there are about 1 , 500 structural variants ( SVs ) above 1 kb in size in the quartet , including about 100 variants larger than 10 kb , and 10 larger than 100 kb in each family member [36] . We took advantage of these naturally occurring variants to ask whether timing differences between the two homologs might be induced by large deletions and duplications . Examination of the ratios of the timing of replication of the homologs in overlapping 500 Mb intervals in regions containing heterozygous SVs revealed genomic regions in close proximity to SVs exhibited a higher prevalence of asynchrony than regions that were far from SVs ( Figures 2C–E , S2C–D ) . This was observed in the two individuals tested . The effect was small in the case of 10 Kb variants but increased progressively as the size of the variants increased . To determine if these observations were statistically significant , we performed randomization of the location of the SVs and calculated the ratios of the timing of replication of the homologs in overlapping 500 Mb intervals . This analysis demonstrated that the effect of SVs on timing of replication was statistically significant ( Figure 2D ) . This suggests that some large SVs alter the timing of replication , maybe because deletions or insertions of multiple origins of replication have an additive effect . This result is consistent with distal regulation of DNA replication initiation events , since large structural mutations might affect chromatin modifications and spatial relationships with chromatin modifiers such as LCRs , enhancers , or insulators , which might in turn affect timing of more than one origin . We then attempted to detect asynchronously replicated domains ( ARDs ) , defined as genomic regions that exhibit asynchronous timing of replication of the same signs over large regions . To automate and optimize the detection of these ARDs and more generally to assess the temporal and spatial resolution of the allele-specific TimEX-seq , we simulated ARDs in our data . These simulations were performed using a binomial random simulator . We first resampled the SNP-containing reads to create simulated maternal and paternal controls data tracks . Various amounts of reads were then added in silico to randomly selected genomic regions of one of these controls to simulate asynchronous regions ranging in size from 125 kb and 2 Mb . Simulated TimEX profiles were then generated by smoothing the data and calculating S/G1 ratios as described above for the actual data . Once these simulated TimEX profiles had been created , we tested a variety of approaches to systematically detect the simulated ARDs . This resulted in the development of a highly efficient two-step method to detect ARDs . In the first step , the island finder of GenPlay is used to detect regions of differential timing in the smoothed profiles; in the second step , the statistical significance of the islands thus defined is assessed by performing chi-square tests on the sum of the reads in each island . Parameters to define the ARDs with the island finder ( TimEX differences threshold , maximum gap size and minimum island size ) were optimized by trial and error to maximize the rate of detection of the simulated islands . These simulations revealed that the sensitivity of the allele-specific TimEX-seq approach is dependent on the size of the asynchronous regions and on the amplitude of the delay ( Figure S3 ) . A 10% excess in S phase DNA ( corresponding to a timing asynchrony of 48 minutes ) , could be detected more than 90% of the time ( with an FDR of 5% ) for 2 Mb asynchronous regions . A similar detection rate for regions of 1 Mb and 500 Kb respectively required an excess of S-phase DNA of 15% ( 72 min . asynchrony ) or 20% ( 96 min . asynchrony ) . Asynchronous regions smaller than 500 , 000 bp were difficult to detect . These simulations also revealed that the smoothing that we applied to the data created some distortion of the replication delay and size of the ARDs . The detected ARDs had a smaller timing delay and a larger size than the simulated asynchronous regions that we had introduced . These distortions were small for regions larger than 500 , 000 bp but quite large for smaller regions ( Figure S3 ) . Since ARDs smaller than 500 , 000 bp were difficult to detect and highly distorted by the smoothing process , we decided to limit our analysis to ARDs larger than 500 , 000 bp . Additional simulations revealed that detection of smaller size ARD would be possible using TimEX-seq but would require sequencing to a higher read-depth . Because the average replication delays in the ARDs were under-estimated using the optimized detection parameters , we also used the Island finder in GenPlay to identify within the ARDs , core regions where the timing difference was highest . Timing delays in these cores are a better approximation of the true delay than the average delay in the entire ARD . Having optimized the method to detect ARDs , we quantified the number of ARDs in individuals FNY01_3_2 and 3_3 . There were 617 statistically significant ARDs in FNY01_3_2 and 611 in 3_3 ( Table S2A–C ) . Two hundred sixty-two ARDs overlapped between the two individuals . The lack of overlap for some of the ARDs likely reflects technical issues ( heterogeneity in the number of SNPs in each region caused by the haplotype structure of the genome ) as well as genetic or epigenetic differences between the two individuals . The average timing delay in the ARDs and in the core ARD were , respectively , approximately 49 and 87 minutes . The maximum delays were around 2 . 5 hours in the ARDs and 3 . 5 hours for the core ARDs . To determine if asynchronous timing of DNA replication was associated with imprinting , we analyzed the distribution of imprinted genes in the ARDs . To increase read-depth we created a combined track in which the reads of both individuals were summed together . One hundred and one known or predicted imprinted genes showed evidence of asynchronous replication . Sixty-three known or predicted imprinted genes were within ARDs in FNY01 3_2 , forty-three in FNY01 3_3 and sixty-five in the combined track ( Figure 3A , Table 1 and S3 ) . Fifty-six genes were in imprinted domains in both individuals or in one individual and in the combined track . In addition , a number of genes , most notably part of the H19 cluster , exhibited evidence of asynchrony ( delays of over 45 minutes over large regions ) but the domains did not reach statistical significance because of low SNP density . The most statistically significant ARD and the most asynchronous ARD were enriched in imprinted genes . When the ARDs were ranked by p-values , respectively eight and seven of the 40 most significant asynchronously replicating domains in individual FNY01_3_2 and 3_3 contained ( or were within 100 kb ) of an imprinted gene . Randomization experiments revealed that this enrichment was highly statistically significant ( p-values = 7 . 90E-4 for FNY01_3_2 and 1 . 68E-4 for FNY01_3_3 ) . Computation of the average length of ARDs that were associated or not associated with imprinted genes revealed that the former category of ARD genes was about 1 . 5-fold larger ( Table 2 ) . These differences were highly statistically significant ( t-test p-values <10−5 ) . The average delays for the ARD and for the core ARD was about 15–30% higher for ARDs that contain imprinted genes when compared to those that did not contain imprinted genes . These differences reached statistical significance for both individuals ( Table 2 ) . We conclude from this analysis that imprinted genes often localize in the longest , most asynchronous ARD . However , many long highly asynchronous ARD did not contain imprinted genes . The erythroblasts profiled in this study are the progeny of thousands of primary stem and progenitor cells seeded in our 2-week erythroid culture system . Because inactivation in placental mammals is random , the TimEX profiles of the paternal and maternal homologs both represent the average timing of replication of the Xa and Xi chromosomes . We therefore did not expect to distinguish between the active ( Xa ) and inactive ( Xi ) X chromosomes unless X inactivation was imbalanced . Imbalanced X chromosome inactivation can occur stochastically or because of selection [41] , [42] . Skewed inactivation have been described in the erythroid lineage , for instance in the case of heterozygous G6PD deficiency [43] . Since the Xi chromosome replicates generally later than the Xa chromosome , which can replicate throughout S phase [44] , [45] , the range of the S/G1 values for the Xi chromosomes is expected to be narrower than that of the autosomes . To verify this prediction in our experimental system , we plotted the histograms of the S/G1 ratios for the autosomes and for the X chromosomes ( Figure 3B ) . As expected , the S/G1 ratios varied from about 0 . 9 to 2 . 2 for the autosomes and only between 0 . 9 and 1 . 5 for the X chromosomes , suggesting that one of the X chromosomes replicated relatively uniformly late in both individuals , consistent with recent whole-genome observations in adult lymphoblastoid cell lines [32] . Further analysis revealed a chromosome X specific difference between individuals FNY01_3_2 and 3_3 , who are both female . The ARDs on the X chromosomes of FNY01_3_2 were similar to those of the autosomes: They represented 10–30% of each chromosome and there was approximately the same number of ARDs exhibiting maternal or paternal delays on each chromosome ( Figure 3C and 3D ) . By contrast , almost 50% of the length of chromosome X of individual FNY01_3_3 replicated asynchronously and almost 90% of the ARDs on chromosomes X of FNY01_3_3 exhibited a maternal delay . Together , these observations strongly suggest that X-chromosome inactivation in the basophilic erythroblasts of FNY01-3_3 that we tested was not random but was skewed toward inactivation of the maternal chromosome . To determine if this skew was due to the lower SNP density on the X chromosomes relative to the autosomes , we masked the SNP-poor regions and repeated the analysis . As shown in Figure 3E , the same pattern was observed , suggesting that the observation is not a technical artifact of the TimEX method . The cellular and molecular mechanisms responsible for this inactivation bias are unclear . The bias might have been acquired in early development , in the hematopoietic stem cells or in culture during erythroid differentiation , maybe because FNY01_3_3 is heterozygous for an X-linked gene such as G6P-D that is selected against in the erythroid lineage . In any case , these observations suggest that allele-specific TimEX will prove useful to assess of inactivation of the X chromosomes in human primary cells . Production of allele-specific timing profiles required a sequencing depth of more than 6 . 1011 bases ( 25–30x coverage per individual ) because only about one read in ten contains an informative SNP . By considering all the reads produced by the experiment , we also constructed high-resolution timing maps that were not allele-specific but that were more precise than any map previously produced . The resolution of these profiles was so high that very little smoothing was required ( Figure 4 ) . Calculating the S/G1 ratios in 1 kb windows and smoothing lightly with a sigma of 20 kb , instead of 5 kb windows and a sigma of 100 kb [34] , revealed reproducible sub-peaks , or ripples within the megabase-size timing domains that have been well-characterized by several labs ( Figure 4 ) [38] , [46]–[49] . By analogy with Raghuraman et al . who have shown in yeast that maxima of timing of replication profiles correspond to the location of origin of replication [50] , we hypothesized that these ripples , which had not been previously reported , might co-localize with zones of origins of replication or particularly dominant origins of replication . To test that hypothesis , we generated maps of replication origins , validated the data by comparison with previously published results , and compared TimEX and origin of replication profiles . To obtain maps of replication origins , we cultured basophilic erythroblasts from FNY01_2_2 , the father of FNY01_3_2 and 3_3 , and generated genome-wide profiles of nascent , newly replicated DNA strands . Because of the reported difficulties involved in reproducing NS profiles between labs , two methods were used . Replication initiation analyses were performed with two technical replicates , using two methods relying on non-overlapping assumptions to isolate nascent DNA strands . Nascent strands were first purified using the classic method which relies on sequencing DNA fragments of small size that contain RNA primers that rendered them resistant to lambda exonuclease digestion [23] . To control for potential biases introduced by the use of lambda exonuclease , we also mapped NS with a method that does not rely on this enzyme . Briefly , the cells were pulse-labeled for 30 minutes with bromo-deoxy-Uridine ( BrdU ) , and short , newly synthesized DNA strands were immuno-precipitated with anti-BrdU antibodies , yielding a population of NS that were also subject to massively parallel sequencing [40] , [51]–[53] . Results from both methods were in good agreement . The genome-wide coefficient of correlation between the profiles obtained with both methods was 0 . 7 and reached 0 . 85 for several chromosomes ( Figure 5A ) . Besnard et al . have reported a two-fold enrichment over randomly simulated data in the number of 4 kb fragments containing lambda exonuclease resistant NS peaks associated with G-quadruplexes and a three-fold enrichment for peaks associated with CpG islands and with transcription sites [24] . Similarly , Messner et al . , using bubble-seq sequencing , have detected enrichments for these three genomic features in origin-containing 3 to 4 kb Eco RI fragments analyzed by Bubble-Seq , albeit of lower magnitude [25] . To compare the BrdU method to the lambda exonuclease method , we called the peaks observed with both methods using the MACS 1 . 42 software [54] . The false discovery rate ( FDR ) was set to 0 . 1% . The BrdU and lambda-exonuclease based methods yielded 249 , 000 and 290 , 000 peaks , respectively . Close examination of the peaks revealed that many of the statistically significant peaks were minor peaks that were barely above background . To simplify the comparison , we increased the size of all peaks to 400 base pairs and only considered the top 100 , 000 peaks for further analysis . 37 . 5 , 6 . 6 and 13 . 1 percent of the BrdU NS peaks were , respectively , associated with G-quadruplexes , transcription start sites and CpG islands ( Figure 5B ) . Enrichments over randomly simulated data were , respectively , 4 . 0 , 2 . 8 and 4 . 7-fold ( Figure 5B ) . About 46% of all peaks were associated with at least one of these three features ( Figure S4A ) . Lambda-exonuclease peaks were associated with the same three features but the enrichment levels were lower ( Figure 5B ) . Most of the prototypical origins [26] , [55]–[58] that have been studied in details contained one or more of these genomic features ( Figure S4B ) . Importantly , only 14 . 2 , 17 . 8 , and 35 . 1% of all G-quadruplexes , transcription start sites and CpG islands were , respectively , associated with BrdU NS peaks ( Figure 5C ) . Only about one G-quadruplex in seven was associated with a BrdU NS peak . Overall , these data are consistent with the notion that subsets of G-quadruplexes , transcription start sites and CpG islands that are associated with origins of replication might have special characteristics . To test whether chromatin accessibility contributed to origin formation , we next determined the percentages of G-quadruplexes , transcription start sites and CpG islands that were associated with BrdU NS peaks produced in our study with the DNase I hypersensitivity generated by the ENCODE consortium in either K562 cells or in fetal liver basophilic erythroblasts . These two cell types were selected because they are very similar to the adult basophilic erythroblasts that we studied here . This analysis revealed that the proportions of G-quadruplexes and transcription start sites associated with both BrdU NS and K562 DNase I hypersensitivity peaks were respectively multiplied by 1 . 52 and 1 . 69 ( Figure 5C ) . Interestingly , chromatin accessibility barely increased the proportion of CpG islands associated with BrdU NS peaks . We therefore conclude that chromatin accessibility facilitates origin formation near G-quadruplexes and transcription start sites . Similar results were obtained when DNase I hypersensitivity data obtained from fetal liver basophilic erythroblasts was used ( Figure S4C ) . It is noteworthy that while DNase I hypersensitivity was strongly associated with origin formation , the majority of origins of replication are not DNase I hypersensitive ( Figure S4D ) . We conclude that chromatin accessibility favors but is not necessary for origin formation . We next analyzed the relationship between origins of replication and the ripples within the large timing domains that we detected in the high-resolution TimEX profiles . As illustrated in Figure 6A , visual examination suggested that the ripples in the timing profiles co-localized with clusters of origins of replication rather than with unique origins . This was not surprising since origins in mammalian cells often occur in zones of replication initiation [59] . Comparison of the high-resolution TimEX-seq with the BrdU NS data windowed at 3 kb revealed a high statistical correlation ( r = 0 . 40 , p<10E-5 ) between the earlier replicating regions of the high-resolution timing map and the NS . Smoothing the BrdU NS track to 20 Kb , to match the resolution of the high-resolution timing data , raised the coefficient of correlation to 0 . 54 ( Table S4 ) . To measure overlap between the timing ripples and the BrdU NS clusters , we called the summits of the ripples and of the clusters of NS in the smoothed BrdU NS data ( Figure S5 ) . We also computationally divided the TimEX profiles into fractions S1 to S5 each containing 20% of the genome , with S1 as the earliest replicating fraction . Analysis of this data revealed that most of the timing ripples and NS clusters were in the S1 and S2 fractions , and that 68% of the ripples in S1 overlapped with the NS clusters ( Figure S5 ) . Since the overlap expected by chance was only 12% , we conclude that the reproducible ripples detected in the high-resolution TimEX profiles reflect the activity of clusters of highly active early-replicating origins of replication . To characterize origin distribution in relation to timing of replication , we computed the density of NS reads per 5 kb intervals , the number of NS peaks , and the distance between called peaks . We also computed the mean area of the NS peaks which is an approximation of origin efficiency . This revealed dramatic differences in the distribution of the NS peaks between the five S phase fractions with almost exponential decreases in the average density of NS per genomic intervals , in the number of the NS peaks , and in origin efficiency ( Figure 6B ) . Conversely , the distance between NS peaks increased as S phase progresses . There were 12 times more NS peaks that were 38 times larger in S1 than in S5 . Finally , the median inter-peak distance was about 500 bp in the S1 fraction and about 14 , 000 bp in the S5 fraction . Therefore , origins of replication are much more numerous and much more efficient in early than in late S phase . Generally , these results were in excellent agreement with the recent results of [22] , [24] . However , since origin usage at the single molecule level is heterogeneous , the average inter-origin distances that we report here are likely smaller than the distances that could be measured on individual molecules . The dramatic changes in origin number and efficiency according to replication time raised the question of whether early and late origins have similar association with G-quadruplexes , CpG Islands and transcription start sites . Analysis revealed that the associations between BrdU NS peaks and G-quadruplexes , transcription start sites , and CpG Islands were similar regardless of replication time ( Figure 6C ) . Besnard et al . have shown that regions that display tissue-specific timing of replication have fewer origins of replication than regions that have generally invariant timing in multiple cell types , suggesting that variation in timing of replication might occur preferentially in origin poor regions . [24] . We therefore hypothesized that a paucity of origins of replication might also be a cause of asynchrony . To test this hypothesis , we computed the number of BrdU peaks and the density of BrdU reads/kb for the synchronous and asynchronous parts of the genome using the data from either FNY01_3_2 , FNY01_3_3 or the combined TimEX data for FNY01_3_2 and 3_3 . The asynchronous part of the genome being defined as all regions located within an ARD . As expected from the analysis of the high-resolution timing profiles depicted in Figure 6B , analysis of allele-specific timing profiles revealed that the average number of BrdU NS peaks and the average density of reads/peak was much higher in early than in late replicating regions in the synchronous part of the genome . The asynchronous regions exhibited a similar pattern . But the number of origins and the NS read density were slightly higher in the asynchronous region ( Figure 6D ) . Therefore , asynchronously replicating regions were not associated with low initiation frequency , but , on the contrary , tended to contain more origins of replication than their synchronous counterparts . These conclusions apply to the population level and might need to be tested at the single molecule level . We then asked whether asynchronous regions contained more G-quadruplexes , transcription start sites , and CpG islands than the synchronous regions . In accordance with the results of Figure 5 , the synchronous regions were on average richer in all three genomic features in early compared to late regions . Again , the asynchronous regions followed the same pattern but contained higher counts of all three genomic features ( Figure 6D ) . Finally , we asked whether asynchronous regions had a different CG and repeat content than the synchronous regions . Synchronous and asynchronous regions had very similar CG and repeat contents: CG and SINE contents were much higher in early compared to late regions , while LINE and LTR followed the inverse pattern ( Figure 6D ) . These variations in CG and repeat content were generally in accordance with previously published results [60] . Very similar results were observed for both individuals tested and when the combined results were analyzed . To determine whether the NSs were differentially distributed on the two homologs in the asynchronously replicated regions , we called and phased the SNPs on the BrdU NS data files , and we generated allele-specific maps of origins of replication for the maternal and paternal homologs . To assess the average NS density over large regions , we plotted the ratio of the number of NS reads in each of the ARDs as a function of the TimEX delays observed in FNY01_3_2 and 3_3 . Results demonstrated that the average NS read density was very similar on the two homologs ( Figure 7 ) . There was no correlation between the sign of the delays between the maternal and paternal homologs TimEX-seq signals and the ratios of the maternal and paternal BrdU NS . We conclude that timing asynchrony is not associated with a generalized change in origin efficiency on the two homologs .
Our results show that the timing of replication is very robustly coordinated in most of the genome . The non–allele specific high resolution profiles of FNY02_3_2 and 3_3 were almost identical . We show for the first time that in about 88% of normal somatic cells , the two chromosome homologs replicate at the same time during S-phase . This conclusion applies to the portion of the genome that we could access with our methodology and can likely be extrapolated to the entire euchromatic genome . Replication of constitutive heterochromatin cannot be assessed with current sequencing technologies . Comparison of the identical and non-identical genomic regions of the two siblings that we analyzed revealed that , at the resolution of this study ( about 400 kb ) , differences in the sequences of individual origins do not generally have a detectable effect at the level of the timing domains . Similarly , recent observations suggest that in human lymphoblastoid cells , replication profiles among autosomes ( and the active X chromosome ) were very similar when different individuals were compared [32] . Taken together , these observations imply either that the primary sequence of replication origins is only a minor determinant of replication initiation efficiency , or that alterations in the firing initiation rate of an origin can be compensated for by nearby origins and therefore do not affect the overall timing of replication in large replication domains . In contrast , we observed that large SVs were associated with timing asynchrony that could be detected in genomic segments up to one Mb suggesting that the replication timing domains can be altered by large genetic differences that affect multiple origins but not by change in the efficiency of single origins . These results support the idea that timing can be regulated by mechanisms acting at the level of entire domains since it demonstrates that replication time is influenced by sequences that are hundreds of kilobases away . To quantify asynchronous replication further , we developed a quantitative method to identify statistically significance ARDs and optimized the detection of these ARDs using simulated data . This approach revealed that there was about 600 ARDs larger than 500 , 000 bp in the genome of the two individuals tested . ARDs could reach several megabases in size . About half of the ARDs overlapped in the 2 individuals tested . About 20% of the genome was located in ARDs . This number is larger than the 12% asynchrony discussed above because our method to call the ARDs tends to overestimate their sizes and because the statistically significantly asynchronous domain encompassed smaller regions that do not appear asynchronous when analyzed in isolation . About 10% of the ARDs contained at least one known or predicted imprinted gene . ARDs that contain imprinted genes were longer and tended to be more highly statistically significant than ARDs that did not contain imprinted genes , suggesting that parental imprinted generates some of the most asynchronous regions in the genome . However , the vast majority of ARDs was not associated with large SVs or with imprinted genes . The mechanism of formation of these ARDs is unclear . The presence of these 600 large asynchronous regions is a strong argument in favor of the existence of regulatory mechanisms that act at the domain level to control the times of initiation , but are difficult to interpret in the context of a strict independent origin model because hundreds of replication origins have to be coordinated to generate megabase-size domains of asynchrony . Analyses of initiation profiles revealed that mapping newly replicated nascent strand DNA with the BrdU approach yields distinct patterns that are correlated to those obtained using lambda exonuclease analyses . Both approaches confirmed and extended previous observations that origins of replication are enriched near G-quadruplexes , CpG islands and transcriptional start sites ( TSS ) . Almost 50% of narrowly defined DNA segments that contain origins of replication detected by the BrdU method were near at least one of these three genomic features . These associations might be caused by the fact that G-quadruplexes , CpG islands and TSSs share sequence or chromatin characteristics that are necessary for origin formation . Alternatively , they might reflect an evolutionary selective advantage that stabilizes distinct genomic features ( e . g . prevent replicative destabilization of G-quadruplexes [61] ) when placed near replication initiation sites . The association of replication events with regulatory chromatin features could also reflect distinct origin types that would each be associated with particular genomic features and coordinate DNA replication with transcription and chromatin condensation [23] . The association between origins of replication and G-quadruplexes , CpG islands , and TSSs was higher in regions of open chromatin , as assessed by DNAse I hypersensitivity , suggesting that the presence of nucleosomes can decrease the probability of origin firing . However , the majority of origins were not located in Dnase I hypersensitive regions suggesting that the replication machinery can overcome these barriers . Importantly , both the BrdU and lambda exonuclease approach revealed that replication initiation takes place near promoters and CpG Islands more often than expected by chance throughout S phase , including during late S phase in regions where no transcription occurs . This implies that the replication machinery recognizes promoters and CpG islands even when they are in a closed chromatin . The association between origins , promoters and CpG islands is therefore not dependent on the transcriptional function of these regions . Whether the replication machinery directly recognize the primary DNA sequence of these regions or epigenetic marks that retain the memory of previous transcription events remains to be determined . Methylation of histone H3 on lysine 79 , which associates with some replication origins might be part of this process [62] . In any case , these data show that the primary DNA sequence and local chromatin structure at the nucleosome levels are critical determinants of where each origin of replication can be located . Therefore , the putative domain-wide epigenetic mechanisms that regulate initiation of replication likely act by modulating the efficiency of origins that are located in DNA regions specified in large part by the primary DNA structure . Analysis of the frequency and efficiency of origins of replication in the ARDs revealed that there were more origins that were in average more efficient in asynchronous than in synchronous regions . Therefore ARDs do not occur preferentially in regions of low origin density . Analysis of the high resolution timing profiles revealed a novel correlation between ripples in the timing of replication profiles and clusters of peaks in the nascent strand profiles . This important observation provides an orthogonal validation for the TimEX-seq and the BrdU and nascent strand approaches , and a novel method to detect initiation zones in timing profiles . The large majority of the nascent strands that we detected were in early replicating genomic regions . The number of reads in NS peaks in late S phase was less than one percent that in early S phase because the peaks were larger and spaced closer together in early regions . Although our measurement of average levels in a population of cells mask the heterogeneity that can be observed at the single molecule level , the much larger quantity of NS that we detected in early replicating regions requires an explanation since DNA replicates only once per cell cycle . One hypothesis is that late replicating regions are in part replicated passively from initiation events that occurred earlier in the cell cycle . This likely contributes to the excess NS in early S phase but the magnitude of this excess and the size of the late replication regions which are often too far from early origins to be replicated passively , suggest that other factors might also be at play . An additional hypothesis is that transcription competes with DNA replication in early S phase since both of these process relies on very large protein complexes that cannot access the DNA strand at the same time . This latter hypothesis is attractive because it also provides a potential explanation for the apparent stochastic nature of initiation within replication initiation zones observed at the single molecule level , since interactions between two independent processes is a well-established deterministic explanation for apparently stochastic events . Interference with transcription and its associated complex chromatin structure could lead to increased origin activity and higher production of NS in early regions by slowing down the replication , thus providing the opportunity for nearby origins to fire rather than being replicated passively . In addition , a slower rate of progression of the replication fork would increase the half-life of short nascent strands and therefore increase their detection by both the lambda exonuclease and the BrdU approaches . The larger number of origins of replication and the higher efficiency of firing that we observed in early versus late S phase might therefore be due to both increased production and increased detection of the nascent strand in early S phase . A higher number of initiation events in early S phase as compared to late S phase would lead to a smaller replicon size in early S . Several studies have shown that this might be the case [63]–[67] . The existence of interference between transcription and replication is also supported by the observation that origins are often located in moderately expressed promoters but are excluded from very active promoters [23] , and by the dramatic changes in the location of origins after the initiation of transcription during development of Xenopus Laevis [68] and mammalian cells [67] . An important question is: how are multi-megabases late domains replicated ? These regions cannot be replicated passively because they are too far from early origins . The fewer origins of replication we observed in later stages of S-phase is consistent with the recent report that late replicating regions ( including the inactive X chromosome ) replicated faster than early regions and exhibit a random order of initiation [32] . This replication pattern might suggest either that replication in these regions does not initiate at consistent origins , or that replication initiation occurs synchronously at many concurrent origins . We detected a large number of low efficiency origins in these regions that are , similarly to early origins , located near G-quadruplexes , transcription start sites and CpG islands more often that would be expected by chance . We propose that large heterochromatic regions might be replicated from a subset of replication origins that are activated late in S phase and that can each replicate much larger regions than early origins because of the regular structure and lack of transcriptional activity of heterochromatin .
This work was performed under approved Einstein IRB protocol number 2011-356-000 . 10 to 20 ml of peripheral white blood cells were harvested by venipuncture from individuals from family FNY01 and mononuclear cells were isolated by density gradient centrifugation on Histopaque ( Sigma-Aldrich ) according to the manufacturer's instructions . The purified cells were frozen in 2 million cell aliquots . Two million mononuclear cells were expanded and differentiated into basophilic erythroblasts in culture for two weeks in serum-free Stemspan media ( Stem Cells Technologies , VA , CA ) containing the cytokine cocktail mix described by Olivier et al . [37] . At the end of the culture , cells were immuno-phenotyped by FACS using antibodies against CD71 and CD235a . Cells were relatively uniform in size and more than 97% of the cells were double-positives demonstrating that the vast majority of cells in the culture were erythroid and at the basophilic stage of differentiation . Genomic DNA was sheared to 300–600 bp size using a focused ultra-sonicator from Covaris ( Woburn , MA ) , size purified by agarose gel electrophoresis , and analyzed on a DNA Chip using a 2100 bio-analyzer ( Agilent Technologies , Inc . ) to verify size distribution . DNA was end-repaired ( End-It kit , Epicenter Biotechnologies ) , an A-overhang was added ( Klenow 3′ to 5′ exo minus , NEB ) , and ligated to Illumina pair-end adapter sequences ( Illumina ) . The ligated libraries were size selected ( 600±50 bp ) , PCR amplified for 10 cycles , and purified with SPRI beads ( Agencourt AMPure XP; Fisher Scientific ) . Pair-end library quality was verified on a bio-analyzer as described above . Each library was sequenced on three lanes on the Illumina HiSeq 2000 yielding about 109 , one hundred bp pair-end reads corresponding to about 30× coverage for each S phase library . After library sequencing , the reads were aligned to hg 19 with bwa [69] using the default parameters and the SNPs were called using the GATK variant caller in the known-allele mode [70] , providing the vcf file that describes the genomes of family FNY01 [36] as a reference file . In this mode , the GATK calls only the SNPs at the positions specified in the user-provided reference vcf file . Once the SNPs had been called , the vcf files for the S phase libraries was phased using the vcf for family FNY01 as a reference and specific functions in GenPlay multi-genome [71] . Once the files were phased files ( in the BED format ) containing the allele depth for the paternal and maternal chromosomes were generated , again using specific functions that are provided in GenPlay . The read depth files were binned in to 500 bp windows , normalized to the total number of reads containing a SNP in each track and smoothed using the GenPlay Gaussian filter ( sigma = 100 kb ) . S/G1 ratios for the paternal and maternal chromosomes were then calculated . The resulting numbers were multiplied by a factor 1 . 4 to take into account the fact that the average copy number in S is higher than in G1 . These resulted in TimEX profiles with value ranging between 0 . 9 and 2 . 1 . No other filter or normalization was applied to the data . Masks were created to eliminate the regions of low SNP-density using GenPlay island finder functions . The resulting masks were applied in selected analysis to ascertain that the results were independent of SNP density . Other smoothing techniques ( Loess , moving average ) yielded essentially the same results . The combined FNY01_3_2&3_3 allele-specific replication profiles were produced by generating four new files: a combined G1 maternal file containing the sum of the allele-depths of the maternal homologs of FNY01_3_2 and 3_2; a combined G1 paternal file containing the sum of the allele depths of the paternal homologs of FNY01_3_2 and 3_2 , and similarly , two combined maternal and paternal S files . These four files were then processed as above to generate combined maternal and paternal S/G1 profiles . Asynchronously replicated regions were detected with the GenPlay island finder on a data file generated by subtracting the paternal S/G1 simulated profiles from the maternal S/G1 simulated profiles . The GenPlay Island finder function is based on the SICER algorithm of Zang et al . [72] and identifies broad regions of enriched read counts rather than peaks . The algorithm allows for gaps in the island and is well suited for the allele-specific TimEX data because of the heterogeneous distribution of the SNPs in the human genome . Implementation of the algorithm in GenPlay is described in the GenPlay documentation in the GenPlay web site . A TimEX difference threshold of 0 . 02 , a maximum gap sizes of 250 , 000 and a minimum island size of 50 , 000 bp were used in the island finder . Statistically significant regions were then identified using a qui-square test for goodness of fit ( chisq . test function in R ) on 2×2 contingency tables built by calculating the total number of reads obtained in the S and G1 fraction of the maternal and paternal chromosomes for each islands . Q-values were then calculated using the p . adjust function in R ( with the fdr parameter ) to control for multiple testing . An FDR rate of 5% was used to define the statistically significant asynchronous replication domains . The parameters for the island finder were determined by optimizing by trial and error , the rate of discovery of simulated regions ( see below ) . Core detection: Core were detected as ARD except that a TimEX difference threshold of 0 . 1 , a maximum gap sizes of 50 , 000 and a minimum island size of 50 , 000 bp were used in the GenPlay Island finder . Statistical significance was determined as above . Simulated maternal and paternal pairs of S and G1 resampled control profiles were generated by imputing in a random number generator based on the binomial function ( similar to the rbinom function in R ) , the number of S reads and the total number of reads ( S+G1 reads ) observed for each informative SNP in the maternal track of FNY01_3_2 . About 1 , 000 randomly located simulated asynchronous genomic regions ( ranging in size from 125 , 000 to 2 , 000 , 000 bp ) were then introduced in the simulated maternal control S profiles by increasing by a fixed percentage ( 5 , 10 , 15 , 20 , 30 or 50% ) , the total ( S+G1 ) number of reads observed for each SNP and by assigning the extra reads to the S phase profile . These S and S+G1 read numbers were then imputed in the random generator to generate resampled read numbers in these simulated asynchronous regions . Once simulated number of reads for the S and G1 profiles of the maternal and paternal homologs had been generated , simulated S/G1 TimEX profiles were generated as described above using Gaussian smoothing ( sigma = 100 kb ) . Bedgraph files were generated to visualize the simulation in GenPlay . Asynchronously replicated regions in these simulated profiles were then detected with GenPlay island finder as described above . Simulations were repeated multiple times . A Java script implementing this algorithm can be found at https://github . com/JulienLajugie/ReplicationTimingSimulation . S and G1 aligned reads obtained after alignment to hg19 were loaded into 1 kb bins in GenPlay , normalized to the total number of reads , S to G1 ratios were calculated , outliers were eliminated ( S/G1>2 . 4 ) , and the profiles were smoothed with the GenPlay Gaussian filter ( sigma = 20 kb ) . Use of other smoothing techniques ( Loess , moving average ) yielded essentially the same results . For more convenient analysis values were then indexed between 1 and 2 . Nascent , newly replicated DNA strands from cultured basophilic erythroblasts were isolated following two protocols . First , we have isolated RNA-primed short newly replicated DNA following the procedure described in Martin et al . [23] . Briefly , DNA was extracted from asynchronous cultures , denatured and fractionated by centrifugation on a neutral sucrose gradient . DNA fractions within the 0 . 5 kb to 1 kb size range were collected and DNA was exposed to lambda exonuclease to remove non-RNA-primed genomic DNA fragments . Purified RNA-primed NS were random-primed and the resulting double-stranded nascent DNA ( 1 µg ) was sequenced using the Illumina ( Solexa ) genome analyzer II . Second , to control for possible lambda exonuclease biases , we have also isolated and sequenced short nascent DNA strands that were isolated based on incorporation of the nucleotide analog BrdU as described by Aladjem et al . [40] . Briefly , we labeled cells with a 30-minute pulse of BrdU , lysed the cells and fractionated small DNA strands ( <1 kb ) using a neutral sucrose gradient . We then employed immune-precipitation with anti-BrdU antibodies to isolate newly replicated DNA strands on the basis of selective incorporation of BrdU . The resulting NS were subject to massively parallel sequencing . For both methods , sheared genomic DNA was sequenced as a standard to control for mapability and other potential biases . Nascent libraries were prepared and sequenced on two lanes of an Illumina GII analyzer . Read length was 36 bp . Reads were aligned to hg19 using bwa and loaded in GenPlay in bins of 100 bp , 3 kb or 20 kb ( depending on the analysis ) . Peaks were called using MACS 1 . 42 using the default parameters , or the GenPlay Island finder . The ratios of the TimEX values of the two chromosome homologs for individuals FNY01 3_2 and 3_3 were calculated in 5 kb windows and averaged over 500 kb intervals ( overlapping every 100 kb ) . Ratios smaller than 1 were inverted , since we expected the SVs to either increase or decrease the timing ratios . SVs present in the family are described in detail in [36] . Similar results were obtained with interval of 1 Mb , with non-overlapping intervals of 0 . 5 or 1 Mb , or by averaging the distance or the absolute values of the differences between the S/G1 ratios of the homologs rather than their ratios . Hi-res TimEX profiles of FNY01 3_2 and 3_3 were averaged and binned at 3 kb . NS were binned at 200 bp and smoothed with the GenPlay Gaussian filter using sigma 3 kb or 20 kb . The tracks were then converted to 3 kb bins after saturating the outliers ( i . e . all bins with scores larger than X with x = mean plus 3 times the standard deviation were set to a score equal to X ) . The averaged TimEX profiles were then divided into 5 equal fractions termed S1 to S5 based on the TimEX values ( with S1 replicating the earliest and S5 the latest ) . Correlation and linear regression were performed in R or in Statgraphics Centurion XVII ( Statpoint Technologies incorporated ) . Removing the outliers rather than saturating them yielded similar results . Peaks were called using MACS v1 . 42 using the default value ( p<10E-5 and FDR<0 . 2% ) . Sheared genomic DNA from the same individual was used as the control . The number of peaks observed with the BrdU and lambda exonuclease methods was similar ( respectively 250 , 000 and 250 , 00 ) , the average widths of the peaks were also similar , about 100–400 bp in both cases . To facilitate the comparison between the methods , all peaks were normalized to a width of 400 bp which resulted in a 50% reduction in peak number and the analysis was performed on the top 100 , 000 peaks only . Peaks were considered to be associated to a genomic features if they overlapping by one base pair or more with a CpG Island , a transcription start site ( defined as 1 kb fragment centered on the start site ) or with 400 bp fragments centered around a G-quadruplexes or a DNase I hypersensitive site . GC content was calculated as the average percent of C and G in 100 kb windows ( in 10 kb intervals ) . Because of the very uneven baseline , MACS or the Genplay island finders were not able to satisfactorily detect and call the ripples within the larger TimEX domains that are illustrated in Figure 4 . To call these peaks we therefore devised a specific strategy illustrated in Figure S5 . Briefly , the TimEX profiles were smoothed with the Gaussian filters and a sigma at 20 kb ( which revealed the ripples ) and at 100 kb ( which smooth out these ripples ) and the 100 kb smooth was subtracted from the 20 kb smooth . The top 10 percent of the peaks thus obtained were then analyzed since they coincided closely with the major ripples that we wished to characterize . The same strategy was used to define the clusters of NS . All statistical analysis were performed using R or using StatGraphics Centurion XVI ( Statpoint technologies ) Most of the data processing and visualization was performed in GenPlay multi-Genome , an application that was written to display , produce and process allele-specific and non-allele-specific sequencing data files . GenPlay is available freely at http://genplay . einstein . yu . edu .
|
DNA replication in mammalian cells proceeds according to a distinct order . Genes that are expressed tend to replicate before genes that are not expressed . We report here that we have developed a method to measure the timing of replication of the maternal and paternal chromosomes separately . We found that the paternal and maternal chromosomes replicate at exactly the same time in the large majority of the genome and that the 12% of the genome that replicated asynchronously was enriched in imprinted genes and in structural variants . Previous experiments have shown that chromosomes could be divided into replication timing domains that are a few hundred thousand to a few megabases in size . We show here that these domains can be divided into sub-domains defined by ripples in the timing profile . These ripples corresponded to clusters of origins of replication . Finally , we show that the frequency of initiation in asynchronous regions was similar in the two homologs .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"developmental",
"biology",
"functional",
"genomics",
"cell",
"biology",
"chromosome",
"biology",
"genomic",
"imprinting",
"gene",
"expression",
"genetics",
"gene",
"regulation",
"biology",
"and",
"life",
"sciences",
"epigenetics",
"molecular",
"genetics",
"genomics",
"chromatin"
] |
2014
|
Allele-Specific Genome-wide Profiling in Human Primary Erythroblasts Reveal Replication Program Organization
|
Currently , malaria rapid diagnostic tests ( RDTs ) are widely used for malaria diagnosis , but test performance and the factors that lead to failure of Plasmodium ovale detection are not well understood . In this study , three pLDH-based RDTs were evaluated using cases in China that originated in Africa . The sensitivity of Wondfo Pf/Pan , CareStart pLDH PAN and SD BIOLINE Pf/Pan in P . ovale detection was 70 , 55 and 18% , respectively . CareStart was worse at detecting P . o . curtisi ( 36 . 5% ) than at detecting P . o . wallikeri ( 75 . 0% ) , and SD could not detect P . o . curtisi . The overall detection ratio of all three RDTs decreased with parasite density and pLDH concentration . Wondfo , CareStart and SD detected only 75 . 0 , 78 . 1 and 46 . 9% of the P . ovale cases , respectively , even when the parasitemia were higher than 5000 parasites/μL . Subspecies of P . ovale should be considered while to improve RDT quality for P . ovale diagnosis to achieve the goal of malaria elimination .
Plasmodium ovale has a wide geographic distribution across tropical countries , especially in Africa , Asia and some Western Pacific islands [1] . P . ovale has been overshadowed by other human malaria parasites in the field of medicine and medical research because of the relatively low morbidity and infections can be easily treated with conventional antimalarial drugs [2] . Although considered mild , P . ovale can cause acute respiratory distress syndrome and acute renal failure [3] . In addition , in the context of the long-term goal of eliminating malaria , it is becoming important to diagnose P . ovale in a timely manner . P . ovale , which shares with P . vivax the ability to form hypnozoites [4 , 5] , can cause chronic infections that may last from months to even years . Rapid and reliable diagnosis is one of the key factors for malaria control and elimination . Accurate identification of Plasmodium infections is critical for administration of a targeted therapy , having a positive impact on patient health , disease management , and preventing transmission risk . Accurate diagnosis of malaria is needed to prevent the emergence and spread of drug resistant strains and to reduce the cost of medicine . The gold standard for malaria diagnosis remains the examination of Giemsa-stained smears by microscopy . This technique requires considerable training and experience . It is very challenging to maintain the capacity for microscopic examination in areas where malaria is being or has been eliminated [6] . In addition , P . ovale infection is difficult to diagnose microscopically owing to the generally low parasite density in patients and the morphology of P . ovale resembles that of P . vivax . Malaria rapid diagnostic tests ( RDTs ) allow many countries to provide access to accurate malaria diagnosis even in the most remote areas by means of a simple-to-use , point-of-care test [7] . RDTs are easy to use and require no specific training or equipment . The results are visually readable as colored lines on a strip , and no special expertise is required . In recent decades , RDTs have replaced microscopy as the method of choice for diagnosing malaria in various settings . Reported sensitivities vary among different RDTs but are generally good for the detection of Plasmodium falciparum [8–11] . However , the sensitivity of RDTs for P . ovale detection is much lower , only 5 . 5% to 80% , with a sharp decrease observed at parasite densities lower than 500 parasites /μL [8–9 , 11–13] . Based on a few assessments with a very limited number ( n = 69–76 ) of tested samples , failure of P . ovale detection by RDTs has been reported [12 , 14–15] . Malaria RDTs exhibited suboptimal performance in the detection of P . ovale infections [16] , but the factors that affect the efficiency of RDTs in the detection of P . ovale have not been systemically investigated . Targets of malaria RDTs are specific antigens of one or more Plasmodium species , such as histidine-rich protein ( HRP2 ) , lactate dehydrogenase ( LDH ) , and aldolase . Among these antigens , Plasmodium-specific lactate dehydrogenase ( pan-pLDH ) is commonly used as a target of RDTs to detect all human Plasmodium species [14] . In addition , P . ovale is comprised of two genetic subspecies , namely , P . ovale curtisi and P . ovale wallikeri , and genetic variations based on P . ovale LDH gene polymorphism could also be involved in RDT failure [17] . We hypothesized that parasite density , level of pLDH , subspecies of P . ovale or polymorphism of pLDH gene might be involved in the failure of P . ovale detection by pLDH-based RDTs . In this study , the performance of three pLDH-based RDTs ( Wondfo diagnostic kit for malaria ( Pf/pan ) ( colloidal gold ) , CareStart Malaria pLDH ( PAN ) , and SD BIOLINE Malaria Ag Pf/Pan ) for P . ovale detection were retrospectively evaluated with blood samples from returned international travelers and laborers from Africa to China . Moreover , the possible factors affecting RDT detection , i . e . , parasite density , pLDH concentration , genetic subspecies and pLDH gene polymorphism , were investigated .
This study was reviewed and approved by the Institutional Ethics Committee of Jiangsu Institute of Parasitic Diseases ( JIPD ) ( IRB00004221 ) . All participants were adults and written informed consent was obtained from all participants before the interview or evaluation . Blood samples were selected from venous EDTA-blood samples stored at -70°C , which were obtained from febrile patients at the clinics of local hospitals in Jiangsu province , China , and sent to the provincial reference laboratory . The patients were international travelers and laborers from African countries . The diagnosis of P . ovale was determined by both microscopy and confirmed by a nested PCR assay using the following commonly used protocols [18] . Thick and thin blood films were prepared from peripheral blood . Blood smears were stained with 3% Giemsa for 30 min at room temperature to identify parasites . Smears were analyzed by experienced microscopists at the JIPD . The standard method recommended by the World Health Organization ( WHO ) was used to estimate the number of circulating parasites per μL of blood . Parasite density was determined by counting the parasites and leucocytes , assuming 8 , 000 leucocytes /μL [19] . All the slides were subjected to double-blind verification by another independent microscopist , and the results were combined . Three pLDH-based RDTs were used in this study: the Wondfo diagnostic kit for malaria ( Pf/pan ) ( colloidal gold ) ( Guangzhou Wondfo Biotech Co . , Ltd . ; lot W05440903WC ) , CareStart Malaria pLDH ( PAN ) ( Access Bio , Inc . ; lot MN13G01 ) and SD BIOLINE Malaria Ag Pf/Pan ( Standard Diagnostics Inc . ; lots 05ED14111 and 05ED15003 ) . The RDTs were selected based on the WHO/FIND malaria RDT performance evaluations and national guidelines of China . All of the RDTs were packed and sealed individually with desiccant and used immediately after opening and were performed based on the instructions of the manufacturers . The major target antigens of these three RDTs were pLDH , which are specific for all human-associated Plasmodium species . Five microliters of stored whole-blood samples were added to the pad , and three to four drops of specific lysis agent were added . The RDT result was read in 15–20 min according to the manufacturer’s instructions and immediately recorded by one person , a second person read the results 5 min later after the first person and was blinded to the initial reading . In the event of discordant results , a third person read the test blindly also and final results was the most common reading . The test was considered valid when the control line on the immunochromatographic test strip was visible . For the Pf/Pan test device ( Wondfo diagnostic kit for malaria ( Pf/pan ) and SD BIOLINE Malaria Ag Pf/Pan ) , the result was recorded as non-falciparum only when the pan-pLDH line was positive . For the pLDH PAN test device ( CareStart Malaria pLDH ( PAN ) ) , the presence of two colored bands ( one band in the control and another band in the test ) indicated a positive result for Plasmodium infection . All three RDTs were performed once . Quantitative levels of pLDH antigens in P . ovale samples were determined using the Quantimal pLDH Malaria CELISA test ( Cellabs ) , a sandwich ELISA for the detection of human Plasmodium pLDH . All P . ovale blood samples were tested in triplicate , and the manufacturer’s instructions were followed . Plates were read on a Zenyth 340 microplate spectrophotometer ( Autobio ) at 450 nm , with a reference wavelength of 620 nm . The reading was completed within 30 min after the stop solution was added . The mean optical density ( OD ) was calculated with the cut-off value as the means plus three standard deviations ( SDs ) of the wells containing healthy human blood alone . Genomic DNA was extracted from 200 μL of whole-blood samples of P . ovale-infected patients using a QIAamp Blood Mini Kit ( Qiagen ) according to the manufacturer’s instructions . DNA extracted from healthy individuals living in nonendemic areas was used as a negative control in the amplification process . The real-time TaqMan PCR ( qPCR ) assay was used to detect P . ovale subspecies as described in a previous publication [20] . Amplification was performed with the following set of primers: POF ( 5′-ATAAACTATGCCGACTAGGTT-3′ ) and POR ( 5′-ACTTTGATTTCTCATAAGGTACT-3′ ) . The probe pPOW HEX-AATTCCTTTTGGAAATTTCTTAGATTG-BHQ1 was used for detection of P . o . wallikeri , and pPOC FAM-TTCCTTTCGGGGAAATTTCTTAGA-BHQ1 was used for detection of P . o . curtisi . The qPCRs were carried out using LightCycler TaqMan Master ( Roche , Germany ) on a Roche LightCycler 480 ( Roche , Germany ) under the following conditions: one step at 95°C for 10 min; 45 cycles at 95°C for 15 sec and 60°C for 60 sec; and a final step at 4°C for 10 sec . Nucleotide sequences corresponding to P . ovale LDH genes were amplified with the primers LDHovD21 ( 5′-GTTCTCGTTGGTCAGGAATGATA-3′ ) and LDHovC915 ( 5′-GGCATCATCAAACATCTTCTTTTCT-3′ ) by conventional PCR using Dream Taq Green PCR Master Mix ( Thermo Scientific ) . Primer design and PCR conditions were based on a previous publication [20] . The PCR products were sequenced by Genescript Biological Technology Co . , Ltd . ( Nanjing , China ) . Nucleotide sequences of P . ovale LDHs were aligned using BioEdit software and compared with available P . ovale sequences in GenBank ( accession number AY486058 ) and a paper by Bauffe et . al . [20] . Moreover , amino acid sequences were derived using GeneDoc software and also compared with available P . ovale sequences . The RDT results for P . ovale detection were compared with each other , and sensitivity was calculated with 95% confidence intervals ( CIs ) by STATA ( version 12 . 0 ) . Categorical variables were determined by Chi-squared tests , with Fisher’s exact correction applied when the expected frequency in any cell was 5 or less . Correlation analyses between parasite density and pLDH OD level were performed as Pearson’s correlation analyses . A Pearson r value greater than 0 . 6 was considered a strong correlation . P values less than 0 . 05 were considered significant for all statistical analyses .
A total of 100 samples containing P . ovale parasites were studied . The samples had been collected from Feb 2012 to June 2015 . The average age of the patients was 41 . 27 years ( range , 22–58 years ) , and all the P . ovale infections were acquired in Africa . A majority ( 38/100 ) of the P . ovale samples were from Equatorial Guinea , followed by Nigeria ( 18/100 ) , Angola ( 17/100 ) and the Republic of Congo ( 9/100 ) ( Table 1 ) . The performances of the three pLDH-based RDTs ( Wondfo Pf/pan , CareStart pLDH ( PAN ) and SD Pf/Pan ) were compared for P . ovale detection . Of all the 100 confirmed P . ovale samples , 70 tested positive with the Wondfo Pf/Pan device , 55 tested positive with the CareStart pLDH device , and only 18 tested positive with the SD Pf/Pan device ( Table 2 ) . All three RDTs exhibited very high false negative rates for P . ovale detection . To investigate whether the subspecies of P . ovale and the variations in P . ovale LDH polymorphism were associated with the sensitivity of the three RDTs for P . ovale detection , a real-time TaqMan PCR ( qPCR ) assay was carried out for subspecies determination , and the pLDH gene was sequenced for polymorphism analysis . Among the 100 P . ovale samples , 52 were P . o . curtisi and 48 were P . o . wallikeri , as determined by qPCR . The two subspecies exhibited no difference in parasitemia and pLDH levels ( P>0 . 05 ) . There was no significant difference in the sensitivity of the detection of the two subspecies by the Wondfo Pf/Pan device ( χ2 = 0 . 49 , P = 0 . 485 ) . A higher sensitivity was observed for P . o . wallikeri than that for P . o . curtisi ( approximately 75% vs 36 . 5% ) with the CareStart Pan device with statistically significance ( χ2 = 14 . 92 , P<0 . 0001 ) . On the other hand , the SD Pf/Pan device could not detect P . o . curtisi at all , and a high false negative rate ( 62 . 5% ) was observed for P . o . wallikeri detection with this device ( χ2 = 20 . 88 , P<0 . 0001 , Fisher’s exact<0 . 0001 and one-sided Fisher’s exact<0 . 0001 ) ( Table 3 ) . The amplified P . ovale LDH gene yielded approximately 890 base pairs , coding for 294 amino acids . A total of 100 of the amplified genes were sequenced to analyze the genetic variations in the P . ovale LDH gene using Clustal W2 software . No nucleotide substitution was detected within the sequences of each subspecies compared to the reference sequences ( P . o . curtisi from GenBank ( AY486058 ) and P . o . wallikeri from a paper by Bauffe et . al . ) . There were twenty-four single nucleotide polymorphisms ( SNPs ) between the two subspecies , and three of these SNPs were nonsynonymous mutations ( S143P , N168K , I204V ) , which was consistent with the results of a previous study [20] . The performance of each RDT was evaluated according to parasite density levels . Based on P . ovale parasite densities , the 100 samples were divided into three groups . For the Wondfo Pf/Pan device , the sensitivity for cases with parasite densities greater than 500 parasites/μL ( 75 . 4% and 75% ) was higher than that for cases with densities lower than 500 parasites/μL ( 27 . 3% ) , and this difference was statistically significant ( χ2 = 10 . 75 , P<0 . 05 , Fisher’s exact = 0 . 007 ) . However , for the SD Pf/Pan device , regardless of parasite density , the sensitivity was less than 50% , and the cases with parasite densities lower than 500 parasites/μL could not be detected with this device and a significant difference was observed ( χ2 = 26 . 76 , P<0 . 0001 , Fisher’s exact<0 . 0001 ) . The sensitivity of the CareStart Pan device reached 78 . 1% for only those cases that had parasite densities greater than 5000 parasites/μL , and a significant difference was observed ( χ2 = 10 . 49 , P = 0 . 005 , Fisher’s exact = 0 . 005 ) ( Table 4 ) . To determine the relationship between pLDH concentration and the sensitivity of the three RDTs for P . ovale detection , the P . ovale samples were divided into three groups according to the pLDH OD levels , and the sensitivity of the three RDTs was evaluated in each group . For the Wondfo Pf/Pan and CareStart Pan device , the sensitivity for P . ovale detection reached 100% and 89 . 7% when the pLDH OD levels in the samples were more than 0 . 5; the sensitivity increased with the pLDH levels; and a significant difference was observed ( χ2 = 82 . 05 , P<0 . 0001 , Fisher’s exact<0 . 0001 for Wondfo; χ2 = 32 . 09 , P<0 . 0001 , Fisher’s exact<0 . 0001 for CareStart ) . On the other hand , the sensitivity of the SD Pf/Pan device increased with the pLDH levels , and a significant difference was also observed ( χ2 = 38 . 52 , P<0 . 0001 , Fisher’s exact<0 . 0001 ) , but the sensitivity reached only 55 . 2% , even when the pLDH level was high ( >0 . 5 OD ) ( Table 5 ) . To determine whether the pLDH levels were associated with parasitemia in these P . ovale samples , the correlation between the pLDH levels and peripheral blood parasitemia was assessed . A moderate correlation was observed between pLDH levels and parasitemia ( r = 0 . 5510 , P value<0 . 0001 ) . Some disagreement was observed between parasitemia and pLDH levels . Ten cases of P . ovale samples with parasitemia greater than 10 , 000 parasites/μL presented low levels of pLDH ( OD<1 ) , and five cases with low parasitemia ( <10 , 000 parasites/μL ) presented high pLDH levels ( OD>1 ) ( Fig 1 ) .
The results of this study suggested that the performance of the three pLDH-based RDTs for P . ovale detection was not optimal . The low parasite density and pLDH concentration contributed to the failure of the RDT test for P . ovale . The subspecies of P . ovale can affect the sensitivity of the detection of P . ovale for the CareStart Pf/Pan and SD PAN RDTs but not the Wondfo Pf/Pan RDT , and the P . ovale LDH gene was relatively well conserved among the subspecies . Therefore , malaria diagnosis might be difficult using only RDTs , especially for P . ovale infections . The present results in the study could provide more aspects for producing better RDTs with significantly improved sensitivity for P . ovale .
|
Plasmodium ovale ( P . ovale ) are under-estimated and overshadowed by other malaria parasites in tropical countries , which can cause chronic infections that last from months to years . The chronic infection caused by P . ovale should be of concern in the context of the long-term goal of eliminating malaria . Rapid diagnostic tests ( RDTs ) is one of the WHO recommended tools to confirm the infection of plasmodium parasites , which can distinguish Plasmodium falciparum and non-falciparum species as well . However , little is known about their performance detecting P . ovale , and the factors that affect the efficiency of RDTs in the detection of P . ovale have not been systemically investigated . This study suggested that the performance of the three pLDH-based RDTs for P . ovale detection was not optimal , the low parasite density and pLDH concentration contributed to the failure of the RDT test for P . ovale . It provided information for the application of malaria RDTs in the field and for research and development to improve RDTs for malaria diagnosis .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"parasite",
"groups",
"medicine",
"and",
"health",
"sciences",
"body",
"fluids",
"respiratory",
"infections",
"plasmodium",
"tropical",
"diseases",
"parasitic",
"diseases",
"parasitic",
"protozoans",
"pulmonology",
"parasitology",
"parasitemia",
"apicomplexa",
"protozoans",
"molecular",
"biology",
"techniques",
"quantitative",
"parasitology",
"research",
"and",
"analysis",
"methods",
"malarial",
"parasites",
"artificial",
"gene",
"amplification",
"and",
"extension",
"molecular",
"biology",
"eukaryota",
"blood",
"anatomy",
"polymerase",
"chain",
"reaction",
"physiology",
"biology",
"and",
"life",
"sciences",
"malaria",
"organisms"
] |
2019
|
Assessment of false negative rates of lactate dehydrogenase-based malaria rapid diagnostic tests for Plasmodium ovale detection
|
Clostridium difficile is a Gram-positive , spore-forming anaerobic bacterium that infects the colon , causing symptoms ranging from infectious diarrhea to fulminant colitis . In the last decade , the number of C . difficile infections has dramatically risen , making it the leading cause of reported hospital acquired infection in the United States . Bacterial toxins produced during C . difficile infection ( CDI ) damage host epithelial cells , releasing erythrocytes and heme into the gastrointestinal lumen . The reactive nature of heme can lead to toxicity through membrane disruption , membrane protein and lipid oxidation , and DNA damage . Here we demonstrate that C . difficile detoxifies excess heme to achieve full virulence within the gastrointestinal lumen during infection , and that this detoxification occurs through the heme-responsive expression of the heme activated transporter system ( HatRT ) . Heme-dependent transcriptional activation of hatRT was discovered through an RNA-sequencing analysis of C . difficile grown in the presence of a sub-toxic concentration of heme . HatRT is comprised of a TetR family transcriptional regulator ( hatR ) and a major facilitator superfamily transporter ( hatT ) . Strains inactivated for hatR or hatT are more sensitive to heme toxicity than wild-type . HatR binds heme , which relieves the repression of the hatRT operon , whereas HatT functions as a heme efflux pump . In a murine model of CDI , a strain inactivated for hatT displayed lower pathogenicity in a toxin-independent manner . Taken together , these data suggest that HatR senses intracellular heme concentrations leading to increased expression of the hatRT operon and subsequent heme efflux by HatT during infection . These results describe a mechanism employed by C . difficile to relieve heme toxicity within the host , and set the stage for the development of therapeutic interventions to target this bacterial-specific system .
Clostridium difficile is a spore-forming , Gram-positive obligate anaerobe that is the most common cause of nosocomial infections in the United States [1] . C . difficile infects the colon , causing a wide range of diseases that vary from infectious diarrhea to pseudomembranous colitis . During infection , C . difficile produces two potent toxins , TcdA and TcdB , which cause severe damage to intestinal epithelial cells resulting in inflammation , fluid secretion , and necrotic cell death [2 , 3] . Perforations in the intestinal epithelial layer lead to bleeding in the gut and subsequent translocation of erythrocytes into the gastrointestinal lumen [4] . Hemolysis due to pathophysiological stress occurs , resulting in the release of hemoglobin-bound heme and free heme at the site of damage [5] . Heme , an iron-containing porphyrin , is the most abundant source of iron in the human body and many invading pathogens have evolved mechanisms to utilize this rich metabolic resource [6–10] . Owing to its reactive nature , heme is toxic to bacteria at high concentrations through a variety of mechanisms [11–14] . In order to defend against the stresses of heme-mediated damage , bacteria encode systems for heme sensing and detoxification [15–21] . In the Gram-positive pathogens Staphylococcus aureus and Bacillus anthracis , the heme stress response is controlled by the heme sensing two-component system , HssRS , which regulates transcription of the ABC transporter HrtAB to reduce heme toxicity through efflux [15 , 16 , 22] . Reducing intracellular heme levels through export is a conserved microbial strategy as heme efflux systems have also been identified in Lactococcus lactis , Streptococcus agalactiae , and Neisseria gonorrhoeae [17 , 18 , 23] . In each example , inactivation of heme detoxification machinery increases heme sensitivity and modulates virulence [15–18 , 23] . Notably , C . difficile does not contain orthologs of known heme detoxification systems , and it is also unknown if this organism encounters heme during infection . The overall goal of this study was to investigate the occurrence of heme exposure to C . difficile within the gastrointestinal lumen during infection . Here , we visualize increased abundance of hemoglobin in the gastrointestinal lumen as a result of CDI using imaging mass spectrometry . A heme-inducible operon was identified that contains a TetR family transcriptional regulator and major-facilitator superfamily transporter . We have named these gene products HatRT for heme activated transporter ( R = regulator , T = transporter ) . The transcriptional regulator HatR responds to intracellular heme concentrations through binding of heme leading to the de-repression and increased transcription of hatRT . Lack of the HatT transporter results in increased intracellular heme concentrations and a decrease in pathogenicity in a murine model of infection . Taken together , these results describe a mechanism by which C . difficile detoxifies heme and establishes a requirement for heme sensing and detoxification for full virulence during C . difficile infection .
To identify host proteins that increase in abundance during C . difficile infection ( CDI ) , we applied matrix-assisted laser desorption ionization imaging mass spectrometry ( MALDI IMS ) to a murine model that induces susceptibility to infection through administration of cefoperazone ( 0 . 5 mg/mL ) [24–27] . Ceca of mice infected with C . difficile R20291 presented with high levels of epithelial damage , edema , and inflammation on day 4 of the infection ( Fig 1A ) . This inflammatory response correlated with a high abundance of the alpha chain of hemoglobin at the sites of pathology and in the luminal space ( Fig 1B and S1 Fig ) . In contrast , mouse ceca mock infected with PBS did not exhibit pathology ( Fig 1A ) and displayed a low abundance of hemoglobin alpha concentrated at the periphery of the intestinal epithelial villi ( Fig 1B and S1 Fig ) . These data demonstrate that CDI leads to high concentrations of hemoglobin at this host-pathogen interface , and considering each hemoglobin protein contains four molecules of heme , support a model whereby C . difficile experiences heme stress during infection . Considering the high concentration of hemoglobin in the infected lumen and the reactive nature of heme , we investigated the sensitivity of C . difficile to heme toxicity [13] . When C . difficile was grown over time in increasing concentrations of heme ( 0–200 μM ) , a dose-dependent increase in toxicity was observed with a complete inhibition of growth at the highest concentration ( Fig 2A ) . To determine if C . difficile can adapt to heme exposure , the growth of C . difficile cells pre-exposed to a low concentration of heme ( 1 μM ) was measured following sub-culturing into media containing varying concentrations of heme ( 0–200 μM; Fig 2B ) . Heme pre-exposure corrected the growth defects of C . difficile cultures not pre-exposed to heme ( Fig 2A and 2B ) , suggesting that C . difficile has an inducible mechanism for heme detoxification . In order to identify the genes that encode proteins responsible for heme adaption , we performed an RNA-sequencing experiment comparing the total relative mRNA transcript abundance of early exponential phase ( OD600 = 0 . 3 ) untreated cultures of C . difficile to cultures grown in 50 μM heme . Heme induced the transcription of 245 genes and decreased the transcription of 146 genes ( Fig 2C; S3 and S4 Tables ) . This dataset was curated by grouping significantly upregulated genes that could function as a mechanism of heme sensing and detoxification . Within this group an operon of two genes encoding a TetR family transcriptional regulator ( CDR20291_1227 ) and a major facilitator super family ( MFS ) transporter ( CDR20291_1226 ) were identified as candidates for further investigation ( Fig 3A ) . These results demonstrate that C . difficile has heme responsive genes that may account for its ability to resist and adapt to heme toxicity . To confirm that CDR20291_1227 and CDR20291_1226 are heme responsive , cultures were grown in equimolar concentrations of NaOH ( vehicle ) , protoporphyrin IX ( porphyrin ring without iron ) , iron sulfate , or heme prior to harvesting RNA at the early exponential phase of growth ( OD600 = 0 . 3 ) . Quantitative reverse transcription PCR ( qRT-PCR ) was performed on cDNA generated from these samples using primers specific for the genes within this operon . Transcription of both genes were minimally increased in the samples treated with NaOH , protoporphyrin IX , and iron sulfate in contrast to a 2–3 log increase in transcript abundance of the heme treated samples compared to the untreated control ( Fig 3B ) . Due to this considerable transcriptional response to heme , as well as data described below , we named CDR20291_1227 heme activated transporter regulator ( hatR ) and CDR20291_1226 heme activated transporter ( hatT ) . To investigate the heme responsive abundance of HatR , polyclonal antiserum was generated against recombinant HatR and immunoblot analyses were performed on whole cell lysates grown in increasing concentrations of heme . The increase in HatR protein abundance correlated with the increase in concentration of heme , further supporting the observation that hatRT is up-regulated upon heme exposure ( Fig 3C ) . To demonstrate the specificity of this antisera , we generated a strain of C . difficile inactivated for hatR ( hatR::CT ) using the ClosTron system [28] . In this strain , HatR is no longer produced in response to heme ( Fig 3C ) . The lack of hatR renders the bacteria more sensitive to heme toxicity , as growth over time in the presence of 50 μM heme is delayed in the mutant compared to wild-type ( WT; Fig 3D ) . A more significant growth delay is observed when hatT is inactivated ( hatT::CT ) using the ClosTron system and exposed to the same concentration of heme ( Fig 3D ) . The growth of hatR::CT and hatT::CT strains are restored to WT levels by expressing hatR or hatT , respectively , in trans under the control of the intergenic region upstream of hatR ( S2 Fig ) . Together these data suggest that HatR and HatT coordinate to sense , respond to , and alleviate heme toxicity . As most members of the TetR family of transcriptional regulators directly bind their effector molecules , we examined the ability of HatR to bind heme [29] . Recombinant HatR ( 10 μM ) was incubated with heme ( 0–25 μM ) , resulting in the appearance of a Soret peak at 413 nm ( Fig 4A ) , indicative of HatR-heme complex formation [30] . Differential absorption spectroscopy at 413 nm over a range of heme concentrations was used to determine that HatR binds heme at a 1:1 ratio using a single site binding model ( kd = 9 . 2 ± 1 . 8 μM; Fig 4A insert ) . To identify the residues responsible for heme binding by HatR , each histidine within HatR was individually mutated . Histidine residues were chosen for substitution due to histidines commonly serving as axial ligands that bind heme [17] . Recombinant proteins containing each individual histidine substitution were purified and heme binding was measured . Substitution of histidine 99 to leucine ( H99L ) was sufficient to abrogate heme binding ( Fig 4B ) . The substitutions of the remaining four histidines to alanine or leucine ( H121A , H126L , H165A , and H180A; S3 Fig ) did not significantly alter heme binding . These data specify histidine 99 as a critical residue in the formation of the HatR-heme complex . An examination into the regulation of the hatRT operon was performed by creating a plasmid containing a fusion of the intergenic region prior to hatR to the reporter gene xylE , and transforming this plasmid into WT C . difficile [22] . Exposure of this reporter strain to 10 μM heme led to a significant increase in XylE activity as compared to an untreated control ( Fig 4C ) , indicating that heme treatment induces the transcription of the hatRT operon in the WT strain . However , upon transformation of the reporter plasmid into the hatR::CT strain , there was no significant difference in XylE activity between the untreated or heme exposed samples ( Fig 4C ) . Moreover , the level of XylE activity of the untreated hatR::CT strain was significantly higher than the heme-exposed WT strain , suggesting constitutive expression of xylE in the absence of HatR . Taken together , these data suggest that HatR functions as a transcriptional repressor of the hatRT operon and that de-repression is achieved through the formation of a HatR-heme complex . One strategy for microbial heme detoxification involves the reduction of intracellular heme concentrations through efflux [15–18 , 23] . To investigate if heme efflux is responsible for HatT-dependent resistance to heme toxicity , we grew the WT and hatT::CT strains in the presence or absence of heme ( 25 μM ) for 16 h and measured intracellular heme concentrations utilizing LC-MS analysis . The WT strain treated with heme exhibited a two-log increase in intracellular heme levels when compared to untreated WT cells ( Fig 5 ) . In contrast , a more dramatic trend was observed in the hatT::CT strain , which exhibited a three-log increase in intracellular heme concentration when compared to the hatT::CT untreated culture ( Fig 5 ) . The intracellular heme concentration was over 40-fold higher in the hatT::CT strain treated with heme when compared to the WT strain treated with heme ( Fig 5 ) . These data , combined with the heme sensitivity of the hatT::CT strain , suggest that the function of HatT is to reduce intracellular heme concentrations to relieve heme toxicity . The abundance of heme in the lumen during infection combined with the observed functions of HatR and HatT to sense and reduce heme concentrations , suggest that strains lacking these proteins may have reduced pathogenicity during CDI . To test this , mice were infected with WT , hatR::CT , or hatT::CT spores and disease was monitored for 4 days . All strains were able to fully colonize the mice as exhibited by ~108 colony-forming units ( CFU ) per gram of stool ( Fig 6A ) . Mice infected with the hatT::CT strain lost significantly less weight than the mice infected with the WT or hatR::CT strains on days 3 and 4 of the infection ( Fig 6B ) , indicating that the mice infected with the hatT::CT strain were partially protected despite similar colonization levels . Furthermore , cecal pathology was significantly reduced in mice infected with the hatT::CT strain compared to mice infected with WT or hatR::CT strains ( Fig 6C ) . To determine whether the reduced virulence of the hatT::CT strain is due to a reduction in toxins TcdA or TcdB , we assessed toxin production in the WT , hatR::CT , and hatT::CT strains using a cell-rounding cytotoxicity assay . These data revealed toxin levels to be equivalent between all tested strains on day 4 of the infection ( Fig 6D ) , suggesting that the reduced virulence of the hatT::CT strain in vivo is independent of C . difficile toxins . Taken together , these data suggest that the hatRT operon senses and detoxifies intracellular heme in C . difficile and is required for full pathogenicity during infection .
C . difficile infection of the colon causes severe epithelial cell damage , inflammation , and edema , which leads to the hallmarks of C . difficile-colitis . Importantly , this damage and subsequent inflammatory response also creates a hostile environment for bacteria within the gut [2 , 31–33] . Highly reactive heme molecules that can be toxic to bacteria are released into the lumen through erythrocyte lysis and necrotic epithelial cell death [3 , 5] . Despite the hazard of heme toxicity , C . difficile thrives in the colon and survives in the presence of high heme levels . Prior to this work , the mechanism by which C . difficile resists heme toxicity were unknown . Herein , we visualized the high abundance of hemoglobin during infection , serving as a proxy for heme , in the murine ceca during CDI . We identified a molecular mechanism encoded by the hatRT operon to sense and detoxify heme in C . difficile . HatR functions as a transcriptional repressor of the hatRT operon and responds to heme concentrations through direct binding of heme . HatR-heme complexes de-repress the hatRT operon , leading to the HatT-mediated reduction in intracellular heme concentrations , presumably through efflux . In support of these data , strains with inactivated hatR or hatT exhibited delayed growth in the presence of heme and the hatT::CT strain conferred reduced pathology in a toxin-independent manner in a mouse model of CDI . While heme sensing and detoxification through efflux is a conserved strategy in multiple Gram-positive organisms , this report is the first to describe an obligate anaerobic pathogen containing such a system [16–18 , 34] . TetR-family transcriptional regulators that bind heme have been identified , including HrtR in Lactococcus lactis , whereby HrtR regulates heme efflux through a system orthologous to HrtAB [17 , 35] . However , HatR shares limited sequence homology ( 38% amino acid identity ) with HrtR . Additionally , the heme binding motifs ( single histidine versus two histidines ) and the heme-complex disassociation constant ( HatR kd = 9 . 2 ± 1 . 8 μM , HrtR kd = 0 . 4 ± 0 . 2 μM ) differ between HatR and HrtR [17] . The significant overexpression of the hatRT operon in the presence of heme but not protoporphyrin IX or iron suggests the formation of the heme-HatR complex involves direct binding to the coordinated iron center of heme . The increased heme sensitivity in the hatR::CT strain despite the constitutive expression of hatT , suggests HatR may also function to reduce heme toxicity through sequestration . The eventual in vitro growth observed when the hatT::CT strain is exposed to high heme suggests the existence of other mechanisms of heme detoxification in C . difficile or the occurrence of suppressor mutations to relieve intracellular heme concentrations through a different transport system . A bioinformatics comparison of HatT with S . aureus HrtAB and the dual S . agalactiae efflux system PefAB/CD , suggests that these systems arose through convergent evolution as there is little homology between these transporters despite their important role in heme detoxification [18 , 34] . The mechanisms of heme toxicity in bacteria are not completely understood . In an anaerobic environment , heme toxicity has been attributed to membrane disruption and DNA damage due to the hydrophobic structure of heme [12–14 , 36] . Bilirubin , the terminal metabolite in heme catabolism in mammals is present in high concentrations in the gastrointestinal tract , and destabilizes the membrane of Gram-positive bacteria , suggesting that heme degradation products may also contribute to toxicity [37] . In C . difficile , heme enters the intracellular compartment through an unknown mechanism . It is also not known if C . difficile utilizes heme as a cofactor or metabolite . Bioinformatic analyses do not reveal heme degradation enzymes of the IsdG or HO enzyme families in C . difficile [38–40] . Additionally , it appears as if C . difficile cannot use heme as a sole iron source [41] . In this study , we demonstrated that heme accumulates in the cytoplasm of C . difficile and is subsequently detoxified through removal by HatT . Results reported in this work demonstrate the importance of heme detoxification in CDI as the hatT::CT strain was less pathogenic in a mouse model of infection . The colonization of the WT , hatR::CT , and hatT::CT strains are at similar levels , supporting a model in which resistance to heme toxicity is important for the end stages of acute infection after serious injury to the intestinal epithelium has occurred . This observation is further supported by the reduction in disease that was only observed on days 3 and 4 following infection in the hatT::CT infected mice . The lack of phenotype of the hatR::CT strain suggests that continual expression of hatT in the absence of HatR in this strain is sufficient to cause full disease . Surprisingly , there were no differences in bacterial burdens at these days or differences in toxin production despite less overall pathology in the hatT::CT infected mice . This suggests C . difficile utilizes either additional heme detoxification operons or compensatory mechanisms to relieve intracellular heme stress outside of HatRT and reveals the importance of toxin-independent mechanisms of virulence . Alternatively , as C . difficile has been shown to occupy different nutritional niches during infection , and heme is heterogeneously distributed throughout the infected ceca , the heme sensitive strains may be able to maintain WT levels of colonization due to occupying niches of reduced heme concentrations at a cost of pathogenicity [42] . These results provide a molecular insight into how C . difficile adapts to the harsh environment of the inflamed gut . Further studies must be performed to elucidate additional mechanisms of protection that C . difficile utilizes to survive during infection .
All animal experiments under protocol M1700053 were reviewed and approved by the Institutional Animal Care and Use Committee of Vanderbilt University . Procedures were performed according to the institutional policies , Animal Welfare Act , NIH guidelines , and American Veterinary Medical Association guidelines on euthanasia . Strains used in this study are listed in S1 Table . C . difficile strains were grown at 37°C in an anaerobic chamber ( 85% nitrogen , 10% hydrogen , 5% carbon dioxide , Coy Lab Products ) in brain-heart-infusion broth ( BD Life Sciences ) supplemented with 0 . 5% yeast extract ( BD Life Sciences ) and 0 . 1% cysteine ( Sigma-Aldrich ) ( BHIS ) or in C . difficile minimal media ( CDMM ) described previously [43] . Escherichia coli strains were grown in lysogeny broth ( LB ) or agar ( LBA ) , supplemented with 50 μg/mL kanamycin or 50 μg/mL carbenicillin when necessary [43] . Bacillus subtilis strains were grown on LBA or in BHI broth supplemented with 5 μg/mL tetracycline or 2 . 5 μg/mL chloramphenicol . All antibiotics were purchased from Sigma-Aldrich . hatR::CT and hatT::CT strain generation . Gene inactivations were achieved using the ClosTron system as described previously [44] . Briefly , gBlocks containing specific modifications for insertion into the genome were generated using the TargeTronics algorithm ( http://www . targetrons . com ) and synthesized by Integrated DNA Technologies . The gBlocks were cloned into pCR-Blunt vector using the Zero Blunt PCR cloning kit ( ThermoFisher Scientific ) followed by restriction digest with BsrgI and HindIII ( NEB ) and ligation ( NEB T4 ligase ) into pJS107 . Plasmids were transformed into the recA+ E . coli MG1655 through a standard heat shock protocol followed by transformation into B . subtilis JH2 using an established method [44] . B . subtilis strains containing the pJS107_hatR or pJS107_hatT plasmids were mated with C . difficile R20291 overnight at 37°C by plating and mixing together 100 μL of each strain onto a BHIS plate in the anaerobic chamber . Plates were scraped and transferred into 2 mL of BHIS prior to plating 200 μL onto BHIS plates containing 20 μg/mL thiamphenicol and 50 μg/mL kanamycin ( BHISthiamp20kan50 ) . Colonies from these plates were patched onto new BHISthiamp20kan50 and BHIS plates containing 5 μg/mL tetracycline ( BHIStet5 ) . Patched colonies that were tetracycline sensitive were patched again onto new BHISthiamp20kan50 and BHIStet5 plates . Colonies that remained tetracycline sensitive were streaked onto BHIS plates containing 20 μg/mL lincomycin ( BHISlinc20 ) . Inactivation of the hatR or hatT gene was confirmed by performing PCR to identify a 1 . 5 kbp shift in gene size using gDNA extracted as previously described on colonies that were lincomycin resistant [44] . xylE reporter and complementation plasmids . Reporter and complementation plasmids ( S1 Table ) were created by GenScript using the pJS116 plasmid as a backbone for the synthesized intergenic region ( 236 bp ) of hatR fused to the xylE reporter gene , the intergenic and full coding region of hatR , and intergenic region of hatR fused to the full coding region of hatT . C . difficile strains were transformed as described above with the removal of the lincomycin selection and were maintained on BHISthiamp20 to ensure plasmid retention . Protein expression plasmids . Protein expression plasmids for HatR were generated by amplifying hatR flanked by BamHI and XhoI and cloning into the multiple cloning site of pLM302 after restriction digest . Point mutant generation in pLM302_hatR was performed with NEB Q5 Site Directed Mutagenesis kit according to the manufacturer’s instructions , using the primers listed in S2 Table . Mutations to Ala or Leu were governed by the surrounding protein motifs and retention of spatial arrangement . Heme toxicity growth assays . Freshly streaked bacterial colonies were used to inoculate 5 mL of BHIS or BHISthiamp20 and grown for 16 h at 37°C . Cultures were subcultured 1:50 into fresh BHIS or BHISthiamp20 and grown for 6 h at 37°C prior to 1:50 inoculation into CDMM or CDMMthiamp20 containing heme at the indicated concentrations . All growth assays were performed in a 96-well plate in 200 μL of media . Optical density at 600 nm ( OD600 ) served as measurement of growth and was measured every 30 min for the indicated total time in an EpochII microplate reader ( BioTek ) . C . difficile were grown anaerobically in triplicate in CDMM in 0 or 50 μM heme . Hemin ( Sigma ) was solubilized in 0 . 1 M NaOH . The cultures were grown at 37°C to an OD600 of 0 . 3 abs . Upon reaching this density , a 1:1 solution of acetone:ethanol was added to an equal volume of the culture . Samples were stored at -80°C until used for RNA extraction . Samples were thawed on ice , pelleted , and resuspended in 750 μL of LETS buffer ( 1 M LiCl , 0 . 5 M EDTA , 1 M Tris pH 7 . 4 ) . Cells were transferred to tubes containing lysing matrix B beads ( MP Biomedicals ) and lysed by a FastPrep-24 ( MP Biomedicals ) bead beater for 45 s at 6 m/s . Lysed samples were heated for 5 min at 55°C and pelleted by centrifugation for 10 min . The supernatant was transferred to a fresh tube and 1 mL TRIzol ( Thermo Scientific ) was added . Chloroform ( 200 μL ) was added to each sample and vortexed prior to separation of the aqueous and organic layers by centrifugation for 15 min . The aqueous ( upper ) layer was transferred to a fresh tube and the RNA was precipitated through the addition of 1 mL isopropyl alcohol . Samples were incubated for 10 min and RNA was pelleted by centrifugation for 10 min . Supernatant was removed and the RNA pellet was washed with 200 μL of 70% ethanol . Samples were air dried for 1 min , then resuspended in 100 μL RNase free water . DNA contamination was removed through the addition of 8 μL RQ1 DNase , 12 μL 10x RQ1 buffer , and 2 μL RNase inhibitor ( Promega ) to the purified RNA . Samples were DNase treated for 2 h and purified using the RNeasy miniprep RNA cleanup kit ( Qiagen ) . RNA concentration was determined using the Synergy 2 with Gen 5 software ( BioTek ) . RNA-seq library preparation and sequencing . RNA-seq library construction and sequencing was performed by HudsonAlpha . Concentration was determined using the Quant-iT RiboGreen RNA assay ( Thermo Scientific ) and integrity was visualized using an RNA 6000 nano chip ( Agilent ) on an Agilent 2100 Bioanalyzer ( Applied Biosystems ) . RNA was normalized to 500 ng of total RNA for each sample and the ribosomal RNA ( rRNA ) was removed using Ribo-Zero rRNA Removal Kit ( Illumina ) . Directly after rRNA removal , the RNA was fragmented and primed for first strand synthesis using the NEBNext First Strand synthesis module ( New England BioLabs Inc . ) followed by second strand synthesis using NEBNext Ultra Directional Second Strand synthesis kit . Library preparation was achieved using NEBNext DNA Library Prep Master Mix set for Illumina with minor modifications . PolyA addition and custom adapter ligation was performed following end–repair . Post-ligated samples were individually barcoded with unique in-house Genomic Services Lab ( GSL ) primers and amplified through 12 cycles of PCR . Library quantity was assessed by Qubit 2 . 0 Fluorometer ( Invitrogen ) , and quality was determined using a DNA High Sense chip on a Caliper Gx ( Perkin Elmer ) . Final quantification of the complete libraries for sequencing applications was measured using the qPCR-based KAPA Biosystems Library Quantification kit ( Kapa Biosystems , Inc . ) . Libraries were diluted to 12 . 5 nM and pooled equimolar prior to clustering . Paired-End ( PE ) sequencing was performed on an Illumina HiSeq2500 sequencer ( Illumina , Inc . ) . Raw sequence data are deposited on the NCBI Sequence Read Archive . Processing of RNA-seq reads . RNA-seq analysis was performed by HudsonAlpha utilizing their unique in-house pipeline . Briefly , quality control was performed on raw sequence data from each sample using FastQC ( Babraham Bioinformatics ) . Curated raw reads were imported into the data analysis platform , Avadis NGS ( Strand Scientifics ) and mapped to the reference C . difficile R20291 genome . Aligned reads were filtered on various criteria to ensure the highest read quality . Replicate samples were grouped and quantification of transcript performed using Trimmed Means of M-values ( TMM ) as the normalization method . Differential expression of genes was calculated using fold change ( using default cut-off ≥ ±2 . 0 ) observed between conditions , and the p-value of the differentially expressed gene list was estimated by Z-score calculations using determined by Benjamini Hochberg FDR correction of 0 . 05 [45] . The genome alignment figure ( Fig 2A ) was created using Circos with a max of 30 , 000 RPKM displayed . RNA was extracted as described above and 2 μg was reverse transcribed by M-MLV reverse transcriptase ( Fisher Scientific ) in the presence of RNase inhibitor ( Promega ) and random hexamers ( Promega ) . Reactions lacking the reverse transcriptase were used to control for DNA contamination . Newly created cDNA was diluted 1:100 and was used in qRT-PCR using iQ SYBR green supermix ( BIO-RAD ) utilizing the primer pairs in S2 Table . Amplification was achieved using a 3-step melt cure program on a CFX96 qPCR cycler ( BIO-RAD ) . Transcript abundance was calculated using the ΔΔCT method normalized by the rpoB gene . HatR was purified as described below and fresh protein was submitted to the Vanderbilt Antibody and Protein Resource core for generation of a rabbit polyclonal antibody against HatR . This antibody was affinity purified for increased HatR specificity . The α-HatR antibody was tested for specificity and reactivity in immunoblot analysis of purified HatR protein in addition to whole cell lysates from heme treated WT and hatR::CT strains . WT or hatR::CT strains were grown in 5 mL of BHIS overnight at 37°C . Cultures were subcultured into fresh BHIS containing 0 , 1 , 5 , 10 or 25 μM heme and grown for 6 h . Cells were pelleted by centrifugation ( 4000 x g for 10 min ) , supernatant was removed and were resuspended in 1 mL of 1 X PBS containing 2 . 5 mg/mL lysozyme ( ThermoFisher Scientific ) . Samples were incubated for 1 h at 37°C , pelleted by centrifugation ( 20 , 000 x g for 5 min ) , then resuspended in 1 X PBS followed by sonication using Ultrasonic dismembrator ( ThermoFisher Scientific ) to lyse the cells . Debris from the lysed cells was pelleted by centrifugation ( 20 , 000 x g for 5 min ) . Supernatant was used in immunoblotting analysis using rabbit polyclonal α-HatR antibodies as previously described [46] . Detection was performed using a goat anti-rabbit IgG ( H+L ) cross-adsorbed secondary antibody with an Alexa Fluor 680 and imaged using a ChemiDoc MP imaging system ( Bio-Rad ) . E . coli BL21 ( DE3 ) pREL containing the pML302_hatR plasmids were grown overnight in 5 mL of LBkan50 at 37°C . Cells were subcultured into Terrific broth ( ThermoFisher Scientific ) containing 50 μg/mL kanamycin and grown to the mid-logarithmic phase of growth ( 0 . 5 abs measured at 600 nm ) at 37°C prior the addition of 1 mM isopropyl-1-thiol-D-galactopyranoside ( IPTG ) . Growth was continued at 16°C for 16 h . Cells were harvested by centrifugation ( 6000 x g for 10 min ) and resuspended in 1 X PBS . Cells were lysed by passage through an EmulsiFlex homogenizer ( Avestin ) three times at 20 , 000 lb/in2 . The insoluble debris was removed by centrifugation at 40 , 000 x g for 1 h and the supernatant was filtered using a 0 . 22-μM-pore sizer filter . Filtered lysate was added to amylose resin ( New England Biolabs Inc . ) and allowed to bind at 4°C for 30 min prior to transfer to a gravity column . The column was washed with four column volumes of wash buffer ( 20 mM Tris-HCl , 500 mM NaCl , 1 mM EDTA , pH 7 . 5 ) three times followed by 2 column volumes of elution buffer ( 20 mM Tris-HCl , 500 mM NaCl , 1 mM EDTA , 10 mM maltose , pH 7 . 5 ) twice . The maltose-binding protein tag ( MBP ) was cleaved using the Pierce HRV 3C Protease Solution kit ( ThermoFisher ) by following the manufacturer’s instructions . Cleaved tag and protease were removed by the addition of HisPur Cobalt Resin ( ThermoFisher ) and allowed to bind at 4°C for 1 h with rotation . Beads were pelleted by centrifugation ( 2000 x g for 2 min ) and the supernatant containing tagless protein was removed . Heme binding by HatR were determined by measuring the absorption spectrum of increasing amounts of hemin ( 0–25 μM ) after addition to a cuvette containing 10 μM recombinant HatR in 1 mL of Tris-buffered saline ( TBS ) and a reference standard containing 1 mL TBS on a Varian Cary 50BIO . Samples were mixed and allowed to incubate at room temperature in the dark for 5 min prior to collecting the spectrum between 300–800 nm with 10 nm increments . Binding ratio of heme to HatR was determined by plotting the change in absorbance at 413 nm between the reference standard and the HatR sample . A curve fit and ratio was obtained by performing the one-site binding model non-linear regression function on Graph Pad Prism 6 . Bacteria harboring the reporter plasmid pJS116_phatR-xylE were grown overnight in BHISthiamp20 and subcultured 1:50 into 10 mLs of fresh BHISthiamp20 containing 0 or 10 μM heme . Cultures were grown for 6 h at 37°C prior to cytoplasmic fraction preparation and analysis of XylE activity as described previously [16] . Absolute XylE activities were determined spectrophotometrically by measuring the formation of 2-hydroxymoconic acid from catechol for C . difficile reporters due to lysozyme interference during protein quantification . Adult ( 8–12 week old ) age-matched male C57Bl/6 ( Jackson Laboratories ) were housed in groups of five and maintained at Vanderbilt University Medical Center Animal Facilities . Mice were subjected to a previously described model of CDI [24 , 27] . Briefly , mice were treated with 0 . 5 mg/mL cefoperazone in their drinking water for 5 days . Mice were given a 2 day recovery period prior to administration of 105 spores of WT , hatR::CT , or hatT::CT C . difficile strains in PBS via oral gavage . Prior to infection , mice were confirmed to be C . difficile negative . After infection , mice were monitored for signs of disease , including diarrhea and weight loss . Mice that displayed severe disease or weight loss greater than 20% were humanely euthanized . Green African monkey kidney epithelial ( Vero , ATCC CCL-81 ) cell-rounding cytotoxicity assays were performed as previously described [27] . Cells were grown to confluence in Dulbecco modified Eagle medium ( DMEM , Gibco Laboratories ) with 1% penicillin-streptomycin ( Gibco Laboratories ) and 10% fetal bovine serum ( Gibco Laboratories ) prior to plating at a total cell density of 105 cells per well in a 96-well plate . Fresh fecal samples were normalized to weight , diluted and homogenized in sterile PBS . Fecal debris was pelleted by centrifugation ( 13 , 000 x g for 5 min ) and tenfold serial dilutions of supernatants were added to the wells of Vero cells . Complete cell-rounding for each dilution was assessed after overnight incubation at 37°C with 5% CO2 . Confirmation of C . difficile toxin A and toxin B were achieved by neutralization of cell rounding with a combined antitoxin antisera ( Techlab ) . Cell rounding cytotoxicity titers are presented as the log10 of the reciprocal value of the highest dilution with complete rounding of cells . All data analysis and statistical tests were performed in GraphPad Prism X software . Specific statistical tests , replicate numbers , calculated errors and other information for each experiment are reported in the figure legends .
|
Clostridium difficile is a pathogenic bacterium that infects the colon and is the leading cause of infectious diarrhea in the United States . C . difficile mediated disease is driven by the production of two toxins , TcdA and TcdB . The toxins cause severe damage to the intestinal epithelial layer of the colon resulting in inflammation and bleeding . Once in the intestinal lumen , red blood cells lyse , resulting in an abundance of extracellular heme at the site of C . difficile infection . Due to the highly reactive nature of heme , elevated concentrations are toxic to bacteria . Here we describe a C . difficile system that allows for survival in the presence of heme intoxication . These results present a mechanism employed by C . difficile to sense and reduce intracellular heme concentrations to relieve toxicity . We also demonstrate the importance of heme detoxification during C . difficile infection , as strains lacking this system display reduced virulence in a murine model of infection . Collectively , these data provide insight into a mechanism that C . difficile utilizes to survive in the host environment during infection .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] |
[
"heme",
"medicine",
"and",
"health",
"sciences",
"gut",
"bacteria",
"pathology",
"and",
"laboratory",
"medicine",
"animal",
"models",
"of",
"disease",
"microbiology",
"operons",
"toxicology",
"animal",
"models",
"toxicity",
"model",
"organisms",
"gastroenterology",
"and",
"hepatology",
"experimental",
"organism",
"systems",
"dna",
"bacteria",
"research",
"and",
"analysis",
"methods",
"clostridium",
"difficile",
"animal",
"models",
"of",
"infection",
"animal",
"studies",
"proteins",
"mouse",
"models",
"gastrointestinal",
"infections",
"biochemistry",
"post-translational",
"modification",
"nucleic",
"acids",
"detoxification",
"genetics",
"biology",
"and",
"life",
"sciences",
"organisms"
] |
2018
|
Heme sensing and detoxification by HatRT contributes to pathogenesis during Clostridium difficile infection
|
Ocular dominance plasticity is a well-documented phenomenon allowing us to study properties of cortical maturation . Understanding this maturation might be an important step towards unravelling how cortical circuits function . However , it is still not fully understood which mechanisms are responsible for the opening and closing of the critical period for ocular dominance and how changes in cortical responsiveness arise after visual deprivation . In this article , we present a theory of ocular dominance plasticity . Following recent experimental work , we propose a framework where a reduction in inhibition is necessary for ocular dominance plasticity in both juvenile and adult animals . In this framework , two ingredients are crucial to observe ocular dominance shifts: a sufficient level of inhibition as well as excitatory-to-inhibitory synaptic plasticity . In our model , the former is responsible for the opening of the critical period , while the latter limits the plasticity in adult animals . Finally , we also provide a possible explanation for the variability in ocular dominance shifts observed in individual neurons and for the counter-intuitive shifts towards the closed eye .
Throughout development , sensory cortex can experience periods of heightened sensitivity to sensory inputs . The rewiring of neuronal networks is very flexible during these periods , but there is less such plasticity otherwise . Having normal sensory experiences during these periods is crucial for a healthy maturation of the brain and they are therefore called critical periods ( CP ) . A well studied example is the critical period for ocular dominance ( OD ) in primary visual cortex ( V1 ) . In the visual pathway , inputs from both eyes usually converge onto the same neuron for the first time in V1 , although a fraction of thalamic neurons already exhibits binocularity in mice [1–3] . The extent to which a neuron’s visually-evoked activity is dominated by one of the eyes is called ocular dominance ( OD ) and is often quantified by the ocular dominance index ( ODI ) . In each hemisphere of mice V1 , the overall response to the contralateral eye is roughly twice as high as that to the ipsilateral eye , but individual neurons display a broad range of ODI values . During a limited period early in the development , closure of one eye for multiple days triggers a shift in neuronal responses towards the open eye . In mice , this critical period spans about ten days , starting around postnatal day 20 . The changes in neuronal responses following this monocular deprivation ( MD ) can be roughly separated into two phases . In a first phase , observed during the first three days of deprivation , the responses to the closed eye are depressed while responses to open-eye inputs remain similar . This effect is often called response depression . For longer deprivations , a second phase follows where the neuronal responses to the open eye are increased , called response potentiation . In this second phase , the neuronal activity caused by the closed eye also increases , but to a lesser extent [4] . Further insights into the working of ocular dominance plasticity are uncovered by studying other deprivation paradigms . Firstly , binocular deprivation ( BD ) does not lead to OD shifts , hinting at some level of competition depending on the strength or coherence of the inputs from both eyes . Secondly , monocular inactivation ( MI ) abolishes the rapid response depression , suggesting that this response depression is activity dependent , relying on spontaneous activity and residual activity caused by light travelling through the closed eyelid during MD [4] . Before and after the critical period , the effects of monocular deprivation on ocular dominance plasticity are either reduced or not observed at all . In pre-CP mice , monocular deprivation leads to a decrease in activity from both eyes , thus not changing their relative strengths and not affecting the overall ocular dominance [5] . In adult mice , the response depression after short monocular deprivation is not observed . However , longer deprivation still leads to the response potentiation of the open eye , and hence a certain shift in ocular dominance can still be observed [6] . A key player in regulating the opening of the critical period is the maturation of inhibition . GAD95-KO mice , which exhibit an impaired γ-aminobutyric acid ( GABA ) release , never experience a critical period and visual cortex remains in a juvenile state . A critical period can be opened in these mice once per lifetime after diazepam infusion , which restore the GABA release . Similarly , diazepam infusion before normal CP onset can accelerate the start of the CP in wildtype mice [7] . Furthermore , a recent experimental study investigated changes in cortical layer II/III excitation and inhibition in juvenile animals after only 24 hours of deprivation [8] . The authors found that the firing rate of parvalbumin-positive ( PV+ ) inhibitory neurons is decreased at that time , while the firing rate of excitatory neurons is increased . Moreover , they show that this decreased inhibition is predominantly mediated by a reduction in excitatory drive from layer IV and V to these PV+ neurons . Interestingly , this reduction of inhibition is not observed in adult animals . The authors then linked this effect to the OD shift , by showing how pharmacological enhancement of inhibition during the critical period prevents any OD plasticity , while pharmacological reduction of inhibition in the adult animals results in an OD shift towards the open eye . It was therefore postulated that OD plasticity depends on the increased firing rate of open eye inputs , caused by a transient reduction in inhibition . The mechanisms behind the closure of the critical period remain more enigmatic . However , several manipulations can reopen a window for OD plasticity in adult mice . Firstly , reducing inhibition was shown to enhance the OD plasticity caused by monocular deprivation [8 , 9] . Related to this , adult mice in enriched environments were shown to have reduced levels of inhibition and OD plasticity [10] . Finally , high contrast stimulation during deprivation also leads to OD shifts [11] , suggesting that enhancing visually evoked responses of the open eye could be functionally similar to reducing cortical inhibition . Other mechanisms that have been implied with the ending of the critical period are changes in the extracellular matrix [12] and the pruning of silent synapses [13] . Taken together , experimental results hint at a visual-experience dependent maturation of V1 , where normal visual stimuli are necessary to shape the network connectivity . In this article , we propose a model for the first phase after deprivation , coinciding with the response depression phase under MD . We follow the hypothesis that a reduced inhibition is the key to allow for plasticity . More specifically , we model a neuronal network and propose synaptic plasticity principles that are able to reproduce many of the phenomena discussed above . In our model , excitatory-to-inhibitory plasticity is responsible for a rapid reduction in inhibition during the CP , which in turn enables a shift in ocular dominance . Our model is consistent with experimental results observed under MD of the contra- and ipsilateral eyes , under MI and under BD . Furthermore , we discuss possible mechanisms underlying the opening and closing of the critical period , and reinstatement of plasticity . Finally , our model provides a possible explanation to why some neurons shift counter-intuitively towards the closed eye and why these neurons tend to have lower firing rates .
For our simplified model , we simulated 1250 presynaptic excitatory neurons , mimicking layer IV inputs , one postsynaptic inhibitory neuron and one postsynaptic excitatory neuron , mimicking layer II/III neurons . The inputs to the layer IV neurons were modelled as two visual inputs and a background input . The visual inputs represent both eyes and are modelled as step currents A input CL and A input IL , and the background input is an additional step current B . The visual input currents are multiplied by a weight in order to generate different ocular dominances . The weights were generated as follows: x = 0 . 30 + 0 . 35 ζ , ζ randomly drawn from a standard normal distribution w ipsi = { 0 if x < 0 x if 0 < x < 1 1 if x > 1 w contra = 1 - w ipsi ( 2 ) In other words , for each layer IV neuron , a random number was generated from a normal distribution with mean 0 . 30 and standard deviation 0 . 35 . The resulting random numbers were rectified , i . e . negative numbers were set to 0 , and numbers larger than 1 were set to 1 . These numbers were chosen as the weights for the ipsilateral inputs to the layer IV neurons , while one minus the ipsilateral weights were the contralateral weights . In this way , each layer IV neuron received an equal amount of input , but with different ocular dominance . An example of the OD distribution for 1000 layer IV neurons is shown in S6 Fig . From the 1250 layer IV neurons , 250 randomly chosen connections are made to the layer II/III excitatory and inhibitory neurons . In this way , the layer II/III excitatory neurons receives inputs with a range of different ODIs . The same random instantiation is used for MD-CL , MD-IL , BD and MI , resulting in an identical initial ODI for each case . All inhibitory neurons received inputs from 150 most contralateral and 100 most ipsilateral neurons . For simulations of Fig 1 the feedforward E-to-E connections from layer IV to the layer II/III neuron ( wEE ) are plastic under the learning rule given by the following equation d w EE d t = η EE · sgn ( ( ρ pre · ρ post ) - θ H ) ( 3 ) The parameter θH is a threshold separating synaptic depression from synaptic potentiation . For simulations of Fig 2 , an extra threshold is introduced d w EE d t = { η EE ·sgn ( ( ρ pre · ρ post ) - θ H ) if ( ρ pre · ρ post ) > θ L 0 if ( ρ pre · ρ post ) < θ L ( 4 ) This learning rule is shown schematically in Fig 3b . Here , the parameter θH is a threshold separating synaptic depression from synaptic potentiation , while θL is a threshold below which no plasticity occurs . In both Eqs 3 and 4 , ηEE is the learning rate and the function sgn ( x ) denotes the sign function and is equal to 1 if x > 0 , equal to 0 if x = 0 and equal to -1 if x < 0 . The plastic weights are constraint by hard lower and upper bounds , and all other weights ( E-to-I , I-to-E ) are static . We start the simulations with the wEE at the upper bound . The duration of the simulations is 10s , with timesteps of 1ms . We alternated activation of the inputs to the layer IV neurons ( 20ms at a constant value A input CL = A input IL = 10 Hz and a background activation B = 10Hz ) with disactivation ( A input CL = A input IL = B = 0 Hz for 30ms ) . After 500ms , we simulated deprivation . In case of monocular deprivation , we reduced the contralateral input to zero , A input , MD CL = 0 Hz . In the case of monocular inactivation , we reduced the contralateral input and the background input to zero , A input , MI CL = B = 0 Hz , and in the case of binocular deprivation , we reduced the contra- and ipsilateral inputs to zero , A input , BD CL = A input , BD IL = 0 Hz . The values of all parameters used during the simulation can be found in Table 1 . We simulated 1000 layer IV neurons , 100 layer II/III excitatory neurons and 20 layer II/III inhibitory neurons . The layer IV neurons are divided into five groups of 200 , each representing a different input feature ( e . g . a different orientation ) . Similarly , the layer II/III excitatory neurons are divided into five groups of 20 neurons . Each layer II/III neuron received 50 connections from each of the five layer IV groups ( hence 250 feedforward synapses in total ) . However , connections from layer IV neurons of the same group were initialized at 0 . 4 * w EE max , while other feedforward connections were initialized ten times smaller . In this way , an initial orientation preference at eye-opening was mimicked [14] . Moreover , the feedforward connections were not chosen randomly . If we would randomly pick these inputs , the probability to have very ipsilaterally or contralaterally dominated neurons in layer II/III is low , and instead all neurons would have an ODI close to the layer IV population ODI ( S6 Fig ) . Therefore , the connections were set to result in a broader ODI distribution in layer II/III . We outline below how we set 50 feedforward connections from one group of 200 layer IV neurons . The same is valid for the four other input groups representing the other orientations . The 200 layer IV neurons representing one orientation are divided into groups according to ocular dominance , O1 to O5 . We then ensured a broad ODI distribution in layer II/III as follows . We choose x layer II/III neurons and randomly make yi connections from Oi to these neurons , with x and yi given in Table 2 . This ensured both a broad distribution in layer II/III neurons ( Fig 4a ) , as well as enough contralaterally and ipsilaterally dominated inputs to each layer II/III neuron . The latter is important to be able to observe an OD shift in our model ( see Results ) . Recurrent E-to-E connections between the 100 layer II/III neurons were initialized randomly , as observed at eye-opening [14] . To this end , we picked random numbers from a normal distribution with mean and standard deviation equal to 5% of the maximum weight , and ensured values below the minimum weight were reset at this minimum weight . For each of the 20 inhibitory neurons we randomly picked one of the input groups representing an orientation , to which it made 50 strong initial connections equal to the maximum weight , while 50 connections from each of the other input groups were initialized at the minimum weight . Similarly to the simplified model , all inhibitory neurons received inputs from 30 most contralateral and 20 most ipsilateral neurons in each input group . Finally , all feedforward input strengths from layer IV to layer II/III excitatory and inhibitory neurons are multiplied by a factor of 5 to mimic a larger input population . The feedforward and recurrent E-to-E plasticity rule is given by Eq 4 . This learning rule is shown schematically in Fig 3b . The thresholds are chosen such that the open eye inputs after deprivation are above the highest threshold , while the closed eye inputs are under this threshold . Furthermore , we choose the low threshold such that all inputs are below this threshold after monocular inactivation . Finally , we show that our results are robust against the exact choice of thresholds ( S5 Fig ) . The E-to-I plasticity rule is a BCM-type rule [15] , given by d w IE d t = η IE · ϕ pre ρ post · ( ρ post - θ BCM ) ( 5 ) Following the results from [8] , the rule should ensure a quick depression of E-to-I inputs from both eyes and a subsequent recovery of mainly the open eye . Therefore , we choose no dependence on presynaptic rate for depression , ϕpre = 2Hz , while we choose a dependence on presynaptic rate for potentiation , ϕpre = ρpre if ρpre > 3 . 2Hz and zero otherwise . Finally , to ensure the inhibition reaches roughly half of its value after deprivation as observed in [8] , we raise the minimum E-to-I weight to 0 . 4 of the maximum value . This E-to-I rule is such as to reproduce experimentally observed inhibitory activity ( Fig 4c ) , but we do not have enough data to further constrain the rule . θBCM is a sliding threshold given by ( < ρpost > ) 2/ρtarget . The average of the peak postsynaptic firing rate < ρpost > is calculated online by low-pass filtering with a long time constant , τ avg d < ρ post > d t = - < ρ post > + ρ post ( 6 ) The ρtarget is a target firing rate for the inhibitory neurons , and was chosen to be 6 [a . u . ] for normal vision , but reduced to 10% of this value after deprivation . Finally , the I-to-E plasticity rule is based on the rule in Vogels et al . [16] . d w EI d t = { η EI ρ pre ( ρ post - ϕ H ) , if ρ post > ϕ L 0 , if ρ post < ϕ L ( 7 ) By choosing the lower threshold for plasticity ϕL at a high value , close to but below ϕH , this rule limits the maximum firing rates of the excitatory neurons but does not homeostatically increase the firing rate when sudden drops occur ( for example after MD ) . To simulate heterogeneous firing rates , for each layer II/III neuron we generate a random number x = 1 + 0 . 1ζ , with ζ drawn from a standard normal distribution . We then multiply all the rates of the presynaptic neurons of a layer II/III neuron by the respective random number x ( thus either increasing or decreasing all presynaptic rates ) , while also multiplying the ϕH of Eq 7 for all I-to-E synapses to this layer II/III neuron with x ( thus increasing or decreasing the maximal postsynaptic rate by an equal amount as the presynaptic rates ) . The threshold ϕL was always a fixed amount lower than ϕH . The first phase of the simulation lasts 50s . In this phase , the excitatory and inhibitory connections develop and reach either the maximum or minimum bound . Furthermore , MD does not lead to OD shifts because initially , the inhibition is weak and reducing inhibition cannot enhance the excitation sufficiently . After this first phase ensured stationary weights and a sufficiently high sliding threshold , we simulate the deprivation . This second phase lasts 50s . Similarly as in the simplified model , we model MD by reducing the closed-eye input to layer IV to zero , BD by reducing both ipsilateral and contralateral inputs to zero , and MI by setting the contralateral input and the background input to zero . To simulate adult animals , we do not allow any plasticity from E-to-I connections . To simulate pre-CP deprivation , we reduce the first phase of the simulation to 2s instead of 50s , ensuring that the excitatory and inhibitory weights are still low . Furthermore , we simulate the developmental shift in visual inputs strength by ( 1 ) increasing the background input from 10 to 15 , and reducing the visual inputs from 10 to 5 in the pre-CP case , ( 2 ) decreasing the background input from 10 to 5 and increasing the visual inputs from 10 to 15 in the adult case . The ocular dominance index ( ODI ) was calculated as ODI = CL - IL CL + IL ( 8 ) where CL stands for the maximum response to a contralateral visual input , and IL stands for the maximum response to an ipsilateral visual input . In this way , a neuron with ODI = 1 is completely monocular for the contralateral eye , while a neuron with ODI = -1 is completely monocular for the ipsilateral eye . The input selectivity index ( SI ) was calculated as one minus the circular variance . We first calculated the maximal response aj of a neuron to each of the N = 5 inputs , and sorted these responses from large to small ( a1 is the largest and a5 the smallest ) . Then , we calculated r = ( ∑ n = 1 N a j · e i 2 π N n ) / a tot SI = | r | ( 9 ) where a tot = ∑ n = 1 N a j . In this way , a neuron that is active for one input but silent for all other inputs will have an SI equal to 1 , while an input that is equally active for all inputs will have an SI equal to 0 .
The experimental observation that increasing inhibition in pre-CP animals [7] and decreasing inhibition in adult animals [9] allows for OD plasticity , naturally leads to a two-level inhibition hypothesis . More specifically , a first increase in inhibition would open the critical period and a further increase of inhibition would close it . However , the results by Kuhlman et al . [8] allow for a different interpretation . Indeed , the authors showed that even during the critical period , a reduction of inhibition is necessary to observe OD plasticity . The authors therefore proposed that the increased levels of inhibition are crucial in opening the critical period because a subsequent reduction of inhibition can amplify the open-eye excitatory activity . Moreover , stimulating adult animals with high contrast gratings also leads to fast OD plasticity [11] . We can then unify all the results by stating that an OD shift towards the open eye is possible when the open-eye responses are transiently increased . This could be either by reducing inhibition , or by enhancing excitation . Thus , we hypothesise the following: To test this hypothesis in simulated networks , we first consider a simplified model where a single neuron representing a layer II/III pyramidal cell receives feedforward excitatory input from a population of layer IV neurons and feedforward inhibition from one inhibitory neuron ( Fig 1a ) . The feedforward excitatory synapses onto the layer II/III neuron are plastic , while other synapses are static . Assuming ρpre and ρpost are the presynaptic and postsynaptic firing rates respectively , we use the following Hebbian excitatory learning rule ( see Methods ) . If the the product ( ρpre ⋅ ρpost ) caused by an input exceeds a threshold θH , the synaptic weight is increased by an amount η . If this value remains below θH , the value is decreased by η . The parameter θH therefore is a constant threshold separating synaptic depression from synaptic potentiation , and η is the learning rate . In this way , layer IV synapses with various ODIs onto the same layer II/III neuron will lead to different values for ( ρpre ⋅ ρpost ) after deprivation . Indeed , a layer IV neuron that is dominated by the deprived eye , will be left with a small value for ρpre after MD , while a layer IV neuron dominated by the open eye will be relatively unaffected and have a high ρpre ( Fig 1b ) . Moreover , MD will also reduce the postsynaptic firing rate ρpost of the layer II/III neuron by an amount depending on its ocular dominance index . Finally we initialize the feedforward excitatory weights at the upper bound . In this simplified model , we first assume that monocular deprivation pushes all possible ( ρpre ⋅ ρpost ) values into the long-term depression ( LTD ) regime by reducing ρpre and ρpost . Since all synapses are depressed equally , this does not affect the relative response strength between the eyes and therefore leaves the ODI unaltered . Secondly , a reduction of inhibition can rescue the original postsynaptic firing rate and hence shifts only the ( ρpre , open ⋅ ρpost ) above the long-term potentiation ( LTP ) threshold θH . Here , by ρpre , open we mean the presynaptic rates corresponding to open-eye dominated neurons . Only after this reduction of inhibition , the ocular dominance shifts by depressing the closed-eye inputs while maintaining the open-eye inputs ( Fig 1b and 1d ) . To simulate this simplified model , we connect 250 presynaptic neurons to one postsynaptic excitatory neuron and one inhibitory neuron , each modelled as rate units . The presynaptic neurons have a broad range of ODIs ( see Methods ) . In this simplified model , only the layer IV to layer II/III excitatory inputs are plastic and initialized at the upper bound . The layer IV neurons are activated by visual inputs from both eyes and a background input ( see Methods ) . When both eyes are open , all these excitatory inputs are in the LTP regime and therefore remain at the upper bound . We then simulate monocular deprivation of the contralateral eye ( MD-CL ) by reducing only the visual part of the input of the contralateral eye to zero . The closure of the eye therefore reduces the firing rates and all synapses undergo LTD ( Fig 1c , top left ) . We subsequently reduce the feedforward excitatory-to-inhibitory connections to a third of the initial value . This reduction of inhibition leads to a recovery of the original postsynaptic excitatory firing rate , consistent with the data from Kuhlman et al . [8] . Now only the feedforward connections from presynaptic neurons dominated by the closed eye are depressed , while the potentiation of the open eye pathway brings the respective synapses back to the upper bound ( Fig 1c , top left ) . This depression of the closed-eye pathway ultimately leads to an OD shift toward the open eye ( Fig 1d and 1e ) . The same model can also reproduce the lack of OD plasticity after binocular deprivation . In this case , both eyes are sutured and therefore all inputs are reduced to a third of the original values . Since all presynaptic firing rates are attenuated by an equal amount , all ( ρpre ⋅ ρpost ) have the same value . This ensures that the open-eye and closed-eye inputs will always have the same direction of plasticity . In our case , they are all in the LTD region and therefore depressed ( Fig 1c top right , Fig 1d ) . In the case of monocular inactivation of the contralateral eye , TTX injection in the retina abolishes all neuronal activity . This is in contrast with MD-CL , where spontaneous activity is present and some light can travel through the sutured eyelid . Experimentally , no ocular dominance shift is observed after monocular inactivation [4] , suggesting that the residual activity is important . Similar to BD , the total amount of neuronal activity is lower under MI than under MD-CL . However , unlike BD the presynaptic inputs strengths are now variable , depending on the ODI of the respective input . This resembles the situation under MD-CL , but with all input strengths shifted to lower values . With an appropriate choice for the threshold θH , we can therefore still obtain that all synapses are depressed , and hence no OD shift is observed in our postsynaptic neuron ( Fig 1c bottom left , Fig 1d ) . Finally , in the case of monocular deprivation of the ipsilateral eye ( MD-IL ) , we follow the same reasoning as in the MD-CL case . Closed-eye inputs fall below θH and open-eye inputs remain above , leading to a shift towards the contralateral eye ( Fig 1c bottom right , Fig 1d ) . This shift is in agreement with [17 , 18] . Recent work by Rose et al . [19] uncovered a substantial degree of heterogeneity in OD plasticity of individual neurons after monocular deprivation . About 40% of the neurons in layer II/III do not show any particular plasticity , while the amount and direction of the shift in the remaining 60% is variable . Indeed , some neurons even shift their responses counter-intuitively towards the closed eye . The latter neurons were shown to have lower visually-evoked activities and the counter-intuitive shift was caused by a depression of the open-eye inputs . Counter-intuitive shifts were also observed in a study with cats , where a global counter-intuitive shift towards the closed eye occurred after increasing the inhibition during the monocular deprivation [20] . Most neurons in layer IV receive inputs coming from both eyes and project to layer II/III neurons . If these layer IV neurons all have very similar ocular dominance , all synapses will be modified in a similar way in our simplified model and no OD shift will be observed . Indeed , it is highly unlikely that the threshold θH will fall somewhere within this narrow distribution ( Fig 2a ) . For layer II/III neurons to show OD plasticity , the difference between the minimum and maximum ODI of incoming layer IV inputs must therefore be large enough ( S2 Fig ) . Assuming a variety of OD distributions for the inputs would therefore suffice to reproduce both non-plastic neurons and neurons shifting towards the open eye with different magnitude , but not the counter-intuitive shifts towards the closed eye . In order to reproduce these counter-intuitive shifters , we adapt our plasticity rule to contain a second threshold . Besides the threshold separating the LTD region from the LTP region , we introduce a lower threshold which separates a no-plasticity region from the LTD region . We can then understand the counter-intuitive shifters as follows . Neurons receiving low-rate input and/or firing at low rates exhibit smaller values of ( ρpre ⋅ ρpost ) compared to Fig 1 . Subsequently , the closed-eye pathway could fall below the lower threshold for plasticity and would not be altered , while the open-eye pathway would be in the depression regime ( Fig 2b ) . This results in a counter-intuitive shift where the closed eye gains strength relative to the open eye . Finally , with this modified learning rule , we set our thresholds so that the ( ρpre ⋅ ρpost ) fall below the threshold for plasticity immediately after deprivation but before the reduction of inhibition . With this choice , the synapses are unaltered immediately after deprivation , as opposed to all being depressed as in Fig 1b . Since both scenarios would not change the relative strength of the two eyes until the inhibition is reduced , they both agree with experiments . In practice , because of the variety of ODIs in the inputs , some inputs dominated by the open eye may fall in the depression regime ( Fig 2c , top left ) . Since we will assume that excitatory-to-inhibitory plasticity has a faster action then excitatory-to-excitatory plasticity , this short period of open-eye LTD has a negligible and transient effect Fig 2e . The simplified models discussed in the previous sections only included feedforward excitatory-to-excitatory ( E-to-E ) plasticity onto a single layer II/III neuron . We now expand this framework to a population of layer II/III neurons , while adding excitatory and inhibitory plasticity in all connections ( Fig 3a ) . The E-to-E plasticity rule remains the same as before , with a low threshold θL below which no plasticity occurs , and a high threshold θH separating synaptic depression from potentiation ( Fig 3b ) . For the E-to-I plasticity rule , we require that it is not too selective and that synaptic depression is induced after monocular deprivation . The first requirement follows from the experimental observation that inhibitory neurons are broadly tuned for orientations [21] , while the second requirement is necessary for ocular dominance plasticity in our model . We therefore choose to model E-to-I plasticity using a modified version of the BCM-rule [15] with hard upper bounds on the synaptic weights . We choose the target firing rate to be dependent on the visual experience . For normal binocular vision , a high target firing rate enables strengthening of inhibition and a reduced selectivity . For all forms of deprivation , we reduce the target firing rate to 10% of its original value , ensuring a depression of all E-to-I connections . In the cases of monocular deprivation , this depression is followed by a recovery of the open-eye inputs ( Fig 4c ) . Raising animals in complete darkness would therefore never lead to a maturation of inhibition , as observed experimentally [22 , 23] . Finally , the I-to-E plasticity rule is a modified version of the rule proposed in Vogels et al . [16] , ensuring each excitatory neuron does not exceed a maximal firing rate ( see details in the Methods section ) . We assume a variety of ocular dominances in both layer IV and layer II/III neurons ( see Methods and S6 Fig ) , and a variety of excitatory firing rates . Moreover , we divide the layer IV neurons into five groups that are activated separately . These groups mimic the encoding of different input features , for example differently oriented lines within a receptive field . After an initial phase of the simulation where excitatory connections reach either upper or lower bounds , we simulate MD-CL , MD-IL , BD and MI . Similar to the simplified model , we simulate MD by abolishing the visual input from the corresponding eye while maintaining the background input . At the end of the simulation , the mean response of layer II/III neurons shifted towards the open eye ( Figs 4a , 4b and 5c ) . This shift is mediated by a depression of closed-eye inputs , while open-eye inputs remain roughly the same ( Fig 4b and 4c ) . Individual neurons show a variety of OD shifts , and some neurons shift counter-intuitively towards the closed eye ( Fig 4b ) . When plotting the individual shifts versus the firing rate , it is clear that the counter-intuitive shifters are neurons with lower-than-average firing rates ( Fig 5a ) . We then simulate MI by completely abolishing both the contralateral input and background input , and BD by abolishing contra- and ipsilateral inputs but not the background . Since the firing rates of all neurons are now substantially lower , no OD shift is observed ( Fig 5c , S4 Fig ) . Finally , in the cases of MD , we observe that inputs driven exclusively by the closed eye do not show any OD shift ( Fig 5d ) . Indeed , for these neurons the firing rates are more significantly reduced and similar to the case of BD . This dependence of OD shift on initial ODI was experimentally observed in [18] . In order to simulate the maturation of the network , we firstly assume that it starts from an ‘immature’ , pre-critical-period state . We assume that the inhibitory and excitatory recurrent connections are still weak and the E-to-I connections start close to the minimum bound . In agreement with experimental data from Hofer et al . [21] , the choice of our learning rules ensure that E-to-E connections are input-selective while E-to-I connections are unspecific . Moreover , during the development , excitatory neurons increase their input selectivity over time while inhibitory neurons broaden their input tuning ( Fig 6c ) . This can be understood as follows . Both inhibitory and excitatory neurons start with a small bias for one input and therefore have a low selectivity index . However , our selective E-to-E rule ensures that excitatory neurons only develop strong connections with similarly tuned neurons ( Fig 6d , S3 Fig ) , while the unselective E-to-I rule ensures that inhibitory neurons strengthen all incoming connections ( Fig 6e , S3 Fig ) . This evolution is in qualitative agreement with experimental observations of input selectivity in juvenile mice [24] . We can then simulate both the pre-CP and adult networks by assuming that the inhibition is unable to amplify the open-eye activity after deprivation . For the pre-CP case , the inhibitory activity cannot be sufficiently reduced after monocular deprivation because of the immature levels of inhibition in the pre-CP period ( Fig 6a ) . For the adult case , one candidate mechanism could be the consolidation of these E-to-I synapses blocking E-to-I plasticity . When simulating both cases , no OD shift towards the open eye is observed . In fact , since some inputs dominated by the open eye fall in the depression regime , we notice a slight global counter-intuitive shift ( Fig 6b and 6f solid lines ) . This shift is eliminated by assuming that the background-to-visual ratio is higher in the pre-CP and lower in the adult , as observed in [25] ( Fig 6b and 6f dashed lines ) .
In this article , we simulated a simplified model of a layer II/III network in primary visual cortex . Our model is able to reproduce several experimentally observed features of the critical period for ocular dominance . In particular , we simulated changes caused by monocular deprivation , binocular deprivation and monocular inactivation . Furthermore , we discuss possible mechanisms for the onset and the end of the critical period . Our model therefore provides possible mechanistic insights into the development of cortical areas and the associated learning rules , which could be tested experimentally . Our aim was to acount for the effects of short deprivation ( up to 3 days ) on the ocular dominance of layer II/III neurons . We did not include longer deprivations in our model ( more than 3 days ) , when response potentiation is observed . In this case , responses to both eyes—but mainly to the open eye—start to increase . Therefore homeostatic plasticity mechanisms are a likely candidate to explain this second phase of OD plasticity [18 , 26 , 27] . Furthermore , the study observing counter-intuitively shifting cells [19] was performed on adult mice . However , these mice were kept in enriched environments and stimulated with high contrast inputs , both known to enable a juvenile-like plasticity [10 , 11] . We only considered plasticity in connections from layer IV to layer II/III and within layer II/III . Therefore , we did not take into account experimentally observed OD shifts in the thalamic relay neurons [1 , 2] and layer IV neurons [28] . Since these areas are upstream of layer II/III , a naive explanation could be that the shift in layer II/III is fully accounted for by the shift in the inputs to this layer . However , Gordon and Stryker [28] described how a larger OD shift is observed in layer II/III compared to layer IV neurons , and similarly a larger shift in layer V/VI is observed compared to layer II/III . Considering the canonical flow of sensory inputs , from thalamus to layer IV , further to layer II/III and finally to layers V/VI , this result suggests that plastic changes happen at each stage and accumulate over layers . An increased inhibition is necessary in our model to open the critical period . This is because weak inhibition cannot be reduced sufficiently to rescue excitatory firing rates after monocular deprivation . Our hypothesis differs from previous theories on the opening of the critical period , which did not take into account the transient reduction of inhibition observed by Kuhlman et al . [8] . For example , one interesting proposal is that the increased inhibition enhances the visual-to-background activity ratio [25] , while another theory proposed that increasing inhibition favoured more coherent inputs over stronger inputs [29] . It is possible that multiple of these mechanisms play a role in OD plasticity . In our model , the background activity is crucial to model the MI since it allows us to increase the impact of MI on the neuronal firing rates . Moreover , assuming immature and adult levels of visual-to-background activity ratio lead to a better agreement of OD shifts between our model and experiments . Finally , adding more spontaneous activity in our model could counteract maturation if we assume that this spontaneous activity predominantly leads to synaptic depression , keeping the weights low and random . In this case maturation of V1 can only happen when the visual-to-spontaneous ratio is sufficiently high . This ratio could be gradually increased by the developmental changes in NMDA-receptor channels [30] , nogo-receptors and myelination [31] , inhibition [25] and changes in recurrent connectivity [14] . The recurrent excitatory connections in LII/III of our model are not critical for our results . These synapses allow us to reproduce the selectivity of excitatory connections , but the recurrence weak ( see Methods ) . Therefore , with slightly different threshold values , similar results are obtained in a static networks without any E-to-E recurrence ( S6 Fig ) . It would be interesting to study the effect of richer recurrent dynamics [32] . Furthermore , the I-to-E plasticity ensures that excitatory rates do not exceed a neuron-specific activity level . After deprivation , I-to-E connections first strengthen to counteract the depression of E-to-I connections and subsequently weaken again once the the inhibition recovers . In layer IV , a strengthening of I-to-E connections is observed after 2 days of MD [33] , however it is unclear whether these connections weaken again for longer deprivations . The end of the critical period is much less understood . Experiments suggest that the adult levels of inhibition are reached during the CP [24] . Furthermore , Kuhlman et al . [8] showed that in adult mice no reduction of inhibition is observed after one day of MD . This readily leads to the assumption that the E-to-I plasticity , which is crucial in our model to observe OD plasticity , is somehow abolished . We therefore implemented the end of the critical period as a consolidation of the E-to-I plasticity , which could be mediated by changes in the extracellular matrix . Indeed , perineuronal nets ( PNNs ) , have been shown to develop around PV+ inhibitory neurons at the end of the critical period [12] , and could affect the plasticity of synapses onto these PV+ neurons [34] . Furthermore , degrading the PNNs in adult animals restored a window for OD plasticity [12] and this removal is related to reduced inhibition [35] . Another possibility is that the E-to-I plasticity rule itself prevents a reduction of inhibition in adults , for example due to changes in firing rates or correlations . Also the amount of silent synapses has been linked to the ability of juvenile-like plasticity [13] . The end of the critical period is characterized by the pruning of most of these silent synapses and a loss of PSD-95 in the adult leads to an increase in silent synapses as well as a recovery of OD plasticity . However , even though it was shown that the AMPA-to-NMDA ratio of excitatory synapses onto PV+ interneurons was similar in wild-type and PSD95-KO mice , it is not clear how a loss of PSD-95 affects E-to-I plasticity . Our model allows for certain predictions that can be experimentally tested . Firstly , we hypothesize that neurons showing a substantial OD shift after MD need to have a sufficient difference between the lowest and the highest ODI of the input synapses ( Fig 2 and ) . It could be possible that the active synapses have a narrower distribution , but that reducing the inhibition uncovers a broader distribution of silent synapses . In this way , both mechanisms discussed previously—the presence of silent synapses [13] and the reduction of inhibition [8]—could contribute to OD plasticity . Moreover , in animals with columnar organisations of OD we would only expect a broad distribution to neurons on the edges between columns . Our model would then predict that only these neurons show a fast OD shift . Secondly , we introduce a low threshold in our plasticity rule separating no-plasticity from plasticity . Such a low threshold has been observed experimentally [36] and was implemented in a modified version of the BCM-rule [37] . The low threshold allows us to reproduce the effects of MI [4] , the dependence of OD shift on ODI [18] and the counter-intuitively shifting cells under MD [19] , while providing an explanation to why the latter tend to have lower firing rates . Thirdly , we assume that the E-to-I plasticity rule depends on the excitatory population activity . In our case , this was implemented using different target firing rates under normal vision and after deprivation . Such a dependence on population activity has been observed in hippocampal I-to-E plasticity [38] , and it would be interesting to investigate whether and how E-to-I plasticity implements similar mechanisms . Finally , our model predicts that increasing all excitatory firing rates during the monocular deprivation leads to a reduced OD shift . To conclude , in this article we describe a theory of the development of cortical layer II/III . We implemented a simplified network with biologically plausible learning rules , which is able to reproduce multiple experimental results . With our model , we propose that:
|
During the development of the brain , visual cortex has a period of increased plasticity . Closing one eye for multiple days during this period can have a profound and life-long impact on neuronal responses . A well-established hypothesis is that the absolute level of inhibition regulates this period . In light of recent experimental results , we suggest an alternative theory . We propose that , in addition to the level of inhibition , synaptic plasticity onto inhibitory neurons is just as crucial . We propose a model which explains many observed phenomena into one single framework . Unlike theories considering only the level of inhibition , we can account for both the onset as well as the closure of this period . Furthermore , we also provide an explanation for the small fraction of neurons that show counter-intuitive behaviour and provide some testable predictions .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion"
] |
[
"medicine",
"and",
"health",
"sciences",
"neural",
"networks",
"nervous",
"system",
"social",
"sciences",
"electrophysiology",
"neuroscience",
"synaptic",
"plasticity",
"vision",
"eyes",
"neuronal",
"plasticity",
"mood",
"disorders",
"developmental",
"neuroscience",
"computer",
"and",
"information",
"sciences",
"animal",
"cells",
"head",
"mental",
"health",
"and",
"psychiatry",
"cellular",
"neuroscience",
"psychology",
"cell",
"biology",
"anatomy",
"synapses",
"physiology",
"neurons",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"ocular",
"system",
"depression",
"sensory",
"perception",
"neurophysiology"
] |
2019
|
Synaptic plasticity onto inhibitory neurons as a mechanism for ocular dominance plasticity
|
Murray Valley encephalitis virus ( MVEV ) is the most serious of the endemic arboviruses in Australia . It was responsible for six known large outbreaks of encephalitis in south-eastern Australia in the 1900s , with the last comprising 58 cases in 1974 . Since then MVEV clinical cases have been largely confined to the western and central parts of northern Australia . In 2011 , high-level MVEV activity occurred in south-eastern Australia for the first time since 1974 , accompanied by unusually heavy seasonal MVEV activity in northern Australia . This resulted in 17 confirmed cases of MVEV disease across Australia . Record wet season rainfall was recorded in many areas of Australia in the summer and autumn of 2011 . This was associated with significant flooding and increased numbers of the mosquito vector and subsequent MVEV activity . This paper documents the outbreak and adds to our knowledge about disease outcomes , epidemiology of disease and the link between the MVEV activity and environmental factors . Clinical and demographic information from the 17 reported cases was obtained . Cases or family members were interviewed about their activities and location during the incubation period . In contrast to outbreaks prior to 2000 , the majority of cases were non-Aboriginal adults , and almost half ( 40% ) of the cases acquired MVEV outside their area of residence . All but two cases occurred in areas of known MVEV activity . This outbreak continues to reflect a change in the demographic pattern of human cases of encephalitic MVEV over the last 20 years . In northern Australia , this is associated with the increasing numbers of non-Aboriginal workers and tourists living and travelling in endemic and epidemic areas , and also identifies an association with activities that lead to high mosquito exposure . This outbreak demonstrates that there is an ongoing risk of MVEV encephalitis to the heavily populated areas of south-eastern Australia .
Murray Valley encephalitis virus ( MVEV ) causes the most serious of the mosquito-borne virus diseases endemic to Australia . It also occurs on the island of New Guinea , and little is known about its epidemiology there [1] . It is a member of the Japanese encephalitis serogroup of flaviviruses and was responsible for four large outbreaks of encephalitis on the east coast of Australia in the early part to the 20th century ( ranging from 21 to 114 cases ) [2] , and subsequently confirmed epidemics in 1951 ( 45 cases ) and 1974 ( 58 cases ) [1] . Since then the virus has been maintained in enzootic foci in the north of Western Australia ( WA ) and the Top End of the Northern Territory ( NT ) , primarily in a cycle between water birds and Culex annulirostris [1] , with possible contributions from other enzootic foci [3] . With the exception of one case in New South Wales ( NSW ) in 2008 , encephalitis due to MVEV between 1975 and 2010 has been confined to these parts of Australia and adjacent areas further south in WA , in central Australia and in northern Queensland ( Qld ) . Spread of MVEV outside these enzootic foci is thought to be due to rainfall and flooding that allowed movement of infected water birds to previously arid environments . Persistence in desiccation-resistant mosquito eggs may also contribute to outbreaks in previously arid areas , and the existence of cryptic enzootic foci has also been postulated [3]–[5] . The majority of infections with MVEV are asymptomatic or cause a non-specific febrile illness usually accompanied by headache , myalgia and occasionally rash [6] . However , in approximately 1∶150 to 1∶1000 infections with MVEV , clinical encephalitis results [6] . After the incubation period of up to 4 weeks , clinical cases usually present with fever ( commonly accompanied by convulsions in children ) , headache , malaise , and altered mental status , which may be followed by progressive neurological deterioration , parkinsonian tremor , cranial nerve palsies , peripheral neuropathy , coma , flaccid paralysis , and death [6] . The reported case fatality rate of encephalitic MVEV is between 15–30% , with long-term neurological sequelae occurring in 30–50% of survivors and only around 40% recovering completely [6] . Sentinel chicken programs , where flocks of flavivirus-naïve chickens are kept specifically for regular testing for MVE infection , are in place in most parts of Australia where MVEV activity has occurred , with the role of providing an early warning system for MVEV activity [7] . The chickens are bled regularly for evidence of seroconversion to MVEV and other flaviviruses . Some states and territories also have mosquito trapping programs to monitor virus activity in mosquito populations , but difficulty of access and technical limitations currently prevent them being used for real-time monitoring [8] . As there are so few human cases of MVEV disease in Australia , there are limited data about the epidemiology and outcomes , and virtually no information about individual case risk factors . In March to May of 2011 , high level MVEV activity with human infections occurred in SE Australia for the first time since 1974 , accompanied by an unusually heavy seasonal MVEV activity in the NT and northern and central WA [9] . During this period there was also a major national outbreak of encephalitis in horses , predominantly in NSW and Victoria . The majority of cases had laboratory evidence of flavivirus infection ( either Kunjin or MVEV ) . This was the first such outbreak since 1974 [10] . In this report we document the human MVEV cases and add to our knowledge about disease outcomes , epidemiology of the disease , individual risk factors and the link between the heightened MVEV activity and environmental factors .
The investigation and reporting of the MVEV cases was undertaken as part of normal Notifiable Diseases follow up by public health officials in state/territory health departments . As such , ethics committee approval for this study was not required . MVEV disease is a ‘notifiable’ communciable disease in all Australian States and Territories , meaning that clinicians and laboratories are required by law to report cases to local health authorities . Cases of MVEV disease , either encephalitic or non-encephalitic , with a date of onset in 2011 were extracted from state and territory surveillance systems . Each case fulfilled the national case definition for a confirmed case of MVEV infection [11] requiring definitive laboratory evidence and clinical evidence . Clinical information was obtained on each case as part of normal case follow up . Additional clinical information for eight cases hospitalised in WA ( seven infected in WA and one in NSW ) was also obtained from a recent publication [12] . Cases or their family members were interviewed about their activities and their location during the incubation period ( three weeks prior to illness onset ) . Activities that involved significant exposure to mosquitoes ( eg outdoor activities , camping , fishing or others describing exposure to large numbers of mosquitoes ) were classified as high-risk activities . Details for an additional case in a tourist who fell ill after returning home and who was not notified in Australia , were accessed via ProMED mail [13] . Cases were classified as Aboriginal according to the National Health Data Dictionary ( Aboriginal but not Torres Strait Islander origin ) [14] , and all others as non-Aboriginal . Cases who lived in the area in which they acquired MVEV infection were classified as resident . Other cases , who were either travelling in or working temporarily in the area in which they acquired MVEV infection , were classified as non-resident . Details of state and territory sentinel chicken programs for MVEV surveillance and mosquito trapping have been described elsewhere [9] , [15] .
A total of 17 cases of MVEV disease were reported in 2011 , including one case who acquired her infection in Australia but became symptomatic in Canada . Of these , nine cases ( one death ) were infected in WA , four cases ( one death ) in the NT , two cases ( one death ) in SA , and two cases in NSW . There were no cases notified in Qld , Tasmania or Victoria in 2011 . All deaths were a direct result of encephalitis . All but one case had a date of onset between March and May 2011 , and the final case occurred in December 2011 in NSW at the beginning of the next summer . See Table 1 and Figure 1 for further details . Fourteen cases were adults , and three cases were children aged two years and under . Nine of the cases were female and eight were male . Of the adult cases , the median age was 37 , ranging from 19–67 years . Fourteen of the 17 cases were non-Aboriginal people . Ten of the cases were residents of the area where they presumably acquired MVEV , the remainder were non-residents; either tourists or people temporarily employed in the MVEV regions . Three cases died , giving a crude case fatality rate of 18% amongst confirmed cases . Of the 14 survivors , four made a full recovery . Of the remaining ten , eight cases had neurological deficits , ( two of which were mild , five severe and one unspecified ) and two cases reported persisting headaches and/or fatigue . All cases occurred in areas known to have had clinical MVEV cases or sentinel chicken seroconversions to MVEV since 1974 . In 2011 evidence of MVEV activity in sentinel chickens and/or horses was recorded in all areas that had clinical cases . Thirteen cases ( 77% ) reported outdoor activities that posed a high risk of mosquito exposure . At the time of diagnosis , four cases had physical evidence of recent mosquito bites , and a further three cases had been in situations where they had observed high levels of mosquitoes during periods of known MVEV activity . In addition , two other cases resided in areas where extensive flooding had occurred prior to the likely time of infection . Much of Australia experienced a very wet 2010/11 summer and autumn with all States and Territories recording above-average rainfall [16] , [17] . Areas of very-much-above-average rainfall were widespread across Australia; with the exception of the southwest of WA [16] . Much of Victoria , southern NSW , eastern SA , and parts of WA , NT and Qld , had falls that ranked as the highest on record [16] . The wet conditions during summer resulted in major flooding in many areas across the country including Qld , parts of NSW , Victoria and northern Tasmania , and the western part of the Midwest region of WA [16] . This flooding extended to large wetland systems throughout the Murray-Darling Basin and the Lake Eyre Basin , where the highest numbers of waterbirds were recorded since 1984 [18] . NSW had further heavy rainfall in November , 2011 , with associated flooding in a number of areas including in proximity to the location of the December case [19] . High numbers of mosquitoes were trapped in inland NSW , with 102 arbovirus isolates , but no MVEV was isolated [9] . Sentinel chicken surveillance programs were in place in WA , NSW , inland Victoria and the NT during the 2010/11 season . The sizes of the sentinel flocks and locations and periods when seroconversions were recorded in 2011 , are shown in Table 2 . Other details of the programs , including testing methods are described elsewhere [9] , [20] . Activity of MVE during the 2011 season was first detected in sentinel chicken flocks in the NT in December 2010 and in WA , Victoria and NSW in February 2011 ( Table 2 ) [21] , [22] . In March , substantial seroconversions were detected across most of northern and central WA [21] , NSW [22] , and in the Central Australian region of the NT . MVEV activity continued through April , and in May , MVEV was detected in WA sentinel chickens located as far south as latitude 29°S . MVEV activity was first detected at this site in 2000 and had not been found that far south since then [21] . Coinciding with the human case in NSW in December 2011 , MVEV seroconversions occurred over a limited period in early December in the same region [22] . Since 1974 , sentinel chicken seroconversions had occurred in inland NSW in three seasons between 2000 and 2010 , and in SA ( not in a formal sentinel chicken surveillance program ) and Victoria in 2008 [2] . Each State and Territory has its own public health response to MVE activity , with actions taken in response to epidemiological analysis of vector numbers , rainfall , historical risk periods , sentinel chicken results or human cases . Victoria implemented mosquito control programs , pumping of flood waters , health alerts to the health system and testing for MVEV . Television , radio and print notices were utilised advising personal protection and mosquito avoidance in conjunction with Tourism Victoria [23] . Western Australia issued four statewide media statements in 2011 . These followed the detection of MVE in chickens in the Kimberley ( February ) , widespread activity across the state ( March ) , a human case in Carnarvon ( April ) , and new detections of MVEV antibodies in sentinel chickens in the Midwest/Wheatbelt and Goldfield regions ( May ) . The statewide media statements were further publicised at a regional and local level by population health units and local governments , and included pictorial warnings distributed to at-risk small communities . Local government mosquito management programs were also escalated in response to detection of seroconversions in sentinel chickens . Every wet season , Western Australia also routinely issues an alert message on a radio station specifically aimed at tourists , which is upgraded following seroconversion of sentinel chickens . SA issued media releases after the first case was notified , as well as providing detailed clinical information to doctors and diagnostic laboratories . Mosquito monitoring and control activities were enhanced and mosquito avoidance health promotion activities were increased . NSW issued a statewide media statement in February 2011 following the first MVEV seroconversions in sentinel chickens . A NSW mosquito control expert panel was convened in March 2011 following the detection of the first human case of MVEV , additional sentinel flock seroconversions and requests to provide advice on mosquito control in flood-affected areas . Four more statewide media statements were issued in March and April 2011 , together with alerts to clinicians and laboratories . Letters were sent to local councils with advice from the expert panel on mosquito control measures and other risk reduction measures , and with recommendations to promote risk prevention through communications and posters . Regional public health units in affected areas also conducted targeted public communications . NSW issued additional statewide media statements in December 2011 and January 2012 following the detection of additional sentinel flock seroconversions and the detection of the second human case of MVEV . In the Northern Territory , a health warning in the form of a media release was issued at the start of the high risk period for MVE . A heightened media alert was issued and health care providers were advised after the sentinel chicken seroconversions . Each case was also followed up with a media release .
2011 saw a dramatic increase in MVEV activity in endemic regions , and the re-emergence of MVEV in south-eastern Australia [6] . There were 17 cases recorded across WA , SA , NSW and NT , which is the largest number of cases since 1974 and the first large multi-state outbreak since 2000 , when there were fourteen cases across central and northern WA , NT and northern SA [7] . The MVEV activity and the resulting human infections arose from two separate but overlapping sets of environmental conditions . The virus activity and human cases in WA and the NT were a result of heavy seasonal rainfall in the northern and central areas of these two jurisdictions . Activity in southeastern Australia followed the extensive rainfall and flooding in the Murray-Darling basin and adjacent areas in Queensland , NSW , Victoria and SA . The case-fatality rate in this outbreak was 18% , which is similar to the 20% case-fatality rate during the last major outbreak of MVE in 1974 [6] . This likely reflects the lack of advancements in specific treatments for MVE over the past four decades beyond supportive therapy in intensive care , which has been available in Australia for some time . The cases in 2011 followed the usual seasonal patterns for MVEV , with sixteen of the cases occurring in autumn ( March–May ) with no cases during winter or spring and a single case in northern NSW in the following summer during a new period of heavy rainfall flooding and MVEV activity in the sentinel chickens . Of the 17 cases , 14 were non-Aboriginal , 14 were adults and almost half of the cases did not reside in the regions where they acquired MVE . This outbreak continues to reflect a change in the demographic pattern of human cases of encephalitic MVEV over the last 15 years from predominantly Aboriginal to predominantly non-Aboriginal and from paediatric to adult disease . This demographic shift was first noted in the 2000 outbreak that included nine cases in WA and five cases in the NT [7] . Early cases in south eastern Australia in 1951 were mainly children ( 25/40 , 62% ) [24] , while in 1974 children comprised only 32% ( 7/22 ) [25] . No information was available about Aboriginality for those outbreaks . Of the cases that occurred between 1975 and 1999 [26]–[28] 23/35 ( 66% ) were children ( Figure 2 ) and 22 ( 63% ) were Aboriginal , compared with 7/34 ( 21% ) and 7/34 ( 21% ) respectively for cases from 2000 to 2011 [12] , [29]–[31] . A similar change has occurred in the age distribution of fatal cases ( Figure 2 ) . This may reflect , in WA and the NT in particular , increasing numbers of non-Aboriginal workers and tourists living and travelling in endemic and epidemic areas [32]–[34] . Serological surveys in the Kimberley region of Western Australia and in the Northern Territory showed increasing levels of immunity to MVE with age in Aboriginal communities in MVE-endemic areas , meaning that children are more susceptible than adults in these communities [28] , [35] . As more non-immune people move into endemic and epidemic regions , either temporarily or permanently , it is perhaps not unexpected that a higher proportion of MVE cases would be non-Aboriginal adults . In both the 2011 and 2000 MVE outbreaks , evidence of MVEV activity was recorded in areas where it has only rarely been recorded previously [29] , [36] . In WA where the majority of cases were acquired , the areas of acquisition were predominantly south of the Kimberley region , whereas prior to 2000 , cases more usually occurred in the Kimberley [37] . This change is likely to have been a major contributor to the shift in the epidemiology of the disease , as these more southerly areas have larger populations with a lower proportion of Indigenous people , mines and mining townships with predominantly temporary or itinerant populations and large amounts of commercial traffic between these areas and metropolitan Perth . The majority of cases were involved in activities that lead to a high likelihood of exposure to mosquitoes . For example , cases reported fishing at dusk , camping near rivers and creeks and attending outdoor evening sports , which are all likely to increase mosquito exposure . Little is known about the predictors of clinically apparent MVEV infection in humans but while there may be a range of factors , animal studies with MVEV [38] and with other flaviviruses including Japanese encephalitis virus [39] , [40] have demonstrated the importance of viral load . Mosquito saliva potentiates West Nile virus infection in mice , and this is thought to be related to local immunomodulation [41] . This suggests another potential mechanism by which multiple mosquito bites could facilitate clinical MVEV disease and emphasises that any reduction of mosquito biting is potentially beneficial . Currently public health warnings advise all people in areas of MVEV activity to avoid all mosquito exposure . In 2011 these had been issued to residents living in areas of MVE activity before human cases were reported [42] , [43] . However , without a formal evaluation , it is not possible to know whether these activities influenced behaviours or prevented additional MVE infections . Refining the warnings to target very high mosquito exposure activities may increase their effectiveness and credibility . Evaluation of the current public health measures is recommended , as well as consideration of additional social marketing activities should high-risk climatic conditions reoccur . Given the small numbers of cases , a MVEV-specific vaccine is unlikely to become available , but a new human chimeric JE vaccine has shown cross-protection against MVEV in mice [44] . Record wet season rainfall was recorded in many areas of northern Australia , central Australia and south-eastern Australia in the summer and autumn of 2011 [16] , [17] . This was associated with significant flooding and increased numbers of the mosquito vector , Culex annulirostris , and subsequent MVEV activity [6] , including widespread seroconversion of sentinel chickens [21] , [22] , [42] . In WA , the level of MVEV activity in sentinel chickens was one of the highest on record and was similar to that in 2000 , when the last large outbreak of MVEV disease occurred in that state . Sentinel chicken seroconversions in Central Australia are relatively rare and are usually associated with southern extension of the annual northwest monsoon activity [42] , which may blow infected mosquitoes into the Barkly or Central Australian regions , and enable local amplification due to increased vector numbers associated with flooding [42] . Alternatively , the increase in southern rainfall and subsequent flowing of inland rivers might result in the southern migration of water birds which , combined with the local increase in vector numbers , could lead to MVEV activity in these areas . While the majority of sentinel chicken seroconversions occurred in areas of previously described MVEV activity , MVEV seroconversions in the Alice Springs area ( Central Australia ) had not been recorded since 2002 when a major local mosquito breeding site was drained [45] . However , the recent seroconversion to MVEV indicates that the local ecology in Alice Springs can still sustain these arboviruses given suitable climatic conditions [42] . In spite of widespread sentinel chicken activity and disease in horses in Victoria , encompassing an area where an estimated population of 270 , 000 people live , there was only one unconfirmed case of MVE [6] , compared to much larger numbers in very sparsely-populated areas of WA and the NT in the same time-period . This was consistent with previous experience in these areas . A serosurvey conducted in Victoria following the 2011 outbreak found very low levels of MVEV antibody , especially in people born since the last epidemic there in 1974 [46] . This is a similar outcome to a serosurvey of people living in high-risk areas of Victoria and NSW conducted in 1991 [47] . These serosurveys suggest that the differences may be due to relatively less human infection in Victoria , possibly due to more effective mosquito control and/or less frequent high mosquito exposure activities in the urban areas of Victoria compared with large remote areas in northern Australia . The reduction in waterbird numbers in the Murray Darling Basin over time , as a result of increased utilisation of water for irrigation and clearing of wetlands may also have reduced the likelihood of mosquitoes being infected with MVEV [5] . The reduced density of waterbirds might explain the fewer MVEV case numbers in South-eastern Australia in 2011 compared to 1974 , in spite of there being similar climatic conditions . This outbreak shows that extensive MVEV activity continues to occur across Australia if the climatic conditions are suitable . The changing demography of human cases , together with the southward spread of MVEV activity in Western Australia shows that changing patterns of human movement and settlement and changes in ecological factors are continuing to influence the epidemiology of MVEV encephalitis in Australia . Continued MVEV surveillance and evidence-based public health responses are warranted given the serious consequences of MVEV encephalitis .
|
An outbreak of Murray Valley encephalitis with 17 confirmed cases occurred across Australia in 2011 . This outbreak involved parts of Australia where cases had not occurred for many decades . The epidemiology in this outbreak reflects a change that has occurred over the past 15 years , with more non-Aboriginal cases , fewer children and more cases that were not resident where they acquired the infection than had been observed prior to 2000 . The outbreak was associated with significant flooding in many parts of Australia and most cases reported either outdoor activities where mosquito exposure was highly likely or significant mosquito exposure .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"medicine",
"infectious",
"diseases",
"mosquitoes",
"zoonoses",
"disease",
"ecology",
"infectious",
"disease",
"epidemiology",
"epidemiology",
"neglected",
"tropical",
"diseases",
"arboviral",
"infections",
"vectors",
"and",
"hosts",
"public",
"health"
] |
2014
|
The Changing Epidemiology of Murray Valley Encephalitis in Australia: The 2011 Outbreak and a Review of the Literature
|
Comparison of elastic network model predictions with experimental data has provided important insights on the dominant role of the network of inter-residue contacts in defining the global dynamics of proteins . Most of these studies have focused on interpreting the mean-square fluctuations of residues , or deriving the most collective , or softest , modes of motions that are known to be insensitive to structural and energetic details . However , with increasing structural data , we are in a position to perform a more critical assessment of the structure-dynamics relations in proteins , and gain a deeper understanding of the major determinants of not only the mean-square fluctuations and lowest frequency modes , but the covariance or the cross-correlations between residue fluctuations and the shapes of higher modes . A systematic study of a large set of NMR-determined proteins is analyzed using a novel method based on entropy maximization to demonstrate that the next level of refinement in the elastic network model description of proteins ought to take into consideration properties such as contact order ( or sequential separation between contacting residues ) and the secondary structure types of the interacting residues , whereas the types of amino acids do not play a critical role . Most importantly , an optimal description of observed cross-correlations requires the inclusion of destabilizing , as opposed to exclusively stabilizing , interactions , stipulating the functional significance of local frustration in imparting native-like dynamics . This study provides us with a deeper understanding of the structural basis of experimentally observed behavior , and opens the way to the development of more accurate models for exploring protein dynamics .
Associated with each protein fold is a set of intrinsically accessible global motions that arise solely from the 3-dimensional geometry of the fold and involve the entire architecture . For a number of systems it has been shown that these intrinsic motions play an important role in protein function [1] , facilitating events such as recognition and binding [2] , [3] , catalysis [4]–[6] and allosteric regulation [1] , [7] , [8] . The time scales of these cooperative motions are usually beyond the reach of conventional MD simulations . They are modeled instead with coarse-grained techniques that omit the finer details of atomic interactions . The elastic network model ( ENM ) is an example of a coarse-grained model that has enjoyed considerable success in predicting global dynamics of proteins and other macromolecules . The central idea behind the ENM is that , in the vicinity of a minimum , the potential energy landscape of a biomolecular system can be approximated by the sum of pairwise harmonic potentials that stabilize the native contacts . In the simplest ENM , the Gaussian network model ( GNM ) [9] , each node of the network is identified by an amino acid , and each edge is a spring that provides a linear restoring force to deviations from the minimum-energy structure . The system's dynamics is therefore expressed in terms of the normal modes of vibration of the many-bodied system about its equilibrium state; and dynamical information about the protein , such as the expectation values of residue fluctuations or cross-correlations , is uniquely defined by the network topology . A few prevalent methods are used for constructing ENMs , but most have at their hearts two underlying assumptions: The springs are all at their rest lengths in the equilibrium ( native ) conformation , and the force constants decrease with the distance between nodes , among other variables . In the earliest models [9] , [10] and the anisotropic network model ( ANM ) [11]–[13] , force constants were taken to be uniform for all nodes separated by a distance less than a specified cutoff distance and zero for greater distances . In parallel , models were proposed in which the force constants decay exponentially [14] , [15] or as an inverse power of distance [16] , [17] , or where stronger interactions are assigned to sequentially adjacent residues [8] , [16] , [18] . Although such modifications can lead to modest improvements in the agreement between ENM predictions and certain experimental data , there is still no clear “best” method for assigning force constants in an ENM . A common approach for assessing the performance of ENMs or estimating their force constants has been to compare the ENM-derived autocorrelations of residue motions to the corresponding X-ray crystallographic B-factors or the mean-square fluctuations ( MSFs ) in residue coordinates observed between NMR models . Because the slow modes have the largest amplitudes , often the focus of study has been a narrow band of the slowest modes . The ENM slow modes have indeed been shown to agree well with those predicted by detailed atomic-level force fields and with experimentally determined dynamics [19] , [20] . However , the majority of the dynamical information conveyed by the ENM is contained in the residue cross-correlations , and this information has been largely overlooked during comparisons of ENM results to experimental data . Further , the subtle and complex dynamics of the structures that lie beneath the gross global motions are ignored when only the slowest modes are considered . Mid- and high-frequency modes are predicted with relatively lower confidence by ENMs , but these modes may be important for coordinating the finer motions of the molecule while the slower modes orchestrate its global rearrangements [21] . Finally , while the ENM-based studies have shown that the network topology is the dominant factor that defines the collective modes , especially those in the low frequency regime , there may be other structural properties ( e . g . secondary structure , hydrogen bond pattern , distance along the sequence/chain between pairs of interacting residues ) that are not accounted for by ENMs but which may provide a more realistic description of equilibrium dynamics , if accurately modeled . Here we examine the ensembles of structural models determined by NMR for 68 proteins and evaluate for each ensemble the covariance in the deviations of residue-positions from their mean values . We present a technique for optimizing ENM force constants within a pre-defined network topology so as to provide the most accurate representation of the experimentally observed covariance data . Our method is based on the concept of entropy maximization: Briefly , when inferring the form of an unknown probability distribution , the one that is least reliant on the form of missing data is that which maximizes the system's entropy subject to constraints imposed by the available data [22] , [23] . This method has been applied to a variety of biological problems , including neural networks [24] , gene interaction networks [25] , and protein folding [26] . The resulting auto- and cross-correlations in residue fluctuations are used to build an ENM-based model with optimal force constants ( OFCs ) . It can be shown ( see [25] and Methods ) that when the constraints of the maximization are pair correlations , the probability distribution takes a Gaussian form . Further , the only terms that contribute to the probability distribution are those that correspond to pairs with correlations that are explicitly considered as constraints on the entropy maximization . In terms of the ENM , this means that for a given network topology , there exists a unique set of force constants that exactly reproduces the experimentally observed cross- correlations between all pairs of interacting residues , along with their autocorrelations ( or MSFs ) . Notably , our technique captures the physical significance of factors such as sequence separation and spatial distance which have been empirically found to influence force constant strengths . Sequence separation is expressed in terms of contact order , i . e . , the number of residues along the sequence between two residues that are connected by a spring in the ENM . Further , our analysis benchmarked against a test set of 41 NMR ensembles of proteins suggests additional factors , including hydrogen bond formation and secondary structure type , which should also be incorporated in the ENMs for a more accurate description of experimental data . It also identifies factors that are of little consequence insofar as the collective dynamics near equilibrium conditions are concerned . Amino acid specificity turns out to be one of them; diffuse , overlapping distributions of OFCs are obtained for different types of amino acids , precluding the assignment of residue-specific OFCs . A modified version of the GNM , mGNM , that accounts for these factors is proposed and is verified to perform better than existing models especially in reproducing cross-correlations . Finally , the study highlights the importance of higher modes and the role of frustration in protein dynamics , the implications of which are discussed with regard to model development and protein design .
The training set of 68 proteins structurally characterized by NMR and deposited in the Protein Data Bank ( PDB ) [27] ( Table S1 ) contains a total of 252 , 775 possible pairwise interactions ( based on the combination of all pairs of residues ) , of which 43 , 118 ( 17 . 1% ) fall within the 10Å cutoff . Upon optimization , a mean force constant of 6 . 23 kcal/mol/Å2 was found , averaged over all pairs and all proteins . Notably , this value is on the same order as typical uniform ENM force constants [8] , [28] , and provides an estimate of the strength of generic inter-residue interactions in native folds . To eliminate environment-specific effects and allow for the compilation and comparative analysis of the results for all proteins , we normalized the force constants such that the average force constant magnitude in each protein is unity . The resulting normalized OFCs range from −10 . 0 to 31 . 1 , in dimensionless units , with a mean of 0 . 430 and a standard deviation of 1 . 831 . Most ( 71% ) of the force constants have absolute magnitude less than 1 . 0 . Figure 1A displays the distribution of OFCs as a function of the distance dij between the interacting pairs of residues i and j , and colored by contact order k . k designates the sequential separation between residues i and j , k = 1 corresponding to bonded pairs . The inset in Figure 1A displays the dependence of the average magnitude <|γij|> on distance . A closer examination of the influence of contact order on the OFCs yields the histograms displayed in Figure 1B . Whereas most OFCs are generally small and distributed evenly around zero , those associated with bonded interactions tend to be positive and large , with a mean value of 2 . 898 and standard deviation of 3 . 009 ( see Figure 1 , black dots ) . These large positive values reflect the almost rigid 3 . 8Å distance restraints on the backbone pseudo-bonds ( virtual Cα-Cα bonds ) , consistent with the fact that the peptide bond dihedral angle ω is confined to the trans state , and consequently , in the absence of rotatable bonds the distance between the consecutive α-carbons is almost fixed . Second-neighbor ( k = 2 ) interactions tend to be negative , with mean −0 . 211±1 . 436 ( red dots in Figure 1A and red histogram in Figure 1B ) . They obey a unique distance dependence ( Figure 1C , red curve ) , suggesting that 2nd neighbors closer than a certain distance are generally too strained . Likewise , those stretched beyond a certain separation exhibit negative force constants . These interactions add frustration to the system: They tend to favor conformational changes away from the equilibrium structure , but only in a manner that does not violate the more magnanimous k = 1 restraints . Taken together , the k = 1 and k = 2 interactions suggest a flexibility of virtual bond angles , which allows adjacent ( first neighboring ) residues along the sequence to retain almost rigidly their separation while second neighbors tend to move with respect to each other . The k = 3 interactions ( blue dots in Figure 1A ) , on the other hand , are positive ( 0 . 385±1 . 366 ) indicating a dynamic correlation between adjacent virtual bond angles . More detailed analysis shows that in this case there is a weak tendency of 3rd neighbors to be destabilized when their distance approaches 10Å ( Figure 1C , blue curve ) . A similar trend is observed in the case of 2nd neighbors , when they approach their maximal separation ( ∼7 . 4 Å ) allowed by chain connectivity . These observations point to the instability of the conformations that strain the backbone . The k = 2 interaction type and strength depend on the distance between residues i and i+2 ( Figure 1C ) . If the residues are separated by 6Å or less , γij tends to be strong and negative , and the correlation between k = 1 and k = 2 force constants is −0 . 386; for distances of more than 6Å , the correlation with k = 1 drops to −0 . 100 . This suggests the importance of secondary structure in protein dynamics , which will be our focus next . In helices , second neighbors tend to be separated by about 5 . 47±0 . 20Å , compared to 6 . 66±0 . 41Å in strands . As can be seen from the red curve in Figure 1C , the former separation coincides with the minimum ( i . e . , largest negative value ) in the OFC curve , which is also consistent with the red histogram displayed in Figure 2B for α-helices . The positioning of α-carbons i and i+2 along an α-helical turn requires the dihedral angles φ and ψ on both sides of Cαi to assume narrowly distributed values in the Ramachandran space and entails relatively tight packing of side chains , which may not be sufficiently stable per se , unless stabilized by hydrogen bonds formed between the adjoining residues on both sides . No such effect is discerned in 2nd neighboring residues on β-strands , given that the corresponding dihedral angles are more broadly distributed , and the backbone conformation allows for favorable interactions between every other side chain . Notably , 3rd neighbors on β-strands tend to exhibit negative OFCs ( Figure 2C ) . The Cαi-Cαi+3 distance of 8 . 796±1 . 408 Å falls in the regime of negative force constants ( see the blue curve in Figure 1C ) . In the case of helices , third neighbors are located at a distance of 5 . 230±0 . 531 Å , and experience favorable interactions on a local scale ( Figures 1C and 2C ) . The flexibility of the β-strand k = 3 contacts and the rigidity of the β-strand k = 1 and k = 2 contacts suggests that strands have a propensity for twisting motions . Hydrogen bond formation is also found to have a strong influence on the OFCs . Using the DSSP [29] algorithm , we determined secondary structures for residues in our dataset and found that the interactions between hydrogen-bonded residues tend to be larger than those between residues that are not hydrogen-bonded ( see Figure 2D ) , which strongly supports the physical realism of the derived OFCs . In α-helices , the average OFC for k = 4 interaction representative of hydrogen-bonded residues on consecutive turns is 0 . 962±1 . 341 , compared to 0 . 137±1 . 008 for all other k = 4 interactions . Similarly , interactions between hydrogen-bonded partners in extended strands or isolated β-bridges have values around 1 . 801±2 . 321 , compared to 0 . 412±1 . 817 for other interactions , thus more than counterbalancing the destabilizing interactions between 3rd neighbors . In both cases , the distributions for hydrogen-bonded and non-hydrogen-bonded interactions overlap significantly but are distinct , with Kolmogorov-Smirnov [30] probabilities of less than 10−44 . This sensitivity to atomic-level details is missing in many coarse-grained ENMs , but it is an essential component of the potential energy . Clearly , despite the existence of destabilizing interactions on a local scale , the overall structure is stable , i . e . , the native structure is a global energy minimum ( as also confirmed mathematically; see Methods ) because these destabilizing pairwise interactions are more than counterbalanced by other stabilizing interactions . For example , there is a weak ( −0 . 274 ) anti-correlation between the k = 1 and k = 2 force constants , and more significant anti-correlations between k = 2 and k = 3 ( −0 . 689 ) and between k = 4 and k = 5 ( −0 . 614 ) ( See Table 1 ) . In particular , when residues i and i+2 are in helices , the force constants corresponding to the interactions between first and second neighbors exhibit a correlation of −0 . 641 ( see also Figure S1 ) . The third and fourth neighbors on α-helices , on the other hand , are distinguished by their strong stabilizing interactions ( Figure 2C and D ) . Similar effects occur between 2nd and 3rd neighbors in β-strands , and in all cases hydrogen bonds appear to make significant contributions to the overall stability . The presence of these ( anti ) correlations suggests that on a local scale there is a subtle balance between favorable and unfavorable interactions that is instrumental in determining the marginal stability of the molecule as well as its collective motions about the equilibrium structure . We analyzed the dependence of the OFCs on amino acid type and coordination number . The distribution of force constant strengths exhibit some variations by amino acid type as can be seen from the heights and widths of the distributions in Figure S2 , but there is no specific correlation of force constant values with amino acid type . Although each amino acid has a unique distribution of force constant strengths , all of these distributions overlap to a large extent , so that accurately predicting interaction strength based on amino acid type is not possible . This observation agrees with the longstanding argument that the global dynamics of solvated proteins are structure-based , and not sequence-based . We note that the insensitivity of force constants to amino acid type does not imply that all contacts contribute equally to the free energy , but that the deviations from their equilibrium positions experience comparable resistance . In terms of energy function , the depths of the energy minima may dependent on amino acid types , but the curvatures of the energy profiles near the minima do not exhibit residue-specific features at this coarse-grained level of representation . As was seen through the large values of the bonded interactions , physical constraints directly impact the interaction values . We therefore expect the OFCs to be greatest in magnitude for the spatially constrained residues in the protein interior , and the mean-square fluctuations to decrease with the coordination number . Indeed , there is a modest ( 0 . 508 ) correlation between the magnitudes of the bonded interactions and the coordination numbers of the nodes they join . There is a stronger ( −0 . 582 ) ( anti ) correlation between the coordination number and self-interaction , and a very strong ( −0 . 909 ) one between a residue's self-interaction and the sum of its interactions with its first neighbors . The weight of the node , defined as the sum of the magnitudes of its edges , relates inversely to its MSF in much the same way as the degree of a node in GNM relates to its MSF ( Figure S3 ) . Although the force constants vary in value at all distances , we were curious to examine in more detail whether there exists an underlying trend that describes the force constant magnitude as a function of distance between residues . We calculated the average absolute magnitude of the force constants as a function of residue separation ( see Figure 1A , inset ) and examined the functional form of this distance dependence . Using a function of the form as proposed by Hinsen [14] , we find the highest correlation of only 0 . 339 when the distance r0 is 6 . 805Å , which is about twice the proposed value of r0 = 3 . 0Å for non-bonded force constants . Fitting the average magnitude to a function of the form , we find the best fit ( cc = 0 . 356 ) using an exponent of α = 1 . 953 , which is remarkably close to the exponent α = 2 suggested by Jernigan and coworkers [17] . Although the trend is for the average magnitude of force constants to decay with distance between nodes , the correlations are not very strong and the abundance of noise in the force constants prohibits the identification of a definitive function with which they universally decay . Figure 1C shows that the distance dependence also varies with contact order . We compared the collective dynamics calculated with GNM to those found via OFCs ( shortly referred to as OFC-GNM ) , with regard to the level of agreement achieved with experimental data . The computed covariance matrix contains three types of elements: diagonal , interacting ( nodes joined with an edge ) and non-interacting . Diagonal elements are representative of the MSFs of individual residues , and off-diagonal terms represent the cross-correlations between the fluctuations of pairs of residues . Table 2 summarizes the level of agreement of the two methods with the experimentally observed covariances . Notably , the optimized model provides a more accurate description of not only MSFs and cross-correlations between connected nodes , but also the cross-correlations between pairs of residues that are located farther apart in the structure . As shown in Table 2 , experimental covariances between non-interacting residues have a correlation of 0 . 759 with the covariances predicted by OFC-GNM , compared to −0 . 014 for GNM . One attractive feature of GNM is its ability to provide results that are robust against minor changes in structure or network topology . To test the resilience of OFC-GNM dynamics , we set small force constants identically to zero and re-calculated the covariance matrix . When the smallest 5% and 10% of the interactions are discarded , the correlation between OFC-GNM and experiment drops from 0 . 967±0 . 020 to 0 . 407±0 . 443 and 0 . 238±0 . 347 , respectively . Unlike the GNM , the optimized model is therefore quite sensitive to the existence or loss of weak interactions . We also examined the robustness of the modes in the low frequency regime . The values in parentheses in Table 2 shows that the top ranking five modes computed with the OFC-GNM yield good agreement with their experimental counterpart , whether the GNM cross-correlations exhibit a considerable decrease in their level of agreement with experiments . We briefly investigated whether the trends observed in the optimized force constants can be used to create a more effective ENM . Using a separate set of 41 proteins ( Table S2 ) , we tested the effects of incorporating bonded interactions , second neighbor interactions and hydrogen bonding into the ENM . The results , summarized in Table 3 and Table S3 , indicate that including these properties mildly improves the agreement of the ENM with observed covariances for the test set . We obtained the best agreement when bonded interactions and hydrogen bonded interactions are increased in magnitude and second-neighbor force constants are negative . One set of parameters for this model , which we refer to as modified GNM or mGNM , is given in Table 3 .
At present , there are copious NMR and X-ray data available from which we can extract information on protein equilibrium dynamics , and the current state of molecular dynamics is such that one can likewise approximate equilibrium ensembles of small proteins in silico . By developing coarse-grained models that reproduce these dynamics , we are able to deepen our understanding of the factors that influence protein folding and function . In the present analysis we selected to use NMR data that provide conformational ensembles based directly on experiments , but any covariance data could have been used , in principle . The REACH algorithm [31] identifies effective ENM force constants through an inversion of a covariance matrix derived from MD simulations . Similarly , the heteroENM [32] utilizes an iterative algorithm to similarly fit the force constants with MD-derived covariances . The advantages to using MD-derived covariances are precision and flexibility . Because the locations of all atoms in an MD run are known to machine precision in each simulation frame , the covariance between even the most distant atoms , such as those separated by several nanometers , can be exactly calculated within the context of the simulation . Further , MD simulations permit in silico alterations to the system under study , allowing one to find effective force constants that are specific to any environment that can be simulated . This is a boon in particular to those who wish to study the global dynamics and interactions of multiple large molecules . On the other hand , there are some shortcomings of MD that make it an unattractive option for developing an ENM . First , MD is itself a theoretical model , and the performance of any MD-based ENM is limited by the accuracy of the force field: Inaccurate MD results beget inaccurate ENM results . Second , MD is stochastic in nature , insofar as simulations of identical systems starting from different initial states may produce different results due to sampling inaccuracies . Finally , MD is generally applicable only for short ( <1µs ) simulations . Covariances calculated over a short time should not be assumed to remain valid when the timescale is increased by several orders of magnitude . Amino acid covariances are calculated here from experiments , specifically NMR structural data . A few well-studied proteins have been crystallized in multiple states – such as those bound to different ligands – allowing residue covariances to be calculated from X-ray data . Although a growing body of work suggests that functional states assumed by the proteins under different conditions are captured in multiple crystal structures [33]–[36] , such multiple X-ray crystallographic structures have been determined for a few well-studied proteins only , and in most cases proteins crystallized in diverse states may not be representative of the native ensembles of conformations accessible to the protein . A more abundant source of protein conformational ensembles is NMR data . The use of various NMR techniques in determining solution dynamics of proteins has been reviewed extensively ( see , for example , [37] , [38] ) , and a number of techniques have been proposed for inferring native-state protein ensembles from NMR data [39]–[43] . Covariances calculated from NMR ensembles have been shown to agree well with MD [44] , X-ray B-factors [45] , [46] and covariances between multiple crystal structures [33]–[36] . NMR data are not , however , without their shortcomings: NMR ensembles may be affected by the sparsity of data and conformational variations found in solution , and as such they necessarily contain noise and do not purely reflect the native state ensemble . As the NOE intensities that are used to define structures decay rapidly with interatomic distance , long-ranged interactions are a likely source of noise in NMR covariance data . Force constant optimization methods that rely on full covariance data [31] , [32] retain this noise . We were able to identify the major determinants of the effective force constants that describe the collective dynamics of proteins by resorting to a rigorous entropy maximization procedure that addresses such uncertainties . Strikingly , a subtle interplay between stabilizing and destabilizing interactions has been disclosed , which depends on contact order , secondary structure and hydrogen-bond-formation properties . Although all of the proteins that we have analyzed are relatively small , the physical basis of the factors impacting force constant strength leads us to believe that our results hold for larger proteins as well . The OFCs are derived from existing structural data , and in this respect our work is similar in spirit to the extraction of knowledge-based potentials from known structures [47]–[53] . The present study differs , however , in four ways: First , previous studies aimed at evaluating the effective potentials of mean force that determine the equilibrium state/energetics of native structures , and they were used in evaluating folded or docked conformations . Here , the goal is to assess the effective force constants that determine the collective fluctuations away from the equilibrium state , which are used in evaluating the equilibrium dynamics . Second , the training dataset consists of distinct proteins' structures in the former approach , whereas here ensembles of conformations corresponding to a given protein are analyzed . Third , the former group of studies counts the probabilistic occurrences of inter-residues pairs ( or pair radial distribution functions ) to derive potentials of mean force using inverse Boltzmann law; here , the departures in coordinates from their mean values are examined , and optimal spring constants are evaluated from an entropy maximization scheme , which is appropriate for sparse data . Fourth , the knowledge-based potentials evaluated in previous studies are residue-specific , whereas the OFCs show no significant dependence on amino acid type . This final observation is in accord with the concept that amino acids influence the fold , and the fold influences the dynamics . In our calculations we intentionally used a slightly longer cutoff distance ( 10Å ) than those determined to optimally reproduce B-factors ( 7–8Å ) [19] , [54] . Our reasoning was that , if a shorter cutoff distance is better , then force constants for residues that are far from each other will tend to be close to zero . Although we find that the average magnitude of the force constants decays with distance , we do not find that the force constants all drop sharply to zero after some distance . GNM consistently predicts global protein motions that agree with experimental observations , using a uniform force constant . It would therefore not have been unexpected to find that the OFCs tend to cluster about a single non-zero value . Instead , we find that the OFCs adopt a range of values centered about zero , and that the strongest indicators of force constant strengths are contact order and backbone hydrogen bond formation propensities . The difference between the predictions of the GNM and observed protein motions is illustrated in the three examples of Figure 3 , selected from the test set ( Table S2 ) . The three curves therein represent the MSFs of residues based on five slowest modes derived from NMR data ( black , solid ) , predicted by the GNM ( red , dashed ) , and predicted by the mGNM ( blue curve ) . As the GNM is based entirely on the protein's folded topology , it tends to instill the most motion in the least connected nodes , e . g . , chain termini or the most exposed loop regions . However , the size of the motion may depart from those indicated by NMR models , and mGNM tends to yield a better agreement with NMR data . Application to the complete test set of NMR ensembles confirmed that the correlation with experiments is improved even when contact order , distance dependence and hydrogen bonding are incorporated into the GNM without laboriously optimizing the force constants ( Table 3 ) . The fact that these physically meaningful effects emerged independently from our entropy maximization calculations validates our approach to some extent . Less expected was the prominence of negative force constants . Overwhelmingly , the methods of ENM construction rely on two assumptions that guarantee physically plausible behavior , but which may be unwarranted . The first is that all springs are at their rest lengths in the equilibrium conformation , and the second is that all spring constants are positive . Taken together , these assumptions are sufficient , but not necessary , to guarantee that any deformations will increase the system's energy . Our optimization procedure naturally produces interactions that are physically equivalent to springs of negative force constant , but so long as the interaction matrix remains nonnegative definite , the system is in a stable equilibrium and negative force constants are acceptable . The existence of negative force constants reflects the implicit frustration of folded proteins; the backbone restrains the protein to certain compact folds , and not all native state contacts are guaranteed , nor should be expected , to be favorable . Negative force constants make the structure prone to certain deformations that may not be preferred when all force constants are positive . Frustration in proteins results in a rough free-energy landscape that gives rise to folding intermediates and alternative conformations [55]–[58] , and calculations involving Go-like potentials , or knowledge-based potentials [49] reveal the requirement to include both stabilizing and destabilizing interactions for an accurate assessment of the folding behavior or stability of proteins . The balance between attraction and repulsion endows proteins with both the sensitivity and the stability that are prerequisite for proper function [59] . We find that the ( i , i+2 ) interactions are the most likely to be at a local maximum , promoting a change in the angle between ( i , i+1 ) and ( i+1 i+2 ) pseudobonds . When we include factors such as hydrogen bonds and negative k = 2 force constants in the GNM , the improved agreement comes in the off-diagonal components of the predicted covariance matrices . Cross-correlations are often overlooked when assessing ENM predictions , but they are essential because they carry information on how the molecule moves as a whole . The autocorrelations that indicate how much individual residues move are each the sum of positive terms and are necessarily dominated by the slower modes . The cross-correlations , on the other hand , are sums of positive and negative terms and are therefore susceptible to the influence of higher modes . Slight modifications to the GNM , such as those that we have introduced in mGNM , do not perturb the network enough to significantly alter the slow modes ( Figure 3 ) , but their effects are captured in the higher modes . Although the slowest modes get the most attention because of their prevailing role in determining the molecule's global motions , the high-frequency modes have shown to be important for identification of conserved residues and folding cores [60]–[63] . Mid- to high-frequency modes are also crucial to all aspects of protein behavior . Allosteric transitions have been shown to occur largely along the slowest modes , but higher modes are essential for the complete transition [64] . Similarly , a protein's response to external perturbations [28] is dependent on all modes , not only the slowest few . An ENM that accurately captures all modes has an enhanced ability to predict large-scale conformational changes , and our technique opens the door to developing better ENMs based on experimental data . Figure 4 shows pairwise comparisons of the eigenspaces spanned by the slowest modes of various models . Panel A shows the correlation of mobilities as a function of the fraction of modes used in the comparison , and panel B shows a similar plot of the overlap of the eigenspaces ( see Methods ) . The green and black curves relate the GNM and mGNM , respectively , to the experimental covariance matrices . The average mobility correlation of GNM with the experimental covariances peaks at 0 . 76 when 12% of the modes are considered and then falls as more modes are taken into account , indicating that the predicted modes in the mid-to-high frequency range introduce errors manifested by departures from experimental data . The modified GNM does not exhibit this decline , but remains steady even as higher modes are considered , indicating that the higher modes of the mGNM do not adversely affect the predicted mobility of the system . Comparison of GNM to mGNM ( blue curves ) shows that the slowest 2% of modes of these models are highly overlapping , but that the similarity decreases as more modes are considered . The modifications of mGNM therefore do not affect the slowest mode , which is presumably determined by the fold topology , but they change the shapes of higher modes . Interestingly , the overlaps of the GNM and the mGNM with the modes of the covariance matrix are almost identical ( compare green and black curves , panel B ) , suggesting that , despite the improved agreement in mobility , the modifications that we have made to the mGNM still fail to precisely capture the system's overall dynamics . Although some additional improvement may be gained by fine-tuning the parameters of the mGNM ( last line , Table 2 ) , the similarity in slow modes of GNM and mGNM once again indicates that fold topology has the dominant influence on the mode shapes .
For our training set , we start with a set of 68 proteins ( Table S1 ) , each of which has at least 40 NMR structures available . The proteins in our set have between 43 and 151 residues . For each protein we calculate the mean structure from the NMR ensemble , and we select as a representative structure the NMR model that has lowest root-mean-square deviation ( RMSD ) from the mean . The test set consists of 41 proteins ( Table S3 ) , each having at least 40 NMR models and no fewer than 50 residues . We seek to determine the pairwise interactions that optimally describe observed covariances between residues while minimizing the assumptions about the form of missing data . For this , we turn to the principle of maximum entropy , which states that when inferring the form of an unknown probability distribution from a limited number of samples drawn from the distribution , the method that is minimally reliant on the form of missing data is entropy maximization . Here the central idea is outlined in terms of the GNM . Consider a protein of N residues for which m structures are known ( e . g . , m models deposited in the PDB for a given protein resolved by NMR spectroscopy ) . The position of residue i in structure k is given by the vector , , the average position of residue i in all structures that have been optimally superimposed ( to eliminate external degrees of freedom ) is defined as , and the vector displacement of residue i in structure k from the average is . In the GNM , we replace the vector displacement ΔRi with the scalar displacement Δri , which is defined such that and . Now define the set π of q pairs of residues such that for all pairs we know the covariances , but for pairs we do not know . We seek the probability distribution that produces the known covariances while remaining minimally presumptive about the form of missing information . According to Jaynes [22] , [23] , this is the distribution that maximizes entropy subject to the constraints that some pair covariances are known and must be reproduced . Defining the N-component vector , , the probability distribution that we seek is ρ ( Δr ) , and it has the properties ( 1 ) ( 2 ) We define the entropy , and impose the above constraints as Lagrange multipliers: ( 3 ) Maximizing ζ with respect to ρ ( Δr ) , we find ( 4 ) or , defining Z = e1+λ . and the matrix K with elements Kij = μij , ( 5 ) Direct integration leads to the result ( 6 ) which is the well-known relationship between covariances and pair interactions . The probability distribution in Equation 5 is of the same Gaussian form as the probability distribution from GNM [9] , but with the interaction matrix K replacing the product of the spring constant γ and the Kirchhoff matrix Γ . Thus , the off-diagonal elements of K correspond to the negative spring constants: Kij = −γij , where γij is the force constant of the interaction between residues i and j . We are claiming knowledge for the covariance information of only the q residue pairs in the set π , so K cannot be found through the simple inversion of the covariance matrix . The matrix K has a well-defined form: the elements are the Lagrange multipliers that have imposed the above constraints on the covariance and may therefore be different from zero; the elements are identically zero . Mathematically , this means that there are no constraints on the covariances of pairs . We then have partial information for both K and K−1: The elements and are known , and the elements and are to be determined . The solution can be found through an N-dimensional minimization as follows . Consider the function ( 7 ) of two symmetric square matrices K and C . Differentiation with respect to each element of K reveals that there exists a single minimum at ( 8 ) Because Cij is undefined for all , we can allow , automatically satisfying the minimization condition for elements not in π . The remaining elements of K can be found by starting with a matrix of the general form of K and iteratively adjusting the non-zero elements against the gradient given in Eq . 8 until the minimum is reached . Optimization is achieved when for all interactions . This criterion appears to be sufficiently strict: Reducing the optimization constant from 0 . 01 to 0 . 005 changes the spring constants by less than 1% , on average . The optimization is somewhat computationally intensive: Each step requires an O ( N3 ) matrix inversion , and the minimization completes after about 104 steps , making this technique best-suited for small proteins . It is noteworthy that only those interactions corresponding to known covariances are optimized , and the rest remains zero . This result stems from the application of entropy maximization . Whereas many networks are capable of exactly accounting for the covariance information in the q known interactions , this is the only one that does so without prior assumptions about other covariances . Each pair interaction carries information on the covariance of two of the N nodes , so a network of more than q interactions carries information on more than q covariances . Nevertheless , all covariances can be calculated with the resultant network . Those covariances that are not known a priori and included in the calculation simply result from the optimized interactions . The matrix C is nonnegative definite by construction , and its inverse K is therefore also nonnegative definite . As a result , no deviation from the native state conformation can lower the system's energy . The interaction matrix K has the dimensions of Å−2 , and physical values for the force constants can be determined by multiplying by 3kBT , where kB is the Boltzmann constant and T is the temperature . Using this conversion , the OFCs vary between −1686 kcal/mol/Å2 and 3868 kcal/mol/Å2 , with a mean of 6 . 23 kcal/mol/Å2 . When K is scaled by a scalar constant , γ , its corresponding covariance matrix is scaled by γ−1 . Thus , the mean element magnitude of the covariance matrix affects the magnitudes of the elements of the interaction matrix , such that large covariances tend to produce weak interactions . The experimental conditions under which the structures are solved influence the magnitudes of the covariances , and therefore also influence the magnitudes of the effective force constants . To reduce the bias on force constants caused by environmental specificity , the OFCs for each protein are scaled by the mean magnitude of the non-zero off-diagonal interactions in that protein . In the GNM , each residue is a node of the network and is represented by its Cα atom . Nodes that are within a cutoff distance , Rc , are considered connected via an elastic edge . Typical values of Rc are between 7Å and 10Å . Using the N-dimensional column vector , , of displacements of the nodes from their equilibrium positions , the potential energy is found to be , where γ is a uniform force constant assigned to all interactions , and Γ is the Kirchhoff adjacency matrix , with off-diagonal elements Γij = −1 if nodes i and j are in contact and Γij = 0 otherwise . The diagonal elements of Γ are such that the sum over all elements in any row or column is identically zero . The elements of the covariance matrix predicted by the GNM are related to Γ as . If U and V are two sets of normal modes for an N-dimensional system under different models , then we define the overlap of the first m modes of the models as , where u ( k ) and v ( p ) are the kth and pth slowest modes of U and V , respectively . Qm ranges from 0 , if none of the space spanned by the slowest m modes of U can be projected onto the first m modes of V , to 1 , if the two spaces overlap exactly . The force constant between residues i and i+k is . The correlation coefficient between force constants corresponding to different contact orders is calculated as follows . First , for a contact order n<k , we define as the average force constant for all pairs between i and i+k that have a contact order of n: ( 9 ) The correlation between force constants and is then ( 10 ) Table S2 lists such correlations for contact orders in the range 1≤k≤5 .
|
As more protein structures are solved , we are able to perform a more critical assessment of the relationship between protein structure and dynamics , and to gain a deeper understanding of the major determinants of structural dynamics . Here we perform a systematic study on a set of proteins structurally determined by NMR spectroscopy . The dynamics are analyzed using elastic network models and a novel method based on entropy maximization to demonstrate that properties such as contact order and secondary structure do play a role in defining the experimentally observed covariance data . Most importantly , an optimal description of observed cross-correlations requires the inclusion of destabilizing , as well as stabilizing , interactions , stipulating the functional significance of local frustration in imparting native-like dynamics .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"computational",
"biology/macromolecular",
"structure",
"analysis",
"biophysics/theory",
"and",
"simulation",
"computational",
"biology"
] |
2010
|
Using Entropy Maximization to Understand the Determinants of Structural Dynamics beyond Native Contact Topology
|
Modulation is essential for adjusting neurons to prevailing conditions and differing demands . Yet understanding how modulators adjust neuronal properties to alter information processing remains unclear , as is the impact of neuromodulation on energy consumption . Here we combine two computational models , one Hodgkin-Huxley type and the other analytic , to investigate the effects of neuromodulation upon Drosophila melanogaster photoreceptors . Voltage-dependent K+ conductances in these photoreceptors: ( i ) activate upon depolarisation to reduce membrane resistance and adjust bandwidth to functional requirements; ( ii ) produce negative feedback to increase bandwidth in an energy efficient way; ( iii ) produce shunt-peaking thereby increasing the membrane gain bandwidth product; and ( iv ) inactivate to amplify low frequencies . Through their effects on the voltage-dependent K+ conductances , three modulators , serotonin , calmodulin and PIP2 , trade-off contrast gain against membrane bandwidth . Serotonin shifts the photoreceptor performance towards higher contrast gains and lower membrane bandwidths , whereas PIP2 and calmodulin shift performance towards lower contrast gains and higher membrane bandwidths . These neuromodulators have little effect upon the overall energy consumed by photoreceptors , instead they redistribute the energy invested in gain versus bandwidth . This demonstrates how modulators can shift neuronal information processing within the limitations of biophysics and energy consumption .
The activity of neurons and neural circuits is modulated to adjust their properties to changes in state , affecting behaviour . Modulation can occur at large spatial scales , encompassing large numbers of neurons , or may be more localised , being restricted to small circuits or single neurons [1–5] . It can be brief or prolonged , and may be intrinsic or extrinsic to the circuits being modulated . A variety of substances from small gases ( e . g . NO ) to biogenic amines ( e . g . serotonin , dopamine ) , peptides ( e . g . FMRFamide ) and other small molecules ( e . g . ATP , PIP2 ) can act as modulators [2] , altering the biophysical properties of neurons and , by doing so , affecting their electrical signalling . Many studies have focussed on the specific effects of modulators on voltage-dependent conductances , quantifying their effects on signalling and , in some cases , on behaviour [2] . For example , by acting on voltage-dependent conductances modulators can switch interneurons and motor neurons from following synaptic inputs directly to generating intrinsic bursting rhythms [6] . Studies of sensory neurons have also shown that modulators can alter their gain and frequency response ( bandwidth ) , as well as increasing their coding precision [7] . But despite extensive characterisation of the effects of neuromodulators upon coding and behavioural consequences , their impact upon the energy consumption has been largely ignored . Numerous lines of evidence suggest that energy consumption is an important factor both in the function and evolution of neurons , neural circuits and the nervous system [8–10] . Energy consumption can be considerable; the human brain is estimated to consume 20% of the basal metabolic rate , whilst the retina of the blowfly is estimated to consume 8% resting metabolic rate [8] . One of the main processes that consumes energy is the movement of ions across the bilipid membrane to support electrical signalling by graded and action potentials [11–14] The properties of the protein channels through which ions move determine the pre- and postsynaptic ion flux across the bilipid membrane . For example , changes in the properties of the voltage-dependent ion channels that generate action potentials can alter their energy consumption by orders of magnitude [11 , 15] . Ion flux is related to energy consumption primarily through the work done by the Na+/K+ pump , which is powered by the hydrolysis of ATP molecules [16 , 17] . Many neuromodulators alter ion flux by adjusting the properties of protein channels through which ions move . This provides a route by which neuromodulators can affect neuronal energy consumption and the energy efficiency of electrical signalling . Here we assess this possibility in the R1-6 photoreceptors of the fruit fly , Drosophila melanogaster , using an analytic model of the photoreceptor to understand and interpret the biophysics accompanied by a computational Hodgkin-Huxley-type model to assess energy costs . The models incorporate the voltage-dependent K+ conductances expressed by these photoreceptors: a rapidly inactivating conductance encoded by the Shaker gene; a more slowly inactivating conductance encoded by the Shab gene; and , a non-inactivating conductance of unknown provenance ( the ‘novel’ conductance ) [18 , 19] . Fruit fly photoreceptors , like those of other insects , encode information contained in light as a graded depolarisation . Consequently , voltage-dependent K+ conductances can change the filter properties of the membrane , altering the gain and bandwidth of electrical signals and have recently been shown to improve energy efficiency [14 , 19–23] . The Shaker and Shab voltage-dependent K+ conductances of fruit fly photoreceptors are modulated by serotonin and PIP2 [24–26] . This provides an opportunity to characterise the changes in electrical signalling wrought by modulators in fruit fly photoreceptors , and determine whether such changes produce substantial effects upon energy consumption and efficiency .
We used a Hodgkin-Huxley ( HH ) formulation [27] to model the dynamics of the three voltage-dependent K+ conductances present in the fruit fly , Drosophila melanogaster , R1-6 photoreceptors . Activation and inactivation rates were obtained from previous studies [19 , 22 , 25] . In the following formulae describing conductance properties , steady-state activation and inactivation gating variables have no units , the voltage , V , is given in mV and the time constants are given in ms . Our model of the fruit fly R1-6 photoreceptor membrane has capacitance 50 pF , intermediate among modern mesurements of 30 pF to 65 pF [19 , 25] . It contains the three types of K+ conductances described above , namely Shaker , Shab and the Novel conductance , plus the light-induced curent ( LiC ) . A Na+/K+ ATPase [29 , 30] and a Na+/Ca2+ exchanger [31–35] maintain homeostasis by reversing the flux of ions . The K+ reversal potential , EK , is -85 mV , and the reversal potential of LiC is EL = 5 mV . To reproduce a dark resting potential of -68 mV and input resistance in the dark of just below 300MΩ ( at the lower end of the in vivo electrophysiological experimental values , reported to be as low as a couple of hundred MΩ [36] , 410MΩ [22] or 1240MΩ [37] ) we included a 0 . 803 nS leak conductance with the reversal potential of the LiC and a 2 . 1 nS K+ leak conductance . From the dark resting potential , the photoreceptor model is depolarised by increasing the LiC . The recordings of the maximum steady-state depolarisation range from -45 mV [22] to almost -30 mV [37 , 38] , and we chose the intermediate value of -36 mV . We calculated the steady-state values of the conductances at different membrane potentials using well-known procedures . The total K+ conductance is the sum of the voltage-dependent K+ conductances and K+ leak conductance g K = g K , Shaker + g K , Shab + g K , Novel + g K , leak ( 16 ) and the total depolarising conductance is the sum of the leak conductance and the light-induced conductance g L = g light + g leak ( 17 ) We assume that the LiC is mediated by Na+ ( 74% ) and Ca2+ ( 26% ) , with a constant current fraction across voltages [34] , and that only K+ flows through the K+ conductances . At each steady-state depolarized membrane potential , V , we assume that all ionic currents , K+ , Na+ and Ca2+ , are restored by the Na+/K+ ATPase and the Na+/Ca2+ exchanger . In each cycle , the Na+/K+ ATPase pumps two K+ ions in and three Na+ ions out at the cost of one ATP molecule . The Na+/Ca2+ exchanger interchanges one Ca2+ by three Na+ ions . The net charge transported is 1 e per cycle in both the Na+/Ca2+ exchanger and the Na+/K+ ATPase . This produces three equations , each equation being the result of zeroing one of the net ionic currents , I ( K ) , I ( Na ) and I ( Ca ) I ( K ) = g K ( V ) ( E K − V ) + 2 I P = 0 ( 18a ) I ( Na ) = 0 . 74g L ( E L − V ) − 3 I P − 3 I E = 0 ( 18b ) I ( Ca ) = 0 . 26 g L ( E L − V ) + 2 I E = 0 ( 18c ) where IP and IE are the net currents produced by the Na+/K+ ATPase and the Na+/Ca2+ exchanger respectively . As we know the steady-state properties of the conductances , for each membrane voltage , V , we know gK ( V ) and we can solve the system of linear equations to obtain glight , IP and IE . The response of a membrane containing voltage-dependent conductances to injected currents is , in the limit of small deflections , equivalent to the response of an electrical circuit composed of resistances , inductors and capacitors [27 , 39 , 40] , sometimes called the equivalent RrLC circuit [14 , 23 , 41] . The equivalent circuit of an inactivating conductance , g = g ¯ m γ h , has two phenomenological branches shunting ( in parallel to ) the membrane capacitance , C . The values of the electrical elements of the circuit , the resistances R , r , rh and the impedances L and Lh are given by R = 1 g ¯ m γ h ( 19a ) r = α + β γ g ¯ m γ − 1 h ( V − E K ) [ α ′ − m ( α ′ + β ′ ) ] = 1 ( V − E K ) g ¯ h d d V m γ ( 19b ) L = r α + β = τ r ( 19c ) r h = α h + β h g ¯ m γ ( V − E K ) [ α h ′ − h ( α h ′ + β h ′ ) ] = 1 ( V − E K ) g ¯ m γ d d V h ( 19d ) L h = r h α h + β h = τ h r h ( 19e ) where α = α ( V ) , β = β ( V ) , τ = τ ( V ) are respectively the activation rate , deactivation rate and time constant of the HH activation variable , m , at the voltage , V . Similarly , αh = αh ( V ) , βh = βh ( V ) , τh = τh ( V ) are respectively the activation rate , deactivation rate and time constant of HH inactivation variable , h , at the voltage , V . Primes ( ′ ) represent derivatives of the functions above with respect to voltage . We then use the values of R , r , L , rh and Lh to compute the impedance of the K+ conductance Z c − 1 ( f ) = 1 R + 1 r + i 2 π f L + 1 r h + i 2 π f L h = 1 R + 1 / r 1 + i 2 π f τ + 1 / r h 1 + i 2 π f τ h ( 20 ) where f is the temporal frequency ( in Hz ) and i is the imaginary unit . Please note that R , r and L are positive quantities , but rh and Lh are negative quantities . rh and Lh can be substituted by a positive resistance r h * and positive capacitance C h * as described elsewhere [42] . As we only use the equivalent circuit to obtain the impedance , we do not perform this substitution . If a conductance has different modes of inactivation or partial inactivation , the conductance can be divided into several conductances either non-inactivating or with only one mode of inactivation , and Eq 20 can be applied to each different component . Once all the impedances of the different conductances , ZShaker ( f ) , ZShab ( f ) and ZNovel ( f ) , have been calculated using Eq 20 , the impedance of the whole photoreceptor is Z ( f ) = 1 Z Shaker − 1 ( f ) + Z Shab − 1 ( f ) + Z Novel − 1 ( f ) + g K , leak + g L , leak + i 2 π f C ( 21 ) where gK , leak and gL , leak are the two leak conductances and i2πfC is the admittance of the membrane capacitance , C . Note that for small deviations around the steady state , we consider IP and IE as fixed currents , which do not affect the photoreceptor impedance . Each photon captured by the photoreceptor produces a transient increase in light-induced conductance and thus a transient increase in LiC ( quantum bump ) . Thus , under physiological conditions , it is useful to describe how the photoreceptor voltage changes with changes in light level . A common measure is the contrast gain function ( e . g . [37 , 43] ) , which is the frequency dependent gain between light contrast and the voltage response of a photoreceptor . In this section we show that , under simplifying assumptions , the contrast gain is the product of LiC and impedance . To account for the effect of voltage-dependent channels upon phototransduction gain , we implemented a simple transduction model . In this simple model , glight follows light contrast , c ( t ) , without any filtering or adaptation in the range of frequencies considered g light = g light ¯ ( 1 + c ( t ) ) ( 22 ) where g light ¯ is the steady-state light induced conductance . When the perturbations around the steady state , g light ¯ , are small , the resulting voltage fluctuations are much smaller than EL − V , and the effect of the change in light contrast is equivalent to an injected current of zero mean I contrast = g light ¯ c ( t ) ( E L − V ) ( 23 ) In the frequency domain , this current injected in a photoreceptor with membrane impedance Z ( f ) produces the following voltage fluctuations V ( f ) = I contrast ( f ) Z ( f ) = g light ¯ c ( f ) ( E L − V ) Z ( f ) ( 24 ) where , abusing the notation , we used the same symbol ( Icontrast ( f ) and c ( f ) ) as in the time domain for the Fourier transforms . From this equation , we deduce that the contrast gain is T ( f ) = V ( f ) / c ( f ) = g light ¯ ( E L − V ) Z ( f ) ( 25 ) In a real photoreceptor , at low frequencies , contrast gain is reduced by adaptation . At high frequencies , light response is limited by the width and latency dispersion of the quantum bump [44 , 45] . As a result , Eq 25 is a theoretical maximum , and will be a valid approximation only at intermediate frequencies . The Na+/K+ pump uses energy from ATP hydrolysis to maintain ionic concentration gradients [17 , 29] . This is the dominant cost of the photoreceptor [14 , 46] . We obtain the net pump current solving Eq 18 . Because the Na+/K+ ATPase hydrolyses an ATP molecule for every 2 K+ ions pumped in and 3 Na+ ions pumped out [29] , we can calculate the pump energy consumption as |IP|/e . Units are hydrolysed ATP molecules/s when the current , IP , is given in ampere and the elementary charge , e , is in coulomb .
The electrical properties of the membrane , including the voltage-dependent activation and inactivation of conductances , determine the voltage response to injected current . We simulated the voltage responses of the photoreceptor to small injected current pulses to separate the contribution of the different voltage-dependent conductances ( Fig 4 ) . We performed these simulations at different light depolarisations , obtained by increasing the light-induced conductance . At all light depolarisations , activation of a voltage-dependent K+ conductance performs negative feedback , while inactivation of a voltage-dependent K+ conductance performs positive feedback . To distinguish the effects of negative and positive feedback , we freeze —i . e . we selectively keep fixed in their steady-state values— the activation and/or inactivation gating variables . When the inactivation gating variables are frozen positive feedback disappears . When all gating variables are frozen both positive and negative feedback disappear , and the voltage response reverts to the RC charging curve of a passive membrane . The voltage responses of photoreceptors with active membranes ( henceforth referred to as active photoreceptors ) initially follow the RC charging curve of the passive photoreceptor ( Fig 4 ) but diverge after a few milliseconds because the negative feedback produced by the activation of voltage-dependent conductances reduces the voltage deflection . Conductance inactivation is slower than activation , so the positive feedback produced from inactivation appears later than the negative feedback produced by activation . The dark adapted photoreceptor produces the largest voltage responses to injected current ( Fig 4a ) because there is little activation of voltage-dependent conductances at -68 mV . There is , however , some negative feedback due to the fast activation of Shaker and the comparatively slower activation of Shab , reducing the voltage signal of the active photoreceptor below that of the passive photoreceptor by about 15% . After a couple of tens of milliseconds , Shaker inactivation produces a small amplification ( approximately 1 . 5% after 150 ms ) . Increasing the light induced conductance , as occurs in low light intensities , depolarizes the simulated photoreceptor to -59 mV . The membrane resistance decreases because of the increased light-induced and K+ conductances . As a consequence , the voltage responses become smaller ( Fig 4b ) . Negative feedback becomes more apparent , subtracting approximately 25% from the voltage response , whilst positive feedback due to Shaker fast inactivation amplifies the depolarisation by approximately 9% at the end of a 150 ms pulse ( Fig 4b ) , and up to 15% in longer pulses . At high light levels , when the light-induced conductance depolarizes the model photoreceptor to -41 mV , the membrane resistance is further decreased and the voltage responses become smaller ( Fig 4c ) . Negative feedback due to conductance activation becomes conspicuous , decreasing gain by approximately 37% , thereby producing an overshoot and an after-potential . Positive feedback due to conductance inactivation is reduced ( less than 4% after a 150 ms pulse ) . At the end of the 150 ms pulse , about 20% of the positive feedback is due to Shab inactivation . This positive feedback is more apparent in longer time scales because of the slow dynamics of Shab inactivation , when it is responsible of about one third of the total inactivation . To analyse the contribution of the voltage-dependent conductances to the membrane’s frequency response , we calculated the impedance , Z ( f ) , which determines the linear filtering characteristics of the membrane in the frequency domain . By solving an equivalent electrical circuit that represents the linearization of the membrane incorporating voltage-dependent conductances , we obtained Eq 21 to calculate the impedance ( see Material and methods ) [14 , 23 , 27 , 39 , 40] . Using Eq 21 , we calculated the impedance at each light level , depolarising the photoreceptor by increasing the light conductance . In the dark , the membrane is a low-pass filter ( Fig 5a ) , not very different from an RC filter with the same resistance and impedance . The feedback produced by voltage-dependent conductances lowers the impedance of the active membrane at every frequency to below that of the passive membrane with the same membrane resistance and capacitance . At low light levels , when the photoreceptor is depolarized to between -59 mV and -50 mV , the impedance shows a conspicuous low frequency amplification below 2-3Hz ( Fig 5a ) that corresponds to the positive feedback due to Shaker inactivation . This amplification is produced solely by positive feedback from Shaker inactivation , and does not depend solely upon a window current as previously suggested [22] . At high light levels the photoreceptor is depolarized to above -41 mV , the negative feedback causes the impedance to become band-passing , and the Shaker-mediated amplification disappears ( Fig 5a ) . Only at the highest light levels does the positive feedback from Shab inactivation amplify frequencies below 1 Hz . At medium and high light levels , the Novel conductance produces negative feedback that decreases impedance at low frequencies diminishing the low frequency amplification of Shab . Many of the mechanisms that adapt photoreceptors to the prevailing light conditions , such as Ca2+ inhibiting TRP and TRPL channels [47] and closing the intracellular pupil [48] , would attenuate low frequencies . The attenuation of low frequencies in the LiC affects the overall photoreceptor frequency response , masking the amplification that is seen in the membrane . Consequently , there is no significant linear amplification in the light response [49] . This suggests that amplification below 2 Hz is unlikely to be the primary function of the Shaker conductance and is merely a side effect of the relatively slow inactivation of fruit fly voltage-dependent conductances . Thus , when calculating gain and bandwidth we only consider frequencies above 2 Hz . The photoreceptor membrane increases its bandwidth from around 26 Hz in the dark adapted state to more than 100 Hz at the highest light levels ( Fig 5d ) . Photoreceptor energy consumption at each depolarisation level can be calculated from the flux of Na+ and K+ ions across the membrane , which must be pumped back across to the opposite side by the Na+/K+ pump ( see Methods ) . The 4-fold increase in photoreceptor bandwidth from the lowest to the highest light levels is produced at the cost of a 4 . 3-fold increase in energy consumption , from 1 . 3 × 108 ATP/s in the dark to 5 . 6 × 108 ATP/s at the highest light levels ( Fig 5e ) . At all light levels , the conductance properties boost the bandwidth of the photoreceptor membrane above that of a passive membrane with identical resistance and capacitance ( Fig 5d ) . In the dark , the active photoreceptor membrane has a 6 Hz higher bandwidth than the corresponding passive membrane , increasing to 10 Hz at low light levels . At the highest light levels , the active photoreceptor membrane has a 40 Hz higher bandwidth than the corresponding passive membrane , an improvement of more than 70% . To achieve the same bandwidth as the active photoreceptor membrane with a passive membrane would increase the energy cost by 10% at low light levels rising to 40% at high light levels ( Fig 5e ) . Using the active and passive membrane impedances , we calculated the gain-bandwidth product ( GBWP ) , the product of bandwidth and peak impedance [23] . The GBWP of the active photoreceptor membrane generally increases with depolarisation , though there is a pronounced ‘dip’ at intermediate voltages around -52 mV ( Fig 6 ) . In contrast , the GBWP of the corresponding passive membranes remains constant . The increase in GBWP in the active photoreceptor is produced by the voltage-dependent K+ conductances . Changes in the membrane potential cause the voltage-dependent ion channels to alter their conformation changing their conductance after a delay , causing them to act like an inductance shunting ( in parallel to ) the membrane capacitance . This is akin to shunt peaking in an electronic amplifier , a technique in which an inductor is placed in series with the load resistance to compensate for capacitive effects and improve the frequency response [50 , 51] . In the photoreceptor membrane the inductive effects of voltage-dependent K+ conductances increase the GBWP [23] . At low light levels shunt peaking increases GBWP by 10-15% , and by 25% at the highest light levels . By selectively freezing the gating variables of the different voltage-dependent conductances , we distinguished the contributions of each conductance to shunt peaking ( Fig 6 ) . Shaker activation is the major contributor to shunt peaking at low light levels ( membrane voltages below -55 mV ) . At higher light levels , however , Shab activation becomes the major contributor to shunt peaking . Activation of the Novel conductance is too slow to significantly contribute to shunt peaking at any light level . The voltage-dependent Shaker and , to a lesser extent , the Shab conductances in Drosophila R1-6 photoreceptors inactivate . To assess the role of inactivation in these voltage-dependent K+ conductances , we froze the inactivation variable at different light levels , and assessed the bandwidth and energy consumption of these photoreceptor membranes across the full operating range of the photoreceptor ( Fig 7 ) . All other features of the photoreceptors remained identical . At low light levels , close to the resting potential , there is little inactivation of the voltage-dependent K+ conductances . Membranes with inactivation frozen at these low light levels have a similar bandwidth and energy consumption to typical photoreceptors . As light levels increase and the photoreceptor depolarises , however , the bandwidth and energy consumption of these membranes is substantially higher than that of a typical photoreceptor ( Fig 7a and 7b ) . When inactivation is frozen at high membrane potentials that correspond to high light levels , photoreceptor bandwidth and energy consumption is unaffected at high and at mid light levels ( Fig 7a and 7b ) . Conversely , at low light levels these membranes have a lower energy consumption but also a lower bandwidth . Although the increase in bandwidth at low light levels is small in absolute terms , it is relatively large . There is almost no change in the energy cost of bandwidth itself , so energy savings are made not through increased energy efficiency but rather by reducing bandwidth at high light levels ( Fig 7c ) . Thus , inactivation of a voltage-dependent K+ conductance permits relatively high bandwidths and energy costs at low light levels but ensures that there is no corresponding increase in bandwidth and energy costs at high light levels . As an alternative to inactivating voltage-dependent K+ conductances , the excessive energy consumption and bandwidth at depolarised potentials could be reduced using non-inactivating conductances with the same steady-state parameters . To study this case , we accelerated the dynamics of inactivation of both Shaker and Shab voltage-dependent K+ conductances to be identical to those of activation . We found that the energy consumption of bandwidth is increased at all depolarised membrane potentials ( Fig 7d ) . This reduction in the energy efficiency of the modified model is a consequence of decreased negative feedback caused by the weaker activation of the voltage-dependent conductances . Thus , the inactivation voltage-dependent K+ conductances improves the energy efficiency of photoreceptors , reducing the energy cost of bandwidth . Photoreceptor contrast gain depends not only on the membrane impedance but also on the light-induced current ( LiC ) . To determine how the membrane filter properties affect photoreceptor contrast gain , we incorporated a simplified , infinitely fast phototransduction cascade ( Eqs 22 and 23 ) . Under these simplifying conditions , the contrast gain is the product of the LiC and membrane impedance ( Eq 25 ) . Feedback from the voltage-dependent K+ conductances changes the impedance but does not change the LiC . Consequently , negative feedback from activation decreases the contrast gain of an active photoreceptor compared to that of a passive photoreceptor with identical membrane resistance and capacitance ( Fig 8a ) . The net consequence is that , in our model , voltage-dependent conductances decrease the photoreceptor peak contrast gain ( i . e . the maximum of contrast gain across different frequencies ) at all light intensities . The contrast gain-bandwidth product ( cGBWP ) is the product of the bandwidth and the peak contrast gain . In our simplified model , the contrast gain is the product of light induced current and membrane impedance ( Eq 25 ) , so the cGBWP is the product of GBWP and LiC . Consequently , the cGBWP of the active membrane exceeds that of the passive membrane by 10-15% at low light levels increasing to 25% at high light levels ( Fig 8b ) . A light-dependent modulation ( LDM ) has been described in Drosophila R1-6 photoreceptors , arising from the action of PIP2 upon microvillar-bound Shab channels , which shifts their activation 10 mV towards negative potentials [26] . To determine the effect of the LDM at different depolarisations , we kept the leak and light conductances unchanged , allowing the membrane potential to change to a new steady-state value . The Na+/K+ pump and Na+/Ca2+ exchanger current change accordingly . The shift in Shab voltage-activation range produced by PIP2 depletion hyperpolarises the photoreceptor and decreases the membrane impedance ( Fig 9a ) . Consequently , the LDM speeds up responses by widening membrane bandwidth ( Fig 9 ) [26] , and causes the effects of Shaker become less conspicuous . The increase in bandwidth is small at low light levels becoming more pronounced at higher light intensities , when the membrane potential is above -55 mV and the Shab current is larger . The Shab mediated band-passing at high light intensities becomes more prominent , and bandwidth increases by up to 18 Hz ( Fig 9b ) . As well as increasing bandwidth , LDM affects photoreceptor energy consumption . At low light levels , LDM has little effect upon photoreceptor energy consumption but at high light intensities energy consumption is increased by as much as ∼10% ( Fig 9c ) . Ca2+ activated calmodulin modulates the maximum conductance of Shab [52] . A decrease of 50% in the maximum Shab conductance hyperpolarises the photoreceptor and increases impedance , reducing the effect of Shab mediated band-passing at the highest light levels and reducing bandwidth by up to ∼20% . It also increases gain by ∼15% and decreases energy cost by up to ∼10% . Conversely , an increase of Shab maximum conductance , the expected effect of an increase in Ca2+/calmodulin upon light adaptation , has the opposite effect . Serotonin ( 5HT ) reversibly shifts the activation and inactivation curves of the Shab conductance by ∼ 30mV and the activation and inactivation curves of the Shaker conductance by 10 to 14 mV [24] . The higher activation voltages of the shifted K+ conductances caused by serotonin depolarises photoreceptors slightly and alters their filter properties ( Fig 10 ) . At each light-induced level of depolarisation photoreceptor impedance increases ( Fig 10a ) [25] . Serotonin also abolishes Shaker-mediated amplification by shifting inactivation [25] . Consistent with the reduction in the negative feedback contributed by Shab , serotonin decreases bandwidth at all depolarisations ( Fig 10b ) , and by as much as 40% at high light levels . By modulating the properties of the Shab and Shaker conductances , serotonin also decreases energy consumption at all light levels by as much as ∼40% ( Fig 10c ) . By altering the voltage-dependent K+ conductances , modulation can adjust the gain-bandwidth product ( GBWP ) of the active photoreceptor membranes . Because the effects of modulation are specific , so too are the effects on shunt peaking and , consequently , the GBWP . Light-dependent modulation ( LDM ) has little effect at low and high light levels ( -60 and -36 mV , respectively ) but increases the GBWP at intermediate light levels ( -52 to -44 mV ) in comparison to the unmodulated active photoreceptor membrane ( Fig 11a ) . The increase in GBWP is a consequence of LDM inducing a 10 mV negative shift in Shab activation so that it occurs at intermediate light levels producing shunt peaking , which extends bandwidth and increases the GBWP . Reduction of the Shab conductance by 50% , as produced by Ca2+ activated calmodulin modulation , reduces the photoreceptor GBWP at high light levels ( -44 to -36 mV ) but has no substantial effect at low and intermediate light levels ( Fig 11b ) . The reduction in GBWP at high light levels is a consequence of the smaller Shab conductance , which reduces the shunt peaking produced by Shab activation and , consequently , the GBWP . Serotonin also reduces the GBWP at high light levels but additionally reduces the GBWP at low light levels and abolishes the ‘dip’ at intermediate voltages ( -52 mV ) producing a monotonic increase GBWP with depolarisation ( Fig 11c ) . The effects on GBWP at low and intermediate light levels are due to the positive shift in the Shaker conductance caused by serotonin; Shaker activation at intermediate light levels causes shunt peaking that increases the GBWP , and simultaneously removes shunt peaking at low light levels . The positive shift in Shab conductance by serotonin reduces shunt peaking produced by Shab , thereby reducing the GBWP at high light levels . The changes in photoreceptor membrane gain and bandwidth wrought by the modulation of voltage-dependent K+ conductances have implications for contrast coding . Light-dependent modulation ( LDM ) of the Shab conductance reduces photoreceptor contrast gain below the corner frequency at all light levels ( Fig 12a ) , increasing photoreceptor bandwidth ( Fig 13a ) . At the highest light levels , LDM enhances the bandpassing of the membrane . Conversely , serotonergic modulation of the Shab and Shaker conductances increases photoreceptor contrast gain below the corner frequency at all light levels ( Fig 12b ) , decreasing photoreceptor bandwidth ( Fig 13a ) . At the highest light levels , serotonin modulation removes the bandpassing of the membrane entirely . To quantify the impact of modulation on the trade-off between contrast gain for bandwidth , we calculated the peak contrast gain and bandwidth at each light level ( Fig 13 ) . The dark adapted photoreceptor membrane has a low peak contrast gain and low bandwidth . At low light levels , peak contrast gain rises markedly though there is almost no change in bandwidth . At intermediate light levels , peak contrast gain rises further accompanied by a substantial increase in bandwidth , whilst at the highest light levels there is almost no change in peak contrast gain but bandwidth nearly doubles . Light-dependent modulation ( LDM ) shifts the relationship between peak contrast gain and bandwidth to higher bandwidths at an equivalent peak contrast gain . Conversely , serotonin modulation shifts the relationship between peak contrast gain and bandwidth to lower bandwidths at an equivalent peak contrast gain . Thus , modulation of the photoreceptor membrane is capable of trading-off contrast gain for bandwidth ( Fig 14a ) . By altering photoreceptor energy consumption , modulation also alters the energy cost of bandwidth and of peak contrast gain . As light level increases , so too does bandwidth and energy cost . The energy cost of the peak contrast gain follows a similar relationship to that of bandwidth and peak contrast gain . Light-dependent modulation ( LDM ) reduces the energy cost of bandwidth but increases that of contrast gain ( Fig 13b and 13c ) . Conversely , serotonin increases the energy cost of bandwidth and decreases that of contrast gain ( Fig 13b and 13c ) . Thus , modulation adjusts the energy cost of peak contrast gain and bandwidth .
We assessed photoreceptor membrane performance by calculating the product of bandwidth and the peak of the impedance function , the gain-bandwidth product ( GBWP ) . This measure combines two biologically germane features of the membrane , gain and bandwidth; a sufficiently high gain can prevent signals from being corrupted by noise , whilst a sufficiently high bandwidth permits fast signals to be transmitted . In both cell membranes and operational amplifiers , the GBWP determines the maximum gain for a given bandwidth , and vice versa [23] . In cell membranes , the GBWP is determined by the membrane capacitance , and specific membrane capacitance is relatively invariant . When ion channels reduce the membrane resistance or produce negative feedback , gain is traded for bandwidth . Only by modifying the one-pole filtering of the membrane ( e . g . by using shunt-peaking ) can the membrane increase the GBWP and thus increase gain for a given bandwidth or bandwidth for a given gain [23] . Thus , the GBWP provides a single variable allowing quantification of the overall photoreceptor membrane performance . We extended this idea to include the light-induced current ( LiC ) . A useful concept when describing the photoreceptor response to light is the contrast gain , the frequency-dependent gain between light contrast and voltage in the photoreceptor soma . As the product of the LiC and the membrane impedance , the contrast gain combines the outputs of the phototransduction cascade and the photoinsensitive membrane . The contrast gain-bandwidth product ( cGBWP ) , is the product of the bandwidth and the peak contrast gain , quantifying the impact of the membrane upon contrast gain and bandwidth . As the cGBWP is also the product of GBWP and LiC , it can be increased partially through shunt-peaking , but also through an increase in LiC at more depolarised membrane potentials . We began by assessing the impact of voltage-dependent K+ conductances themselves upon the graded electrical signalling and energy consumption of fruit fly photoreceptor membranes . By activating upon depolarisation , the Shaker , Shab and Novel conductances produce negative feedback . The main role of this feedback is to reduce the impedance to below that of a passive photoreceptor with identical membrane resistance and capacitance , as is the case for delayed rectifier conductances in blowflies [14] . Activation of the Shaker conductance at low depolarizations and the Shab conductance at high depolarizations produces shunt peaking , which is again produced by delayed rectifier conductances in blowflies [23] . Shunt peaking is produced by voltage-dependent ion channels changing their conductance with membrane potential after a delay , causing them to act like an inductance . This increases bandwidth without decreasing membrane gain , thereby increasing the photoreceptor membrane gain-bandwidth product ( GBWP ) . The combination of the decrease in gain by negative feedback and the increase in GBWP by shunt-peaking caused by voltage-dependent K+ conductances produces an increase in bandwidth without incurring the full cost that reducing the membrane resistance of a passive membrane would produce . Consequently , voltage-dependent conductances modify the trade-offs between membrane bandwidth , gain and energy cost by making it cheaper to operate a membrane with lower gain and higher bandwidth . At the resting potential and at low light levels a window current [40] is present in D . melanogaster photoreceptors produced by the overlap between the activation and inactivation of the Shaker K+ conductances [22 , 25 , 49] . The reduction in the window current upon depolarisation has been assumed by previous studies to produce amplification at low frequencies in D . melanogaster photoreceptors [22 , 25 , 49] . Our biophysical models show that this increase in the gain of the membrane at low frequencies occurs against a background of negative feedback produced by the activation of other conductances as well as that of Shaker itself . In some D . melanogaster photoreceptors , amplification may be sufficiently prominent so as to produce an impedance that at low frequencies is above the membrane resistance ( Frolov , pers . comm . ) . However , similarly to our model , in most photoreceptors the impedance is unlikely to exceed the membrane resistance , and so it is rather a reduction in attenuation . Amplification by the Shab conductance at high light levels has not , to our knowledge , been described before . The amplification is restricted to lower frequencies than that of Shaker , because of the slower inactivation dynamics of Shab . It is smaller because Shab inactivation occurs at high light levels when input resistance is low , reducing the relative effect of feedback . Moreover , the Shab steady-state inactivation curve has a relatively shallow slope and is centred at -25 . 7 mV , outside the range of typical steady-state depolarisations . The functional significance of low frequency amplification remains unclear . Indeed , low frequency attenuation by the light current [37] makes it likely that the amplification produced by the Shaker and Shab conductances has little or no impact upon signalling in fruit fly photoreceptors , and is merely a side effect of inactivation . Thus , the role of Shaker in fruit fly photoreceptors is similar to that of A-type conductances in the dendrites of vertebrate interneurons: decreasing the amplitude and time-to-peak of depolarising postsynaptic potentials rather than amplifying them [54] . Our modelling suggests that by increasing bandwidth at low light levels whilst ensuring that high bandwidths do not occur at high light levels , Shaker conductance inactivation reduces energy consumption in comparison to an equivalent non-inactivating conductance ( Fig 7 ) , as hypothesised by Niven et al . [21] . This raises the possibility that the inactivation of A-type conductances in dendrites [54] may also be linked to reducing energy consumption . Other molecular components of the photoreceptors can also produce low frequency amplification . For example , the Na+/Ca2+ , which has a time constant of 350 ms [34] , can produce positive feedback amplifying frequencies below 1 Hz . However , the fraction of Ca2+ in the LiC is only about 26% [34] , and thus the exchanger net current is only 13% of the LiC , making the contribution of this positive feedback small . Moreover , this positive feedback is unlikely to affect photoreceptor bandwidth and gain because the light response filters low frequencies [37] . To maintain consistency with other choices of our model , such as only considering frequencies above 2 Hz in bandwidth and gain calculations , we considered the Na+/Ca2+ exchanger as a constant current at each steady-state depolarisation , while studying the response to small signals . Our models also incorporated the Na+/K+ pump as a constant current at each steady-state depolarisation , rather than incorporating the actual dynamics . This approximation is accurate for the Na+/K+ pump current because its slow dynamics ( seconds , e . g . [29] ) restricts its negative feedback to extremely low frequencies . Neuromodulators alter specific properties of the voltage-dependent K+ conductances , shifting the trade-off between contrast gain and bandwidth in fruit fly photoreceptors . Light-dependent modulation ( LDM ) of the Shab conductance by PIP2 depletion , which shifts Shab activation to lower potentials [26] , moves the trade-off to favour higher bandwidths and lower peak contrast gains . Conversely , the positive shift in the voltage operating range of both the Shaker and Shab conductances caused by serotonin [24] , moves the trade-off to favour lower bandwidths and higher peak contrast gains . Likewise , Ca2+/Calmodulin antagonists that decrease the Shab conductance [52] also favour lower bandwidths and higher peak contrast gains . Conversely , an increase in the Shab conductance would favour higher bandwidths and lower peak contrast gains . So , modulators enable fruit fly photoreceptors to explore a greater region of the trade-off between peak contrast gain and bandwidth . This enables photoreceptors to adjust gain or bandwidth depending on the prevailing conditions . Both shifts produced by products of the photocascade , PIP2 depletion and an increase in Shab conductance produced by an increase in Ca2+ [26 , 52] , sacrifice gain for bandwidth thereby contributing to light adaptation ( Figs 14a and 15 ) . Serotonin [24] has the opposite effect , increasing gain at the cost of bandwidth potentially adjusting photoreceptors to low light levels ( Figs 14a and 15 ) . By adjusting voltage-dependent conductances , these neuromodulatory changes also have implications for photoreceptor energy consumption . If the light-induced conductance is kept constant , the neuromodulatory effects of phototransduction cascade products [26 , 52] that increase bandwidth also increase in energy consumption . Thus , they reinforce the ability of the membrane to produce high bandwidth only when it is needed . This allows photoreceptors to save energy by avoiding investing in high bandwidth when it is not needed . Conversely , serotonin [24] reduces bandwidth and decreases the energy consumption if the light-induced conductance is kept constant . For a given energy consumption , light-dependent modulation ( LDM ) by PIP2 depletion decreases the cost of bandwidth but increases the cost of gain ( Fig 13b and 13c ) . Conversely , serotonin decreases photoreceptor bandwidth but increases gain ( Fig 13b and 13c ) . Likewise , Ca2+/Calmodulin antagonists also increase the energy cost of bandwidth and decreases the energy cost of gain . An increase in Ca2+/Calmodulin also decreases the cost of bandwidth but increases the cost of gain . The presence of multiple voltage-dependent conductances combined with multiple neuromodulators allows the costs of gain and bandwidth to be adjusted in independent ways . By shifting the balance between gain and bandwidth ( Fig 14a ) neuromodulators can change the contrast-gain-bandwidth product ( cGBWP ) . However , they have little effect on the cost of the cGBWP ( Fig 14b ) . This can be understood by considering that the cGBWP is the product of the membrane GBWP times the light current , while the energy cost is proportional to the depolarising current —which is dominated by the light current at all but the lowest light levels . Thus , the cost of cGBWP is proportional to the membrane GBWP . As GBWP tends to be restricted between 1 . 2 and 1 . 5 for most choices of conductances , the cost of cGBWP varies little upon depolarisation or conductance shifts . We used a simplified model of phototransduction in which the photocurrent follows photon absorption . This model fails to capture low frequency filtering by light adaptation and high frequency filtering by the finite width and latency dispersion of the quantum bump [44 , 45] . Indeed , in Drosophila photoreceptors the phototransduction cascade current has a bandwidth between two and five times lower than the membrane bandwidth [37] . Incorporating the light-induced current ( LiC ) into the model allowed us to study photoreceptor gain in a more relevant way than with membrane impedance alone . If membrane impedance is decreased using negative feedback , no increase in the LiC is needed to keep the membrane voltage steady . The decreased impedance reduces the membrane gain but , because the LiC is unchanged , the net result is a decrease in contrast gain . Conversely , if the membrane impedance were decreased by reducing the membrane resistance , an increase in the LiC would be necessary to keep the membrane voltage steady . The net result would be that contrast gain is unchanged . This illustrates the importance of the interplay between the LiC and the membrane impedance . Within the arthropods , neuromodulation of photoreceptors is widespread . Some of these neuromodulators target the voltage-dependent conductances , whereas others act on molecular components of photoreceptor phototransduction cascade . In the fruit fly ( Drosophila melanogaster ) , both octopamine and dopamine can increase the latency of quantum bumps without changing the membrane properties [55] . Likewise , in the desert locust ( Schistocerca gregaria ) serotonin shifts photoreceptors from the day state in which they express a voltage-dependent K+ conductance with delayed rectifier-like properties to the night state in which they express a rapidly inactivating voltage-dependent K+ conductance [56] . Just as in D . melanogaster photoreceptors , the modulation of voltage-dependent K+ conductances by serotonin in desert locusts is likely to boost gain at low light intensities , increasing quantum bump amplitude thereby making them easier to detect . Increasing the expression of a rapidly inactivating voltage-dependent K+ conductance will also likely reduce locust photoreceptor energy investment in bandwidth , producing a similar redistribution of energy investment in gain and bandwidth that our modelling shows is produced by serotonergic modulation in D . melanogaster photoreceptors . The light response of desert locust photoreceptors is also increased by nitric oxide and cGMP [57] , again improving detectability . The combination of neuromodulators that adjust elements of the phototransduction cascade and the photoinsensitive membrane separately provides considerable flexibility within photoreceptors to match prevailing environmental conditions , thereby ensuring that energy is invested in high gains or high bandwidths only when it is needed . Avoiding expenditure on high bandwidth when it is unnecessary is an important strategy for reducing the cost of vision . Such changes may be predicted rather than being generated by prevailing environmental conditions , such as in the horseshoe crab ( Limulus ) in which photoreceptors undergo 24-hour cyclic changes in sensitivity caused by octopamine ( for a review see [58] ) . The effects of neuromodulators upon the dynamics of single neurons and neural circuits are well documented in both invertebrates and vertebrates [1 , 2 , 59] . Whilst neuromodulatory effects on the performance of single neurons and neural circuits have been described extensively , their impact upon the energy consumption has been largely ignored ( but see [60 , 61] ) . Our results show that the changes wrought by neuromodulators to voltage-dependent conductances are likely to affect the energy consumption of neurons and neural circuits . Indeed , many changes caused by neuromodulators , such as to sensory or synaptic inputs/outputs , or the reconfiguration of neural circuits , have implications for neuronal energy consumption . Although some changes may increase or decrease overall energy consumption [60 , 61] , our findings suggest that the impact may be more subtle; neuromodulators may redistribute energy investment in different aspects of neuronal performance without substantially altering overall energy consumption . Mechanoreceptors demonstrate the potential for neuromodulators to adjust or redistribute neuronal energy consumption . Neuromodulators can adjust numerous properties of invertebrate mechanoreceptors including their firing rate , dynamics , sensitivity and spike timing precision [7 , 62–69] . Although in many cases the underlying causes of these changes are unknown , in some they are linked to changes in conductance . For example , in the crab ( Carcinus maenas ) neuromodulation by allatostatin or serotonin of mechanoreceptors can adjust their spike rates and spike timing precision through changes in conductance; high conductance states are associated with lower spike rates and higher spike timing precision [7] . In such cases , there will be implications both for the energy consumption of the neuron and for the cost of information coding . High conductance states to increase spike timing precision caused by allatostatin will likely increase the energy consumption of spikes and of maintaining the resting potential but the accompanying reduction in spike rate may reduce energy consumption [7] . In these mechanoreceptors , serotonin has the opposite effect producing a low conductance , high spike rate state . Thus , as in fruit fly photoreceptors , neuromodulators in crab mechanoreceptors may redistribute the energy invested in different aspects of neuronal coding . The ability of neuromodulators to redistribute the energy invested in different aspects of neuronal performance , and potentially to adjust the overall energy expenditure , has implications not only for single neurons but also for neural circuits . There are potential energy savings to the reconfiguration of neural circuits by neuromodulators to generate different behavioural outputs . The reuse of neurons to generate different behaviours reduces the need to build and maintain additional neurons , which would reduce energy consumption [70] . Neuromodulators may also redistribute energy investment in different parts of neural circuits . Indeed , neuromodulation may allow concerted changes in neurons and across neural circuits to co-ordinate investment in different aspects of neural processing to match prevailing behavioural and environmental circumstances .
|
The properties of neurons and neural circuits can be adjusted by neuromodulators , molecules that alter their ability to respond to future activity . Many neuromodulators target voltage-dependent ion channels , molecular components of cell membranes that influence the electrical activity of neurons . Because of their importance , the action of neuromodulators upon voltage-dependent ion channels and the subsequent changes in neural activity has been studied extensively . However , the properties of voltage-dependent ion channels also influence the energy that neural signalling consumes . Here we assess the impact of neuromodulators upon neuronal energy consumption . We use analytical and computational models to determine the impact of different neuromodulators upon the signalling properties and energy consumption of fly photoreceptors . Our models uncover previously unknown properties of voltage-dependent ion channels in fly photoreceptors , showing how they adjust the membrane properties , gain and bandwidth , to prevailing light levels . Neuromodulators alter voltage-dependent ion channel properties , adjusting the gain and bandwidth . Although neuromodulators do not substantially alter the overall energy consumption of photoreceptors , they redistribute energy investment in gain and bandwidth . Hence , our models provide novel insights into the functions that neuromodulators play in neurons and neural circuits .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion"
] |
[
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"neurochemistry",
"engineering",
"and",
"technology",
"signal",
"processing",
"membrane",
"potential",
"social",
"sciences",
"light",
"electrophysiology",
"neuroscience",
"animals",
"electromagnetic",
"radiation",
"animal",
"models",
"serotonin",
"drosophila",
"melanogaster",
"model",
"organisms",
"experimental",
"organism",
"systems",
"neurotransmitters",
"cellular",
"structures",
"and",
"organelles",
"drosophila",
"signal",
"bandwidth",
"research",
"and",
"analysis",
"methods",
"animal",
"cells",
"animal",
"studies",
"sensory",
"receptors",
"insects",
"cell",
"membranes",
"arthropoda",
"physics",
"biochemistry",
"signal",
"transduction",
"cellular",
"neuroscience",
"psychology",
"neuromodulation",
"eukaryota",
"cell",
"biology",
"physiology",
"neurons",
"photoreceptors",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"afferent",
"neurons",
"sensory",
"perception",
"physical",
"sciences",
"biogenic",
"amines",
"organisms"
] |
2018
|
Modulation of voltage-dependent K+ conductances in photoreceptors trades off investment in contrast gain for bandwidth
|
Numerous studies are currently underway to characterize the microbial communities inhabiting our world . These studies aim to dramatically expand our understanding of the microbial biosphere and , more importantly , hope to reveal the secrets of the complex symbiotic relationship between us and our commensal bacterial microflora . An important prerequisite for such discoveries are computational tools that are able to rapidly and accurately compare large datasets generated from complex bacterial communities to identify features that distinguish them . We present a statistical method for comparing clinical metagenomic samples from two treatment populations on the basis of count data ( e . g . as obtained through sequencing ) to detect differentially abundant features . Our method , Metastats , employs the false discovery rate to improve specificity in high-complexity environments , and separately handles sparsely-sampled features using Fisher's exact test . Under a variety of simulations , we show that Metastats performs well compared to previously used methods , and significantly outperforms other methods for features with sparse counts . We demonstrate the utility of our method on several datasets including a 16S rRNA survey of obese and lean human gut microbiomes , COG functional profiles of infant and mature gut microbiomes , and bacterial and viral metabolic subsystem data inferred from random sequencing of 85 metagenomes . The application of our method to the obesity dataset reveals differences between obese and lean subjects not reported in the original study . For the COG and subsystem datasets , we provide the first statistically rigorous assessment of the differences between these populations . The methods described in this paper are the first to address clinical metagenomic datasets comprising samples from multiple subjects . Our methods are robust across datasets of varied complexity and sampling level . While designed for metagenomic applications , our software can also be applied to digital gene expression studies ( e . g . SAGE ) . A web server implementation of our methods and freely available source code can be found at http://metastats . cbcb . umd . edu/ .
The increasing availability of high-throughput , inexpensive sequencing technologies has led to the birth of a new scientific field , metagenomics , encompassing large-scale analyses of microbial communities . Broad sequencing of bacterial populations allows us a first glimpse at the many microbes that cannot be analyzed through traditional means ( only ∼1% of all bacteria can be isolated and independently cultured with current methods [1] ) . Studies of environmental samples initially focused on targeted sequencing of individual genes , in particular the 16S subunit of ribosomal RNA [2]–[5] , though more recent studies take advantage of high-throughput shotgun sequencing methods to assess not only the taxonomic composition , but also the functional capacity of a microbial community [6]–[8] . Several software tools have been developed in recent years for comparing different environments on the basis of sequence data . DOTUR [9] , Libshuff [10] , ∫-libshuff [11] , SONs [12] , MEGAN [13] , UniFrac [14] , and TreeClimber [15] all focus on different aspects of such an analysis . DOTUR clusters sequences into operational taxonomic units ( OTUs ) and provides estimates of the diversity of a microbial population thereby providing a coarse measure for comparing different communities . SONs extends DOTUR with a statistic for estimating the similarity between two environments , specifically , the fraction of OTUs shared between two communities . Libshuff and ∫-libshuff provide a hypothesis test ( Cramer von Mises statistics ) for deciding whether two communities are different , and TreeClimber and UniFrac frame this question in a phylogenetic context . Note that these methods aim to assess whether , rather than how two communities differ . The latter question is particularly important as we begin to analyze the contribution of the microbiome to human health . Metagenomic analysis in clinical trials will require information at individual taxonomic levels to guide future experiments and treatments . For example , we would like to identify bacteria whose presence or absence contributes to human disease and develop antibiotic or probiotic treatments . This question was first addressed by Rodriguez-Brito et al . [16] , who use bootstrapping to estimate the p-value associated with differences between the abundance of biological subsytems . More recently , the software MEGAN of Huson et al . [13] provides a graphical interface that allows users to compare the taxonomic composition of different environments . Note that MEGAN is the only one among the programs mentioned above that can be applied to data other than that obtained from 16S rRNA surveys . These tools share one common limitation — they are all designed for comparing exactly two samples — therefore have limited applicability in a clinical setting where the goal is to compare two ( or more ) treatment populations each comprising multiple samples . In this paper , we describe a rigorous statistical approach for detecting differentially abundant features ( taxa , pathways , subsystems , etc . ) between clinical metagenomic datasets . This method is applicable to both high-throughput metagenomic data and to 16S rRNA surveys . Our approach extends statistical methods originally developed for microarray analysis . Specifically , we adapt these methods to discrete count data and correct for sparse counts . Our research was motivated by the increasing focus of metagenomic projects on clinical applications ( e . g . Human Microbiome Project [17] ) . Note that a similar problem has been addressed in the context of digital gene expression studies ( e . g . SAGE [18] ) . Lu et al . [19] employ an overdispersed log-linear model and Robinson and Smyth [20] use a negative binomial distribution in the analysis of multiple SAGE libraries . Both approaches can be applied to metagenomic datasets . We compare our tool to these prior methodologies through comprehensive simulations , and demonstrate the performance of our approach by analyzing publicly available datasets , including 16S surveys of human gut microbiota and random sequencing-based functional surveys of infant and mature gut microbiomes and microbial and viral metagenomes . The methods described in this paper have been implemented as a web server and are also available as free source-code ( in R ) from http://metastats . cbcb . umd . edu .
To account for different levels of sampling across multiple individuals , we convert the raw abundance measure to a fraction representing the relative contribution of each feature to each of the individuals . This results in a normalized version of the matrix described above , where the cell in the ith row and the jth column ( which we shall denote fij ) is the proportion of taxon i observed in individual j . We chose this simple normalization procedure because it provides a natural representation of the count data as a relative abundance measure , however other normalization approaches can be used to ensure observed counts are comparable across samples , and we are currently evaluating several such approaches . For each feature i , we compare its abundance across the two treatment populations by computing a two-sample t statistic . Specifically , we calculate the mean proportion , and variance of each treatment t from which nt subjects ( columns in the matrix ) were sampled:We then compute the two-sample t statistic:Features whose t statistics exceeds a specified threshold can be inferred to be differentially abundant across the two treatments ( two-sided t-test ) . The threshold for the t statistic is chosen such as to minimize the number of false positives ( features incorrectly determined to be differentially abundant ) . Specifically , we try to control the p-value—the likelihood of observing a given t statistic by chance . Traditional analyses compute the p-value using the t distribution with an appropriate number of degrees of freedom . However , an implicit assumption of this procedure is that the underlying distribution is normal . We do not make this assumption , but rather estimate the null distribution of ti non-parametrically using a permutation method as described in Storey and Tibshirani [21] . This procedure , also known as the nonparametric t-test has been shown to provide accurate estimates of significance when the underlying distributions are non-normal [22] , [23] . Specifically , we randomly permute the treatment labels of the columns of the abundance matrix and recalculate the t statistics . Note that the permutation maintains that there are n1 replicates for treatment 1 and n2 replicates for treatment 2 . Repeating this procedure for B trials , we obtain B sets of t statistics: t10b , … , tM0b , b = 1 , … , B , where M is the number of rows in the matrix . For each row ( feature ) , the p-value associated with the observed t statistic is calculated as the fraction of permuted tests with a t statistic greater than or equal to the observed ti:This approach is inadequate for small sample sizes in which there are a limited number of possible permutations of all columns . As a heuristic , if less than 8 subjects are used in either treatment , we pool all permuted t statistics together into one null distribution and estimate p-values as: Note that the choice of 8 for the cutoff is simply heuristic based on experiments during the implementation of our method . Our approach is specifically targeted at datasets comprising multiple subjects — for small data-sets approaches such as that proposed by Rodriguez-Brito et . al . [16] might be more appropriate . Unless explicitly stated , all experiments described below used 1000 permutations . In general , the number of permutations should be chosen as a function of the significance threshold used in the experiment . Specifically , a permutation test with B permutations can only estimate p-values as low as 1/B ( in our case 10−3 ) . In datasets containing many features , larger numbers of permutations are necessary to account for multiple hypothesis testing issues ( further corrections for this case are discussed below ) . Precision of the p-value calculations is obviously improved by increasing the number of permutations used to approximate the null distribution , at a cost , however , of increased computational time . For certain distributions , small p-values can be efficiently estimated using a technique called importance sampling . Specifically , the permutation test is targeted to the tail of the distribution being estimated , leading to a reduction in the number of permutations necessary of up to 95% [24] , [25] . We intend to implement such an approach in future versions of our software . For complex environments ( many features/taxa/subsystems ) , the direct application of the t statistic as described can lead to large numbers of false positives . For example , choosing a p-value threshold of 0 . 05 would result in 50 false positives in a dataset comprising 1000 organisms . An intuitive correction involves decreasing the p-value cutoff proportional to the number of tests performed ( a Bonferroni correction ) , thereby reducing the number of false positives . This approach , however , can be too conservative when a large number of tests are performed [21] . An alternative approach aims to control the false discovery rate ( FDR ) , which is defined as the proportion of false positives within the set of predictions [26] , in contrast to the false positive rate defined as the proportion of false positives within the entire set of tests . In this context , the significance of a test is measured by a q-value , an individual measure of the FDR for each test . We compute the q-values using the following algorithm , based on Storey and Tibshirani [21] . This method assumes that the p-values of truly null tests are uniformly distributed , assumption that holds for the methods used in Metastats . Given an ordered list of p-values , p ( 1 ) ≤p ( 2 ) ≤…≤p ( m ) , ( where m is the total number of features ) , and a range of values λ = 0 , 0 . 01 , 0 . 02 , … , 0 . 90 , we computeNext , we fit with a cubic spline with 3 degrees of freedom , which we denote , and let . Finally , we estimate the q-value corresponding to each ordered p-value . First , . Then for i = m-1 , m-2 , … , 1 , Thus , the hypothesis test with p-value has a corresponding q-value of . Note that this method yields conservative estimates of the true q-values , i . e . . Our software provides users with the option to use either p-value or q-value thresholds , irrespective of the complexity of the data . For low frequency features , e . g . low abundance taxa , the nonparametric t–test described above is not accurate [27] . We performed several simulations ( data not shown ) to determine the limitations of the nonparametric t-test for sparsely-sampled features . Correspondingly , our software only applies the test if the total number of observations of a feature in either population is greater than the total number of subjects in the population ( i . e . the average across subjects of the number of observations for a given feature is greater than one ) . We compare the differential abundance of sparsely-sampled ( rare ) features using Fisher's exact test . Fisher's exact test models the sampling process according to a hypergeometric distribution ( sampling without replacement ) . The frequencies of sparse features within the abundance matrix are pooled to create a 2×2 contingency table ( Figure 2 ) , which acts as input for a two-tailed test . Using the notation from Figure 2 , the null hypergeometric probability of observing a 2×2 contingency table is: By calculating this probability for a given table , and all tables more extreme than that observed , one can calculate the exact probability of obtaining the original table by chance assuming that the null hypothesis ( i . e . no differential abundance ) is true [27] . Note that an alternative approach to handling sparse features is proposed in microarray literature . The Significance Analysis of Microarrays ( SAM ) method [28] addresses low levels of expression using a modified t statistic . We chose to use Fisher's exact test due to the discrete nature of our data , and because prior studies performed in the context of digital gene expression indicate Fisher's test to be effective for detection of differential abundance [29] . The input to our method , the Feature Abundance Matrix , can be easily constructed from both 16S rRNA and random shotgun data using available software packages . Specifically for 16S taxonomic analysis , tools such as the RDP Bayesian classifier [30] and Greengenes SimRank [31] output easily-parseable information regarding the abundance of each taxonomic unit present in a sample . As a complementary , unsupervised approach , 16S sequences can be clustered with DOTUR [9] into operational taxonomic units ( OTUs ) . Abundance data can be easily extracted from the “* . list” file detailing which sequences are members of the same OTU . Shotgun data can be functionally or taxonomically classified using MEGAN [13] , CARMA [32] , or MG-RAST [33] . MEGAN and CARMA are both capable of outputting lists of sequences assigned to a taxonomy or functional group . MG-RAST provides similar information for metabolic subsystems that can be downloaded as a tab-delimited file . All data-types described above can be easily converted into a Feature Abundance Matrix suitable as input to our method . In the future we also plan to provide converters for data generated by commonly-used analysis tools . Human gut 16S rRNA sequences were prepared as described in Eckburg et al . and Ley et al . ( 2006 ) and are available in GenBank , accession numbers: DQ793220-DQ802819 , DQ803048 , DQ803139-DQ810181 , DQ823640-DQ825343 , AY974810-AY986384 . In our experiments we assigned all 16S sequences to taxa using a naïve Bayesian classifier currently employed by the Ribosomal Database Project II ( RDP ) [30] . COG profiles of 13 human gut microbiomes were obtained from the supplementary material of Kurokawa et al . [34] . We acquired metabolic functional profiles of 85 metagenomes from the online supplementary materials of Dinsdale et al . ( 2008 ) ( http://www . theseed . org/DinsdaleSupplementalMaterial/ ) .
As outlined in the introduction , statistical packages developed for the analysis of SAGE data are also applicable to metagenomic datasets . In order to validate our method , we first designed simulations and compared the results of Metastats to Student's t-test ( with pooled variances ) and two methods used for SAGE data: a log-linear model ( Log-t ) by Lu et al . [19] , and a negative binomial ( NB ) model developed by Robinson and Smyth [20] . We designed a metagenomic simulation study in which ten subjects are drawn from two groups - the sampling depth of each subject was determined by random sampling from a uniform distribution between 200 and 1000 ( these depths are reasonable for metagenomic studies ) . Given a population mean proportion p and a dispersion value φ , we sample sequences from a beta-binomial distribution Β ( α , β ) , where α = p ( 1/φ−1 ) and β = ( 1−p ) ( 1/φ−1 ) . Note that data from this sampling procedure fits the assumptions for Lu et al . as well as Robinson and Smyth and therefore we expect them to do well under these conditions . Lu et al . designed a similar study for SAGE data , however , for each simulation , a fixed dispersion was used for both populations and the dispersion estimates were remarkably small ( φ = 0 , 8e-06 , 2e-05 , 4 . 3e-05 ) . Though these values may be reasonable for SAGE data , we found that they do not accurately model metagenomic data . Figure 3 displays estimated dispersions within each population for all features of the metagenomic datasets examined below . Dispersion estimates range from 1e-07 to 0 . 17 , and rarely do the two populations share a common dispersion . Thus we designed our simulation so that φ is chosen for each population randomly from a uniform distribution between 1e-08 and 0 . 05 , allowing for potential significant differences between population distributions . For each set of parameters , we simulated 1000 feature counts , 500 of which are generated under p1 = p2 , the remainder are differentially abundant where a*p1 = p2 , and compared the performance of each method using receiver-operating-characteristic ( ROC ) curves . Figure 4 displays the ROC results for a range of values for p and a . For each set of parameters , Metastats was run using 5000 permutations to compute p-values . Metastats performs as well as other methods , and in some cases is preferable . We also found that in most cases our method was more sensitive than the negative binomial model , which performed poorly for high abundance features . Our next simulation sought to examine the accuracy of each method under extreme sparse sampling . As shown in the datasets below , it is often the case that a feature may not have any observations in one population , and so it is essential to employ a statistical method that can address this frequent characteristic of metagenomic data . Under the same assumptions as the simulation above , we tested a = 0 and 0 . 01 , thereby significantly reducing observations of a feature in one of the populations . The ROC curves presented in Figure 5 reveal that Metastats outperforms other statistical methods in the face of extreme sparseness . Holding the false positive rate ( x-axis ) constant , Metastats shows increased sensitivity over all other methods . The poor performance of Log-t is noteworthy given it is designed for SAGE data that is also potentially sparse . Further investigation revealed that the Log-t method results in a highly inflated dispersion value if there are no observations in one population , thereby reducing the estimated significance of the test . Finally , we selected a subset of the Dinsdale et al . [6] metagenomic subsystem data ( described below ) , and randomly assigned each subject to one of two populations ( 20 subjects per population ) . All subjects were actually from the same population ( microbial metagenomes ) , thus the null hypothesis is true for each feature tested ( no feature is differentially abundant ) . We ran each methodology on this data , recording computed p-values for each feature . Repeating this procedure 200 times , we simulated tests of 5200 null features . Table 1 displays the number of false positives incurred by each methodology given different p-value thresholds . The results indicate that the negative binomial model results in an exceptionally high number of false positives relative to the other methodologies . Student's t-test and Metastats perform equally well in estimating the significance of these null features , while Log-t performs slightly better . These studies show that Metastats consistently performs as well as all other applicable methodologies for deeply-sampled features , and outperforms these methodologies on sparse data . Below we further evaluate the performance of Metastats on several real metagenomic datasets . In a recent study , Ley et al . [35] identified gut microbes associated with obesity in humans and concluded that obesity has a microbial element , specifically that Firmicutes and Bacteroidetes are bacterial divisions differentially abundant between lean and obese humans . Obese subjects had a significantly higher relative abundance of Firmicutes and a lower relative abundance of Bacteriodetes than the lean subjects . Furthermore , obese subjects were placed on a calorie-restricted diet for one year , after which the subjects' gut microbiota more closely resembled that of the lean individuals . We obtained the 20 , 609 16S rRNA genes sequenced in Ley et al . and assigned them to taxa at different levels of resolution ( note that 2 , 261 of the 16S sequences came from a previous study [36] ) . We initially sought to re-establish the primary result from this paper using our methodology . Table 2 illustrates that our method agreed with the results of the original study: Firmicutes are significantly more abundant in obese subjects ( P = 0 . 003 ) and Bacteroidetes are significantly more abundant in the lean population ( P<0 . 001 ) . Furthermore , our method also detected Actinobacteria to be differentially abundant , a result not reported by the original study . Approximately 5% of the sample was composed of Actinobacteria in obese subjects and was significantly less frequent in lean subjects ( P = 0 . 004 ) . Collinsella and Eggerthella were the most prevalent Actinobacterial genera observed , both of which were overabundant in obese subjects . These organisms are known to ferment sugars into various fatty acids [37] , further strengthening a possible connection to obesity . Note that the original study used Students t-test , leading to a p-value for the observed difference within Actinobacteria of 0 . 037 , 9 times larger than our calculation . This highlights the sensitivity of our method and explains why this difference was not originally detected . To explore whether we could refine the broad conclusions of the initial study , we re-analyzed the data at more detailed taxonomic levels . We identified three classes of organisms that were differentially abundant: Clostridia ( P = 0 . 005 ) , Bacteroidetes ( P<0 . 001 ) , and Actinobacteria ( P = 0 . 003 ) . These three were the dominant members of the corresponding phyla ( Firmicutes , Bacteroides , Actinobacteria , respectively ) and followed the same distribution as observed at a coarser level . Metastats also detected nine differentially abundant genera accounting for more than 25% of the 16S sequences sampled in both populations ( P≤0 . 01 ) . Syntrophococcus , Ruminococcus , and Collinsella were all enriched in obese subjects , while Bacteroides on average were eight times more abundant in lean subjects . For taxa with several observations in each subject , we found good concordance between our results ( p-value estimates ) and those obtained with most of the other methods ( Table 2 ) . Surprisingly , we found that the negative binomial model of Robinson and Smyth failed to detect several strongly differentially abundant features in these datasets ( e . g the hypothesis test for Firmicutes results in a p-value of 0 . 87 ) . This may be due in part to difficulties in estimating the parameters of their model for our datasets and further strengthens the case for the design of methods specifically tuned to the characteristics of metagenomic data . For cases where a particular taxon had no observations in one population ( e . g . Terasakiella ) , the methods proposed for SAGE data seem to perform poorly . Targeted sequencing of the 16S rRNA can only provide an overview of the diversity within a microbial community but cannot provide any information about the functional roles of members of this community . Random shotgun sequencing of environments can provide a glimpse at the functional complexity encoded in the genes of organisms within the environment . One method for defining the functional capacity of an environment is to map shotgun sequences to homologous sequences with known function . This strategy was used by Kurokawa et al . [34] to identify clusters of orthologous groups ( COGs ) in the gut microbiomes of 13 individuals , including four unweaned infants . We examined the COGs determined by this study across all subjects and used Metastats to discover differentially abundant COGs between infants and mature ( >1 year old ) gut microbiomes . This is the first direct comparison of these two populations as the original study only compared each population to a reference database to find enriched gene sets . Due to the high number of features ( 3868 COGs ) tested for this dataset and the limited number of infant subjects available , our method used the pooling option to compute p-values ( we chose 100 permutations ) , and subsequently computed q-values for each feature . Using a threshold of Q≤0 . 05 ( controlling the false discovery rate to 5% ) , we detected 192 COGs that were differentially abundant between these two populations ( see Table 3 for a lisitng of the most abundant COGs in both mature and infant microbiomes . Full results are presented as supplementary material in Table S1 ) . The most abundant enriched COGs in mature subjects included signal transduction histidine kinase ( COG0642 ) , outer membrane receptor proteins , such as Fe transport ( COG1629 ) , and Beta-galactosidase/beta-glucuronidase ( COG3250 ) . These COGs were also quite abundant in infants , but depleted relative to mature subjects . Infants maintained enriched COGs related to sugar transport systems ( COG1129 ) and transcriptional regulation ( COG1475 ) . This over-abundance of sugar transport functions was also found in the original study , strengthening the hypothesis that the unweaned infant gut microbiome is specifically designed for the digestion of simple sugars found in breast milk . Similarly , the depletion of Fe transport proteins in infants may be associated with the low concentration of iron in breast milk relative to cow's milk [38] . Despite this low concentration , infant absorption of iron from breast milk is remarkably high , and becomes poorer when infants are weaned , indicating an alternative mechanism for uptake of this mineral . The potential for a different mechanism is supported by the detection of a Ferredoxin-like protein ( COG2440 ) that was 11 times more abundant in infants than in mature subjects , while Ferredoxin ( COG1145 ) was significantly enriched in mature subjects . A recent study by Dinsdale et al . profiled 87 different metagenomic shotgun samples ( ∼15 million sequences ) using the SEED platform ( http://www . theseed . org ) [6] to see if biogeochemical conditions correlate with metagenome characteristics . We obtained functional profiles from 45 microbial and 40 viral metagenomes analyzed in this study . Within the 26 subsystems ( abstract functional roles ) analyzed in the Dinsdale et al . study , we found 13 to be significantly different ( P≤0 . 05 ) between the microbial and viral samples ( Table 4 ) . Subsystems for RNA and DNA metabolism were significantly more abundant in viral metagenomes , while nitrogen metabolism , membrane transport , and carbohydrates were all enriched in microbial communities . The high levels of RNA and DNA metabolism in viral metagenomes illustrate their need for a self-sufficient source of nucleotides . Though the differences described by the original study did not include estimates of significance , our results largely agreed with the authors' qualitative conclusions . However , due to the continuously updated annotations in the SEED database since the initial publication , we found several differences between our results and those originally reported . In particular we found virulence subsystems to be less abundant overall than previously reported , and could not find any significant differences in their abundance between the microbial and viral metagenomes .
We have presented a statistical method for handling frequency data to detect differentially abundant features between two populations . This method can be applied to the analysis of any count data generated through molecular methods , including random shotgun sequencing of environmental samples , targeted sequencing of specific genes in a metagenomic sample , digital gene expression surveys ( e . g . SAGE [29] ) , or even whole-genome shotgun data ( e . g . comparing the depth of sequencing coverage across assembled genes ) . Comparisons on both simulated and real dataset indicate that the performance of our software is comparable to other statistical approaches when applied to well- sampled datasets , and outperforms these methods on sparse data . Our method can also be generalized to experiments with more than two populations by substituting the t-test with a one-way ANOVA test . Furthermore , if only a single sample from each treatment is available , a chi-squared test can be used instead of the t-test . [27] . In the coming years metagenomic studies will increasingly be applied in a clinical setting , requiring new algorithms and software tools to be developed that can exploit data from hundreds to thousands of patients . The methods described above represent an initial step in this direction by providing a robust and rigorous statistical method for identifying organisms and other features whose differential abundance correlates with disease . These methods , associated source code , and a web interface to our tools are freely available at http://metastats . cbcb . umd . edu .
|
The emerging field of metagenomics aims to understand the structure and function of microbial communities solely through DNA analysis . Current metagenomics studies comparing communities resemble large-scale clinical trials with multiple subjects from two general populations ( e . g . sick and healthy ) . To improve analyses of this type of experimental data , we developed a statistical methodology for detecting differentially abundant features between microbial communities , that is , features that are enriched or depleted in one population versus another . We show our methods are applicable to various metagenomic data ranging from taxonomic information to functional annotations . We also provide an assessment of taxonomic differences in gut microbiota between lean and obese humans , as well as differences between the functional capacities of mature and infant gut microbiomes , and those of microbial and viral metagenomes . Our methods are the first to statistically address differential abundance in comparative metagenomics studies with multiple subjects , and we hope will give researchers a more complete picture of how exactly two environments differ .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] |
[
"computational",
"biology/metagenomics"
] |
2009
|
Statistical Methods for Detecting Differentially Abundant Features in Clinical Metagenomic Samples
|
Loss-of-function mutations in PINK1 and Parkin cause parkinsonism in humans and mitochondrial dysfunction in model organisms . Parkin is selectively recruited from the cytosol to damaged mitochondria to trigger their autophagy . How Parkin recognizes damaged mitochondria , however , is unknown . Here , we show that expression of PINK1 on individual mitochondria is regulated by voltage-dependent proteolysis to maintain low levels of PINK1 on healthy , polarized mitochondria , while facilitating the rapid accumulation of PINK1 on mitochondria that sustain damage . PINK1 accumulation on mitochondria is both necessary and sufficient for Parkin recruitment to mitochondria , and disease-causing mutations in PINK1 and Parkin disrupt Parkin recruitment and Parkin-induced mitophagy at distinct steps . These findings provide a biochemical explanation for the genetic epistasis between PINK1 and Parkin in Drosophila melanogaster . In addition , they support a novel model for the negative selection of damaged mitochondria , in which PINK1 signals mitochondrial dysfunction to Parkin , and Parkin promotes their elimination .
Parkinson disease is a common neurodegenerative disorder with no disease-modifying therapy presently available for its treatment [1] . Study of recessive forms of familial Parkinson disease , such as those resulting from mutations in the E3 ubiquitin ligase Parkin ( GeneID: 5071 ) or the mitochondrial kinase PINK1 ( GeneID: 65018 ) , may reveal disease mechanisms important to the development of disease in these families as well as those suffering from sporadic Parkinson disease . Although the cause of sporadic Parkinson disease is likely complex , several lines of evidence link mitochondrial dysfunction to its pathogenesis . Mitochondria within the substantia nigra ( SN ) , a midbrain region that is preferentially affected in Parkinson disease , have a higher somatic mitochondrial DNA ( mtDNA ) mutation rate than all other regions of the brain examined [2] . Increased mitochondrial damage in the SN , particularly to mtDNA , has been associated with sporadic Parkinson disease [3]–[5] , and mitochondrial dysfunction is sufficient to cause parkinsonism in patients with rare multiple mtDNA deletion syndromes and in animal models with decreased mtDNA expression [6]–[8] . In addition , toxins such as MPTP and rotenone , which are believed to increase reactive oxygen species from complex I of the electron transport chain , can induce a parkinsonian syndrome in humans and animal models [9] , [10] . Since neurons in the SN are postmitotic , any mitochondrial damage they acquire could accumulate over an organism's lifetime , leading to progressive mitochondrial dysfunction—including increased oxidative stress , decreased calcium buffering capacity , loss of ATP , and , eventually , cell death—unless quality control processes eliminate the damaged mitochondria . Recent studies have linked Parkin and PINK1 in a pathway critical for the maintenance of mitochondrial integrity and function . Loss of either protein in Drosophila results in a similar phenotype , with mitochondrial damage preceding muscle degeneration , as well as disrupted spermatogenesis and death of dopaminergic neurons [11]–[15] . Interestingly , overexpression of Parkin can partially compensate for PINK1 loss , but PINK1 overexpression cannot compensate for Parkin loss , suggesting that PINK1 functions upstream of Parkin in a common pathway . Additionally , mice null for either Parkin or PINK1 exhibit increased oxidative damage and decreased mitochondrial function in the striatium ( which receives projections from dopaminergic neurons ) [16] , [17]; and primary cells from patients with loss-of-function mutations in Parkin or PINK1 have similar abnormalities [18]–[20] . Together these findings suggest that Parkin and PINK1 may function in an evolutionarily conserved pathway critical for the maintenance of mitochondrial integrity and function . We recently reported that Parkin is selectively recruited to dysfunctional mitochondria with low membrane potential and , subsequently , promotes their autophagic degradation [21] . This suggests that Parkin may limit mitochondrial damage by acting in a pathway that identifies and eliminates damaged mitochondria from the mitochondrial network . How mitochondrial dysfunction is signaled to Parkin , however , is unknown . Here , we show that full-length PINK1 accumulates selectively on dysfunctional mitochondria , and that Parkin recruitment to depolarized mitochondria and subsequent Parkin-induced mitophagy are strictly dependent on PINK1's mitochondrial targeting signal and depolarization-induced accumulation . Together , these results strongly support a novel model for signaling between PINK1 and Parkin in response to mitochondrial damage . In this model , mitochondrial PINK1 is rapidly turned over on bioenergetically well-coupled mitochondria by proteolysis , but is selectively stabilized on mitochondria with low membrane potential . Selective accumulation of PINK1 on the impaired mitochondria recruits Parkin , and Parkin , in turn , induces the degradation of the damaged mitochondria . In this model , PINK1 and Parkin form a pathway for sensing and selectively eliminating damaged mitochondria from the mitochondrial network . Disease-causing mutations in PINK1 and/or Parkin disrupt this pathway at distinct steps , consistent with the pathway's importance for preventing early-onset parkinsonism .
Parkin is selectively recruited to damaged mitochondria that have lost their membrane potential , but how Parkin distinguishes dysfunctional mitochondria with low membrane potential from healthy mitochondria is unknown . Since PINK1 is genetically upstream of Parkin , we tested whether PINK1's activity might be activated by mitochondrial depolarization . Remarkably , levels of endogenous mitochondrial PINK1 respond robustly to changes in mitochondrial membrane potential . When HeLa cells are treated with CCCP , which depolarizes mitochondria by increasing membrane permeability to H+ , a large increase in endogenous full-length PINK1 ( ∼63 kDa ) is seen beginning by 30 min and continuing for at least 3 h ( Figure 1A ) . This ∼63-kDa band also increases in the mitochondria-rich membrane fraction following treatment with valinomycin , which , unlike CCCP , depolarizes mitochondria by permeabilizing the membrane to K+ ( Figure S1A ) . By contrast , no band increases in the cytosolic fraction following depolarization with CCCP ( Figure S1B ) . To verify that the ∼63-kDa band is in fact PINK1 , we immunoblotted for endogenous PINK1 in M17 cells stably transduced with control short hairpin RNA ( shRNA ) or PINK shRNA . We found that the ∼63-kDa band increases following CCCP treatment in control shRNA cells , but does not increase in the PINK1 shRNA cells , demonstrating that this ∼63-kDa band is endogenous PINK1 ( Figure 1B ) . Similar results were found in PINK1−/− cells transfected with PINK1-myc or left untransfected ( Figure S1C ) . We also tested whether PINK1 similarly accumulates in primary rat cortical neurons following depolarization with CCCP . Although we ( and others ) failed to detect endogenous rat or mouse PINK1 with the available commercial antibodies ( [22] and unpublished data ) , we observed PINK1-V5 increases in cortical neurons following treatment with 1 µM of CCCP for 6 h ( Figure 1C ) . With CCCP treatment , PINK1 may accumulate more slowly in primary neurons than in HeLa cells , because , unlike HeLa cells [23] , neurons rely almost exclusively on oxidative phosphorylation for ATP production [24] . To explore the kinetics of PINK1 accumulation at the single-cell level , we fused YFP to PINK1 and imaged cells live following depolarization with CCCP . Consistent with results obtained by Western blotting , we found that PINK1-YFP expression steadily increases from 1–5 min , when an increase is first detectable , until at least 70 min ( Figure 1D and Video S1 ) . To examine the selectivity of PINK1 accumulation on uncoupled mitochondria within single cells , we first investigated its expression in mouse embryonic fibroblasts ( MEFs ) null for mitochondrial fusion proteins mitofusin-1 and mitofusin-2 ( Mfn1/2 ) . The Mfn1/2 null MEFs have a heterogeneous population of mitochondria , some of which are bioenergetically uncoupled and some of which are well coupled [25] . We found that , similar to YFP-Parkin [21] , PINK1-YFP accumulates selectively on mitochondria with low membrane potential , demonstrating that PINK1 is selectively stabilized on the depolarized mitochondria within a bioenergetically diverse population of mitochondria ( Figure 1E and 1F ) . Treatment with paraquat , a pesticide that has been linked to Parkinsonism , also results in a heterogeneous population of mitochondria , likely due to stochastic damage of mitochondria by reactive oxygen species [26] . We treated HeLa cells overnight with a high dose of paraquat ( 2 mM ) . Similar to results with Parkin reported previously [21] , we found that PINK1-YFP accumulates preferentially on damaged mitochondria with low membrane potential ( Figure 1G ) . Although PINK1-YFP colocalizes with cytochrome c , which is present in all mitochondria ( average Pearson coefficient = 0 . 58±0 . 11 ) , PINK1-YFP does not colocalize with MTR ( average Pearson coefficient = 0 . 26±0 . 13 ) , which accumulates only in bioenergetically active mitochondria ( p-value <0 . 001 for PINK1/cytochrome c vs . PINK1/MTR , paired Student t-test ) . These data suggest that PINK1-YFP accumulates selectively on depolarized mitochondria that have been damaged by oxidative stress ( Figure 1H ) . Next , we examined whether Parkin is recruited to the same depolarized mitochondria that accumulate PINK1 following treatment with paraquat . This relationship is difficult to test directly , because overexpression of PINK1 appears to accelerate the kinetics of Parkin recruitment to mitochondria ( as shown later ) [27] , and so we used a kinase-deficient version of PINK1 ( PINK1 KD ) [28] , as a reporter for endogenous PINK1 accumulation in the HeLa cells . We find PINK1 KD expression is regulated by mitochondrial voltage similarly to wild-type PINK1 ( Figure S1D ) ; but unlike wild-type PINK1 , PINK1 KD does not enhance Parkin recruitment when overexpressed ( as shown later ) . After treatment with paraquat overnight , we find that PINK1 KD accumulates selectively on depolarized mitochondria , in a pattern similar to wild-type PINK1 ( Figure 1I ) . In addition , a substantial subset of the mitochondria that accumulated PINK1 KD ( and so likely accumulated endogenous PINK1 , as well ) recruited Parkin ( Figure 1I ) . Although Parkin and PINK1 KD colocalize in paraquat-treated cells ( average Pearson coefficient = 0 . 45±0 . 13 ) , PINK1KD does not colocalize with MTR ( average Pearson coefficient = 0 . 22±0 . 13; p-value = 0 . 002 for PINK1 KD/Parkin vs . PINK1 KD/MTR ) ( Figure 1J ) . Considered together , these results demonstrate that PINK1 selectively accumulates on dysfunctional mitochondria with low membrane potentials . Regulation of PINK1 expression at the level of transcription or translation would likely not be selective for a subpopulation of mitochondria , and so we assessed whether increased PINK1 expression on damaged mitochondria is achieved by the selective removal of PINK1 from functional mitochondria . Full-length PINK1 ( ∼63 kDa ) , which is anchored in the mitochondrial membrane , is proteolytically cleaved into an ∼52-kDa cytosolic fragment that can be degraded by the proteasome [22] , [28]–[30] . To test whether PINK1 accumulation following CCCP treatment is due to inhibition of its proteolytic cleavage , we assessed the effect of CCCP washout on PINK1 cleavage . HeLa cells were treated with vehicle ( DMSO ) or CCCP for 3 h , after which CCCP was either washed out or left in for an additional 30 min . Cycloheximide was either added or left out during the final hour of treatment to control for de novo PINK1 synthesis during the washout period . Following PINK1 accumulation in the continuous presence of CCCP for 3 h , the addition of cycloheximide for 1 h has little effect on the abundance of full-length PINK1 , suggesting that once it has accumulated , the ∼63-kDa PINK1 is relatively stable on depolarized mitochondria ( Figure 2A , lanes 4 vs . lane 6 ) . However , within 30 min of CCCP washout , ∼63-kDa PINK1 abundance falls dramatically , consistent with its being cleaved and maintained at low abundance on polarized , undamaged mitochondria ( Figure 2A , lanes 4–7 vs . lanes 8–11 ) . The residual full-length PINK1 seen following CCCP washout largely represents PINK1 that had accumulated during the 3-h CCCP treatment , as the addition of cycloheximide prior to washout has little effect on its level ( Figure 2A , lane 8 vs . lane 10 ) . To further assess the stability of PINK1 under depolarizing conditions , we performed the same set of experiments in the presence of MG132 , an inhibitor of proteasomal degradation . When MG132 is added during the final hour of treatment in HeLa cells treated with vehicle , an ∼52-kDa band appears , consistent with the cleavage product of full-length Parkin described in previous reports [22] , [28] , [30] ( Figure 2A , lane 1 vs . lane 2 ) . The accumulation of this short form of PINK1 following treatment with MG132 suggests that it is unstable under basal conditions , as has been observed previously [22] , [30] ( Figure 2A , lane 1 vs . lane 2 ) . Interestingly , levels of the short form of PINK1 in the presence of MG132 decrease following depolarization with CCCP for 3 . 5 hrs , as levels of full-length PINK1 rise ( Figure 2A , lane 2 vs . lane 5 ) ; but increase following CCCP washout , as levels of full-length PINK1 fall ( Figure 2A , lane 2 vs . lane 9 ) . This pattern indicates that the cleavage of full-length PINK1 into the unstable short form is blocked by mitochondrial depolarization and reinstated upon CCCP washout . Taken together , these results support a two-step model for the processing of PINK1: first , full-length PINK1 is cleaved into the ∼52-kDa short form in a voltage-dependent , proteasome-independent manner , and second , the short form of PINK1 is rapidly degraded by the proteasome ( Figure 2B ) . The voltage-dependent processing of PINK1 maintains low levels of PINK1 on healthy polarized mitochondria , but allows for the rapid accumulation of PINK1 on depolarized mitochondria . Although these experiments suggest that the increased expression of PINK1 is due at least in part to inhibition of PINK1 cleavage , it is possible that increased transcription of PINK1 following depolarization might also be contributing to the increase in PINK1 abundance . To assess whether PINK1 transcription is also regulated by membrane potential , we performed quantitative RT-PCR ( qRT-PCR ) of PINK1 levels in HeLa cells treated with DMSO or CCCP for 1 h . We found that whereas exogenous expression of PINK1 causes a significant increase in PINK1 transcription relative to untransfected HeLa cells , PINK1 transcription does not significantly increase following depolarization with CCCP ( p = 0 . 4499 ) . These data confirm that the increase in PINK1 expression following depolarization is not driven by an increase PINK1 transcription ( Figure 2C ) . Finally , to test the localization of accumulated PINK1 on depolarized mitochondria , we performed a protease protection assay , using an antibody raised against PINK1's kinase domain . Consistent with results from a recent study of PINK1's topology when it is ectopically expressed [22] , we found that the kinase domain of endogenous PINK1 faces the cytosol following depolarization ( Figure S1E ) . The protease responsible for PINK1 cleavage in mammalian cells is unknown , but in Drosophila cells , the intramembrane serine protease , Rhomboid-7 , appears to be required for PINK1 cleavage [31] . To examine whether the mammalian ortholog of Rhomboid-7 , PARL , is responsible for PINK1 cleavage in mammalian cells , we tested whether PINK1-V5 accumulates in HeLa cells transfected with PARL shRNA and treated with CCCP . Although endogenous PARL could not be detected in HeLa cells , PARL shRNA inhibited expression of overexpressed PARL ( Figure S2A and S2B ) . Knockdown of PARL did not appreciably change basal levels of endogenous PINK1 or augment the depolarization-induced accumulation of endogenous PINK1 in HeLa cells ( Figure S2B ) . Likewise , PINK1-V5 levels were similar in PARL−/− and PARL+/+ MEFs , under basal conditions and following depolarization with CCCP ( Figure S2C ) . Together , these results suggest that PARL is dispensable for PINK1 cleavage . Previous studies in Drosophila and mammalian cells indicate that PINK1 functions genetically upstream of Parkin [11]–[13] , [20] , [32] , although the molecular mechanism of this genetic interaction remains unexplained . To test whether PINK1 accumulation on mitochondria is upstream of Parkin recruitment to depolarized mitochondria , we assessed the dependence of PINK1 accumulation on Parkin expression . Endogenous PINK1 accumulates similarly in HeLa cells , which display little or no endogenous Parkin expression , and HeLa cells stably expressing YFP-Parkin ( Figure S2D ) . Consistent with these findings , we observed that exogenous PINK1-myc accumulates similarly in immortalized Parkin−/− and Parkin+/+ MEFs ( Figure S2E ) . Together , these results show that PINK1 accumulation is upstream of Parkin recruitment to depolarized mitochondria and independent of Parkin expression . Next , we tested whether Parkin recruitment to depolarized mitochondria is dependent on PINK1 expression . We found that , although YFP-Parkin is recruited to mitochondria in 43 . 3±8 . 1% ( mean ± standard deviation [SD] ) of PINK1+/+ primary MEFs after 3-h exposure to 20 µM CCCP , it is not detectably recruited to mitochondria in PINK1−/− MEFs , as assessed by confocal microscopy ( Figure 3A and 3B ) . We also failed to detect YFP-Parkin recruitment at 24 h following CCCP in PINK1−/− MEFs ( unpublished data ) , suggesting that little or no recruitment of YFP-Parkin to depolarized mitochondria occurs in the absence of PINK1 . YFP-Parkin recruitment could be reconstituted in PINK1−/− MEFs by expression of wild-type PINK1 , but not by PINK1 ΔN lacking its mitochondrial targeting N-terminus ( 1–155 ) [22] , suggesting that mitochondrial targeting of PINK1 is required for Parkin recruitment to mitochondria ( Figure 3A and 3B ) . A kinase-deficient ( KD ) version of PINK1 [28] also failed to reconstitute Parkin recruitment to mitochondria ( Figure 3A and 3B ) . We further tested the dependence of Parkin recruitment on PINK1 in a SV40-transformed MEF cell line , which was derived from an independently generated PINK1−/− mouse [29] ( Figure 3C and Figure S3A ) . Similar to the primary PINK1−/− MEFs , no recruitment is seen in the transformed PINK1−/− cells , whereas Parkin is recruited to mitochondria in 60 . 7±7 . 7% of PINK1+/+ cells upon CCCP treatment . Likewise , Parkin recruitment in the transformed PINK1−/− cells is reconstituted following exogenous expression of PINK1 ( 72 . 8±7 . 7% vs . 0 . 0±0 . 0% , p-value <0 . 001 ) , but not PINK1 ΔN or PINK1 KD . Finally , we tested the dependence of Parkin recruitment in a human neuroblastoma cell line ( M17 ) [33] , [34] . In M17 cells stably transduced with PINK1 shRNA , YFP-Parkin translocated to mitochondria in 4 . 7±1 . 2% of CCCP-treated cells , whereas 67 . 3±3 . 1% of control shRNA M17 cells displayed mitochondrial YFP-Parkin after treatment with 10 µM CCCP for 3 h ( p-value <0 . 001 ) ( Figure 3D and 3E ) . Vehicle treatment failed to induce YFP-Parkin translocation to mitochondria in both cell lines . The necessity of PINK1 expression for Parkin recruitment to membranes was also examined in the M17 cell line by immunoblotting . In control shRNA cells , YFP-Parkin levels increase in the mitochondria-rich membrane fraction and decrease in the supernatant following treatment with CCCP , consistent with Parkin translocation to mitochondria ( Figure 3F , upper panel , and Figure S3B ) . YFP-Parkin was expressed less in the PINK1 shRNA cells compared to control shRNA cells , possibly because the transfection efficiency is lower in these cells and/or because Parkin is less stable in the absence of PINK1 , as has been observed previously [29] . Nonetheless , we failed to see Parkin increase in the membrane fraction either under equal loading conditions or when loading was adjusted so that total Parkin was approximately equal in the two cell populations , further indicating that Parkin is not recruited to uncoupled mitochondria in the absence of PINK1 ( Figure 3F , lower panel , and Figure S3B ) . We previously reported that ectopic Parkin can induce the autophagy of depolarized mitochondria [21] . To test whether PINK1 is necessary for Parkin-induced mitophagy , we treated primary PINK1−/− and PINK1+/+ MEFs transiently expressing YFP-Parkin with 20 µM CCCP for 24 h ( Figure 4A and 4B ) . Whereas no mitochondria can be detected in 66 . 1±16 . 8% of PINK1+/+ MEFs , all PINK−/− MEFs retain their mitochondria . Parkin-dependent mitophagy is reconstituted by exogenous PINK1 expression in the PINK−/− MEFs , with 65 . 5±5 . 0% of reconstituted PINK−/− cells displaying undetectable mitochondria following CCCP treatment . We found that Parkin-induced mitophagy is also dependent on PINK1 expression in the M17 human neuroblastoma cell line . Whereas in 27 . 1±8 . 6% of control shRNA M17 cells displayed complete loss of mitochondria after 24 h , less than 5% of cells lost mitochondria in the PINK1 shRNA cells ( Figure 4C and 4D ) . These results suggest that PINK1 is necessary for the mitophagy of depolarized mitochondria following overexpression of Parkin . To test whether PINK1 expression affects mitochondrial turnover in the presence of endogenous levels of Parkin , we treated the control shRNA and PINK1 shRNA M17 cells ( which express moderate levels of Parkin ) with DMSO or CCCP for 24 h and measured their relative mitochondrial mass by Mitotracker Green ( MTG ) staining and flow cytometry . MTG , a sensitive measure of mitochondrial mass , stains mitochondrial lipid in a membrane potential–independent manner and has been used to measure mitochondrial mass of depolarized mitochondria previously [35] , [36] . Control shRNA M17 cells exhibit a decrease in mitochondrial mass ( CCCP vs . DMSO , −22 . 4±12 . 6% ) following CCCP treatment , whereas PINK1 shRNA M17 cells exhibit an increase in mitochondrial mass ( CCCP vs . DMSO , 43 . 5±20 . 0% ) following depolarization ( p-value = 0 . 008 for change in mitochondrial mass control shRNA vs . PINK shRNA ) ( Figure 4E ) . These results are consistent with endogenous PINK1 promoting mitochondrial degradation in the context of continued ( or increased ) mitochondrial biogenesis [37] , [38] . To more directly assay mitochondrial turnover in control and PINK1 shRNA M17 cells , we pulsed the cells with MTG and tracked loss of MTG intensity at 0 , 16 , and 24 h in the presence of CCCP . Consistent with the hypothesis that endogenous PINK1 promotes the degradation of depolarized mitochondria , MTG intensity decreases more slowly in PINK1 shRNA cells when compared with control shRNA cells , treated with CCCP ( 0 . 58±0 . 07 vs . 0 . 33±0 . 07 relative MTG intensity at 24 h ) ( Figure 4F ) . These data suggest that PINK1 promotes mitophagy in the context of endogenous levels of Parkin . Additionally , these results suggest that the selective turnover of dysfunctional mitochondria may be balanced by the biogenesis of new mitochondria , allowing exchange of damaged , dysfunctional mitochondria for healthy , functional mitochondria . Consistent with genetic studies in Drosophila , these findings show that Parkin translocation to depolarized mitochondria and Parkin-induced mitophagy are downstream of PINK1 expression , whereas PINK1 accumulation in response to depolarization is upstream of Parkin recruitment . The expression of mitochondrial PINK1 is necessary for recruitment of Parkin to mitochondria . Next , we tested whether PINK1 overexpression is sufficient for Parkin recruitment to mitochondria . Using live-cell imaging , we found that moderate overexpression of PINK1 dramatically accelerates the kinetics of Parkin recruitment following depolarization with CCCP ( time to translocation 5 . 0±1 . 5 min vs . 32 . 0±5 . 4 min , p-value <0 . 001 ) ( Figure 5A and 5B; Video S2 vs . S3 ) . Consistent with necessity of PINK1's mitochondrial localization and kinase activity , exogenous expression of PINK1 KD or PINK1 ΔN fails to accelerate the kinetics of Parkin recruitment ( Figure 5B ) . In cells with high expression of PINK1 , Parkin is recruited to mitochondria even in the absence of CCCP ( Figure 5C and 5D ) , as has been reported previously [27] . YFP-Parkin colocalizes with the potentiometric mitochondrial dye TMRE in 45 . 3±7 . 6% of cells coexpressing YFP-Parkin and PINK1 vs . 0±0% of cells expressing Parkin alone ( p-value <0 . 001 ) ( Figure 5C and 5D ) . Together , these results demonstrate that the kinetics of Parkin recruitment is exquisitely sensitive to PINK1 levels in the cell . In addition , they indicate that increased PINK1 expression is sufficient for Parkin recruitment independent of membrane potential . To test whether stable expression of PINK1 on the mitochondria is sufficient for Parkin recruitment , we constructed a fusion protein that would be predicted to lack PINK1's proteolytic cleavage site and therefore exhibit greater stability on mitochondria . Based on the ∼11-kDa difference between the full-length form and the cleaved form , the cleavage site likely lies before residue 110 ( residues 1–110 have a predicted molecular weight of 11 . 54 kDa ) , and so we replaced residues 1–110 of PINK1 with the outer mitochondrial membrane anchor from OPA3 ( 1–30 ) ( Figure 6A ) . We found that removing the first 110 amino acids of PINK1 prevents targeting of PINK1 to mitochondria , consistent with previous reports [39] ( Figure 6B , middle panel ) ; whereas the fusion of OPA3 ( 1–30 ) to PINK1 Δ1-110 restores mitochondrial targeting , likely to the outer mitochondrial membrane ( Figure 6B , right panel ) . As predicted by the proteolytic cleavage results ( Figure 2A ) , OPA3-PINK1 Δ1-110-YFP exhibits increased stability compared to PINK1-YFP ( Figure 6C ) . In addition , OPA3-PINK1 Δ1-110-YFP levels do not respond to mitochondrial depolarization with CCCP , indicating that stabilization of PINK1 by depolarization depends on its first 110 amino acids . When coexpressed with mCherry-Parkin , PINK1-YFP recruits mCherry-Parkin to mitochondria in 57 . 9±1 . 8% of cells in the absence of CCCP; whereas PINK1 Δ1-110-YFP , which is not expressed on mitochondria , fails to recruit mCherry-Parkin in the absence of CCCP . However , OPA3-PINK1 Δ1-110-YFP , which does not display voltage dependent proteolysis , recruits mCherry-Parkin to mitochondria in 98±1 . 8% of cells in the absence of CCCP ( Figure 6D and 6E ) . Together , these data demonstrate that stable expression of PINK1 on mitochondria is sufficient for Parkin recruitment to mitochondria , regardless of membrane potential . To verify that increased expression of PINK1 on the outer mitochondrial membrane is sufficient to induce Parkin recruitment , we used a regulated heterodimerization system [40] , in which the modified FRB domain was fused to PINK1 Δ1-110-YFP and the FKBP domain was fused to the outer mitochondrial membrane anchor of TOM20 ( residues 1 through 33 ) ( Figure S4A ) . In the presence of the rapamycin derivative AP21967 , the FRB domain and the FKBP domain heterodimerize , but only if they are in the same compartment . Thus , FRB-PINK1Δ1-110-YFP should be recruited from the cytosol to mitochondria if the FKBP domain of TOM20-FKBP faces the cytosol but not if it faces the inter membrane space or the matrix . We found that FRB-PINK1Δ1-110-YFP is in the cytosol in the absence of AP21967 , but is quickly recruited to the outer mitochondrial membrane following the addition of AP21967 ( Figure S4B and S4C ) . To assess whether PINK1 expression on the outer mitochondrial membrane is sufficient to recruit Parkin , we cotransfected mCherry-Parkin , FRB-PINK1Δ1-110-YFP , and TOM20-FKBP . In the absence of AP21967 , mCherry-Parkin is in the cytosol , but following incubation with AP21967 mCherry-Parkin is recruited to mitochondria in 96 . 7±4 . 1% of cells , in the absence of CCCP ( Figure 6F and Figure S4D ) . Thus , increased expression of PINK1 on the outer mitochondrial membrane is sufficient to recruit Parkin to mitochondria . Next , we tested whether Parkin recruitment following increased PINK1 expression on the mitochondria is sufficient to induce mitophagy in the absence of depolarization with CCCP . Cotransfection of PINK1 and Parkin results in a substantial proportion of cells ( 42 . 1±7 . 3% ) with no mitochondria after 96 h . By contrast , cotransfection of cytosolic PINK1 Δ1-110-YFP with Parkin produces no cells lacking mitochondria after 96 h . Fusing the outer membrane anchor of OPA3 to PINK1 Δ1-110-YFP , which results in stable expression of PINK1 on mitochondria , restores the ability of PINK1 and Parkin to induce mitophagy , with 76 . 4±2 . 2% of cells lacking mitochondria at 96 h ( Figure 6G and 6H ) . These data demonstrate that stable expression of PINK1 on the outer mitochondrial membrane is sufficient to induce mitophagy in the presence of Parkin , irrespective of membrane potential . To test whether accumulation of endogenous PINK1 following depolarization is necessary for Parkin recruitment , we treated HeLa cells with CCCP alone ( for 60 min ) or with CCCP plus cycloheximide , a general inhibitor of protein synthesis ( cycloheximide was added 30 min before CCCP and maintained throughout the 60-min CCCP treatment ) . Treatment of HeLa cells for 90 min with cycloheximide blocked the depolarization-induced accumulation of endogenous PINK1 in whole-cell lysates as well as in the mitochondria-rich membrane fraction ( Figure 7A and 7B ) . A 90-min treatment with cycloheximide , likewise , blocked Parkin recruitment to depolarized mitochondria by confocal microscopy ( 96 . 0±3 . 5% vs . 11 . 3±4 . 2% ) ( Figure 7C and 7D ) . By contrast , 90-min treatment with actinomycin D , an inhibitor of transcription , had a modest effect on Parkin recruitment to uncoupled mitochondria by confocal microscopy ( Figure 7C and 7D ) , suggesting that new transcription of PINK1 is not required for Parkin recruitment . This is consistent with the absence of PINK1 mRNA up-regulation following uncoupling ( Figure 2C ) . Cycloheximide likewise blocked YFP-Parkin accumulation in the mitochondria-enriched heavy membrane fraction by immunoblotting ( Figure 7E ) . Although these findings do not prove new PINK1 synthesis is required for Parkin recruitment ( cycloheximide blocks de novo synthesis of all proteins and thus may inhibit Parkin recruitment independently of PINK1 accumulation ) , they suggest that PINK1 accumulation and Parkin recruitment may be casually related . It has been proposed that PINK1 may induce mitochondrial recruitment of Parkin through phosphorylation of threonines 175 and 217 in a highly conserved region/domain of Parkin , which has been recently named RING0 ( Figure S5A ) [27] , [41] . We found that although mutation of T175 and T217 to alanine blocked recruitment of Parkin to mitochondria , as was reported previously , the phosphomimetic mutants T175E , T217E , and T175 , 217E do not translocate to mitochondria spontaneously . In addition , these phosphomimetic mutants appear to inhibit CCCP-induced recruitment of Parkin . Although these findings do not rule out the possibility that phosphorylation of these sites by PINK1 or another kinase induces Parkin recruitment , they suggest that these threonines are more likely to play an important structural role ( Figure S5B and S5C ) . We assessed the ability of disease-causing mutations in PINK1 to reconstitute YFP-Parkin recruitment to mitochondria in PINK1−/− primary MEFs . Following exogenous PINK1 WT expression in PINK1−/− MEFs , YFP-Parkin was recruited to mitochondria in 78 . 6±3 . 9% of cells after 20 µM CCCP treatment for 3 h ( Figure 8A and 8B ) . We found that the L347P patient mutant of PINK1 is unstable ( Figure 8C and Figure S6A ) , as was reported previously [28] , and that L347P failed to reconstitute YFP-Parkin recruitment to depolarized mitochondria ( Figure 8A and 8B ) . Of the patient mutations that exhibited stable expression , A168P and H271Q also failed to reconstitute YFP-Parkin recruitment at 3 h , whereas G309D only partially reconstituted YFP-Parkin recruitment ( 30 . 7±16 . 7% ) ( Figure 6A and 6B ) . The polymorphism G411S , which to date has only been found in cases heterozygous for the mutation [42] , reconstituted YFP-Parkin recruitment to a similar extent as wild-type PINK1 ( 74 . 2±5 . 4% ) , suggesting that PINK1 containing this polymorphism may be functional in the PINK1/Parkin pathway ( Figure 8A and 8B ) . This is consistent with the idea that G411S may represent a natural variant and may not be a true disease-causing mutation . Protein levels of all PINK1 mutants accumulated upon exposure of cells to CCCP ( Figure 8C and Figure S6A ) . Next , we tested patient mutations in Parkin to see if they would affect Parkin recruitment to mitochondria and/or Parkin-induced mitophagy . Parkin has an N-terminal ubiquitin-like domain ( UBL ) and a C-terminal RING-between-RING ( RBR ) superdomain , which consists of three atypical RING domains ( Figure 9A ) . The fold of the N-terminal RING1 most closely resembles that of traditional RING domains , such as that of c-CBL , whereas the In-Between-RING ( IBR ) and the C-terminal RING2 likely have unique folds [43]–[45] . The RBR domain is responsible for Parkin's ubiquitin ligase activity , whereas its UBL domain is thought to mediate interactions between Parkin and proteins with ubiquitin-binding domains ( UBDs ) [46] . As was reported previously , wild-type YFP-Parkin is recruited to mitochondria in the majority of HeLa cells ( 94 . 7±5 . 8% ) by confocal microscopy , following treatment with 10 µM CCCP for 1 h ( Figure 9A and 9B ) . Pathogenic mutations in the UBL domain ( R42P and R46P ) , deletion of the UBL , or mutation of a key residue ( I44A ) in the interaction of UBLs with UBDs [47] , all cause a moderate deficit in Parkin recruitment to depolarized mitochondria ( 34±5 . 3% and 26 . 5±6 . 6% for R42P and R46P , respectively ) ( Figure 9A and 9B and Figure S7A–S7E ) . Mutations in conserved cysteines of the RING domains ( the patient mutations C253Y , C289G , and C441R and the engineered mutation C332S ) completely disrupt recruitment at 1 h of CCCP treatment , as do mutations ( patient mutation Q311X and engineered mutation T415X ) that result in loss of RING2 ( Figure 9A and 9B ) . Mutations K211N and C212Y , which lie within a highly conserved region of Parkin that is likely a novel RING-like domain [41] ( Figure S5A ) , similarly blocked the recruitment of Parkin to mitochondria ( Figure 9B ) , consistent with the importance of this region for Parkin's activity . Mitochondrial recruitment was seen for several of the conserved cysteine RING mutants ( C289G , C332S , and C441R ) after 24 h of CCCP exposure , suggesting that recruitment is not completely disrupted with these mutations ( Figure S8A and S8B ) . Interestingly , the R275W mutation in RING1 exhibited only a mild deficit in recruitment ( 81 . 7±2 . 1% ) ( Figure 9A and 9B ) . The recruitment of YFP-Parkin R275W was verified in a live-cell imaging experiment ( Figure S8C ) . Although under control conditions some mutants formed visible aggregates ( Figure S8D ) , no mutant , including R275W , colocalized with mitochondria ( Figure 9B and Figure S8E ) . Next , we assessed recruitment of Parkin mutants to depolarized mitochondria by immunoblotting . As with our previous results [21] , we found some background YFP-Parkin signal in the membrane fraction under control conditions . Following treatment with CCCP for 1 h , levels of wild-type Parkin increase in the mitochondria-rich membrane fraction and decrease in the supernatant ( Figure 9C ) . Although expression of Parkin R275W was moderately less than wild type , it also increases localization in the membrane fraction and decreases in the supernatant upon CCCP treatment , consistent with the mitochondrial translocation seen for this mutation by confocal microscopy ( Figure 9C ) . No membrane translocation was detectable by Western blotting for either R42P or C332S , however , suggesting that translocation of these mutants is substantially lower than wild type and consistent with the deficit in mitochondrial translocation seen by confocal microscopy ( Figure 9C ) . We assessed the ability of Parkin mutants to induce mitophagy . As we found previously [21] , expression of wild-type Parkin in HeLa cells that do not detectably express endogenous Parkin completely eliminates mitochondria in greater than half of the cells ( 59 . 0±15 . 1% ) following treatment with CCCP for 24 h ( Figure 9D and 9E ) . Mutations in the UBL of Parkin exhibit a moderate loss in mitophagy activity ( 22 . 0±2 . 0% and 23 . 1±8 . 4% of cells exhibited no mitochondria for R42P and R46P , respectively ) ; whereas mutations in the conserved cysteines of the RBR or truncations that resulted in loss of RING2 exhibit a severe mitophagy deficit ( 0±0% to 5 . 3±2 . 3% , depending on the mutation ) ( Figure 9D and 9E ) . In addition , patient mutations R211N and C212Y caused a similar deficit in mitophagy , supporting the notion that this may be an atypical RING domain similar to RING1 , IBR , and RING2 ( Figure 9D ) . Interestingly , the R275W mutation in RING1 also exhibited a severe deficit in mitophagy ( 4 . 0±4% ) ( Figure 9D and 9E ) even though it appears to largely retain its ability to translocate to uncoupled mitochondria ( Figure 9A and 9B ) . This pattern of findings suggests that recruitment of Parkin to mitochondria and its induction of mitophagy are dissociable events .
We recently reported that the Parkinson disease-linked E3 ubiquitin ligase , Parkin , is selectively recruited to dysfunctional mitochondria with low membrane potential to promote their autophagic degradation , suggesting that a deficiency of mitochondrial quality control may underlie the observed mitochondrial dysfunction in Parkin knockout Drosophila and mice [11]–[15] , [17] , [21] . How Parkin is able to distinguish damaged , depolarized mitochondria from healthy , polarized mitochondria , however , was unknown . Here , we show that PINK1 selectively accumulates on depolarized mitochondria that have sustained damage . This selective accumulation is achieved by a novel mechanism , in which PINK1 is constitutively synthesized and imported into all mitochondria , but cleaved from healthy mitochondria by voltage-sensitive proteolysis ( Figure S9 ) . On damaged mitochondria that have lost their membrane potential , however , PINK1 cleavage is inhibited , leading to high PINK1 expression on the dysfunctional mitochondria . Expression of mitochondrial PINK1 is required for the recruitment of Parkin to the dysfunctional mitochondria and for their selective elimination by Parkin . In addition , increased expression of PINK1 on the outer mitochondrial membrane is sufficient for Parkin recruitment and Parkin-induced mitophagy , suggesting that loss of membrane potential activates Parkin recruitment primarily through the up-regulation of mitochondrial PINK1 . This model offers a parsimonious explanation for several observations that have been made previously . Full-length mitochondrial PINK1 ( ∼63 kDa ) is cleaved into a short ∼52-kDa form , but the short , primarily cytosolic form is unstable , raising the questions: why is PINK1 found both on the mitochondria and in the cytosol , and which form of PINK1 is active in the PINK1/Parkin pathway [22] , [28]–[30] . Our results suggest that full-length mitochondrial PINK1 is the active form in the PINK1/Parkin pathway , and that cleavage of PINK1 into an unstable cytosolic form maintains low levels of PINK1 on healthy mitochondria in order to suppress the PINK1/Parkin pathway in the absence of mitochondrial damage . Additionally , this model provides an explanation for the observation that the uncoupler valinomycin ( which can inhibit the TIM22/23 mitochondrial import pathway ) blocks PINK1 processing but fails to block PINK1 import [30] . Our model suggests that membrane potential is not required for PINK1 import but is required to selectively maintain low PINK1 expression on healthy mitochondria . This mechanism couples the collapse of mitochondrial voltage potential following mitochondrial damage to selective PINK1 accumulation on damaged mitochondria . At present , it is unclear which protease ( s ) mediate the cleavage of PINK1 in mammalian cells . Although the intramembrane serine protease Rhomboid-7 appears to be required for PINK1 cleavage in Drosophila [31] , our results suggest that PARL , its mammalian ortholog , is not required for PINK1 cleavage in mammalian cells . This situation is similar to that of OPA1 , which also requires Rhomboid-7 for cleavage in Drosophila , but does not require PARL for cleavage in mammalian cells [48] . In addition , determining how PINK1 cleavage is modulated by membrane potential will require further study . The protease itself may be sensitive to membrane potential and/or the PINK1 cleavage site may be available to the protease only in the presence of a membrane potential . Alternatively , the regulation of PINK1 cleavage by membrane potential may be indirect . That inhibition of PINK1 cleavage by mitochondrial depolarization up-regulates the PINK1/Parkin mitophagy pathway also raises the possibility that inhibitors of PINK1's protease might up-regulate the pathway and have some therapeutic benefit . Our results suggest that PINK1 induces Parkin recruitment to a particular subset of mitochondria , following its accumulation , and there are several models for how PINK1 might induce Parkin recruitment . In the simplest , as PINK1 accumulates , Parkin may be recruited to mitochondria through a direct interaction with the accumulated PINK1 . In support of this model , PINK1 appears to directly bind Parkin at least in some contexts [29] . Alternatively , PINK1 may need to phosphorylate Parkin , a substrate of Parkin , or an adaptor between PINK1 and Parkin , and thereby increase Parkin's affinity for a substrate or receptor on mitochondria . Consistent with a role for phosphorylation in the activation of Parkin , we found a kinase-deficient version of PINK1 fails to rescue Parkin recruitment to mitochondria in PINK1 null MEFs ( even though PINK1 KD appears to be processed identically to wild-type PINK1 ) . We were unable to replicate findings suggesting that phosphorylation of two threonines in a conserved region of Parkin is sufficient to induce Parkin recruitment to mitochondria [27] , but it is possible that Parkin may be phosphorylated by PINK1 elsewhere . If direct phosphorylation is sufficient to induce Parkin recruitment to mitochondria , however , it seems difficult to explain how Parkin can be targeted to a particular subset of mitochondria , as appears to occur in cells with bioenergetically diverse populations of mitochondria [21] . Mutations in Parkin and PINK1 are inherited primarily in a recessive manner , and loss of their function is thought to cause early-onset Parkinson disease . We find that patient mutations in PINK1 and Parkin disrupt the PINK1/Parkin mitochondrial turnover pathway at distinct steps , consistent with the potential relevance of this pathway for the development of Parkinson disease . Mutations in Parkin's UBL or its deletion caused a moderate deficit in Parkin recruitment to depolarized mitochondria and induction of mitophagy . That deletion of the UBL only partially inhibited the recruitment of Parkin to mitochondria suggests that whereas this domain promotes the recruitment of Parkin to mitochondria , it is not absolutely necessary for recruitment or subsequent mitophagy . The UBL likely promotes recruitment of Parkin through interaction with a protein containing a ubiquitin-binding domain , as mutating residue isoleucine 44 , which is critical for the interaction between UBLs and UBDs [47] , to alanine resulted in a recruitment deficit similar to that caused by deletion of the UBL domain . The disease-causing mutations R42P , which causes global unfolding by NMR [49] , and A46P lie on either ends of the beta-pleated sheet containing I44A , suggesting that these mutations may inhibit Parkin recruitment by disrupting the interaction between Parkin and UBD-containing proteins ( Figure S7A and S7B ) . Mutations in key cysteine residues in the RBR domain or deletion of RING2 , which is responsible for Parkin's ubiquitin ligase activity , severely disrupt both the recruitment of Parkin to mitochondria and its induction of mitophagy . Interestingly , the R275W mutation in RING1 of Parkin causes only a minor disturbance of Parkin recruitment to depolarized mitochondria but severely disrupts mitophagy , suggesting that recruitment and mitophagy can be experimentally disassociated . The R275W polymorphism in Parkin and the G411S polymorphism in PINK1 have only been identified as heterozygous polymorphisms in cases of Parkinson disease [42] , [50] . For this reason , the pathogenicity of these polymorphisms has been a matter of controversy . Our results show that the R275W Parkin mutation , which affects a highly conserved arginine residue , causes a significant loss of Parkin function in our mitophagy assay . This is consistent with in vivo data in Drosophila melanogaster , demonstrating that Parkin R275W , unlike wild-type Parkin , fails to compensate for loss of endogenous Parkin . By contrast , we found that PINK1 containing the G411S polymorphism , which is conserved in vertebrates , but not invertebrates , could compensate for loss of endogenous PINK1 , consistent with the view that PINK1 G411S may be a natural variant and not a disease-causing mutation . The stringent dependence of Parkin recruitment on PINK1 under depolarizing conditions is a little surprising given that , when overexpressed , Parkin can partially compensate for PINK1 loss in Drosophila and in mammalian cells [11]–[13] , [20] . How Parkin overexpression compensates for PINK1 loss is not known , but there are several possible explanations . First , there may be mechanisms independent of PINK1 and depolarization that can recruit Parkin to dysfunctional mitochondria . Alternatively , Parkin may serve other functions in the cell that are independent of PINK1 and protect against mitochondrial dysfunction indirectly; or Parkin may function to some degree upon overexpression independently of mitochondrial docking , perhaps effecting mitophagy or other mitochondrial changes from the cytosolic compartment . Stable loss or knockdown of PINK1 in mammalian cellular models and mice leads to a number of mitochondria-related abnormalities . Mitochondria in these cells or tissues exhibit electron transport chain ( ETC ) dysfunction , diminished membrane potential , increased reactive oxygen species production , mitochondrial fragmentation , and calcium dysregulation , among other abnormalities [20] , [32] , [34] , [51] . Although some of these abnormalities may be a reversible consequence of others—for instance , mitochondrial fragmentation may be due to low membrane potential [34] , and ETC dysfunction and decreased membrane potential may be , in part , a functional consequence of calcium dysregulation [51]—other abnormalities may be due to irreversible dysfunction of specific mitochondrial proteins or protein complexes . For instance , Complex I and the putative Na+/Ca2+ transporter seem to be dysfunctional in cultured cells following PINK1 knockdown [51] , whereas Complex I and II appear to be dysfunctional in the striatum of mice lacking PINK1 [16] . Although the proximate cause of these abnormalities in PINK1 null cells remains obscure , one explanation may be the failure of PINK1/Parkin pathway to eliminate oxidatively damaged mitochondria , which accumulate over time as a natural consequence of metabolism and other cellular stresses . That Parkin null cells and tissues appear to share some of the same mitochondrial defects as PINK1 null cells and tissues supports the view that these abnormalities may be due to loss of a common PINK1/Parkin pathway [17] , [18] . We cannot rule out that PINK1 may actively prevent mitochondrial damage and dysfunction , in addition to its signaling role in the PINK1/Parkin pathway . PINK1's interaction with HtrA2/OMI , for instance , appears to be independent of Parkin function in Drosophila [31] , [52] , [53] . Loss of PINK1 and Parkin affects some cell populations , like substantia nigral neurons , more than others , even though PINK1 and Parkin appear to be more widely expressed . Why some tissues are more vulnerable to loss of PINK1/Parkin than others is unclear , but it may relate to the degree of damage mitochondria sustain within that tissue ( e . g . , mitochondria in the SN are subject to greater oxidative stress than those in other neural tissues [2] ) ; the existence of redundant mitophagy pathways ( e . g . , mammalian tissues may contain pathways orthologous to those recently identified in yeast [54] , [55] ) ; the ability of the tissue to mitigate the damage by other means ( a tissue composed of mitotic cells may be able to manage mitochondrial damage through cellular turnover rather than mitochondrial turnover ) ; and mitochondrial demand within a particular tissue ( neurons have high , local metabolic demands , and dopaminergic neurons are subject to especially high calcium fluxes that need to be buffered by mitochondria [56] ) . Some or all of these factors may contribute to the special reliance of SN neurons on PINK1 and Parkin . PINK1 and Parkin are a significant cause of autosomal recessive parkinsonism and have been genetically linked to a pathway that protects against progressive mitochondrial damage and dysfunction . We have found that PINK1 levels and subsequently Parkin recruitment to mitochondria are dramatically regulated by the bioenergetic state of individual mitochondria , and that this unique regulation may allow PINK1 and Parkin to promote the selective and efficient turnover of mitochondria that have become damaged . Loss of PINK1 or Parkin function due to pathogenic mutations can disrupt this mitochondrial turnover pathway which may lead to the accumulation of dysfunctional mitochondria in vulnerable tissues—with a resultant increase in oxidative stress , depression of metabolism , and , eventually , accelerated cell death , which has been observed in Drosophila and , to a lesser extent , in mouse models of the disease [11]–[17] . Together , these findings provide a biochemical explanation for the genetic epistasis between PINK1 and Parkin observed in Drosophila , and support a novel , testable model of how loss of PINK1 and Parkin function may lead to autosomal recessive parkinsonism .
HeLa YFP-Parkin , E18 Rat cortical neurons , PINK1+/+ SV40-transformed MEF cells , PINK1−/− SV40-transformed MEF , M17 neuroblastoma control shRNA , M17 neuroblastoma PINK1 , Mfn1/2−/− MEF , and Parl−/− MEF cell lines have been described previously [21] , [25] , [29] , [33] , [57] , [58] . PINK1+/+ and PINK1−/− primary MEFs were isolated from embryos using a standard protocol [16] . Parkin+/+- and Parkin−/−-transformed MEFs were created by isolation of primary cells from embryos of B6 . 129S4-Park2tm1Shn/J mice ( Jackson Labs ) , using a standard protocol [16] , followed by retroviral transduction of SV40 ( Applied Biological Materials ) . YFP-Parkin , YFP-Parkin mutants , mCherry-Parkin , PINK1-YFP , PINK1KD-YFP , PINK1 Δ1-110-YFP , and Opa3-PINK1 Δ1-110-YFP are in C1 or N1 Clontech vectors . PINK1WT-V5 , PINK1KD-V5 , and PINK1 Δ1-156-V5 are in pDest40 vector ( Invitrogen ) . PINK1 patient mutations are in the pLenti-V5 vector ( Invitrogen ) . PINK1-myc is in a pCMBTNT vector ( Promega ) . The PARL shRNA construct targeting ( 5′-CCAACTTGGAGCTTCTAGTAAGTTCTCTACTAGAAGCTCCAAGTTGG-3′ ) is in the pSuper-GFP vector ( OligoEngine ) . To make FRB-PINK1 ( 111–581 ) -YFP and Tom20 ( 1–33 ) -FKBP , PCR fragments containing PINK1 ( 111–581 ) -YFP and Tom20 ( 1–33 ) were cloned into the BamHI site of the pC4-RHE vector and the EcoRI and XbaI sites of pC4M-F2E vectors , respectively ( ARIAD Pharmaceuticals ) . The rapamycin analog AP21967 was obtained from ARIAD Pharmaceuticals . Confocal microscopy of fixed samples , scoring of Parkin recruitment and Parkin-induced mitophagy , and live-cell imaging were performed as described previously [21] . Experiments in Mfn1/2 null cells were performed as described previously , with minor modifications , as described in the supplemental materials and methods ( Text S1 ) [21] . For PINK1 experiments , cells were fractionated using the Mitochondria Isolation Kit ( Pierce ) , according to manufacturer's specifications , with slight modifications , as described in the supplemental methods ( Text S1 ) . To isolate integral membrane proteins , membrane fractions obtained as above were carbonate extracted with 0 . 1 M Na2CO3 fresh cold buffer , and membranes were pelleted , as described in the supplemental materials and methods ( Text S1 ) . For Parkin experiments , cells were fractionated as described previously , with minor modifications , as described in the supplemental materials and methods ( Text S1 ) [21] . The protease protection assay was performed as described previously [22] . Cells were fixed and immunostained as described previously [21] . The following primary antibodies were used: anti-Parkin ( PRK8 ) monoclonal ( Santa Cruz Biotechnology ) , anti-Tom20 polyclonal ( Santa Cruz Biotechnology . ) , anti-cytochrome c monoclonal ( BD Biosciences ) , anti-PINK1 polyclonal ( Novus Biologicals ) , anti-VDAC monoclonal ( Calbiochem ) , anti-GAPDH polyclonal ( Sigma-Aldrich ) , anti-Tubulin monoclonal ( Sigma-Aldrich ) , anti-V5 monoclonal ( Invitrogen ) , anti-GFP polyclonal ( Invitrogen ) , anti-TIM23 monoclonal ( BD Biosciences ) , and anti-Hsp60 monoclonal ( Stressgen ) . qRT-PCR of PINK1 mRNA levels was performed as described in detail previously [34] .
|
Mutations in the PINK1 or Parkin genes lead to an inherited form of Parkinson disease . Understanding how the products of these genes work may give us insights into what goes wrong in these patients and in Parkinson disease more generally . Previous studies in flies and mice , and in human cells suggest that PINK1 and Parkin are part of a common pathway that protects against damaged mitochondria; these organelles power the cell when healthy but can produce harmful reactive oxygen species when damaged . Exactly how PINK1 and Parkin work together to protect against damaged mitochondria is unclear . The findings we report in this paper suggest a new model in which PINK1 and Parkin together sense mitochondria in distress and selectively target them for degradation . In this pathway , PINK1 acts as a flag that accumulates on dysfunctional mitochondria and then signals to Parkin , which tags these mitochondria for destruction . Since disease-causing mutations in PINK1 or Parkin disrupt this pathway , patients with these mutations may not be able to clean up their damaged mitochondria , leading to the neuronal damage typical of parkinsonism .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"biochemistry",
"cell",
"biology/cell",
"signaling",
"cell",
"biology",
"neurological",
"disorders/movement",
"disorders",
"neuroscience/neurobiology",
"of",
"disease",
"and",
"regeneration"
] |
2010
|
PINK1 Is Selectively Stabilized on Impaired Mitochondria to Activate Parkin
|
Cutaneous Leishmania major has affected many travelers including military personnel in Iraq and Afghanistan . Optimal treatment for this localized infection has not been defined , but interestingly the parasite is thermosensitive . Participants with parasitologically confirmed L . major infection were randomized to receive intravenous sodium stibogluconate ( SSG ) 20mg/kg/day for ten doses or localized ThermoMed ( TM ) device heat treatment ( applied at 50°C for 30 seconds ) in one session . Those with facial lesions , infection with other species of Leishmania , or more than 20 lesions were excluded . Primary outcome was complete re-epithelialization or visual healing at two months without relapse over 12 months . Fifty-four/56 enrolled participants received intervention , 27 SSG and 27 TM . In an intent to treat analysis the per subject efficacy at two months with 12 months follow-up was 54% SSG and 48% TM ( p = 0 . 78 ) , and the per lesion efficacy was 59% SSG and 73% TM ( p = 0 . 053 ) . Reversible abdominal pain/pancreatitis , arthralgias , myalgias , headache , fatigue , mild cytopenias , and elevated transaminases were more commonly present in the SSG treated participants , whereas blistering , oozing , and erythema were more common in the TM arm . Skin lesions due to L . major treated with heat delivered by the ThermoMed device healed at a similar rate and with less associated systemic toxicity than lesions treated with intravenous SSG . ClinicalTrials . gov NCT 00884377
Cutaneous leishmaniasis is a parasitic infection transmitted by the bite of an infected sand fly , with an estimated worldwide incidence of 1 . 5 million cases [1] . Generally self resolving within months , infection with Leishmania major can leave disfiguring scars or chronic ulcers , usually in areas not covered with clothing . Since 2003 , the U . S . military has reported >1 , 300 cases of cutaneous leishmaniasis in Army soldiers [2] . While systemic use of the pentavalent antimonial sodium stibogluconate ( Pentostam , Glaxo Smith Kline , United Kingdom ) showed efficacy with treatment responses >90% in cutaneous leishmaniasis , field ready treatment of short duration using limited supplies or equipment was considered preferable [3] . Dermatotrophic Leishmania species such as L . major , L . tropica , and L . mexicana are thermosensitive with higher temperatures limiting amastigote replication [4]–[6] . In mice infected with L . mexicana , environmental therapy at 37°C for 4–6 weeks was compared to ten day cycles of intralesional meglumine antimoniate . Thermotherapy was associated with more rapid resolution compared to antimonial treatment but both had similar cure rates at 10 weeks ( 93% and 88% respectively ) . All untreated control animals had large lesions by week 19 [7] . Heat treatment of leishmaniasis has been delivered by various methods , including hot water soaks , circulating hot water in heating pads , alternating ultraviolet and infrared heat , and ultrasonically induced hyperthermia of the skin [8]–[11] . Clinical trials used several devices which provided localized current from radiofrequency instruments to generate heat in affected tissue [12]–[19] . One of these instruments , ThermoMed Model 1 . 8 , ( Thermosurgery Technologies , Inc . , Phoenix , AZ ) has received 510 ( K ) clearance from the Food and Drug Administration ( FDA ) for treatment of cutaneous leishmaniasis . Published experience suggests thermotherapy efficacy was similar to or better than intralesional or parenteral pentavalent antimonial treatment , but the studies used differing entry criteria , parasite species , follow-up , and outcome definitions [12] , [15]–[19] . Our controlled trial provides detailed information about the course and toxicity in well characterized cases of dermatotrophic L . major in “traveling” soldiers without prior immunity . Our subjects left the endemic region by the time of treatment and were adherent with follow-up , thereby eliminating factors that plagued prior studies .
The trial was performed at Walter Reed Army Medical Center ( WRAMC ) in Washington , DC in 2004–6 . Eligible participants were Department of Defense health care beneficiaries with parasitologically confirmed cutaneous leishmaniasis . Based on travel history , all participants were likely infected in Iraq or Kuwait . All were treatment naive . Randomization criteria included L . major species confirmation . Exclusionary criteria were >20 lesions , pregnancy , lactation , hypersensitivity to pentavalent antimonials or local anesthetics , serious medical illness , lesion in proximity to mucous membranes , face , or cartilage , implanted metallic devices , or unwillingness to avoid procreation for at least two months . The protocol received Institutional Review Board approval at Walter Reed Army Medical Center and all participants provided written informed consent . A sample size of 27 participants per treatment group was planned , assuming a 73% cure rate of TM [12] , 99% for SSG [20] , controlling for a probability of a Type I error at alpha = 0 . 05 and was predicted to have 80% power to determine a 26% difference in outcome . This controlled trial compared 10 days of 20 mg/kg/day intravenous sodium stibogluconate ( SSG , GlaxoSmithKline , United Kingdom ) to one treatment session using the ThermoMed Model 1 . 8 device ( TM ) . SSG was infused over 10–50 minutes for a total of 10 doses . Participants randomized to thermotherapy received oral antibiotics for secondary bacterial infections of the leishmaniasis lesion ( s ) prior to treatment ( SSG arm participants were treated concurrently ) . Prior to thermotherapy , each lesion was cleansed , anesthetized , moistened , and overlying eschar was removed . Two trained physicians performed TM treatments . The TM probe was placed on the skin , covering the lesion with 50°C TM treatments applied for 30 seconds in a grid fashion extending 4mm into border skin . The lesion size determined the number of applications . Each lesion was then covered with a dressing ( Coverlet , Beiersdorf-Jobst , Wilton CT ) that was changed daily . Participants were seen daily for 10 days with physical examination , laboratory testing ( complete blood count , creatine phosphokinase , amylase , lipase , complete metabolic profile ) , and electrocardiograms at baseline , days 3 , 7 , 10 . Interview follow-up by telephone , email , letter , or infrequently in person , occurred at 2 , 6 , and 12–24 months post treatment . Photographs were taken at baseline , day 10 , and requested with 2 , 12 month follow-ups . Sequential photographs were independently assessed by blinded leishmaniasis experts , who were clinicians experienced in the treatment of CL , with a tiebreaker assessment when needed . Participants with clinical failure at 2 months were offered crossover treatment . The primary study objective was to compare efficacy of local thermotherapy with the TM device to a 10 day course of parenteral SSG , ( IND# 14150 ) , to treat cutaneous L . major ( 2 months post treatment ) . Secondary objectives included comparison of the treatment toxicity profiles . Clinical cure was defined as complete epithelialization or visually healed at 2±1 month after completion of therapy and no reactivation in 12 months after the start of treatment . Clinical failure was less than complete epithelialization or visually not healed at 2±1 month after treatment completion . Relapse failure was defined as skin lesion persistence at the treatment site or elsewhere in the period up to 12 months after start of therapy , regardless of appearance at 2±1 month after treatment completion . The overall treatment responses are reported as determined by: subject , blinded expert photograph review , and consensus response where photograph review was the prioritized outcome but subject assessment was used when photographs were not provided or were inadequate for review . The statistician ( RH ) generated the randomization plan in blocks of 4 subjects using www . randomization . com . The research pharmacist made assignments using the randomization plan in sequential order . The allocation sequence was unavailable to investigators until completion of the trial . Diagnosis was confirmed prior to enrollment using skin scrapings or biopsy from one characteristic lesion per subject . Histology was reviewed at the Armed Forces Institute of Pathology , Washington , DC . Leishmania culture and genus/species PCR were done at the Walter Reed Army Institute of Research , Silver Spring , MD . Speciation was determined by isoenzyme analysis of positive cultures and a glucose phosphate isomerase ( GPI ) based RT-PCR assay for Leishmania genus PCR positive samples [21] . Subject data were entered into Access ( Microsoft ) software and analyzed in SPSS ( Windows v17 , SPSS Inc . , Chicago , IL ) . Groups were compared using Fisher's exact test and the Wilcoxon rank sum test . Logistic regression and the Mantel Haenszel test were used to examine the effect of multiple risk factors on healing . Serial laboratory assessments were compared using repeated measures analysis of variance . Data not satisfying assumptions of normality were logarithmically transformed . All p values are two-sided and p<0 . 05 was considered statistically significant . Study data were analyzed in an intent to treat ( ITT ) basis with missing data handled as missing . The four treatment crossover subjects were considered failed in the analysis , the twelve month outcome was not evaluable given multiple treatment modalities .
We enrolled 56 participants but two were subsequently excluded: one lacked confirmed speciation and one for olecranon bursitis under a lesion which might have been associated with excessive burn injury . Thus , 27 participants in each arm completed study prescribed treatment . Figure 1 shows good adherence with study follow-up visits . End points and follow-up data were available for all but one SSG treated participant who withdrew from follow-up after the 6 month visit . Participants receiving crossover treatment and the withdrawn subject were unable to be assessed for the 12 month outcome ( 3 in TM , 2 in SSG ) , although all eligible were contacted ( n = 4 ) and endorsed being healed . The baseline demographic characteristics ( Table 1 ) were similar between treatment arms with the exception of lesion location . In the SSG treatment group there was a higher number of arm and chest lesions and fewer neck , leg , and back lesions . All treated participants had parasitologic confirmation of L . major infection . In an ITT analysis at two months after the completion of treatment , 12/25 evaluable TM participants ( 48% ) and 14/26 SSG treated participants ( 54% ) were assessed as clinically cured with complete and durable healing of all lesions ( p = 0 . 78 , odds ratio: 0 . 79 , ( 95% CI: 0 . 26–2 . 38 ) ) . At 2 months , 63 TM treated lesions ( 73% ) versus 50 SSG treated lesions ( 59% ) were assessed as clinically cured ( p = 0 . 053 , odds ratio: 1 . 92 ( 95% CI: 1 . 01–3 . 65 ) ) . There were two relapse failures . Relapses occurred prior to the six month follow-up and in the TM arm included activation/ulceration at a site not treated , subcutaneous nodules forming around the treated area , or breakdown within the edge of the TM treated area . After the two month follow-up , three persons in the TM assigned arm crossed over to SSG and one person in the SSG assigned arm was subsequently retreated with TM . Other participants with clinical failure at 2 months elected for no further treatment . The survival analysis of time to healing for the two treatment arms ( Figure 2 ) suggested that participants in the SSG treated arm appreciated a more rapid time to healing as compared to thermotherapy treated ( p = 0 . 058 ) when healing was based solely on patient evaluation . This difference was not found in the analysis using blinded reading of photographs where the efficacy by treatment was statistically similar ( p = 0 . 24 ) . The demographics of participants with clinical cure at 2 months versus failure ( Table 2 ) suggested no difference in race , duration of lesion , lesion size , or number of TM treatments/lesion . Subjects with clinical cure were 3 years younger ( p = 0 . 019 ) . There was a difference in the participant response between the two TM operators ( p = 0 . 019 ) , but not a per lesion response difference , ( p = 0 . 38 ) . The operator with the higher failure rate treated subjects with more lesions ( median = 3 , range: 1 to 14 ) compared to a median of 2 ( range: 1 to 4 ) for the other operator ( p = 0 . 06 ) . Among those participants with clinical failure , the median number of lesions was higher ( p = 0 . 026 ) and lesion location also differed with more arm , chest involvement in those failing ( p = 0 . 001 ) . When the number of lesions and age were included in a multivariate analysis , the difference between treatment groups was not significant ( p = 0 . 85 , odds ratio: 0 . 89 ( 95% CI: 0 . 27–2 . 91 ) ) . When location was taken into account in the analysis of lesions , treatment groups did not differ significantly ( p = 0 . 30 , odds ratio: 1 . 44 ( 95% CI: 0 . 72–2 . 86 ) ) . Complete cure at two months after completion of treatment was a very stringent outcome assessment , so we compared clinical outcome where participants were graded as <50% lesions healed , ≥50% lesions healed and 100% lesions healed . This demonstrated no significant differences in efficacy between treatment arms at the 2 or 12 month follow-ups ( Figure 3 ) . A secondary objective was comparison of toxicity profiles . Daily physician evaluations occurred during treatment with frequent laboratory and electrocardiogram evaluation . There were seven serious adverse events , four in the TM arm and three in the SSG arm . All were assessed as no or remote relationship to study treatment , consisting of a basal cell carcinoma diagnosis prior to treatment and hospitalizations during the follow-up period for gastroenteritis , elective repair of the left acromioclavicular joint , pneumonia , surgery for rectal carcinoid , inflammatory bowel disease , and trauma from a motor vehicle accident . Serial laboratory values over the 10 day treatment , segregated by treatment modality , show statistically significant decreases in white blood cells , hematocrit , platelets , and increases in amylase , lipase , ALT , AST in those receiving SSG ( Table 3 ) . Of note , changes in creatinine , corrected QT interval , and creatine kinase ( CK ) were similar . After treatment initiation , there were five lesion-related wound infections ( 19% ) in the TM treated arm and one ( 4% ) in the SSG treated group ( p = 0 . 19 ) . In the TM arm , the median time to wound infection after treatment was 10 days ( range 3–75 ) . Lesion reactions associated with TM included blister 25 ( 93% ) , oozing 21 ( 78% ) , and keloid 2 ( 7% ) . Participant description of their lesions showed no qualitative difference between treatment arms at the 12–24 months follow-up . Chest pain , abdominal discomfort , diarrhea , nausea , vomiting , arthralgias/myalgias , rash , headache , dizziness , fatigue , and pyrexia were more common in the SSG treated arm , ( Table 4 ) . Among SSG treated participants , 82% developed elevated amylase or lipase at some time versus none in the TM treated arm . 46% SSG and 7% TM treated developed elevated ALT/AST . CK was increased over baseline values in 25% SSG and 32% TM .
Standard treatment practice in the U . S . for cutaneous L . major includes no treatment ( generally for <5 uncomplicated skin lesions ) , use of investigational drugs including SSG , miltefosine , paromomycin , or off label use of azoles or amphotericin products . These present an array of adverse effects to consider in each individual with a cosmetically concerning/scarring dermal infection generally not associated with systemic complications . Our results confirm that a single heat treatment of lesions with TM was as effective as 10 days of intravenous SSG at 20mg/kg/day in L . major infection in American soldiers . Our primary endpoint of healing at two months with no reactivation over 12 months showed 48% TM and 54% of SSG treated participants achieved a clinical cure . Additionally , per lesion analysis showed 73% of TM treated lesions and 59% of SSG treated lesions at 2 months follow-up were clinical cures ( p = 0 . 053 ) . We had excellent adherence; all receiving study treatment completed the course with daily physician follow-up for ten days . Unexpectedly , interview responses from participants regarding outcome at 2 , 6 , 12–24 months suggested a perception that SSG was associated with more rapid healing ( p = 0 . 058 ) . This was not corroborated by the expert ( blinded ) committee assessment of lesion photographs . We speculate a perception bias developed among participants that TM was a less potent treatment ( SSG is a standard treatment at WRAMC and does not initially enlarge the lesion like TM ) . Participants generally endorsed no perceived cosmetic outcome difference between the treatments at the latest follow-up . We monitored toxicity of TM and SSG with laboratory and electrocardiogram testing , daily physician assessments for 10 days after treatment and long follow-up ( 12–24 months ) . The thermotherapy arm provided insight into the differential frequency of adverse effects of a 10 day course of intravenous pentavalent antimony . Systemic effects including pancreatitis ( 67% ) , arthralgias , myalgias ( 59% ) , headache ( 44% ) and fatigue ( 67% ) were more likely in the SSG treated group and local effects of lesion blistering ( 93% ) , oozing ( 78% ) , and erythema ( 26% ) were more likely in the TM ( which causes a burn ) treated arm . Interestingly , there was no statistical difference in electrocardiogram findings , including changes in QTc , between the two treatment groups . Compared to the typical SSG regimen of 20 days in a similar military population , the magnitude of change in laboratory values was lower in the 10 day SSG infusion course [3] . Previous studies of thermotherapy in Old World cutaneous leishmaniasis have suggested reasonable but variable efficacy . A small observational study in U . S . soldiers showed 88% efficacy of TM treatment with a 6 months follow-up [16] . A large randomized trial in Afghanistan of presumed L . tropica using the ThermoMed device found 69% cure at 100 days follow-up and lesser efficacy ( 45% ) with intramuscular SSG ( Albert Davis Ltd , India ) with a maximum dose of 850 mg per day [17] . In Iran ( presumed L . major species ) a different radiofrequency heating device was used in four weekly sessions compared to intralesional meglumine antimoniate with 6 month efficacy of 81% for thermotherapy and 55% for intralesional antimonial [18] . Studies of thermotherapy in New World cutaneous leishmaniasis have also suggested efficacy: 73% in Guatemala [12] , 90% at eight weeks in Mexico [14] , 71% in Brazil at 28 days [15] , and 100% per protocol but 19% in ITT analysis in Colombia at 100 days [19] . Our study has an early outcome measure at 2 months after the start of treatment , although information at 6 months was also obtained . It is difficult to identify another OWCL study where similar species ( often mixed/not specified L tropica and L . major ) and outcome definition were applied at 2 months . We tested a shorter course using full dose intravenous SSG ( no upper limit to the daily weight based dose ) which has not been studied head to head in a randomized trial of other agents . While the standard duration of intravenous SSG is 20 days , a previous study at our center suggested similar efficacy using a 10 day course at 6 month follow-up , with less toxicity [21] . From 2003–6 , in 141 L . major patients treated at our center we found 92% clinical cure with SSG 10 versus 95% in SSG 20 patients ( n = 273 ) at six months follow-up [personal communication Aronson] . Importantly , this thermotherapy study findings suggest that a single heat treatment of cutaneous leishmaniasis with a FDA ‘approved’ device suitable for resource limited areas has similar healing rates to SSG for 10 days . We further described the efficacy and lower toxicity seen with short duration intravenous SSG therapy in L . major infection . The strengths of our study include a randomization , similar detailed assessments of both treatment arms , prolonged follow-up , excellent adherence , parasitologic species confirmation , and rigorous outcome assessment with treatment blinded experts reviewing lesion photographs supplemented with participant's impression . A universal measure of healing in Old World cutaneous leishmaniasis is lacking but the standard timepoint seems to be 3 months after treatment [22] . We chose an earlier timepoint to maximize our observation of the effect of our interventions in a self healing infection . Regarding study limitations , our study lacks a placebo arm to address the self healing observed with time and variably described as 10–68% in a systematic review in L . major [23] . A Cochrane review suggested L . major self heals in 2–3 months , but may persist up to 5 years [22] . In our study , the median duration of skin lesions prior to treatment exceeded 120 days; there likely was a referral bias for more persistent and severe L . major infection being sent back to the U . S . for treatment . Regarding diagnosis , only one lesion was sampled per participant and unresponsive skin lesions may not always have been cutaneous leishmaniasis , however in a randomized clinical trial this is unlikely to have a differential effect . The generalizability of our results is limited to L . major lesions excluding the face , of moderate size ( mean 205 mm2 ) , in otherwise healthy men who constituted our study population . The outcome assessment was restricted by the lower quality of some and incomplete provision of returned photographs as well as subjective information relayed by the participants rather than in person evaluations during the follow up . The clinical implications of our study are that a single thermotherapy treatment showed similar healing rates and was as well tolerated as a 10 day infusion of SSG for L . major infection . Thermotherapy devices are portable , multi-use , and easy to operate in resource limited environment . This contrasts to the time , labor , and cost involved using an intravenous investigational new drug such as SSG under FDA regulations which required transfer to the U . S . for our subjects . Recognizing that L . major often results in a self healing infection , the choice of treatment intervention should weigh patient wishes , and consider the number , location , duration , size of lesions , and signs of local dissemination . Thermotherapy is generally well tolerated and may limit the size of subsequent scar . The effect is local so use in disseminated infection is not recommended . Large numbers of lesions and very large lesions require a significant amount of time to treat with the device . To mitigate the risk of secondary wound infection of the burn , others have used topical bacterial ointment during the initial post treatment period [16] . Analyses have not found antimonial treatment of cutaneous leishmaniasis as cost-effective . While thermotherapy cost has not been included in these models , it is likely to be cost-effective in endemic areas [17] , [24] , [25] . Future research could assess whether thermotherapy treated areas showed persistent Leishmania organisms as seen after clinical healing post SSG; if not , then the implication of lessened risk of later reactivation but also the role of potentially lessened protection from reinfection might be better understood [26] , [27] .
|
Cutaneous leishmaniasis , a parasitic skin infection transmitted by the bite of a sand fly , can result in chronic skin sores and is estimated to affect more than 1 . 5 million persons worldwide . While the infection generally heals on its own in months to years , treatment can be expensive and difficult . We compared a heat treatment using the ThermoMed device to an ( abbreviated ) ten day course of intravenous Pentostam ( a pentavalent antimony drug ) in a population of U . S . soldiers who acquired their infections in Iraq . We found no statistically significant difference between the two treatments in the healing rate at two months . The heat treatment had less associated toxicity . Heat therapy is a ruggedized , battery operated method that could be adapted to humanitarian situations and less developed health care settings , likely with less cost and side effects than local treatment alternatives .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"infectious",
"diseases/neglected",
"tropical",
"diseases",
"infectious",
"diseases/tropical",
"and",
"travel-associated",
"diseases"
] |
2010
|
A Randomized Controlled Trial of Local Heat Therapy Versus Intravenous Sodium Stibogluconate for the Treatment of Cutaneous Leishmania major Infection
|
Digestion of blood in the midgut of Aedes aegypti results in the release of pro-oxidant molecules that can be toxic to the mosquito . We hypothesized that after a blood meal , the antioxidant capacity of the midgut is increased to protect cells against oxidative stress . Concomitantly , pathogens present in the blood ingested by mosquitoes , such as the arboviruses Dengue and Zika , also have to overcome the same oxidative challenge , and the antioxidant program induced by the insect is likely to influence infection status of the mosquito and its vectorial competence . We found that blood-induced catalase mRNA and activity in the midgut peaked 24 h after feeding and returned to basal levels after the completion of digestion . RNAi-mediated silencing of catalase ( AAEL013407-RB ) reduced enzyme activity in the midgut epithelia , increased H2O2 leakage and decreased fecundity and lifespan when mosquitoes were fed H2O2 . When infected with Dengue 4 and Zika virus , catalase-silenced mosquitoes showed no alteration in infection intensity ( number of plaque forming units/midgut ) 7 days after the infectious meal . However , catalase knockdown reduced Dengue 4 , but not Zika , infection prevalence ( percent of infected midguts ) . Here , we showed that blood ingestion triggers an antioxidant response in the midgut through the induction of catalase . This protection facilitates the establishment of Dengue virus in the midgut . Importantly , this mechanism appears to be specific for Dengue because catalase silencing did not change Zika virus prevalence . In summary , our data suggest that redox balance in the midgut modulates mosquito vectorial competence to arboviral infections .
Arthropod-borne viral ( arboviral ) diseases , such as Dengue , Chikungunya , and especially Zika , have recently occupied a central spot in the global discussions concerning infectious diseases due to the rapid spread of cases worldwide and the associated increase in syndromes such as microcephaly and Guillain–Barré , prompting the World Health Organization to declare Zika a public health emergency [1] . Over two million people live in areas where Zika has been reported , highlighting the risk of major epidemics , especially in Central and South Americas , as well as Southeast Asia [2] . Concerning Dengue , more accurate epidemiological data are available , and the number of annual infections could be as high as half a billion [3] . Currently , there is no vaccine to prevent new infections and no effective treatment options for sick individuals . Strategies targeting Aedes aegypti , the mosquito vector , such as utilizing the anti-viral effects of Wolbachia [4 , 5] and the sterile insect technique [6 , 7] , are attractive possibilities under implementation . To stop mosquito spread of arboviral diseases , we need to further elucidate the molecular interactions between the virus and the vector . This knowledge will help to explain , for example , the observed differences in susceptibility to viral infections of mosquito strains/populations[8 , 9] . Reactive oxygen species ( ROS ) have emerged as central molecules in a wide array of pathological as well as physiological processes , including signaling , immunity , cell proliferation and differentiation [10] . The biological actions of ROS are based on their ability to donate or receive electrons from biomolecules , triggering a diverse set of events associated with the normal function of cells . However , under oxidative stress , elevated levels of ROS may disrupt redox signaling pathways , leading to a non-homeostatic state commonly associated with disease [11] . Therefore , the correct balance between ROS-generating systems ( such as mitochondria , endoplasmic reticulum or NADPH oxidases ) and ROS-detoxifying reactions ( including antioxidant enzymes such as catalase , which detoxifies H2O2 into water and oxygen ) is critical for maintaining homeostasis in virtually all studied organisms . ROS metabolism influences critical parameters of insect physiology , including fecundity [12 , 13] , immune response [14 , 15] and vector competence in the interaction between Anopheles and Plasmodium [16–22] . In hematophagous arthropods , such as the Aedes aegypti mosquito , blood digestion in the midgut releases heme , a pro-oxidant molecule . Cells subjected to high concentrations of heme , such as gut epithelial cells after a blood meal , must maintain redox balance to avoid oxidative stress . Blood-sucking organisms have evolved a series of adaptations against the deleterious effects of excess heme [23–29] . One such mechanism is the activation of antioxidant enzymes , such as catalase , which reduces H2O2 and prevents its contact with heme/iron , a reaction known to generate highly toxic ROS [30 , 31] . An interesting possibility concerning hematophagous insect vectors is that the antioxidant protection induced upon blood feeding could also defend human pathogens being carried by the mosquitoes , such as arboviruses , from blood-induced oxidative challenge . Our results demonstrated that a blood meal up-regulated catalase mRNA and activity in the midgut epithelium and that silencing of catalase through RNAi reduced mosquito fecundity and resistance to hydrogen peroxide feeding . Catalase silencing had no effect on the intensity of infection with Dengue or Zika viruses . Interestingly , it reduced the prevalence of Dengue infection , but had no effect on the prevalence of Zika-infected females . This indicates that the redox environment of the midgut can alter mosquito susceptibility to infection with some flaviviruses .
All animal care and experimental protocols were conducted in accordance with the guidelines of the Committee for Evaluation of Animal Use for Research of the Universidade Federal do Rio de Janeiro ( CEUA-UFRJ ) . The protocols were approved under the registry CEUA-UFRJ 155/13 . Dedicated technicians at the animal facility at the Institute of Medical Biochemistry ( UFRJ ) carried out all aspects related to rabbit husbandry under strict guidelines to ensure careful and consistent handling of the animals . Two– 10-day-old Aedes aegypti females ( red-eye strain ) were used in all the assays and were maintained in 12-h light-dark periods at 28°C and 80% relative humidity . Females were fed ad libitum with cotton pads soaked in a 5% sucrose solution or allowed to feed on rabbit blood . Mosquitoes were cold-anesthetized and dissected in 50% ethanol . The midgut epithelia was separated from the blood bolus and collected in PBS ( 10 mM sodium phosphate buffer and 150 mM NaCl , pH 7 . 4 ) supplemented with a protease inhibitor cocktail ( 50 μg/mL SBTI , 1mM benzamidine , 1mM PMSF ) . Samples were mechanically homogenized with a pestle and stored at -80°C until use . Catalase activity was measured following H2O2 absorbance ( 240 nm for 1 min ) according to the protocol described by Aebi [32] in the presence of mosquito homogenates . The protein concentration was determined according to Lowry [33] . For in vivo inhibition experiments , mosquitoes were fed blood supplemented with different doses of 3-amino-1 , 2 , 4-triazole ( AT ) , a catalase inhibitor . For in vitro inhibition , AT was incubated for 30 minutes at 4°C with tissue homogenates before measurements of enzymatic activity . Total RNA ( pools of 5–10 midgut epithelia ) was extracted with TRIzol reagent ( Invitrogen ) according to the protocol suggested by the manufacturer and treated with DNAse I . cDNA was synthesized using a High-Capacity cDNA Reverse Transcription Kit ( Applied Biosystems ) according to standard protocols . The sequence CAT1A –AAEL013407 –RB ( Transcript ID from VectorBase ) was used to design the catalase primers . The primer sequences used for qPCR experiments were CAATGAACTGCACCGACAAC ( forward ) and AGCCTCATCCAGAACACGAC ( reverse ) . The sequence AAEL003396-RA ( Transcript ID from VectorBase ) was used as a housekeeping gene and corresponded to the ribosomal gene rp49 [34] . The corresponding primers for the housekeeping gene were GCTATGACAAGCTTGCCCCCA ( forward ) and TCATCAGCACCTCCAGCT ( reverse ) . qPCR was performed using the SYBR Green PCR Master Mix ( Applied Biosystems ) . Relative gene expression was calculated using the method described by Livak and Schmittgen [35] . A 1158-base-pair fragment of the catalase gene ( AAEL013407 –RB ) was amplified using cDNA from the midgut epithelia of blood-fed mosquitoes ( 24 h after feeding ) using primers F-TTCAAGGAGTCCCAGAAGGA and R-AACCGGAATCAGAGGGAACT . This amplicon was subjected to a nested PCR with catalase primers that also contained a T7 binding sequence ( which is necessary for RNA polymerase binding , see underlined ) . The primers used were TAATACGACTCACTATAGGGACTCCACTTGCTGTGCGTTT ( forward ) and TAATACGACTCACTATAGGGTCTCCCTTAGCAATAGCGTTG ( reverse ) . A 773-base-pair fragment was generated , purified and used as the template for an in vitro transcription reaction for the synthesis of double-stranded RNA ( dsRNA ) for catalase ( dsCat ) with a MEGAscript RNAi kit ( Ambion ) . RNAi experiments were performed via injection of 69 nL of dsCat ( 3 μg/μL ) ( or dsLacZ as an unrelated dsRNA ) in the thoraxes of 2-day old female mosquitoes . Mosquitoes were used 2–3 days after the injections . A 218-bp LacZ fragment was amplified using the primers GAGTCAGTGAGCGAGGAAGC ( forward ) and TATCCGCTCACAATTCCACA ( reverse ) and was cloned into the pCRII-TOPO vector . This plasmid was used for a subsequent PCR in which the T7 RNA polymerase promoter was also inserted , using the primer sequences GTAAAACGACGGCCAGT ( M13F ) and CTCGAGTAATACGACTCACTATAGGGCAGGAAACAGCTATGAC ( M13R ) . This PCR product was used for the synthesis of the dsLacZ performed with a MEGAscript RNAi kit ( Ambion ) . H2O2 was measured with Amplex Red ( Invitrogen ) following the recommendations of the manufacturer with minor modifications . The midgut epithelia of sugar-fed mosquitoes were dissected in 2 . 5% BSA in PBS , the gut contents were washed out and the epithelia ( pools of 5 organs ) were incubated in PBS under dim light at room temperature in the presence of Amplex Red ( 40 μM ) and 4 U horseradish peroxidase ( HRP , Sigma ) . After a 30 min incubation , the epithelia were centrifuged , and the supernatants were collected and evaluated for fluorescence emission at 530/590 nm ( Ex/Em ) in a Varian Cary Eclipse Fluorescence Spectrofluorometer . The resulting values were subtracted from fluorescence readings generated by nonspecific Amplex Red oxidation by the midgut epithelia ( pools of 5 organs ) in the absence of HRP . Dengue-4 was kindly provided by Dr . João Trindade Marques ( UFMG–Universidade Federal de Minas Gerais , Brazil ) . Zika virus was obtained from Dr . Laura Helena Vega Gonzales Gil ( Centro de Pesquisas Aggeu Magalhães , Fundação Oswaldo Cruz , Brazil ) . Viral stocks were propagated in C6/36 cells maintained in Leibovitz-15 ( L-15 ) media ( Gibco #41300–039 ) pH 7 . 4 supplemented with 5% fetal bovine serum , triptose 2 . 9 g/L , 10 mL of 7 . 5% sodium bicarbonate/L; 10 mL of 2% L-glutamine/L , 1% of non-essential amino acids ( Gibco #11140050 ) and 1% penicillin/streptomycin at 28°C . Culture supernatants containing viral particles were harvested , centrifuged , aliquoted and stored at -80°C until use . Plaque assays ( see section 2 . 8 ) were performed to determine viral titers . The Dengue-4 titer used was 1 x 107/mL . The Zika titer used was 2 x 107/mL . Mosquitoes were starved from sucrose ( but not water ) for 18–24 h and were offered a meal containing a 1:1 mix of rabbit red blood cells and L-15 media containing different amounts of Dengue-4 or Zika virus . ATP pH 7 . 4 at a final concentration of 1 mM was included as a phagostimulant . Viral stocks were thawed immediately before use . Mosquitoes were allowed to ingest the infectious blood through a membrane attached to an artificial feeder kept at 37°C for approximately 40 min inside a BSL-2 insectary facility . Mosquitoes were quickly cold-anesthetized , and fully engorged females were separated and housed as indicated in section 2 . 1 until use . Dengue-4 plaque assays were performed in BHK-21 cells and Zika plaque assays were performed in Vero cells maintained in DMEM ( Gibco #12100–046 ) supplemented with sodium bicarbonate , 1% L-glutamine ( 200 mM , Gibco #25030081 ) , 10% fetal bovine serum and 1% penicillin/streptomycin and seeded as monolayers ( approximately 70% confluency ) onto 24-well plates 12–24 h before the experiment . Seven days after the infectious meal , mosquitoes were surface-sterilized with 70% ethanol ( 20 seconds ) and rinsed twice with sterile PBS . Midguts were dissected using clean glass slides and forceps in sterile PBS and transferred to sterile Eppendorf tubes contain 200 μl of DMEM ( same as above ) and 50–100 mg of sterile glass beads ( Scientific Industries SI-BG05–0 . 5 mm diameter ) . Midguts were individually stored at -80°C until use . Midgut tissue was disrupted to liberate viral particles by vortexing the tubes for 10 minutes at room temperature . The samples were then centrifuged at 10 , 000 x g at 4°C and serially diluted in DMEM . One hundred μl of each sample was added to Vero ( Zika ) or BHK-21 ( Dengue-4 ) cell culture monolayers and gently shaken for 15 minutes at room temperature , followed by an additional 45 minutes without shaking at 37°C and in a 5% CO2 incubator . Subsequently , 700 μl of DMEM containing 2% FBS and 0 . 8% methylcellulose ( Sigma #M0512 . Viscosity 4 , 000 cP ) was added to each well . Plates were incubated at 37°C and 5% CO2 for five days . Samples were stained with a 1% crystal violet solution in a 1:1 ( v:v ) mixture of methanol/acetone for 1 h at room temperature and washed with water to remove excess dye . Then , individual plaque forming units ( PFU ) were visually counted . All experiments were carried out independently at least two times , and statistical analyses were performed with GraphPad Prism software . The appropriate tests are described in the figure legends .
To investigate the role of catalase in the midgut of Aedes aegypti in response to blood feeding , we compared gene expression in the epithelia of sugar-fed ( SF ) and blood-fed ( BF ) females dissected at 12 , 24 , 36 , 48 and 72 h after blood intake . Catalase mRNA levels increased 6-fold at 24 and 36 h after a meal and decreased to SF levels at 72 h ( Fig 1A ) . Enzyme activity was also monitored throughout the digestion process , which spans approximately 48 h . Fig 1B shows that H2O2-removing capacity also increased in the epithelia after feeding , reaching its maximal induction at 24 h , near the peak of blood digestion [36] , and returned to initial levels at 44 h . We tested the sensitivity of catalase to 3-amino-1 , 2 , 4-triazole ( AT ) , a well-known inhibitor [37] . Using an in vitro assay exposing midgut samples collected 24 h after a blood meal to different concentrations of AT , we showed that 100% of the epithelial H2O2 detoxification could be abrogated with AT concentrations below 20 mM and that 50% enzyme inhibition occurred close to 1 mM . Using a similar approach , we fed females blood supplemented with 15 mM AT and showed that catalase activity accounted for more than 90% of H2O2 removal 24 h after the meal ( Fig 1D ) . We injected Aedes aegypti females with dsRNA against catalase and evaluated the mRNA levels in the midgut epithelia . Fig 2A shows that the catalase transcripts were reduced by 93% and 86% in SF and BF mosquitoes 24 h after feeding . Consistent with the reduced mRNA levels , we observed a decrease in catalase activity in the epithelia of blood-fed mosquitoes ( Fig 2B ) . To determine whether catalase silencing affected midgut redox metabolism , we measured H2O2 ( the substrate of catalase and a diffusible ROS ) released by epithelial cells . Fig 2C shows that catalase knockdown increased hydrogen peroxide levels leaked to the supernatant . Together , Fig 2A–2C confirm that the RNAi approach negatively impacted catalase activity and redox metabolism of the midgut . To address its physiological significance , we demonstrated a reduction in lifespan of both sugar-fed ( SF ) and blood-fed ( BF ) mosquitoes challenged with sucrose supplemented with H2O2 ( Fig 3A and 3B ) . The median time to death was anticipated in 1 day in the dsCatalase group , which represents 15–20% of the lifespan of mosquitoes feeding on hydrogen peroxide under the conditions tested . We also observed a small but statistically significant 22% reduction in oviposition ( Fig 3C ) , similar to what was reported for Anopheles gambiae and Lutzomyia longipalpis [12 , 13] . To test the hypothesis that antioxidant protection triggered by a blood meal could influence a mosquito’s infection status with different flavivirues , we challenged catalase-silenced Aedes aegypti with two doses of Zika virus and measured the number of PFU per midgut ( infection intensity ) and the number of infected midguts ( infection prevalence ) seven days after administration of virus-contaminated blood . Catalase silencing did not change any of the parameters evaluated in Zika-infected females . ( Fig 4A–4D ) . Interestingly , the highest dose offered , 107/mL ( corresponding to our maximal titer obtained from C6/36 cells supernatants ) , and its 100-fold dilution , ( 105/mL ) , resulted in only a four-fold change viral loads after 7 days ( 107/mL mean PFU ~ 10000; 105/mL mean PFU ~ 3000 ) ( Fig 4A and 4C ) . However , the same doses resulted in a change in infection prevalence ( 107/mL Zika viral particles produced 100% prevalence in both dsLacZ and dsCat while 105/mL viral particles produced ~50% prevalence in both dsLacZ and dsCat ) ( Fig 4B and 4D ) . When we challenged dsCatalase-treated mosquitoes with the maximal Dengue-4 infectious dose ( 5 x 106/mL ) , there was no alteration in the median infection intensity ( Fig 4E ) . However , we observed a significant reduction in infection prevalence ( 70% of dsLacZ mosquitoes were infected vs 46% of dsCatalase; p = 0 . 0006 ±chisquare , demonstrating that a reduction in epithelial H2O2-removing capacity through catalase knockdown reduced the ability of Dengue-4 , but not Zika virus , to infect the midgut of Aedes aegypti .
Overall , we showed that a blood meal induced antioxidant protection in the midgut of Aedes aegypti and that RNAi-mediated knockdown of catalase resulted in reduced oviposition and lifespan when mosquitoes were challenged with H2O2 and decreased midgut virus prevalence after infection with Dengue-4 , but not Zika . The levels of pro-oxidant molecules , including hydrogen peroxide , in a given tissue must be carefully monitored to maintain normal cellular functions , with deviations from the optimal concentration being potentially harmful for the organism . The increase in H2O2 levels may be particularly deleterious to hematophagous arthropods because blood digestion releases heme , a pro-oxidant molecule , which may interact with cellular ROS , leading to oxidative stress [30 , 38] , decreased reproductive output [12 , 13] and possibly death of the insect [39 , 40] . We recently demonstrated that after a blood meal , Aedes aegypti inhibits the metabolic generation of ROS [24] . In that report , we showed that after blood intake , heme triggers a protein kinase C-dependent mechanism that inhibits dual oxidase ( Duox ) activity ( a source of ROS in the midgut ) , maintaining low ROS production during blood digestion compared with SF levels . The pattern of antioxidant enzyme expression observed in most organisms shows that they typically respond to increased levels of ROS , such as catalase in Rhodnius prolixus [29] . It is peculiar , then , that Aedes aegypti increases catalase expression and activity after a blood meal , especially when ROS levels were reduced compared with sugar-fed mosquitoes . We hypothesize that hematophagous mosquitoes evolved a redundant protection strategy to prevent oxidative stress following blood intake , which may explain the simultaneous decrease in ROS production and increase in antioxidant capacity . This is a major departure from most studies on ROS metabolism , where antioxidant enzymes are regulated by previous oxidant stress . Redox metabolism has been implicated in the response of Anopheles mosquitoes to plasmodium infection [20 , 41] . While in Anopheles gambiae catalase knockdown reduced Plasmodium berghei oocyst counts , supposedly through augmented concentration of toxic H2O2 , it increased oocysts of the human malaria P . vivax in its natural vector , Anopheles aquasalis [22] , revealing a complex and species-specific role of catalase in the gut of mosquitoes during malaria infection . Regarding the interaction of ROS and microorganisms , high levels of free radicals are believed to be detrimental . However , an emerging concept posits that similar to the role of catalase in P . vivax–An . aquasalis , other parasites , such as Trypanosoma cruzi , thrive under host oxidative stress [42–45] . In the case of Aedes aegypti , altering redox homeostasis in the gut through catalase silencing reduced Dengue prevalence in the midgut ( Fig 4F ) , suggesting that ROS could antagonize infection by this specific arbovirus . However , catalase RNAi did not alter the percentage of infected midguts in females challenged with Zika ( Fig 4B and 4D ) , indicating a differential sensitivity of flaviviruses to ROS produced by Aedes aegypti . Our understanding of mosquito immunity to dengue virus has predominantly focused on classical immune genes and the roles of the Toll , Stat and RNAi pathways have been firmly established [46–49] . Additionally , genes that conventionally are not considered ( or labeled ) immune genes , such as lipid and redox metabolism , are also known to influence mosquito-arbovirus interactions and should not be overlooked [8 , 27 , 50] . Little is known about the molecular aspects of the interaction between Aedes aegypti and Zika , and antiviral mechanisms described for other flaviviruses will likely be involved [51] . In agreement with our result using Dengue-4 , it was recently shown for Dengue-2 that RNAi-silencing of the ROS-producing enzymes Duox and NoxM , as well as treatment of mosquitoes with the antioxidant vitamin C , enhanced viral infection in Aedes aegypti [52] . The data presented here suggested that catalase silencing altered the so-called midgut infection barrier ( MIB ) for Dengue-4 . MIB is a concept that refers to mechanisms involved in the inhibition of the initial contacts between virions and the intestinal epithelial cells , preventing the establishment of the infection [53 , 54] . These mechanisms may involve physical barriers , such as the peritrophic matrix , or physiological and immunological mechanisms , such as the midgut microbiota and/or the RNAi pathway , to name a few [55] . Here , catalase silencing was shown to reduce Dengue-4 prevalence possibly through an alteration in the midgut threshold of infection . If the virus is able to pass this bottleneck , then it establishes a successful cycle of replication , which was seen by the similar infection intensities of the catalase-silenced and control groups of insects . Importantly , this mechanism did not alter Zika prevalence in the midgut , indicating significant differences between how these two flaviviruses establish infections in the mosquito gut .
|
Mosquitoes ingest large amounts of blood , a rich and abundant source of energy to sustain egg production . Blood digestion offers challenges to the insect , like managing high concentrations of heme and iron , pro-oxidant and potentially toxic molecules derived from hemoglobin . Mosquitoes and other blood-feeding arthropods have evolved adaptations to overcome this problem , such as the activation of catalase , an antioxidant enzyme that protect tissues against toxic free radicals . Mosquitoes act as important vectors of human diseases because during a blood meal they might also ingest microorganism circulating in our blood , such as dengue and zika virus . The adaptive antioxidant program that protects mosquito tissues against the oxidative challenge imposed by a blood meal might also influences the ability of virus to establish infection and disseminate from the midgut to the salivary glands . We show here that catalase differentially influences the number of infected midguts after mosquitoes were challenged with blood contaminated with virus , being beneficial to Dengue-4 but neutral do Zika , suggesting that redox metabolism may have distinct roles on mosquito vector competence towards different arbovirus .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion"
] |
[
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"body",
"fluids",
"rna",
"interference",
"pathology",
"and",
"laboratory",
"medicine",
"enzymes",
"pathogens",
"enzymology",
"microbiology",
"animals",
"viruses",
"rna",
"viruses",
"epigenetics",
"insect",
"vectors",
"infectious",
"diseases",
"aedes",
"aegypti",
"proteins",
"medical",
"microbiology",
"genetic",
"interference",
"gene",
"expression",
"microbial",
"pathogens",
"arboviral",
"infections",
"catalases",
"disease",
"vectors",
"insects",
"arthropoda",
"biochemistry",
"rna",
"mosquitoes",
"blood",
"anatomy",
"flaviviruses",
"nucleic",
"acids",
"physiology",
"viral",
"pathogens",
"genetics",
"biology",
"and",
"life",
"sciences",
"species",
"interactions",
"viral",
"diseases",
"antioxidants",
"organisms",
"zika",
"virus"
] |
2017
|
Catalase protects Aedes aegypti from oxidative stress and increases midgut infection prevalence of Dengue but not Zika
|
Licensed human papillomavirus ( HPV ) vaccines provide near complete protection against the types of HPV that most commonly cause anogenital and oropharyngeal cancers ( HPV 16 and 18 ) when administered to individuals naive to these types . These vaccines , like most other prophylactic vaccines , appear to protect by generating antibodies . However , almost nothing is known about the immunological memory that forms following HPV vaccination , which is required for long-term immunity . Here , we have identified and isolated HPV 16-specific memory B cells from female adolescents and young women who received the quadrivalent HPV vaccine in the absence of pre-existing immunity , using fluorescently conjugated HPV 16 pseudoviruses to label antigen receptors on the surface of memory B cells . Antibodies cloned and expressed from these singly sorted HPV 16-pseudovirus labeled memory B cells were predominantly IgG ( >IgA>IgM ) , utilized diverse variable genes , and potently neutralized HPV 16 pseudoviruses in vitro despite possessing only average levels of somatic mutation . These findings suggest that the quadrivalent HPV vaccine provides an excellent model for studying the development of B cell memory; and , in the context of what is known about memory B cells elicited by influenza vaccination/infection , HIV-1 infection , or tetanus toxoid vaccination , indicates that extensive somatic hypermutation is not required to achieve potent vaccine-specific neutralizing antibody responses .
The quadrivalent HPV ( qHPV ) vaccine provides near-complete protection against sexually transmitted HPV infections that most commonly cause anogenital and oropharyngeal cancers ( HPV types 16 and 18 ) and genital warts ( HPV 6 and 11 ) when administered to individuals naive to these types [1]–[6] . It is thus recommended as an adolescent vaccine or before the onset of sexual activity . The vaccine is comprised of virus-like particles ( VLPs ) assembled from the major capsid L1 protein of each of these four HPV types in alum . Although no correlate of protection has been confirmed for the qHPV vaccine due to low numbers of disease cases in vaccinees [7] , passively transferred immune sera have been shown to be sufficient for protection against papillomavirus challenge in a number of animal models [8]–[10] . These findings suggest that the qHPV vaccine , like most prophylactic vaccines , protects by generating antibody ( Ab ) [11] . There is also evidence that qHPV vaccination elicits plasma cells , which help sustain antigen ( Ag ) -specific Ab levels over time by secreting Ab for extremely long periods; and memory B cells ( Bmem ) , which renew Ab levels by rapidly differentiating into short-lived Ab-secreting plasmablasts upon re-exposure to Ag [12] . Classic plasma cells and Bmem are the products of germinal centers , which are transient structures that develop within secondary lymphoid tissues during a T-cell dependent immune response . It is also within germinal centers that B cell immunoglobulin genes undergo class-switching and somatic hypermutation , and where the resulting B cell receptors , or membrane tethered immunoglobulins , are selected for increased antigen affinity . Evidence that qHPV vaccination elicits both plasma cells and Bmem derives from studies that have observed sustained Ab levels out to 5-years post-vaccination and boosts in Ab responses upon re-vaccination or ex vivo Ag exposure [13]–[15] . However , these studies have never directly identified or characterized HPV-specific Bmem . Such information would not only enable us to evaluate whether there are differences in the quality of B cell memory between different vaccine formulations or schedules , but would also advance our basic understanding of the immunological memory elicited by a highly efficacious vaccine in the absence of pre-existing immunity . The latter is particularly valuable to vaccine development , given that the target populations for candidate HIV-1 or hepatitis C virus vaccines have no pre-existing immunity to these infections . Therefore , we developed an Ag-labeling method that uses fluorescently conjugated HPV 16 pseudoviruses ( psV ) to identify and isolate HPV 16-specific Bmem from the blood of female adolescents and young women who had no pre-existing HPV 16 immunity when they received the qHPV vaccine . Similar Ag-specific labeling approaches have been successfully employed by many groups to identify Bmem ex vivo [16]–[18] . We then cloned Abs from singly sorted HPV 16-psV labeled Bmem in order to evaluate whether these Bmem encoded functional Abs and/or exhibited hallmarks of classical Bmem ( e . g . , somatic hypermutation and class-switching ) . We chose to focus our studies on HPV 16 , as it causes the vast majority of HPV-associated cancers [4] .
In order to identify qHPV-specific Bmem , we generated Alexa Fluor 488 ( AF488 ) -conjugated HPV 16 psV to fluorescently label Ag receptors on the surface of Bmem . These psV are comprised of the L1 and L2 ( major and minor ) capsid proteins of HPV 16 and a reporter plasmid , where L1 is the qHPV vaccine Ag . We chose psV as our ‘bait’ Ag not only because they contain the vaccine Ag in a polyvalent geometry that closely resembles the vaccine VLPs [19] , but also because they could be evaluated for proper folding and function following AF488-conjugation , using a reporter-based neutralization assay [20] . In this assay , psV bound and neutralized by type-specific monoclonal Abs ( mAbs ) are prevented from entering 293TT cells and expressing an encapsidated reporter plasmid . Ab neutralization thus results in a decrease in reporter signal , which can be quantified . Both AF488-conjugated HPV 16 psV ( AF488-HPV 16 ) and unconjugated HPV 16 psV are equally neutralized by an anti-HPV 16 murine mAb ( Fig . S1A ) . This result indicates that AF488-HPV 16 are in the appropriate conformation , and that covalently linked AF488 moieties do not prevent Ab recognition . We observed the same result for AF488-conjugated and unconjugated bovine papillomavirus ( BPV ) psV , which were generated as negative controls for non-specific binding in our flow cytometry and neutralization assays , respectively . In addition , we found that AF488-HPV 16 and AF488-BPV psV exhibit essentially identical fluorescence intensities when applied to 293TT cells in optimized amounts ( Fig . S1B ) . After confirming our AF488-psV reagents , we tested their ability to identify HPV 16-specific Bmem in peripheral blood mononuclear cell ( PBMC ) samples collected at month 7 ( one month post-vaccination ) from 12 qHPV vaccinees . As Ag-specific Bmem are rare [21] , each sample was first enriched for B cells . These B cells were then further divided into two parts and separately stained with a multicolor flow cytometry panel to detect classic CD27+IgD− Bmem [22] , [23] and either AF488-HPV 16 or AF488-BPV . The frequency of AF488+ Bmem in each sample part was then analyzed by flow cytometry , and AF488-HPV16+ Bmem were single cell sorted into 96-well PCR plates . Representative flow cytometry data for 4 of 12 subjects post-vaccination show that AF488-HPV 16+ Bmem were clearly distinguishable from background , despite having very low cell numbers ( only 1–10 cells sorted for each of nine samples; Fig . 1A ) . In contrast , almost no AF488-BPV+ Bmem were observed . Indeed , a significantly higher frequency of AF488-HPV 16+ Bmem than AF488-BPV+ Bmem were detected in all samples ( Fig . 1B ) , thus confirming that AF488-HPV 16+ Bmem were not simply due to non-specific psV binding . Altogether , HPV 16-specific Bmem comprised ∼0 . 2% of total Bmem at one-month post-qHPV vaccination , and 29 AF488-HPV 16+ Bmem were singly sorted from nine different subjects' samples . To evaluate the clonal diversity of HPV 16-specific Bmem elicited by the qHPV vaccine , we cloned full-length variable region sequences from the 29 singly sorted AF488-HPV 16+ Bmem described above . We designed primers that provide complete coverage of all known functional human leader and constant region 5′ exons that flank the variable region , as these regions exhibit low somatic hypermutation rates . The resulting materials and methods clone both the leader and variable regions from the heavy and light chains of single B cells as described in the Methods , Figure S2 , and Table S1 . For comparison , the in silico coverage of other commonly used primer sets that target the leader region are shown in Figures S3 , S4 , S5 . In total , 3 IgA , 15 IgG , and 1 IgM leader-variable regions were amplified from five different subjects' Bmem and confirmed through traditional DNA sequencing , of which approximately half were kappa and half lambda ( Fig . 2A ) . This represents a PCR efficiency of ∼66% ( 19 Abs from 29 Bmem ) . The IgG antibodies were evaluated further , because they could be assessed for expression , folding , and function in vitro . Of these 15 clones , 14 generated high quality sequence reads that could be accurately analyzed for germline V , J , and D gene usage and somatic mutations using IMGT's V-QUEST [24] and IgBlast ( Table 1 ) . We found that these 14 Abs derived from diverse V genes ( Fig . 2B ) and were somatically mutated ( Fig . 2C ) , with an average of 33±19 NT mutations and 19±9 AA changes per clone . In cases where more than one IgG clone utilized a given V gene , it should be noted that the clones were derived from different subjects . Moreover , no clonally related IgG ( or IgA , IgM ) were observed in any subject , potentially due to low cell numbers . Heavy , kappa , and lambda chain sequences corresponding to 12 human monoclonal Abs ( mAbs ) were successfully cloned into their respective AbVec vectors using gene-specific forward and reverse cloning primers . Ab 1 was isolated from subject EK1006; Abs 2 , 3 , and 4 from EK1027; Ab 5 from EK1042; Ab 6 from EK1377; Abs 8 , 12 , and 13 from EK1073; and Abs 16 , 17 , and 18 from EK1078 . We screened the resulting clones to identify those with 100% homology to the original PCR product , or , in rare cases where the original PCR product sequence was ambiguous , we used these clones to derive a consensus sequence . After co-transfecting corresponding heavy and light chain vectors into 293F cells to produce full-length IgG1 , we found that IgG1 expressed with their native leader sequences gave comparable yields as an irrelevant IgG1 expressed with the murine leader encoded by the AbVec vectors ( Fig . S6 ) . In addition , we found that Ab 4 did not express , as predicted by V-QUEST . To demonstrate that the low number of AF488-HPV 16+ Bmem found in the subjects' samples expressed functional Abs that were , in fact , HPV 16-specific , we evaluated our human mAbs in the aforementioned neutralization assay . We also included murine mAbs H16 . V5 and 5B6 as well-characterized positive controls for HPV 16 and BPV neutralization , respectively [25]–[27] . A series of dilutions of each Ab were tested against both unconjugated HPV 16 psV ( pink solid lines ) and BPV psV ( black solid lines ) ( Fig . 3A ) . From the neutralization curves it is immediately apparent that none of our human Abs neutralized BPV psV . In contrast , most of the human Abs neutralized HPV 16 psV with potent IC50 values ranging from 0 . 44 pM–2 . 98 nM ( Fig . 3B ) . This result thus demonstrates the utility of our Ag-labeling approach for identifying HPV 16-specific Bmem . Four of the twelve human mAbs ( 5 , 13 , 16 , and 18 ) did not neutralize HPV 16 even when tested at dilutions up to 450 nM . Therefore , to learn whether there are differences in the abilities of these Abs to bind vs . neutralize HPV 16 , as has been observed for Abs elicited against other viruses ( e . g . , HIV-1 ) , we evaluated their binding to psV in an ELISA . In this assay , antibodies in serum or plasma are tested for binding to psV immobilized on the surface of polystyrene microtiter plates . Antibody binding is then detected and quantified using a secondary antibody against human IgG . Similar to the neutralization results , we found that Abs 5 , 13 , 16 , and 18 did not bind to HPV 16 L1 in the ELISA , even when tested at concentrations up to 100 nM ( Fig . 4A ) . Therefore , it appears that these Abs are not HPV 16-specific and that the AF488-HPV 16+ Bmem from which they derived represented background or false positives . Such non-specific staining has been observed with other Ag-labeling methods [28] . We also found that the EC50 values for binding were higher than the IC50 values for neutralization , indicating that Ab affinities were lower in the ELISA than in the neutralization assay ( Fig . 4B ) . Lower apparent Ab affinities in the ELISA than in the neutralization assay have also been observed for the anti-HPV 16 murine mAbs [25] , [29] . This discrepancy was subsequently suggested to result from the ELISA conditions not satisfying the law of mass action , unlike the neutralization assay; therefore , the ELISA may be a less accurate measure of Ab affinity [25] , [30] .
In order to characterize Bmem elicited by the qHPV vaccine , we first needed to establish an Ag-labeling method to identify qHPV-specific Bmem . The AF488-psV-labeling method described here both successfully and selectively identifies Ag-experienced HPV 16-specific Bmem as evidenced by the findings that significantly more AF488-HPV 16+ than AF488-BPV+ Bmem were observed in the post-vaccination samples , and that Abs cloned from these cells potently neutralized HPV 16 psV . Such Ag-specificity is perhaps not surprising , given that HPV Ab responses are almost uniformly type-specific , and HPV 16 L1 only shares ∼77% amino acid identity with BPV L1 [31] , [32] . Still , the use of a negative control ( AF488-BPV ) was likely instrumental in our high success rate for identifying HPV 16-specific Bmem . Taken all together , these results show that we can both find and characterize HPV 16-specific Bmem in qHPV vaccinees . The observed frequency of HPV 16-specific Bmem is within the range of Ag-specific Bmem frequencies observed by other studies that have used Ag-labeling [33] , [34] . However , it is nearly an order of magnitude lower than the frequency of HPV 16-specific Ab-secreting cells reported for these samples by Smolen et al . [35] . This discrepancy may be attributed to differences in both methodology and sampling . Specifically , Smolen et al . utilized an ELISPOT assay to quantify HPV 16-specific IgG secreting cells , and ELISPOT has been shown to be a more sensitive method for detecting Ag-specific responses than Ag-labeling [36] . It is also possible that we unintentionally excluded some HPV 16-specific Bmem by positively gating for singlet cells . In addition , the authors evaluated more samples than we did ( 166 vs . 12 samples ) ; therefore , the frequency of HPV 16-specific Bmem reported here may actually underrepresent the total vaccine response . We recognize that the low numbers of HPV 16-specific Bmem observed in these samples may also have limited our findings . These low numbers derive in part from the low number of starting cells in each sample ( 7 . 8×106±2 . 3×106 PBMCs ) and in part from diminished viability of these samples . In fact , based upon the observed average number of PBMCs , B cell frequency ( ∼5% ) , B cell purity ( ∼90% ) , sample viability ( ∼80% ) , Bmem frequency ( ∼10% ) , and frequency of HPV 16-specific Bmem ( ∼0 . 2% ) , we anticipated detecting at most 25 cells per sample . We also needed to establish methods for cloning the Ag-specific receptors , or Abs , from these cells . We opted to clone Abs from single B cells in order to retain heavy and light chain pairing information . For the latter , we sought to identify primers that would amplify heavy and light chain variable regions with minimal bias by targeting region that exhibit low somatic hypermutation rates . However , in our search we discovered that many such published primer sets had gaps in coverage , which are particularly notable in the case of the lambda chain primers and have resulted in documented amplification biases [37] , [38] . Therefore , we developed new heavy , kappa , and lambda chain primers to provide complete coverage of known human leader and constant region sequences . Although we have not directly compared all of these primer sets and their associated methods , we find that the efficiency of our primers and method for amplifying paired Ab sequences from Bmem ( ∼70% ) is greater than that reported by other groups ( ∼20–50% ) , all of which are lower than the efficiencies reported for amplifying Ab sequences from plasmablasts ( ≥80% ) [34] , [39] . Such gross differences in amplification efficiencies likely reflect the disparate levels of Ab transcripts in these B cell subsets [40] . Isolated AF488-HPV 16+ Bmem were predominantly IgG ( 79% ) , but also included IgA ( 16% ) and IgM ( 5% ) Bmem . Although we did not observe any somatic variants in this study , Huo et al . have described the appearance of both HPV 16-specific IgA and IgG Ab-secreting cells ( i . e . , plasmablasts ) in blood following qHPV vaccination [41] . Therefore , it will be interesting to see if we find clonally related HPV 16-specific IgA and IgG in our future work . We also found that the IgG clones were somatically mutated ( ∼33 NT mutations and ∼19 AA changes from germline , on average ) and utilized diverse germline V genes . This level of Ab somatic hypermutation is not remarkably high . Indeed , it is more similar to the level observed for IgG cloned from healthy donors , influenza vaccinees , or tetanus toxoid ( TT ) vaccinees than the broadly neutralizing anti-HIV-1 Abs described in the literature [34] , [37] , [42]–[45] . The former studies found an average of 18 . 0 and <10 NT mutations in the heavy and light chains of healthy donor IgG+ B cells; 25 . 3 and 16 . 2 NT mutations in the heavy and light chains of TT-specific Bmem; and 19 . 4 NT mutations in the heavy chain of plasmablasts responding to seasonal influenza vaccination . Moreover , Klein et al . have shown that anti-HIV-1 Abs with limited neutralization breadth have ∼40 NT mutations and ∼24 AA changes on average . In contrast , the most potent and broadly neutralizing anti-HIV-1 Abs have ∼118 NT mutations and ∼56 AA changes [46] . Our data thus appear to corroborate an emerging trend that extensive somatic hypermutation is not required to achieve potent , albeit narrow or type-specific , neutralizing Abs responses . Such differences may also reflect differences in antigen immunogenicity and/or the cumulative Ag- and germinal center-experience of the Bmem . Indeed , we know that broadly neutralizing Ab responses develop later in HIV-1-infected individuals than isolate-specific responses [47] . The correlation between somatic hypermutation levels and neutralization breadth may also indirectly relate to diverse V gene usage as opposed to the convergent V usage reported for both broadly neutralizing HIV-1 and influenza Abs [37] , [44] , [48]–[50] . Namely , there are many solutions ( germline V genes ) to a simple problem ( neutralizing one virus ) ; but fewer solutions to a more complex one ( e . g . , V genes that can tolerate framework mutations , which appear to be key for neutralizing many different viruses; [46] ) . It will thus be interesting to see if there are any differences between the extent of somatic hypermutation or V gene usage in Bmem repertories elicited by vaccination versus that observed in Bmem repertories generated following natural HPV infection . Finally , it should be noted that we do not know whether the Abs expressed by these HPV 16-specific Bmem are present in serum . This may be important if pre-existing Bmem are not able to detect HPV and generate a de novo Ab response in time to neutralize the incoming virus . The kinetics of HPV infection occurs in hours [51] , [52] , whereas Bmem to plasmablast differentiation takes 4–8 days post-vaccination when measured in peripheral blood [39] , [45] , [53] . Therefore , it would appear that the concentration and affinity ( i . e . , composition ) of pre-existing Abs may be more likely determinants of protection against HPV infection . At the same time , Bmem to plasmablast differentiation appears to occur more rapidly for certain vaccines ( e . g . , antibodies to the hepatitis B virus ( HBV ) surface Ag have been detected in sera at 3–4 days post-HBV vaccine boosting [54] ) and may occur more rapidly at the site of infection or in draining lymph node ( s ) than can be measured systemically . A recently published study may provide some indication of the overlap between Bmem and serolological Ab repertoires: Lavinder et al . showed that in the case of the tetanus toxoid ( TT ) vaccine , only a minor fraction of the Bmem and plasmablast Ab repertoires at day 7 post-boosting ( <1% and <5% respectively ) comprise the Ag-specific serological Ab repertoire found using proteomics [55] . This study thus indicates that not all plasmablast and/or Bmem clones are selected ( or survive ) to populate the bone marrow niche . Two potential confounders of this interpretation , however , were that Bmem analyzed in this study were not selected for Ag-specificity and were not sampled at the peak of the Bmem response , which occurs on day 14 post-TT boosting [39] . Frölich et al . assessed the kinetics and repertories of both TT-specific Bmem and TT-specific plasmablasts and found that the composition of their Ab repertoires were highly similar . Taken together , these two studies suggest that a low frequency of the Ag-specific Bmem repertoire ( <5% ) makes it into the serological Ab repertoire . Similarly , Purtha et al . showed that the Ab specificities of Bmem following west nile virus infection in mice were broader than the Ab specificities expressed by plasma cells or found in serum . However , these authors did not show the frequency of Bmem clones among the plasma cell and/or serological repertoires . Similar studies comparing Bmem and serological repertoires are warranted for other vaccines and infections . For although the TT vaccine is highly efficacious , its Ab levels decline steadily over time , suggesting that it does not elicit the same quality of long-lived plasma cells as the smallpox or yellow fever vaccines , for example [56] .
De-identified PBMC samples were collected as part of a phase III , post-licensure , randomized , controlled , multi-center trial ( NIH registry number NCT00501137 ) to compare the immunogenicity of a reduced dose qHPV vaccine schedule to that of the licensed 3-dose qHPV vaccine [57] . This trial included three study groups: female adolescents ( aged 9–13 years ) who received a 2-dose qHPV vaccine at months 0 and 6; female adolescents who received the 3-dose qHPV vaccine at months 0 , 2 , and 6; and young women ( aged 16–26 years ) who received the 3-dose qHPV vaccine . PBMC samples were isolated from pre-vaccination and post-vaccination blood draws as described [35] . HPV 16 and BPV psV were generated using plasmids and protocols described on the National Cancer Institute ( NCI ) 's Laboratory of Cellular Oncology ( LCO ) website ( http://home . ccr . cancer . gov/lco/default . asp ) with the following modifications: 45 million 293TT cells ( NCI Developmental Therapeutics Program ) were seeded overnight in each of three 225 cm2 tissue culture flasks ( Corning; Sigma-Aldrich , St . Louis , Missouri ) with 60 ml of DMEM-10 [high-glucose DMEM ( Life Technologies , Carlsbad , California ) supplemented with 10% FBS ( Gemini Bioproducts , West Sacramento , California ) , L-glutamine , and 1× non-essential amino acids ( Life Technologies ) ] . The next morning flasks between 70–90% confluency were each co-transfected with 57 µg of pYSEAP ( plasmid expressing secreted alkaline phosphatase , or SEAP ) and either 57 µg of p16L1L2 ( plasmid expressing HPV 16 L1 and L2 ) or pBPVL1L2 ( plasmid expressing BPV L1 and L2 ) using 255 µl of Lipofectamine2000 transfection reagent ( Life Technologies ) . After ∼72 hours , cells were recovered by centrifugation and transferred to a siliconized eppendorf tube in phosphate buffered saline ( PBS ) -Mg buffer [PBS supplemented with 1× antibiotic-antimycotic ( Life Technologies ) and 9 . 5 mM MgCl2] . Cells were then re-suspended in PBS-Mg solution ( 1 . 5× the volume of cells ) , 0 . 5% Triton X-100 ( Thermo Fisher Scientific , Rockford , Illinois ) , 40 mM sodium phosphate buffer pH 7 . 5 , and 20 µg/ml RNAse A; and incubated for 24 hours at 37°C . Lysate was stored at −80°C until use . AF488 dye ( Life Technologies ) was re-suspended in 200 µl molecular biology grade ( MB ) water , aliquoted in eppendorf tubes , lyophilized , and stored at −20°C until use . PsV were conjugated with AF488 using protocols described on the LCO website with the following specifications: Lysates prepared above were clarified by centrifugation ( 10 min at 5 , 000× g , 4°C ) , transferred into clean siliconized eppendorf tubes , re-clarified by centrifugation , and then pooled in polypropylene tubes . Lysate was diluted to 3 . 5 mg/ml protein in PBS and divided into 1 ml fractions in clean siliconized eppendorf tubes containing 100 µl of 1M sodium bicarbonate buffer , pH 8 . 5 . Pre-aliquoted AF488 dye was thoroughly re-suspended in DMSO at 10 mg/ml and then 34 µg was added to each tube of diluted lysate and immediately vortexed . Samples were rotated end-over-end for one hour at room temperature while protected from light . Each tube of lysate was then brought to neutral pH with 40 µl of 1M sodium phosphate buffer , pH 6 . 5 . Pooled tubes of neutral conjugated lysate were purified by density gradient ultracentrifugation using layered gradients of 27% , 33% , and 39% Optiprep ( Sigma-Aldrich ) and PBS as a diluent , as previously described [20] . Fractions collected following gradient ultracentrifugation were screened in duplicate for the presence of psV using a direct ELISA with H16 . V5 and 5B6 ( Methods S1 ) . Fractions containing the highest signals above background were pooled , aliquoted , and stored at −80°C until use . To determine the relative amounts of L1 , and thus psV , in each AF488-HPV 16 and AF488-BPV stock , sample aliquots were reduced with 6 . 8% ( v/v ) 2-mercaptoethanol ( Sigma-Aldrich ) for 5 minutes at 100°C and separated on SDS-PAGE gels . Gels were then stained with Coomassie blue and L1 band ( ∼55 kDa ) intensities quantified using ImageJ [58] . The average ratio of AF488-HPV 16:AF488-BPV L1 band intensities from duplicate gels was used to normalize AF488-psV amounts for subsequent flow cytometry experiments . AF488-HPV 16 and AF488-BPV were separately titrated on 293TT cells to identify optimal amounts of fluorescent Ag for B cell staining . PBMCs were quickly thawed in pre-warmed , heat inactivated FBS , re-suspended in ice-cold MACS separation buffer ( Milteyni Biotec , Auburn , California ) , counted , and enriched for B cells using the B Cell Isolation Kit II ( Milteyni Biotec ) . Enriched B cells were then washed and re-suspended in PBS , divided in half , and stained with Live/Dead Violet dead cell dye ( Life Technologies ) for 30 minutes [59] . Each of the two samples were then washed and re-suspended in 2% FBS-PBS solution and stained with either AF488-HPV 16 psV or AF488-BPV psV , as well as the following fluorescent mAbs for 30 minutes: anti-CD38 APC and anti-IgD PE ( Milteyni Biotec ) , anti-CD3 V500 , anti-CD19 APC-Cy7 , anti-CD27 PE-Cy7 , and anti-CD20 PerCP-Cy5 . 5 ( BD Biosciences , San Jose , California ) . All staining was conducted with optimized amounts of staining reagents , cells on ice , and minimal light exposure . Stained cells were washed and re-suspended in 2% FBS-PBS , maintained on ice , and protected from light until fluorescence activated cell sorting ( FACS ) . FACS was conducted using an Aria II ( BD Biosciences ) in single cell sort mode . Just prior to FACS , sort buffer was prepared using a modified formulation of the buffer described by Wardemann et al . : 0 . 425× RNase-free PBS ( Life Technologies ) , 10 mM dithiothreitol ( dTT ) , and 16 U RNasin ( Promega , Madison , Wisconsin ) [60] . Eight µl of this buffer were added to each well of a 96-well AB-2800 PCR plate ( Thermo Fisher Scientific ) , sealed with adhesive PCR films ( Thermo Fisher Scientific ) , and kept on ice until sorting . Cells were gated for size ( SSC-A vs . FSC-A ) , to exclude doublets ( SSC-W vs . SSC-H; FSC-W vs . FSC-H ) and dead cells ( Live/Dead Violet− ) , and to include Bmem ( CD3−CD19+CD20+CD27+IgD− ) with a high AF488-HPV 16+ fluorescence intensity above background ( AF488− cells in samples stained with AF488-BPV ) . For each well containing a single sorted cell , an equal number of wells were kept empty as non-template controls . Immediately after sorting , plates were sealed with foil PCR films ( Thermo Fisher Scientific ) , placed onto dry ice , and stored at −80°C . Sort plates were thawed on ice and briefly centrifuged before adding 11 . 7 µl per well of ice-cold RT-PCR master mix containing 5 . 8 µl nuclease free water , 4 . 8 µl 5× FS buffer ( Life Technologies ) , 1 µl 25 mM dNTP mix ( Roche Applied Science , Indianapolis , Indiana ) , and 0 . 1 µl random primers ( Life Technologies , catalog no . 48190-011 ) per well . Plates were again briefly centrifuged , incubated at 65°C for 5 minutes , and returned to ice . Then 8 . 3 µl of ice-cold RT-PCR master mix ( 2 ) containing 4 . 8 µl nuclease-free water , 0 . 8 µl 5× FS buffer , 0 . 2 µl RNasin , 2 µl 0 . 1M dTT , and 0 . 5 µl SuperScript III RT ( Life Technologies ) were added per well , and cDNA was generated with the following PCR program: 1 cycle for 5 minutes at 25°C , 1 cycle for 60 minutes at 50°C , 1 cycle for 15 minutes at 70°C , and 4°C forever . cDNA was stored at 4°C ( short-term ) or −20°C ( long-term ) . In order to amplify full-length variable regions from singly sorted AF488-HPV 16+ Bmem with minimal bias , we designed forward amplification primers that anneal within the leader region , which precedes the variable region , and reverse amplification primers that anneal within the constant region , which follows the variable region and possesses substantially less germline and somatic variation than the variable region , as its name implies . In the case of the forward primers , we specifically designed them to anneal within the first of two exons in the Ab leader region , as the 5′ boundary of somatic hypermutation lies within the leader intron [61]–[64] . Where possible , up to two degenerate bases were used to minimize the number of primers in each set . Primers were evenly distributed across primer sets on the basis of similar melting temperatures ( within ∼3°C of variation , according to the default analysis settings of Integrated DNA Technologies' OligoAnalyzer 3 . 1 , Coralville , Iowa ) and minimal primer-primer interactions ( delta G≤−10 kcal/mol , base pairs ≤4; OligoAnalyzer 3 . 1 ) . The resulting primer sets and all other primers used in this study are provided in Table S1 . For example , we also designed forward and reverse cloning primers based upon the above amplification primers , so that the amplification PCR products could be recombinantly expressed as full-length Abs using the AbVec vectors [65] . Importantly , our forward amplification primers possess complete homology to all known functional human leader sequences for the heavy , kappa , and lambda chains from IMGT ( Fig . S2 ) . Our reverse primers also possess complete homology to all known human IgA , IgG , IgM , kappa , and lambda constant region alleles . To amplify full-length variable regions from the bulk cDNA generated above , six forward primer sets based on the sequences of all known functional Ab leaders were used: three for the heavy chain , two for the lambda chain , and one for the kappa chain ( details of set-up are also included in Table S1 ) . Each forward primer set was added separately to a given master mix from the reverse primer and other PCR components . The final concentration of the forward primer set within the master mix was 0 . 5 µM . Some of the forward primer sets contain degenerate primers . In this case , each degenerate primer was treated as n primer parts , where n represents the number of degenerate bases . For example , if a primer contains only one degenerate position , and that position is a ‘D’ nucleotide ( D = A , C , or G ) , such a degenerate primer contains three degenerate bases and would be treated as three primer parts . There were equal nmoles of each primer part in the final master mix . The reverse primer was added in 3-fold excess of this amount ( i . e . , 3-fold the nmoles of each primer part ) . In addition to the variable volumes of the forward primer set and reverse primer , the other PCR components per PCR plate well included: 3 µl of cDNA , 4 . 375 µl HotStar Taq Plus buffer , 2 . 5 µl MgCl2 , 0 . 5 µl 25 mM dNTP mix , 0 . 4375 µl HotStar Taq Plus DNA Polymerase ( Qiagen , Valencia , California ) , and nuclease-free water to 40 µl . The PCR program used was: 1 cycle for 5 minutes at 95°C; 50 cycles at 94°C for 30 seconds , 54–59°C for 30 seconds ( as indicated in Table S1 ) , and 72°C for 55 seconds; 1 cycle at 72°C for 10 minutes , and 4°C forever . The IgG heavy and kappa chain PCR reactions were conducted first , and the resulting products analyzed on DNA agarose gels for the presence of ∼500 or ∼630 bp fragments , respectively . If no PCR product was observed for a given well or wells with the kappa chain primer set , additional PCR reactions were carried out for these wells using the two lambda chain primer sets ( ∼470 bp band expected ) . PCR products in wells with clearly visible Ab bands were then purified from the other PCR components using a QIAquick PCR purification kit ( Qiagen ) , resuspended in MB water , and submitted for traditional DNA sequencing with the described sequencing primers . After the variable region sequences of these products were confirmed and assigned to their respective germline genes using IgBlast and V-QUEST , single gene-specific forward and reverse cloning primers were employed to re-amplify these variable regions for molecular cloning , using the same PCR conditions described above and 1 pg of template . The resulting products were again purified from other PCR components and resuspended in MB water . Abvec-hIgG ( FJ475055 . 1 ) , Abvec-hIglambda ( FJ517647 . 1 ) , and Abvec-hIgKappa ( FJ475056 . 1 ) vectors were kindly provided by Patrick Wilson's lab ( University of Chicago ) . Similar to other Ab expression vectors , the AbVec vectors encode IgG1 , Igκ1 , and Igλ2 constant regions in-frame and downstream of cloning sites for exogenous variable regions , as well as a murine leader in-frame and upstream of the variable region cloning sites . In their current configurations , these cloning sites trim off the very 5′ , and in some cases also the 3′ , end of inserted variable regions . However , as we wished to preserve the entire variable region sequence , we utilized alternative cloning sites and introduced a new cloning site in AbVec-hIgKappa at bps 1417–18 using site-directed mutagenesis . These sites allow both the leader and variable region sequences to be cloned into these vectors , including the first 4 , 33 , and 14 amino acids of the corresponding IgG CH1 , Cκ , or Cλ . The leader-specific forward primers and IgG/κ/λ specific reverse primers for the cloning PCR reactions thus contain restriction enzyme sites corresponding to these alternative 5′/3′ cloning sites , which for AbVec-hIgG1 , AbVec-hIgKappa , and AbVec-hIgLambda correspond to EcoRI/ApaI , EcoRI/XhoI , and EcoRI/XhoI , respectively . It should be noted that only in one case does the inserted coding sequence of the constant region allele ( IGLC1*02 ) upstream of the cloning site differ from that of the vector allele ( IGLC2*02 ) . However , we have confirmed that in this case , the altered AAs do not alter the expression , folding , or neutralization potency of the resulting hybrid Ab clone when compared to ‘corrected’ clone ( i . e . , a clone where the constant region coding sequence upstream of the cloning site was reverted to the vector coding sequence; Fig . S7 ) . Heavy , kappa , and lambda chain PCR products generated above were inserted into each respective AbVec vector using traditional cloning techniques , and the resulting colonies were screened for the correct insert size and sequence . In-Fusion cloning technology ( Clontech , Mountain View , California ) was utilized for sequences containing internal restriction sites . To generate full-length Abs from the above cloning products , 15 µg of each heavy and light chain vector were co-transfected into Freestyle 293-F cells according to the manufacturer's protocols ( Life Technologies ) . Cell supernatants were collected by centrifugation 72 hours post-transfection . IgG1 were purified from sterile-filtered supernatants using 0 . 3 ml of protein G agarose ( 0 . 6 ml slurry ) and 0 . 5–2 . 0 ml capacity disposable plastic columns according to manufacturer's recommendations ( Thermo Fisher Scientific ) , except that PBS was used as a binding buffer; 0 . 1M citric acid , pH 3 . 0 was used as an elution buffer; and 1M Tris base , pH 9 . 0 was used as a neutralization buffer . Following dialysis in PBS , Abs were sterile-filtered and stored at 4°C . The 293TT psV neutralization assay was conducted using protocols described on the LCO website , with the following specifications: Unconjugated HPV 16 and BPV psV were generated and purified as described for AF488-psV above , omitting the dilution of clarified lysate and AF488 conjugation steps . Four to six hours before the addition of Ab and psV mixtures to cells , 30 , 000 293TT cells were seeded in the inner 60 wells of a 96-well flat bottom tissue culture plate in 100 µl of DMEM-10++ [DMEM-10 supplemented with 1× pen strep ( Gemini Bioproducts ) and 400 µg/ml hygromycin B ( Mediatech , Manassas , Virginia ) ] . The outer 36 wells were filled with 200 µl DMEM-10+ ( DMEM-10 supplemented with 1× pen strep ) , and plates were returned to 37°C . Approximately two hours before Ab and psV mixtures were added to cells , three-fold Ab dilution series were prepared in triplicate in DMEM-10++ in 96-well polypropylene plates on ice , including H16 . V5 and 5B6 . Initially , the starting concentration of purified human mAbs was 167 nM ( 25 µg/ml ) per well , but after subsequent experiments showed that these Abs were extremely potent , starting human mAb concentrations were reduced to 50 nM . PsV were also diluted in DMEM-10++ to a pre-determined titer ( i . e . , the titer at which the same assay conducted without Ab was within the linear range of the assay ) . Twenty-four µl of pre-diluted psV and 96 µl of pre-diluted Ab were then mixed in the wells of a 96-well polypropylene plate on ice . Additional controls included wells without Ab or psV ( ‘background’ wells ) and wells without Ab that had psV ( ‘no Ab’ wells ) . All plates were incubated at RT for 1 hour before transferring to 293TT cells and then all were incubated for an additional 68 hours at 37°C . At the end of this incubation period , 30 µl of supernatant were removed from each well , transferred to an Immulon 2 HB plate containing 100 µl of AP substrate [4 . 3 mg/ml Sigma 104 phosphatase substrate ( Sigma-Aldrich ) in 100 mM sodium bicarbonate buffer , 10 mM magnesium chloride , pH 9 . 5] , and incubated 30 minutes at RT . Absorbance was read at 405 nm . Sample signals were corrected for background and percent neutralization determined , where percent neutralization = [ ( mean signalno Ab wells−signalAb wells ) / ( mean signalno Ab wells ) ]*100 . HPV 16 psV were incubated on Immunlon II plates ( Thermo Fisher Scientific ) overnight at 4°C in PBS . After washing 3× with PBS , plates were blocked for 1 hour with blocker ( PBS plus 0 . 05% Tween-20 , 2% non-fat dry milk ) . Human Abs were diluted in blocking buffer starting at 100 nM or 10 nM , followed by 1∶3 serial dilutions . Plates were incubated at room temperature with gentle shaking for 1 hour followed by washes . Alkaline phosphatase-conjugated goat anti-human IgG , Fcγ specific ( Jackson ImmunoResearch Labs , Inc . , West Grove , PA; 1∶1000 dilution in blocking buffer ) , was added to all wells and incubated as before . After washing , plates were developed by the addition of AP substrate and read at 30 minutes at 405 nm . Reagents were added to the plate in 50 µl volumes except for blocker and developer ( 200 µl ) . Leader-variable region sequences obtained by primer-specific PCR amplification of heavy and light chain cDNA and subsequent DNA sequencing were submitted to V-QUEST and IgBlast using the default settings , except that for V-QUEST , the number of diversity ( D ) genes accepted was increased to three . If there was a discrepancy between V gene assignments or number of nucleotide ( NT ) mutations/amino acid ( AA ) changes , upon closer inspection V-QUEST typically identified more closely related germline sequences than IgBlast . It should be noted that NT mutations and AA changes in the V gene that overlapped with ‘n’ nucleotides or a given D or joining ( J ) gene were not counted in the somatic hypermutation statistics . Flow cytometry data were analyzed and plotted using FlowJo software ( Tree Star , Ashland , Oregon ) . All other graphs were plotted and prepared using Prism , including any statistical analyses ( GraphPad Software , La Jolla , California ) . IC50 neutralization values were determined using Prism's non-linear regression analysis , specifically the dose-response inhibition model and log ( inhibitor ) vs . response equation . EC50 binding values were determined using the sigmoidal ( variable slope ) equation .
|
There is an urgent need to better understand how to reliably generate effective vaccines , particularly subunit vaccines , as certain pathogens are considered to pose too great of a safety risk to be developed as live , attenuated or killed vaccines ( e . g . , HIV-1 ) . The human papillomavirus ( HPV ) vaccines are two of the most effective subunit vaccines ever developed and have continued to show protection against HPV associated disease up to and beyond five years post-vaccination . Moreover , the target population for these vaccines have essentially no pre-existing immunity to the HPV types covered by the vaccine; therefore , these vaccines provide an excellent model for studying the immunity elicited by a highly effective subunit vaccine . As the HPV vaccines , like most vaccines , protect by generating antibodies , we are interested in characterizing the memory B cells elicited by the HPV vaccine . Memory B cells help to sustain antibody levels over time by rapidly differentiating into antibody secreting cells upon pathogen re-exposure . Although previous studies have provided evidence that the HPV vaccines elicit memory B cells , they did not characterize these cells . Here , we have isolated HPV-specific memory B cells from adolescent females and women who received the quadrivalent HPV vaccine and have cloned antibodies from these cells . Importantly , we find that these antibodies potently inhibit HPV and that the memory B cells from which they derive exhibit hallmarks of long-lived memory B cells .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"blood",
"cells",
"humoral",
"immunity",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"immunology",
"vaccines",
"infectious",
"disease",
"immunology",
"sexually",
"transmitted",
"diseases",
"class",
"switching",
"vaccination",
"and",
"immunization",
"antibody",
"response",
"infectious",
"diseases",
"white",
"blood",
"cells",
"human",
"papillomavirus",
"infection",
"animal",
"cells",
"clonal",
"selection",
"immune",
"response",
"antibody-producing",
"cells",
"cancer",
"vaccines",
"cell",
"biology",
"b",
"cells",
"clinical",
"immunology",
"immunity",
"antigen",
"processing",
"and",
"recognition",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"viral",
"diseases",
"humoral",
"immune",
"response",
"recombinant",
"vaccines"
] |
2014
|
Characteristics of Memory B Cells Elicited by a Highly Efficacious HPV Vaccine in Subjects with No Pre-existing Immunity
|
The snake Bothrops atrox is responsible for the majority of envenomings in the northern region of South America . Severe local effects , including hemorrhage , which are mainly caused by snake venom metalloproteinases ( SVMPs ) , are not fully neutralized by conventional serum therapy . Little is known about the immunochemistry of the P-I SVMPs since few monoclonal antibodies ( mAbs ) against these molecules have been obtained . In addition , producing toxin-neutralizing mAbs remains very challenging . Here , we report on the set-up of a functional screening based on a synthetic peptide used as a biosensor to select neutralizing mAbs against SVMPs and the successful production of neutralizing mAbs against Atroxlysin-I ( Atr-I ) , a P-I SVMP from B . atrox . Hybridomas producing supernatants with inhibitory effect against the proteolytic activity of Atr-I towards the FRET peptide Abz-LVEALYQ-EDDnp were selected . Six IgG1 Mabs were obtained ( named mAbatr1 to mAbatr6 ) and also two IgM . mAbatrs1 , 2 , 3 and 6 were purified . All showed a high specific reactivity , recognizing only Atr-I and B . atrox venom in ELISA and a high affinity , showing equilibrium constants in the nM range for Atr-I . These mAbatrs were not able to bind to Atr-I overlapping peptides , suggesting that they recognize conformational epitopes . For the first time a functional screening based on a synthetic biosensor was successfully used for the selection of neutralizing mAbs against SVMPs .
Snakebites cause up to 1 , 800 000 envenomations per year , mainly in tropical areas [1]–[4] . Snakebites might be considered as a daily occupational hazard since rural subsistent farming communities are the main population suffering from this condition [3] , [4] , considered as a Neglected Tropical Condition by WHO ( World Health Organization ) since 2008 [1] . In Brazil , nearly 30 , 000 snakebite envenomings occur per year and the incidence is about 14 cases/100 , 000 people/year , a number as high as those found in many other Latin American countries [1] , [5]–[8] . Moreover , in the Brazilian Amazon region , 9 , 000 snakebites occur per year with an incidence fourfold higher than that found in the rest of Brazil . Bothrops atrox is found in tropical lowlands and rainforests in the north of South America and is responsible for the majority of envenomations in this area , causing approximately 80% of snake bites [8]–[10] . B . atrox envenoming is characterized systemically by headache , severe coagulopathy , with consumption of coagulation blood factors , generalized hemorrhage and renal failure . Locally , severe tissue lesions may be observed , including swelling , blisters , inflammatory response , erythema , ecchymosis , local hemorrhage and necrosis [11] , [12] . Immunotherapy by antivenoms is the only efficacious treatment approved by WHO for snakebite accidents . Antivenoms are produced by hyper immunization of animals ( generally horses , sheeps or goats ) with a pool of venoms from the most important species of snakes found in each country/region [13] . It is known that serum therapy is effective against several of the systemic noxious effects of snake envenomings , when administered early enough [14]–[16] . However , the local effects are not fully neutralized , being clinically important [17]–[20] due to complications related to local hemorrhage and tissue necrosis that can permanently provoke a disability and morbidity among patients , causing a very important socio-economic impact [21] . B . atrox venom is a rich mixture of bioactive components belonging to few protein families [22]–[24] . Proteomic characterization of toxin composition of B . atrox venom used in this study indicates that the main components of this venom are represented by SVMPs ( Snake Venom Metalloproteinases ) ( 58 . 2% ) , including P-III and P-I classes , SVSP ( Snake Venom Serine Proteinases ) ( 11 . 17% ) , PLA2 ( Phospholipase A2 ) ( 11 . 0% ) and others [25] . Although these molecules act synergistically in a typical “pit viper envenoming” clinical picture , it is well established that SVMPs are responsible for the most severe local effects ( i . e . hemorrhage and its variable consequences ) [11] , [26]–[31] . SVMPs are zinc-dependent proteinases representing up to 70% of venom dry weight , and can be classified into three classes ( P-I to P-III ) and several subclasses , according to their domain organization [22]–[24] , [32]–[35] . The P-I class are endopeptidases possessing the metalloproteinase catalytic domain only . The P-II class presents both metalloproteinase and disintegrin domains and the P-III class SVMPs contain disintegrin-like and cysteine-rich domains , in addition to the proteinase domain . Although no P-III class SVMPs from B . atrox have yet been characterized at the protein level , evidence supporting the presence of this class of enzymes has been provided by proteomic and transcriptomic studies [23]–[25] , [36] . In addition , three P-I class SVMPs from B . atrox venom have already been purified and characterized [37]–[39] . Atr-I ( Atroxlysin-I ) , Batx-I and Batroxase are P-I enzymes isolated from B . atrox venom from the Amazonian regions of Peru , Colombia and Brazil , respectively . They are hemorrhagic and fibrinogenolytic and do not bear any pro-coagulant activity . These molecules are hemorrhagins that can act proteolytically upon extracellular matrix components , contributing to the local damages following bite . They can also play a systemic role , causing myotoxic effects [38] and inhibition of platelet aggregation [37] , which can contribute to hemorrhagic , necrotic and blood-clotting disturbances [37]–[39] . Considering the important role played by P-I class SVMPs in the B . atrox poisoning , polyclonal and monoclonal antibodies against these molecules constitute highly useful tools to investigate the structural determinants of toxicity , for patient diagnosis and may have great potential for the preparation of more efficient antivenoms for passive immunotherapy or even for vaccination . Thus , in this study we used a functional screening based on a synthetic biosensor to produce neutralizing mAbs against Atr-I . Specificity assays of Atr-I using oxidized insulin B-chain as substrate showed that the enzyme cleaves the Ala14-Leu15 peptide bond [37] . Inhibition of enzymatic-dependent cleavage of the Atr-I peptide substrate [Abz-LVEALYQ-EDDnp , containing a FRET ( Fluorescence Resonance Energy Transfer ) system] , determined by means of fluorescence emission , was used as a tool to select specific monoclonal antibodies . Four monoclonal antibodies that fully neutralize the proteolytic and hemorrhagic in vivo activities of Atr-I and crude B . atrox venom were obtained using this functional screening .
This study was performed in accordance with the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health ( NIH Publication No . 85-23 , revised 1996 ) ( A5452-01 ) and were approved by the Animal Experimentation Ethics Committee of the Universidade Federal de Minas Gerais ( License number 200/2010 ) . Animals were maintained at the Centro de Bioterismo of the Instituto de Ciências Biológicas of the Federal University of Minas Gerais , Brazil and received water and food under controlled environmental conditions . Venoms pooled from at least 5–6 adult B . barnetti , B . brazili , B . castelnaudi , B . chloromelas , B . hyoprora , B . microphthalmus , B . peruvianus , B . pictus , B . taeniata and Peruvian B . atrox specimens were generously donated by the Instituto Nacional de Salud ( Lima , Perú ) . Venom from Brazilian B . atrox , Lachesis muta muta , Crotalus durissus and Micrurus frontalis were ceded by Fundação Ezequiel Dias ( FUNED – Belo Horizonte , Brazil ) . Atroxlysin-I was purified as described earlier [37] , using 1 . 252 mg of B . atrox crude venom collected in the Amazon region of Ucayali – Peru . The purity and the molecular mass of 23 kDa were assessed by SDS-PAGE . BCA Kit ( Pierce ) was used to determine protein concentration following the manufacturer's instructions . Atroxlysin-I ( UniProtKB/SwissProt P85420 ) ; BaP1 ( UniProtKB/SwissProt P83512 ) ; Batroxstatin-1 ( UniProtKB/SwissProt C5H5D2 ) ; Batroxstatin-2 ( UniProtKB/SwissProt C5H5D3 ) ; Batroxstatin-3 ( UniProtKB/SwissProt C5H5D4 ) ; Batx-I ( UniProtKB/SwissProt P0DJE1 ) ; B . atrox myotoxin I ( UniProtKB/SwissProt Q6JK69 ) ; Leucurolysin-a ( UniProtKB/SwissProt P84907 ) ; Mutalysin-II ( formely named LHF-II – UniProtKB/SwissProt P22796 ) were used/cited in this work .
To try to bias the selection of mAbs towards antibodies with Atr-I neutralizing activity , we devised an hybridoma screening assay based on the capacity of hybridoma supernatants to block the proteolytic activity of Atr-I towards the synthetic substrate Abz-LVEALYQ-EDDnp ( Figure 1 ) . After immunization of BALB/c mice with Atr-I , a panel of twenty-one anti-Atr-I secreting hybridomas was selected on the basis of their capacity to neutralize Abz-LVEALYQ-EDDnp hydrolysis by Atr-I . Only hybridomas that presented at least 50% of inhibition of the Atr-I proteolytic activity were selected . These clones were then subcloned and a new round of selection was performed . Eight clones presenting the highest inhibition of the proteolytic activity of Atr-I were finally chosen for production , of which six were IgG1 ( named mAbatr1 to mAbatr6 ) and two were IgM . mAbatr1 to 6 were purified on a protein A-Sepharose column and appeared as homogeneous bands on SDS-PAGE ( Figure 2 ) . The eventual cross-reactivity of the selected mAbatrs with several P-I SVMPs ( i . e . Atr-I , BaP1 , Leuc-a and Mut-II ) and B . atrox crude venom was tested in an ELISA format . Figure 3A shows that mAbatrs were not able to recognize the heterologous P-I SVMPs tested . On the other hand , mAbatr1 , 2 , 3 and 6 presented a high reactivity against Atr-I and a moderate reactivity with B . atrox whole venom . mAbatr4 weakly recognized Atr-I and presented low reactivity against the crude venom , while mAbatr5 reacted neither with Atr-I nor B . atrox venom . mAbatr1 , 2 , 3 and 6 were also tested against several South American snake venom antigens ( Figure 3B ) . All mAbs tested showed high specific reactivity against B . atrox venom in our ELISA conditions . The other venoms used as antigens coated to ELISA plates were not recognized by the mAbatrs . Based on the specificity of mAbatrs1 , 2 , 3 and 6 , we decided to evaluate their potential application as diagnostic tools for B . atrox simulating experimental envenoming . In sandwich ELISA using plates coated with polyspecific anti-bothropic antivenom from FUNED ( Brasil ) , mAbatrs recognized B . atrox venom exclusively with an absorbance signal significantly higher ( p<0 . 001 ) compared to all other venoms tested ( Figure 3C ) . mAbatr1 , 2 , 3 and 6 were also tested against B . atrox venom in Western Blot . All mAbatrs tested recognized four bands around 15 , 23 , 30 and 55 kDa ( Figure 4 ) . In an attempt to understand mAbatrs epitope recognition on Atr-I , octa- and pentadecapeptides frameshifted by 1 or 3 residues , respectively , covering the amino acid sequence of Atr-I were synthesized by the SPOT technique and tested either with mAbatr1 , 2 , 3 and 6 ( Figure 5 ) or IgG anti-Atr-I from rabbit as positive control . None of the mAbatrs was capable of reacting with the linear peptides covering the Atr-I primary sequence . However , rabbit anti-Atr-I polyclonal IgGs exhibited reactivity against linear epitopes of Atr-I ( not shown - manuscript in preparation ) , suggesting that epitope recognition by mAbs requires folding of Atr-I into its native structure . The kinetic parameters of mAbatrs interaction with Atr-I were measured on a ProteOn system ( BioRad ) . This equipment measures up to 36 interactions simultaneously , allowing comparisons of affinity among mAbatrs , since they are tested at the same time . Association ( ka ) , dissociation ( kd ) and equilibrium ( KD ) constants for Atr-I binding to mAbs are shown in figure 6 . mAbatr1 , 2 and 6 showed high affinity to Atr-I with equilibrium constants in the 10−9 M range , whilst mAbatr3 and mAbatr4 showed slightly lower affinities ( KD = 9 . 67×10−8 M and 9 . 40×10−8 M ) . mAbatr5 was not able to react with Atr-I , corroborating the ELISA's results . The interaction of BaP1 , Leucurolysin-a and Mutalysin-II with mAbatr1 , 2 , 3 and 6 were also tested on ProteOn , but no binding was detected ( not shown ) . The mAbatrs were tested both in vitro and in vivo in order to assess their neutralizing ability against purified Atr-I or B . atrox whole venom . Inhibition of the proteolytic activity of Atr-I or B . atrox crude venom on Abz-LVEALYQ-EDDnp substrate cleavage is shown in figure 7 . mAbatr1 and 6 exhibited around 85% neutralization of the maximal effect ( mAbatr1 – 83 . 56%±0 . 40; mAbatr6 – 84 . 67%±0 . 78 ) . mAbatr2 showed a weaker blocking of Atr-I activity ( 75 . 30%±1 . 5 ) and mAbatr-3 only demonstrated a moderate neutralizing effect ( 37 . 55%±3 . 67 ) . When tested against B . atrox whole venom in vitro , mAbatr1 , 2 , 3 and 6 presented a weaker neutralization ability ( 51 . 24%±0 . 01; 43 . 63%±1 . 62; 29 . 79%±0 . 93 and 39 . 58%±4 . 45 , respectively ) . On the other hand , mAbatr4 and 5 did not neutralize enzymatic proteolysis of the synthetic substrate induced by Atr-I in the tested conditions . To reduce testing in living animals , only the three strongest in vitro neutralizing antibodies ( mAbatr1 , 2 and 6 ) were tested against Atr-I or B . atrox venom induced hemorrhage . One MHD of Atr-I was pre-incubated with mAbatrs at different concentrations and then injected subcutaneously in mice . mAbatr1 and 6 fully neutralized the hemorrhagic activity induced by atr-I at a molar ratio of 5∶1 ( Atr-I : mAbatr ) , while mAbatr2 only neutralized the hemorrhage at a 2 . 5∶1 molar ratio or higher ( Atr-I : mAbatr2 ) ( Figure 8A ) . 1 . 8 MHD of B . atrox crude venom was completely neutralized by mAbatr1 at 50 µg ( 0 . 73∶1 molar ratio Atr-I: mAbatr1 ) or higher , while mAbatr2 and 6 inhibited B . atrox hemorrhagic activity at 100 µg ( 0 . 37∶1 molar ratio Atr-I: mAbatr ) ( figure 8B ) . However , when injected without preincubation with B . atrox venom , none of the mAbatrs was able to neutralize hemorrhage induced by B . atrox venom at any dose tested ( not shown ) .
The bothropic envenoming induces severe local symptoms in human victims , including hemorrhage and tissue necrosis . Complications arise in up to 40% of the cases , in which permanent damage can necessitate the amputation of limbs . The local noxious effects of bothropic venoms ( e . g . hemorrhage and necrosis ) are mainly due to the action of SVMPs [11] , [26]–[30] , which seem to be not well neutralized by current therapeutic anti-bothropic antivenoms [17]–[20] . In this present work , we describe a new rational and functional method to produce neutralizing monoclonal antibodies against P-I SVMPS , using as hybridoma selection criterion the capability of mAbs to block the proteolysis induced by SVMPs . Atr-I , a 22 . 3 kDa P-I class SVMP isolated from the Peruvian B . atrox venom , is able to hydrolyze the peptide bond between ala14 and leu15 present in insulin B-chain [37] , as well in the fluorogenic peptide ( Abz- LVEALYQ-EDDnp ) . We decided to synthesize this peptide sequence coupled to a fluorescent donor and its respective quencher and to use the FRET technique to measure the inhibition of Atr-I-induced hydrolysis of this biosensor by mAbs . Hybridoma supernatants abolishing fluorescent emission were selected as potentially neutralizing mAbs . Based on this method , we have obtained six IgG1 and two IgM monoclonal antibodies against Atr-I . mAbs previously produced against BaP1 or Mut-II were selected by ELISA and presented cross-reactivity against heterologous venoms and P-I class SVMPs [45]–[47] . Although our mAbatrs were not selected on an antigen-binding capacity basis , they showed a very high specificity to Atr-I and B . atrox venom . We demonstrated that they do not recognize either other P-I SVMPs from Latin American pit viper venoms that present high degrees of similarity to Atr-I ( i . e . Mut-II and BaP1 , 57% of identity and Leuc-a , 52% of identity ) , or any other South American whole venoms from species that share the Amazonia forest as habitat with B . atrox . In the Amazonian forest , the snake Lachesis muta muta is responsible for approximately 10% of all snakebites , and the symptoms of this accident are very similar to B . atrox envenoming . Currently , there is no laboratory diagnostic able to differentiate B . atrox from L . muta muta envenoming . Therefore , we have developed a simple test for discriminating B . atrox envenoming from envenoming caused by other genera . To avoid unnecessary animal suffering , we have bled Swiss mice and prepared a mixture of their sera with different venoms , simulating experimental envenoming . ELISA plates were coated with polyspecific anti-bothropic antivenom ( FUNED – Brasil ) to capture antigens from different venoms . Using a pool of mAbatrs , only antigens from B . atrox venom were recognized , presenting an absorbance signal double that of pre-immune sera , suggesting that our mAbatrs could be useful in the development of a differential diagnostic for B . atrox envenoming . Western blot assays using mAbatr and purified Atr-I or B . atrox venom were done under reducing ( not shown ) and non-reducing conditions . Under reducing conditions , no reactivity was observed . In non-reducing conditions , mAbatr1 , 2 , 3 and 6 recognized four bands at approximately 15 , 23 , 30 and 55 kDa . Earlier works have shown the presence of several toxins in B . atrox venom [23] , [24] , [48] , [49] , including: B . atrox myotoxin I , a secreted Lys49 PLA2 with an calculated MW of 13 , 826 [50] , which possesses 37% of identity compared to Atr-I; a P-I SVMP named Batroxase [39] , which contains 90% of identical residues compared to Atr-I; as well as three SVMPs from P-III class , called batroxstatin-1 , -2 and -3 , which possess up to 60% of identity at the proteinase domain compared to Atr-I [36] , [39] . Due to the likely evolution of toxins by gene duplication and diversification , some epitopes may be kept unchanged during evolution [24] , [51] , [52] and common epitopic motifs recognized by mAbatrs might be shared by some different classes of SVMPs in B . atrox venom , which could explain the cross-reacting bands around 30 and 55 kDa in Western Blot assay , as observed before [46] , [47] , [53] , [54] . However , the recognition of the band at ∼15 kDa is probably not the result of an interaction between mAbatrs and PLA2 since the overall shape of PLA2 molecules differ from SVMPs structures . Thus it is reasonable to assume that the reacting band at 15 kDa might be either an artifact or result of SVMPs autolysis/degradation . None of our mAbatrs recognized overlapping synthetic peptides from the Atr-I primary sequence , confirming that the conformation of Atr-I is very important in recognition by mAbatrs and suggesting that all mAbatrs bind to conformational epitopes . Few works have reported on the molecular interaction of monoclonal antibodies with their respective epitopes in SVMPs . Apparently , neutralizing mAbs against P-I SVMPs interact predominantly via conformational structures [45]–[47] . Further studies are necessary to better characterize the functional epitopes recognized by mAbatrs and to clarify their role in the biological activity of Atr-I . However , recognition of loops adjacent to the methionine-turn near to the catalytic region , which is an important region for the catalytic activity and determines substrate specificity , might explain the neutralizating activities of mAbatrs . Moreover , this region presents the highest variability in SVMPs , which could account to the high mAbatrs' specificity for Atr-I [55] , [56] . mAbatr6 ( KD = 8 . 52×10−9 M ) , mAbatr1 ( KD = 12 . 0×10−9 M ) and mAbatr2 ( KD = 15 . 1×10−9 M ) , showed the highest affinities for Atr-I . Despite the fact that the hybridoma selection method we designed was based on function and not on binding , mAbatrs presented nanomolar equilibrium constants for their binding to Atr-I . This affinity is in the same range as that found in mAbs against others SVMPs , which were selected through affinity-based ELISA assays . When selected conventionally by binding assays , mAbs against SVMPs do not present a correlation between affinity and inhibitory action [45] , [47] . However , it is interesting to note that there is a clear correlation between the measured affinity of mAbatrs to Atr-I and their neutralizing efficacy . Three mAbs ( mAbatr1 , mAbatr2 and mAbatr6 ) efficiently neutralized proteolysis induced by both Atr-I and B . atrox venom upon Abz-LVEALYQ-EDDnp . As hemorrhagic activity of SVMPs is dependent on their enzymatic activities , we decided to test whether mAbatr1 , 2 and 6 could prevent hemorrhage induced by either Atr-I or B . atrox venom in vivo . Although mAbatr1 , mAbatr2 and mAbatr6 presented a slightly weaker inhibitory activity on synthetic substrate compared to anti-BaP1 monoclonal antibodies MABaP1-3 and MABaP1-6; mAbtr1 , 2 and 6 fully neutralized in vivo hemorrhage induced by Atr-I isolated or B . atrox whole venom when preincubations with mAbs were performed . Brazilian and Peruvian B . atrox venoms are composed mainly by SVMPs , including P-I and P-III classes [21]–[25] , which are the main molecules responsible for the hemorrhagic activity in B . atrox envenoming . The possible cross-reactivity against some other SVMPs present in B . atrox venom , but not all of them , may play a key role in neutralization of B . atrox whole venom in hemorrhage and could explain the weaker neutralizing ability of mAbatrs against B . atrox whole venom tested in vitro . The neutralization of other bothropic South American venoms was not tested , since all mAbatrs were very specific and recognized only B . atrox venom . Currently , the standard procedure used , even for monoclonal antibodies , to measure the neutralization capacity of an antivenom against the hemorrhagic activity of snake venoms consists in the preparation of a mixture of venom and antivenom , followed by injection of this preincubated mixture in animals [45] , [46] , [53] , [57] , [58] . Thus , it is reasonable to assume that once mAbatrs bind to Atr-I and/or Atr-I-like molecules in preincubation step of our hemorrhagic assay these hemorrhagins are sterically hindered and not able to bind to their in vivo molecular targets , leading to the abolishment of hemorrhage . On the other hand , when mAbatrs are not preincubated with B . atrox venom , hemorrhage is still observed , suggesting that when hemorrhagins are first injected in animals , they bind to and degrade their in vivo molecular targets and become inaccessible to mAbatrs . Further studies are needed to clarify the efficacy of preincubation steps of mabs and SVMPs in studies of neutralization of hemorrhage induced by SVMPs . In conclusion , we developed an efficient method for functional antibody screening , based on a synthetic biosensor to produce mAbs specifically neutralizing P-I SVMPs in vitro and in vivo . To the best of our knowledge , this is the first time that a functional screening has been used in order to select monoclonal antibodies able to block the toxic effects of SVMPs . It is also the first description of mAbs against Atr-I , isolated from B . atrox venom , with inhibitory potential against toxic activities of purified Atr-I and B . atrox crude venom . It is still unknown where neutralizing mAbatrs bind to Atr-I . Further , mAbatrs are highly specific to B . atrox antigens and may be useful as diagnostic tools for B . atrox envenoming . These very encouraging results open the way for a wider utilization of synthetic biosensors in functional screening aiming at the production of neutralizing monoclonal antibodies for further therapeutic approaches or diagnostic assays against B . atrox envenoming .
|
In this work , we propose a new screening strategy to produce monoclonal antibodies against Atr-I , a P-I class SVMP from Bothrops atrox , which is the snake responsible for the majority of the accidents in South America . SVMPs are the main toxic factors in Bothrops venom causing systemic and local hemorrhage , which may evolve to inflammation and/or necrosis . Since the toxic effects of SVMPs are related to their proteolytic activity , we have produced a peptide which was used as a biosensor for Atr-I hydrolysis . Hydrolysis of this substrate was monitored and the clones possessing inhibitory activity against the proteolytic activity of Atr-I upon the peptide were selected . Using our new approach , we have obtained four monoclonal antibodies highly specific and with neutralizing capacity against the hemorrhagic activity of either Atr-I alone or Bothrops atrox whole venom . To the best of the authors' knowledge , this is the first time where a functional screening is used for the selection of neutralizing mAbs against SVMPs . It is also the first description of mAbs anti-Atr-I , with inhibitory potential against its toxic activities which may be useful for diagnosis and treatment in the future .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"humoral",
"immunity",
"biochemistry",
"immunochemistry",
"clinical",
"immunology",
"immunity",
"biology",
"and",
"life",
"sciences",
"immunology",
"vaccination",
"and",
"immunization",
"immunotherapy"
] |
2014
|
Use of a Synthetic Biosensor for Neutralizing Activity-Biased Selection of Monoclonal Antibodies against Atroxlysin-I, an Hemorrhagic Metalloproteinase from Bothrops atrox Snake Venom
|
Prolific sheep have proven to be a valuable model to identify genes and mutations implicated in female fertility . In the Lacaune sheep breed , large variation in litter size is genetically determined by the segregation of a fecundity major gene influencing ovulation rate , named FecL and its prolific allele FecLL . Our previous work localized FecL on sheep chromosome 11 within a locus of 1 . 1 Mb encompassing 20 genes . With the aim to identify the FecL gene , we developed a high throughput sequencing strategy of long-range PCR fragments spanning the locus of FecLL carrier and non-carrier ewes . Resulting informative markers defined a new 194 . 6 kb minimal interval . The reduced FecL locus contained only two genes , insulin-like growth factor 2 mRNA binding protein 1 ( IGF2BP1 ) and beta-1 , 4-N-acetyl-galactosaminyl transferase 2 ( B4GALNT2 ) , and we identified two SNP in complete linkage disequilibrium with FecLL . B4GALNT2 appeared as the best positional and expressional candidate for FecL , since it showed an ectopic expression in the ovarian follicles of FecLL/FecLL ewes at mRNA and protein levels . In FecLL carrier ewes only , B4GALNT2 transferase activity was localized in granulosa cells and specifically glycosylated proteins were detected in granulosa cell extracts and follicular fluids . The identification of these glycoproteins by mass spectrometry revealed at least 10 proteins , including inhibin alpha and betaA subunits , as potential targets of B4GALNT2 activity . Specific ovarian protein glycosylation by B4GALNT2 is proposed as a new mechanism of ovulation rate regulation in sheep , and could contribute to open new fields of investigation to understand female infertility pathogenesis .
Women , cattle , goats and ewes have generally one or two offspring , whereas other mammals , such as sows , rodents and dogs are prolific and produce more than three offspring . It relies on the number of ovulations at each estrus cycle i . e . the ovulation rate ( OR ) , for which the underlying genetic mechanism was puzzling until the identification of fecundity genes in sheep , bone morphogenetic protein-15 ( BMP15 ) , growth and differentiation factor-9 ( GDF9 ) and BMP receptor-1B ( BMPR1B ) [1] . Following the discovery of sheep fecundity genes , several research groups have focused on BMP15 and GDF9 and they have found numerous mutations associated with human ovarian pathologies such as premature ovarian failure or polycystic ovary syndrome [2] . Thus , prolific sheep are now considered as valuable models for identifying genes and mutations involved in mechanisms controlling the ovarian function , for agronomical purposes such as genetic selection of prolificacy , and for clinical purposes in the case of female infertility or subfertility . In the meat strain of the Lacaune sheep breed , large variation in litter size has been observed and genetic studies explained this variation by the segregation of at least two major genes influencing OR and prolificacy , one being X-linked and named FecX , the second being autosomal and named FecL [3] , [4] . FecX is known as BMP15 and in the Lacaune breed , the mutant allele ( FecXL ) associated with high prolificacy was identified as a pCys321Tyr substitution altering the BMP15 protein function [3] . Heterozygous FecXL mutation is associated with a twofold increase in OR , but homozygous FecXL/FecXL ewes are sterile , thus mimicking the phenotype observed for the other 5 mutations described in the ovine BMP15 gene [5]–[7] . The influence of the autosomal FecLL mutation on OR is additive with one copy increasing OR by about 1 . 5 and two copies by about 3 . 0 [4] , [8] . We have recently established that the FecL locus influences both the ovarian activity and the endocrine profiles [9] . Indeed , increased OR in homozygous FecLL/FecLL ( thereafter named L/L ) ewes is associated with an increased number of gonadotropin-dependent follicles with a diameter greater than 3 mm , an increase in plasma estradiol concentrations , and an increase in the frequency of Luteinizing Hormone ( LH ) pulsatility during the follicular phase , leading to a precocious LH surge . In contrast , plasma concentrations of Follicle Stimulating Hormone ( FSH ) were not different compared to wild-type ewes . Based on ovarian phenotype and endocrine profiles , these findings suggest that the FecLL mutation affects ovarian function in a different way compared to other known hyperprolificacy-associated mutations , all affecting genes of the bone morphogenetic protein signaling system , BMP15 , GDF9 and BMPR1B [1] , [10] . The FecLL mutation associated with increased OR has not yet been identified . In a previous work , a full genome scan localized the FecL locus on sheep chromosome 11 ( OAR11 ) . Fine mapping reduced the interval containing FecL to markers BM17132 and FAM117A , corresponding to a synteny block of 1 . 1 megabases on human chromosome 17 ( HSA17 ) , which encompasses 20 genes [8] . With the aim to identify the FecL gene and its hyperprolificacy-associated mutation , we combined different approaches based on genetic fine mapping ( both classical development of genetic markers and high throughput Roche 454 sequencing strategy ) , gene expression analysis , histochemistry and protein identification by mass spectrometry . From our results , we propose B4GALNT2 , encoding the glycosylation enzyme beta-1 , 4-N-acetyl-galactosaminyl transferase 2 , as the FecL gene . Finally , the FecL sheep model of prolificacy-associated mutation leads to the discovery of a new pathway involved in the regulation of folliculogenesis and ovulation rate .
The interval of localization previously published corresponded to a synteny block of 1 . 1 megabases on HSA17 [8] . Genotyping additional markers further reduced it across the whole experimental Lacaune pedigree ( F1 , BC and F1xBC representing 189 animals ) . This reduced interval was comprised between markers GNGT2-M2 ( OAR11:36899194 ) and Ms162 ( OAR11:37387389 ) encompassing 488 kb on the ovine chromosome 11 ( OAR11 , ovine genome version 3 . 1 released October 2012 ) and corresponded to a block of synteny of 479 kb on bovine chromosome 19 ( BTA19 , bovine genome version 4 . 6 . 1 released October 2011 ) . This entire region was sequenced with the Roche 454 sequencing technology , using long-range PCR , in one heterozygous L/+ animal and two homozygous +/+ and L/L animals . Sixty-two polymorphisms were evidenced and an appropriate subset was genotyped on the recombinant animals allowing the reduction of the locus . This new interval of localization , comprised between two SNP markers on OAR11 g . 36910171T>C ( recombinant ewe n°990855 ) and g . 37107627G>C ( recombinant ewe n°60718 ) , was estimated at 197 kb based on ovine genome OARv3 . 1 ( Figure 1 ) . This region encompasses 3 predicted protein-coding genes on the ovine genome , named B4GALNT2 ( beta-1 , 4-N-acetyl-galactosaminyl transferase 2 ) , EZR ( ezrin ) and IGF2BP1 ( insulin-like growth factor 2 mRNA binding protein 1 ) . In order to identify all the polymorphisms contained within this 197 kb interval , we analyzed separately the sequences of the two homozygous L/L and +/+ animals . Sequence information coming from the Roche 454 sequencing technology was contained within 13 and 17 independent sequence contigs in L/L and +/+ animals , respectively . We have then completed these sequences by Sanger sequencing to link contigs with each other . By comparing the two sequences , we identified 49 polymorphisms ( 43 SNP , 4 microsatellites and 2 Insertion/Deletion , Table 1 ) . None of the detected variants affected the coding sequence of the annotated genes . Polymorphisms were first tested on a set of one L/L , one L/+ and 2 to 6 +/+ animals for allele sharing . If the allele associated with the L-haplotype was not found on a wild chromosome , then the marker was tested on successive subsets of wild chromosomes from the Lacaune families and other sheep populations , as described in the materials and methods section . There were only two polymorphisms segregating as the FecLL mutation , i . e . fully associated with the hyperprolific phenotype: the SNP g . 36938224T>A , localized in the intron 7 of B4GALNT2 , and the SNP g . 37034573A>G localized in the intergenic sequence between B4GALNT2 and EZR , 10 . 4 kb upstream of EZR ( Figure 1 ) . The sequencing of the locus indicated a real interval of 194 . 6 kb ( GenBank:KC352617 ) and allowed to correct and complete the ovine reference genome sequence in this region . The locus effectively contained the full sequence of the B4GALNT2 gene based on the ovine B4GALNT2 mRNA ( sequenced from Lacaune sheep , GenBank:KC175557 ) , the full sequence of the IGF2BP1 gene based on the bovine IGF2BP1 mRNA ( GenBank:NM_001192454 ) and a BLAST hit with the bovine EZR mRNA ( GenBank:NM_174217 ) that we assumed to be a pseudogene . Indeed , the bovine EZR gene is located in a region of BTA9 that was not syntenic of the ovine FecL locus on OAR11 and the predicted sequence of this EZR annotation on OAR11 carried a premature STOP codon limiting the predicted protein to 223 amino-acids instead of 581 . Consequently , only B4GALNT2 and IGF2BP1 were considered as positional candidate genes for FecL . The mRNA expression of the 2 positional candidate genes within the FecL locus was checked by real-time quantitative PCR in different tissues of the reproductive axis ( hypothalamus , pituitary gland and ovarian granulosa and theca cells ) isolated from +/+ and L/L ewes . In the hypothalamus and the pituitary gland , B4GALNT2 and IGF2BP1 mRNA amounts were similar between genotypes , but they were significantly higher in the ovarian cells of the L/L , compared to +/+ ewes ( Figure 2 ) . Interestingly , B4GALNT2 exhibited a 1000-fold higher expression level in the L/L granulosa and theca cells , whereas IGF2BP1 expression was enhanced 6-fold only in granulosa cells and was found unchanged in theca cells . To check if this over-expression might concern other genes of the locus , the expression of the flanking genes within the “one recombinant zone” ( Figure 1 ) , i . e . PHOSPHO1 , ABI3 and GNGT2 on the one side and GIP , SNF8 , UBE2Z , ATP5G1 and CALCOCO2 on the other side , was also studied . None of these genes showed an altered expression in the different L/L tissues of the reproductive axis . Given that GIP was not expressed in those tissues , its expression was checked in intestine and no difference was found between genotypes ( supplemental figure S1 ) . Moreover , as IGF2BP1 is a RNA binding protein controlling the steady-state level of the MYC oncogene mRNA [11] and the ß-actin ( ACTB ) gene translation [12] we checked for MYC mRNA expression and ACTB protein accumulation in Lacaune granulosa cells as a possible consequence of IGF2BP1 overexpression ( supplemental figure S2 ) . Real-time PCR analysis showed a large decreased expression of the MYC gene in large follicles compared to small ones ( 18-fold , p<0 . 001 ) , but no alteration associated with elevated IGF2BP1 mRNA level in L/L granulosa cells . ACTB accumulation checked by western blotting was not affected by the L/L genotype either . Thus , the localization of the B4GALNT2 gene within the minimal locus , the presence of the SNP g . 36938224T>A on OAR11 , localized in the intron 7 of this gene as the possible causal mutation , and the presence of a very high overexpression of the gene in the ovarian cells of L/L ewes led us to further investigate B4GALNT2 ectopic expression and activity in the ovary of the FecLL carrier animals . B4GALNT2 is a β1 , 4-N-acetylgalactosaminyltransferase which was previously shown to be involved in the synthesis of the Sd ( a ) antigen , a carbohydrate expressed on erythrocytes , colonic mucosa and other tissues . B4GALNT2 transfers a beta-1 , 4-linked GalNAc to the galactose residue of an alpha-2 , 3-sialylated chain found on both N- and O-linked glycans [13] . Immunohistochemistry experiments using an antibody raised against B4GALNT2 showed that the enzyme was detectable in the granulosa cells of L/L ovarian follicles only ( Figure 3 ) . The use of the Dolichos Biflorus Agglutinin ( DBA ) lectin and the KM694 antibody raised against the Sd ( a ) antigen , both specific of the glycosylation activity of B4GALNT2 , clearly localized the targets of the enzyme only in the granulosa cells and the antral follicular fluid of L/L ovaries ( Figure 4 ) . To confirm the glycosylation activity of B4GALNT2 , +/+ ovine granulosa cells were transiently transfected by a B4GALNT2 expressing construct . Specific staining with DBA was observed in B4GALNT2-transfected +/+ granulosa cells ( Figure 5 ) , indicating that overexpression of the B4GALNT2 gene is directly related to positive DBA staining . The presence of a B4GALNT2 activity in the granulosa cells of the FecLL carrier ewes was confirmed by the results of DBA lectin and KM694 antibody precipitation and subsequent western-blot experiments using L/L and +/+ granulosa cell extracts and follicular fluids ( Figure 6 ) . Indeed , DBA or KM694 staining revealed very different glycoprotein profiles between L/L and +/+ ewes . In both granulosa cells and follicular fluids , at least 7 glycoprotein forms of various molecular weights ( ranging from 40 to over 250 kDa ) were retained by DBA lectin precipitation in the L/L ewes and were absent in non-carrier ewes . The KM694 immunoprecipitation of the proteins contained in the follicular fluids of the FecLL carrier ewes evidenced a subset of these glycoproteins , with a high molecular weight . We aimed at identifying the glycoproteins which are the targets of B4GALNT2 in ovary after their purification from L/L follicular fluids using DBA lectin affinity , followed by high-resolution mass spectrometry . Peptides and proteins purified from L/L and +/+ follicular fluids were identified with at least four independent peptides and a high probability threshold ( >95% ) . This allowed identifying , on two repetitions of the experiment , 10 glycoproteins only present in FecLL follicular fluids , and then suspected to be the main substrates of B4GALNT2 activity ( Table 2 ) . Identified proteins were of a large molecular weight range ( ranging from 39 to 613 kDa ) and most of them were extracellular matrix or membrane-associated proteins . Interestingly , the inhibin α and βA subunits , leading to the production of Activin A and Inhibin A , and the chondroitin sulfate proteoglycan Versican are well known to be directly involved in female reproductive function [14]–[16] . These proteins represent promising physiological candidates to understand the mechanism by which the FecLL mutation affecting B4GALNT2 expression increases the ovulation rate in Lacaune sheep population .
The fine mapping strategy combining marker development through high-throughput sequencing and genotyping of selected recombinant animals allowed the delimitation of the FecL locus on OAR11 between SNP markers g . 36910171T>C and g . 37107627G>C . With the double aim to densify the locus with more informative polymorphisms and to directly identify the causal polymorphism , we developed a systematic high-throughput sequencing of the locus with the Roche 454 technology on finely chosen animals . The targeting of the locus was done by overlapping long-range PCR fragments ( around 10 kb ) , a method successfully used to detect genomic variations in BRCA1/BRCA2 locus in human [17] . We experienced some bias in detecting polymorphisms , especially SNP , in the reads depth of the L/+ animal due non-independent reads [18] and PCR-dependent allele disequilibrium . Nevertheless the use of the two independent sequences L/L and +/+ allowed the elimination of this latter bias . The polymorphisms evidenced were further studied for allele sharing . The different appropriate subsets of markers that were genotyped on recombinant ewes led to the reduction of the minimal interval to 194 . 6 kb . Within this interval of localization , 2 SNPs have been found fully associated with the FecLL mutation , namely SNP g . 36938224T>A and SNP g . 37034573A>G . The SNP g . 36938224T>A was localized in the intron 7 of B4GALNT2 , in a sequence portion that was only conserved in the bovine orthologous gene , and not in other species . The SNP g . 37034573A>G was localized in the intergenic sequence between B4GALNT2 and IGFBP1 in a non-conserved region enriched with LINES and SINES repetitive elements . The finding of non-coding SNPs as causal mutations contrasts with other known mutations affecting ovulation rate through alteration of the coding sequence of BMP15 , GDF9 and BMPR1B genes impairing the protein function [10] . To go further in the fine mapping study , different strategies can be proposed . The first one is to find additional recombinant animals within the genetic families . A second one is to find recombinant animals at the population level , i . e . animals carriers of a shorter L-haplotype and finely phenotyped . However the probability to find such animal recombining within an interval of 195 kb is very low . If only one of these two SNP is the causal mutation , one way to prove it is to find an animal carrying a recombination between the two markers . However , the 2 SNPs are 96 kb apart and the probability to find such animal is nearly null . A third possibility is to increase the number of wild chromosomes ( i . e . wild haplotypes ) tested for allele sharing . Animals coming from 9 different breeds have been genotyped ( representing 180 wild haplotypes ) but did not allow the elimination of one or the other putative causal mutation . Annotation available on the last release of the sheep genome assembly ( OARv3 . 1 ) indicated that the minimal 194 . 6 kb interval , that we entirely sequenced , contained only 3 potential protein encoding genes , B4GALNT2 , EZR and IGF2BP1 . The EZR annotation is very recent ( October 2012 ) and this gene was not considered as an expressional candidate for FecL at the beginning of the work . Moreover , if the localizations of B4GALNT2 and IGF2BP1 on OAR11 were coherent with their syntenic location on BTA19 and HSA17 , the presence of the EZR gene annotation is intriguing . Indeed , in bovine ( UMD3 . 1 ) and human ( GRCh37 ) genomes , this gene is positioned on BTA9 and HSA6 , respectively ( www . Ensembl . org ) . Moreover , Blastn analysis ( www . livestockgenomics . csiro . au ) of the bovine EZR cDNA ( NM_174217 ) indicated sequences producing significant alignments on Ovis aries breed Texel contig_127582 , OAR11 and OAR8 . Only matched sequences on OAR8 presented an exon/intron structure in a syntenic region of BTA9 and HSA6 , and then should correspond to the ovine EZR gene . The high identity match we found between bovine EZR cDNA and both the ovine reference genome on OAR11 and our own FecL locus sequence ( 98% coverage , 95% identity ) evidenced the presence of an EZR spliced pseudogene with a premature stop codon impairing the translation of an EZR protein . For those reasons , only IGF2BP1 and B4GALNT2 were considered as positional candidate genes for FecL in this study . Due to the absence of polymorphism in the coding sequence of IGF2BP1 and B4GALNT2 genes , we searched for an expressional candidate to discriminate between the two genes . Real-time PCR analysis showed that both IGF2BP1 and B4GALNT2 expression was significantly affected by FecLL , specifically in ovarian cells indicating a tissue-specific regulation by the FecLL mutation . Interestingly , no differential expression was observed for genes outside of the minimal locus , reinforcing the genetic fine mapping result . For follicles collected at the same size class , the differential expression observed between +/+ and L/L follicles could not be attributed to the existence of a difference in follicle maturity as already observed for the hyperprolificacy-associated Booroola mutation [19] . Indeed , IGF2BP1 and B4GALNT2 expression in granulosa cells was not different between small and large antral follicles that were at different maturity stages as attested by marker expression , such as LHR , INHA , CYP19A1 and CYP11A1 as previously shown [9] or MYC ( present study ) . SNP g . 36938224T>A , localized in the intron 7 of B4GALNT2 , and SNP g . 37034573A>G in the inter-genic region , were the only two polymorphisms segregating as the FecLL mutation . They were then thought to be directly the cause of the huge overexpression of B4GALNT2 and in a lesser extent of IGF2BP1 in the ovaries of FecLL carrier ewes , through a molecular mechanism that remains to be determined . In order to explain this differential expression we searched in silico for transcriptional factors able to bind these SNP locations . However , we failed to find matches with consensus sequences binding known transcriptional factors at each site . Experimentally , electromobility shift assays with ovine granulosa cell nuclear extracts failed also to discriminate between SNPs and between prolific and wild type alleles . As a second in silico approach , we also searched matches in Patrocles , the database of polymorphic microRNA-target interactions ( www . patrocles . org ) . Polymorphism in miRNA target sites might be important effectors of phenotypic variation as attested by the hypermuscularity phenotype in Texel sheep [20] . Interestingly , the B4GALNT2 intronic SNP g . 36938224T>A was found to create a motif GTGTGAGA in FecLL which is a recognition site for MIR342 that is conserved among bovine and human . However , we cannot reconcile this finding with the current knowledge on gene regulation by microRNAs , indicating that those small noncoding RNAs usually targeted matured mRNA in the 3′-untranslated regions within cytoplasmic protein complexes to repress protein synthesis [21] . Another hypothesis is that the intronic location of the SNP g . 36938224T>A may be associated with alternative splicing of the B4GALNT2 mRNA , but analysis by RT-PCR amplification between exon 6 and terminal exon 11 failed to evidence such alternative splicing between L/L and +/+ mRNA . In human gastrointestinal cancer cells the expression of B4GALNT2 is dependent on the promoter methylation status [22] . As in other species , a large CpG island is present in the vicinity of the ovine B4GALNT2 promoter . By bisulfite sequencing or restriction site analysis by methyl sensitive enzymes of granulosa cell derived DNA from +/+ and L/L animals , we failed to detect any differential methylation status . Thus , it remains to be determined how the g . 36938224T>A and/or g . 37034573A>G SNPs are able to distantly regulate B4GALNT2 and IGF2BP1 expression at the mRNA level specifically in ovarian cells , maybe through yet unknown tissue-specific DNA-protein interactions . The IGF2BP1 gene was part of the minimal FecL locus and its expression was increased under the influence of the FecLL prolific allele ( being g . 36938224A and/or g . 37034573G ) in granulosa cells . This expression in ovine ovary is consistent with the IGF2BP1 expression already observed in mouse and human ovaries [23] . Elevated IGF2BP1 expression was also associated with ovarian carcinoma and proliferation deregulation through a c-myc ( MYC ) dependent mechanism [24] . Indeed , IGF2BP1 possesses RNA binding motifs and can stabilize the MYC mRNA [11] and enhance the translation of several other genes such as IGF2 and ß-actin ( ACTB ) by binding to their mRNA [25] . Interestingly , all these known target genes of IGF2BP1 action play an important role in follicle function . IGF2 is present in follicular fluid [26] , expressed by granulosa [27] and theca cells [28] and participates in the maturation of ovarian follicles [29] . Actin is associated with granulosa cell shape changes along antral folliculogenesis [30] that may consequently affect steroid synthesis and proliferation [31] . MYC participates in the control of granulosa cell proliferation under gonadotropin and insulin stimulation [32] , [33] . However , checked by MYC mRNA and ACTB protein accumulation , it seemed that the 6-fold overexpression of IGF2BP1 had no biological consequences in the granulosa cells of L/L ovaries . The FecLL prolific allele was clearly associated with an ectopic overexpression ( at least 1000-fold compared to wild-type ) of B4GALNT2 mRNA in ovarian granulosa and theca cells from antral follicles . This ectopic expression seemed to occur specifically in the ovary , since expression checked by real-time qPCR in pituitary , hypothalamus ( Figure 2 ) , intestine ( Figure S1 ) and adrenals ( data not shown ) showed no difference between genotypes , indicating a tissue-specific mechanism for B4GALNT2 expression regulation by the FecLL mutation . Such ectopic expression of B4GALNT2 has never been demonstrated in the ovary before . However , ectopic expression of B4galnt2 is the molecular basis for the action of the Mvwf locus , a major modifier of plasma von Willebrand factor ( VWF ) level in RIIIS/J mice [34] . Indeed a switch of B4galnt2 gene expression from intestine epithelial cells to vascular endothelial cells , resulting in aberrant VWF glysosylation , explained the phenotypic characteristics of the RIIIS/J mice similar to human type 1 von Willebrand disease . The region responsible for the Mvwf locus regulatory switch lies within a 30-kb genomic interval upstream to the B4galnt2 gene [35] . In the present study , one of the potential causal SNP ( g . 37034573A>G ) lies 42 kb upstream to B4GALNT2 in a region which is not conserved between sheep and mouse . The mRNA huge overexpression of B4GALNT2 observed in the L/L ovaries was accompanied by increased protein expression level as shown by specific immunohistochemistry , mainly in granulosa cells . As previously stated , B4GALNT2 is involved in the synthesis of the Sd ( a ) antigen on various protein targets [13] with the transfer of a terminal GalNAc , specifically recognized by DBA lectin [34] and the KM694 antibody [36] , [37] . Using DBA lectin staining , we evidenced the GalNac transferase activity exclusively in the granulosa cells of the FecLL ovaries . The glycosylated targets of B4GALNT2 were mainly secreted in the follicular fluid as shown by histochemistry and western-blotting experiments . However , the pattern of target proteins revealed by DBA lectin and KM694 antibody was different . This discrepancy could be explained by the strict recognition specificity of the Sd ( a ) antigen ( GalNac transfer to terminal α2 , 3-sialylated galactose residue in the ß1 , 4 linkage ) by the KM694 antibody but a wider spectrum of terminal GalNac recognition by DBA lectin , indicating that B4GALNT2 could create other carbohydrate structures than the Sd ( a ) antigen . Moreover , we demonstrated that the in vitro overexpression of B4GALNT2 was responsible for the DBA lectin staining . Regarding these results , we assume that an atypical glycosylation of proteins within the granulosa cells of FecLL ewes , which does not occur in wild-type cells , is due to the overexpression of B4GALNT2 . This could represent the initiating mechanism of the increased ovulation rate which characterizes the FecLL Lacaune sheep . In transgenic mice , manipulating glycosylation at the oocyte level led to slight increased fecundity or primary ovarian insufficiency depending on the glycosyltransferase being invalidated [38] , [39] . Anyway , it proves the importance of glycosylation in the control of ovarian function . In order to go further in the role of B4GALNT2 in the FecLL ovaries , we tried to identify the target proteins recognized by the DBA lectin using lectin affinity purification . Through comparative mass spectrometry analysis , we identified several atypically glycosylated proteins secreted in the follicular fluid of L/L ewes . Among the glycoproteins identified , were the inhibin subunits ( INHA and INHBA ) participating in the inhibin A and the activin A hormone formation [16] . Inhibin A and activin A are dimeric glycoproteins belonging to the transforming growth factor-beta ( TGFβ ) superfamily . They are produced by granulosa cells and can accumulate in high concentrations in follicular fluid . Activin A is considered to act mainly through auto/paracrine signaling in granulosa and theca cells [16] , while inhibin A acts mainly through endocrine negative feedback regulation of pituitary FSH secretion . However , inhibin A has also been shown to exert a blocking action on the activin A and other TGFß member dependent regulation of steroidogenesis and proliferation within the ovary [16] . Immunization against inhibin can promote an increase in ovulation rate and prolificacy [40] through enhanced follicular development in sheep [41] , goat [42] and water buffalo [43] . It has been suggested that inhibin antibodies may act primarily by an intraovarian paracrine action rather than by reducing the suppressive action of inhibin on pituitary FSH release [43] , [44] . Inhibin A is a heterodimer of α- ( INHA ) and ßA- ( INHBA ) subunits , whereas activin A is a homodimer of ßA subunits . Interestingly , the differential subunit association ( α-ßA or ßA-ßA ) was dependent on glycosylation events implicating N-linked oligosaccharide [45] . The B4GALNT2 atypical glycosylation of INHA and INHBA could alter the subunits association , and then change the biological activity or the physiological ratio of activin A and inhibin A produced by the FecLL ovaries . Given that no difference in FSH plasmatic concentration during the follicular phase was observed between +/+ and L/L Lacaune ewes [9] , one might suspect a more direct consequence on activin A signaling within the ovary . Experiments are ongoing to demonstrate the direct effect of B4GALNT2-dependent glycosylation on inhibin A and activin A biological activity at the auto/paracrine and endocrine levels . Other good physiological candidates to explain increased OR in Lacaune sheep were the proteoglycans versican ( VCAN ) and inter-alpha trypsin inhibitor ( heavy chain H1 , ITIH1 ) , implicated in follicular fluid formation and osmotic gradient , cumulus expansion , follicular remodeling and finally fertility [15] , [46] . The coagulation factor F5 , the serine protease inhibitor of the serpin family SERPINE2 and the heparan sulfate proteoglycan sulfotransferase ( HS6ST2 ) would be other important regulators of coagulation , controlling antithrombin , plaminogen or fibrinogen activities present in follicular fluid [47]–[51] . Clusterin ( CLU ) , a sulfated glycoprotein can also act on the ovarian function through its protective effect on granulosa cells against apoptosis during follicular atresia [52] , and it is well known that a reduction in atresia is associated with increased ovulation rate [53] . Clusterin is a binding protein of the low-density lipoprotein receptor-related protein-2 ( LRP2 ) [54] , and may participate in cholesterol delivery to steroidogenic cells as LRP8 does in the bovine ovary [55] . Interestingly , in Lacaune sheep follicular fluids , we evidenced the presence of LRP1 that could have the same function on cholesterol uptake but can also act as a TGFß receptor ( type 5 ) and regulate its signaling in ovarian cells [56] . Hemicentins are extracellular matrix proteins implicated in cell contacts , adhesion and migration [57] . Hemicentin-1 ( HMCN1 ) carries a von Willebrand A domain that may explain its glycosylation by B4GALNT2 . Hemicentins interact with fibulins that are estrogen regulated and overexpressed in ovarian cancer cells [58] , [59] , but no direct role of hemicentins has been described in the ovarian function . Finally , all the glycoproteins identified could have a role in the ovarian follicle function , may be through the control of intra-follicular activity of hormones or growth factors and/or their transfer outside of the follicle to the general blood circulation . In conclusion , the present study reports strong evidence for the B4GALNT2 gene to be the FecL fecundity gene in Lacaune sheep . We propose that its overexpression in granulosa cells under the influence of only 1 or 2 non-coding regulatory SNP can induce an atypical glycosylation of follicular target proteins such as inhibin subunits and that is the starting point of the mechanism explaining increased ovulation rate and prolificacy in this breed . For the first time a fecundity gene in sheep does not belong to the TGFß/BMP signaling genes and it opens new fields of investigation regarding ovarian glycosylation and the pathogenesis of fertility disorders in women .
The presence of the Lacaune autosomal fecundity locus and its prolific allele FecLL was checked in our experimental Lacaune meat strain flock ( n = 189 ) as described [8] . The three genotypes at the FecL locus are called +/+ , L/+ and L/L representing FecL+/FecL+ , FecLL/FecL+ and FecLL/FecLL , respectively . Briefly , as a unique haplotype is associated with the FecLL mutation , the presence of this particular mutant haplotype was established by the genotyping of four close markers , including the DLX3:c . *803A>G SNP that alone provides accurate classification of animals ( 99 . 5% ) as carriers or non-carriers of the mutation . The absence of the FecXL mutation was checked in the studied Lacaune ewes by direct genotyping of the mutation [3] . The estrus cycles of all adult Lacaune ewes were synchronized with intravaginal sponges impregnated with fluorogestone acetate ( FGA , 40 mg , Intervet , Angers , France ) for 14 days . All procedures were approved by the “Direction Départementale des Services Vétérinaires de Haute-Garonne” ( approval number C31-429-01 ) for the agricultural and scientific research agency INRA ( French National Institute for Agricultural Research ) , and conducted in accordance with the Guide for the Care and Use of Agricultural Animals in Research and Teaching . Genomic DNA was extracted from blood samples following a salt-based DNA extraction [60] . Firstly , 45 long-range PCR fragments spanning 488 kb were amplified on genomic DNA from the heterozygous L/+ ewe n°982140 , chosen as a dam of a double-recombinant ewe able to reduce the locus and because of its high level of homozygosity within the FecL locus . On a second time , 25 long-range PCR fragments spanning 250 kb were amplified from a homozygous L/L ewe and a homozygous +/+ ewe . The L/L ewe n°991012 was chosen at random , as it exists only one L-haplotype in the studied population . The +/+ ewe n°011182 was chosen for its homozygosity all along the locus because several wild-type haplotypes segregated in the population . Long-range PCR fragments were amplified on an ABI 9700 thermocycler ( Applied Biosystems ) using the Long PCR Enzyme Mix ( Fermentas ) . Independently for each animal , the resulting fragments were purified and pooled all together at equal concentrations . These samples were then sequenced using the Roche 454 Life Sciences Genome Sequencer FLX ( 454 Life Science , Roche ) , following the manufacturer's instructions . Three shotgun libraries were prepared with 1 µg of pooled PCR product DNA using the Titanium General Library Preparation Kit . Nebulized , purified , and adaptor-linked DNA fragments were amplified using the GS FLX Titanium LV emPCR Kit , and sequencing on the FLX Genome Sequencer was performed using the GS FLX Sequencing Kit , Titanium Reagents XLR70 . L/+ , L/L and +/+ sequencing data from fastq files generated by 454 sequencer were cleaned using an in-house algorithm . A total of 356 873 reads with an average length of 366 bases were aligned on the sheep genome ( OAR v2 . 0 - released March 2011 - draft sheep reference genome ) with bwa software ( bwasw algorithm [61] , indicating a mean sequencing deepness of 103X . The resulting SAM format files were processed using samtools view , sort and merge functions [62] . Screening of polymorphisms was done “manually” using IGV software [63] . The position of all described polymorphisms within this manuscript is given according the last release of the sheep genome ( OAR v3 . 1 - released October 2012 - sheep reference genome [64] ) . Prior to 454 sequencing , the search for single nucleotide polymorphisms ( SNPs ) was performed from ovine ESTs and BAC end sequence information , on a set of 2 L/L and 8 +/+ animals by single strand conformation polymorphism ( SSCP ) and silver staining as described [8] . Identified polymorphic fragments , after SSCP or 454 sequencing , were amplified by PCR on ABI 9700 thermocycler ( Applied Biosystems ) . Microsatellites genotyping through fluorescent SSCP was performed on an ABI 3100 sequencer ( Applied Biosystems ) . Depending on markers , SNP genotyping was done by SSCP and silver staining , or SSCP with fluorescent primers as described in Applied Biosystems Publication 116AP01-02 , or by direct Sanger sequencing using the ABI Prism BigDye Terminator v3 . 1 Cycle Sequencing kit ( Applied Biosystems ) on ABI 3100 sequencer ( Applied Biosystems ) . The particular FecXL mutation was genotyped by SSCP as described [3] . After identification , the allelic sharing of a given marker was tested first on a subset of at least 2 to 6 +/+ Lacaune ewes , then the marker was tested on successive subsets of wild chromosome providing from: the Lacaune genetic family ( n = 103 ) , the dairy Lacaune GEBRO population ( n = 173 ) and the Blanche du Massif-Central ( BMC ) population ( n = 148 ) that shared common ancestors . Additional wild chromosomes from 9 different breeds genetically unlinked to the Lacaune breed were also used ( n = 180 , Limousine , Bizet , Rava , Suffolk , Causse du Lot , Charmoise , Préalpes du Sud , Berrichon du Cher , Noire du Velay , 10 animals of each breed ) . Estrus-synchronized ewes ( +/+ , n = 13 and L/L , n = 14 ) were slaughtered during the follicular phase 36 h after FGA sponge removal . Ovaries , pituitary gland and hypothalamus were collected from each animal . Pituitary gland and hypothalamus were immediately frozen in liquid nitrogen and stored at −80°C for further RNA extraction . Some ovaries were immediately placed in fixative Bouin's solution and embedded in paraffin for further histochemistry . Other ovaries were finely dissected to isolate individual antral follicles >1 mm in diameter . Once dissected , follicles were classified according to their size , small ( 1–3 mm ) and large ( ≥6 mm ) and with respect to genotype , independent of atresia . Granulosa cells and follicular fluids were recovered from small and large follicles as described previously [65] , and theca layer was gently detached from the follicular wall of large follicles with forceps and washed in PBS to eliminate residual granulosa cells . Pools of each category of cells were established per animal and stored at −80°C for further RNA or protein extraction . Total RNA from ovarian cells and from pituitary gland was isolated using Nucleospin RNA II or Nucleospin RNA L kit , respectively , according to the manufacturer's protocol ( Macherey-Nagel ) . Total RNA from hypothalamus was isolated by urea/LiCl precipitation and phenol extraction [66] . All RNA samples were DNAse-treated to avoid genomic DNA contamination and diluted at 0 . 5 µg/µl in RNAse-free water . RNA ( 1 µg ) was reverse-transcribed using Superscript II reverse transcriptase ( Invitrogen ) . Real-time quantitative PCR was run on a LightCycler 480 system ( Roche Diagnostic ) using Power SYBR Green PCR Master Mix ( Applied Biosystems ) in 384 wells-plate as described [8] . The specific primer sequences ( Table 3 ) used for each gene were designed using the Beacon Designer 7 software ( Premier Biosoft International ) . For each primer pair , efficiency curves were generated using serial dilutions of cDNA in abscissa and the corresponding cycle threshold ( Ct ) in ordinate . The slope of the log-linear phase reflects the amplification efficiency ( E ) derived from the formula E = e ( −1/slope ) . Amplification efficiency obtained for each primer pair is indicated in Table 3 . For quantification analysis , the Ct of target gene was compared with the internal reference gene RPL19 encoding an ubiquitous ribosomal protein , according to the ratio R = [EL19CtL19/EtargetCt target] expressed in percentage . Paraffin embedded ovaries were serially sectioned at a thickness of 7 µm . For immuno-histochemistry , deparaffinized sections were subjected to antigen unmasking solution ( Vector Laboratories ) boiled for 3 min . After two 5-min washes with 0 . 1% saponin in PBS , sections were incubated at 4°C for 30 min with PBS containing 0 . 1% saponin and 0 . 3% H2O2 to remove endogenous peroxidase activity . After three 5-min washes in 0 . 1% saponin in PBS , sections were incubated at 4°C overnight with goat polyclonal anti-B4GALNT2 antibody ( sc-107334 , Santa Cruz Biotechnology ) diluted 1∶50 or mouse monoclonal KM694 ( anti-Sda ) antibody kindly provided by Shigeyuki Yamano ( Kyowa Hakko Kirin Co . , Japan ) diluted 1∶1000 in PBS containing 0 . 1% saponin and 0 . 1% BSA . After three 5-min washes in PBS containing 0 . 1% saponin , sections were incubated for 2 hours at room temperature with the biotinylated secondary antibody ( donkey anti-goat , Santa Cruz Biotechnology; donkey anti-mouse , Jackson ImmunoResearch Laboratories ) diluted 1∶800 in PBS containing 0 . 1% saponin and 0 . 1% BSA , and washed thereafter . For lectin-histochemistry , deparaffinized sections were subjected to Carbo-Free blocking solution ( Vector Laboratories ) for 30 min . After a 5-min wash in PBS containing 0 . 05% Tween-20 ( Sigma ) , sections were incubated with PBS containing 0 . 5 µg/mL of biotinylated DBA lectin ( Dolichos Biflorus Agglutinin , Vector Laboratories ) for 2 hours at room temperature , with or without 0 . 1 M of N-Acetyl-D-galactosamine ( Sigma ) to check for its specificity and then washed in PBS containing 0 . 05% Tween-20 . For both approaches , stained sections were incubated with avidin-peroxidase conjugate from Vectastain Elite ABC kit ( Vector Laboratories ) and developed with 0 . 4 mg/ml DAB ( 3 , 3′-diaminobenzidine tetrahydrochloride dehydrate; Sigma ) and 0 . 012% H2O2 in 50 mM Tris-HCl ( pH 7 . 8 ) for 1 to 5 min at room temperature . Negative control sections involved omission of the primary antibody or the lectin from the procedure . Granulosa cells from small antral +/+ follicles were seeded at 100 000 viable cells/chamber on Lab-Tek 8 chambers slide system ( Thermo-Scientific ) and cultured for 48 h at 37°C with 5% CO2 , in McCoy's 5a medium ( Sigma ) supplemented with 3% fetal ovine serum ( FOS ) . Cells were transiently transfected with 1 µg of empty pCDNA3 . 1 or pCDNA-hB4GALNT2 ( kindly provided by A . Harduin-Lepers ) using jetPEI transfection reagent ( Polyplus Transfection ) for 18 h with a DNA/JetPEI ratio of 1/2 ( w/w ) as specified by the manufacturer , and thereafter , medium was changed with fresh McCoy's 5a medium supplemented with 3% FOS for an extra 30 h . At the end of the culture period , cells were fixed in 4% paraformaldehyde for 10 min at 4°C . Slides were washed in PBS 0 . 1% saponin , and then stained with biotinylated DBA lectin ( 0 . 5 µg/µL ) as described above . Granulosa cell whole cell extracts were obtained by resuspension in RIPA lysis buffer as previously described [67] . Granulosa cell lysates and follicular fluids from large antral follicles were centrifuged at 15000 g for 20 min at 4°C , and the protein concentration in the supernatant was determined by a colorimetric assay ( BC Assay kit; Uptima Interchim ) . Protein samples corresponding to 50 µg of granulosa cell extract or 200 µg of follicular fluid were incubated with 30 µl of agarose-coupled DBA lectin ( Vector Laboratories ) in 300 µL of RIPA buffer for 2 hours at room temperature on a rotating platform . Alternatively , 250 µg of protein from follicular fluid were incubated with 25 µL of protein A-sepharose ( Vector Laboratories ) and KM694 antibody ( 1∶300 ) . After a brief centrifugation ( 10000 g , 2 min ) , supernatant was discarded and the precipitated protein complex linked to agarose or sepharose beads was washed 3 times with 1 mL RIPA buffer . After wash , the complex was resuspended in Laemmli sample buffer ( Bio-Rad ) , fractionated using SDS-PAGE in 10% polyacrylamide gels and transferred onto nitrocellulose membranes . Membranes were either blocked for 30 min at room temperature in Carbo-Free solution , and stained with 0 . 5 µg/mL biotinylated-DBA lectin ( Vector laboratories ) for 1 h at room temperature in PBS , or blocked with 5% non-fat dry milk , 0 . 1% Tween 20 ( Sigma ) in PBS , stained for 2 h with KM694 antibody ( 1∶1000 ) , followed by incubation with peroxidase-conjugated anti-mouse IgG ( 1∶2500; Jackson ImmunoResearch Laboratories ) for 1 h at room temperature . Glycoproteins precipitated by DBA lectin were revealed by ABC reaction ( Vectastain , Vector Laboratories ) , followed by ECL Plus detection ( GE Healthcare ) . Glycoproteins precipitated by KM694 were revealed by direct ECL Plus detection . Luminescence was captured using an Image Master VDS-CL box imager ( Amersham Pharmacia Biotech ) . For mass spectrometry identification of glycoproteins purified by DBA lectin affinity , proteins ( 1 mg ) from a mix of follicular fluid of large and small follicles ( one mix for each genotype ) were incubated with 60 µL agarose-coupled DBA lectin in RIPA buffer , as indicated above . After the last wash step , purified glycoproteins were eluted from the precipitated complex through incubation with 0 . 2 M of N-Acetyl-D-galactosamine ( Sigma ) for 90 min at room temperature under stirring . After centrifugation ( 10000 g , 2 min ) , the supernatant was collected and frozen for subsequent analysis . Eluted samples were submitted to SDS-PAGE in 10% polyacrylamide gels ( 12 min at 90 V ) and stained by Coomassie blue . Proteins were in-gel digested with trypsin as previously described [68] . Each peptide mixture from +/+ and L/L genotype was analyzed in triplicate by nanoflow liquid chromatography tandem mass spectrometry ( nanoLC-MS/MS ) . All experiments were performed on a LTQ Orbitrap Velos Mass Spectrometer ( Thermo Fisher Scientific , Bremen , Germany ) coupled to an Ultimate 3000 RSLC chromatographer ( Dionex , Amsterdam , The Netherlands ) . Samples were loaded on an LCPackings trap column ( Acclaim PepMap 100 C18 , 100 µm i . d×2 cm long , 3 µm particles ) and desalted for 10 min at 5 µL/min with 4% solvent B . Mobile phases consisted of ( A ) 0 . 1% formic acid , 97 . 9% water , 2% acetonitrile ( v/v/v ) and ( B ) 0 . 1% formic acid , 15 . 9% water , 84% acetonitrile ( v/v/v ) . Separation was conducted using a LCPackings nano-column ( Acclaim PepMap C18 , 75 µm i . d×25 cm long , 3 µm particles ) at 300 nl/min by applying gradient consisted of 4–45% B for 120 min . The mass spectrometer was operated in data dependent scan mode . Survey full scan MS spectra ( from 300–1800 m/z ) were acquired in the Orbitrap analyser with R = 30 000 . The 20 most intense ions were fragmented in the high-pressure linear ion trap by collision-induced dissociation . Dynamic exclusion was active during 30 s with a repeat count of 1 . Polydimethylcyclosiloxane ( m/z , 445 . 1200025 ) ions were used for internal calibration . MS/MS ion searches were performed using Mascot search engine v 2 . 2 ( Matrix Science , London , UK ) against the mammalian section of Uniprot_sprot database ( 2012_07 ) . The search parameters included trypsin as a protease with allowed two missed cleavages , carbamidomethylcysteine , methionine oxidation and acetylation of N-term protein as variable modifications . The tolerance of the ions was set to 10 ppm for parent and 0 . 8 Da for fragment ion matches . Mascot results obtained from the target and decoy databases searches were subjected to Scaffold 3 software ( v 3 . 4 . 3 , Proteome Software , Portland , USA ) . The raw data may be downloaded from ProteomeCommons . org linked to the Tranche data repository using the “follicular_fluid_sheep” keywords . Peptide and protein identification was done by the Peptide and Protein Prophet algorithms [69] , [70] with a probability >95 . 0% . Only proteins with greater than four identified peptides were considered . Differentially expressed proteins were determined using the spectral counting quantitative module of Scaffold 3 Q+ software ( version 3 . 4 , Proteome Software , Portland , USA ) . To eliminate quantitative ambiguity into protein groups , we ignored all the spectra matching any peptide that is shared between proteins . Thereby , quantification performed with normalized spectral counts was carried out on distinct proteins identified from two biological replicates . T-test was performed to differentiate the significantly changed proteins with a p-value <0 . 05 between the two genotypes . All experimental data are presented as means ± SEM . The genotype effect on gene expression was analyzed using t-test for comparisons between two means . For all analyses , differences with P>0 . 05 were considered as not significant .
|
Prolific sheep have proven to be a valuable model to identify genes and mutations implicated in ovarian function and female fertility . Indeed , fecundity genes of the Bone Morphogenetic Protein ( BMP ) family discovered in sheep was evidenced as genetic candidates to explain female infertility pathologies . Studying French Lacaune sheep breed , we discovered another fecundity gene named B4GALNT2 , encoding a glycosylation enzyme that is not related to the BMP family . The high prolificacy of Lacaune sheep was explained by overexpression of B4GALNT2 in the ovary leading to atypical glycosylation of inhibin , an important hormone regulating ovarian function . Our findings open promising fields of investigation to better understand female fertility disorders .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[] |
2013
|
The Highly Prolific Phenotype of Lacaune Sheep Is Associated with an Ectopic Expression of the B4GALNT2 Gene within the Ovary
|
Dengue vaccines will soon provide a new tool for reducing dengue disease , but the effectiveness of widespread vaccination campaigns has not yet been determined . We developed an agent-based dengue model representing movement of and transmission dynamics among people and mosquitoes in Yucatán , Mexico , and simulated various vaccine scenarios to evaluate effectiveness under those conditions . This model includes detailed spatial representation of the Yucatán population , including the location and movement of 1 . 8 million people between 375 , 000 households and 100 , 000 workplaces and schools . Where possible , we designed the model to use data sources with international coverage , to simplify re-parameterization for other regions . The simulation and analysis integrate 35 years of mild and severe case data ( including dengue serotype when available ) , results of a seroprevalence survey , satellite imagery , and climatological , census , and economic data . To fit model parameters that are not directly informed by available data , such as disease reporting rates and dengue transmission parameters , we developed a parameter estimation toolkit called AbcSmc , which we have made publicly available . After fitting the simulation model to dengue case data , we forecasted transmission and assessed the relative effectiveness of several vaccination strategies over a 20 year period . Vaccine efficacy is based on phase III trial results for the Sanofi-Pasteur vaccine , Dengvaxia . We consider routine vaccination of 2 , 9 , or 16 year-olds , with and without a one-time catch-up campaign to age 30 . Because the durability of Dengvaxia is not yet established , we consider hypothetical vaccines that confer either durable or waning immunity , and we evaluate the use of booster doses to counter waning . We find that plausible vaccination scenarios with a durable vaccine reduce annual dengue incidence by as much as 80% within five years . However , if vaccine efficacy wanes after administration , we find that there can be years with larger epidemics than would occur without any vaccination , and that vaccine booster doses are necessary to prevent this outcome .
Dengue is currently the most important arboviral disease of humans and has an increasing global public health burden [1] . Worldwide , the combined annual number of infections by the four dengue serotypes has been estimated to be close to 400 million , of which 96 million develop symptomatic disease [2] . Globally , dengue incidence has consistently increased for the last five decades due to geographic expansion and transmission intensification in endemic tropical and subtropical regions [3–6] . Since individuals may be infected multiple times with different viral serotypes , and because re-infection is associated with an increased risk for severe disease , dengue presents unique challenges for prevention and control [7] . Vector control is the only option currently practiced to reduce dengue transmission , with most efforts targeting Aedes aegypti and Ae . albopictus , but these programs provide limited protection and may not be sustainable . Most communities undertaking vector control lack the budget , personnel , and expertise needed to effectively reduce mosquito populations . Although the use of DDT as a vector control measure substantially reduced dengue transmission in the 1960’s and 70’s , vector control efforts in the post-DDT era have not been sufficient to prevent invasion of dengue into new regions [8–10] . Vaccination may soon be available as an additional option for dengue intervention . Six vaccines are in clinical development , but to date only the Sanofi-Pasteur vaccine , Dengvaxia , has completed phase III trials [11] . Phase III trials conducted in Latin America estimated vaccine efficacy of 64 . 7% ( 95%CI [58 . 7 , 69 . 8] ) , while the estimate from a trial in South East Asia was 56 . 5% ( 95%CI [43 . 8 , 66 . 4] ) . Pooled analysis of these two trials indicates vaccine efficacy is significantly higher for participants with pre-existing dengue neutralizing antibodies ( 81 . 9%; 95%CI [67 . 2 , 90 . 0] ) compared to those who were seronegative at the time of vaccination ( 52 . 5%; 95%CI [5 . 9 , 76 . 1] ) . Vaccine efficacy against hospitalization for dengue in Latin America was 80 . 3% ( 95%CI [64 . 7 , 89 . 5] ) and in South East Asia was 67 . 2% ( 95%CI [50 . 3 , 78 . 6] ) , and vaccine efficacy estimates varied by serotype in both trials [12–14] . Overall , these are promising results for Dengvaxia , but trial outcomes have been mixed . Efficacy appears to decline in recipients that have not had a previous natural infection , including an apparent increased risk of hospitalization in pre-school age children . Vaccine rollout plans will need to be carefully evaluated , particularly the age targeted for routine vaccination relative to the age by which people in the target region have typically had an infection [15–18] . To accurately model the impact of potential vaccination strategies , we must account for the primary mechanisms that influence dengue transmission: mosquito population dynamics and behavior [10 , 19 , 20] , seasonal influences on these factors [21 , 22] , human movement and demography [23 , 24] , the build-up of strain-specific immunity in the population through time , and the immune response following re-exposure [6 , 15 , 19 , 25] . Consistent with the global trend , dengue incidence and severity have increased significantly in Mexico over the past four decades , with transmission regularly reported in 28 of the 32 Mexican states . Incidence is particularly well-documented in the state of Yucatán . Dengue was reintroduced to Yucatán in 1979 after widespread DDT use in the region ceased; the virus had not been detected in the state for the previous two decades [26 , 27] . We use an agent-based simulation model fitted to data on dengue occurrence to examine the possible effectiveness of deploying a Dengvaxia-like vaccine in Yucatán . We compare potential vaccine rollout strategies under varying assumptions regarding the duration of vaccine-induced immunity . In particular , we consider routine vaccination targeting different age groups ( 2 , 9 , or 16 year olds ) , with and without one-time catch-up campaigns , and with durable or waning vaccine-induced immunity . Our transmission model extends previously published work that examined dengue vaccination in Thailand with a hypothetical vaccine [28] .
Uninfected mosquitoes acquire dengue virus by biting infectious humans . Infected mosquitoes that survive the extrinsic incubation period ( EIP ) can infect new hosts on subsequent blood-feeding attempts ( Fig 1; see Section S1 in S1 Text ) . Infected humans also incubate the pathogen before they become infectious . Dengue virus comprises four serologically distinguishable lineages , called serotypes . Infection produces lifelong immunity to the infecting serotype and induces temporary cross-protection against other serotypes , but can later enhance severity in a subsequent infection , an effect referred to as antibody-dependent enhancement ( ADE ) . ADE also occurs in infants due to maternal antibodies [37] . We represent infection with all four serotypes , with different disease outcome probabilities depending on number of previous infections . Different disease outcomes ( asymptomatic , mild , and severe ) also influence transmission , since they have shorter ( asymptomatic ) to longer ( severe ) infectious periods . In our model , seasonal rain reliability determines mosquito population size and temperature determines EIP . With these seasonality drivers , after fitting we find the dengue basic reproduction number ( R0 ) falls below 1 for roughly four months each year , indicating that trans-seasonal maintenance via transmission in the local human population is unlikely . The mechanism that causes dengue to re-emerge after the winter in Yucatán is unknown . Plausible explanations include but are not limited to human movements from regions with on-going transmission , infected mosquitoes introduced via e . g . freight , and vertical transmission in the vector . We address seasonal re-emergence using a small , fixed daily exposure probability to represent whatever the real processes are . The serotypes for these exposures are based on the observed serotypes in Yucatán . We classify infections as asymptomatic ( “inapparent” ) or symptomatic ( a “case” ) . Cases are further separated into mild or severe , and for comparisons with empirical data , we assume that mild and severe cases can have different reporting rates . We model the pathogenicity ( probability an infection is symptomatic ) as a reference probability combined with relative-risk factors for particular serotypes , past infection history , and age . We use a similar approach for the probability of severe disease . Finally , in our model infants ( people <1 year old ) may have maternal antibodies , which causes them to either resist infection or have enhanced pathogenicity ( see Section S1 in S1 Text ) . We explicitly model individuals and their activity in a synthetic population within Yucatán , based on census and household data [38 , 39] , and national statistics on local economies and schools [40 , 41] . Individuals have a fixed age , gender , and household , and may travel to school or work during the day . We use gender to determine a mother-child relationship when an infant is exposed so that we can consider maternal antibodies , because the model population structure specifies cohabitation but not familial relationship . For practical reasons , we do not consider changing population size or age structure: fixing these across all simulations dramatically reduces the model’s computational complexity . Household composition and location , as well as associated schools or workplaces , are also static and identical across all simulations . These locations and the distances between them provide all the spatial information in the model , as humans and mosquitoes strictly exist at and move between these places . Using such locations for spatial distribution is convenient because household , school , and workplace data are available from national and international sources , and because it is a natural approximation of where people and mosquitoes interact . Representing how acquired immunity changes over time in a population is critical for modeling long-term epidemic dynamics . To avoid the complexity of a dynamic demographic model , instead of having individuals age , immune histories are annually shifted from younger to older individuals . Thus , the population matures by accumulating immunity as epidemics occur , while maintaining a realistic household age distribution . Because the size of age cohorts trends downward with increasing age , mature immune histories are implicitly lost due to mortality and replaced with newborns who have no acquired immunity , though they may have temporary maternally-derived immune responses ( see Section S2 in S1 Text ) . We model the mosquito population in two parts: aggregate populations for uninfected ( susceptible ) mosquitoes at each location , and mobile agents for infected mosquitoes . The aggregate mosquito populations have location-specific sizes drawn from an exponential distribution with a fitted mean ( see Section S5 in S1 Text ) that varies seasonally as a function of rainfall ( Fig 2 ) . Upon infection , individual mosquitoes are separated from the aggregate population . The mosquito’s age is determined by sampling from the total mosquito age distribution . Her EIP is then drawn from the day’s EIP distribution ( based on effective temperature ) and added to her age to determine at what age she will become infectious . Finally , the mosquito’s age-at-death is determined by sampling from the distribution of mosquito ages greater than or equal to the current age ( see Section S3 in S1 Text ) . Mosquitoes that will die before becoming infectious cannot contribute to disease transmission and effectively die instantaneously . Individual mosquitoes may move between houses , workplaces and schools; if they do move , they randomly select from adjacent locations weighted by the inverse distance-squared distance to those locations . Simulation time has important features at several scales: daily day-night cycles , seasonal change in temperature and rainfall by day-of-year , annual population turnover , and multi-year eras . Each day bridges a day-night cycle , 7 a . m . –7 a . m . People go to work or school ( or stay home if not employed ) during the day and are at their homes during the night; there is no representation of weekends or holidays . Individuals’ daily movement patterns sometimes change in response to disease ( see Section S2 in S1 Text ) , but not otherwise . Mosquitoes are more likely to bite during the day than at night . Seasonal effects on the mosquito population ( driven by precipitation patterns ) and on the EIP ( driven by temperature ) change daily , based on a time series generated from 35 years of historical temperature [42] and precipitation [43] data . Each year , the human population ages on the same day of the year ( Julian day 99; i . e . , April 9 ) , which is roughly the nadir in transmission . Leap-days are not modeled: each year is 365 days long . We do not adjust the observed data to match this assumption . At the multi-year scale , there are four time periods: ( 1 ) a priming period , to establish a stable distribution of acquired immunity in the population; ( 2 ) an intense vector control period , corresponding to the use of DDT in Yucatán ( 1956 to 1978 ) ; ( 3 ) a fitting period ( 1979 [first recorded Yucatán dengue epidemic] to 2013 [last year with complete data] ) ; and finally ( 4 ) a 20 year forecast period ( 2014 to 2033 ) where we consider vaccination strategies . All periods ( priming , DDT era , fitting , and forecast ) are simulated with the same parameters in a particular run , except that during the DDT period , both mosquito populations and external introductions are reduced . We represent vector control by reducing the mosquito populations by 77% and the rate of dengue introductions by 90% , based on the fraction of households that were treated with DDT and other insecticides in Yucatán and neighboring states , respectively , from 1956 to 1978 [44] . In our model , vaccination reduces , but does not eliminate , the probability of infection with dengue given exposure , and has no other direct effect on transmission or disease . Vaccination status is added to an individual’s immune history but is distinct from immunity acquired by natural infection . We consider vaccines that confer durable ( lifelong ) protection , as well as the possibility that vaccine-induced protection declines linearly ( “wanes” ) with the number of days since vaccination . We consider three possible waning half-lives . We assume vaccine efficacy ( VE ) values consistent with the the phase III trial results for Dengvaxia in Latin America [12 , 13] ( see Section S4 in S1 Text ) . Trial results indicated that prior dengue infection approximately doubles the vaccine efficacy . Table 1 gives the values used in the model . Dengvaxia has a three dose schedule , each six months apart . We also model a 3 dose regimen , with the first dose providing full efficacy , and all vaccinees receiving all three doses . For scenarios where the vaccine wanes , it does so between doses as well , but each dose is assumed to return efficacy back to the initial level . During the forecast period ( 2014–2033 ) , individuals in an age category ( 2 , 9 , or 16 year-olds ) are targeted for routine vaccination annually on Julian day 100 , one day after immune histories are transferred . We also consider scenarios in which routine vaccination is supplemented with a one-time catch-up campaign , where vaccination occurs across many age groups ( from one year older than the routine age , up to age 30 ) in the first year of the forecast period . When considering booster strategies for a waning vaccine , booster doses are provided to all previously vaccinated individuals , every two years ( irrespective of waning period ) from the final dose date . Because the sizes of our age groups differ , we cannot simply hold the fraction of individuals vaccinated in an age group ( i . e . , coverage ) constant while assessing the outcomes for targeting different ages: at the population level , this would be assessing different vaccination rates and confound with other differences due to age . Instead , we hold the number of doses constant across routine strategies , and across catch-up campaigns when used . However , we base the number of doses on attaining 80% coverage in 9 year olds when targeting that age ( 30 , 100 vaccinees ) , and 60% coverage in the catch-up cohort associated with 9 year olds ( 448 , 500 vaccinees ) . While we have represented some of the features of the Sanofi-Pasteur vaccine , our intent is to represent a generic , moderate-efficacy vaccine . We assume that vaccine performance is affected by serostatus but not age of vaccinee per se , as that has not been specifically tested in the trials . We do not address the potential complexities indicated by trial results in Southeast Asia , particularly any potential for disease enhancement [14] . We assess vaccination strategies by contrasting the projected dengue burden with and without vaccine deployment over a 20 year forecast period . Matched baseline and vaccination scenario runs are simulated for the forecast period , with these comparison runs sharing parameters produced by the fitting procedure ( see Section S5 in S1 Text ) as well as simulated history for the priming , vector control , and fitting periods . We average across 1000 runs ( 100 parameter combinations times 10 samples each ) of the baseline and each scenario to get an expected number of cases each year . In a time interval Δt , the total vaccine effectiveness ( Veff , Δt ) is 1 minus the proportion of the symptomatic cases in the vaccination scenario ( VSΔt ) relative to the number in a baseline with no intervention ( BΔt ) : V eff , Δ t = 1 - VS Δ t B Δ t ( 1 ) We calculate this value for an annual ( Δt = 1 year , beginning 0 , 1 , 2 , … , 19 years after initiation of the intervention ) and cumulative ( Δt = 1 , 2 , … , 20 years , all from initiation of the intervention ) basis , across parameter combinations and replicates . The model uses parameters from a wide range of sources ( see Section S5 in S1 Text ) . Our synthetic human population was constructed using satellite imagery , microcensus , workplace and school data ( see Synthetic Human Population ) . Seasonality in mosquito population and EIP is based on empirical temperature and precipitation time series ( see Section S3 in S1 Text ) . When data were not available to inform or fit the model directly , as with certain vaccine performance parameters , we made assumptions that simplified the model implementation . Epidemiological , entomological , and vaccine parameters were taken from the literature , or fit using Approximate Bayesian Computation ( see Table D in S1 Text ) . We fit our model to reported case and seroprevalence data collected between 1979 and 2013 ( the fitting period , main text Fig 3 and Fig . P in S1 Text ) . We retained the 100 best-performing parameter combinations ( out of 70 , 000–10 , 000 per set , 7 sets ) from the fitting procedure . In our forecasts , we used each of those 100 parameter combinations 10 times , for a total of 1000 replicates . Since transmission parameters vary seasonally , we estimate the basic reproduction number , R0 , by day of year . For a vector-borne disease , R0 may be thought of as the number of additional human infections that are expected to result from a single infected human in a naïve population , after exactly one human-mosquito-human transmission cycle . R0 does not take into account existing immunity in a population , but it nonetheless can provide some intuition about the seasonal timing and peak size of an epidemic . We estimate R0 as follows: for each day of the year , we randomly infect an individual in an otherwise completely susceptible population , allow that person but no other people to infect mosquitoes , run the simulation forward until all infections clear , and count all human infections after the first . We do this for the same 1000 samples used in the forecasting and average the number of secondary infections across the samples to compute R0 for that day .
Using the best parameter combinations from our fitting procedure , we simulated the historical and fitted period outbreaks to establish background immune profiles . Our model generally predicts the size and timing of epidemics during most of the fitting period ( 1979–2008 ) , but not the large epidemics since 2009 . We also forecast transmission from 2014 through 2033 ( the 20 year forecast period ) without any intervention to provide a baseline to compare interventions against . Median results are reported here , and prediction intervals can be found in the supplement ( see Section S7 in S1 Text ) . To generally characterize dengue transmission in the region , we calculate the seasonally-varying R0 for DENV1 . We estimated a seasonal peak R0 of 5 . 2 , occurring in August ( Fig . A in S1 Text , panel C ) . From late December through mid April , R0 is below 1 . 0 . Dengue introductions to the population in the model can happen throughout the year , so stuttering transmission chains are still observed in those months . We did not repeat this analysis for all serotypes , but the others would have generally lower R0 given our assumption that DENV1 has the highest risk of severe disease , and thus longer infectious periods than the other serotypes . We evaluate the performance of the fitted model using seasonality and seroprevalence data that were not used in the fitting procedure . Precipitation and temperature seasonality in the model ( see Section S3 in S1 Text ) drives changing mosquito populations and changing EIPs ( Fig . A in S1 Text ) . These seasonal effects successfully reproduce the overall shape and timing of average weekly dengue incidence ( Fig 2 , average of years for which weekly data are available ) . Seroprevalence , or the fraction of individuals with at least one past dengue infection , is a general indicator of whether an epidemic is possible and how large one might be if it occurs . This relationship is more complicated for dengue given the temporary nature of cross-protection between serotypes and subsequent disease-enhancement , but still provides some insight . To qualitatively assess our model fit , we compared age-stratified results from a recent serosurvey of Mérida [45 , 46] , the largest city in Yucatán , with simulated seroprevalence among Mérida residents in the synthetic population ( Fig 4 ) . Our model results overlap with confidence intervals for the measurements , but are generally low for individuals below age 20 and high for those over 30 . For scenarios that assumed vaccine-induced immunity did not wane over time ( a durable vaccine ) , annual effectiveness gradually increased for routine-only strategies ( Fig 5 ) and spiked early with catch-up campaigns followed by short increasing trend , then a gradual decline to roughly the same level as achieved with routine-only vaccination . Routine-only vaccination started near 0% and increased to 65% annual effectiveness , while strategies with catch-up started near 65% and quickly increased to 75% , but after about 7 years , began to decrease to 65% by end of the forecast period . Varying the target age for routine vaccination had a modest impact on annual effectiveness ( Table 2 ) , although strategies targeting older children generally out-performed those targeting younger children . We expect a positive correlation between target age and effectiveness , based on the anticipated trend in seroprevalence with age ( confirmed for this population at the outset of the forecast period; see Fig 4 ) and our assumption that antibody-primed vaccinees benefit from enhanced vaccine efficacy . However , annual effectiveness for all strategies appeared to be converging by the end of the forecast period . We also considered vaccine-induced immunity that wanes linearly over time ( Fig . D in S1 Text ) , with three different half lives . Under the 2 year half-life waning model , for example , vaccinees have immunity based on Table 1 immediately after each dose , and that efficacy declines linearly each day until it is 0 at 4 years post-vaccination . Waning substantially reduced the long-term effectiveness of routine strategies that did not include additional booster vaccinations ( dashed lines , Fig 6A ) . When vaccinees were re-vaccinated every 2 years , performance improved , reaching 50% effectiveness after 20 years . Waning had a more dramatic effect on strategies with catch-up ( Fig 6B ) . Catch-up campaigns with waning vaccines but no booster vaccination all resulted in negative annual effectiveness—performance worse than baseline—at some point within 20 years . That outcome is the effect of delaying cases in the catch-up cohort: annual effectiveness initially looks good as cases are temporarily prevented relative to the baseline , but when the vaccine effect fades , vaccinees that have avoided natural immunizing infections soon experience infections that have already happened in the baseline , resulting in excess cases in later years . However , cumulative effectiveness shows net case reduction–i . e . some of the cases are actually eliminated rather than just delayed ( see Section S7 in S1 Text ) . Booster vaccination prevented negative annual effectiveness ( lighter solid lines ) , but overall performance was worse than the catch-up scenario with a durable vaccine ( dark solid line ) . The rate at which immunity wanes was an important factor for strategies without booster vaccination , but not for those with it .
We fit an agent-based dengue transmission model to empirical data from Yucatán , Mexico , and then used this model to evaluate a range of vaccination scenarios . For our effectiveness analysis , we used efficacy values based on phase III clinical trial results for Dengvaxia [11–13] . We concluded that a Dengvaxia-like vaccine can be an effective tool for reducing the dengue burden , although a vaccine with waning efficacy would require a booster program . We estimated a cumulative reduction in cases of 74% over 20 years for the most favorable scenario ( Table 2 ) , and scenarios with a durable vaccine converged near 65% reduction in annual case burden after 20 years ( Fig 5 ) . Scenarios with waning vaccines required booster vaccination programs to maintain appreciable effectiveness; however , with boosting , they converge at around 50% annual effectiveness by the end of the forecast ( Fig 6 ) . In general , vaccination strategies that include only routine vaccination at a particular age are much less effective in the first 10 years than those with a one-time catch-up ( Fig 5 ) . However , as the initial catch-up cohort shrinks as a share of all vaccinees ( due to mortality and on-going routine vaccination ) , we expect the annual effectiveness of routine-only and catch-up strategies to converge . For a durable vaccine , our model forecasts that effectiveness will converge around 65% after roughly 20 years . The results for a waning vaccine are more complicated , particularly when there is a catch-up campaign . To address concerns raised in a recently published analysis of long-term follow-up data from the phase III trials [14] , we considered vaccines that provide protective immunity that wanes with a half-life of 2 , 5 , or 10 years . In that study , researchers found no significant reduction in dengue hospitalization risk for vaccinated versus control groups during the third year post-vaccination , suggesting that vaccine-induced protective immunity may begin to wane . Without other adjustments to deployment strategies , we found that vaccination in these waning scenarios provides minimal long-term benefit . Furthermore , if there was an initial catch-up campaign , some years have increased incidence relative to no vaccination , though there are still small cumulative benefits ( see Section S7 in S1 Text ) . For these scenarios , the vaccine initially prevents large epidemics , leading to a decline in naturally acquired immunity compared to baseline scenarios . When the relatively large cohort of catch-up vaccinees then collectively loses its vaccine-induced immunity over a short period of time , larger-than-baseline epidemics can result , which leads to years with expected negative annual effectiveness . However , adding booster doses to the vaccination strategy can substantially offset waning , and results in annual effectiveness around 50% at the end of the forecast period . As expected , vaccines with longer half-lives produce better effectiveness , but all waning vaccines had low long-term effectiveness without a booster program . As a supplementary analysis , we also considered the effect of projected temperature increase associated with climate change on these results ( see Section S6 in S1 Text ) . This sensitivity study suggests that increasing temperature would increase the projected dengue burden , but that estimates of annual vaccination effectiveness are robust to the increasing force of infection . For other public health considerations , such as adherence to dose schedules and compliance with booster campaigns , we anticipate that changing these factors would have obvious directional effects ( e . g . lower coverage will lead to lower effectiveness ) , but there are not sufficient data at this time to make meaningful quantitative predictions . While compliance rates to inform such analyses might reasonably be inferred from other vaccine programs , the most critical issue is unclear: the appropriate model of vaccine action . Until there are appropriate data on Dengvaxia performance , attempts to quantify the nuanced effects of vaccine delivery are premature . For all of our scenarios , we assumed that the vaccine efficacy for individuals who have not had a natural infection ( i . e . , antibody-naïve ) is half of that for those who have had one ( i . e . , antibody-primed; see Table 1 ) . Thus , as the vaccine drives down natural infection rates , it will become less effective , lowering the long-term benefit . Previous analyses have suggested that interventions ( either vector control [47] or vaccination [48] ) in a population that historically experienced high force of infection would initially look effective , but then have declining benefit . We observed this effect for the scenarios with a catch-up cohort: the initially high effectiveness declines after about 10 years to what appears to be a new steady state that reflects both routine vaccination coverage and a reduced level of natural infections associated with reduced transmission ( Fig 5B , solid lines ) . In addition to long-term outcomes , relative vaccine efficacy also influences the effectiveness of vaccination strategies based on the target age for routine vaccination: older vaccinees are more likely to be antibody-primed . This results in higher effectiveness for these strategies , but the effect is modest in the modeled Yucatán population . Therefore , other considerations such as distribution logistics might reasonably take precedence when choosing which age group to vaccinate . In general , our fitting procedure reproduces several features of the observed data well ( Table F in S1 Text ) , but the model is not fit to and is not intended to replicate the exact historical time series . Dengue epidemics in Yucatán are highly variable , undoubtedly influenced by factors we do not consider ( e . g . the circulating serotypes in adjacent regions , inter-annual environmental variation ) . We predict the approximate timing of peaks in 1980 , 1984 , and 1997 due to the introduction of serotypes that had not recently circulated , but we do not predict the large epidemics observed near the end of the fitting period ( Fig 3 ) . As a consequence , we under-predict seroprevalence in young people ( Fig 4 ) , and thus may be under-predicting short-term performance of the vaccine . These recent large epidemics are unlikely to be driven by gradual trends , such as might be captured by improved data on natural history of dengue generally , mosquito ecology in the region , and demographic and economic trends in the region . Large epidemics after quiet years are historically associated with the introduction of novel serotypes . Thus , a more substantial change to the model , such as introducing a novel strain of DENV2 capable of re-infecting people with past DENV2 infections as suggested in [49] , may be necessary to replicate the end of the fitted period and the age distribution of seroprevalence . Despite the model’s inability to reproduce the most recent large epidemics , we believe it is informative for forecasting vaccine performance for two reasons . First , even though the vaccine has twice the efficacy for seropositive versus seronegative recipients , average efficacy is relatively insensitive to changes in seroprevalence: e . g . if 50% of vaccinees are seropositive , overall efficacy to DENV1 is 0 . 45 , while reducing seroprevalence by half to 25% only reduces average efficacy to 0 . 38 , a ∼15% reduction . Second , the increased efficacy in seropositive vaccinees produces a stabilizing effect: if epidemics become large , the vaccine performs better thus driving epidemics smaller , while if epidemics have been small , overall efficacy decreases , permitting larger epidemics . These offsetting effects make population-level effectiveness relatively robust . Our ability to forecast vaccination impact is primarily limited by the current uncertainty regarding whether and how vaccine efficacy wanes over time and how vaccine efficacy is affected by prior infection . Nevertheless , our model provides a useful perspective on how vaccine properties and strategic choices affect the relative size and severity of projected epidemics . The long-term effectiveness of a strategy is a function of vaccination effort , the efficacy and duration of vaccine-induced immunity , and the interaction of these factors with the generation of natural immunity through the underlying transmission process . A Dengvaxia-like vaccine will be a great advance in dengue control in the short-term , but will not be a complete , long-term solution: even under the most optimistic scenarios , our model suggests vaccination alone cannot eliminate dengue in Yucatán . Many different dengue mitigation strategies are actively being developed [50–52] , and in the long term these may supplement or replace Dengvaxia . As the vaccine is deployed to control epidemics in the near future , development of additional mitigation strategies should continue alongside further study of vaccine performance , to provide the data needed to inform new strategies and , potentially , identify one capable of eliminating dengue .
|
Dengue is a mosquito-transmitted viral disease that is common throughout the tropics . Despite a long history in humans and extensive efforts to control dengue transmission in many countries , the number , severity , and geographic range of reported cases is increasing . Most control efforts have focused on controlling mosquito populations , but the main vector , Aedes aegypti , flourishes in human-disturbed and indoor environments . Because the mosquitoes prefer to bite during the day when people are active and potentially moving around high-risk locations , fixed barriers like bed nets are not effective . Several dengue vaccines are being actively developed and may become valuable tools in dengue control . Using historical dengue data from Yucatán , Mexico , we fit a detailed simulation of human and mosquito populations to project future transmission , then used efficacy data from vaccine trials to evaluate the benefit of potential vaccination deployment strategies in the region . For a durable vaccine , we find that population-level , annual vaccine effectiveness approaches 65% by the end of the 20-year forecast period . For waning vaccines , however , effectiveness is greatly reduced–and sometimes negative–unless booster vaccinations are used .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"infectious",
"disease",
"epidemiology",
"immunology",
"geographical",
"locations",
"animals",
"vaccines",
"preventive",
"medicine",
"north",
"america",
"infectious",
"disease",
"control",
"vaccination",
"and",
"immunization",
"insect",
"vectors",
"public",
"and",
"occupational",
"health",
"infectious",
"diseases",
"booster",
"doses",
"epidemiology",
"disease",
"vectors",
"insects",
"arthropoda",
"people",
"and",
"places",
"mosquitoes",
"mexico",
"immunity",
"biology",
"and",
"life",
"sciences",
"organisms"
] |
2016
|
Projected Impact of Dengue Vaccination in Yucatán, Mexico
|
Global warming and ocean acidification are forecast to exert significant impacts on marine ecosystems worldwide . However , most of these projections are based on ecological proxies or experiments on single species or simplified food webs . How energy fluxes are likely to change in marine food webs in response to future climates remains unclear , hampering forecasts of ecosystem functioning . Using a sophisticated mesocosm experiment , we model energy flows through a species-rich multilevel food web , with live habitats , natural abiotic variability , and the potential for intra- and intergenerational adaptation . We show experimentally that the combined stress of acidification and warming reduced energy flows from the first trophic level ( primary producers and detritus ) to the second ( herbivores ) , and from the second to the third trophic level ( carnivores ) . Warming in isolation also reduced the energy flow from herbivores to carnivores , the efficiency of energy transfer from primary producers and detritus to herbivores and detritivores , and the living biomass of detritivores , herbivores , and carnivores . Whilst warming and acidification jointly boosted primary producer biomass through an expansion of cyanobacteria , this biomass was converted to detritus rather than to biomass at higher trophic levels—i . e . , production was constrained to the base of the food web . In contrast , ocean acidification affected the food web positively by enhancing trophic flow from detritus and primary producers to herbivores , and by increasing the biomass of carnivores . Our results show how future climate change can potentially weaken marine food webs through reduced energy flow to higher trophic levels and a shift towards a more detritus-based system , leading to food web simplification and altered producer–consumer dynamics , both of which have important implications for the structuring of benthic communities .
Forecasting the effect of global change on ecosystem functioning is a major challenge in ecology [1] , partly because future climate shifts are likely to reorganize complex food webs , generating novel communities composed of new combinations of species [2] . In marine ecosystems , multiple anthropogenic stressors are already eroding biodiversity by changing the composition of species [3] and affecting rates of biomass transfer through ecological networks , resulting in altered food web organisation and dynamics [4 , 5] . While overexploitation is largely responsible for altering the structure and functioning of many ecosystems [6] , global warming is forecast to amplify these effects , having serious consequences for the health and sustainability of marine ecosystems [7] . Despite many studies showing a potential detrimental effect of climate change on biodiversity [2] , we still lack a strong and coherent theoretical and empirical foundation for understanding how species communities are likely to respond to global change [8] . Marine ecosystem functioning is maintained by the flow of energy from primary producers at the base of food webs through intermediate consumers to top predators , as well as via cycling of materials such as nutrients within the ecosystem [7 , 9] . Disturbances such as habitat modification can decouple , alter , or concentrate energy flows towards a smaller number of species and erode resilience by removing alternative feeding pathways in the food web [10] . This can drive trophic food webs to shift states and potentially collapse [11] . In this context , climate change can independently affect , or synergistically amplify , the effect of other disturbances such as habitat degradation , species overexploitation , and species invasions [12] . This can reconfigure future food webs through major structural changes , like shifts in the number of trophic groups and links that connect species at the top of the food chain to basal species , which can result in altered flows of energy and shifts in the biomass of key functional groups , leading to biodiversity loss , with potentially serious implications for ecosystem functioning [13] . Global warming and ocean acidification are already affecting the physiology , behaviour , phenology , demography , abundance , and distribution of many marine species [14] . Elevated temperature ( T ) affects fish performance and growth through increasing metabolic rates and respiratory demands , leading to a reduced aerobic scope for important life-supporting activities such as feeding , somatic growth , maturation , and predator avoidance [15–17] . Ocean acidification raises the energetic costs involved with calcification and acid-base regulation [18 , 19] and can impair neural functioning [20] . Besides such direct effects , there is a suite of indirect effects that can impact species persistence and diversity under global change [21] . The survival of a species or a group of species in an ecosystem depends on how well they respond to dynamic changes in productivity , in terms of direction ( i . e . , positive or negative ) and strength . However , the responses of species to global change are not individual-based; they are connected through biotic interactions within and across trophic levels [22 , 23] . Importantly , the energy flow ( biomass fluxes ) to higher trophic levels is determined by various biological interactions ( e . g . , predator–prey relationships , competition , facilitation , and mutualism ) of species that are directly or indirectly linked to proximate trophic levels [7 , 24] . For example , climate change can weaken the energy transfer efficiency between primary producers and consumers by reducing feeding performance with potential impacts further up the food chain [25] . Although ocean warming and acidification can boost basal productivity [26 , 27] , it does not necessarily result in an increase in secondary productivity [28] . The propagation of production through trophic levels could be modified by food chain length [29] or with an increase in the dominance of herbivory-resistant primary producers [30 , 31] . Cyanobacteria , in the form of toxin-producing deleterious phytoplankton , can potentially divert productivity into alternate food web pathways , which are unavailable to higher trophic levels . Cyanobacteria have been forecast to increase in dominance as a result of global warming [29] . Conversely , in line with metabolic theory [32] , increased food demand in predators can intensify top-down control of their prey populations [16] , influencing body size-driven metabolic differences in food web structure [33] . Thus , a dichotomy between enhanced bottom up control through an increase in herbivory-resistant primary producers and enhanced top-down control due to higher metabolism-driven energetic demand in predators can jeopardise intermediate trophic levels and weaken food web stability . Such a mismatch can alter trophic energy flow dynamics [25 , 34] , with the consequences expected to be greater in food webs with three trophic levels or more [29] . Forecasting the effects of climate change at the ecosystem level requires holistic approaches that consider complex ecological communities with multiple functional groups or species of different trophic levels [35] . Large-scale mesocosm experiments are ideal for such approaches and have the potential to enhance our understanding of the ecological consequences of climate change on the sudden expansion of opportunistic species , species extinction risk , community structure , and ecosystem functions [36] . We did a community-level manipulation of a temperate marine food web , consisting of 17 functional groups ( ranging from primary producers to herbivores to carnivores across 3 broad trophic levels ) . We maintained these groups in an indoor mesocosm experiment divided into 4 treatments: elevated CO2 ( OA ) , elevated temperature ( T ) , elevated CO2 and temperature ( OAT ) , and ambient controls ( present-day levels of pCO2 and temperature ) , each with 3 replicate mesocosms per treatment . We achieved an elevated pCO2 of approximately 900 ppm ( pH = 7 . 89 ) and warming of +2 . 8°C , which represent the conditions predicted for the end of this century , following a business-as-usual emission scenario ( RCP8 . 5 ) [37] . We used an ecosystem modelling tool ( Ecopath ) widely used to characterise quantitative food web structures and pathways of energy flows in aquatic ecosystems [38] . The Ecopath model is built on a system of linear equations and creates a static mass-balanced snapshot of the resources in a given ecosystem according to biomass estimates and food consumption relationships of functional groups that represent the organisms in a food web . The quantitative description of food web properties is essential to advance our understanding of ecosystem structure and functioning at a community level [39] . Using the Ecopath approach , we tested whether: ( 1 ) global warming and ocean acidification enhance energy fluxes through bottom-up effects that stimulate primary productivity , or ( 2 ) global warming and ocean acidification allow opportunistc groups to proliferate and divert productivity into alternative pathways , as well as ( 3 ) biomass of lower trophic levels and detritus will dominate the future structure of marine food webs due to reduced energy transfer efficiencies to higher trophic levels . We also test whether synergies between ocean acidification and increasing temperature are likely to amplify the effect of global warming on marine ecosystems . Thus , through a combination of experimental and modelling approaches , we provide new evidence for altered energy flow ( biomass fluxes ) and transfer efficiency through food webs due to global change stressors , which is crucial for understanding the potential effects of climate change on marine food web structure and functioning .
Neither warming nor acidification affected the energy flow originating from primary producers and detritus at trophic level 1 ( Fig 1; S1 Table ) . The combined effect of warming and acidification ( OAT ) ( p = 0 . 003; post hoc energy flow: OAT < control ) reduced the energy flow from trophic level 1 to trophic level 2 . In contrast , energy flow was higher in the OA-only treatment compared to the controls ( p = 0 . 011; post hoc energy flow: high CO2 > control ) . Warming ( T and OAT ) also reduced the energy flows from trophic level 2 to trophic level 3 ( ANOVA , F1 , 8 = 43 . 06 , p < 0 . 001 ) . The individual functional groups at trophic level 2 and trophic level 3 showed variable responses to warming and acidification . For example , functional groups such as meiobenthos , copepods , small epifaunal invertebrates , and filter feeders experienced significantly lower energy flow from trophic level 1 to trophic level 2 under warming ( T ) , while macroinvertebrates experienced reduced flow under the combination of warming and acidification ( OAT ) ( S1 Fig; S4 Table ) . Furthermore , warming significantly reduced the capacity of transferring energy flows of omnivorous , filter feeding , and benthic carnivorous fish , while benthic carnivorous and carnivorous fish experienced an increase in flow under acidification from trophic level 2 to trophic level 3 ( S2 Fig , S5 Table ) . The reduced energy flow from trophic levels 1 to 2 under the combined warming-acidification treatment ( OAT ) coincided with a negative effect of warming ( T and OAT ) on energy transfer efficiency between levels 1 and 2 ( ANOVA , F1 , 8 = 11 . 22 , p = 0 . 010 ) ( Fig 1; S1 Table ) . In contrast , OA had no effect on energy transfer efficiency between these levels . Energy transfer efficiency between trophic levels 2 to 3 was not affected by either warming or acidification . Whilst the combined effect of warming and acidification enhanced the biomass of primary producers ( p = 0 . 001; post hoc living biomass: OAT > control ) and acidification enhanced secondary consumer biomass ( p = 0 . 034; post hoc living biomass: OA > control ) , warming ( T and OAT ) , irrespective of ocean acidification , caused a decrease in living biomass at trophic levels 2 and 3 ( Fig 2; S2 Table ) . Warming ( T and OAT ) induced higher cyanobacterial biomass ( ANOVA , F1 , 8 = 19 . 90 , p = 0 . 002 ) , replacing palatable turf algae at trophic level 1 ( Fig 3a , S3a Table ) . Consequently , energy was not transferred to successive trophic levels through consumption but accumulated as detrital biomass ( ANOVA , F1 , 8 = 9 . 12 , p = 0 . 017 ) ( Fig 3b; S3b Table ) at the bottom of the food web . The system became less efficient in recycling the accumulated detrital biomass under warming ( S3 Fig; S3c Table; ANOVA , F1 , 8 = 9 . 31 , p = 0 . 016 ) , suggesting a collapse at the base of the food web .
Our study provides strong empirical evidence that global warming has the capacity to drive a collapse in marine food webs by altering energy flows between successive trophic levels . In our ecologically complex benthic mesocosm experiment , the combination of warming and acidification enhanced the biomass of primary producers , but reduced energy flow to herbivores , while warming ( irrespective of acidification ) reduced energy flow to carnivores at higher trophic levels . Warming also decreased the trophic transfer efficiency between primary producers and herbivores , consequently reducing standing biomass of herbivores and carnivores . Other studies based on the metabolic theory of ecology ( MTE ) [32] , however , suggest that temperature-driven increased primary production is likely to propagate through food webs via strong top-down control [40 , 41] , resulting in greater levels of heterotrophic biomass , relative to autotrophic biomass [4] . In our case , the combination of warming and acidification decoupled increased basal productivity from herbivore production , while warming in isolation reduced predator production , making most of the primary production unavailable further up the food chain . Thus , energy from enhanced primary producer biomass under future climate conditions may not always transfer through to successive trophic levels , but instead can decouple food demand and supply . Such a decoupling may alter dietary preferences of consumers , modifying consumer–prey relationships within food webs . Our current inability to capture more realistic features of food web responses to global climate change is mostly due to a reliance on short-term , small-scale experiments harbouring single species and lower trophic levels , which provide an ambiguous approximation of naturally complex food webs [4] . Moreover , few studies of ecological responses to climate change include predator–prey dynamics and competitive interactions or allow for the potential proliferation of opportunistic groups of species , all of which can greatly influence or reverse many of the predicted responses of species and communities to climate change [26] . In contrast , our experimental results account for complex multispecies interactions and biotic processes , providing an improved representation of natural systems and how these are likely to respond to global warming [42 , 43] . Our experimental data provide insights into how anthropogenic climate change can potentially affect food web dynamics for relatively short-lived taxa . This is because large scale mesocosm experiments such as ours bridge the gap between simplified experimental conditions and the real world [44] , providing important opportunities to better understand the likely mechanisms by which primary productivity under certain future climate conditions propagates through the food web . Increased standing biomass in our benthic mesocosm experiment was most evident for cyanobacteria under warming ( irrespective of acidification ) . Warming is known to enhance the primary productivity of some taxa , particularly of weedy species such as turf algae [45] . Cyanobacteria form a major component of turf algae [46 , 47] and are predicted to proliferate and expand under eutrophication and climate change [48] . The potential for cyanobacterial dominance in turf algal assemblages under future warming and acidification has been demonstrated previously using experimental approaches [49] . The effect of warming sea temperatures is expected to be greater in areas that are shallow and nutrient rich [50] , providing ideal environmental conditions for cyanobacteria to invade other benthic primary producer groups [51 , 52] . Some cyanobacteria are known to be toxic and cause localized anoxia and mortality in marine organisms [48] . Cyanobacteria also produce potent allelochemicals that deter feeding and are difficult to control by grazers [53] . Herbivores like macroinvertebrates and small epifaunal invertebrates predominantly feed on mat-forming turf algae rather than cyanobacteria . Metabolic theory suggests that consumers will show a greater response to temperature than producers [4 , 54 , 55] . Therefore , we hypothesised a priori that warming would drive an increase in metabolic and consumption rates for macroinvertebrates in our experiment . However , reduced food availability , brought about by palatable types of turf algae being replaced by unpalatable cyanobacteria , caused food limitation , preventing increased metabolic rates for macroinvertebrates at higher temperatures , suppressing the flow of energy to the second trophic level [16 , 56] . Furthermore , biomass of major prey groups such as copepods , small epifaunal invertebrates , and filter feeders , which largely form the diet of consumers at the third trophic level , collapsed under the warming treatments , resulting in significantly less energy available for the third trophic level . One of the reasons for this collapse is an excessive predation pressure on primary consumers by species at the third trophic level ( i . e . , predators ) due to their higher energetic demand [28] , which was not matched by any increase in productivity rates of primary consumers under warming . Alternatively , under warming , the consumer–producer relationship at the base of the food web could be nonsynchronous if consumption rates of herbivores peak earlier than the growth rates of producers , creating a mismatch between production and consumption [45] . This means that under certain conditions , even in the absence of herbivory-resistant primary producers , warming-induced metabolic stress of organisms can effectively decouple the consumer–producer relationship if consumer metabolism and consumption cannot keep pace with increasing production . In our experiment , the collapse of biomass of major functional groups under warming played a major part in reducing the transfer efficiency of energy between trophic levels in a food web . A lower transfer efficiency of energy was evident between trophic levels 1 and 2 . This is because transfer efficiency depends on both the availability of food , the biomass of all consumers , and their consumption rates . The lower standing biomass at trophic level 2 and reduced palatable food abundance at trophic level 1 collectively brought about lower overall transfer efficiency between these trophic levels under warming . Metabolic theory states that the structure and dynamics of ecological communities are based on the individual metabolism of organisms where the individual metabolic rate is primarily controlled by body size , body temperature , and resource availability [57] . Metabolic theory does not specifically consider the standing biomass of consumers , which is an important component of our model , allowing us to calculate transfer efficiency . Therefore , food web properties , such as transfer efficiency , are difficult to interpret from the perspective of metabolic theory , yet they can have important community-level effects [58] . The replacement of turf algae by cyanobacteria further resulted in a more detritus-based food web under warming . Detritus can be very important for sustaining food webs and ecosystem stability [59] , but only when proper recycling occurs within the system [60] . Decreasing detritus recycling in ecosystems correlates with decreased system resilience to perturbations , with lower rates of recycling resulting in slower recovery [60 , 61] . We show ( based on the Finn's cycling index ) that warming significantly reduced the detritus recycling capacity of the system . The inability of herbivores to consume enhanced primary producer biomass and the simultaneous failure of detritivores to transfer excessive detritus production to the successive trophic levels resulted in an accumulation of detrital biomass at the base of the food chain . Therefore , warming-induced detritus accumulation , as observed in our study , might have far-reaching ecosystem consequences for future oceans , such as the spreading of ‘dead zones’ through increased microbial activity and consumption of dissolved oxygen in bottom waters [30] . Reduced secondary and tertiary production has been forecast under future acidification [2] . However , this pattern was not detected in our experiment in energy flows , a proxy for production at different trophic levels . We show that acidification can in fact exert positive bottom-up effects on energy flow towards secondary producers ( trophic level 2 ) . Ocean acidification could increase secondary productivity in situations where strong indirect positive effects dampen the direct negative effects of elevated CO2 , i . e . , through increased habitat and food , as well as reduced predator abundances [62 , 63] . Only benthic carnivores and carnivorous fish at trophic level 3 experienced increased energy flows under acidification , which supports the finding of increased productivity of a carnivorous fish ( Favongobius lateralis ) in a previous study done in a similar ecological setting [28] . In contrast , omnivorous fish showed a decrease in flow and this group was the major contributor of energy flow to trophic level 3 under control conditions . Taken together , the positive effect of acidification on the energy flows of some functional groups and negative or lack of effects on other groups resulted in no overall significant increase in energy flow to trophic level 3 under acidification . Here we quantify secondary and tertiary production , using a more complex ( and ecologically realistic ) food web , which better captured real-world community structure and important species interactions , and how these are likely to change in response to future global warming and ocean acidification [8 , 39 , 64] . This broader food web model also allows us to quantify energy transfer efficiency across multiple trophic levels; an important ecosystem function which can regulate many ecological processes ( i . e . , trophic structure , food chain length ) and mediate ecosystem services ( i . e . , fisheries production ) [65–67] . Our results indicate that the response of future food webs to ocean acidification and temperature are likely to depend on the localized community composition and consumer–resource interactions of the specific ecosystem . Although we used one of the most complex benthic mesocosm experiments to date , our approach is not devoid of caveats . For example , difficulties in separating the functional roles of turf and cyanobacteria meant that they were modelled as one functional group ‘mat-forming algae’ . In our model , we did not consider regular bacteria ( other than cyanobacteria ) as a separate functional group , but rather considered them under detritus . Thus , detritivores are considered to mainly feed on detritus and its associated bacteria . We opted for not using an extra bacterial compartment because bacteria would largely overshadow any other trophic flows of the system [68] . Our study showed a relatively larger biomass flux between trophic levels 1 to 2 compared to between trophic levels 2 and 3 due to the presence of relatively large-bodied primary consumers ( such as herbivorous macroinvertebrates: Bulla quoyii ) , which were too big to handle for the gape-limited predators in our system . The presence of a wider range of higher-order invertebrate predators , as is the case in natural ecosystems , would have reduced this disproportionally high flux between primary producers and primary consumers by stronger top-down control . However , since the focus of our study was to show relative difference among the climate treatments , the results still provide a valuable quantitative insight into the potential future of some benthic marine ecosystem under two co-occurring global climate stressors . One of the weaknesses of earlier applications of the Ecopath model were assumptions of ‘steady-state’ or equilibrium conditions , meaning that the model outputs should only be considered for the period across which the model input parameters are deemed valid [69] . Ecopath modelling approaches now no longer assume steady-state conditions but instead the model parameterizations are based on a mass-balance assumption over a chosen arbitrary period . The mass-balance approach in Ecopath filters for mutually incompatible estimates of flow [69] . Moreover , under the Ecopath modelling approach , we assumed that mortality for a prey equals consumption of a predator and that all prey are equal in terms of energetic content . Additional care needs to be taken when inferences are drawn from ecosystem models built for highly dynamic systems due to likely nonlinearities in important food web properties of some functional groups , operating at fine temporal scales . Nevertheless , since we used multiple sampling points through time and averages based on multiple replicates for our model input , especially for taxonomic groups that have the potential to show large oscillations with environmental fluctuations , the model outputs are likely to be indicative of near-future ecological states . Lastly , model outputs were tested using the PREBAL diagnostic and pedigree index ( see Supplementary Methods for details on model and data quality ) , and confirmed a stable model that is ecologically robust . In summary , our results suggest that warming has the capacity to drive an energetic collapse at the base of marine food webs , and this effect can propagate to higher trophic levels—subsequently leading to a collapse in species biomass of the entire food web . The underlying mechanism for this collapse is the replacement of preferred turf algae by cyanobacterial biomass that drives the system towards food limitation for herbivores , with detrimental effects on their predators , combined with a switch towards a less efficient detritus-driven system . Several studies have reported an apparent increase in the occurrence of cyanobacteria in marine waters globally as a result of increasing temperatures [52] , and regionally in temperate [70] , tropical [51 , 71] , and subtropical [72 , 73] areas . Thus , these findings are particularly important in the context of climate change , as mismatches in trophic dynamics can decouple linkages between trophic levels driving ecosystems towards simplified , less productive systems , with cascading effects on ecosystem resilience and functioning .
This research was carried out under the approval of the University of Adelaide animal ethics committee ( approval S-2012-193A ) . All the habitats and organisms collections were permitted by the Minister for Transport and Infrastructure and the Government Department of Primary Industry and Regions SA ( exemptions: 9902676 and 9902752 ) . An indoor mesocosm experiment was maintained from February 2015 to July 2015 . A total of 12 circular mesocosms , each holding 1 , 800 L of water were set up inside a large temperature-controlled room to simulate shallow temperate coastal ecosystems typical of the Gulf St . Vincent , South Australia ( S4 Fig ) . All habitats and organisms used in the mesocosms were collected at a depth of 1–5 m within 60 km distance of the mesocosm facility . Each mesocosm comprised of a mosaic of the 3 primary local habitats ( with 2 replicate patches per habitat per mesocosm ) : rocky reef , seagrass , and open sand [74] . Rocky reefs consisted of natural rocks collected in situ and included attached macrophytes dominated by an assemblage of fucoids ( Order Fucales; mainly species belonging to the families Fucaceae and Sargassaceae ) and benthic invertebrates . Rocks were selected to be as similar as possible in terms of presence and cover of major fucoid species . Seagrass habitat was mimicked by artificial green polypropylene ribbon harbouring epiphytes and planted into fine silica sand at a depth of 6 cm . The seagrass habitat resembled the most abundant local seagrass Posidonia spp . [74] and was incubated in situ for 2 weeks to allow for epiphytic colonization . The circular ‘rocky reef' patches and ‘artificial seagrass’ patches were of equal size ( 0 . 42 m diameter ) . The space in between and around these patches was ‘open sand’ habitat , comprising fine silica sand with a depth range between 6–25 cm . The open sand and sand within the seagrass patches were additionally seeded with 0 . 025 m3 natural sediment collected in situ between patches of live seagrass and included all infauna and flora . Fish and invertebrates were introduced into the mesocosm and represented different feeding guilds ( see S6 Table for a list of species associated with the mesocosms , their stocking density , and mean sizes ) . The fishes were selected based on their high local juvenile abundances in shallow coastal waters during summer , while the gastropods came from the rocks used to build the rocky reef patches and were redistributed evenly among all mesocosms . In the flow-through system , unfiltered seawater from 1 . 5 km off-shore and 8 m depth continuously supplied nutrients and planktonic propagules to each mesocosm at 2 , 300 L day−1 . A diffuser was used to form a light circular current in the mesocosms to simulate tidal water movement alternating direction in 6-hour intervals ( S4 Fig ) . A lamp was mounted above each mesocosm with a spectrum close to sunlight , which is roughly equivalent to 72 . 83 ± 24 . 78 μmoles/m2/second Photon flux corresponding to a local water depth of 6–7 m ( 14/10 light-dark cycle , 30-minute dawn and dusk dimming ) . We applied a control temperature of 21 . 0°C in our mesocosm experiment corresponding to the average summer temperature based on a 5-year dataset of 2 local temperature loggers ( 5 m depth , 2010–2015 , SA Water ) . OA at each mesocosm was achieved through a header tank preconditioned to elevated pCO2 levels using pure CO2 ( control system ACQ110 Aquatronica , Italy ) . Additionally , each mesocosm was supported by a 60-L bin bubbled heavily with enriched air at 1 , 000 ppm pCO2 ( PEGAS 4000 MF Gas Mixer , Columbus Instruments , Columbus , Ohio ) or ambient air at 400 ppm pCO2 , to maintain target levels . Submersible titanium heaters inside the 60-L bins were used for the future warming treatments . Temperature and pH were measured daily ( Mettler Toledo SevenGo SG2 , calibrated daily; S5 Fig ) , while salinity was measured fortnightly ( SR6 refractometer; Vital Sine ) . The total alkalinity was also measured fortnightly by Gran titration from water samples ( 888 Titrando , Metrohm , Switzerland ) . The diurnal variability in pH ( S6 Fig ) confirms that our mesocosms were autonomous systems that mimicked natural day-night fluctuations . For a description of other seawater properties , see S7 Table . The fish community , herbivorous macroinvertebrates , small epifaunal invertebrates , filter feeders , and macro-crustaceans were all sampled and counted at the end of the experiment and their biomass measured as wet weight . Biomass of tanaids , copepods , and meiobenthos was determined using benthic samplers ( 6 . 5 cm in diameter and 2 cm depth filled with 1 . 5 cm of mesocosm sand , with 2 replicate samplers per mesocosm ) , which were placed at the bottom of the tanks for about a month , allowing colonization of these species . Samples were collected twice from each mesocosm during the experimental period and pooled for each mesocosm prior to processing . Tanaids , meiobenthos , and copepods were extracted from the sand within the benthic samplers via floatation using a Ludox TM colloidal solution with a specific gravity of 1 . 18 . The animals were then collected using a 120 μm sieve . Microzooplankton biomass was measured following volumetric method [75] by filtering 400 L of water from each mesocosm through a plankton sampler at the end of the experiment . Phytobenthos biomass was measured using the same benthic samplers as above . Two samplers were placed in each mesocosm for biomass measurements . A micro-spatula was used to carefully scrape the thin phytobenthic layer from the upper surface ( approximately 1 mm thick ) of the sand . The remaining sand was filtered through a pre-combusted and pre-weighed Whatman GF/C glass fiber filters to determine the detritus biomass . In the laboratory , photosynthetic pigments were extracted from freeze-dried sand samples ( 0 . 3–0 . 6 g ) with 10 ml 90% acetone . After 48 hours of darkness at –20°C , the samples were stirred in a vortex , centrifuged at 3 , 500 rpm for 15 minutes , and extracts were analyzed in a 6 , 405 UV/Vis , Jenway spectrophotometer and their concentration calculated [76] . Phytoplankton biomass ( measured as Chlorophyll a ) was quantified based on photosynthetic pigment concentration . Four litres of water were filtered from each mesocosm with Whatman GF/C glass fiber filters of 4 . 7 cm diameter , and ground and extracted [76] . Samples for both phytobenthos and phytoplankton were collected twice during the experimental period , and the average of both was used as the model input . To estimate the biomass of macrophytes and mat-forming algae , all habitats ( rock , seagrass , and open sand ) were sampled at the end of the study period . Their wet weight was determined to the nearest 0 . 1 mg . Mat-forming algae were defined as a mix of turf and cyanobacteria . The relative cover of cyanobacteria in mat-forming algae was estimated using the Coral Point Count with Excel Extensions ( CPCe ) Software [77] . In addition , community metabolism was measured as gross oxygen production ( mgO2/m3/min1 ) once per mesocosm at the end of the study . Oxygen concentration was measured in 1 minute intervals over at least 30 minutes ( HQ40d Portable Meter , sensor LDO101 , HachTM ) . A linear regression model of O2 production rate ( where O2 concentration was modelled as function of time ) was fitted and confirmed a high level of precision in the measurement of O2 concentrations across the 12 mesocosms ( mean ± SD , R2 = 0 . 94 ± 0 . 04 ) . We built mass-balanced food web models . Trophic links were weighted by material fluxes among functional groups using Ecopath [78] . We modelled energy or mass flow over a 4 month time step based on local summer conditions today and in the future . We then converted and expressed the model produced energy flow results from the experimental period to values/month to make it more comparable to other systems . Our model used the following input parameters: biomass , represented by the value B; production per unit of biomass , represented by the value P/B; consumption per unit of biomass , represented by the value Q/B; diet matrix; and the model-estimated ecotrophic efficiency , represented by the value EE . The latter is a parameter that is derived from the model , describing the fraction of the productivity that is used in the system . Most of these model parameters were calculated using our empirical mesocosm data , including final biomass ( end-of-experiment ) and diet composition of consumers of various functional groups ( see S8 , S9 and S10 Tables ) . The Q/B ratio for most of the functional groups was calculated using stomach content analysis and in situ feeding trials that incorporated treatment effects . Therefore , it incorporates the direct effects of temperature on metabolism , accounting for estimates of trophic biomass fluxes and efficiencies . In cases where data were not available , data were derived from empirical equations and published information ( see S1 Text ) . Flow and transfer efficiency were based on the trophic aggregation routine [69] that aggregates the entire system into discrete trophic levels sensu [79] . The discrete trophic levels start with level I , corresponding to the primary producers and the nonliving , detrital compartments . Strict herbivores or detritivores consequently occupy a position of level II . This is followed by higher-order consumers that are allocated to several discrete trophic positions according to the type and amounts of food that reach them along feeding pathways . Energy flows were calculated for different trophic levels following [80] , where—for example—if a group obtains 40% of its food as an herbivore and 60% as a carnivore , the corresponding fractions of the flow through the group are attributed to the herbivore level and the carnivore level , respectively . The relative flows ( these are proportions adding up to 1 ) were converted to absolute amounts and shown as the net amount of energy that flows to higher trophic levels through consumption ( g/m2/month ) . The ‘transfer efficiency’ is the percentage of trophic flows at trophic level n that is converted into flows at level n + 1 . The transfer efficiency of a given trophic level ( trophic level = n ) not only depends on the available energy at a given trophic level ( trophic level = n ) but also the standing biomass at the next trophic level ( trophic level = n + 1 ) . The transfer efficiencies between successive discrete trophic levels were calculated as the sum of the flow that is transferred from any given level to the next higher level , plus exports from the original level relative to the throughput ( or input ) of the given ( originating ) trophic level [69 , 78] . The throughput is the sum of all flows ( such as consumption , export , respiration , and flows to detritus ) in a given trophic level and represents the ‘size of the trophic level in terms of flow’ [81] . Calculation of transfer efficiency from trophic level 1 to 2 is not possible without having information on gross primary production or respiration [82] . We measured net productivity and respiration in each mesocosm and used them to estimate the transfer efficiency between trophic level 1 and 2 for each mesocosm food web [69 , 78] . The initial output for both energy flow and transfer efficiency was obtained for discrete trophic levels I to V . However , for the simplicity of the model output and better visualization , we pooled the data for trophic levels III to Vand showed this as 1 integrated trophic level ( i . e . , trophic level 3 ) . We used the Ecopath pedigree routine to quantify the uncertainty associated with the model parameters by recording the origin and quality of the input data and assigning a value of uncertainty or a confidence interval to each input ( e . g . , biomass , P/B , Q/B , diets ) , which are then used to calculate the overall model pedigree index . The pedigree index varies between 0 ( low precision information ) and 1 ( high quality , i . e . , obtained from modelled system and highly precise ) , allowing a description of the quality of the model [78] . Our overall individual model pedigree index of 0 . 71 , and a measure of fit of t = 3 . 819 , indicates a very high quality and robust model compared to 393 previously constructed models from habitats from around the world , for which pedigree values ranged between 0 . 164 and 0 . 676 [83] . More details on the parameterization and model computation are given in the Supplementary Methods . The effects of warming and ocean acidification on food web properties ( response variables: absolute energy flow , transfer efficiency , and standing biomass ) were analysed using two-way ANOVAs . Both climate factors were treated as fixed and orthogonal . Before analysis , normality was checked for all response variables using the Shapiro-Wilk test , and homogeneity of variances was assessed using a Levene’s test as well as by evaluating plots of residuals against predicted values . Response variables were log10 transformed prior to analysis if they did not conform to a normal distribution . For significant interactions , posthoc multiple comparisons adjusted by false discovery rate were performed [84] . All data analyses were done with the software package R version 3 . 2 . 3 [85] .
|
Healthy marine ecosystems are crucial for people’s livelihoods and food production . Global climate stressors , such as warming and ocean acidification , can drastically impact the structure and function of marine food webs , diminishing the production of goods and services . Our ability to predict how future food webs will respond to a changing environment is limited by our understanding of species responses to climate change , which are often tested in isolation or in simplified experimental designs . More realistic predictions of the impacts of climate change on ecosystems requires consideration of entire species communities , including the species interactions that can buffer or exacerbate these impacts . We experimentally tested the effects of warming and acidification , both individually and in combination , on a benthic marine food web in a near-natural ecological setting . Energy flow from the first trophic level ( primary producers and detritus ) to the second ( herbivores ) , and from the second to the third trophic level ( carnivores ) was quantified under these different regimes . We show that warming , either alone or in combination with acidification , can constrain productivity to the bottom of the food web by enhancing cyanobacterial biomass and reducing energy flow to higher trophic levels , thus lowering energy transfer efficiency between producers and consumers . In contrast , increased ocean acidification alone showed a positive effect on herbivores and carnivores . Our finding is important because it demonstrates that future warming could drive marine food web collapses to potentially simplified and less productive coastal systems .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] |
[
"energy",
"flow",
"ecology",
"and",
"environmental",
"sciences",
"marine",
"ecosystems",
"marine",
"biology",
"biomass",
"(ecology)",
"plant",
"science",
"herbivory",
"plant",
"ecology",
"research",
"facilities",
"research",
"and",
"analysis",
"methods",
"ecological",
"metrics",
"mesocosms",
"ecosystems",
"food",
"web",
"structure",
"marine",
"and",
"aquatic",
"sciences",
"plant-animal",
"interactions",
"community",
"ecology",
"ecology",
"trophic",
"interactions",
"marine",
"ecology",
"earth",
"sciences",
"biology",
"and",
"life",
"sciences"
] |
2018
|
Climate change could drive marine food web collapse through altered trophic flows and cyanobacterial proliferation
|
Mycobacterium tuberculosis heparin-binding hemagglutinin ( HBHA ) , a virulence factor involved in extrapulmonary dissemination and a strong diagnostic antigen against tuberculosis , is both surface-associated and secreted . The role of HBHA in macrophages during M . tuberculosis infection , however , is less well known . Here , we show that recombinant HBHA produced by Mycobacterium smegmatis effectively induces apoptosis in murine macrophages . DNA fragmentation , nuclear condensation , caspase activation , and poly ( ADP-ribose ) polymerase cleavage were observed in apoptotic macrophages treated with HBHA . Enhanced reactive oxygen species ( ROS ) production and Bax activation were essential for HBHA-induced apoptosis , as evidenced by a restoration of the viability of macrophages pretreated with N-acetylcysteine , a potent ROS scavenger , or transfected with Bax siRNA . HBHA is targeted to the mitochondrial compartment of HBHA-treated and M . tuberculosis-infected macrophages . Dissipation of the mitochondrial transmembrane potential ( ΔΨm ) and depletion of cytochrome c also occurred in both macrophages and isolated mitochondria treated with HBHA . Disruption of HBHA gene led to the restoration of ΔΨm impairment in infected macrophages , resulting in reduced apoptosis . Taken together , our data suggest that HBHA may act as a strong pathogenic factor to cause apoptosis of professional phagocytes infected with M . tuberculosis .
Tuberculosis remains a serious global problem , although many researchers have made a persistent effort for several decades . Mycobacterium tuberculosis , a major causative agent of pulmonary tuberculosis , is responsible for 1 . 8 million deaths per year worldwide [1] . Innate immune system plays a critical role in antimicrobial host response during the early stage of M . tuberculosis infection . Alveolar macrophages mediate innate immunity by phagocytosing pathogens and are the chief defense against M . tuberculosis , which can survive and replicate within phagocytes [2] . The course of tuberculosis rests on the outcome of the interaction between the bacterium and host macrophage . Therefore , a better understanding of these complex interactions is critical to controlling mycobacterial infection . Many bacterial and viral pathogens utilize various strategies to manipulate host machinery to serve their own needs . Apoptotic cell death has been regarded as an innate cellular response to limit the multiplication of intracellular pathogens [3] , although the precise mechanism of the direct antimicrobial action in infected macrophages undergoing apoptosis is unclear . Generally , infectious intracellular pathogens tend to prevent host cell apoptosis during an early stage of infection . However , they may also induce host cell apoptosis with a specific aim to subvert the host attack , such as immune and inflammatory response , at later stages [4] , [5] . A number of reports have indicated that M . tuberculosis does indeed inhibit host cell apoptosis , while at the same time it induces pro-apoptotic signals . Recent studies showed that only virulent mycobacterial species can inhibit apoptosis induction in primary human alveolar macrophages [6] , THP-1 [7] , [8] , and J774 macrophage cell lines [9] . Virulent M . tuberculosis reportedly induced the apoptotic death of host cells . For example , enhanced apoptotic response was detected in alveolar macrophages recovered from patients with pulmonary tuberculosis [10] , [11] . Extensive apoptosis was also observed in caseating granulomas from lung tissue samples obtained from patients with tuberculosis [12] , [13] . Several apoptosis-inducing factors of M . tuberculosis , such as 19-kDa glycolipoprotein ( Rv3763 ) [14] , PE_PGRS33 ( Rv1818c ) [15] , ESAT6 ( Rv3875 ) [16] , and 38-kDa lipoprotein ( Rv0934 ) [17] are reported . Heparin-binding hemagglutinin adhesin ( HBHA ) is a 28-kDa multifunctional protein found on the surface and culture filtrates of mycobacteria . It has hemagglutination activity and binds to sulfated glycoconjugates such as heparin and dextran sulfate [18] . HBHA interacts specifically with non-phagocytic cells and is essential for the infection of lung epithelial cells and extrapulmonary dissemination of M . tuberculosis [18] , [19] . Protective immunity induced by HBHA is observed in M . tuberculosis-infected mouse models , indicating that HBHA is a protective antigen [20] . Recent studies suggest that HBHA is a useful diagnostic marker for tuberculosis [21] . We also identified and characterized HBHA as a serologically active mycobacterial antigen in a previous study , whereby HBHA binds strongly to the immunoglobulin M of patients with tuberculosis [22] . Although HBHA function in mycobacterial pathogenesis has been extensively studied , the role of HBHA on professional phagocytes , such as macrophages , is still poorly understood . The aim of the present study was to characterize the biological effects of M . tuberculosis HBHA on macrophages . We found that HBHA induced apoptosis in murine macrophages and investigated its underlying mechanism . Here , we show that HBHA treatment caused a loss of mitochondrial transmembrane potential ( ΔΨm ) and the release of cytochrome c from purified mitochondria in vitro , as well as mitochondria of intact cells , and HBHA was efficiently targeted to mitochondria of macrophages .
We first sought to determine whether HBHA could induce macrophage apoptosis . Apoptosis was assessed by quantifying DNA fragmentation , which is considered a hallmark of apoptosis , in the cytoplasmic fractions of dying cells using a commercially available ELISA kit . The incubation of RAW 264 . 7 cells with HBHA resulted in a significant increase in the release of oligonucleosomal fragments into the cytoplasm in both dose- and time-dependent manners as compared to compared to control cells ( Figure 1A and 1B ) . Cell death was significantly greater in cells treated with HBHA as compared to buffer-treated control cells . As lactate dehydrogenase was not detected in the cell culture supernatant during HBHA treatment , the possibility that HBHA-induced death is necrosis was excluded ( Figure S1 ) . We used native antigen 85 complex ( Ag85 ) as an unrelated control protein . The Ag85 of M . tuberculosis is the major secreted protein and fibronectin-binding protein , and shows strong immunoreactivity [23] , [24] . Similar results were observed in bone marrow-derived macrophages ( BMDMs ) ; like PBS-treated BMDMs DNA fragmentation was not detected in Ag85-treated cells , whereas dramatic DNA fragmentation was observed in HBHA-treated cells ( Figure 1C ) . HBHA-induced apoptosis was further confirmed by examining the nuclear morphology of dying cells using a fluorescent DNA-binding agent , 4′-6-diamidino-2-phenylindole ( DAPI ) . As shown in Figure 1D , control cells treated with buffer had intact nuclei . In contrast , within 48 h of HBHA treatment , RAW 264 . 7 cells clearly exhibited condensed or fragmented nuclei indicative of apoptotic cell death . We further analyzed the caspase dependency of HBHA-induced apoptosis . Western blot analysis showed that the cleavage of caspase-3 , caspase-9 , and poly ( ADP-ribose ) polymerase ( PARP ) was evident in cells incubated with HBHA for 48 h ( Figure 1E ) . Inhibition of caspases by a pan-caspase inhibitor , zVAD-fmk , attenuated the HBHA-induced DNA fragmentation , indicating that HBHA induces caspase-dependent apoptosis ( Figure 1F ) . These results suggest that macrophages treated with HBHA undergo caspase-dependent apoptosis . The mitochondrion acts as a central executioner in response to apoptotic stimuli , allowing signals from various inputs to converge [25] . We investigated whether HBHA treatment affected the structural and biochemical integrity of mitochondria . Mitochondrial damage was assessed by examining mitochondrial ΔΨm , which was determined by staining cells with 3 , 3′-Dihexyloxacarbocyanine ( DiOC6 ) , a dye that incorporates into mitochondria with intact membrane potential [26] , for flow cytometric analysis . As shown in Figure 2A , a significant loss of ΔΨm was observed in RAW 264 . 7 cells incubated with HBHA as indicated by a decrease in DiOC6 intensity . Analysis of the time course for examination of ΔΨm onset showed a noticeable dissipation of ΔΨm after 18 h of HBHA treatment , which further decreased with time . A similar result was obtained in BMDMs incubated with HBHA ( Figure 2B ) . These results suggest that mitochondrial damage appears as a subsequent event in the intracellular action of HBHA . Apoptosis at the mitochondrial level involves the oligomerization of the pro-apoptotic protein Bax [27] , leading to permeabilization of the outer mitochondrial membrane ( MOMP ) and release of cytochrome c [28] . We performed immunocytochemistry to detect Bax translocation and cytochrome c release . An antibody recognizing the Bax N-terminus , which is exposed by the activation of Bax and its insertion into the mitochondrial membrane , was used . Figure 3A shows the translocation of Bax distributed evenly in the cytoplasm to the mitochondria in macrophages as evident by the colocalization of Bax with Mitotracker , a potential-sensitive dye specific for mitochondria . In PBS-treated cells , cytochrome c showed a punctate pattern that colocalizes with Mitotracker , whereas the faint signal for cytochrome c and the decreased colocalization with Mitotracker were detected in HBHA-treated cells , indicating cytochrome c release . These results were confirmed by performing subcellular fractionation and Western blot analysis ( Figure 3B ) . HBHA caused a decrease in cytochrome c immunoreactivity in the mitochondrial fraction with a concomitant increase in the cytosolic fraction and vice versa for Bax immunoreactivity . Collectively , these findings suggest that the apoptotic effect of HBHA on macrophages is associated with cytochrome c release and Bax translocation . To determine whether Bax activation is necessary for HBHA-induced apoptosis , we knocked down the level of Bax by transfecting RAW 264 . 7 cells with Bax siRNA . The Bax protein level was significantly reduced in cells transfected with Bax siRNA; Bax protein in control siRNA-transfected cells was unchanged ( Figure 3C ) . We then determined the effect of knockdown Bax on HBHA-induced apoptosis in RAW 264 . 7 cells . As shown in Figure 3D and 3E , HBHA-induced increase in DNA fragmentation was blocked and ΔΨm loss was restored by Bax knockdown , suggesting that Bax activation is required for HBHA-induced macrophage apoptosis . Enhanced reactive oxygen species ( ROS ) production , characteristic of early apoptotic events , can be both a cause and a consequence of changes in ΔΨm [26] , [29] . To examine the involvement of ROS generation on HBHA effects in macrophages , the oxidation of DCF was monitored by flow cytometry and fluorescent microscopy ( Figure 4A ) . Compared to PBS or Ag85 , HBHA significantly induced the increase of intracellular hydroperoxide in macrophages . To determine the requirement of ROS increase in HBHA-induced apoptosis , the effect of HBHA alone or in combination with N-acetylcysteine ( NAC ) , a general ROS scavenger , on DNA fragmentation was assessed . NAC pretreatment effectively inhibited HBHA-induced DNA fragmentation ( Figure 4B ) as well as ROS production , suggesting that ROS increase is essential for the apoptotic response caused by HBHA . Studies have suggested that some infectious intracellular pathogens regulate apoptosis of their host cells by targeting proteins to mitochondrial membranes that either induce or inhibit MMP [30] . We addressed the question of where HBHA is localized in mitochondria of HBHA-treated cells . Therefore , the possibility that HBHA interacts with the mitochondrial compartment was examined . Confocal microscopic analysis revealed the presence of HBHA in the mitochondria of HBHA-treated cells , as evidenced by a significant overlap between HBHA and Mitotracker ( Figure 5A ) . Subcellular fractionation and Western blot analysis consistently showed that large amounts of HBHA were detected in the mitochondrial fraction , but not in the cytosolic fraction , where little HBHA was observed ( Figure 5B ) . In contrast , the minimum of Ag85 were detected in cytoplasmic fraction of macrophage treated with Ag85 , suggesting that it is not able to pass through plasma membrane . Furthermore , to determine whether HBHA was imported into mitochondria , we isolated mitochondria from cells treated with HBHA . The purified mitochondria were subsequently digested with proteinase K . As shown in Figure 5C , HBHA disappeared in mitochondria digested with proteinase K , indicating that HBHA adheres to the outer membrane of the mitochondria . We next determined whether HBHA induced cytochrome c release from isolated mitochondria . As shown in Figure 6A , isolated mitochondria from RAW 264 . 7 cells released cytochrome c after HBHA treatment , whereas the buffer control or Ag85 did not stimulate this release in a cell-free assay . We also examined the effect of HBHA on the collapse of membrane potential in purified mitochondria . For this , mitochondria incubated with HBHA were stained with DiOC6 , and the fluorescence intensity was monitored by flow cytometry ( Figure 6B ) . A significant shift to a lower intensity was observed in mitochondria treated with HBHA as compared to buffer control or Ag85 , indicating the decrease in ΔΨm . These data provide evidence that similar to the event that occurs in macrophages , HBHA can solely induce mitochondrial damage in a cell-free system , indicating that Bax translocation to mitochondria is not essential for of ΔΨm loss and cytochrome c release . HBHA is a secreted protein in M . tuberculosis as well as a surface-associated protein [18] . To examine whether HBHA is also transported to mitochondria during M . tuberculosis infection , BMDMs were infected with H37Rv wild type and mutant disrupted in hbhA . Immunofluorescence microscopy of infected cells revealed that a part of HBHA colocalized with mitochondria ( Figure 7A ) . Purified mitochondrial fraction of these cells contained a considerable amount of HBHA protein , although a large portion of HBHA were observed in cytosolic fraction ( Figure 7B ) . These findings demonstrate that HBHA is efficiently transported to mitochondria of infected macrophages . To analyze the effects of HBHA on macrophages in the context of the bacterium as a whole , we compared the relative ability of M . tuberculosis H37Rv wild type and mutant disrupted in hbhA to induce apoptosis and ΔΨm collapse in macrophages . A reduced DNA fragmentation and an increased intact mitochondria were observed in BMDMs infected with mutant strain compared to its parent ( Figure 7C and 7D ) , which was noticeable when macrophages were infected at MOIs of 5 and 10 but not at an MOI of 25 ( Figure S2A ) . On the other hand , there was no significant difference in LDH release between cells infected with two strains ( Figure S2B ) . Similarly , more significant DNA fragmentation and ΔΨm loss were detected in cells infected with M . smegmatis ectopically expressing HBHA compared to cells infected with the M . smegmatis control . HBHA is involved in the interaction of mycobacteria with alveolar epithelial cells [19] . To determine whether these cells exposed to HBHA undergo apoptosis , human type II A549 pneumocytes were treated with purified HBHA for 48 h . As shown in Figure 8A and 8B , neither DNA fragmentation nor ΔΨm collapse was observed in HBHA-treated A549 cells . Immunofluorescent microscopy showed that a very faint green signal was detected in A549 cells incubated with HBHA , indicating that HBHA enters A549 cells much less efficiently ( Figure 8C , upper panels ) . To confirm this issue , A549 cells were infected with M . tuberculosis wild type and hbhA deficient strains . Like experiments conducted in macrophages , M . tuberculosis infection led to severe ΔΨm dissipation , accompanied by the partial presence of HBHA in mitochondrial compartments ( Figure 8B and 8C , lower panels ) . In contrast , a decrease in the percentage of cells displaying loss of ΔΨm was observed in A549 cells infected with the hbhA deficient strain ( Figure 8B ) . These data suggest that cell entry and targeting to mitochondria of HBHA are essential for ΔΨm loss and apoptotic response .
Programmed cell death is emerging as a major effect of bacterial pathogenesis . Numerous studies have shown that M . tuberculosis infection can increase the rate of macrophage apoptosis [31] , [32] . Pro-apoptotic activities of a growing number of mycobacterial components have recently been described [14]–[17] . Nevertheless , data regarding the identities of the mycobacterial molecules involved and the underlying apoptotic mechanism are still scarce . We showed here that intracellular HBHA is targeted to mitochondria in murine macrophages , which leads to ΔΨm dissipation and eventual apoptosis . Although the possibility that HBHA may interact with cytosolic molecules or other cell compartments cannot be ruled out completely , these connections clearly appear to be insignificant . To our knowledge , the present study is the first description of a mycobacteria-encoded protein stimulating apoptotic cell death via a mitochondria-dependent pathway in macrophages . M . tuberculosis HBHA is a protein that is both surface-associated and secreted . HBHA is involved in the binding of M . tuberculosis to type II pneumocytes , but not to professional phagocytes such as macrophages , and is required for the dissemination of tubercle bacilli from the lungs to other tissues [19] . In this respect , its impact on macrophages has received relatively little attention . However , HBHA was recently demonstrated to have the capacity to bind to complement component C3 , and recombinant HBHA was found to mediate the attachment of latex beads to murine macrophage-like cells in both C3-dependent and -independent manners [33] . M . tuberculosis can bind to the complement receptors and is subsequently introduced into the phagocytic cell [34] . These results raise the possibility of the interaction between HBHA and macrophages during mycobacterial infection . Mitochondria are central organelles in which a variety of key events in apoptosis occur , including the release of cytochrome c , changes in electron transport , ΔΨm collapse , altered cellular oxidation–reduction , and participation of pro- and anti-apoptotic Bcl-2 family proteins [26] . Presently , mitochondria are regarded as the targets for the manipulation of many bacterial and viral pathogens determining the fate of infected host cells [35] . Moreover , mitochondrial damage has been suggested to play a critical role in the outcome of macrophage infection with M . tuberculosis [36] . These findings offer the potential of mycobacterial components for the regulation of programmed cell death at the mitochondrial level . MMP is regulated by endogenous molecules , including Bcl-2 family members such as Bax [37] . The Bax present in the cytosol under normal conditions fosters the loss of ΔΨm and releases cytochrome c and apoptosis-inducing factor ( AIF ) from mitochondria after its introduction into the mitochondrial compartment [26] . Indeed , mitochondrial translocation of Bax was observed in macrophages treated with HBHA , and the interaction of HBHA with mitochondria resulted in cytochrome c release in murine macrophages . However , Bax translocation may not be essential for mitochondrial dysfunction by HBHA , as evidenced by a mitochondrial cell-free assay in which HBHA caused ΔΨm loss and cytochrome c release in vitro . Not surprisingly , we observed the activation of caspases 3 and 9 and subsequent cleavage of PARP after incubation of macrophages with HBHA . In contrast , we found no evidence of cytosolic or nuclear translocation of AIF induced by HBHA ( data not shown ) , indicating that it is not involved in HBHA-induced cell death . ROS generation with ΔΨm modulation and caspase-9 activation is known to be a major component of the mitochondrial pathway of apoptosis [38] . ROS are predominantly produced in the mitochondria and lead to the modulation of ΔΨm , which finally results in apoptosis [39] . Our results indicate that HBHA induces macrophage apoptosis through ROS generation and ΔΨm collapse , suggesting that these play an essential role in HBHA-induced apoptosis . Our results indicate that cellular entry is essential for mitochondria-mediated apoptotic effect of HBHA , although the mechanism by which HBHA internalized by host cells remains unresolved . In A549 cells infected with M . tuberculosis but not cells incubated with purified HBHA , the severe ΔΨm collapse and the presence of intracellular HBHA in mitochondrial compartment were observed . There was a significant increase in the percentage of cells with intact ΔΨm , when A549 cells were infected with the mutant strain lacking HBHA gene . We cannot rule out that these results might come from decreased number of mycobacteria in cells , because invasion of A549 cells , but not macrophages , by HBHA-deficient strain compared with parental strain was reduced [19] . Moreover , HBHA induced ΔΨm loss and cytochrome c release in purified mitochondria from not only RAW 264 . 7 cells but also mouse liver ( data not shown ) . Thus , no impact on viability of epithelial cells treated with HBHA might be due to the absence of intracellular this protein . What host molecules physically and functionally interact with intracellular HBHA and how do they then induce mitochondrial dysfunction ? In the present study , proteinase K digestion in vitro showed that intracellularly inserted HBHA is attached to the mitochondrial surface but is not imported into mitochondria , indicating that HBHA probably interacts with integral outer membrane molecules . Several mitochondria-targeted proteins encoded by pathogens interact with voltage-dependent anion channel ( VDAC ) . The porin B from N . meningitidis is a VDAC-targeted protein [40] . Hepatitis B virus X protein also co-localizes to mitochondria where it interacts with a particular VDAC isoform , HVDAC3 [41] . Anti-apoptotic members of the Bcl-2 family , such as Bcl-2 and Bcl-xL , are located in mitochondrial membranes where they inhibit cytochrome c release from mitochondria and thereby prevent downstream caspase activation . Pro-apoptotic members of the Bcl-2 family , such as Bax , can translocate into mitochondria and induce MMP [42] . These Bcl-2-like proteins can be prominent targets of bacterial proteins [30] , [43] . Recombinant HBHA used in the present study was a His-tagged fusion protein . To determine the interaction between HBHA and VDAC or the Bcl-2 family proteins , HBHA and interacting molecules were purified by Ni-NTA affinity chromatography , followed by immunoblotting against them . However , HBHA showed no direct interaction with VDAC or Bcl-2 family members ( data not shown ) . In addition , the possibility of HBHA nonspecific binding to mitochondria cannot be excluded . The C-terminal region of HBHA contains several cationic lysine-rich repeats where methylation can occur [44] . This region may work like natural antibiotic peptides which form cationic residues on one end and interact with anionic molecules such as phospholipids to disrupt negatively charged membranes and result in apoptosis [45] . Virulent M . tuberculosis induces necrosis of the infected macrophages by inhibiting the repair process of plasma membrane; this leads to cellular lysis and reinforces the spreading to the adjacent infection sites [46]–[48] . Recent reports suggested that high intracellular burden of virulent M . tuberculosis induces host cell death via a new caspase-independent apoptotic pathway involved in the bacterial escape and extracellular replication [49] , [50] . Because gain of function mutation in HBHA enhanced the apoptogenic potency of M . smegmatis ( Figure 7C ) , it is plausible that HBHA may be the factor that allows M . tuberculosis to escape from the infected macrophages at high intracellular burden . However , similar levels of apoptosis were observed between macrophages infected with M . tuberculosis H37Rv wild type and mutant disrupted in hbhA at an MOI of 25 but not low MOIs ( Figure S2A ) . Further , HBHA deficiency had no influence on the macrophage necrosis caused by M . tuberculosis at both low and high MOIs ( Figure S2B ) , indicating no involvement of HBHA in bacterial escape from the macrophages at an early stage of infection . Studies on the comparison of virulent and attenuated mycobacterial strains have demonstrated that the latter has much stronger apoptotic activity in macrophages . This concept is supported by the identification of genes that inhibit apoptosis of host cells [51]–[53] . In this sense , our claims that HBHA targets to the mitochondria of host cells in the induction of apoptosis may be confusing . However , there is cumulative evidence suggesting that virulent M . tuberculosis induces host cell apoptosis . Furthermore , the transcriptional profiling of cells infected with virulent M . tuberculosis showed increases in the expression of both pro- and anti-apoptotic genes [54] , [55] . Collectively , it is highly likely that M . tuberculosis infection results in pro- and anti-apoptotic response of host cells . The final outcome may depend on the nature and activation status of the host cell . Although the pro-apoptotic response is inarguably beneficial to the host , it may provide a favorable circumstance for the induction of necrotic cell death and subsequent bacterial escape to the adjacent cells , which may provide a clue for HBHA function during M . tuberculosis infection [46] , [49] , [50] . Taken together , the present study suggests the possibility that the M . tuberculosis HBHA may be an apoptosis-inducing factor of mycobacteria , although the molecular mechanism by which HBHA causes loss of ΔΨm remains unknown . Future work should focus on the exploration of host targets of HBHA and the mechanism by which HBHA modulates ΔΨm and cytochrome c release in detail , as well as identification of the HBHA domain essential for its activity in mitochondrial dysfunction .
All animal procedures were approved by the Institutional Animal Care and Use Committees of Chungnam National University ( Permit Number: 2010-3-9 ) . All animal experiments were performed in accordance with Korean Food and Drug Administration ( KFDA ) guidelines . Antibodies against caspase-3 , caspase-9 , and VDAC were purchased from Cell Signaling Technology Inc ( Beverly , MA ) . The anti-PARP and anti-β-actin , anti-Bax , and anti-Tom40 antibodies were obtained from Santa Cruz Biotechnology ( Santa Cruz , CA ) . Antibodies against cytochrome c ( for immunofluorescence , clone 6H2 . B4; for Western blot analysis , clone 7H8 . 2C12 ) were acquired from BD Pharmingen ( San Diego , CA ) , and the anti-cytochrome oxidase subunit IV ( COX IV ) antibody was purchased from Abcam ( Cambridge , UK ) . Dichlorodihydrofluorescein diacetate ( H2DCFDA ) , DAPI , and DiOC6 were obtained from Molecular Probes ( Eugene , OR ) and zVAD-fmk and NAC were purchased from Calbiochem ( San Diego , CA ) . Mycobacterium smegmatis strains , recombinant HBHA protein from M . smegmatis , and antiserum to HBHA were produced and prepared as described previously [22] . Ag85 was purified from the culture filtrate protein of M . tuberculosis H37Rv ( ATCC 27294 ) , as previously described by Lim et al [24] . Parental and mutant ( hbhA deletion ) Mycobacterium tuberculosis 103 were kindly provided by Dr . Camille Locht ( Institut Pasteur de Lille , Lille , France ) [19] . HBHA proteins were used in experiments after lipopolysaccharide ( LPS ) inactivation with polymyxin B ( Invivogen , San Diego , CA ) , a known pharmacological antagonist of LPS . RAW 264 . 7 murine macrophage cell line and A549 human alveolar epithelial cell line were cultured in Dulbecco's modified Eagle's medium ( DMEM; Lonza , Walkersville , MD ) supplemented with 10% fetal bovine serum ( FBS; Hyclone , Logan , UT ) , 1% HEPES , and 1% l-glutamine at 37°C with 5% CO2 . BMDMs were obtained from 6–8-week-old female C57BL/6 mice . Briefly , bone marrow cells from the femur and tibia were cultured in DMEM that contained 2 mM l-glutamine , 100 U/mL penicillin , 100 µg/mL streptomycin , 10% FBS , and 25 ng/mL recombinant mouse M-CSF ( R&D system , Minneapolis , MN ) at 37°C with 5% CO2 . After 4 days , non-adherent cells were removed and differentiated macrophages were incubated in antibiotic-free DMEM until use . One day before transfection , RAW 264 . 7 cells were plated and grown at 37°C to 70% confluency in complete medium without antibiotics in 6 well plates . One micrograms of a Bax siRNA ( Bioneer , Deajeon , Korea , sense: CCGGCGAAUUGGAGAUGAA; anti-sense: UUCAUCUCCAAUUGGCCGG ) or a noncomplementary siRNA were transiently transfected into RAW 264 . 7 using Lipofectamine 2000 transfection reagent ( Invitrogen , Carlsbad , CA , USA ) , according to the manufacturer's instructions . Cells were seeded in 96-well flat-bottom culture plates . After incubation with recombinant HBHA proteins , cells were collected , washed with PBS , and processed for quantification of cytoplasmic histone-associated DNA fragments formed during apoptosis using an enzyme-linked immunosorbent assay ( Cell Death Detection ELISA PLUS; Roche Diagnostic ) according to the manufacturer's instructions . The release of LDH from RAW 264 . 7 cells incubated with recombinant HBHA or from BMDMs infected with M . tuberculosis was measured using a Cytotoxicity Detection Kit plus ( Roche , Indianapolis , IN ) according to the manufacturer's protocol . Relative cytotoxicity was calculated using the following equation: Cytotoxicity ( % ) = % of LDH released from the infected cells/maximum LDH released . ΔΨm was assessed by measuring retention of the lipophilic cationic dye DiOC6 in mitochondria . Cells were harvested and incubated in a DiOC6 solution ( 10 nM in fresh medium ) for 20 min at 37°C in the dark . The cells were then washed and resuspended in PBS . Immediately after PBS washing , ΔΨm was measured by sorting the cells using FACSCanto ( BD Biosciences ) . Dead cells were excluded by forward and side-scatter gating . Data were acquired by analyzing an average population of 10 000 cells using CELLQuest software ( BD Biosciences ) . Cells were seeded onto glass coverslips in 12-well plates . Nuclear changes were analyzed by DAPI staining . After cells were incubated with HBHA for the indicated times , they were fixed with 4% paraformaldehyde and incubated with DAPI ( 10 µg/mL ) for 10 min in the dark . The nuclei of stained cells were visualized using an Olympus BX50 fluorescence microscope ( Olympus Optical Co . , Hamburg , Germany ) . To determine the localization of cytochrome c or Bax , cells treated with HBHA were incubated in pre-warmed medium containing 100 nM of Mitotracker Red ( Molecular Probes ) , fixed in 4% paraformaldehyde , permeabilized with 0 . 1% Triton X-100 , and then stained with anti-cytochrome c or anti-Bax and Alexa-488-conjugated secondary antibody ( Jackson Immuno Research Laboratories ) before confocal microscopy . The subcellular localization of HBHA was analyzed using a confocal microscope ( LSM510 META; Carl Zeiss ) . The cells incubated with Mitotracker Red were fixed , permeabilized , and stained with an anti-HBHA antibody followed by a fluorophore-conjugated antibody ( anti-mouse IgG Alexa-488 ) . After DAPI staining , cells were imaged with a confocal microscope . Subcellular fractionation was performed as previously described [56] . Briefly , cells were incubated on ice for 5 min in 100 µL of ice cold CLAMI buffer ( 200 mM sucrose , 70 mM KCl , 200 µg/mL digitonin in PBS ) and centrifuged at 1 , 000 × g for 5 min at 4°C . The supernatants ( cytosolic fractions ) were stored at −80°C and the pellets were resuspended in 50 µL of IP buffer ( 50 mM Tris-Cl , pH 7 . 4 , 150 mM NaCl , 2 mM EDTA , 2 mM EGTA , 0 . 2% Triton X-100 , 0 . 3% NP-40 ) containing protease inhibitor cocktail ( Roche Diagnostics Corporation , Indianapolis , IN ) and incubated on ice for 10 min . The samples were centrifuged at 10 000 × g for 5 min at 4°C and the supernatants ( mitochondrial fractions ) were stored at −80°C until use in further experiments . Cells were detached , centrifuged , and lysed in lysis buffer ( 10 mM Tris , pH 7 . 4 , 5 mM EDTA , 150 mM NaCl , 1% Triton X-100 , 1 mM PMSF , protease inhibitor cocktail ) . Protein concentrations were determined with the Bradford assay and 30 µg of protein was separated with SDS-PAGE , followed by electrotransfer to a nitrocellulose membrane ( Hybond-ECL; Amersham Pharmacia Biotech ) . The blots were probed with primary antibodies at optimized concentrations followed by horseradish peroxidase-conjugated secondary antibodies . The enhanced chemiluminescence system ( ECL; Amersham/GE Healthcare ) followed by exposure to chemiluminescence film was used to visualize proteins . Intracellular ROS were evaluated through staining cells with H2DCFDA . Cells were incubated in 10 µM H2DCFDA for 30 min at 37°C , washed , and detached . Resuspended cells were washed and immediately analyzed by flow cytometry using FACSCanto . At least 10 , 000 cells per sample were analyzed using CellQuest Pro acquisition and analysis software . Mitochondria were isolated from 1 × 108 RAW 264 . 7 cells as described previously [57] . Briefly , cells were harvested by centrifugation at 600 × g and resuspended in ice-cold IB buffer ( 10 mM Tris-MOPS , 200 mM sucrose , 1 mM EGTA/Tris , pH 7 . 4 ) . All subsequent centrifugations were performed at 4°C . The cells were then homogenized with 35 strokes in a glass potter after incubation for 10 min on ice . Cell debris was removed by centrifugation at 600 × g for 10 min , and then the supernatant was centrifuged for 10 min at 7 000 × g to precipitate mitochondria . The pellet was then resuspended in EB buffer ( 10 mM Tris-MOPS , 125 mM KCl , 100 µM EGTA/Tris , 1 mM KH2PO4 , pH 7 . 4 ) . An aliquot of the preparation was incubated with HBHA for 1 h at 37°C and centrifuged for 10 min at 7 000 × g . The pellet containing mitochondria was resuspended in the same buffer and stained with DiOC6 . An average population of 50 000 mitochondria was analyzed by flow cytometry . Alternatively , the proteins contained in the supernatant were concentrated with by ultrafiltration using a 3-kDa cutoff Centricon device ( Amicon , Millipore , Bellerica , MA ) . Immunoblot analysis for cytochrome c was performed as described above . The data represent the mean ± standard deviation ( SD ) from at least three independent experiments . Statistical analyses were performed using unpaired Student's t tests with Bonferroni adjustment . A P-value of <0 . 05 was considered significant .
|
Cell death is a common outcome during infection with a number of pathogenic microorganisms . Therefore , defining the factors responsible for killing of host cells is important to uncovering mechanisms of pathogenesis . World-wide , two billon people are latently infected with Mycobacterium tuberculosis , which is still killing 2–3 million people each year . Heparin-binding hemagglutinin ( HBHA ) protein of M . tuberculosis is known to interact specifically with non-phagocytic cells and to be involved in dissemination from lungs to other tissues . Nevertheless , the role of HBHA in phagocytic cells such as macrophages , which are the first cells of the immune system to encounter inhaled pathogens , has been unknown . In the present study , we suggest HBHA as a critical bacterial protein for macrophage cell death . After M . tuberculosis infection or HBHA treatment of macrophages , HBHA targeted to mitochondria and then caused mitochondrial damage and oxidative stress , which eventually lead to apoptosis . A mutant of M . tuberculosis lacking HBHA induced less apoptosis with moderated mitochondrial damage . These experiments provide a candidate virulence factor which may be a novel target for tuberculosis treatment .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"medicine",
"bacterial",
"diseases",
"infectious",
"diseases",
"mycobacterium",
"immunity",
"innate",
"immunity",
"microbial",
"pathogens",
"biology",
"microbiology"
] |
2011
|
Targeting of Mycobacterium tuberculosis Heparin-Binding Hemagglutinin to Mitochondria in Macrophages
|
Many studies have presented virus sequences which suggest the existence of a variety of putative new phleboviruses transmitted by sandflies in the Old World . However , in most of these studies , only partial sequences in the polymerase or the nucleoprotein genes were characterised . Therefore to further our understand of the presence and potential medical importance of sandfly-borne phleboviruses that circulate in southern Anatolia , we initiated field campaigns in 2012 and 2013 designed to identify , isolate and characterise phleboviruses in sandflies in this region An entomological investigation encompassing 8 villages in Adana , Mediterranean Turkey was performed in August and September 2012 and 2013 . A total of 11 , 302 sandflies were collected and grouped into 797 pools which were tested for the presence of phleboviruses using specific primers for RT-PCR analysis and also cell culture methods for virus isolation . Seven pools were PCR positive , and viruses were isolated from three pools of sandflies , resulting in the identification of two new viruses that we named Zerdali virus and Toros virus . Phylogenetic analysis based on full-length genomic sequence showed that Zerdali virus was most closely related with Tehran virus ( and belongs to the Sandfly fever Naples species ) , whereas Toros virus was closest to Corfou virus . The results indicate that a variety of phleboviruses are co-circulating in this region of southern Anatolia . Based on our studies , these new viruses clearly belong to genetic groups that include several human pathogens . However , whether or not Toros and Zerdali viruses can infect humans and cause diseases such as sandfly fever remains to be investigated .
The genus Phlebovirus ( family Bunyaviridae ) currently contains 9 viral species Sandfly fever Naples ( SFNV ) , Salehabad ( SALV ) , Rift Valley fever ( RVFV ) , Uukuniemi ( UUKV ) , Bujaru ( BUJV ) , Candiru ( CRUV ) , Chilibre ( CHIV ) , Frijoles ( FRIV ) and Punta Toro ( PTV ) including 33 distinct serotypes , and 32 tentative serotypes as defined in the 9th Report of the International Committee on Taxonomy of Viruses ( ICTV ) [1] . Nevertheless , the past decade has witnessed the discovery of many new phleboviruses that remain to be classified: some are transmitted to vertebrates by sandflies ( Fermo ( FERV ) , Granada ( GRAV ) , Punique ( PUNV ) ) [2 , 3 , 4] , others by ticks ( Heartland ( HRTV ) , Hunter island group ( HIGV ) ) [5 , 6] , whereas some do not have recognised vectors and appear to be transmitted directly between vertebrates ( Malsoor ( MALV ) , Salanga ( SGAV ) ) [7 , 8] . In the Old World , sandfly-borne phleboviruses are transmitted between vertebrates mainly by female sandflies ( genus Phlebotomus ) when they take a blood meal . Some Old World sandfly-borne phleboviruses may cause self-limiting febrile illnesses ( sandfly fever ) or neuro-invasive infections . They are widely distributed in the Mediterranean Basin , in Africa , in the Indian subcontinent , in the Middle-East , and in far-eastern former USSR republics [9] . Annually , Toscana virus ( TOSV ) , a serotype of SFNV is the leading cause of meningitis from May to October in central Italy [10] and one of the most prevalent human pathogenic phleboviruses in other southern European countries . Forty years ago , seroprevalence studies showed that Sandfly fever Sicilian virus ( SFSV ) and SFNV were present in the Mediterranean and Aegean regions of Turkey [11 , 12] . Recently , serological investigations were carried out in the Mediterranean , Aegean , and Central Anatolian regions , where outbreaks have occurred and circulation of SFSV and a SFS-like virus ( Sandfly Fever Turkey virus ( SFTV ) ) were reported [13 , 14 , 15] . The presence of TOSV was confirmed serologically and through RNA detection and sequencing [16 , 17 , 18 , 19] . Despite the publication of many articles , virus isolations were reported only for SFTV from a patient [13] and Adana virus ( ADAV ) [20] from sandflies . To further understand of the dynamics of sandfly-borne phleboviruses and sandfly fever in the Mediterranean region in the vicinity of Adana city , we organized sandfly trapping campaigns .
Sandflies were captured during August and September in 2012 and in 2013 in Adana city located in the Mediterranean region of Turkey ( Fig 1 ) using CDC Miniature Light Traps as previously reported [21] . Live sandflies were pooled according to sex , trapping site and day of capture , with up to 30 individuals per pool and placed in 1 . 5mL tubes , and stored at -80°C . In order to reduce the time between capture and storage and therefore to increase the likelihood of virus isolation , morphological identification of the sandflies was not performed . Sandfly pools were processed as previously described [22] in a final volume of 600μL , of which 200μL were used for total nucleic acid extraction using the Virus Extraction Mini Kit the BioRobot EZ1-XL Advanced ( both from Qiagen ) . Elution was performed in 90μL of extraction buffer of which 5μL were used for RT-PCR and nested-PCR assays using primers targeting the polymerase gene and the nucleoprotein gene as previously described [20 , 23 , 24] . PCR products of the expected size were column-purified ( Amicon Ultra Centrifugal filters , Millipore ) and directly sequenced . Two real-time RT-PCR assays ( Rt-RT-PCR ) were designed for specific detection of the new strains in the N gene: TORV-N-FW ( AACTCTGACTCGTGTGGCTG ) , TORV-N-REV ( GCCTTGGGTATGTCTGACCA ) , and TORV-N-Probe ( 6FAM-AGGCAATAGAAGTTGTGGAGAAC-TAMRA ) ; ZERV-N-FW ( ACTTCCTGTTACTGGAACAACAAT ) , ZERV-N-REV ( CCATGAGCATCTGCAATAACTTC ) , and ZERV-N-Probe ( 6FAM-ATGATGCATCCTAGTTTTGCAGGA-TAMRA ) . Reaction conditions and cycling programs were previously described [20] . Fifty μL ( derived from sandflies ground in the 600μL of EMEM as aforementioned ) were inoculated onto a 12 . 5 cm2-flask of Vero cells , incubated at room temperature for 1 hr , and supplemented with 3mL of EMEM ( 5% FBS , 1% Penicillin/Streptomycin , 1% L-Glutamine 200 mM , 1% Kanamycin , and 3% Fungizone ) . The flasks were incubated at 37°C in 5% CO2 atmosphere and examined daily for cytopathic effects ( CPE ) . For detailed characterisation , Zerdali virus ( ZERV ) passage 5 , Toros virus ( TORV ) strain 292 , passage 3 , and TORV strain 213 , passage 7 were subjected to complete genome characterisation using Next Generation Sequencing ( NGS ) . Briefly , 140μL of infectious cell culture supernatant medium was incubated with 30 U of Benzonase ( Novagen 70664–3 ) for 7 hr at 37°C . This material was then purified using the Viral RNA Mini Kit ( Qiagen ) . Tagged random primers for reverse transcription ( RT ) and tag-specific and random-primers were used for PCR ( Applied Biosystems ) . The resulting PCR products were purified ( Amicon Ultra Centrifugal filters , Millipore ) ; 200ng of DNA were used for sequencing using the Ion PGM Sequencer ( Life Technologies SAS , Saint Aubin , France ) . NGS reads , of 30 nucleotides minimum length , were trimmed using CLC Genomic Workbench 6 . 5 , with a minimum of 99% quality per base and mapped to reference sequences . Parameters were set such that each accepted read had to map to the reference sequence for at least 50% of its length , with a minimum of 80% identity to the reference . From the contigs obtained , viral sequences were identified by best BLAST similarity against reference databases . Sequence gaps were completed by amplification and sequencing overlapping regions using either Sanger sequencing or NGS . The 5' and 3' extremities of each segment were sequenced using a primer including the 8-nt conserved sequence as previously described [25] . Complete genome sequencing was also performed for Corfou virus ( CFUV ) strain PaAr814 using the frozen cell culture supernatant medium following the methods above for comparison with the newly discovered TORV strains since they were shown to be closely related but only the complete S [26] , the partial L [4] and M [27] genome sequences of the CFUV were known . Ultimately , all complete sequences obtained using NGS were verified by amplification and Sanger sequencing of overlapping regions spanning the entire genome . Complete sequences of each of the 5 genes ( L , Gn , Gc , N , Ns ) were aligned without indels together with homologous sequences of selected phleboviruses retrieved from the Genbank database using CLUSTAL within the MEGA 5 program [28] . Nucleotide ( nt ) and amino acid ( AA ) distances were calculated with the p-distance method . Neighbor-joining ( p-distance model ) and Maximum likelihood analyses were carried out with AA sequences using MEGA version 5 , with 1000 bootstrap pseudoreplications . The Recombination Detection Program v . 4 . 27 ( RDP4 ) was used for recombination analysis using the nucleotide alignments . Recombination events , likely parental isolates of recombinants , and recombination break points were analyzed using RDP , GENECONV , Chimaera , MaxChi , BOOTSCAN , and SISCAN algorithms implemented in the RDP4 program with default settings [29] . To attempt identification of the sandfly species present in the TORV and ZERV positive pools , PCR was performed using 3-μL of nucleic acid extract of the pool to amplify the cytochrome c oxidase I ( COI ) gene using the following primers; LCO1490: GGTCAACAAATCATAAAGATATTGG and HCO2198: TAAACTTCAGGGTGACCAAAAAATCA [30] . The PCR products were processed and sequenced through NGS as described above . NGS reads were compared with available sequences in Genbank by Blastn using the CLC Genomic Workbench 6 . 5 . For the final determination of the species the sequences were aligned with the reference sequences of regional populations of the sandfly species . However , we would like to acknowledge that a valid protocol would be to cut off the male genitalia using a cold-stage microscope in the laboratory , so that the specimens can be identified morphologically . This would be faster and cheaper than PCR amplification followed by NGS or Sanger sequencing of the COI gene .
A total of 11 , 302 ( 4 , 513 females and 6 , 789 males ) sandflies were collected in August and September 2012 and 2013 from eight villages ( Fig 1 ) located in the surroundings of Adana city ( Mediterranean Turkey ) . They were organized as 797 pools ( 494 females , 303 males ) ( Table 1 ) . Two pools , #213 and #292 , were positive with primers N-phlebo2S/2R and N-phlebo1S/1R [24] , respectively . The 245-nt sequence obtained from pool #213 was most closely related with CFUV ( Genbank no: GQ165521; 95% AA identity , 78% nt identity ) . The 513-nt sequence corresponding to pool #292 was also closely related with CFUV ( Genbank no: GQ165521; 88% and 78% identity at the AA and nt level , respectively ) . These 2 pools consisted each of 20 male sandflies trapped in Damyeri village in 2012 ( 36S0733357 North and 4140570 East , altitude 194m ) . The pool #37 ( 20 males trapped in Zerdali village in 2013; 36S732947 North and 4142749 East , altitude 238m ) was positive using primers SFNV-S1/S2 [23] . The corresponding 390-nt sequence was most closely related to THEV ( Genbank no: JF939848; 96% AA identity , 85% nt identity ) . The TORV specific rt-RT-PCR confirmed that pools #213 and #292 were positive ( Ct values < 26 ) . Pool #10 ( 20 females collected in Damyeri in 2013 ) was also positive for TORV . Four-fold dilutions of the RNA were positive until the dilution 1:256 for the pools #213 , #292 , and #10 . The ZERV specific rt-RT-PCR confirmed pool #37 was positive ( Ct value = 26 . 33 ) , and detected ZERV RNA in 3 additional pools ( #128–20 females , #342–20 males , and #374–29 females ) . Four-fold dilutions of the pool #37 RNA on the one hand and of pools #128 , #342 and #374 on the other were positive until dilutions 1:4 , 096 and 1:1 , 1024 , respectively . Pools#128 , #342 and #374 consisted of sandflies trapped in 2012 in the respective villages of Damyeri , Camili and Koyunevi ( Fig 1 ) . The rate of infection for TORV was 0 . 026% , for ZERV 0 . 035% , and for both TORV and ZERV 0 . 062% assuming that only one sandfly was infected in each pool . Vero cells inoculated with pool #292 showed a clear CPE at day 6 post-inoculation ( pi ) . Pool #213-inoculated Vero cells did not produce CPE during 4 serial passages . However , virus replication was demonstrated by RT-PCR ( N-phlebo1 system , 24 ) starting from passage 3 . CPE appeared at day 4 pi at passages 4 and 5 and virus replication was confirmed by RT-PCR . In a similar manner , pool #37- inoculated Vero cells provided a clear CPE at day 4 pi of passage 3 ( RT-PCR was positive at passage 2 ) . Neither virus isolation nor positive RT-PCR was obtained after 5 serial passages for pools #10 , #128 , #342 , and #374 . Freeze-dried suspensions of ZERV-strain #37 ( passage 8 ) , TORV-strain #292 ( passage 5 ) and TORV-strain #213 ( passage 8 ) have been included in the collection of the European Virus Archive ( www . european-virus-archive . com/ ) where they are publicly available for academic research at non-profit costs . The complete genomes of both strains ( #213 and #292 ) of the TORV consisted of 6 , 456 nts , 4 , 326 nts and 1 , 702 nts for the L , M and S segment , respectively ( Genbank acc . no of the strain 213; KP966619 , KP966620 , andKP966621; Genbank acc . no of the strain 292;KP966622 , KP966623 , and KP966624 ) . The polymerase gene encoded a 6 , 270-nt long ORF ( 2 , 090 AA ) , whereas the glycoprotein gene encoded a 4 , 077-nt long ORF ( 1 , 359AA ) . The small segment encoded a 738-nt and a 780-nt long ORF which when translated corresponded to the nucleocapsid protein ( 246 AA ) and a non-structural protein ( 260 AA ) , respectively . The complete genome of the ZERV ( strain #37 ) consisted of 6 , 403 nts , 4 , 202 nts and 1 , 907nts for the L , M and S segment , respectively ( Genbank acc . KP966616 , KP966617 , and KP966618 ) . The polymerase gene encoded a 6 , 285-nt long ORF ( 2 , 095 AA ) , whereas the glycoprotein gene encoded a 4 , 002-nt long ORFs ( 1 , 334 ) . The small segment encoded a 942-nt and a 762-nt long ORF which were translated to a nucleocapsid protein ( 314AA ) and a non-structural protein ( 254AA ) , respectively . The complete genome of the CFUV consisted of 6 , 453nts , 4 , 329nts , and 1 , 704nts for the L , M and S segment , respectively ( Genbank acc . no KR106177 , KR106178 , and KR106179 ) . The polymerase gene encoded a 6270-nt long ORF ( 2 , 090AA ) , whereas the glycoprotein gene encoded a 4 , 077-nt long ORFs ( 1 , 359 AA ) . The small segment encoded a 738-nt and a 780-nt long ORF which were translated to a nucleocapsid protein ( 246AA ) and a non-structural protein ( 260AA ) , respectively . Pairwise distances of the nt- and AA- sequences are presented in S1 Table . The alignment of each gene is also available in S2 Table . AA distances between TORV and SFSV-like viruses ( SFSV , SFTV , SFCV , CFUV ) were ≤25 . 2% ( N ) , ≤37 . 8% ( NS ) , ≤43 . 3% ( M ) , ≤40 . 7% ( Gn ) , ≤33 . 7% ( Gc ) and ≤20 . 6% ( L ) , whereas AA distances between TORV and other phleboviruses were much higher: ≥ 42 . 9% ( N ) , ≥71 . 4% ( NS ) , ≥58 . 7% ( M ) , ≥53 . 3% ( Gn ) , ≥ 47 . 4% ( Gc ) and ≥43 . 9% ( L ) . AA pairwise distances between ZERV and viruses of the SFNV species ( TOSV , THEV , SFNV , PUNV , MASV , GRAV ) were ≤17 . 3% ( N ) , ≤58 . 0% ( NS ) , ≤42 . 7% ( M ) , ≤41 . 8% ( Gn ) , ≤29 . 5% ( Gc ) and ≤17 . 4% ( L ) , whereas AA distances between ZERV and other phleboviruses were much higher: ≥ 40 . 6% ( N ) , ≥80 . 9% ( NS ) , ≥66 . 3% ( M ) , ≥64 . 8% ( Gn ) , ≥55 . 0% ( Gc ) and ≥44 . 5% ( L ) . Gene by gene comparative analysis showed that distances observed between ZERV and viruses belonging to the SFNV species were consistently lower than the lowest distances observed between ZERV and non SFNV-phleboviruses . The same relationship was observed with distances between TORV and SFSV-like viruses on the one hand , and TORV and non-SFSV-like viruses on the other . These findings are supportive for ( i ) the inclusion of TORV in the SFSV species complex ( SFSV , SFTV , SFCV , CFUV ) , ( ii ) the inclusion of ZERV in the SFNV species complex ( TOSV , THEV , SFNV , PUNV , MASV , GRAV ) . Regardless of the gene used for phylogenetic analysis , and the tree-building programme ( i . e . NJ or ML ) TORV clustered with SFSV , CFUV and the other SFS-like viruses ( from Turkey , Cyprus , Ethiopia ) with bootstrap values ≥ 99% ( Figs 2 , 3 , 4 , 5 and 6 ) . TORV consistently grouped together with CFUV ( ≥ 99%bootstrap ) forming a subgroup within the SFS-like viruses that is distinct from the second subgroup including SFSV and related genotypes originating from Italy , Turkey , Cyprus , and Ethiopia . The stability of the topology and relationships between TORV and most closely related viruses suggested that the TORV genome did not contain evidence of genetic recombination or reassortment . Likewise , no recombination events were detected using any of the 6 algorithms implemented in RDP4 . ZERV consistently grouped together with THEV and Naples virus strain Yu_8–76 ( subgroup I ) , with bootstrap values at ≥ 99 for L , Gn and Gc , and with lower values for N and Ns . Within this species , there were 3 other subgroups: ( i ) subgroup II: Toscana viruses; ( ii ) subgroup III: Naples viruses ( except for Yu_8–76 included in subgroup I ) ; ( iii ) : subgroup IV: Granada , Massilia and Punique viruses . Partial sequences were obtained in the polymerase gene for pool #128 ( 505 nt ) and #10 ( 379 nt ) , and in the nucleoprotein for pools #128 ( 280 nt ) , #342 ( 438 nt ) , and #374 ( 245 nt ) . Partial polymerase sequences of the pools of #128 and #10 were identical ( except 2 nts and 1 nt , respectively but 100% identical in AA ) with ZERV and TORV , respectively . The partial nucleoprotein sequence from the pool #128 was also identical with the ZERV ( except 1 nt but 100% identical in AA ) . However , the partial nucleoprotein sequences from the pool #342 ( 6 nt and 3 AA different ) and #374 ( 49 nt and 2 AA different ) were not identical with ZERV although originating from neighbouring localities ( Fig 1 ) ; this suggests that there may be topotypes that remain to be identified and characterised . Genotyping was performed for 7 pools . The species composition of the pools and number of reads are shown in Table 2 . NGS reads were compared with available sequences in Genbank ( Genbank accession numbers: KT634318 , KF483675 , KR349298 , JQ769142 , KF137560 , KJ481126 ) by Blastn using the CLC Genomic Workbench 6 . 5 . The species were determined when the consensus sequences had ≥98% similarity with the regional reference sequences except Sergentomyia sp . sequences which had ≥85% similarity with S . dentata from Adana ( Genbank accession numbers: KU659595 , KU659596 , KU659597 , KU659598; release date 01 July 2016 ) . Therefore we indicated these sequences as Sergentomyia sp .
Although there are published serological data [13] and a recent report of detection of TOSV [19] in the Adana , Mediterranean region of Turkey there have been no previous reports of virus isolation . To further understand the presence of sandfly-borne phleboviruses that circulate in this endemic region for leishmaniasis [31] , close to the border of Syria where many refugee camps are settled , we organised field-study campaigns in 2012 and 2013 . The TORV and ZERV that were isolated during our study were most closely related to but distinct from SFS- and SFNV- like viruses , respectively . Studies conducted in 2012 led to the isolation of ADAV a novel putative member of the Salehabad species [20] . These results demonstrate that 3 phleboviruses belonging to 3 different genetic lineages co-circulate in the population of sandflies in this geographic area . Cumulative data resulting from this study and that of [20] enabled estimation of infection rates in sandflies ( 0 . 07% for sandfly-borne viruses in this region of Turkey ) which is in the same order of magnitude as previously reported in France , Tunisia , Spain and Italy [2 , 23 , 32 , 33 , 34] . Regardless of the gene used for analysis TORV and CFUV were grouped together in a sublineage that is clearly distinct from that including all other SFS-like viruses . CFUV was isolated from Phlebotomus major sensu lato [35] trapped in the eponymous Greek island . Interestingly , TORV has been isolated from two pools that contained P . perfiliewi sensu lato and P . tobbi , both belonging to the Larroussius subgenus as P . major sensu lato . Similarly , SFTV was detected in P . major sensu lato [21] . In contrast , other SFSV strains were isolated from P . papatasi that belongs to the Phebotomus subgenus which is distinct from the Larroussius subgenus [36] . Therefore , the two subgroups of viruses might reflect vector properties , with CFUV/ TORV associated with the Larroussius subgenus whereas other SFSV viruses are associated with P . papatasi with the exception of SFTV association with P . major sensu lato . Vector-virus association needs to be studied in a more detailed manner in order ( i ) to determine unambiguously the sandfly species transmitting these newly described viruses , ( ii ) and to verify our hypothesis that virus subgroups within viral species might be linked to specific vectors belonging to distinct taxonomic entities . Additional field studies combined with experimental studies using sandfly colonies need to be initiated to understand the parameters driving vector capacity and competence for different strains of viruses . ZERV was consistently grouped together with THEV and Naples virus strain YU-8-76 . Interestingly , THEV and the Serbian isolate Yu 8/76 apparently do not require expression of the NSs ORF , since their replication is not impaired by the presence of either an early stop codon or a large truncation [37] , whereas there is no such impairment or truncation in the ZERV genome which has a complete NSs ORF . Similar observations were also reported for a naturally attenuated RVFV strain ( clone 13 ) that has a large in-frame deletion in the NSs coding region [37 , 38] . Within the SFNV species , it is possible to discriminate 4 sublineages ( I to IV ) ; we propose to assign ZERV to sublineage I , where it was most closely related with THEV ( Figs 2 , 3 , 4 , 5 and 6 ) . THEV was isolated from P . papatasi sandflies in Iran in 1959 [39] , whereas YU-8-76 strain of SFNV was isolated from P . perfiliewi sensu lato trapped in Serbia in 1976 [37] . Subgroup III appears to be associated with P . papatasi , whereas subgroups II and IV appear to be associated with vectors belonging to the subgenus Larroussius . Subgroup I may be associated with Larroussius except THEV isolated in P . papatasi . Only P . tobbi was found to be present in all of the four ZERV positive pools . The same comment formulated above concerning the need for experimental studies to understand species-related competence and specificity of sandflies applies here . Genetic and phylogenetic analyses support the fact that both ZERV and TORV should be considered as new strains within pre-existing SFNV species and the yet to be recognised species including SFSV and CFUV , respectively . To date , SFSV and CFUV are listed as tentative species by the ICTV [1] . This study , based on complete genome sequences , suggests that all these viruses should be considered as members of the same species which could be further subdivided into CFUV / TORV , and SFSV / SFS-like viruses . A written proposal will be submitted to the Bunyaviridae Study Group of the ICTV . During this two-year study , 48% of the sandflies were trapped from Damyeri village where the ecological conditions were no different from those observed in other sampling stations . However , in Damyeri , the number of domestic animals ( sheep , goats , and cows ) was much higher than in other localities and these animals were constantly in the close vicinity of houses producing droppings which are known to be favoured breeding sites for sandflies or we set the traps very close to the possible breeding sites by chance therefore we got higher numbers of sand flies for this village . Thus , human exposure to sandflies might be greater in Damyeri than other sampling stations . Whether TORV and ZERV can infect humans and cause diseases such as sandfly fever is currently unknown and remains to be investigated . After the outbreak of sandfly fever occurred in Adana in 2008 , specific IgM against SFSV and/or SFCV was detected in acute cases by mosaic-immunofluorescence test , although the cause of the epidemic was not formally established through virus isolation or molecular detection with sequence confirmation [13] . The region where this outbreak occurred is located less than 25 km away from our trapping stations . Despite the fact that this study does not provide results supporting that both newly discovered viruses are human or animal pathogens , both TORV and ZERV belong to genetic groups that include several human pathogens . There is no doubt that SFSV "historic strains" were causing massive outbreak of debilitation and incapacitating disease [12 , 40 , 41] . Moreover , SFCV was also isolated from human cases [42 , 43] , as well as SFTV [13] . Antibodies against CFUV / SFSV were reported in humans living in mainland Greece and on Corfu Island using the immunofluorescence assay ( IFA ) [44] . Viral RNA of Chios virus , closely related to CFUV , was detected in the CSF of a patient presenting with severe encephalitis ( Papa and Pavlidou , 2003 , Genbank no , AY293623 ) . In contrast , specific antibodies were never described in humans for THEV [12] , although SFNV a close relative of THEV and subsequently ZERV was undoubtedly the cause of explosive outbreaks in newcomers to endemic areas during summertime [9] . Importantly , although serological studies have not yet been reported we do have serological data to support the concept that the newly isolated ZERV and TORV can infect vertebrates . However , whether or not these vertebrates do play a reservoir or amplifying role is not yet clear . Interestingly , although TORV and ZERV genomic RNA was detected in female pools of sandflies , both viruses were only isolated from male pools . Based on current knowledge it is not known how male sandflies become infected . Whilst , transovarial transmission seems a likely possibility it is not yet known how significant or efficient this mechanism of transmission is in natural habitats . However , laboratory experiments have shown that the rates of infection amongst offspring are low and show a decline from the first generation to ongoing generations . Other studies suggest that venereal ( horizontal ) transmission from infected males to uninfected females by mating and transstadial transmission of TOSV in diapausing Phlebotomus perniciosus larvae [9] may also contribute to long-term virus survival . From what is currently known and in the absence of defined vertebrate reservoirs , maintenance and transmission of sandfly-borne phleboviruses appears to depend on the abundance and accessibility of appropriate vector species . This lack of available knowledge of virus transmission and the virus maintenance mechanisms clearly need to be investigated both in natural habitats and under experimental conditions . Future studies are planned to examine female sandfly salivary glands and heads to look for the presence of infectious virus . Recent studies have detected sequences compatible with the existence of many putative new phleboviruses transmitted by sandflies in the Old World . In most of the cases , they were only partial sequences in the polymerase or the nucleoprotein genes . To date these limited genetic data , preclude classification by ICTV . This situation applies for viruses that may belong to ( i ) the SFNV species such as FERV [2] , Provincia virus [48] , Girne1 virus [18] , and Saddaguia virus [45] , ( ii ) the SALV species such as Adria virus [46 , 47] Olbia virus [48] and Edirne virus [18] , and to ( iii ) the SFSV / CFUV complex such as Chios virus ( Papa and Pavlidou , 2003 , Genbank no , AY293623 ) , Utique virus [4] , Girne2 virus [18] , SFS-like viruses [4 , 42 , 43 , 49] . Accordingly , although our knowledge of sandfly-borne phleboviruses is more extensive than it was a half-decade ago; efforts to isolate virus strains and determine their complete sequence should continue . Since virus taxonomy for the Phlebovirus genus still relies on neutralisation-based antigenic relationships , virus isolation is also essential . Nevertheless , the criteria for taxonomy appear to be evolving towards full-length genome comparative analysis . In conclusion , the results obtained in this study together with previously published results [20] demonstrate that ( i ) at least 3 different phleboviruses are co-circulating in phlebotomine sandfly populationsin the Adana region of Mediterranean Turkey; ( ii ) these new viruses belong to 3 distinct but closely related phylogenetic groups or species ( SFNV , SALV , SFSV / CFUV ) ; ( iii ) all the closely related viruses are known to be sandfly-borne arboviruses; ( v ) we have evidence of vertebrate seroprevalence for this group of viruses . Thus whilst it is indirect , the evidence for TORV and ZERV being arboviruses is compelling . It is also important to emphasise that although >10 , 000 sandflies were tested in this study; TOSV was neither detected nor isolated from the field samples . This challenges recent reports of TOSV-specific antibodies in blood donors from this region of Turkey [17] , and TOSV RNA / TOSV IgG in dogs from Mersin and Adana [19] . However , we recognise that the sampling points in these earlier studies do not overlap with those selected for our investigations .
|
We provide evidence that sandfly-borne phleboviruses belonging to 3 distinct genetic and phylogenetic groups ( Sandfly fever Naples virus [SFNV] , Sandfly fever Sicilian virus [SFSV] , and Salehabad virus [SALV] ) co-circulate in Adana city , in Mediterranean Turkey . While Adana virus was recently described as a new member of the SALV species , Zerdali and Toros viruses are described here as new phleboviruses genetically closely related to SFNV and SFSV , respectively . In this study , isolated and characterised these two new viruses by determining their complete genome sequence and by phylogenetic analysis . This study demonstrates that 3 distinct viruses can co-circulate in the same geographic area and based on their phylogenetic relationships and association with sandflies are likely to be transmitted by these arthropod vectors . Our molecular and phylogenetic data are important for establishing group-specific molecular detection assays in order to further understand of the possible impact of these viruses in animal and human health in this region of Turkey .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] |
[
"turkey",
"(country)",
"sequencing",
"techniques",
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"pathogens",
"population",
"genetics",
"geographical",
"locations",
"microbiology",
"sand",
"flies",
"gene",
"pool",
"viruses",
"signs",
"and",
"symptoms",
"phylogenetic",
"analysis",
"genome",
"analysis",
"molecular",
"biology",
"techniques",
"population",
"biology",
"insect",
"vectors",
"research",
"and",
"analysis",
"methods",
"europe",
"sequence",
"analysis",
"genomics",
"medical",
"microbiology",
"epidemiology",
"microbial",
"pathogens",
"gene",
"prediction",
"molecular",
"biology",
"disease",
"vectors",
"molecular",
"biology",
"assays",
"and",
"analysis",
"techniques",
"people",
"and",
"places",
"asia",
"fevers",
"viral",
"pathogens",
"genetics",
"biology",
"and",
"life",
"sciences",
"computational",
"biology",
"evolutionary",
"biology",
"organisms"
] |
2016
|
Sandfly-Borne Phlebovirus Isolations from Turkey: New Insight into the Sandfly fever Sicilian and Sandfly fever Naples Species
|
Agents living in volatile environments must be able to detect changes in contingencies while refraining to adapt to unexpected events that are caused by noise . In Reinforcement Learning ( RL ) frameworks , this requires learning rates that adapt to past reliability of the model . The observation that behavioural flexibility in animals tends to decrease following prolonged training in stable environment provides experimental evidence for such adaptive learning rates . However , in classical RL models , learning rate is either fixed or scheduled and can thus not adapt dynamically to environmental changes . Here , we propose a new Bayesian learning model , using variational inference , that achieves adaptive change detection by the use of Stabilized Forgetting , updating its current belief based on a mixture of fixed , initial priors and previous posterior beliefs . The weight given to these two sources is optimized alongside the other parameters , allowing the model to adapt dynamically to changes in environmental volatility and to unexpected observations . This approach is used to implement the “critic” of an actor-critic RL model , while the actor samples the resulting value distributions to choose which action to undertake . We show that our model can emulate different adaptation strategies to contingency changes , depending on its prior assumptions of environmental stability , and that model parameters can be fit to real data with high accuracy . The model also exhibits trade-offs between flexibility and computational costs that mirror those observed in real data . Overall , the proposed method provides a general framework to study learning flexibility and decision making in RL contexts .
The adaptation of learning to contingency changes and noise has numerous connections to various scientific fields from cognitive psychology to machine learning . A classical finding in behavioural neuroscience is that instrumental behaviours tend to be less and less flexible as subjects repeatedly receive positive reinforcement after selecting a certain action in a certain context , both in animals [5–8] and humans [9–13] . This suggests that biological agents indeed adapt their learning rate to inferred environmental stability: when the environment appears stable ( e . g . after prolonged experience of a rewarded stimulus-response association ) , they show increased tendency to maintain their model of the environment unchanged despite reception of unexpected data . Most studies on such automatization of behaviour have focused on action selection . However , weighting new evidence against previous belief is also a fundamental problem for perception and cognition [14–16] . Predictive coding [17–22] provides a rich , global , framework that has the potential to tackle this problem , but an explicit formulation of cognitive flexibility is still lacking . For example , whereas [22] provides an elegant Kalman-like Bayesian filter that learns the current state of the environment based on its past observations and predicts the effect of its actions , it assumes a stable environment and cannot , therefore , adapt dynamically to contingency changes . The Hierarchical Gaussian Filter ( HGF ) proposed by Mathys and colleagues [23 , 24] provides a mathematical framework that implements learning of a sensory input in a hierarchical manner , and that can account for the emergence of inflexibility in various situations . This model deals with the problem of flexibility ( framed as expected “volatility” ) by building a hierarchy of random variables: each of these variables is distributed with a Gaussian distribution with a mean equal to this variable at the trial before and the variance equal to a non-linear transform of the variable at superior level . Each level encodes the distribution of the volatility of the level below . Although it has shown its efficiency in numerous applications [25–30] , a major limitation of this model , within the context of our present concern , is that While the HGF accommodates a dynamically varying volatility , it assumes that the precision of the likelihood at the lowest level is static . To understand why it is the case , one should first observe that in the HGF the variance at each level is the product of two factors: a first “tonic” component , which is constant throughout the experiment , and a “phasic” component that is time-varying and controlled by the level above . These terms recall the concepts of “expected” and “unexpected” uncertainty [31 , 32] , and in the present paper , we will refer to these as variance ( of the observation ) and volatility ( of the contingency ) . Now consider an experiment with two distinct successive signals , one with a low variance and one with a high variance . When fitted to this dataset , the HGF will consider the lower variance as the first tonic component , and all the extra variance in the second part of the signal will be assigned to the “phasic” part of the volatility , thus wrongfully considering noise of the signal as a change of contingency ( see Fig 1 ) . In summary , the HGF will have difficulties accounting for changes in the variance of the observations . Moreover , the HGF model cannot forget past experience after changes of contingency , but can only adapt its learning to the current contingency . This contrasts with the approach we propose , where the assessment of a change of contingency is made with the use of a reference , naive prior that plays the role of a “null hypothesis” . This way of making the learning process gravitate around a naive prior allows the model to actively forget past events and to eventually come back to a stable learning state even after very surprising events . These caveats limit the applicability of the HGF to a certain class of datasets in which contingency changes affect the mean rather than the variance of observations and in which the training set contains all possible future changes that the model may encounter at testing . As will be shown in detail below , in the model proposed in the present paper , volatility is not only a function of the variance of the observations: if a new observation falls close enough to previous estimates then the agent will refine its posterior estimate of the variance and will decrease its forgetting factor ( i . e . will move its prior away from the fixed initial prior and closer to the learned posterior from the previous trial ) , but if the new observation is not likely given this posterior estimate , the forgetting factor will increase ( i . e . will move closer to the fixed initial prior ) and the model will tend to update to a novel state ( because of the low precision of the initial prior ) . In the results of this manuscript , we show that our model outperforms the HGF in such situations . In Machine Learning and in Statistics , too , the question of whether new unexpected data should be classified as outlier or environmental change is important [33] . This problem of “denoising” or “filtering” the data is ubiquitous in science , and usually relies on arbitrary assumptions about environmental stability . In signal processing and system identification , adaptive forgetting is a broad field where optimality is highly context ( and prior ) -dependant [2] . Bayesian Filtering ( BF ) [34] , and in particular the Kalman Filter [35] often lack the necessary flexibility to model real-life signals that are , by nature , changing . One can discriminate two approaches to deal with this problem: whereas Particle Filtering ( PF ) [36–38] is computationally expensive , the SF family of algorithms [2 , 39] , from which our model is a special case , usually has greater accuracy for a given amount of resources [36] ( for more information , we refer to [35] where SF is reviewed ) . Most previous approaches in SF have used a truncated exponential prior [40 , 41] or a fixed , linear mixture prior to account for the stability of the process [37] . Our approach is innovative in this field in two ways: first , we use a Beta prior on the mixing coefficient ( unusual but not unique [42] ) , and we adapt the posterior of this forgetting factor on the basis of past observations , the prior of this parameter and its own adaptive forgetting factor . Second , we introduce a hierarchy of forgetting that stabilizes the learning when the training length is long . We therefore intend to focus our present research on the very question of flexibility . We will show how flexibility can be implemented in a Bayesian framework using an adaptive forgetting factor , and what prediction this framework makes when applied to learning and decision making in Model-Free paradigms .
Classical RL [43] , or Bayesian RL [44 , 45] cannot discriminate learners that are more prone to believe in a contingency change from those who tend to disregard unexpected events and consider them as noise . To show this , we take the following example: let p ( ρ|r≤j ) = Beta ( α , β ) be the posterior probability at trial j of a binary reward rj ∼ Bern ( ρ ) with prior probability ρ ∼ Beta ( α0 , β0 ) . It can be shown that , at the trial j = vj + uj , where vj is the number of successes and uj the number of failures , the posterior probability parameters read {αj = α0 + vj , βj = β0 + uj} . This can be easily mapped to a classical RL algorithm if one considers that , at each update of v and u , the posterior expectation of ρ is updated by E [ ρ | r ≤ j ] = v j - 1 + r j v j - 1 + r j + uj - 1 + ( 1 - r j ) ( 1 ) = j - 1 j E [ ρ j - 1 ] + r j j ( 2 ) = E [ ρ j - 1 ] + η ( r j - E [ ρ j - 1 ] ) ( 3 ) where η ≜ 1 j ( 4 ) which has the form of a classical myopic Q-learning algorithm with a decreasing learning rate . The drawback of this fixed-schedule learning rate is that , if the number of observed successes outnumbers greatly the number of failures ( v ≫ u ) at the time of a contingency change in which failures become suddenly more frequent , the agent will need v − u + 1 failures to start considering that p ( rj = 0|r≤j ) > p ( rj = 1|r≤j ) . This behaviour is obviously sub-optimal in a changing environment , and Dearden [44] suggests adding a constant forgetting factor to the updates of the posterior , making therefore the agent progressively blind to past outcomes . Consider the case in which E [ ρ | r ≤ j ] = w v j - 1 + r j w v j - 1 + r j + w u j - 1 + ( 1 - r j ) with w ∈ [0; 1] being the forgetting factor . We can easily see that , in the limit case of α0 = 0 and β0 = 0 , α j + β j → 1 1 - w as j → ∞ . We can define 1 1 - w as the efficient memory of the agent , which provides a bound on the effective memory , represented by the total amount of trials taken into account so far ( e . g . αj+ βj in the previous example ) . This produces an upper and lower bound to the variance of the posterior estimate of p ( ρ|r≤j ) . This can be seen from the variance of the beta distribution V a r [ ρ | r ≤ j ] = ( α j ) ( β j ) ( α j + β j ) 2 ( α j + β j + 1 ) which is maximized when αj = βj , and minimized when either αj = α0 or βj = β0 . In a steady environment , agents with larger memory are advantaged since they can better estimate the variance of the observations . But when the environment changes , large memory becomes disadvantageous because it requires longer time to adapt to the new contingency . Here , we propose a natural solution to this problem by having the agent erase its memory when a new observation ( or a series of observations ) is unlikely given the past experience . Our framework is based on the following assumptions: Assumption 1 The environment is fully Markovian: the probability of the current observation given all the past history is equal to the probability of this observation given the previous observation . Assumption 2 At a given time point , all the observations ( rewards , state transitions , etc . ) are i . i . d . and follow a distribution p ( x|z ) that is issued from the exponential family and has a conjugate prior that also belongs to the exponential family p ( z|θ0 ) . For conciseness , the latent variables z ( i . e . action value , transition probability etc . ) and their prior θ will represent the natural parameters of the corresponding distributions in what follows . Assumption 3 The agent builds a hierarchical model of the environment , where each of the distributions at the lower level ( reward and state transitions ) are independent , i . e . the reward distribution for one state-action cannot be predicted from the distribution of the other state-actions . Assumption 4 The agent can only observe the effects of the action she performs . Finally , an important assumption that will guide the development of the model is that the evolution of the environment is unpredictable ( i . e . transition probabilities are uniformly distributed for all states of the environment ) with the notable exception that it is more or less likely to stay in the same state than to switch to another state . Formally: Assumption 5 Let { z a } a = 1 A be a set of environment states , with A ≫ 0 and a , b , c ∈ {1 , 2 , … , A} , a ∉ {b , c} . We assume that the transition probabilities are uniformly distributed for b , c ∈ {1: A}¬a , which reads: p ( z j = z b | z j - 1 = z a ) = p ( z j = z c | z j - 1 = z a ) ≠ p ( z j = z a | z j - 1 = z a ) . Assumption 5 implies that any attempt to learn new transitions from state to state based on a uniform prior over these transitions will harm the performance of the predictive model , and the best strategy one could adopt is to learn the probability of staying in the same state and group the probabilities of changing to any other state together . Then , the only two transition probabilities to learn are: p ( z j = z b | z j - 1 = z a ) for b ≠ a p ( z j = z a | z j - 1 = z a ) . This is what the “critic” part of the AC-HAFVF we propose achieves . Of course , the model could be improved by learning the other transition probabilities , if needed , but we leave this for future work ( see for instance [35] ) . Let us now analyze the expected behaviour of an agent using a model similar to the one just described , in a steady environment: if all x = { x j } j = 1 J belong to the same , unknown distribution p ( x|z ) , the value of E p j ( z ∣ x ≤ j ) [ z ] will progressively converge to its true value as the prior ( or the previous posterior ) over w will eventually put a lot of weight on the past experience ( i . e . it favours high values of w ) , since the distribution from which x is drawn is stationary . We have shown that such models rapidly tend to an overconfident posterior over w [1] . In practice , when the previous posterior of w is confident on the value that w should take ( i . e . has low variance ) , it tends to favor updates that reduce variance further , corresponding to values of pj ( w|x≤j ) that match pj−1 ( w|x<j ) , even if this means ignoring an observed mismatch between pj−1 ( z|x<j ) and pj ( z|x≤j ) . In order to deal with this issue , we enrich our model by introducing a third level in the hierarchy . We re-define the prior over w as a two-component mixture of priors: p j ( w | x < j ; b , ϕ 0 ) ≜ p ( w | x < j ) b p ( w | ϕ 0 ) 1 - b Z ( b , x < j , α 0 ) and the full joint probability has the form p ( x j , z , w , b | x < j ; θ 0 , ϕ 0 , β 0 ) = p ( x j | z ) p ( z | x < j ) w p ( z | θ 0 ) 1 - w Z ( w , x < j , θ 0 ) p ( w | x < j ) b p ( w | ϕ 0 ) 1 - b Z ( b , x < j , ϕ 0 ) p ( b | x < j , β 0 ) . ( 10 ) This additional hierarchical level allows the model to forget w as a function of observed data ( i . e . not at a fixed rate ) providing it with the capacity to adapt the approximate posterior distribution over w with greater flexibility [1] . The latent variable b can be seen as a regulizer for pj ( w ∣ x≤j ) . The prior parameters of the HAFVF and their interpretation is outlined in Table 1 . Eqs 6–10 involve the posterior probability distributions of the parameters given the previous observations . When these quantities have no closed-form formula , two classes of methods can be used to estimate them . Simulation-based algorithms [46] such as importance sampling , particle filtering or Markov Chain Monte Carlo , are asymptotically exact but computationally expensive , especially in the present case where the estimate has to be refined at each time step . The other class of methods , approximate inference [47 , 48] , consists in formulating , for a model with parameters y and data x , an approximate posterior q ( y ) , that we will use as a proxy to the true posterior p ( y|x ) . Roughly , approximate inference can be partitioned into Expectation Propagation and Variational Bayes ( VB ) methods . Let us consider in more detail VB , as it is the core engine of our learning model . In VB , optimizing the approximate posterior amounts to computing a lower-bound to the log model evidence ( ELBO ) L ( q ( y ) ) ≤ logp ( x | x < j ) , whose distance from the true log model evidence can be reduced by gradient descent [49] . Hybrid methods , that combine sampling methods with approximate inference , also exist ( e . g . Stochastic Gradient Variational Bayes [50] or Markov Chain Variational Inference [51] ) . With the use of refined approximate posterior distributions [52–54] , they allow for highly accurate estimates of the true posterior with possibly complex , non-conjugate models . We define a variational distribution over y with parameters υ: q ( y|υ ) , which we will use as a proxy to the real , but unknown , posterior distribution p ( y|x ) . The two distribution match exactly when their Kullback-Leibler divergences are equal to zero , i . e . DKL[q ( y ) ∥p ( y|x ) ]=0⇔DKL[p ( y|x ) ∥q ( y ) ]=0⇔p ( y|x ) =q ( y ) where we have omitted the approximate posterior parameters υ for sparsity of the expressions . Given some arbitrary constraints on q ( y ) , we can choose ( for mathematical convenience ) to reduce DKL[q ( y ) ||p ( y|x ) ] wrt q ( y ) . Formally , this can be written as q * ( y ) = arg min q ( y ) D K L [ q ( y ) ∥ p ( y | x ) ] = arg min q ( y ) ∫ q ( y ) logq ( y ) p ( y | x ) d y = arg min q ( y ) ∫ q ( y ) logq ( y ) d y - ∫ q ( y ) logp ( y | x ) d y . We can now substitute log p ( y|x ) by its rhs in the log-Bayes formula D K L [ q ( y ) ∥ p ( y | x ) ] = ∫ q ( y ) logq ( y ) d y - ∫ q ( y ) ( logp ( x , y ) - logp ( x ) ) d y ⇔ logp ( x ) = ∫ q ( y ) logp ( x , y ) d y - ∫ q ( y ) logq ( y ) d y ︸ L ( q ( y ) ) + D K L [ q ( y ) ∥ p ( y | x ) ] . ( 11 ) Because log p ( x ) does not depend on the model parameters , it is fixed for a given dataset . Therefore , as we maximize L ( q ( y ) ) in Eq 11 , we decrease the divergence DKL[q ( y ) ||p ( y|x ) ] between the approximate and the true posterior . When a maximum is reached , we can consider that ( 1 ) we have obtained the most accurate approximate posterior given our initial assumptions about q ( y ) and ( 2 ) L ( q ( y ) ) provides a lower bound to log p ( x ) . It should be noted here that the more q ( y ) is flexible , the closer we can hope to get from the true posterior , but this is generally at the expense of tractability and/or computational resources . The ELBO in Eq 11 is the sum of the expected log joint probability and the entropy of the approximate posterior . In order for the former to be tractable , one must carefully choose the form of the approximate posterior . The Mean-field assumption we have made allows us to select , for each factor of the approximate posteriors , a distribution with the same form as their conjugate prior , which is the best possible configuration in this context [55] . Applying now this approach to Eq 10 , our spherical approximate posterior looks like: q ( z , w , b ) ≜ q ( z | θ ) q ( w | ϕ ) q ( b | β ) . In addition , in order to recursively estimate the current posterior probability of the model parameters given the past , we make the natural approximation that the true previous posterior can be substituted by its variational approximation: p ( z | x < j ) ≈ q j - 1 ( z ) ( 12 ) and similarly for p ( w|x<j ) and p ( b|x<j ) . The use of this distribution as a proxy to the posterior greatly simplifies the optimization of qj ( z , w , b ) . The full , approximate joint probability distribution at time j therefore looks like p ( x j , z , w , b | x < j , θ 0 , ϕ 0 ) ≈ p ( x j | z ) q j - 1 ( z | θ j - 1 ) w p ( z | θ 0 ) 1 - w Z ( w , θ j - 1 , θ 0 ) q j - 1 ( w | ϕ j - 1 ) b p ( w | ϕ 0 ) 1 - b Z ( b , ϕ j - 1 , ϕ 0 ) q ( b | β j - 1 ) where θj−1 , ϕj−1 and βj−1 are the variational parameters at the last trial for z , w and b respectively . A further advantage of the approximation made in Eq 12 is that the prior of z and w simplifies elegantly: p ( x j , z , w , b | x < j ) ≈ p ( x j | z ) p ( z | w ( θ j - 1 - θ 0 ) + θ 0 ) × p ( w | b ( ϕ j - 1 - ϕ 0 ) + ϕ 0 ) p ( b | β j - 1 ) ( 13 ) ( see Appendix A for the full derivation ) . A conjugate distribution for p ( w ) is hard to find . Šmìdl and Quinn [56] propose a uniform prior and a truncated exponential approximate posterior over w . They interpolate the normalizing constant between two fixed value of w , which allows them to perform closed-form updates of this parameter . Here , we chose p ( w|ϕ0 ) and q ( w|ϕj ) to be both beta distributions , a choice that does not impair our ability to perform closed-form updates of the variational parameters as we will see in the Update equation section . In this model ( see Fig 2 ) , named the Hierarchical Adaptive Forgetting Variational Filter [1] , specific prior configurations will bend the learning process to categorize surprising events either as contingency changes , or as accidents . In contrast with other models [57] , w and b are represented with a rich probability distribution where both the expected values and variances have an impact on the model’s behaviour . For a given prior belief on z , a confident prior over w , centered on high values of this parameter , will lead to a lack of flexibility that would not be observed with a less confident prior , even if they have the same expectation . Application of this scheme of learning to the RL case is straightforward , if one considers x = { r j ( s j , a j ) j } j = 1 J as being the observed rewards and z as the parameters of the distribution of these rewards . In the following , we will assume that the agent models a normally distributed state-action reward function xj = r ( s , a ) , from which she tries to estimate the posterior distribution natural parameters z ≜ η ( μ ( s , a ) , σ ( s , a ) ) where η ( ⋅ ) is the natural parameter vector of the normal distribution . In this context , an intuitive choice for the prior ( and the approximate posterior ) of these parameters is a Normal Inverse-Gamma distribution ( N G - 1 ) : for the prior , we have μ ( s , a ) ∼ N ( μ 0 μ ( s , a ) , σ 2 ( s , a ) κ 0 μ ( s , a ) ) σ 2 ( s , a ) ∼ G - 1 ( α 0 σ ( s , a ) , β 0 σ ( s , a ) ) and the approximate posterior can be defined similarly with a normal component N ( μ j μ ( s , a ) , σ 2 ( s , a ) κ j μ ( s , a ) ) and a gamma component G - 1 ( α j σ ( s , a ) , β j σ ( s , a ) ) . So far , we have provided all the necessary tools to simulate behavioural data using the AC-HAFVF . It is now necessary to show how to fit model parameters to an acquired dataset . We will first describe how this can be done in a Maximum Likelihood framework , before generalizing this method to Bayesian inference using variational methods . The problem of fitting the AC-HAFVF to a dataset can be seen as a State-Space model fitting problem . We consider the following family of models: {Ω j , n = f ( Ω j - 1 , n , Ω 0 , n , x j - 1 , n ) ︸ H A F V F y j , n ∼ E p ( μ 1 , μ 2 , σ 1 2 , σ 2 2 | Ω j , n ) [ W i e n e r ( μ 1 - μ 2 , σ 1 2 + σ 2 2 , ζ n , z 0 n , τ n ) ] ︸ N I G D M where Ω j , n = { μ j , n μ , κ j , n μ , α j , n σ , β j , n σ , ϕ j , n , β j , n} j , n = 1 J , N and y j , n = { t j , n , a j , n } j , n = 1 J , N ( 21 ) and tj , n stands for the reaction time associated with the state-action pair ( s , a ) of the subject n at the trial j . Unlike many State-Space models , we have made the assumption in Eq 21 that the transition model Ωj , n = f ( Ωj−1 , n , Ω0 , n , xj−1 , n ) is entirely deterministic given the subject prior Ω0 , n and the observations x<j , which is in accordance with the model of decision making presented in the The actor: Decision making under the HAFVF section . Note that the Bayesian procedure we will adopt hereafter is formally identical to considering that the drift and noise are drawn according to the rules defined in the The actor: Decision making under the HAFVF section , making this model equivalent to: y j , n ∼ W i e n e r ( ξ ∼ , ς ∼ , ζ n , z 0 n , τ n ) ξ ∼ , ς ∼ ∼ f ( Ω j - 1 , n , Ω 0 , n , x j - 1 , n ) . Quite importantly , we have made the assumption in Eq 21 that the threshold , the non-decision time and the start-point were fixed for each subject throughout the experiment . This is a strong assumption , that might be relaxed in practice . To simplify the analysis , and because it is not a mandatory feature of the model exposed above , we do not consider this possibility here and leave it for further developments . We can now treat the problem of fitting the AC-HAFVF to behavioural data as two separate sub-problems: first , we will need to derive a differentiable function that , given an initial prior Ω0 , n and a set of observations xn produces a sequence Ω n = {Ω j , n} j = 1 J , and second ( Maximum a Posteriori estimate of the AC-HAFVF section ) a function that computes the probability of the observed behaviour given the current variational parameters .
In order to compare the performance of our model to the HGF , we generated a simple dataset consisting of a noisy square-wave signal of two periods of 200 trials , alternating between two normally distributed random variables ( N ( 1 , 0 . 33 ) and N ( - 1 , 0 . 33 ) ) . We fitted both a Gaussian HGF and the HAFVF to this simple dataset by finding the MAP parameter configuration for both models . The default configuration of the HGF was used , whereas in our case we put a normal hyperprior of N ( 0 , 1 ) on the parameters ( with inverse softplus transform for parameters needing positive domains ) . This fit constituted the first part of our experiment , which is displayed on the left part of Fig 4 . We compared the quadratic approximations of the Maximum Log-model evidences [88] for both models , which for the HAFVF reads logp ( θ 0 * , ϕ 0 * , β 0 * ∣ x ) = ∑ j = 1 J logp ( x j | z ) + w logq j - 1 ( z | θ j - 1 ) + ( 1 - w ) logp ( z | θ 0 ) - logZ ( w , θ j - 1 , θ 0 ) + b logq j - 1 ( w | ϕ j - 1 ) + ( 1 - b ) logp ( w | ϕ 0 ) - logZ ( b , ϕ j - 1 , ϕ 0 ) + logq ( b | β j - 1 ) + logp ( θ 0 , ϕ 0 , β 0 ) + M 2 log2 π + 1 2 log| - H | - 1 where H is the hessian of the log-joint at the mode and M is the number of parameters of the model . We found a value of -186 . 42 for HAVFV and -204 . 73 for the HGF , making the HAVFV a better model of the data , with a Bayes Factor [89] greater than 8 * 107 . The second part of the experiment consisted in adding to this 400-trial signal a 1200-trial signal of input situated at y = 5 . We evaluated for both models the quality of the fit obtained when using the parameter configurations resulting from the fit of the first part of the experiment ( first 400 trials , or training dataset ) ( Fig 4 , right part ) , on the remaining dataset ( following 1200 trials , i . e . testing dataset ) . An optimal agent in such a situation should first account for the surprise associated with the sudden contingency change , and then progressively reduce its expected variance estimate to reflect the steadiness of the environment . We considered the capacity of both models to account for new data for a given parameter configuration as a measure of their flexibility . This test was motivated by the observation that a change detection algorithm has to be able to detect changes at test time that might be qualitatively different from changes at training time . A financial crisis is for instance an event that is in essence singular and unseen in the past ( otherwise it would have been prevented ) . The algorithm should nevertheless be able to detect it efficiently . The HGF was unable to exhibit the expected behaviour: it hardly adapted its estimated variance to the contingency change and did not adjust it significantly afterwards . This contrasted with the HAFVF , in which we observed initially an increase in the variance estimate at the point of contingency change ( reflecting a high surprise ) , followed by progressively decreasing variance estimate , reflecting the adaptation of the model to the newly stable environment . Together , these results are informative of the comparative performance of the two algorithms . The Maximum Log-model Evidence was larger for the HAFVF than for the HGF by several orders of magnitude , showing that our approach modelled better the data at hand than the HGF . Moreover , the lack of generalization of the HGF to a simple , new signal not used to fit the parameters , shows that this model tended to overfit the data , as can be seen from the estimated variance at the time of the contingency change . Importantly , this capability of the HAFVF to account for unseen volatility changes did not need to be instructed through the selection of the model hyperparameters: it is a built in feature of the model . In the following datasets , we simulated the learning process of four hypothetical subjects differing in their prior distribution parameters ϕ0 , β0 , whereas we kept θ0 fixed for all of them ( Table 2 ) . The choice of the subject parameters was made to generate limit and opposite cases of each expected behaviour . With these simulations , we aimed at showing how the prior belief on the two levels of forgetting conditioned the adaptation of the subject in case of contingency change ( CC , Experiment 1 ) or isolated expected event ( Experiment 2 ) . In both experiments , these agents were confronted with a stream of univariate random variables from which they had to learn the trial-wise posterior distribution of the mean and standard deviation . In Experiment 1 , we simulated the learning of these agents in a steady environment followed by an abrupt CC , occurring either after a long ( 900 trials ) or a short ( 100 trials ) training . The signal r = {r1 , r2 , … , rn} was generated according to a Gaussian noise with mean μ = 3 before the CC and μ = −3 after the CC , and a constant standard deviation σ = 1 . Fig 5 summarizes the results of this first simulation . During the training phase , the subjects with a long memory on the first level learned the observation value faster than others . Conversely , the SS subject took a long time to learn the current distribution . More interesting is the behaviour of the four subjects after the CC . In order to see which strategy reflected best the data at hand , we computed the average of the ELBOs for each model . The winning agent was the Long-Short memory , irrespective of training duration , because it was better able to adapt its memory to the contingency . The two levels had a different impact on the flexibility of the subjects: the first level indicated how much a subject should trust his past experience when confronted with a new event , and the second level measured the stability of the first level . On the one hand , subjects with a low prior on first-level memory were too cautious about the stability of the environment ( i . e . expected volatile environments ) and failed to learn adequately the contingency at hand . On the other hand , after a long training , subjects with a high prior on second-level memory tended to over-trust environment stability , compared to subjects with a low prior on second level memory , impairing their adaptation after the CC . The expected forgetting factors also shed light on the underlying learning process occurring in the four subjects: w ^ grew or was steady until the CC for the four subjects , even for the SL subject which showed a rapid growth of w ^ during the first trials , but failed to reduce it at the CC . In contrast , the LS subject did not exhibit this weakness , but rapidly reduced its expectation over the stability of the environment after the CC thanks to her pessimistic prior belief over b . In Experiment 2 , we simulated the effect of an isolated , unexpected event ( rj = −3 ) after long and short training with the same distribution as before . For both datasets , we focused our analysis on the value of the expected forgetting factors w ^ and b ^ , as well as the effective memory of the agents , represented by the parameter κμ . As noted earlier , the value of w ^ sets an upper bound ( the efficient memory ) on κμ , which represented the actual number of trials kept in memory up to the current trial . Fig 6 illustrates the results of this experiment . Here , the flexible agents ( with a low memory on either the first or second memory level , or both ) were disadvantaged wrt the low flexibility agent ( mostly LL ) . Indeed , following the occurrence of a highly unexpected observation , one can observe that the LS learner memory dropped after either long or short training . The LL learner , instead , was able to cushion the effect of this outlier , especially after a long training , making it the best learner of the four to learn these datasets ( Table 3 ) . The model we propose has a large number of parameters , and overfitting could be an issue . To show that inference about the latent variables of the model depicted in the Fitting the AC-HAFVF section is possible , we simulated a dataset of 64 subjects performing a simple one-stage behavioural task , that consisted in trying to choose at each trial the action leading to the maximum reward . In any given trial j = { j } j = 1 1000 , the two possible actions ( e . g . left or right button press ) were associated to different , normally distributed , reward probabilities with a varying mean and a fixed standard deviation σ 1 , 2 2 = 1 : for the first action ( a1 ) the reward had a mean of 0 for the first 500 trials , then switched abruptly to + 2 for 100 trials and then to −2 for the rest of the experiment . The second action value was identically distributed but in the opposite order and with the opposite sign ( Fig 7 ) . This pattern was chosen in order to test the flexibility of each agent after an abrupt CC: after the first CC , an agent discarding completely exploration in favour of exploitation would miss the CC . The second and third CC tested how fast did the simulated agents adapt to the observed CC . Individual prior parameters Ω 0 , n ≜ { N G - 1 prior θ 0 , n , Beta priors ϕ 0 , n , β 0 , n } and thresholds ζn were generated as follows: we first looked for the L2-regularized MAP estimates of these parameters that led to the maximum total reward: Ω 0 * , ζ * = arg max Ω 0 , ζ ∏ j = 1 J p ( a j = arg max a jr j | r < j , a < j , Ω , ζ ) p ( Ω 0 , ζ ) using a Stochastic Gradient Variational Bayes ( SGVB ) [90] optimization scheme . With a negligible loss of generality , the prior mean μ0 was considered to be equal to 0 for all subjects for both the data generation and the fitting procedures . With a negligible loss of generality , the prior mean μ0 was considered to be equal to 0 for all subjects for both the data generation and the fitting procedures . We then simulated individual priors centered around this value with a covariance matrix arbitrarily sampled as Σ Ω 0 , ζ ∼ W 10 - 1 ( [ 1 0 ⋯ 0 0 1 ⋮ ⋮ ⋱ 0 0 ⋯ 0 0 . 1 ] ) where W n - 1 ( · ) is an inverse-Wishart distribution with n degrees of freedom . This choice of prior lead to a large variability in the values of the AC-HAFVF parameters , except for the NIGDM threshold whose variance was set to a sufficiently low value ( hence the 0 . 1 value in the prior scale matrix ) to keep the learning performance high . This method ensured that each and every parameter set was centered around an unknown optimal policy . This approach was motivated by the need to prevent strong constrains on the data generation pattern while keeping behaviour close to optimal , as might be expected from healthy population . The other DDM parameters , νn and τn , were generated according to a Gaussian distribution centered on 0 and 0 . 3 , respectively . Simulated subjects with a performance lower than 70% were rejected and re-sampled to avoid irrelevant parameter patterns . Learning was simulated according to the Continuous Learning strategy ( see Counterfactual learning ) , because it was supposed to link more comprehensively the tendency to explore the environment with the choice of prior parameters Ω0 . Choices and RT where then generated according to the decision process described in the The actor: Decision making under the HAFVF section using the algorithm described by [91] . Fig 7 shows two examples of the simulated behavioural data . The behavioural results showed a clear tendency of subjects with large expected variance in action selection to act faster and less precisely than others . This follows directly from the structure of the NIGDM: larger variance of the drift-rate leads to faster but less precise policies . More interesting is the negative correlation between the expected stability and the reward-rate and average reaction time: this shows that the AC-HAFVF was able to encode a form of subject-wise computational complexity of the task . Indeed , large stability expectation leads subjects to trust more their past experience , thereby decreasing the expected reward variance after a long training , but it also leads to a lower capacity to adapt to CCs . For subjects with low expectation of stability , the second level memory was able to instruct the first-level to trust past experience when needed , as the positive correlation between accuracy and upper level memory shows . To study the TD learning described in the Temporal difference learning section , we built two similar Markov Decision Processes ( MDP ) which are described in Fig 11 . In brief , they consist of 5 different states where the agent has to choose the best action in order to reach a reward ( r = 5 ) delivered once a specific state has been reached ( hereafter , we will use “rewarded state” and “rewarded state-action” interchangeably ) . The task consisted for the agent to learn the current best policy during a 1000-trial experiment where the contingency was fixed to the first MDP during the first 500 trials , and then switched abruptly to the second MDP during the second half of the experiment . The same prior was used for all subjects: the mean and variance prior was set to μ0 = 0 , κ0 = 0 . 5 , α0 = 3 , β0 = 0 . 5 . The forgetting factors shared the same flat prior α 0 w , b = 1 , β 0 w , b = 1 . The priors on the discounting factor were set to a high value α 0 γ = 9 , β 0 w , b = 1 in order to discourage myopic strategies . The policy prior ( see Appendix D ) was set to a high value ( π0 = 5 . ) in order to limit the impact of initial choices on the computation of the state-action value . Results are displayed in Fig 12 . In both experiments , agents had a similar behaviour during the first phase of the experiment: they both learned well the first contingency by assigning an accurate value to each state-action pair in order to reach the rewarded state more often . As expected , after the contingency change , the agents in experiment ( i ) took a longer time to adapt than they took to learn the initial contingency , which can be seen from the steeper slope of the reward rate during the first half of the experiment wrt the second . This feature can only be observed if the agent weights its belief by some measure of certainty , which is not modelled in classical , non-Bayesian RL . In experiment ( ii ) , the CC changes less the environment structure than the CC in experiment ( i ) . The agents were able to take this difference into account: the value of the effective memories dropped less , and so did the reward rate . An important feature observed in these two experiments is that the expected value of γ adapted efficiently to the contingency: although we used a prior skewed towards high values of γ , its value tended to be initially low as the agents had no knowledge of the various action values . Also , this value increased afterwards , reflecting a gain in predictive accuracy . The drop of E q [ w ] had the effect of pushing E q [ γ ] towards its prior , which in this case increased the posterior expectation of γ . At the same time , the uncertainty about γ increased , thereby enhancing the flexibility of this parameter .
In this paper , we propose a new Bayesian Reinforcement Learning ( RL ) algorithm aimed at accounting for the adaptive flexibility of learning observed in animal and human subjects . This algorithm adapts continuously its learning rate to inferred environmental variability , and this adaptive learning rate is optimal under some assumptions about statistical properties of the environment . These assumptions take the form of prior distributions on the parameters of the latent and mixing weight variables . We illustrate different types of behaviour of the model when facing unexpected contingency changes by taking extreme case scenarios . These scenarios implemented four types of assumptions on the tendency of the environment to vary over time ( first-level memory ) and on the propensity of this environmental variability to change itself over time ( second-level memory ) . This approach allowed us to reproduce the emergence of inflexible behaviour following prolonged experience of a stable environment , similar to empirical observations in animals . Indeed , it has long been known that extensive training leads to automatization of behaviour , called habits [5 , 8 , 9 , 93–96] , or procedural “system 1” actor ( [95 , 97] ) , which is characterized by a lack of flexibility ( i . e . failure to adapt to contingency changes ) and by reduction of computational costs , illustrated by the capacity to perform these behaviours concomitantly to other tasks [98–100] . These automatic types of behaviour are opposed to Goal-Directed behaviours ( [99–101] ) and share the common feature of being inflexible , either in terms of planning ( for Model-Free RL for instance ) or in terms of adaptation in general . Regarding the actor part , we implemented a general , Bayesian decision-making algorithm that reflects in many ways the Full-DDM proposed by [3] , as it samples the reward distribution associated to each action and selects at each time step the best option . These elementary decisions are integrated until a decision boundary is reached . Therefore , the actor maps the cognitive predictions of the critic onto specific behavioural outputs such as choice and reaction time ( RT ) . Importantly , this kind of Bayesian evidence accumulation process for decision making is biologically plausible [102 , 103] , well suited for decision making in RL [104–107] and makes predictions that are in accordance with physiological models of learning and decision making [108] . Other noteworthy attempts have been made to integrate sampling-based decision-making and RL [109 , 110] using the DDM . However , the present work is the first , to our knowledge , to frame the DDM as an optimal , Bayesian decision strategy to maximize long-term utility on the basis of value distributions inferred from a RL algorithm . We show that this RL-DDM association , and especially in the framework of the Full DDM [3] , finds a grounded algorithmic justification in a Bayesian perspective , as the resulting policy mimics the one of an agent trying to infer the best decision given its posterior belief about the reward distribution . Interestingly , under some slight modifications ( i . e . assuming that the sampled rewards are not simulated but retrieved from the subject memory ) , it is similar to the model recently proposed by Bornstein et al . [111 , 112] . More specifically , while our decision making scheme used a heuristic based on the asymptotic property of MCMC , the scheme of decision making proposed by Bornstein and colleagues might recall other approximation techniques such as Approximate Bayesian Computation ( ABC [113] ) . Following this approach , data samples are generated according to some defined rule , and only the samples that match the actual previous observations are kept in memory to approximate the posterior distribution or , in our case , to evaluate the option with the greatest reward . This simple trick in the decision making process keeps the stochastic nature of the accumulation process , while directly linking the level of evidence to items retrieved from memory . The HAFVF also recalls the learning model proposed by Behrens and colleagues [32] who studied the variations of human learning rate in volatile environments and showed that activity in Anterior Cingulate Cortex reflected their model estimate of environmental volatility . The AC-HAFVF exhibits several differences with respect to this model: first , and crucially , it uses a Stabilized Forgetting framework to modify the belief that the agent has in the parameter values at each level , unlike Behrens et al . who used a purely forward model , similar in this sense to the HGF . Second , our model used Mean-field VB to make inference about the parameter values , allowing it to approximate the posterior distribution at low cost . The drawback of this approach , however , is that the AC-HAFVF presented here does not allow us to compute posterior covariance of their parameters at each trial , in contrast to Behrens and colleagues . Given these differences , it would interesting to compare how these various models of adaptation to volatility fit actual behavioural data and how well their parameters follow recorded neurobiological signals . An important feature of the HAFVF is that it can account for unstructured changes of contingency . In other words , it allows the agent to learn anew the state of the environment even if the transition that leads to this state has never been experienced before . This is an important feature that contrasts with Kalman filters and Hidden Markov Models [1] . Both approaches have their pros and cons: learning state transition probabilities makes sense in environment that enjoy specific regularity conditions , but if the environment is chaotic , they can lead to poor adaptation performance . The approach we adopted here makes sense in situations in which one expects that the environment may change in an unstructured way , e . g . in which an environment that has remained stable for a long period of time may ( suddenly or progressively ) change in a random and hence unpredictable manner . An intermediate approach could however be developed [35] . One major advantage of our model is that its parameters are easily interpretable as reflecting hidden behavioural features such as trial-wise effective memory , prior and posterior expected stability , etc . We detail how model parameters can be fitted to data in order to recover these behavioural features at the trial , subject and population levels . This approach could be used , for instance , to cluster subjects in high-stability seeking and low-stability seeking sub-populations , and correlate these behaviours to health conditions , neurophysiological measures or training condition ( stress , treatment etc . ) We show that , for a simulated dataset , the fitted posterior distribution of the parameters correlate well with their original value . Moreover , each of the layers of the model ( learning , first level memory and second level memory ) have interesting behavioural correlates in terms of accuracy and RT . These results show that the model is identifiable and that there is a low redundancy in the various layers of the model . Also , different priors over the expected distribution of rewards and environment stability led also to different outcomes in terms of reward rate and RT: subjects whom assumed large environmental stability were likely to act faster , but also to be less flexible and to gain less on average , than subjects whom assumed high likelihood of contingency changes . The second-level memory had different effect , in the sense that large memory tended to be associated with flexible or inflexible behavior , depending on whether past experience corresponded to volatile or stable environment , respectively . This finding also resonates with the habitual learning literature , in which decreased flexibility is also typically associated with short average RT , reflecting lower cognitive cost ( see also [74] ) . However , in contrast to our approach , Keramati and colleagues suggest that adaptations to changes in volatility would rely on switching between two alternative decision strategies: an information-seeking goal-directed controller and a greedy habitual system . Habits would consist in bypassing computationally expensive inference steps in order to maximize reward rate when information gathering is too costly . The AC-HAFVF does not require to achieve model selection prior to making a decision: on the contrary , computational cost of the decision process is optimized automatically during learning and inference . For instance , VPI-guided evidence accumulation ( see Appendix E ) is achieved at a cost virtually identical of inference using a Q-value sampling strategy . Also , the switch from information-gathering to pure value-based selection strategy is natural when the posterior variance of the action values decreases , and the VPI vanishes as the average return becomes certain . Furthermore , using Q-value sampling only , we can see that the behaviour turns from being stochastic and explorative to being deterministic through training , again confirming that the learning and decision scheme we propose can account for the emergence of automatic behaviours . In turn , this could only happen if the environment is considered as stable by the agent . Further developments of the model could show how the threshold ( e . g . [114] ) and the start point ( e . g . [115] ) could be adapted to optimize the exploration policy and the cost of decisions . Finally , we extended our work to MDP and multi-stages tasks . More than adding a mere reparameterization of TD learning , we propose a framework in which the time discount factor is considered by the agent as a latent variable: this opens the possibility of studying how animals and humans adapt their long-term / short-term return balance in different conditions of environmental variability . We show in the results that deep CC provoke a reset of the parameters , and lead subjects to erase their acquired knowledge of the posterior value of γ . We also show that this ability to adapt γ to the uncertainty of the environment can also be determinant at the beginning of the task , where nothing is known about the long term return of each action , and where individuals might benefit of a low expectation of γ that contrasts with the high expectation of subjects that have a deeper knowledge of the environment . The present model is based on forgetting , which is an important feature in RL that should be differentiated from learning . In signal processing and related fields where online learning is required , learning can be seen as the capacity to build on previous knowledge ( or assess a posterior probability distribution in a Bayesian context ) to perform inference at the present step . Forgetting , on the contrary , is the capacity to erase the learning to come back at a naive , initial state of learning . The algorithm we propose learns a posterior belief of the data distribution and infers how likely this belief is to be valid in the next time step , on the basis of past environmental stability . This allows the algorithm to decide when and how much to forget its past belief in order to adapt to new contingencies . This feature sets our algorithm apart from previous proposals such as HGF , in which the naive prior looses its importance as learning goes on , and where the learner has no possibility of coming back to his initial knowledge . This lack of capacity to forget implies that the agent can be easily fooled by its past experience , whereas our model is more resistant in such cases , as its point of reference is fixed ( which is the common feature of SF algorithms , see above ) . We have shown in the results that our model outperforms the HGF both in its fitting capability and its capacity to learn new observations with a given prior configuration . The AC-HAFVF should help to flexibly model learning in these contexts , and find correlates between physiological measures , such as dopaminergic signals [116 , 117] , and precise model predictions in term of memory and flexibility . The SF scheme we have used , where the previous posterior is compared with a naive prior to optimize the forgetting factor , is widely diffused in the signal processing community [2 , 35–37 , 40 , 42 , 118 , 119] and finds grounded mathematical justifications for error minimization in recursive Bayesian estimation [120] . However , it is the first time , to our knowledge , that this family of algorithms is applied to the study of RL in animals . We show that the two algorithms ( RL and SF ) share deep common features: for instance , the HAFVF and other similar algorithms ( [57] ) can be used with a naive prior θ0 set to 0 , in which case the update equations reduce somehow to a classical Q-learning algorithm ( [1] and Appendix B ) . Another interesting bound between the two fields emerges when the measure of the environment volatility is built hierarchically: an interesting consequence of the forgetting algorithm we propose is that , when observations are not made , the agent erases progressively its memory of past events . This leads to counterfactual learning schedule that favors exploration over exploitation at a rate dictated by the learned stability of the environment ( see Appendix C for a development ) . Crucially , this updating scheme , and the consecutive exploration policy , flows directly from the hierarchical implementation of the SF scheme . This work provides a tool to investigate learning rate adaptation in behaviour . Previous work has shown , for instance that the process of learning rate adaptation can be decomposed into various components that relate to different brain areas or networks [32 , 121] . Nassar and colleagues [122] have also looked at the impact of age on learning rate adaptation , and found that older subjects were more likely to have a narrow expectation of the variance of the data they were observing , impairing thereby their ability to detect true CC . The AC-HAFVF shares many similarities with the algorithm proposed by Nassar and McGuire: it is designed to detect how likely an observation is to be caused by an abrupt CC , and adapts its learning rate accordingly . Also , this detection of CC depends in both models not only on the first and second moment of the observations , but also on their prior average and variability . We think , however , that our model is more flexible and biologically plausible than the model of Nassar and McGuire for three reasons . First , it is fully Bayesian: when fitting the model , we do not fit an expected variance of the outcome observed , but a prior distribution on this variance . This important difference is likely to predict better the observed data , as the subjects performing an experiment have probably some prior uncertainty about the variability of the outcome they will witness . Second , we considered the steadiness of the environment as another Bayesian estimate , meaning that the subjects will have some confidence ( and posterior distribution ) in the fact that the environment has truly changed or not . Third , we believe that our model is more general ( i . e . less ad hoc ) than the model of Nassar , as the general form of Eq 13 encompasses many models that can be formulated using distributions issued from the exponential family . This includes behavioural models designed to evolve in multi-stage RL tasks , such as TD learning or Model-Based RL [123] . It is important to emphasize that the model we propose does not intend to be universal . In simple situations , fitting a classical Q-learning algorithm could lead to similar or even better predictions than those provided by our model . We think , however , that our model complexity makes it useful in situations where abrupt changes occur during the experiment , or where long sequences ( several hundreds of trials ) of data are acquired . The necessity to account for this adaptability could be determined by comparing the accuracy of the HAFVF model ( i . e . the model evidence ) to the one obtained from a simpler Bayesian Q-Learning algorithm without forgetting . The richness and generality of the AC-HAFVF opens countless possibilities of future developments . The habitual/goal-directed duality has been widely framed in terms of Model-Free/Model-Based control separation ( e . g . [4 , 11 , 96 , 124–128] ) . Although we do not model here a Model-Based learning algorithm , we think our work will ultimately help to discriminate various forms of inflexibility , and complete the whole picture of our understanding of human RL: in the usual Model-Free/Model-Based control duality , overconfidence in a Model-Free controller means that the subject will need to go through the same sequences of state-action transitions over and over to downweight actions situated early in a sequence . This contrasts with Model-Based control , which can immediately adapt its policy when an action situated far from the current state is devaluated [129] . The current implementation of the AC-HAFVF can model a lack of flexibility due to an overconfidence in the volatility of the environment , whereas adding a Model-Based component to the model might help to discriminate a lack of flexibility due to the overuse of a Model-Free strategy that characterizes the Model-Free/Model-Based paradigm . This balance could be learned and , in turn , be subject to forgetting . In short , implementation of Model-Based RL in an AC-HAFVF context might enrich greatly our understanding of how the balance between Model-Based and Model-Free RL works . This is certainly a development we intend to implement in the near future . In conclusion , we provide a new Model-Free RL algorithm aimed at modelling behavioural adaptation to continuous and abrupt changes in the environment in a fully Bayesian way . We show that this model is flexible enough to reflect very different behavioural predictions in case of isolated unexpected events and prolonged change of contingencies . We also provide a biologically plausible decision making model that can be integrated elegantly in our learning algorithm , and completes elegantly the toolbox to simulate and fit datasets .
|
In stable contexts , animals and humans exhibit automatic behaviour that allows them to make fast decisions . However , these automatic processes exhibit a lack of flexibility when environmental contingencies change . In the present paper , we propose a model of behavioural automatization that is based on adaptive forgetting and that emulates these properties . The model builds an estimate of the stability of the environment and uses this estimate to adjust its learning rate and the balance between exploration and exploitation policies . The model performs Bayesian inference on latent variables that represent relevant environmental properties , such as reward functions , optimal policies or environment stability . From there , the model makes decisions in order to maximize long-term rewards , with a noise proportional to environmental uncertainty . This rich model encompasses many aspects of Reinforcement Learning ( RL ) , such as Temporal Difference RL and counterfactual learning , and accounts for the reduced computational cost of automatic behaviour . Using simulations , we show that this model leads to interesting predictions about the efficiency with which subjects adapt to sudden change of contingencies after prolonged training .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[] |
2019
|
Learning and forgetting using reinforced Bayesian change detection
|
This paper presents the development of an agent-based model ( ABM ) to incorporate climatic drivers which affect tsetse fly ( G . m . morsitans ) population dynamics , and ultimately disease transmission . The model was used to gain a greater understanding of how tsetse populations fluctuate seasonally , and investigate any response observed in Trypanosoma brucei rhodesiense human African trypanosomiasis ( rHAT ) disease transmission , with a view to gaining a greater understanding of disease dynamics . Such an understanding is essential for the development of appropriate , well-targeted mitigation strategies in the future . The ABM was developed to model rHAT incidence at a fine spatial scale along a 75 km transect in the Luangwa Valley , Zambia . The model incorporates climatic factors that affect pupal mortality , pupal development , birth rate , and death rate . In combination with fine scale demographic data such as ethnicity , age and gender for the human population in the region , as well as an animal census and a sample of daily routines , we create a detailed , plausible simulation model to explore tsetse population and disease transmission dynamics . The seasonally-driven model suggests that the number of infections reported annually in the simulation is likely to be a reasonable representation of reality , taking into account the high levels of under-detection observed . Similar infection rates were observed in human ( 0 . 355 per 1000 person-years ( SE = 0 . 013 ) ) , and cattle ( 0 . 281 per 1000 cattle-years ( SE = 0 . 025 ) ) populations , likely due to the sparsity of cattle close to the tsetse interface . The model suggests that immigrant tribes and school children are at greatest risk of infection , a result that derives from the bottom-up nature of the ABM and conditioning on multiple constraints . This result could not be inferred using alternative population-level modelling approaches . In producing a model which models the tsetse population at a very fine resolution , we were able to analyse and evaluate specific elements of the output , such as pupal development and the progression of the teneral population , allowing the development of our understanding of the tsetse population as a whole . This is an important step in the production of a more accurate transmission model for rHAT which can , in turn , help us to gain a greater understanding of the transmission system as a whole .
The tsetse fly ( genus: Glossina ) is the vector for human African trypanosomiasis ( HAT ) or sleeping sickness , a neglected tropical disease caused by two sub-species of the protozoan parasite Trypanosoma brucei s . l . : T . b . rhodesiense , in eastern and southern Africa and T . b . gambiense in West Africa [1] . T . b . rhodesiense HAT ( rHAT ) is a zoonosis , affecting a wide range of wildlife [2 , 3] and domestic animals , particularly cattle [4] , presenting in humans as an acute disease [5] . The history of HAT in sub-Saharan Africa is characterised by long periods of endemicity where the disease self-sustains at low background levels , with periodic epidemics in regional foci [6] . As sleeping sickness is a neglected tropical disease , treatments are often out-of-date , difficult to administer , physically invasive and partially validated , with the prospect for future developments of more effective treatments being limited ( e . g . [7–11] ) . Furthermore , where tools are available , HAT is rarely prioritised due to competing public health interests [12] . In terms of disease prevention , there is currently no immunological prophylaxis to stop infection in humans [13] , made difficult to produce due to the parasite being able to evade the host's immune response by altering the antigenic character of its glycoprotein surface coat [14] . Given these difficulties with preventing and treating HAT infection in humans , it is not surprising that mitigation strategies focused on vector control have seen success ( e . g . [15–18] ) , given that the tsetse fly is not only required for transmission , but also for several stages of parasite development [19 , 20] . Despite such efficacy , the control of the disease in tsetse ( and , therefore , wildlife ) in game reserves and other protected areas is complicated by ecological , conservationist and environmental considerations [21–23] . Gaining a greater understanding of the population dynamics in a tsetse population appears to be an attractive goal , considering that such an understanding could lead to the development of more targeted vector control strategies which have a less adverse ecological impact , while also allowing a more plausible understanding of the rHAT transmission system . For the latter , demographic growth ( through the availability of food and habitat ) and climate changes ( affecting tsetse development and mortality rates ) are two factors which could affect tsetse population dynamics , and ultimately affect the transmission system [24 , 25] . As a result of the significant role that a tsetse population has in determining the rate and distribution of rHAT transmission , this paper considers the tsetse sub-component of the larger rHAT transmission system in detail , with the ultimate goal being the creation of a more accurate representation of the transmission system as a whole . Collecting comprehensive data on populations of tsetse in the field is expensive , complex and time consuming and , consequently , numerous attempts have been made to model tsetse populations as part of vector control or HAT transmission studies ( e . g . [26–29] ) . Some models incorporate climatic drivers which create fluctuations in the tsetse population through the seasons ( e . g . [30–33] ) . One recent example used agent-based modelling ( ABM ) techniques to simulate a simple fluctuation in tsetse population size through different seasons by altering the length of a predetermined lifespan for tsetse , depending on whether the tsetse emerges in the dry ( 2 months ) or wet season ( 3 months ) [31] . Incorporating more detail , [33] used known relationships between temperature and different life events and processes , such as mortality and the length of the pupation period , as parameters when constructing a population model for vector control . ABMs are “a computerized simulation of a number of decision-makers or agents , and institutions , which interact through prescribed rules” [34] . ABMs have been described as a “third way” of conducting scientific research , incorporating both deductive since ABMs start with basic assumptions , and inductive approaches , as they produce simulation data to analyse [35] . However , Epstein [36] suggests that rather than inductive or deductive , ABMs should be considered as “generative” tools in that , through the initialisation of a population of autonomous agents in a relevant spatial environment , one can allow the agents to interact given a simple set of local rules , and generate , from the bottom up , the macroscopic behaviour and regularity of the population as a whole . Such an approach lends itself well to both the investigation of the HAT transmission system as a whole and the tsetse populations and their dynamics as a component . Starting with tsetse population dynamics , much is written about how varying climatic conditions have different impacts on various tsetse life events and processes e . g . : pupal period duration ( e . g . [37] ) , probability of pupal death ( e . g . [38 , 39] ) , and time between oviposition ( e . g . [40 , 41] ) . Representing observations made from samples acquired both in the field and laboratory studies , these patterns provide us with a solid framework to model the larger population , for which comprehensive data are much more difficult , if not impossible , to acquire . By initialising a tsetse population as individuals , each abiding by rules set by the above behavioural patterns ( and others relating to feeding , mating and age-dependent mortality ) , plausible population level outcomes such as fluctuations in population size should be observable as the simulation progresses . When the HAT transmission system is incorporated into an ABM for acquiring preliminary knowledge of the disease transmission system , the constructed model becomes a representation of a complex system ( e . g . [42–44] ) , given that the prevalence of the disease is a complicated emergent phenomenon produced by relatively simple , individual specific rules ( both vector and host ) concerning movement and resource acquisition . In a complex system , the causes of emergent phenomena cannot easily be decoupled and explained by specific parts of the system [45] with , in this case , the model landscape and agent behaviour creating variation in the timing , location and probability of infection as a result of their influence on variability in contact patterns between vector and host [46 , 47] . In this way , ABMs could be considered the most appropriate way to investigate both the HAT transmission system , and tsetse fly dynamics as a sub-component , allowing the representation of interdependent processes such as how individuals interact with each other and their environment through space and time more easily than is possible through more traditional epidemiological techniques [48] . In previous work , an ABM of rHAT transmission was produced using a spatialized approach , incorporating factors often overlooked ( e . g . human behaviour and activity-based movement; density and mobility of vectors; and the contribution of additional hosts ) [27] . This paper presents the first ABM which considers the effect of climatic factors on individual tsetse and their life processes in detail , while also considering the effect this has on rHAT transmission in a large study area in Eastern Province , Zambia . Through the incorporation of seasonality parameters into an existing fine spatial and temporal scale ABM of rHAT transmission in the region [27] , the aim was to develop a greater understanding of tsetse population dynamics through simulation , and subsequently produce a more plausible model of rHAT transmission . The incorporation of such data is vital where transmission rates , and indeed the transmission system as a whole , are to be explored over multiple years . The existing model provided a suitable starting point for the simulation of these seasonal parameters by modelling tsetse flies at the individual level , along with different life events for which durations and probability of occurrence can be climatically constrained . Ultimately , the modified model was implemented with the aim of answering the following research questions: throughout the year , how does the tsetse fly population fluctuate both as a whole , and within different life stages ( e . g . pupal , teneral , mature ) ? Under the caveat that a plausible model has been produced , what rates of disease transmission are observed , and how do these vary seasonally ? Such a model will allow for future exploration of long-term mitigations strategies , alterations to the demographic make-up of the study area , and climate change scenarios .
Eastern Province , Zambia is situated in southern Africa , sharing borders with Malawi ( to the East ) and Mozambique ( to the South ) . The Luangwa Valley is an extension of the Great Rift Valley of East Africa , traversing the Zambian Eastern , Northern and Muchinga Provinces . The valley is characterised as a flat bottomed valley bounded by steep , dissected escarpments which rise to a plateau at approximately 900–1000 m [49] . Different types of vegetation are observed at different altitudes , with valley areas consisting mainly of mopane woodland and patches of grassland , while the natural vegetation on the escarpment and plateau is miombo woodland , interspersed with munga woodland [50] . The study area spans a sparsely populated region of the Luangwa Valley . Villages are small ( between 5 and 20 households ) and inhabitants are predominantly subsistence farmers . The data collection area and region to be modelled consists of a 75 km transect which starts close to Mfuwe airport in the north , and runs southwards along the Lupande River and its distributaries ( Fig 1 ) . Average monthly temperature and rainfall measurements collected at the Mfuwe airport ( 1982–2012 ) weather station are reproduced in Fig 2 [51] . There are three main seasons in Zambia’s tropical climate: the rainy season spans November to April ( wet and warm ) with mean monthly rainfall peaking at 210 mm in January . After the rains , a cold and dry period occurs prior to August , in which May is the hottest and wettest month , with mean temperatures below 23°C and mean rainfall below 3 mm . The hot and dry season usually spans August , September and October , with mean temperatures reaching 28°C in October accompanied by 17 mm of rainfall on average , the first after four dry months in succession [49 , 51] . The Luangwa River and its main tributaries are perennial , and although flash flooding occurs in all rivers during the wet season , the smaller rivers which drain the valley floor dry out during the dry season and flow during the rains [52] . rHAT is endemic in the Luangwa Valley , first being reported in 1908 [53] . G . m . morsitans was not originally considered a vector of rHAT in the valley , despite 50% of domestic and game animals in the Valley having been observed to harbour trypanosomes [54] . In the early 1970s , a large rHAT outbreak occurred in Isoka ( 241 case in 3 years ) attributed to fly encroachment from Luangwa [55] . Wildlife had been observed to reside in Isoka for several months during the rainy season , migrating away during the dry season . In 1973 , early diagnosis and improved treatment methods were introduced , and case numbers fell [56] . Today , cases of rHAT continue to be reported in the Luangwa Valley . Mid-Luangwa Valley has recently experienced increased immigration of people seeking fertile land . Land pressure has resulted in human settlement in increasingly marginal , tsetse-infested areas , previously avoided for fear of disease risk to introduced livestock . Households grow cotton as a cash crop and maize and groundnuts for home consumption [49] . These anthropogenic changes have the potential to destabilise current trypanosomiasis transmission cycles , resulting in increasing prevalence of trypanosomiasis in both human and animal hosts , and the spread of rHAT into previously unaffected areas . Risk factors include human proximity to the large wildlife reservoir in the South Luangwa National Park to the north-west [2] , and ever-increasing livestock and human density on the plateau . Little is known concerning tsetse-trypanosome-human interaction in the region . Therefore , the ABM has the potential to enable exploration of contact risk within communities . Furthermore , with climate changes expected to occur in the near future , such as reduced annual rainfall , increased storm events and increased temperature [57][58] , it is becoming increasingly important to understand how climate factors can affect tsetse populations , particularly in areas such as this , where increases in temperature could see the tsetse habitat spreading further up the valley to more populous areas . This paper describes a new , seasonally sensitive ABM for rHAT/animal African trypanosomiasis ( AAT ) , based on an earlier , non-seasonal model that was constructed using data derived from a detailed rHAT , AAT , and G . m . morsitans ecological survey , undertaken in 2013 , in Eastern Province , Zambia [27] . Due to the fine spatial and temporal scales used to model the system , and the number of mechanisms incorporated ( e . g . , tsetse reproduction , tsetse feeding , human agent movements using real-world routines and pathfinding techniques [59] ) , the model was complex and its data inputs were numerous . As a result , only new data and modifications to the original model are described here . A detailed description of the original , non-seasonal model framework , and the data used to construct it , can be found in [27] . The previous iteration of the model included a longer pupal duration in males than in females , as suggested in the literature ( e . g . [37 , 60] ) , and so for each larva deposited during the simulation , a 35 and 30 day pupal period was included for males and females , respectively , represented as a period of inactivity . However , pupation is known to be temperature sensitive with pupal periods decreasing with increasing temperature , a relationship observed by Phelps and Burrow’s laboratory experiments at constant temperatures [37] . Hargrove [41] utilised the data to present a near perfect fit for pupal duration at temperatures between 16°C and 32°C ( r2 = 0 . 998 ) ( see Fig 5 ) , represented by Eq 2: r=k31+e ( a+bt ) , pupalduration=1r , Eq 2 Where: t = temperature , for males: a = 5 . 3 , b = -0 . 24 and k3 = 0 . 053 and for females: a = 5 . 5 , b = -0 . 25 and k3 = 0 . 057 . Given the excellent fit to the data and the large variation in pupal periods expected within the temperature range found in the study area ( 19°C = ~60 days , 28°C = ~20 days ) , variation in pupal duration with temperature is clearly an important factor to incorporate in the model . The previous , non-seasonal ABM provided the majority of the methods and data used in the current version of the simulation , and so readers are referred to [27] for greater detail and only a summary is provided here . Census data were used to locate and initialise the human and animal populations living in the households shown in Fig 1 . A sample of resource-seeking routines sorted by gender and age was taken in the field ( see supplementary information of [27] ) , and a set of plausible paths from each village to each resource was created using a pre-processing A* pathfinding technique [59] . For tsetse , an estimate of the total apparent population size , density and distribution was provided . Four agent types were included in the ABM , together with an areal representation of wildlife . Humans , cattle , other domestic animals and tsetse used in the ABM were constructed as four separate classes , with populations modelled 1:1 with the data collected in the census ( e . g . 16 , 024 human agents ) and the estimated tsetse population discussed previously . Each class had its own initial information and storage structures for events that occurred through the simulation . The ABM was written in Python 2 . 7 using an object-oriented framework , and run on the Lancaster University High End Computing ( HEC ) Cluster , with all spatial data being processed using Quantum GIS 1 . 8 . 0 . The subsequent sections draw attention to any modifications between the original , non-seasonal modelling framework and the new ABM model , while also describing how the climatic drivers affecting the tsetse population were incorporated into the model . The initial iteration of the model was split into 2 , 400 time-step ( or tick ) days , as the more frequent the tick , the smaller the jumps made by agents as the simulation updates , and the less chance of missing potential interactions . However , this method was restrictive in terms of memory usage and CPU time required to run just six months of the simulation . Further tests were carried out to establish how coarse the temporal resolution could be made before the number of simulated domestic host-vector contacts was reduced , and a greater daily probability of wildlife feed was required to maintain the tsetse population levels . It was established that 600 ticks per day ( 2 . 4 minutes per tick ) allowed the simulation to progress with no obvious effect on human , cattle and other domestic animal bite numbers , while requiring a very similar daily wildlife bite probability to produce a stable tsetse population ( 37% chance per day of a hungry tsetse taking a wildlife bite , compared with 35% in the previous version ) . As a result , 600 ticks per day were used to produce the results of this investigation , which required approximately 4 . 5 GB of RAM per simulation run on the high performance machine , and 24 hours of CPU time per simulated year . To capture the effect of seasonality on the tsetse fly population , daily temperature was calculated every 24 hours using the interpolation method discussed previously , and set as a global variable for the simulation . For each female , once mated , the number of days since mating was compared with the birth interval calculated using Eq 1 and the daily temperature . If and when the number of days since mating exceeded the interval calculated on a given day , a pupa was deposited . A count of the number of days since mating was replaced with a count of the number of days since last offspring , and Eq 1 was used again on a daily basis ( using the alternative constants for further births ) , until another birth occurred . This process was repeated for the duration of a female tsetse fly’s lifespan . There was an equal chance of each tsetse offspring being male or female , and each pupa was deposited in a bush area in which the female tsetse rested during the previous night . A rolling average of the temperature that each pupa has experienced since birth was calculated and attributed to each individual . This temperature was used to determine each individual’s pupal duration , given that if a pupa’s age exceeded the pupal duration calculated using Eq 2 , the pupa would emerge as a teneral fly . It was considered important to use a rolling average of temperature here as the length of a pupal period can span months with quite different temperatures . As described previously , rather than a single probability used to decide whether a pupa would die during its entire pupal period , a variable daily probability of pupal death was included , increased in some months to account for losses observed in the rainy season . Should the probability be exceeded for a pupa , that tsetse was removed from the simulation . Death could result from pupal mortality , starvation , or if a tsetse fly exceeded the daily mortality rate calculated by sex , age and temperature ( Eq 6 , Figs 7 and 8 ) . The mortality rate was calculated individually for each teneral and mature fly , and if the probability was exceeded , the tsetse was removed from the simulation . Starvation occurred if a tsetse tried and failed to feed before a given period of time had elapsed . The starvation element was more strict for teneral flies ( 3 days instead of 5 days ) highlighting their increased vulnerability and reduced flight strength . In the previous version of the simulation , 75 teneral files were added to the simulation for the first 35 days to account for pupae deposited prior to the start of the simulation . As this version of the simulation started in August , and the simulated climate quickly became hostile for teneral flies as temperature increased , 500 teneral tsetse were required per day for the first 45 days , which is representative of average simulated pupal maturation rates during September as the simulation progressed ( see Results ) . In the original model , in the absence of climatic factors , a scaling factor for adult fly mortality was required to offset fly starvation within the simulation . This value was set at 55% . Although this scaling factor is still required in this iteration of the model due to the same starvation element , the incorporation of temperature dependent mortality , and more detailed mechanisms for modelling pupae , has reduced the required level of scaling to 80% To allow the model to initialise and stabilise , the simulation was run for a year before the results for this paper were produced , allowing a ‘burn-in’ period . For example , the results presented below are representative of years 2–4 of the simulation . 100 repeat simulations were used to produce the results presented here .
At the end of the three year simulation , a relatively stable population record was observed in both the male and female tsetse populations , with both exhibiting a double peak in response to the climatic driver ( Fig 9 ) . Each year , until peak temperature was reached in October and November , the population slowly increased , with each gender’s population size increasing by approximately 2000 flies . Such population increases during this hot and dry season could be attributable to the absence of a boosted pupal mortality which is observable during the rainy season [61] , with increasing temperatures having a greater impact in reducing pupal duration and the period between births , than increasing tsetse teneral and mature tsetse mortality . During the rainy season ( November-April ) , this population gradually fell to an annual low , a result of peaks in pupal mortality at the start of the rainy season , and high temperatures causing increased mortality in the annual peak population of teneral flies ( see Fig 10 ) ( now emerged after a high period of births discussed previously—birth numbers can be seen in Fig 11 ) . During this period , with a reduced number of pupae to develop , and teneral tsetse to mature and start reproducing , the higher temperatures no longer aided a growth in population as there were fewer pupal maturations and birth rates to ‘accelerate’ ( Fig 11 ) At the end of the rainy season , the tsetse population gained a small boost due to a plateau in temperature , and the drop in population slowed through the cool and dry season ( May to July ) , although recovery did not start during this period as temperatures were too low to aid rapid repopulation of the tsetse , and the pupal population was still recovering ( Fig 10 ) . Fig 12 presents the different possible modes of tsetse death included in the model , and how the rates varied as the simulation progressed . Non-starvation death represented the deaths attributable to the age-temperature dependent mortality model defined by Eq 6 , and was consistently responsible for the largest number of daily deaths , peaking in the period of highest temperature with approximately 350 deaths per day . Unsurprisingly , given its temperature dependency , the mortality shape closely aligned to mean monthly temperature , except for a period in February and March after the pupal population was reduced by a period of high pupal mortality during the rainy season , resulting in a reduced teneral population and , therefore , fewer adult deaths . Deaths due to starvation followed closely the general pattern of population size , with teneral starvations being particularly low–likely a result of the low daily teneral population size ( ranges between 100 and 400 –Fig 10 ) and the teneral tsetse population having the highest age-temperature dependent mortality rate . Using the Ackley and Hargrove model [61] for pupal mortality produced peaks prior to the rainy season and , to a lesser extent , after the wettest months ( Fig 12 ) . The ratio of pupae to mature tsetse was approximately 2:1 at any given time , with the mature to teneral population ranging between 15:1 at the peak of population size and 25:1 when population sizes were generally lower . Across the three year simulation , the approximate incidence rate for human and cattle rHAT infections was 0 . 355 per 1000 person-years ( SE = 0 . 013 ) , and 0 . 281 per 1000 cattle-years ( SE = 0 . 025 ) . There were 11 human infections each year on average ( i . e . per year , per run ) , and 2 cattle infections . Fig 13 illustrates how these infections clustered spatially and by season . The aggregate number of infections across all years and each of the 100 repeats was used to produce this heat map due to the low infection numbers . There was not much spatial variability through the seasons despite the variation in tsetse population size . However , the number of infections reduced during the second half of the rainy season with the lowest density of infections observed during the cool and dry months . Two hotspots are visible in each of the seasons , each with elongated elements suggesting that frequently used paths were sources of interaction between vector and human host . This is possibly most visible in the north as east-to-west movement here could represent movement between villages and the river , a hypothesis which is given support by observations of infection by activity ( Table 1 ) which suggest that in each season , water collection accounted for approximately 25% of human infections , second only to school trips which accounted for 49% to 51% of infections . No human infections were acquired whilst watering or grazing cattle , while the third highest number of infections occurred when farming . There was little variation in infections by activity between the seasons . With the observed high proportion of infections coming from school trips , it is unsurprising that 5–10 year olds and 10–18 year olds had the highest infections rates ( Table 2 ) . Infection rates were generally lower in the cool and dry season , peaking in the hot and dry season . Table 3 shows that the highest incidence rates were observed amongst immigrant tribes , with the only indigenous tribe ( the Kunda ) exhibiting one of the lowest infection rate across each time period , despite making up over 70% of the population . Infection rates observed by gender and cattle ownership were comparable across time periods , with males and cattle owning households exhibiting marginally higher infections rates in comparison to females and households without cattle ( Table 4 ) . Infections acquired and matured within the tsetse population fluctuated as the three year simulation progressed , with a small year-on-year increase in average infections both in the midgut and salivary gland ( Fig 14 ) . On average , the peak time of salivary gland infection development was at the beginning of the rainy season , which reflects the period of highest tsetse densities plus a time-lag for development of mature infections in the fly .
The first plausible individual-based model representation of a real world tsetse population was created allowing a simulation of the system over multiple years . The model was specified using temperature-dependent parameters derived from the literature , detailed human and animal information from acquired datasets , and expert opinion , and an estimate of the initial tsetse population size and distribution . For example , the pupal population which was completely emergent from the model ( as no initial pupae data were inputted ) corresponded with literature findings that pupae are comparatively difficult to find in the rainy season , and that the pupal population will be greater than that of the developed flies [38] , unsurprising considering that the parameters suggest that pupae are ‘safer’ than teneral flies , pupal duration is at least 3 weeks , and a constant flow of developing pupae is required to replace teneral files which are dying or maturing . In addition , the ratio of female-to-male tsetse fluctuated around 2:1 , a change from the simpler , non-seasonal model [27] , but more in line with estimates in the literature [67] , possibly as a result of running the simulation for longer , and with the addition of climate-driven parameters . The shape of the mature population was comparable to samples of tsetse collected in the region of the South Luangwa National Park ( Regional Tsetse and Trypanosomiasis Control Programme ( RTTCP ) data reported in [22] ) , Eastern Province , Zambia [68] , G . pallidipes in neighbouring Zimbabwe [61] , and similar , yet less detailed , ABM studies [31] . The peak adult population of around 6500 flies suggests that the relatively crude technique used to extrapolate sample data from tsetse surveys for initial model construction ( see [27] for more detail ) produced a reasonable estimate with 5250 flies . Furthermore , the small teneral population observed is perhaps not a surprise , given that the teneral stage is a brief transition with a gradual input of developing pupae , and high mortality rates coupled with maturation to adult fly on first feed as outputs . The decrease in pupal population during the rainy season , combined with a consistently small teneral population highlights how one or two years with a very hot and wet rainy season could have serious consequences for a tsetse population , with a reduction in pupal development during periods of high mortality , and high temperatures killing more teneral tsetse reducing the birth rate over subsequent months . Similarly , such a relationship could occur over the coming years in response to climate change , with IPCC reports suggesting that more extreme rainfall events could occur , along with a rise in temperature over the next 50 years ( e . g . [57 , 58] ) . As a result , it is not surprising that some studies have suggested that certain tsetse fly populations could face extinction within the next 50 years [69] . Future studies will consider using the present model as the basis to test future climate change scenarios and examine the response in the tsetse population to such perturbations . The model suggested similar incidence rates for rHAT infection in humans and cattle , which is likely to be a response to both the fact that the majority of the cattle were in households at the south of the transect , away from the tsetse zone ( only approximately 550 of 2925 cattle were within close proximity of the tsetse zone ) [27] , and that humans were modelled to be much more active than cattle in the simulation , travelling more frequently away from the home . The latter point is corroborated by similar observations of human incidence rate in both cattle owning and non-cattle owning households , particularly as no human infections occurred while tending to cattle in the field or by the river . As with observations in the previous study , collecting water and school attendance provided the highest proportion of infections by some margin , and is likely to be in response to the high frequency of both trips within the simulation and , for schools , the longer distances travelled to a sparse resource , and the time of day of the trips coinciding with tsetse activity . In support of these simulated observations , a recent study of rHAT infections in Zambia found that almost half of the observed female infections were found in school-age children [70] . The data for males suggested fewer infections in children . This perhaps reflects that school attendance in the model is overestimated for the male population , and , in reality , young men may be needed to work to provide for the family at a younger age . The high incidence rates observed in immigrant tribes gives weight to the suggestion that as populations move down the plateau and into the valley , people are increasingly occupying marginal land , and increasing their exposure to the tsetse fly . As a result , future studies using the model will look to investigate how influxes of people into the region and the associated development affects the tsetse fly population in terms of habitat availability , but also how infection patterns respond to the perturbation of the system . Six cattle infections were used to seed the model at the beginning of the simulation ( along with five goats , one dog and two pigs ) , to reflect the estimated prevalence of T . b . rhodesiense in the sample of animals from the study transect . The model was also implemented with 10 humans infected at the simulation start , to take into account information from medical teams in the region , who suggested that there had been two reported cases in the past year , and the known high levels of under-reporting and under-detection in the area , and further afield ( e . g . [22] ) . For example , one study suggested that levels of under-detection of rHAT could be as high as 12 cases for every one identified [71] . Furthermore , the recent study in Zambia found that , when a period of more active surveillance was adopted , the number of diagnoses increased dramatically , suggesting high levels of under-detection in the region . In addition , the investigation found that no action was taken by approximately one quarter of people showing symptoms of rHAT infection prior to diagnosis in the study , and less than half sought medical care from a health facility on first sign of symptoms [70] . As a result , given that there is no under-detection in a simulation , two cattle and 11 human infections on average per year appears plausible , especially when considering that there is currently no removal of infection from the simulation ( and no reduction in activity when infected ) , creating a gradually increasing reservoir of infection , and an increase in tsetse infections ( Fig 14 ) . Despite extensive effort to incorporate seasonality accurately into the simulated system , there are some omissions which were largely unavoidable here , but which should be noted . Firstly , in reality , the spatial distribution of tsetse will change through the seasons , with tsetse concentrated in the dense woodland vegetation in the hot dry season , and more widely dispersed in the wet and cool seasons since tsetse use microhabitats to evade extremes in temperature [60 , 72] . Using an interpolated temperature gradient across a study area through time may allow this behaviour to be simulated , although there would be limitations as temperatures would not reflect sheltered areas utilised by tsetse . As a result , such an implementation should be used in conjunction with a variable land classification , highlighting changes in vegetation with seasons . In addition , no data were available on how human movements vary seasonally in this region at the temporal resolution being modelled , and therefore , the daily routines used are consistent through the year . Finally , it is understood that maturation rate and transmissibility of trypanosomes in tsetse varies with temperature [60] , with early work in Zambia suggesting that higher trypanosome infection rates occurred in G . morsitans in the hot season than in the cold season [54] . However , very little research has been carried out on this subject and , within this study , transmission rates should be low enough for this to have little impact . For the first time in the field of rHAT transmission research , data produced and relationships identified in different studies , focusing on different aspects of the tsetse life-cycle and tsetse-climate interactions , have been incorporated into a single detailed ABM , creating a plausible , stable model , which can ultimately produce a reasonable estimate for transmission rates . While providing an element of validation to these individual entomological studies , through the production of tsetse population curves which closely follow those produced from data collected in the nearby South Luangwa National Park , the model represents a step towards a greater understanding of disease transmission for rHAT in this case , while also being adaptable to gHAT foci in the future . As with all models , the ABM is not without its limitations , for example , variability in tsetse feeding behaviour and preference has been incorporated , but at a basic level . However , through working towards an accurate model representation of the disease landscape one can expect to achieve a greater understanding of the rHAT transmission system , which in turn can help the devising of spatially and temporally targeted mitigation strategies in the future , to help those in need with sustainable solutions , and are more appropriate for spatially marginal communities susceptible to neglected tropical diseases . The dynamics of a tsetse population are difficult to model due to difficulties in acquiring data , and the complexity of the system , but are important to understand due to their importance in rHAT transmission . Gaining a greater understanding of tsetse population dynamics may lead to greater understanding of rHAT transmission and aid future mitigation strategies . This paper presented the first seasonally-varying rHAT transmission model , defined at a fine resolution and modelling directly individual flies , with the full tsetse life cycle as a sub-component . By incorporating numerous parameters estimated from the literature , from data and from expert opinion into such a detailed model , a range of outputs were created which can be used by scientists to analyse and evaluate our current understanding of tsetse fly dynamics and the rHAT disease transmission system , and by decision-makers to investigate alternative mitigation strategies . In its current state , including seasonally varying effects , the model lends itself to modelling future scenarios , including insecticide application and other vector control strategies , the incorporation of a changing climate , the effects of landcover change and human development adjacent to , and within , the biodiverse tsetse habitat .
|
African trypanosomiasis is a parasitic disease which affects humans and other animals in 36 sub-Saharan African countries . The disease is transmitted by the tsetse fly , and the human form of the diseases is known as sleeping sickness . In an attempt to improve our understanding of the mechanisms which contribute to sleeping sickness transmission , a detailed , seasonally driven model of the tsetse fly has been produced , with the theory that a greater understanding of the disease vector’s life cycle will allow developments in our knowledge of disease transmission . The model incorporates previously developed spatial data for the Luangwa Valley case study , along with demographic data for its inhabitants . Tsetse and potential human and animal hosts are modelled at the individual level , allowing each contact and infection to be recorded through time . Through modelling at a fine-scale , we can incorporate detailed mechanisms for tsetse birth , feeding , reproduction and death , while considering what demographics , and which locations , have a heightened risk of disease .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"death",
"rates",
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"ruminants",
"african",
"trypanosomiasis",
"tropical",
"diseases",
"vertebrates",
"parasitic",
"diseases",
"animals",
"mammals",
"parasitic",
"protozoans",
"glossina",
"simulation",
"and",
"modeling",
"seasons",
"developmental",
"biology",
"protozoans",
"pupae",
"neglected",
"tropical",
"diseases",
"tsetse",
"fly",
"population",
"biology",
"insect",
"vectors",
"research",
"and",
"analysis",
"methods",
"infectious",
"diseases",
"zoonoses",
"life",
"cycles",
"protozoan",
"infections",
"trypanosomiasis",
"insects",
"disease",
"vectors",
"arthropoda",
"population",
"metrics",
"trypanosoma",
"eukaryota",
"earth",
"sciences",
"biology",
"and",
"life",
"sciences",
"species",
"interactions",
"cattle",
"amniotes",
"bovines",
"organisms"
] |
2018
|
An agent-based model of tsetse fly response to seasonal climatic drivers: Assessing the impact on sleeping sickness transmission rates
|
What is the mechanism through which transcription factors ( TFs ) assemble specifically along the enhancer DNA ? The IFN-β enhanceosome provides a good model system: it is small; its components' crystal structures are available; and there are biochemical and cellular data . In the IFN-β enhanceosome , there are few protein-protein interactions even though consecutive DNA response elements ( REs ) overlap . Our molecular dynamics ( MD ) simulations on different motif combinations from the enhanceosome illustrate that cooperativity is achieved via unique organization of the REs: specific binding of one TF can enhance the binding of another TF to a neighboring RE and restrict others , through overlap of REs; the order of the REs can determine which complexes will form; and the alternation of consensus and non-consensus REs can regulate binding specificity by optimizing the interactions among partners . Our observations offer an explanation of how specificity and cooperativity can be attained despite the limited interactions between neighboring TFs on the enhancer DNA . To date , when addressing selective TF binding , attention has largely focused on RE sequences . Yet , the order of the REs on the DNA and the length of the spacers between them can be a key factor in specific combinatorial assembly of the TFs on the enhancer and thus in function . Our results emphasize cooperativity via RE binding sites organization .
Cellular response to environmental signals relies on tight gene regulation . Specific recognition of response elements ( REs ) by transcription factors ( TFs ) [1]–[4] and their combinatorial assembly [1] , [5] , [6] on promoters and enhancers is crucial for functional , gene-specific transcription initiation [7] . However , how TFs recognize specific REs along the genome which contains hundreds of thousands of similar RE sequences , how the TFs and their co-regulators assemble to form the enhanceosome which is the functional unit , and how the RE organization on the enhancer DNA ( the order of the REs on the DNA stretch and the spacer sizes between consecutive REs ) play a role in the specificity are still open questions . It has been argued that the cell is populated by a large number of copies of the TF [1] , [4] , [8] . Consequently , all chromatin-exposed REs will be bound by their corresponding TF , if the TF can be favorably accommodated on the enhanceosome [1] , [6] , out-competing other TFs . Conformational ensembles of the RE-bound TFs will undergo allosteric , DNA-induced population shifts , which would alter the TFs' co-factor binding sites to binding-favored states [1] , [9] . Whether the RE-bound TF will affect function depends on factors such as co-factor availability and post-translational modification state , which relate to the cellular environment . RE availability is governed by chromatin packaging and re-modeling [10] , which is determined by the organism's developmental state and cellular environment . Selective RE recognition and TF activation on chromatin-exposed DNA were proposed to reflect three factors [1]: ( i ) the cellular network ( or environment ) which determines the post-translational modification states , co-factor concentration , etc; ( ii ) protein and DNA which exist as dynamic conformational ensembles that re-distribute allosterically upon binding , post-translational modification , external conditions , etc; and ( iii ) tight packing of multiple TFs and co-regulators in enhanceosomes ( or promoters ) . This last factor relates to TFs shapes and sizes , and lengths of intervening DNA stretches between neighboring REs [1] . Although dubbed in the literature as ‘combinatorial assembly’ , the implications as specificity-determining factor in RE recognition have largely been overlooked . Enhanceosomes often involve tens of TFs [1] , [2] , [11] packed along a DNA stretch of several hundreds of bps [1] , [2] , [12] , [13] . REs typically occur in clusters with spacers of variable lengths where REs can also overlap [6] , [14] . Given the large number of possible REs , and RE nucleotide sequence redundancy , the question of how specific TFs prevail over others for given REs is crucial since each RE is associated with a different gene and thus a different function [1] , [2] , [9] . The IFN-β enhanceosome has been a model system for transcription regulation due to its small size . While a typical enhanceosome functions through long-range interactions [15] , the IFN-β enhanceosome sits only tens of bps upstream of the IFN-β gene transcription initiation site and recruits co-factors such as p300 [16] which acetylates histone H1 [17] . The acetylation of histone ‘loosens’ the nucleosomes at the TATA box region , exposing the promoter , thus promoting assembly of the general transcription factor TFIIB and RNA polymerase II [18] which leads to transcription initiation [8] . IFN-β gene expression requires a minimal number of 8 proteins on the enhancer ( Figure 1 ) : ATF-2/c-Jun dimer , four IRF-3 and/or IRF-7 proteins , and an NFκB dimer ( typically p50 and p65 ) [19] , that are activated through three different pathways [20]–[22] . The synergistic [23] , thus orderly [8] assembly is assisted by the HMG I ( Y ) protein [24] , [25] . Once the IFN-β protein is expressed to a certain level , it dramatically increases IRF-7 expression , which further promotes the re-assembly of the enhanceosome with the IRF-7 incorporated [26] . The IFN-β enhancer is composed of four positive regulatory domains ( PRDs ) , IV , III , I , and II from positions −99 to −55 with respect to the transcription initiation site ( Figure 1 ) . Several crystal structures are available [27]–[30] , each of which encompasses part of the enhanceosome ( Figure 1 ) . p50 has been shown to bind to the IFN-β enhancer prior to viral entry , while completion of the assembly of all 8 TFs on the DNA occurs after infection [31] . Of interest , binding of IRF-3 at PRDIII depends on the ATF-2/c-Jun heterodimer orientation on the DNA [32] . PRDIV is composed of two components , the consensus for ATF-2 binding and non-consensus for c-Jun ( Figure 1 ) ; similarly , PRDII is also divided into two non-symmetric parts: the 5′ site is recognized by p50 and the 3′ site by RelA [33] . The four IRF-3 binding sites within PRDI and PRDIII are also arranged in alternative consensus and non-consensus motifs ( Figure 1 ) . Crystal structures of the DNA/IRF-3/IRF-7 complex indicated that IRF-3 binds site C ( and/or A ) and IRF-7D ( and/or B ) . Understanding how these loosely packed TFs communicate with each other and the role of the REs organization in TF selectivity is important for deciphering the mechanism of cooperative assembly . Using MD simulations and modeling we show that despite the sparseness of protein-protein interactions within the enhanceosome , packing along the DNA is already maximized: binding of each of the four enhanceosome TF dimers to their respective REs cooperatively influences the association of a neighboring pair , by partially pre-configuring the overlapped segment of the neighboring binding sites . We also show that the arrangement of consensus and non-consensus binding sites on the DNA facilitates the optimization of the binding of TF partners . The emerging picture from our results is that overlap of REs leads to specificity by enhancing binding of one TF and restricting others . Together , our results can provide an explanation for how specific assembly on enhancer DNA can be achieved despite the limited protein-protein interactions within the assembly .
Molecular dynamics ( MD ) simulations were performed on various combinations of the structural motifs from the 1t2k crystal structure ( Table 1 ) . Figure 2 shows the conformational changes of each simulated system with average structures from the respective trajectories superimposed onto the crystal structure . Several observations were made: 1 ) the full complex was unexpectedly flexible , with the DNA deviating significantly from the crystal structure ( Figure 2a ) . However , the local DNA conformations at the sites where the proteins were bound were relatively stable ( Figure 2b ) ; 2 ) when the two IRF-3 proteins were removed , the DNA bent toward the ATF-2/c-Jun motif with large magnitude , while the DNA conformation in the ATF-2/c-Jun region was reasonably retained ( Figure 2c ) . When the ATF-2/c-Jun motif was removed , the DNA conformation deviated less from the crystal ( Figure 2d ) ; 3 ) when one IRF-3 was removed , the conformation of the DNA at that IRF-3 site drifted away while the IRF-3 bound region still conserved the crystal conformation ( Figure s 2e , f ) . As expected , when simulated alone , the DNA relaxed and lost its unique conformational features such as kinks present in the crystal structure ( Figure 2j ) , and the ATF-2/c-Jun hetero dimer demonstrated high flexibility during the 60-ns trajectory ( data not shown ) . Further analysis showed that IRF-3A anchored well into the major groove throughout the trajectory while IRF-3B was ejected from the major groove to some extent ( data not shown ) . This may have to do with binding specificity and tightness of each IRF molecule . IRF-3A binding was more specific ( more hydrogen bonds ( HBs ) with bases ) while IRF-3B was less so , as further discussed later . These results show that the overall complex is quite flexible due to the sparse protein-protein interactions , and in the absence of protein binding the DNA conformation easily deviates from the protein-bound crystal structure . Although it is expected that the DNA conformation will fluctuate due to the lack of significant interactions between the proteins , the extent of DNA bending in the DNA/ATF-2/c-Jun simulation was still surprising . Inspection of the crystal structure revealed that the DNA conformation at the c-Jun site appeared unusual as it had few contact with the c-Jun arm on the right hand-side ( Figure 1 , Figure 2a in Text S1 ) . To quantitatively characterize the DNA conformation , we calculated groove parameters . Because the four DNA groove parameters are inter-correlated ( larger major groove width corresponds to smaller major groove depth; smaller minor groove width to larger minor groove depth ) , table 2 presents only the minor groove depths . The largest are at -93T and -87A , where His40 and Leu42 from IRF-3A and IRF-3B interact with the minor groove . Comparison with a similar crystal structure illustrates that the uniqueness of this conformation ( Figures 2a , b in Text S1 ) is due to the presence of IRF-3A . This explains the dramatic DNA conformational change in the DNA/ATF-2/c-Jun complex simulation , because upon removal of IRF-3A , the DNA/ATF-2/c-Jun motif had to adjust its conformation to optimize the interactions , resulting in large changes . Further analysis of the binding specificity and experimental biochemical data shed some light on the nature of the cooperativity . The interaction of ATF-2 with the consensus site TGAC ( Figure 1 ) involved specific HBs with bases and electrostatic interactions with the DNA backbone , with an Asn344 side-chain HB with T-99 and G-98 ( C of the complementary strand ) , and Arg352 HB with C ( G of the complementary strand ) . On the other hand , c-Jun interacts with ( non-consensus ) DNA backbone without any specific HB with the bases . Interestingly two other similar structures involving c-Jun ( 1JNM and 2H7H ) were found to have no HBs with bases either , suggesting that indeed binding of c-Jun could be of lower DNA sequence stringency compared to ATF-2 . Combined , these results suggest that ATF-2/c-Jun binding orientation and DNA conformational change were dominated by the requirement to selectively favor IRF-3 binding because IRF-3a and c-Jun share two nucleotides . This also explains the previous experimental observation that in the absence of IRF-3 , c-Jun/ATF-2 were able to bind their respective sites even when the order of the two sites was reversed [32] . However , when IRF-3 was present , the ternary complex was formed only when the two sites had the wild type sequence . Reversing the order of the DNA binding sites for ATF-2 and c-Jun will put the ATF-2 binding sequence next to the IRF site , hampering native IRF-3 binding . Although the sequence of binding events between ATF-2/c-Jun and IRF-3 dimers is unclear , MD simulations revealed that the effect of dimer binding on the DNA conformation is local and limited . It does not appear that one dimer binding pre-configures the entire adjoining RE for the next dimer binding except the overlapped segments . This is evidenced by the relaxation of DNA conformations following removal of either ATF-2/c-Jun or the IRF-3 dimer . Details of DNA conformational changes upon removal of the proteins are given in Figures 2g–j . In the full complex , the groove parameters were dramatically different from site to site ( Figure 2g ) . Upon removal of IRF-3A and IRF-3B , the minor groove next to the ATF-2/c-Jun binding site immediately became larger ( Figure 2h ) although it partially recovered later in the trajectory . When ATF-2 and c-Jun were removed , minor groove widths at IRF sites were reasonably retained , and the c-Jun binding site conformation was partially preserved , particularly near the IRF-3A end ( Figure 2i ) , suggesting that IRF-3 binding can keep the DNA in favorable conformation for c-Jun binding . Although the DNA organization seems to be loose which allows very limited protein-protein interactions between the ATF-2/c-Jun and the IRF-3 motifs , modeling a conformation with IRF-3 binding one-bp upstream revealed that there would be extensive steric clashes between IRF-3 and ATF-2 and c-Jun ( Figure 3 in Text S1 ) . This clarifies why IRF-3A binds to the non-consensus AAAA site , particularly in the presence of ATF-2/c-Jun , even though a consensus site is available one-bp upstream ( GAAA ) . This result shows that binding site overlap was already maximized . Taken together , this suggests that binding cooperativity is achieved largely via overlapped DNA and via limited protein-protein interactions , as evidenced in Figure 4 in Text S1 . As revealed in crystal structures 2O6G and 2PI0 , the apparent conformations of the four IRF-3 ( IRF-3A , -3B , -3C and -3D ) bound to PRDIII and I , respectively , are very similar and are similarly bound to DNA ( Figures 1 , 4 ) . Only one protein-protein interaction occurred among the IRFs ( between IRF-3A and IRF-3C ) ( Figure 1 in Text S1 ) . However , interestingly the protein-DNA interactions are distinct: for example , those for IRF-3A and IRF-3C ( chains e and g from 2O6G ) were more extensive , involving both HBs with bases and electrostatic interactions with DNA backbones ( Table 3 ) , while those for IRF-3B and -3D were mainly with the DNA backbone ( Table 3 ) . Each monomer interacted similarly with the DNA at the minor groove via conserved residues His40 and Leu42 [11] . The significance of the minor groove interaction by these two residues is that the base pairs involved were the two central pairs of the upstream IRF binding site; that is , the two consecutive IRF-3 proteins shared part of the binding site , with one binding from the major groove side and the other from the minor groove . The differences between sites A/C and B/D with respect to the association with DNA lie in the interactions of IRF-3 at the major groove . Arg78 of IRF-3A and IRF-3C formed 3-center HBs with two consecutive G bases at positions −91 and −90 relative to the transcription initiation site ( Figure 3a , c ) ; by interacting with these two G bases , IRF-3A also shared a couple of bps with c-Jun . Arg86 formed HBs with the next two A bases and a C on the complementary chain; Arg81 interacted with the DNA backbone; and interestingly , of the three arginines , only Arg81 was conserved in the IRF-3 family . Ser8 also formed HB with the T that forms a bp with one of the two consecutive As bound to Arg86 . In contrast , in IRF-3B , Arg78 interacted with the DNA backbone , Arg81 interacted weakly with a G base without forming a HB , and Arg86 formed a HB with an A base . For IRF-3D , Arg78 formed HB with a T base; Arg81 interacted with DNA backbone while Arg86 was not in close contact with any DNA bases or backbone . These data show that while all IRF-3 proteins were able to form some HBs with DNA bases , and thus render some specificity , the extent of the specificity varied due to differences in HBs . IRF-3A and -3C were more specific and IRF-3B and -3D were less so . The binding patterns for crystal structure 2PI0 [28] , which differ by one base pair each in sites A and C and by having a 3 bp spacer instead of 2 between binding sites IRF-3C and -3D , were essentially the same in terms of the general specificity trend ( Table 3 ) ; that is , interactions for IRF-3A and -3C were more extensive and specific than for IRF-3B or -3D . Alignment of partial structures revealed that both the protein and DNA segment involved in direct contact matched very well between the two structures ( Figures 2D , E in Text S1 ) . The only difference is that Leu42 and His40 interacted at the minor groove with two terminal bps instead of the two central ones . MD simulations were performed on both 2O6G and 2PI0 which are only slightly different in DNA sequence and complex conformation as described earlier . Simulations of the full complex 2O6G revealed that as expected , DNA fluctuation was smaller in the IRF-3 bound region than at the terminal ( Figures 4a , b ) . When only the DNA/IRF-3A/-3B or the DNA/IRF-3C/-3D complex were simulated , DNA conformation at the IRF-3 bound region was again relatively conserved ( Figures 4c , d ) ; however , the DNA region now deprived of IRF-3 relaxed and deviated from the starting structure . The complexes with the motif combinations DNA/IRF-3A/IRF-3C and DNA/IRF-3B/IRF-3D were also simulated to evaluate whether binding of a dimer on the same DNA side ( AC or BD ) would be different from that on opposite sides ( AB , or CD ) since experimentally , cooperative binding of IRF-3 dimers exists only when both PRDI and PRDIII sites are present [34] . The results from these simulations were similar in terms of DNA conformational dynamics ( data not shown ) . Since binding of two IRF-3 molecules at sites A and C and two IRF-7 molecules at sites B and D is the functionally relevant mode , the DNA/IRF-3A , -3C/IRF-7B , -7D was modeled and simulated as well . While the global conformational changes of the full complex were similar to those of the DNA/IRF-3A/-3B/-3C/-3D simulation results , the protein-DNA interaction profile did reveal some differences . In the DNA/IRF-3ABCD complex , the IRF-3 interaction energies with DNA were more spread while for the DNA/IRF-3AC-7BD complex these interactions were closer to each other ( Figures 5A , B ) , although this feature is not obvious for the 2PI0 complex simulation ( Figures 6A , B ) . Furthermore , the interaction energy for IRF-3BD with DNA was less favorable than that of the IRF-7BD for both 2O6G and 2PI0 complexes ( Figures 5C , D , E and 6C , D ) . Other interaction energies were also calculated and presented in Figure 5 in Text S1 . These results indicate that positions B and D prefer IRF-7 while A and C favor IRF-3 . Analysis of the DNA groove parameters confirmed the limited impact of one IRF binding on the other . In the 2O6G complex , there was significant minor groove narrowing between binding sites ( Figure 4e ) . After removal of IRF-3C and IRF-3D or IRF-3A and IRF-3B , these structural features completely disappeared in the region where IRF-3 was removed , whereas the IRF-3 bound region still remained close to that of the crystal structure ( Figures 4f , g ) . In the DNA/IRF-3C/-3D complex , the minor groove for the spacer region between sites B and C remained narrow , suggesting that much of the binding site for IRF-3B was in a ‘ready’ state because the sites overlap ( Figure 4g ) . Comparison of DNA parameters between the full and partial complexes shows that there is some impact on the overall DNA conformation when the two dimers were bound together ( Figures 4e , f and g ) . In the absence of proteins , the groove parameters were characteristic of free DNA ( Figure 4h ) . Similar to the 1T2K complex , the simulations of 2O6G did not show that the binding of one IRF-3 dimer was able to keep the neighboring DNA full sites in the crystal structure conformation . However , it did show that the DNA conformations in the IRF-3 bound region were well retained . Because the binding sites for the two IRF-3 dimers ( or monomers ) overlap significantly , cooperativity can take place through a pre-organization of the overlapped DNA concomitant with the binding of one dimer . The DNA conformation in the full complex differed from that of the DNA/IRF-3 dimer , suggesting cooperative strengthening of the interaction of each with the DNA . Above , we showed that the interactions of Arg78 were different at the four IRF binding sites , with sites A and C similar to each other , and different from B and D ( Table 3 , Figure 3 ) . The main reason why Arg78 oriented differently at sites B and D relates to the T base preceding the consensus sequence ( Figure 1 ) . Due to the protruding methyl group from the T , Arg78 could not form stable HB with the G within the binding sites and was forced to turn away ( Figures 3b , d ) . When IRF-7 was bound at these two positions , such steric conflict did not exist , fitting snugly at the sites . Figure 7 shows that the binding of IRF-7 at the B site was different from that of IRF-3 at the same site because the residue at the Arg78 position was Thr93 which has a shorter side chain and thus able to make hydrophobic interactions with the otherwise unfavorable methyl group of T ( Figures 7a , b ) . As a result , IRF-7 binds DNA more tightly at sites B or D than IRF-3 ( Figures 7c , d ) . In the 2o61 crystal structure , interactions between IRF-3C and IRF-7D are sparse , with only one HB between Arg60 of IRF-3C and Ser125 of IRF-7D , which is the C-terminal residue ( Figure 1 in Text S1 ) . Interactions between p50 and RelA are extensive ( Figure 6 in Text S1 ) . Analysis of the protein-DNA interactions again revealed an interesting phenomenon . IRF-3C interacts with DNA in a pattern similar to what was described for the 2O6G complex . However , the IRF-7D interaction is more extensive and specific than IRF-3B and IRF-3D in 2O6G ( Table 3 ) . Thr , which replaced the IRF-3 Arg78 , did not need to bend or re-orient to avoid the steric conflict with the underneath T base . Instead , it made van der Waals/hydrophobic contact through the methyl group . In the full complex simulation , both the local conformations and the overall structure were retained relatively well compared with the 1t2k complex , although the conformational difference from the crystal structure was still noticeable ( Figures 8a , b ) : the simulations of DNA/IRF-3C/IRF-7D and DNA/p50/RelA complexes show that DNA conformations were minimally perturbed at the binding sites ( Figures 8c , d ) while the overall structures significantly drifted from the crystal conformation , which was expected . This result illustrated again that the DNA conformation fluctuation and the relatively large movement between the segments was the consequence of the sparseness of protein-protein interactions on different DNA segments . Details of the protein-DNA interaction energies are presented in Figures 8e , f . A few interesting observations can be noted: 1 ) The interaction energies between the DNA and p50/RelA were very similar in the full complex and in the p50/RelA-DNA motif , indicating stable interactions for this association ( data not shown ) ; 2 ) the interaction energies for IRF-3 and IRF-7 with DNA were very similar to each other ( Figure 8f ) , suggesting that the IRF-7 binding at the D ( and the B ) site was more favorable than the IRF-3 binding at the same sites . This observation is consistent with the simulations results of the 2O6G and 2PI0 full complexes with IRF-7 bound at the B and D sites . DNA groove parameter analysis also revealed limited yet observable DNA conformational impact by protein binding at the neighboring site ( Figures 8g–j ) . When the p50/RelA dimer was removed from the complex , the minor groove width downstream of IRF-7D did not change significantly ( Figure 8h ) . However , the conformation in the IRF-3C and IRF-7D bound region did not maintain well in the crystal structure , suggesting that IRF binding was not as tight in the absence of p50/RelA . On the other hand , when IRF-3C and IRF-7D were removed , the p50 and RelA bound portion retained well the crystal structure conformation , highlighting the stability of this protein-DNA motif ( Figure 8i ) . In this case , the DNA conformation for the IRF-7D binding site was also similar to the crystal structure , confirming the impact of p50/RelA binding on the DNA conformation at the IRF-7 site .
The combinatorial assembly mechanism of TFs in the enhanceosome is of paramount importance . Even for the small IFN-β enhanceosome , despite considerable cell biology , biophysics , and structural characterization work , it is still unclear how the three modules are selectively recognized and come together to lead to transcription initiation . From the functional standpoint , the IFN-β enhancesome complex can be roughly divided into three modules: ATF-2/c-Jun , IRF , and p50/RelA sites listed from upstream to downstream ( Figure 1 ) . While we have shown that packing has reached maximum tightness , the complexes demonstrated high flexibility , higher than typically observed in protein-DNA complexes where there exist extensive protein-protein interactions . DNA can be very flexible , capable of forming sharply looped DNA-protein complexes [35] . However , complexes where two proteins bind shoulder to shoulder on a DNA segment with high specificity and extensive protein-protein interactions , allow very limited DNA fluctuations . For example , the complex of the p53 tetramer with DNA presents very limited DNA conformational change or DNA bending , with a maximum of 30 degrees of curvature only when the DNA sequence is optimized [36] , which is evidenced in low resolution experiments . Such dynamic properties can be demonstrated through MD simulations , and is not always captured in crystal structures possibly due to crystal effects . The salient feature that the IFN-β enhanceosome harbors few protein-protein interactions suggests that assembly cooperativity could stem from DNA conformational changes following protein binding; that is , TF binding-induced conformational changes may propagate along the DNA , pre-configure neighboring REs for optimal binding by a second TF , and this could be a key factor in RE recognition . Yet , our results show that the direct effect on DNA conformation by binding of a TF dimer is limited to only the neighboring sites . This is supported by our simulation results that removing a protein molecule from the complexes will cause the DNA conformation to drift away from that in the crystal structure , with only a few bps next to the binding sites reasonably retaining the crystal conformation . Thus , instead of long range DNA allosteric effects , our results suggest that overlap of binding sites is the mechanism of enhanceosome binding cooperativity , between ATF-2/c-Jun and IRF-3A , among IRFs , and between IRFs and p50/RelA proteins . Overlap of binding sites is reasonable and likely to be a broadly utilized enhanceosome mechanism . Constructs with different overlaps of REs and abolished protein-protein interactions may help in delineating the impact of these conformational factors on transcription . Hetero-dimerization of TFs is widely recognized and known to be important for binding specificity and consequently function [37] . Experimental data show that pairs of the enhanceosome TFs are often expressed together . For example , the RelA/p50 and RelB/p50 data suggest that they are synthesized at the same time , and are found in complex with p100 in the nucleus [38] , [39] and bind DNA first [40] . The question is why unique combinations of ATF-2/c-Jun , IRF-3/IRF-7 and p50/p65 ? NF-kB ( p50/RelA ) is a ubiquitous eukaryotic TF which plays critical roles in transcription of numerous genes [41] and is often modified [42] . Like the ATF-2/c-Jun dimer , it is present in most cells and involved in many biological processes including proliferation , differentiation , and apoptosis [43]–[45] . p50/RelA dimerization is important for transcription . Since the binding specificity is high and the dimerization interface is stable , the binding of this motif is expected to contribute significantly to the stability of the enhanceosome . Interestingly , when the spacer between p50/RelA and IRF-7D changes from 2 to 3 nucleotides the transcriptional activity is only slightly affected . Because the two binding sites still overlap by 3–4 bps with the 3-bp spacer , it is understandable that cooperativity , and thus function , is only minimally changed . ATF-2 and c-Jun belong to a super-family of TFs that share the basic-region Leucine-zipper motif but have different DNA binding specificities . The ATF-2/c-Jun heterodimer is more populated and binds DNA tighter than either homodimer [46] . c-Jun by itself recognizes the so-called AP-1/TRE site with the symmetrical sequence TGACTCA while ATF-2 recognizes the ATF/CRD consensus site TGACGTCA , which is also symmetric [47] . The difference is in one bp . This difference may suggest that c-Jun dimer binding is not as specific as the ATF-2 since it binds to smaller sites ( TGA ) while ATF-2 needs two TGAC sites . Combining previous work which shows that the assembly of ATF-2/c-Jun/IRF-3 complex occurs only when the DNA sites were in the ‘right’ order [32] and our simulation results , it is likely that the non-consensus site is only for c-Jun binding since structural analysis demonstrates that it has few specific interactions with the DNA . Thus , nature has designed the DNA sequence and the ATF-2/c-Jun dimer for optimized binding specificity of each TF and cooperativity between neighboring partners . IRF-3 activation requires dimerization through phosphorylation [48] which appears controlled by acetylation [49] . However , the IRF-3 dimerization benefit is not obvious , as there is almost no interaction between the DNA binding domains either on the same or opposite sides of the DNA . In addition , it seems that IRF-3 at sites B and D can be easily replaced by IRF-7 , since IRF-7 binding at these two positions is more stable than IRF-3 binding . Therefore , the initial binding mode of dimeric IRF-3 ( same- or opposite-side of the DNA ) may not be as important as previously thought and IRF-3/IRF-7 dimerization should also be favorable . Because the binding of the IRF DNA binding domain was weak when the other proteins were absent [29] , dimerization may allow concurrent binding , which enhances not only the binding affinity , but also the specificity , excluding other TFs from binding to the same sites . Interestingly , the IRF-5/IRF-7 dimer is a repressor of IFN genes [50] . Further study is needed to gain insight into the structural basis of this difference between IRF-3 and IRF-5 binding . Assembly of a unique enhanceosome depends on factors such as the chromatin state , i . e . , whether the enhancer is available , the TFs concentration and post-translational modification states , and TFs affinity to their respective REs [1]–[3] , [51] , [52] . Specificity also relates to binding of partners ( and cofactors ) since allostery and structural reorganization are always involved in conformational perturbation during binding [53] . A recent analysis of 8mer REs [54] suggested that while each TF has sequence preferences , just about half of the TFs bind to distinct DNA motifs . TFs from even the same family may show large differences in affinity and site preference [2] , [3] , [9] . Related to our case , IRF-4 and IRF-5 both bind strongly to DNA containing CGAAAC segments but weakly to TGAAAG and CGAGAC; and specifically , IRF-3 prefers sites A and C while IRF-7 has higher affinity toward B and D . Although there is distinct sequence preference [55] and some correlation between binding affinity and specificity [56] , RE sequences are not the only factor that determines what will bind . As shown in table 3 , various binding patterns were observed in complexes with similarities at specific positions . For example , binding patterns of Arg78 and Arg86 were different in two crystal structures ( PDB 1T2K and 2PI0 ) at identical non-consensus sites , while other residues including Arg81 , Ser82 and Ala83 interacted with DNA in almost the same way . In one case ( 2PI0 ) , both arginines formed HB with respective bases , while in the other ( 1T2K ) Arg78 only interacted with the methyl groups of two thymines . One of the major differences between the two complexes is that in 1T2K , ATF-2/c-Jun dimer bound upstream of the IRF-3A , which forced Arg78 to point inward and to interact with bases within its own binding sites . As a result , Arg86 adjusted its interactions as well . Similarly , although IRF-3 binding at sites B and D was not optimal relative to IRF-7 , it was able to bind at these sites with adjusted orientations , resulting in transcription upon viral infection . Of interest , TFs from the same family that share similar DNA binding domains often have different functions [57] . These could reflect altered cofactor binding sites , the outcome of RE-induced allosteric propagation . To conclude , our work emphasizes the crucial , yet largely overlooked role of the organization of successive REs along regulatory DNA stretches , such as enhancers and promoters , in specifying TF binding selectivity . To date , efforts have largely focused on analysis of binding sites and derivation of consensus sequences . Yet , the order of REs and the spacers between consecutive REs can also play a critical role ( Figure 9 ) . Spacer sizes determine the TF shape and dimensions: TFs which are too large or too small are disfavored due to either steric effects ( Figure 9a ) or lack of interactions with the adjoining TFs ( Figure 9a ) . Overlapping REs ( Figure 9b ) can function via cooperative effects through the binding of TFs to complementary bases , excluding disfavored TFs or enhancing those with relatively low affinity . We propose that overlap of REs is a general mechanism in enhanceosome assembly , beyond the IFN-β . Finally , the order of the binding sites can also be expected to have a functional significance , with a reversed order ( Figure 9c ) functioning as a repressor . It will be interesting to test the role of spacers by in vivo experiments , where other TFs are also present . Genome searches for identical binding sites but with reversed order are expected to uncover additional occurrences of such a functional mechanism which could be tested experimentally . Combined with current experimental data , our results lead us to propose key factors in RE selectivity and functional TF assembly: exposed ( i . e . not covered by nucleosomes ) enhancer DNA , available for TF binding; RE sequence and order; the length ( positive or negative ) of spacers between REs; the TFs concentration and post-translational modification states; and proteins and DNA conformational ensembles . Here , our study emphasizes the key role of cooperativity in making the REs a functionally unique gene regulation site . RE organization along the DNA and the intervening spacers play a key role in selective combinatorial assembly , and as such , in the regulation of gene expression .
MD simulations were performed on four partial enhanceosome crystal structures and their components [28]–[30] . The composition of each simulation is listed in table 1 . Each system was solvated with a TIP3P water box [58] with a margin of at least 10 Å from any edge of the box to any protein or DNA atom . Solvent molecules within 1 . 6 Å of the DNA or within 2 . 5 Å of the protein were removed . The systems were then neutralized by adding sodium ions . The resulting systems were subjected to a series of minimizations and equilibrations using the CHARMM program ( academic version ) [59] , [60] and the CHARMM 22 and 27 force field for the protein [61] and nucleic acid [62] , [63] , respectively . The production MD simulations were performed at temperatures of 300 degrees Kelvin using the NAMD program [64] and the CHARMM force field . Periodic boundary conditions were applied and the non-bonded lists were updated every 20 steps . The NPT ensemble [65] was applied and the pressure kept at 1 atom using Langevin-Nose-Hoover coupling [66] . SHAKE constraints [67] on all hydrogen atoms and a time step of 2 fs and a nonbonded cutoff of 14 Å with force shift algorithm were used in the trajectory production . Electrostatic interactions were treated with particle mesh Ewald algorithm [68] , [69] . The sizes of the systems were about 110 , 000 atoms and the duration for each simulation was 60 ns . Two complexes were modeled that constituted the DNA IRF-3ac/IRF-7bd with both the 2O6G and 2PI0 templates . In addition , because some of the residues were missing in the crystal structure of 2PI0 , IRF-3 structure at position B was used to model IRF-3 at positions A and D . These complexes were constructed by superimposing the backbone of IRF-3 or IRF-7 onto the proteins that were originally there . The systems were minimized for 2000 steps with the ABNR algorithm . The obtained structures were then solvated and further minimized as described in the previous procedures . DNA parameters were calculated with the CURVES program [70] , [71] .
|
An enhanceosome is a functional unit that consists of DNA segment called enhancer; its transcription factors ( TFs ) ; and their interacting cofactors . To function , the TFs must assemble on their corresponding response elements ( REs ) cooperatively . Understanding how TFs assemble is important because the TF combination on the enhancer spells gene-specific activation ( or repression ) . Traditional studies focused mainly on the derivation of consensus DNA sequences , and the TF interaction with its respective RE . This yielded limited success in deciphering the mechanism of selective TF binding . Here , in addition to the conventional roles of protein and DNA , we studied the organization of REs . The IFN-β enhanceosome is a good example because there are limited protein-protein interactions between consecutive TFs . Our molecular dynamics simulations revealed that cooperativity is achieved via overlap of REs , in addition to sparse protein-protein interactions . That is , because the REs overlap , binding of neighboring TFs affect each other through DNA conformation perturbation . In addition , alternation of consensus and non-consensus REs along the enhancer allows more efficient binding of TFs , while the order of the REs excludes unwanted TFs , and enhances selective TF binding . Our findings emphasize the overlooked role of the order and organization of REs , and the length of spacers between consecutive REs .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"biology"
] |
2011
|
The Role of Response Elements Organization in Transcription Factor Selectivity: The IFN-β Enhanceosome Example
|
Family with sequence similarity 20 , -member C ( FAM20C ) is highly expressed in the mineralized tissues of mammals . Genetic studies showed that the loss-of-function mutations in FAM20C were associated with human lethal osteosclerotic bone dysplasia ( Raine Syndrome ) , implying an inhibitory role of this molecule in bone formation . However , in vitro gain- and loss-of-function studies suggested that FAM20C promotes the differentiation and mineralization of mouse mesenchymal cells and odontoblasts . Recently , we generated Fam20c conditional knockout ( cKO ) mice in which Fam20c was globally inactivated ( by crossbreeding with Sox2-Cre mice ) or inactivated specifically in the mineralized tissues ( by crossbreeding with 3 . 6 kb Col 1a1-Cre mice ) . Fam20c transgenic mice were also generated and crossbred with Fam20c cKO mice to introduce the transgene in the knockout background . In vitro gain- and loss-of-function were examined by adding recombinant FAM20C to MC3T3-E1 cells and by lentiviral shRNA–mediated knockdown of FAM20C in human and mouse osteogenic cell lines . Surprisingly , both the global and mineralized tissue-specific cKO mice developed hypophosphatemic rickets ( but not osteosclerosis ) , along with a significant downregulation of osteoblast differentiation markers and a dramatic elevation of fibroblast growth factor 23 ( FGF23 ) in the serum and bone . The mice expressing the Fam20c transgene in the wild-type background showed no abnormalities , while the expression of the Fam20c transgene fully rescued the skeletal defects in the cKO mice . Recombinant FAM20C promoted the differentiation and mineralization of MC3T3-E1 cells . Knockdown of FAM20C led to a remarkable downregulation of DMP1 , along with a significant upregulation of FGF23 in both human and mouse osteogenic cell lines . These results indicate that FAM20C is a bone formation “promoter” but not an “inhibitor” in mouse osteogenesis . We conclude that FAM20C may regulate osteogenesis through its direct role in facilitating osteoblast differentiation and its systemic regulation of phosphate homeostasis via the mediation of FGF23 .
FAM20C is a member of the “family with sequence similarity 20” . In mammals , this evolutionarily conserved protein family consists of three members: FAM20A , FAM20B and FAM20C . FAM20A was originally observed in the lung and liver and displays obvious differential expression in hematopoietic cells undergoing myeloid differentiation [1] . A viral mRNA transgenic mouse line with an accidental deletion of a 58-kb fragment in chromosome 11E1 encompassing part of the Fam20a gene and its upstream region showed growth disorder [2] . Recently , it was found that FAM20A is also expressed in ameloblasts and its mutations are associated with human amelogenesis imperfecta and gingival hyperplasia syndrome [3] . More recently , FAM20B was shown to be involved in cartilage matrix production and the ultimate regulation on the timing of skeletal development [4] . FAM20C is highly expressed in the mineralized tissues and identified as the causal gene for lethal osteosclerotic bone dysplasia ( Raine Syndrome , OMIM 259775 ) [1] , [5]–[7] . Given the high level of conservation in the C-terminal domains among the three FAM20 members and the their roles observed in the hard tissues , it is tempting to speculate that this evolutionarily conserved family might be a new cluster of molecules performing important functions in the development of the mineralized tissues . Mouse FAM20C , also known as “dentin matrix protein 4” ( DMP4 ) [5] , contains 579 amino acid residues , including a putative 26-amino acid signal peptide at the N-terminus . A C-terminal region of approximately 350 amino acids ( corresponding to residue218-residue569 in the mouse FAM20C sequence ) has been named the “conserved C-terminal domain” ( CCD ) , which is highly conserved among different species [1] . In a previous study , we systematically analyzed the expression and distribution of FAM20C in mouse bone and tooth using in situ hybridization ( ISH ) and immunohistochemistry ( IHC ) methods [7] , which showed that FAM20C was highly expressed in the mineralized tissues; it was detected in the osteoblasts/osteocytes , odontoblasts , ameloblasts , and cementoblasts , as well as in the matrices of bone , dentin , and enamel . FAM20C was also detected in the epithelium of early-stage tooth germs and in the chondrogenic cells of long bones . The high expression levels of FAM20C in the mineralized tissues strongly suggest that it may play an important role in the formation and/or mineralization of these tissues . Hao et al . showed that overexpression of mouse FAM20C accelerated the odontoblast differentiation process and silencing this molecule by siRNA inhibited cell differentiation , implying that this protein may be a factor promoting odontoblast differentiation [5] . Subsequently , Simpson et al . reported that the loss-of-function mutations in the FAM20C gene were associated with lethal/non-lethal osteosclerotic bone dysplasia ( Raine Syndrome ) [6] , [8] , an autosomal recessive disorder characterized by a generalized increase in the density of all bones; these data indicated that FAM20C might be a down-regulator of biomineralization , which apparently contradicts the mineralization-promoting properties of FAM20C observed by Hao et al . In this study , we sought to determine the biological functions of FAM20C via generation and characterization of Fam20c conditional knockout ( cKO ) mice . Our data showed remarkable skeletal defects , along with a significant reduction of serum phosphate and a dramatic elevation of serum fibroblast growth factor 23 ( FGF23 ) in the homozygous Fam20c cKO mice . The phenotypic profiles of the Fam20c-deficient mice resemble those of hereditary hypophosphatemic rickets in humans and rodents resulting from mutations in molecules affecting the regulation of FGF23 [9]–[15] .
The mouse Fam20c gene consists of 10 exons and spans approximately 55-kb . To generate a conditional knockout allele for Fam20c , we constructed a targeting vector with loxP sites floxing exons 6∼9 which are highly conserved across species ( Figure 1A ) ; a number of mutations were identified in this region of the human FAM20C gene in patients with lethal osteosclerotic bone dysplasia [6] . The correct targeting events were confirmed by polymerase chain reaction ( PCR ) screening , and the presence of 5′ and 3′ loxP sites was determined by PCR product sequencing . Two correctly targeted ES cell clones were identified ( Figure 1B , Clones 286 and 297 ) , and both went through germline transmission . F1 Fam20cflox/+ heterozygous mice were crossbred with Sox2 promoter-Cre transgenic mice to generate “Sox2-Cre-Fam20cΔ/Δ” mice , in which exons 6∼9 were removed from both alleles of the Fam20c gene in the epiblasts at post coitum day 6 . 5 ( E6 . 5 ) . The presence of the floxed alleles and the absence of exons 6∼9 in the null alleles were confirmed by PCR genotyping ( Figure 1C ) . The lack of Fam20c mRNA in the Sox2-Cre-Fam20cΔ/Δ mice was shown by reverse transcription PCR ( RT-PCR ) performed with two sets of primers using mRNA extracted from the long bones ( Figure 2A ) , as well as by in situ hybridization ( ISH ) carried out on the long bones ( Figure 2B ) . The lack of FAM20C protein was determined by immunohistochemistry ( IHC ) analyses ( Figure 2C ) performed on the long bones using an affinity-purified anti-FAM20C polyclonal antibody [7] . Both male and female Sox2-Cre-Fam20cΔ/Δ ( homozygous cKO ) mice are infertile , while the Sox2-Cre-Fam20cΔ/+ ( heterozygous cKO ) mice have normal fertility . The Fam20cflox/flox mice and the heterozygous cKO mice did not demonstrate any phenotypic changes compared with their wild type ( WT ) littermates ( data not shown ) , while the homozygous cKO mice displayed remarkable skeletal defects , indicating that the haploinsufficiency of Fam20c has no significant effects on the bone formation . We also bred the Fam20cflox/flox mice with the 3 . 6 kb Col 1a1-Cre mice to generate Col1a1-Cre-Fam20cΔ/Δ mice , which displayed skeletal defects similar to those observed in the Sox2-Cre-Fam20cΔ/Δ mice . In this report , we described in detail the analyses of phenotypic changes in the Sox2-Cre-Fam20cΔ/Δ mice while the X-ray and histology data of the long bone from the Col1a1-Cre-Fam20cΔ/Δ mice were included in one set of the figures to show the similarity between the global and mineralized tissue-specific cKO mice . The data regarding the Fam20c cKO mice refer to the analyses of the Sox2-Cre-Fam20cΔ/Δ mice unless otherwise stated . The osteocytes in Fam20cΔ/Δ mice lost normal morphology and appeared immature as shown by resin-casted scanning electron microscopy ( SEM ) analyses ( Figure 8A and 8B ) , indicating a faulty maturation process from osteoblasts to osteocytes . Backscatter SEM analyses revealed periosteocytic lesions ( “halo” ) surrounding the osteocytes in the Fam20c cKO mouse bone ( Figure 8C and 8D ) . To determine the molecular changes associated with the immaturity of osteoblasts/osteocytes , we examined their terminal differentiation markers: type Ia collagen , dentin matrix protein 1 ( DMP1 ) , and osteocalcin ( OCN ) . ISH ( Figure 8E–8J ) , and real-time PCR analyses ( Table 2 ) revealed a significant downregulation of these markers in the Sox2-Cre-Fam20cΔ/Δ mice . Microarray analyses using total RNA extracted from the calvaria of 3-week-old Sox2-Cre-Fam20cΔ/Δ mice and their WT littermates indicated that among the ∼45 , 000 molecules evaluated , 350 genes were upregulated by over 2 . 0 folds and 185 were downregulated . Real-time PCR analyses on selected genes confirmed the significant changes in a number of biomineralization regulators and key players in the Wnt and TGF-β signaling pathway associated with cell differentiation ( Table 2 ) [9]–[15] , [19]–[27] , suggesting an essential role of FAM20C in the differentiation and mineralization of osteogenic cells . Notably , the most striking transcriptional alteration was FGF23 ( upregulated by ∼110 folds ) , a phosphorus regulator mainly produced by osteoblasts/osteocytes [11] , [28] , [29] . Immunohistochemistry against FGF23 confirmed the dramatic elevation in the bone cells and bone matrix of Fam20c cKO mice ( Figure 8K and 8L ) . The transcript levels of the above genes in the Sox2-Cre-Fam20cΔ/+ ( heterozygous cKO ) mice showed no difference from the WT mice ( data not shown ) . Given the many similarities among the Fam20c cKO mice , Dmp1 KO mice and Hyp mice , we examined the expression levels of Fam20c in the Dmp1- and Phex- deficient mice , and the levels of Dmp1 and Phex in the Fam20c cKO mice by real-time PCR analyses . The Fam20c expression was not altered in the Dmp1 KO mice and Hyp mice ( data not shown ) . The expression of Dmp1 was significantly downregulated ( Table 2 , Figure 8 ) while that of Phex was not affected ( data not shown ) in the Fam20c cKO mice . Ectonucleotide pyrophosphatase/phosphodiesterase ( Enpp1 ) , another molecule involved in regulating phosphorus homeostasis was slightly downregulated in the bone of the Fam20c cKO mice , but the change ( ∼1 . 4 folds ) was not statistically significant from the WT ( data not shown ) . The bone phenotypes in the Fam20c cKO mice appear opposite to those observed in the patients associated with the human FAM20C mutations [6] . These contradictory results raise the question of whether FAM20C functions differently between the two species . The lentiviral shRNA-mediated “knockdown” of FAM20C in mouse preosteoblasts MC3T3-E1 cells , human Saos-2 cells ( osteoblasts isolated from human osteosarcoma ) and human mesenchymal stem cells ( hMSC ) revealed a remarkable downregulation of DMP1 ( Figure 9A–9C ) , along with a significant upregulation of FGF23 in both the human and mouse cell lines ( Figure 9D–9F ) , indicating that FAM20C may function similarly in humans and mice . To examine the role of FAM20C during osteoblast proliferation and differentiation , recombinant mouse FAM20C protein was generated by insect cells using a Bac-to-Bac baculovirus system . The recombinant FAM20C added to the culture of MC3T3-E1 preosteoblasts promoted the mineral deposition ( nodule formation ) in a dose-dependent manner ( Figure 10A ) , and significantly enhanced the transcription of DMP1 , osteocalcin ( OCN ) , and bone sialoprotein ( BSP ) ( Figure 10B ) . Adding recombinant FAM20C to MC3TC-E1 cells did not alter the expression of FGF23 and the proliferation rate of the cells at all tested concentrations ( data not shown ) . Seeing the significant elevation of FGF23 in the bone cells of Fam20c cKO mice , we performed serum biochemistry analyses in 18-day- and 42-day-old mice ( Table 3 ) . The circulating FGF23 level was remarkably elevated in both the 18-day-old ( ∼200 folds ) and 42-day-old ( ∼60 folds ) cKO mice . Accordingly , the serum phosphorus level significantly decreased at both ages ( ∼2 . 5 folds in 18-day-old cKO , ∼2 folds in the 42-day-old cKO mice ) . The circulating PTH level was significantly elevated in both the18-day-old ( ∼8 folds ) and the 42-day-old ( ∼5 folds ) cKO mice . The serum 1 , 25 ( OH ) 2D3 level was significantly reduced ( ∼2 folds ) in the 18-day-old cKO mice , while the serum 1 , 25 ( OH ) 2D3 level in the 42-day-old cKO mice was slightly higher than that of the WT , but the change in the older mice was not statistically significant . The serum calcium level slightly decreased in the 18-day- and 42-day-old cKO mice , but the reduction was not statistically significant from the WT mice . The blood urea nitrogen ( BUN ) level in cKO mice had no statistic difference from that of the WT mice at both ages , indicating that no renal failure was occurring in the cKO mice . The serum biochemistry results of the Sox2-Cre-Fam20cΔ/+ ( heterozygous cKO ) mice were not different from their WT literates ( data not shown ) . The transcriptional levels of renal Klotho and NaPi-2a were slightly lower in the 18-day-old Fam20c cKO mice , and significantly downregulated ( ∼3 folds ) in the 42-day-old cKO mice . The renal 1α-hydroxylase level was significantly reduced ( ∼2 folds ) in the cKO mice at both ages . We also observed remarkable upregulation of renal 24-hydroxylase in the 18-day-old cKO mice ( ∼30 folds ) as well as in the 42-day-old cKO mice ( ∼7 folds ) ( Table 4 ) . The transcriptional levels of these genes in the heterozygous Fam20c cKO mice had no difference from their WT littermates ( data not shown ) . Conditional transgenic ( cTg ) mice expressing the full length FAM20C were generated to test the gain of function in vivo . We obtained 15 lines of cTg mice , and three of them with the transgene expression levels of 4∼8 folds over those of the WT littermates were further analyzed . One line with the highest expression level of the transgene ( approximately 8 folds over the WT , Figure 11A ) was characterized in detail . No abnormalities were observed in the bone of any of the cTg mice by postnatal 6 weeks ( Figure 11B ) . In addition , by breeding the cTg mice with the Fam20c cKO mice , we obtained mice expressing the transgene in the Fam20c knockout background ( designated “Sox2-Cre-Fam20cΔ/Δ-cTg mice” ) . The Sox2-Cre-Fam20cΔ/Δ-cTg mice had no abnormalities in the skeleton by postnatal six weeks ( Figure 11B and 11C ) , indicating that expressing the transgene fully rescued the defects of the Fam20c-deficient mice . Additionally , IHC staining using the anti-FGF23 antibodies showed that the expression of the Fam20c transgene rescued the altered expression of FGF23 in the Fam20c-deficient bone ( Figure 11D ) . The fact that overexpressing FAM20C in the WT background did not cause defects in the bone , along with the observation that overexpressing the transgene rescued the Fam20c-deficient abnormalities , has provided further evidence that the defects in the hard tissues of the Sox2-Cre-Fam20cΔ/Δ mice resulted from the loss-of-function , and were not due to the gain-of-function . In summary , the multipronged approaches in this study demonstrated that inactivation of FAM20C in mice led to rickets/osteomalacia , along with altered levels of serum phosphate and FGF23 . The manifestations of the Fam20c-deficient mice are consistent with a diagnosis of hypophosphatemic rickets . The Fam20c-deficient cells in the mineralized tissues appeared immature and incapable of forming healthy tissues . It is likely that a combination of cell differentiation failure and hypophosphatemia resulting from the FGF23 excess led to the skeletal defects in the Fam20c-deficient mice .
Little is known about FAM20C , a new molecule . In vitro studies have shown that it promotes the differentiation and mineralization processes of undifferentiated mesenchymal cells and odontoblasts [5] , whereas human genetic studies suggested that FAM20C might be a down-regulator ( inhibitor ) of bone formation and/or mineralization [6] . To answer critical questions regarding the biological roles of FAM20C , we generated Fam20c conditional knockout mice , in which exons 6–9 ( majority of the conserved CCD region ) were ablated . It is worth noting that most of the mutations identified in the patients with lethal osteosclerotic bone dysplasia were in exons 6–9 [6] . The Fam20c conditional knockout mice developed hypophosphatemic rickets but not osteosclerosis . We believe that the abnormalities in the Fam20c cKO mice resulted from the loss-of-function and were not due to the gain-of-function for this protein . This belief is based on the following observations: 1 ) deleting exons 6–9 ( majority of the CCD ) in the Fam20c cKO mice was most likely to inactivate this molecule; 2 ) the inheritance of the phenotypic changes in Fam20c cKO mice occurred in an autosomal recessive trait , while the gain-of-function is usually inherited in an autosomal dominant manner; 3 ) transgenic mice overexpressing the Fam20c transgene were normal; 4 ) the overexpression of the Fam20c transgene fully rescued the phenotypic changes in the Fam20c cKO mice; and 5 ) recombinant FAM20C promoted the differentiation and mineralization of MC3T3-E1 cells in a dose-dependent manner . These data combined with a significant downregulation of osteoblast differentiation markers in cKO mice suggest that FAM20C is essential to the differentiation of mineralizing cells and promotes the formation and mineralization of hard tissues , and thus , inactivation of this molecule leads to differentiation failure of the cells forming these tissues . Additionally , the Fam20c cKO mice developed hypophosphatemia with a remarkable elevation of the serum FGF23 level . We believe that a combination of cell differentiation failure and hypophosphatemia caused by the increase of serum FGF23 led to the skeletal defects in the Fam20c conditional knockout mice . In 2007 , Simpson et al . reported that the mutations of human FAM20C are associated with an osteosclerotic phenotype in some patients [6] . In a later study by the same group [8] , Simpson et al . identified FAM20C mutations in two patients whom they believed were suffering from a “different type of Raine Syndrome”; these two patients did not show a generalized increase in bone density , with one case showed “manifestations consistent with a diagnosis of hypophosphatemic rickets” , as the authors stated . The osteosclerotic phenotype in some patients with FAM20C mutations appears opposite to that observed in the Fam20c-deficient mice . These contradictory results raise questions of whether different domains/fragments of FAM20C protein have different functions or if their functions are different between humans and mice . In previous reports , the human FAM20C mutations in the osteosclerotic patients include point missense mutations and “splicing” mutations [6] , [8] . The point mutations were located in different regions of the gene , including the region encoding the N-terminal portion of the protein and that corresponding to the C-terminal part . The 1309G→A mutation ( D437N , in exon 7 ) observed in the “hypophosphatemic rickets”-like patient ( Case 1 in [8] ) is located between the two missense mutations 1121T→G ( L374R , in exon 6 ) and 1603C→T ( R535W , in exon 10 ) and was very close to a splicing mutation C1322-2A→G ( in intron 7 ) . The latter three mutations initially reported by Simpson et al . were associated with a generalized hypermineralization in the patients [6] , while the former one was associated with “hypophosphatemic rickets” [8] . The secreted form of mouse FAM20C contains 553 amino acid residues ( excluding a putative 26-amino acid signal peptide ) , and its calculated molecular mass is approximately 63 kDa . In our previous study [7] , Western immunoblotting analyses of the culture medium from HEK-293 cells transfected with a pMES construct containing full-length mouse FAM20C cDNA demonstrated a single protein band at approximately 65 kDa , consistent with the expected mass of full-length mouse FAM20C . We did not observe any lower molecular weight protein bands that could be recognized by the anti-FAM20C antibodies . Similar results were documented in the analyses of the mouse C3H10T1/2 cells and MC3T3-E1 cells transfected with the FLAG-tagged FAM20C [5] . These observations indicate that mouse FAM20C may not be proteolytically processed into fragments . Taken together , these human and mouse data do not support the contention that different domains or fragments of FAM20C may perform different functions . In this study , the lentiviral shRNA-mediated knockdown of FAM20C in the human mesenchymal stem cells and human osteoblasts led to a remarkable downregulation of DMP1 , along with a significant upregulation of FGF23 ( Figure 9B , 9C , 9E , and 9F ) . The findings in the human cells are consistent with the results in the shRNA-knockdown of FAM20C in mouse MC3T3-E1 cells ( Figure 9A and 9D ) and with the observations in the Fam20c conditional knockout bone ( Figure 8 , Table 2 ) ; these results indicate that FAM20C is likely to function similarly in humans and mice . Clearly , more studies are warranted to further clarify the discrepancy between the human and mouse data . As a growth factor , FGF23 principally functions as a phosphaturic hormone via binding to the Klotho/FGF receptor ( FGFR ) complexes in the kidney [30] , [31] . The binding of FGF23 to FGFR accelerates phosphate excretion into the urine , thereby inducing a negative phosphate balance , which helps maintain the serum phosphate levels in the normal range under physiological conditions ( Figure 12 ) . Elevation of the FGF23 plasma level is known to lead to renal phosphate-wasting and hypophosphatemia [11]–[15] , [28] , [32]–[35] . The main sources of FGF23 are the osteoblasts and osteocytes in the skeleton [11] , [28] , [29] , and a number of studies have shown that inactivating mutations in certain molecules expressed by these bone cells increase the plasma level of FGF23 , which leads to hereditary hypophosphatemic rickets [10] , [11] , [13] , [14] . Inactivating mutations in the phosphate-regulating gene with homologies to endopeptidases on the X chromosome ( PHEX ) cause X-linked hypophosphatemic rickets ( OMIM 307800 ) [10] , and loss of DMP1 activity results in autosomal-recessive hypophosphatemic rickets ( OMIM 241520 ) [11] , [13] . The phenotypic changes in the Fam20c-conditional knockout mice share many similarities ( hypomineralization , elevation of FGF23 , hypophosphatemia ) with those observed in the PHEX- and DMP1-deficient subjects . As in the cases of PHEX- and DMP1-deficiency , FGF23 was overexpressed in the bones of the Fam20c cKO mice ( Figure 8K and 8L , Table 2 ) . The overproduction of FGF23 by the bone cells is likely to be responsible for the elevation of this protein in the serum . Mutations in four genes , FGF23 itself , PHEX , DMP1 and ENPP1 , have been reported to remarkably increase the plasma levels of FGF23 , leading to hereditary hypophosphatemic rickets [9]–[11] , [13]–[15] . Dmp1- , Phex- and Fam20c-deficient mice shared similarities in osteomalacia , hypophosphatemia and the remarkable elevation of FGF23 in the circulation and skeleton . Interestingly , an alteration of Fam20c expression was not observed in the Dmp1 KO mice or Hyp mice , while remarkable downregulation of Dmp1 ( but not Phex ) was observed in the Fam20c cKO mice . An up-regulation of Dmp1 was observed in MC3T3-E1 cells treated with recombinant FAM20C . On the other hand , a remarkable down-regulation of Dmp1 was seen in human and mouse osteogenic cell lines treated with FAM20C-shRNA . These findings , along with the similarities of skeletal and serum changes between the Fam20c-deficient and Dmp1-deficient mice , raise the question of whether FAM20C regulates DMP1 ( Figure 12 ) . Clearly , further studies are warranted to answer this question . Additionally , there are still a pool of patients with hereditary hypophosphatemic rickets whose etiology is unknown [35] , [36] , and our discovery that the inactivation of Fam20c in mice results in hypophosphatemic rickets necessitates a consideration of screening FAM20C in such patients . High level FGF23 reduces the expression of renal vitamin D 1α-hydroxylase and increases the expression of the catabolic 25-hydroxyvitamin D 24-hydroxylase , thus leading to decreased levels of 1 , 25 ( OH ) 2D3 in the serum [15] , [37] , [38] . In the 18-day-old Fam20c cKO mice , the 1 , 25 ( OH ) 2D3 level was significantly lower , while in the 42-day-old cKO mice the 1 , 25 ( OH ) 2D3 level managed to return to the normal range ( or a not significantly higher level ) . Similar shifts in the serum 1 , 25 ( OH ) 2D3 level with aging have been observed in other hypophosphatemic models such as the Fgf23 transgenic mice , Hyp mice and Dmp1-KO mice that have high levels of circulating FGF23 [11] , [12] , [39] . While elevation of serum FGF23 reduces the expression of the renal 1α-hydroxylase , hypophosphatemia is normally a stimulator for renal 1α-hydroxylase expression to increase circulating 1 , 25 ( OH ) 2D3 [40]; the stimulating effect of hypophosphatemia on the 1α-hydroxylase expression has been well illustrated in the NaPi2a knockout mice ( with lower phosphate and lower FGF23 levels in the serum ) , in which the serum 1 , 25 ( OH ) 2D3 level is elevated due to the increased 1α-hydroxylase expression stimulated by hypophosphatemia [41] . The Fam20c-cKO mice displayed a decreased level of renal 1α-hydroxylase in the presence of hypophosphatemia , indicating that in these mutant mice , the negative modulation of FGF23 on the expression of the 1α-hydroxylase may outweigh the stimulating effect of hypophosphatemia on the 1α-hydroxylase expression . In comparison with the Fgf23 transgenic mice , Dmp1-KO mice and Hyp mice , a more remarkable upregulation of 24-hydroxylase was observed in the kidney of the Fam20c-cKO mice , which may be due to the fact that Fam20c-cKO mice have a higher serum FGF23 level than the former three [11] , [12] , [33] . A higher level of 24-hydroxylase in the 18-day-old cKO mice than that in the 42-day-old Fam20c cKO mice ( ∼30-fold versus ∼7-fold elevation ) may be responsible for the significantly lower serum 1 , 25 ( OH ) 2D3 level in the younger animals . In the older Fam20c cKO mice , a significant reduction of FGF23 co-receptor Klotho , along with a relatively lower serum FGF23 level than that in the younger cKO mice , may attenuate the FGF23-elevation effects on circulating 1 , 25 ( OH ) 2D3 and thus may help maintain a relatively normal serum 1 , 25 ( OH ) 2D3 level in the older cKO mice . More likely , the lower serum level of 1 , 25 ( OH ) 2D3 in the younger Fam20c KO mice may have triggered the overproduction of PTH , as in the cases of Fgf23 transgenic mice and Hyp mice . The secondary hyperparathyroidism may play a critical role for reversing the 1 , 25 ( OH ) 2D3 level to the normal range in the older Fam20c cKO mice [42] , [43] . In addition , the elevated PTH may synergize with the high level of serum FGF23 to increase renal phosphate excretion by reducing the expression of NaPi2a in the proximal tubules [38] , [44]; a significant reduction of NaPi2a was observed in the kidney of the Fam20c cKO mice . In the end , 1 , 25 ( OH ) 2D3 may maintain a relatively normal level in the older Fam20c cKO mice at the expense of a significant phosphorus wasting . The skeletal defects of the PHEX- and DMP1-deficient subjects are believed to be due to the combined effects of two factors: 1 ) the intrinsic defects of the PHEX- and DMP1-deficient cells that prevent them from forming and mineralizing ECM properly and 2 ) hypophosphatemia [11] , [33] , [45] , [46] . The Fam20c-deficient cells responsible for forming the mineralized tissues appeared immature and showed altered expression levels for molecules associated with cell differentiation . While hypophosphatemia in the Fam20c conditional knockout mice can be attributed to the overproduction of FGF23 in the abnormal skeleton , the direct cause of cell differentiation failure may be complicated . As stated above , the defects in the mineralized tissues of the Fam20c conditional knockout mice could be the combined results of cell differentiation failure and hypophosphatemia . Although the way FAM20C regulates cell differentiation has not yet been defined in this study , our data suggest that FAM20C may regulate the differentiation and function of the mineralizing cells by participating in certain signaling pathways . Several lines of evidence suggested that FAM20C might be associated with the canonical Wnt signaling pathway . The Wnt canonical pathway inhibitors , secreted frizzled related protein 1 ( Sfrp1 ) and Sfrp3 , were upregulated in the Fam20c-deficient mice . Accordingly , the downstream target genes of Wnt pathway , leucine-rich repeat-containing G protein-coupled receptor 5 ( Lgr5 ) and lymphoid enhancer-binding factor 1 ( Lef1 ) were significantly downregulated in the Fam20c-deficient bone . Lgr5 is a stem cell marker and a Wnt pathway regulator which has been identified in multiple tissues including bone marrow cells [23] , [24] . Lef1 is a Wnt-responsive transcription factor that associates with β-catenin and has been documented to increase osteoblast activity and trabecular bone mass [25] . However , not all of the findings support the postulation that FAM20C is a participant in the Wnt signaling pathway . For example , the level of axis inhibition protein 2 ( Axin2 ) , a putative Wnt downstream target gene , was unchanged in the Fam20c-deficient bone . In addition to the molecules in the Wnt signaling pathway , Follistatin ( Fst ) , a potent inhibitor of Activin and the TGF-β pathway , was significantly upregulated in the bones of 3-week and 6-week-old Fam20c-deficient mice . Fst has been reported to inhibit ameloblast and osteoblast differentiation [19] , [20] , suggesting a possible association between FAM20C and the TGF-β pathway . Type II and Type X collagen were downregulated in the growth plates of the Fam20c conditional knockout mice . The downregulation of these genes may arise from the intrinsic defects of chondrocytes or occur as a systematic consequence of hypophosphatemia . It has been well documented that hypophosphatemia significantly decreases programmed cell death in growth plates by impairing caspase-mediated apoptosis of hypertrophic chondrocytes [17] , [18] . In conclusion , the results in this investigation have demonstrated the crucial role of FAM20C in osteogenesis . Our findings indicate that FAM20C is essential to the differentiation of osteoblasts/osteocytes and is involved in the regulation of phosphate homeostasis via the mediation of FGF23 .
All animal procedures were performed in accordance with the NIH Guide for the Care and Use of Laboratory Animals and approved by the Institutional Animal Care and Use Committee of Texas A&M Health Science Center , Baylor College of Dentistry ( Dallas , TX , USA ) . To generate the Fam20c conditional knockout mice , a 2 . 2 kb targeting fragment spanning exons 6∼9 of mouse Fam20c was produced by PCR using the genomic DNA of WW6 ES cells as a template ( forward primer: 5′-CTCTCGGGTGAGGCTGTAAG-3′; reverse primer: 5′-AGATCTCTTAGGGAAGAGGGGTCAGG-3′ ) . The fragment was subcloned into a floxed BamHI site upstream of an Frt flanked mcl-Neo cassette in the conditional targeting vector pFlox-Frt-Neo [47] . A 2 . 5 kb 5′ homologous arm was generated by PCR ( forward primer: 5′-CTCGAGTGGGTGTGTCAGGAATCGTA-3′; reverse primer: 5′-CTCGAGACCCGAGAGCAACCACATAC-3′ ) , and subcloned into the XhoI site of the targeting vector . A 4 . 2 kb 3′ homologous arm with EagI sites at both ends was generated by PCR ( forward: 5′-TCGGCCGTTGGACATAGGCTCCCAAAG-3′; reverse: 5′-TGTGCAGGATTGAGAACCAG-3′ ) , and subcloned into the NotI site of the targeting vector . Finally , a negative selection PGK-DTA ( diphtheria toxin A ) cassette was subcloned into the NotI site downstream of the 3′ homologous arm in the targeting vector ( Figure 1A ) . The final targeting construct was linearized with SacII and electroporated into W4 ES cells ( Transgenic Mouse Core Facilities , University at Albany , Rensselaer , NY , USA ) . Clones were picked after G418 selection . Genomic DNA was extracted from the ES cells in duplicate plates , and PCR analyses were performed to screen the targeted clones ( 5′ screen primers: 5′S-F: 5′-TTTCTGTCCTAGGTAAGGGTGAAG-3′ , 5′S-R: 5′-ACTGCTCGATGAAGTTCCTATTCT-3′; 3′ screen primers: 3′S-F: 5′-TGTTCGGATCGAAGTTCCTATACT-3′ , 3′S-R: 5′-ACAGCTTCTTGAATTGGGATAAAG-3′ ) ( Figure 1A and 1B ) . The 5′ and 3′ screening PCR products were sequenced to confirm the correct targeting and the presence of 5′ and 3′ loxP sites . Two correctly targeted ES cell clones ( Clone 286 and 297 , Figure 1B ) were identified . The neo cassette was removed from the targeted ES clones by transient transfection of pCAGGS-flpE-puro vector ( Addgene plasmid 20733 ) [48] . The removal of the neo cassette was confirmed by PCR ( forward: 5′-GCATCTGCAGACCGAGCCCA-3′ , reverse: 5′-CCCCCTGTCCTGAGGGCTGA-3′ ) . Random integration of pCAGGS-flpE-puro DNA into ES cell genome was excluded by PCR analyses ( forward: 5′-GCATGGCCGAGTTGAGCGGT-3′ , reverse: 5′-GGTGACGGTGAAGCCGAGCC-3′ ) . The targeted ES clones recovered from the master plates were injected into the blastocysts of C57BL/6 mice in the Transgenic Core Facility of the University of Texas Southwestern Medical Center at Dallas . Male chimeras were crossbred with C57BL/6 females to produce F1 agouti offspring . The floxed alleles of F1 agouti mice were genotyped by PCR analyses ( see below ) . To generate Fam20c conditional knockout mice , the F1 heterozygous mice ( Fam20cflox/+ ) were first crossbred with Sox2 promoter-driven Cre transgenic mice ( Jackson Laboratory ) that express the Cre recombinase transgene in the epiblasts at E6 . 5; this breeding gave rise to Sox2-Cre-Fam20cΔ/+ mice in which exons 6∼9 were removed from one allele of the Fam20c gene . The Sox2-Cre-Fam20cΔ/+ mice were inbred to produce Fam20c conditional knockout ( cKO ) mice designated as “Sox2-Cre-Fam20cΔ/Δ mice” , in which exons 6∼9 were removed from both alleles of the Fam20c gene . Genotypes were determined by PCR analyses using genomic DNA extracted from the mouse tails . The floxed allele was distinguished from the wild type ( WT ) allele by PCR analyses using a mixture of three primers ( a: 5′-TCCAGCTTGCTAGGGCTCTGACC-3′ , b: 5′-CTATGTCCAACGGCCGCAGCTT-3′ , and c: 5′-GTCCTGAGGGCTGACCCAAGACTA-3′ ) ( Figure 1A and 1C ) . The null allele ( i . e . , that with exons 6∼9 removed ) arising from Cre-loxP recombination events was detected by PCR using primers Rec-F: 5′-GTGGTCTCTGCCGCTGATGTACC-3′ and Rec-R: 5′-TTTGGGAGCCTATGTCCAACGGCC-3′ ( Figure 1A and 1C ) . Genotyping for the Cre transgene was determined by PCR using primers Cre-F: 5′-CCCGCAGAACCTGAAGATG-3′ and Cre-R: 5′-GACCCGGCAAAACAGGTAG-3′ . We also used the 3 . 6 kb Col 1a1 promoter-Cre transgenic mice ( The Jackson Laboratory ) to generate “Col1a1-Cre-Fam20cΔ/Δ mice” , in which FAM20C was inactivated in tissues expressing type I collagen . We first crossbred the Fam20cflox/+ mice with 3 . 6 kb Col 1a1 -Cre mice , and then inbred the offspring of the Col1a1-Cre-Fam20cΔ/+ mice to get Col1a1-Cre-Fam20cΔ/Δ mice . To generate Fam20c conditional transgenic mice , the coding sequence of mouse Fam20c cDNA was subcloned into a bicistronic pMES vector [49] downstream to a chicken β-actin promoter and upstream to an IRES-EGFP cassette as previously described [7] . A floxed STOP cassette was inserted between the β-actin promoter and the Fam20c sequence to block the transcription of the transgene . The Fam20c transgene can be activated only after the floxed STOP cassette is removed by Cre recombinase [7] . This Fam20c conditional transgenic construct was linearized and injected into the pronuclei of fertilized eggs from the C57BL/6 mice in the Transgenic Core Facility of the University of Texas Southwestern Medical Center at Dallas . Fifteen lines of transgenic mice with the conditional Fam20c transgene were identified by PCR genotyping using primers located in the exogenous EGFP sequences ( GFP-forward: 5′-ACGTAAACGGCCACAAGTTC-3′ and GFP reverse: 5′-TGCTCAGGTAGTGGTTGTCG -3′ ) . The mice carrying the conditional transgene were crossbred with the Sox2-Cre transgenic mice to remove the floxed STOP cassette between the chicken β-actin promoter and the Fam20c cDNA , thereby allowing the Fam20c transgene to be transcribed . These conditional transgenic mice ( cTg mice ) were genotyped using the aforementioned GFP primers and Cre primers . The expression level of the Fam20c transgene in the long bones of each line was evaluated by quantitative real-time PCR . Three lines with the transgene expression levels of 4∼8 folds over that of the WT mice were further analyzed . To generate mice expressing the transgene in the Fam20c conditional knockout background , we crossbred Sox2-Cre-Fam20cΔ/+ mice with cTg mice expressing the highest level of the trangene to obtain Sox2-Cre–Fam20cΔ/+-cTg mice , which were then inbred to produce mice expressing the transgene in the Fam20c conditional knockout background ( designated “Sox2-Cre-Fam20cΔ/Δ-cTg mice” ) . PCR analyses with primers used in the identification of the Fam20c conditional knockout mice and cTg mice were employed in the genotyping of the Sox2-Cre-Fam20cΔ/Δ-cTg mice . All mice in this study were fed with Teklad 6% fat mouse/rat diet ( Harlan , IN ) and some of the chow contents are as follows: Calcium 2 . 4% , phosphorus 1 . 5% , vitamin D 3 . 0 IU/g . Femurs from the Fam20c conditional knockout mice , Sox2-Cre-Fam20cΔ/Δ-cTg mice and the WT littermates were dissected , and total RNA was extracted using an Rneasy Mini Kit ( Qiagen ) according to the manufacturer's instructions . The total RNAs were converted into cDNAs using a Reverse Transcription Kit ( Qiagen ) . RT-PCR was performed to examine the lack of Fam20c mRNA in the cKO mice using two sets of primers: Set 1-F: 5′-TGCGGAGATCGCTGCCTTCC-3′ , Set 1-R: 5′-GCCACTGTCGTAGGGTGGCG-3′; Set 2-F: 5′-GAGAGCAGGAGACGCCGCCT-3′ , and Set 2-R: 5′-CCACCACACTGCTCAGCCCG -3′ ( Figure 2A ) . One-week-old Sox2-Cre-Fam20cΔ/Δ mice and WT littermates were skinned , eviscerated and fixed in 95% ethanol . Alizarin Red/Alcian Blue staining of the skeletons was performed to visualize the skeleton and the overall mineralization levels , as described previously [50] . The narcotized mice or the dissected jaws and long bones from hind legs were analyzed using X-ray radiography ( Faxitron MX-20DC12 ) . Micro-computed tomography ( Micro-CT ) analyses were performed using a Scanco micro-CT35 imaging system ( Scanco Medical ) with a medium-resolution scan ( 7 . 0 µm slice increment ) on the dissected tissues , as previously reported [51] . The images were reconstructed with the EVS Beam software using a global threshold at 240 Hounsfield units . Tibia and jaw tissues dissected from the mice were fixed with 4% paraformaldehyde in 0 . 1% diethyl pyrocarbonate ( DEPC ) -treated PBS solution at 4°C overnight and then were decalcified in 0 . 1% DEPC-treated 15% EDTA ( pH 7 . 4 ) at 4°C for 8 days . The tissues were processed for paraffin embedding , and serial 5 µm sections were prepared for histological analyses . H&E staining was performed as previously described [17] . BrdU was administrated to 3-week-old Sox2-Cre-Fam20cΔ/Δ mice and WT littermates at a dosage of 1 ml per 100 g body weight by intraperitoneal ( i . p . ) injection according to the manufacturer's instructions ( Invitrogen ) . Two hr after the injection , the mice were sacrificed . Tibias were dissected and processed for paraffin embedding , and 5 µm sections were prepared for BrdU detection using a Zymed BrdU staining kit ( Invitrogen ) following the manufacturer's instructions . Apoptosis in growth plates was examined by TUNEL assay using the ApopTag Plus Fluorescein In Situ Apoptosis Detection Kit ( Millipore ) according to the manufacturer's instructions . Six serial sections from each of six individual samples of Sox2-Cre-Fam20cΔ/Δ mice and WT littermates were counted , and the data were analyzed statistically . The IHC experiments were carried out using an ABC kit and a DAB kit ( Vector Laboratories ) according to the manufacturer's instructions . A polyclonal C-terminal anti-FAM20C antibody was used at a concentration of 1 µg IgG/ml for the IHC experiments , as previously described [7] . A monoclonal FGF23 antibody ( Cell Essentials ) was used at a dilution of 1∶400 following the manufacturer's instruction . A polyclonal biglycan antibody ( LF-159 ) was kindly provided by Dr . Larry Fisher ( NIDCR , National Institutes of Health ) [52] . Methyl green was used for counterstaining . A 380 bp fragment from the region of exons 6–9 of Fam20c cDNA was obtained by PCR using forward primer 5′- CCGAGCATGCCCTGTGTGGG -3′ and reverse primer 5′- TGCAGCACTGATGAAGAGGAGCG -3′ . The PCR product was subcloned into the pCRII-TOPO vector ( Invitrogen ) and then linearized with EcoRV to synthesize the antisense RNA probes using the Sp6 RNA polymerase or with HindIII to synthesize the sense RNA probes using the T7 RNA polymerase . The constructs used to generate RNA probes for DMP1 , osteocalcin , collagen type I , collagen type II , and collagen type X were provided by the laboratory of Dr . Jian Q . Feng . The constructs were linearized and labeled with digoxigenin ( DIG ) using a RNA Labeling Kit ( Roche , Indianapolis , IN ) as previously described [7] . DIG-labeled RNA probes were detected by an enzyme-linked immunoassay with a specific anti-DIG-AP antibody conjugate ( Roche ) and an improved substrate ( Vector Laboratories ) , which produces a red color for positive signals , according to the manufacturer's instructions . Methyl green was used for counterstaining . Tibias dissected from 6-week-old mice were fixed in 4% paraformaldehyde overnight . The specimens were dehydrated through a graded series of ethanol ( 70–100% ) and embedded in methylmethacrylate ( MMA ) without prior decalcification , as previously described [53] . Ten µm sections were prepared for Goldner staining and double-labeling fluorescent analysis . Double fluorescence labeling was performed as previously described [54] . Briefly , calcein ( 5 mg/kg i . p . ; Sigma-Aldrich ) was administered to the 5-week-old mice , followed by injection of an Alizarin Red label ( 20 mg/kg i . p . ; Sigma-Aldrich ) 7 days later . The mice were sacrificed 48 hr after the injection of the second label and the tibias were embedded in MMA; 10 µm sections were then prepared . The unstained sections were viewed under epifluorescent illumination using a Nikon E800 microscope , interfaced with Osteomeasure histomorphometry software ( version 4 . 1 , Atlanta , GA ) . The mean distance between the two fluorescent labels was determined and divided by the number of days between labels to calculate the mineral deposition rate ( µm/day ) . Ten µm undecalcified sagittal sections from the tibias were stained using Goldner-Masson trichrome assay , as previously described [55] . The cortical bone areas in the midshaft were photographed using a Nikon microscope at 10× with Bioquant OSTEO v . 7 . 20 . 10 ( R&M Biometrics ) software . Unmineralized osteoid stains red , and mineralized bone stains green/blue . For resin-casted osteocyte lacunocanalicular SEM , the surface of the MMA embedded tibia was polished , acid-etched with 37% phosphoric acid for 2–10 s , washed with 5% sodium hypochlorite for 5 min and then coated with gold and palladium and examined by FEI/Philips XL30 Field emission environmental SEM . Backscattered SEM was performed as we previously described [56] Total RNAs were isolated from the calvaria bones and decapsulated kidneys of 3-week-old mice and cultured cells . The kits for RNA extraction and reverse transcription were the same as in the RT-PCR experiments . Quantitative real-time PCR was performed on a Bio-Rad CFX96 system ( Bio-Rad ) using SYBR Green Master Mix ( Stratagene ) . The Ct values were normalized to the reference gene 18s rRNA ( SABiosciences ) , and then expressed as fold changes compared with the experimental controls . The primers for human 18s rRNA , mouse 18s rRNA , mouse Fam20c ( NM_030565 ) and human FGF23 ( NM_020638 ) were bought from SABiosciences . All other primers were synthesized by Integrated DNA Technologies ( Table S1 ) . Microarray analyses were performed in the Microarray Core Facility of University of Texas Southwestern Medical Center at Dallas , using total RNA extracted from the calvaria of 3-week-old mice . GeneChip Mouse Genome 430 2 . 0 Array ( Affymetrix ) was employed in the microarray analyses , following the manufacturer's instructions . Data analyses were performed using GeneSpring software ( Agilent Technologies ) . The microarray results were uploaded to MAGE-TAB ArrayExpress database ( accession number E-MTAB-772 ) . Serum phosphorus was measured using the phosphomolybdate-ascorbic acid method , as previously described [57] . Serum calcium was measured using a colorimetric calcium kit ( Stanbio Laboratory ) . The serum FGF23 and PTH levels were measured using a full-length FGF23 ELISA kit ( Kainos Laboratories ) and a mouse intact PTH ELISA kit ( Immutopics ) . Serum 1 , 25 ( OH ) 2D3 was measured using a 1 , 25 Dihydroxy Vitamin D EIA Kit ( Immunodiagnostic Systems ) . Blood urea nitrogen ( BUN ) was measured using a BUN Reagent Kit ( BQ Kits ) . The recombinant mouse FAM20C was expressed by insect cells using a Bac-to-Bac baculovirus expression system ( Invitrogen ) . Briefly , the N-terminal of mouse FAM20C ( signal peptide removed ) was fused with a baculovirus signal sequence gp67 ( envelope glycoprotein ) , 6xHis ( tag ) , SUMOstar ( Small Ubiquitin-like Modifier ) , and a TEV ( Tobacco Etch Virus ) cleavage site . The fusion gene was inserted into the pFastDual vector ( Invitrogen ) downstream of a polyhedrin promoter , and a GFP cDNA was inserted downstream of a P10 promoter serving as a baculovirus indicator . The construct was transformed into DH10Bac E . Coli cells ( Invitrogen ) , in which the fusion gene was introduced into BacMid via homologous recombination . The BacMid was extracted from DH10Bac E . Coli cells and transfected into Sf21 insect cells ( Invitrogen ) to produce the baculovirus . The insect cells infected with the baculovirus secreted the recombinant FAM20C into the SFX-insect cell culture medium ( Hyclone ) . After two rounds of scale-up , the cell culture medium was collected and subjected to a one-step Ni-NTA purification . Turbo-TEV ( Eton Bioscience ) was used to release mouse FAM20C from the fusion protein , and reverse Ni-NTA purification was performed to remove Turbo-TEV , His tag , and SUMOstar . Mouse MC3T3-E1 cells were grown in α-MEM medium ( Gibco ) supplemented with 10% fetal bovine serum ( Hyclone ) and antibiotics ( Gibco ) ; human Saos-2 cells ( ATCC , osteoblasts from human osteosarcoma ) were grown in McCoy's 5a medium ( Gibco ) supplemented with 15% fetal bovine serum; human mesenchymal stem cells ( hMSC ) ( Lonza ) were maintained in MSCGM BulletKit medium ( Lonza ) . Cells were treated with recombinant FAM20C or the viruses when they reach 80% confluence . For the gain-of-function analyses , MC3T3-E1 cells were treated with mouse recombinant FAM20C at the concentrations of 200 ng/ml , 400 ng/ml and 800 ng/ml . For the loss-of-function studies , MC3T3-E1 , Saos-2 and hMSC cells were infected with the mouse or human shRNA-lentiviruses ( all from Santa Cruz ) containing a mixture of three target-specific shRNA sequences against the mouse or human FAM20C . A lentivirus expressing the scrambled shRNA ( with no specific target in the genome ) served as the control virus ( Santa Cruz ) . The infection rate was monitored by infecting these cells with another control virus ( Santa Cruz ) expressing the GFP indicator . After 1-week selection with 5–8 µg/ml concentrations of puromycin ( Santa Cruz ) , the control virus with GFP showed a nearly 100% infection rate . To induce osteogenic differentiation for the MC3T3 cells and Saos-2 cells , the culture medium was supplemented with 100 µg/ml ascorbic acid , 10 mM β-glycerophosphate and 30 nM dexamethasone; the osteogenic differentiation of human MSC was induced using the Osteogenic BulletKit ( Lonza ) following the manufacturer's instruction . Total RNA was extracted from the cells at different time points . For the gain-of-function analyses , RNA was extracted after 3 weeks of induction . For the “shRNA knockdown” analyses , RNA was extracted at 3 days after the lentiviral infection and before the osteogenic medium ( inducing the osteogenic differentiation ) addition , and at 1 , 2 and 3 weeks after the start of osteogenic induction . Real-time PCR was performed to evaluate the mRNA levels of the selected genes . The mineral deposition rate was determined by nodule formation and Alizarin red concentration in each well were measured using the Osteogenesis Quantitation kit ( Millipore ) for MC3T3-E1 cells treated with recombinant FAM20C ( the gain-of-function experiments ) . MC3T3-E1 cells and hMSC cells infected with the FAM20C-shRNA viruses became unhealthy ( showing lot of cell death ) after 4-week culture . The Saos-2 cells could not survive in the osteogenic medium for longer than 2 weeks . The data collected from the culture of cells at the “unhealthy stages” were discarded . The data are expressed as the mean ± SD of at least 6 individual determinations in all experiments unless otherwise indicated . We statistically evaluated the data employing ANOVA to test for any differences among the sample groups . When a difference was determined , 2-sample t tests were employed to evaluate all possible pairs of samples .
|
A recent study demonstrated that the inactivating mutations in the FAM20C gene were associated with lethal osteosclerotic bone dysplasia characterized by a generalized hardening of all bones; this observation implied an inhibitory role of FAM20C during bone formation . However , in vitro studies revealed a contradictory finding that FAM20C accelerated the differentiation of cells forming the mineralized tissues . Here we generated Fam20c conditional knockout ( cKO ) mice , in which the gene was inactivated either in all tissues or specifically in the mineralized tissues . We also generated recombinant FAM20C protein and Fam20c transgenic mice . The cKO mice did not mimic the human skeleton abnormalities of osteosclerotic bone dysplasia , but exhibited rickets ( softer bone ) along with a significant reduction of serum phosphate level and a remarkable elevation of serum FGF23 , a hormone known to promote phosphate wasting . A number of differentiation markers of the bone-forming cells were downregulated in the cKO mice . Recombinant FAM20C promoted the differentiation of mouse preosteoblasts . Introducing the Fam20c transgene did not lead to any abnormalities but rescued the bone defects of the cKO mice . Taken together , we conclude that FAM20C promotes the differentiation of osteoblast lineages and regulates phosphate homeostasis via the mediation of FGF23 .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"musculoskeletal",
"system",
"bone",
"physiology",
"genetics",
"biology",
"anatomy",
"and",
"physiology",
"genetics",
"and",
"genomics"
] |
2012
|
Inactivation of a Novel FGF23 Regulator, FAM20C, Leads to Hypophosphatemic Rickets in Mice
|
Infectious disease is a leading threat to public health , economic stability , and other key social structures . Efforts to mitigate these impacts depend on accurate and timely monitoring to measure the risk and progress of disease . Traditional , biologically-focused monitoring techniques are accurate but costly and slow; in response , new techniques based on social internet data , such as social media and search queries , are emerging . These efforts are promising , but important challenges in the areas of scientific peer review , breadth of diseases and countries , and forecasting hamper their operational usefulness . We examine a freely available , open data source for this use: access logs from the online encyclopedia Wikipedia . Using linear models , language as a proxy for location , and a systematic yet simple article selection procedure , we tested 14 location-disease combinations and demonstrate that these data feasibly support an approach that overcomes these challenges . Specifically , our proof-of-concept yields models with up to 0 . 92 , forecasting value up to the 28 days tested , and several pairs of models similar enough to suggest that transferring models from one location to another without re-training is feasible . Based on these preliminary results , we close with a research agenda designed to overcome these challenges and produce a disease monitoring and forecasting system that is significantly more effective , robust , and globally comprehensive than the current state of the art .
Infectious disease remains extremely costly in both human and economic terms . For example , the majority of global child mortality is due to conditions such as acute respiratory infection , measles , diarrhea , malaria , and HIV/AIDS [1] . Even in developed countries , infectious disease has great impact; for example , each influenza season costs the United States between 3 , 000 and 49 , 000 lives [2] and an average of $87 billion in reduced economic output [3] . Effective and timely disease surveillance — that is , detecting , characterizing , and quantifying the incidence of disease — is a critical component of prevention and mitigation strategies that can save lives , reduce suffering , and minimize impact . Traditionally , such monitoring takes the form of patient interviews and/or laboratory tests followed by a bureaucratic reporting chain; while generally considered accurate , this process is costly and introduces a significant lag between observation and reporting . These problems have motivated new surveillance techniques based upon internet data sources such as search queries and social media posts . Essentially , these methods use large-scale data mining techniques to identify health-related activity traces within the data streams , extract them , and transform them into some useful metric . The basic approach is to train a statistical estimation model against ground truth data , such as ministry of health disease incidence records , and then apply the model to generate estimates when the true data are not available , e . g . , when forecasting or when the true data have not yet been published . This has proven effective and has spawned operational systems such as Google Flu Trends ( http://www . google . org/flutrends/ ) . However , four key challenges remain before internet-based disease surveillance models can be reliably integrated into an decision-making toolkit: Our paper draws upon prior scholarly and practical work in three areas: traditional patient- and laboratory-based disease surveillance , Wikipedia-based measurement of the real world , and internet-based disease surveillance .
Access logs for all Wikipedia articles are available in summary form to anyone who wishes to use them . We used the complete logs available at http://dumps . wikimedia . org/other/pagecounts-raw/ . Web interfaces offering a limited view into the logs , such as http://stats . grok . se , are also available . These data are referred to using a variety of terms , including article views , article visits , pagecount files , page views , pageviews , page view logs , and request logs . These summary files contain , for each hour from December 9 , 2007 to present and updated in real time , a compressed text file listing the number of requests for every article in every language , except that articles with no requests are omitted . ( This request count differs from the true number of human views due to automated requests , proxies , pre-fetching , people not reading the article they loaded , and other factors . However , this commonly used proxy for human views is the best available . ) We analyzed data from March 7 , 2010 through February 1 , 2014 inclusive , a total of 1 , 428 days . This dataset contains roughly 34 , 000 data files totaling 2 . 7TB . 266 hours or 0 . 8% of the data are missing , with the largest gap being 85 hours . These missing data were treated as zero; because they were few , this has minimal effect on our analyses . We normalized these request counts by language . This yielded , for each article , a time series containing the number of requests for that article during each hour , expressed as a fraction of the hour's total requests for articles in the language . This normalization also compensates for periods of request undercounting , when up to 20% fewer requests were counted than served [96] . Finally , we transposed the data using Map-Reduce [97] to produce files from which the request count time series of any article can be retrieved efficiently . Our goal was to evaluate a broad selection of diseases in a variety of countries across the world , in order to test the global applicability and disease agnosticism of our proposed technique . For example , we sought diseases with diverse modes of transmission ( e . g . , airborne droplet , vector , sexual , and fecal-oral ) , biology ( virus , bacteria , protozoa ) , types of symptoms , length of incubation period , seasonality , and prevalence . Similarly , we sought locations in both the developed and developing world in various climates . Finally , we wanted to test each disease in multiple countries , to provide an opportunity for comparison . These comprehensive desiderata were tempered by the realities of data availability . First , we needed reliable data establishing incidence ground truth for specific diseases in specific countries and at high temporal granularity; such official data are frequently not available for locations and diseases of interest . We used official epidemiological reports available on websites of government public health agencies as well as the World Health Organization ( WHO ) . Second , we needed article access counts for specific countries . This information is not present in the Wikipedia article access logs ( i . e . , request counts are global totals ) . However , a proxy is sometimes available in that certain languages are mostly limited to one country of interest; for example , a strong majority of Thai speakers are in Thailand , and the only English-speaking country where plague appears is the United States . In contrast , Spanish is spoken all over the world and thus largely unsuitable for this purpose . Third , the language edition needs to have articles related to the disease of interest that are mature enough to evaluate and generate sufficient traffic to provide a reasonable signal . With these constraints in mind , we used our professional judgement to select diseases and countries . The resulting list of 14 disease-location contexts , which is designed to be informative rather than comprehensive , is enumerated in Table 1 . These incidence data take two basic forms: ( a ) tabular files such as spreadsheets mapping days , weeks , or months to new case counts or the total number of infected persons or ( b ) graphs presenting the same mapping . In the latter case , we used plot digitizing software ( Plot Digitizer , http://plotdigitizer . sourceforge . net ) to extract a tabular form . We then translated these diverse tabular forms to a consistent spreadsheet format , yielding for each disease-location context a time series of disease incidence ( these series are available in supplemental data S1 ) . The goal of our models is to create a linear mapping from the access counts of some set of Wikipedia articles to a scalar disease incidence for some disease-location context . To do so , a procedure for selecting these articles is needed; for the current proof-of-concept work , we used the following: This procedure has two potential complications . First , an article may not exist in the target language; in this case , we simply omit it . Second , Wikipedia contains null articles called redirects that merely point to another article , called the target of the redirect . These are created to catch synonyms or common misspellings of an article . For example , in English , the article “Flu” is a redirect to “Influenza” . When a user visits http://en . wikipedia . org/wiki/Flu , the content served by Wikipedia is actually that of the “Influenza” article; the server does not issue an HTTP 301 response nor require the reader to manually click through to the redirect target . This complicates our analysis because this arrangement causes the redirect itself ( “Flu” ) , not the target ( “Influenza” ) , to appear in the access log . While in principle we could sum redirect requests into the target article's total , reliably mapping redirects to targets is a non-trivial problem because this mapping changes over time , and in fact Wikipedia's history for redirect changes is not complete [98] . Therefore , we have elected to leave this issue for future work; this choice is supported by our observation below that when target and redirect are reversed , traffic to “Dengue fever” in Thai follows the target . If we encounter a redirect during the above procedure , we use the target article . The complete selection of articles is available in the supplementary data S1 . Our goal was to understand how well traffic for a simple selection of articles can nowcast and forecast disease incidence . Accordingly , we implemented the following procedure in Python to build and evaluate a model for each disease-location context . In order to test forecasting potential , we repeat the above with the article time series time-shifted from 28 days forward to 28 days backward in 1-day increments . For example , to build a 4-day forecasting model — that is , a model that estimates disease incidence 4 days in the future — we would shift the article time series later by 4 days so that article request counts for a given day are matched against disease incidence 4 days in the future . The choice of ±28 days for lag analysis is based upon our a priori hypothesis that these statistical models are likely effective for a few weeks of forecasting . We refer to models that estimate current ( i . e . , same-day ) disease incidence as nowcasting models and those that estimate past disease incidence as anti-forecasting models; for example , a model that estimates disease incidence 7 days ago is a 7-day anti-forecasting model . ( While useless at first glance , effective anti-forecasting models that give results sooner than official data can still reduce the lead time for action . Also , it is valuable for understanding the mechanism of internet-based models to know the temporal location of predictive information . ) We report r2 for each time-shifted multi-article model . Finally , to evaluate whether translating models from one location to another is feasible , we compute a metric rt for each pair of locations tested on the same disease . This meta-correlation is simply the Pearson's r computed between the correlation scores r of each article found in both languages; the intent is to give a gross notion of similarity between models computed for the same disease in two different languages . A value of 1 means that the two models are identical , 0 means they have no relationship , and -1 means they are opposite . We ignore articles found in only one language because the goal is to obtain a sense of feasibility: given favorable conditions , could one train a model in one location and apply it to another ? Table 2 illustrates an example .
Model and official data time series for selected successful contexts are illustrated in Figure 1 . The method's good performance on dengue and influenza is consistent with voluminous prior work on these diseases; this offers evidence for the feasibility of Wikipedia access as a data source . Success in the United States is somewhat surprising . Given the widespread use of English across the globe , we expected that language would be a poor location proxy for the United States . We speculate that the good influenza model performance is due to the high levels of internet use in United States , perhaps coupled with similar flu seasons in other Northern Hemisphere countries . Similarly , in addition to Brazil , Portuguese is spoken in Portugal and several other former colonies , yet problems again failed to arise . In this case , we suspect a different explanation: the simple absence of dengue from other Portuguese-speaking countries . The case of dengue in Brazil is further interesting because it highlights the noise inherent in this social data source , a property shared by many other internet data sources . That is , noise in the input articles is carried forward into the model's estimate . We speculate that this problem could be mitigated by building a model on a larger , more carefully selected set of articles rather than just 10 . Finally , we highlight tuberculosis in China as an example of a marginally successful model . Despite the apparently low r2 of 0 . 66 , we judged this model successful because it captured the high baseline disease level excellently and the three modest peaks well . However , it is not clear that the model provides useful information at the time scale analyzed . This result suggests that additional quantitative evaluation metrics may be needed , such as root mean squared error ( RMSE ) or a more complex analysis considering peaks , valleys , slope changes , and related properties . Forecasting and anti-forecasting performance of the four selected contexts is illustrated in Figure 2 . In the case of dengue and influenza , the models contain significant forecast value through the limit of our 28-day analysis , often with the maximally effective lag comprising a forecast . We offer three possible reasons for this . First , both diseases are seasonal , so readers may simply be interested in the syndrome for this reason; however , the fact that models were able to correctly estimate seasons of widely varying severity provides counterevidence for this theory . Second , readers may be interested due to indirect reasons such as news coverage . Prior work disagrees on the impact of such influences; for example , Dukic et al . found that adding news coverage to their methicillin-resistant Staphylococcus aureus ( MRSA ) model had a limited effect [58] , but recent Google Flu Trends failures appear to be caused in part by media activity [94] . Finally , both diseases have a relatively short incubation period ( influenza at 1–4 days and dengue at 3–14 ) ; soon-to-be-ill readers may be observing the illness of their infectors or those who are a small number of degrees removed . It is the third hypothesis that is most interesting for forecasting purposes , and evidence to distinguish among them might be obtained from studies using simulated media and internet data , as suggested by Culotta [82] . Tuberculosis in China is another story . In this case , the model's effectiveness is poorer as the forecast interval increases; we speculate that this is because seasonality is absent and the incubation period of 2–12 weeks is longer , diluting the effect of the above two mechanisms . Figure 3 illustrates the three contexts where the model was not effective because , we suspect , it was not able to discern meaningful patterns in the official data . These suggest a few patterns that models might have difficulty with: ( Note that we distinguish noisy official data from an unfavorable signal-to-noise ratio , which is discussed below . ) Both HIV/AIDS and tuberculosis infections progress quite slowly . A period of analysis longer than three years might reveal meaningful patterns that could be captured by this class of models . However , the social internet is young and turbulent; for example , even 3 years consumes most of the active life of some languages of Wikipedia . This complicates longitudinal analyses . In all three patterns , improvements such as non-linear models or better regression techniques could lead to better results , suggesting that this is a useful direction for future work . In particular , noise suppression techniques as well as models tuned for the expected variation in a particular disease may prove fruitful . Figure 4 illustrates the three contexts where we suspect the model failed due to a signal-to-noise ratio ( SNR ) in the Wikipedia data that was too low . That is , the number of Wikipedia accesses due to actual observations of infection is drowned out by accesses due to other causes . In the case of Ebola , there are relatively few direct observations ( a major outbreak has tens of cases ) , and the path to these observations becoming internet traces is hampered by poor connectivity in the sub-Saharan countries where the disease is active . On the other hand , the disease is one of general ongoing interest; in fact , one can observe on the graph a pattern of weekly variation ( higher in midweek , lower on the weekend ) , which is common in online activity . In combination , these yield a completely useless model . The United States has good internet connectivity , but plague has even lower incidence ( the peak on the graph is three cases ) and this disease is also popularly interesting , resulting in essentially the same effect . The cholera outbreak in Haiti differs in that the number of cases is quite large ( the peak of the graph is 4 , 200 cases in one day ) . However , internet connectivity in Haiti was already poor even before the earthquake , and the outbreak was a major world news story , increasing noise , so the signal was again lost . Table 3 summarizes the performance of our models in the 14 disease-location contexts tested . Of these , we classified 8 as successful , producing useful estimates for both nowcasting and forecasting , and 6 as unsuccessful . Performance roughly broke down along disease lines: all influenza and dengue models were successful , while two of the three tuberculosis models were , and cholera , ebola , HIV/AIDS , and plague proved unsuccessful . Given the relatively simple model building technique used , this suggests that our Wikipedia-based approach is sufficiently promising to explore in more detail . ( Another hypothesis is that model performance is related to popularity of the corresponding Wikipedia language edition . However , we found no relationship between r2 and either a language's total number of articles or total traffic . ) At a higher level , we posit that a successful estimation model based on Wikipedia access logs or other social internet data requires two key elements . First , it must be sensitive enough to capture the true variation in disease incidence data . Second , it must be sensitive enough to distinguish between activity traces due to health-related observations and those due to other causes . In both cases , further research on modeling techniques is likely to yield sensitivity improvements . In particular , a broader article selection procedure — for example , using big data methods to test all non-trivial article time series for correlation , as Ginsberg et al . did for search queries [36] — is likely to prove fruitful , as might a non-linear statistical mapping . Table 4 lists the transferability scores rt for each pair of countries tested on the same disease . Because this paper is concerned with establishing feasibility , we focus on the highest scores . These are encouraging: in the case of influenza , both Japan/Thailand and Thailand/United States are promising . That is , it seems plausible that careful source model selection and training techniques may yield useful models in contexts where no training data are available ( e . g . , official data are unavailable or unreliable ) . These early results suggest that research to quantitatively test methods for translating models from one disease-location context to another should be pursued .
Human activity on the internet leaves voluminous traces that contain real and useful evidence of disease dynamics . Above , we pose four challenges currently preventing these traces from informing operational disease surveillance activities , and we argue that Wikipedia data are one of the few social internet data sources that can meet all four challenges . Specifically: This preliminary study has several important limitations . These comprise an agenda for future research work: Wikipedia has its own data peculiarities that can also cause difficulty . For example , during preliminary exploration for this paper in early July 2013 , we used the inter-language link on the English article “Dengue fever” to locate the Thai version , “” ( roughly , “dengue hemorrhagic fever” ) ; article access logs indicated several hundred accesses per day for this article in the month of June 2013 . When we repeated the same process in March 2014 , the inter-language link led to a page with the same content , but a different title , “” ( roughly , “dengue fever” ) . As none of the authors are Thai speakers , and Google Translate renders both versions as “dengue fever” , we did not notice that the title of the Thai article had changed and were alarmed to discover that the article's traffic in June 2013 was essentially zero . The explanation is that before July 23 , 2013 , “” was a redirect to “”; on that day , the direction of the redirect was reversed , and almost all accesses moved over to the new redirect target over a period of a few days . That is , the article was the same all along , but the URL under which its accesses were recorded changed . Possible techniques for compensation include article selection procedures that exclude such articles or maintaining a time-aware redirect graph so that different aliases of the same article can be merged . Indeed , when we tried the latter approach by manually summing the two URLs' time series , it improved nowcast r2 from 0 . 55 to 0 . 65 . However , the first technique is likely to discard useful information , and the second may not be reliable because complete history for this type of article transformation is not available [98] . In general , ongoing , time-aware re-training of models will likely be helpful , and limitations of the compensation techniques can be evaluated with simulation studies that inject data problems . Finally , it is important to recognize the biases inherent in Wikipedia and other social internet data sources . Most importantly , the data strongly over-represent people and places with good internet access and technology skills; demographic biases such as age , gender , and education also play a role . These biases are sometimes quantified ( e . g . , with survey results ) and sometimes completely unknown . Again , simulation studies using synthetic internet data can quantify the impact and limitations of these biases . Despite these limitations , we have established the utility of Wikipedia access logs for global disease monitoring and forecasting , and we have outlined a plausible path to a reliable , scientifically sound , operational disease surveillance system . We look forward to collaborating with the scientific and technical community to make this vision a reality .
|
Even in developed countries , infectious disease has significant impact; for example , flu seasons in the United States take between 3 , 000 and 49 , 000 lives . Disease surveillance , traditionally based on patient visits to health providers and laboratory tests , can reduce these impacts . Motivated by cost and timeliness , surveillance methods based on internet data have recently emerged , but are not yet reliable for several reasons , including weak scientific peer review , breadth of diseases and countries covered , and underdeveloped forecasting capabilities . We argue that these challenges can be overcome by using a freely available data source: aggregated access logs from the online encyclopedia Wikipedia . Using simple statistical techniques , our proof-of-concept experiments suggest that these data are effective for predicting the present , as well as forecasting up to the 28-day limit of our tests . Our results also suggest that these models can be used even in places with no official data upon which to build models . In short , this paper establishes the utility of Wikipedia as a broadly effective data source for disease information , and we outline a path to a reliable , scientifically sound , operational , and global disease surveillance system that overcomes key gaps in existing traditional and internet-based techniques .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"public",
"and",
"occupational",
"health",
"infectious",
"diseases",
"computer",
"and",
"information",
"sciences",
"data",
"management",
"ecology",
"and",
"environmental",
"sciences",
"computer",
"modeling",
"nature-society",
"interactions",
"medicine",
"and",
"health",
"sciences",
"disease",
"ecology",
"data",
"acquisition",
"behavioral",
"and",
"social",
"aspects",
"of",
"health",
"global",
"health",
"computerized",
"simulations",
"geographic",
"information",
"systems",
"geoinformatics"
] |
2014
|
Global Disease Monitoring and Forecasting with Wikipedia
|
Face cleanliness is a core component of the SAFE ( Surgery , Antibiotics , Facial cleanliness , and Environmental improvements ) strategy for trachoma control . Understanding knowledge , attitudes , and behaviors related to face washing may be helpful for designing effective interventions for improving facial cleanliness . In April 2014 , a mixed methods study including focus groups and a quantitative cross-sectional study was conducted in the East Gojjam zone of the Amhara region of Ethiopia . Participants were asked about face washing practices , motivations for face washing , use of soap ( which may reduce bacterial load ) , and fly control strategies . Overall , both knowledge and reported practice of face washing was high . Participants reported they knew that washing their own face and their children’s faces daily was important for hygiene and infection control . Although participants reported high knowledge of the importance of soap for face washing , quantitative data revealed strong variations by community in the use of soap for face washing , ranging from 4 . 4% to 82 . 2% of households reporting using soap for face washing . Cost and forgetfulness were cited as barriers to the use of soap for face washing . Keeping flies from landing on children was a commonly cited motivator for regular face washing , as was trachoma prevention . Interventions aiming to improve facial cleanliness for trachoma prevention should focus on habit formation ( to address forgetfulness ) and address barriers to the use of soap , such as reducing cost . Interventions that focus solely on improving knowledge may not be effective for changing face-washing behaviors .
Trachoma is the leading cause of infectious blindness globally . [1–3] Caused by the bacterium Chlamydia trachomatis , trachoma is thought to be transmitted by direct contact from infected persons and clothing , as well as the moisture-seeking fly Musca sorbens . [4 , 5] Currently endemic in 53 countries[6] , trachoma is estimated to result in blindness or severe vision loss in more than 2 million people[1] , with the majority of cases found in sub-Saharan Africa . [1] Despite large reductions in the burden of trachoma in the past several decades[1] , trachoma remains an important cause of blindness primarily among individuals living in poor , predominantly rural areas . [6–9] The cornerstone of trachoma control is the SAFE ( Surgery , Antibiotics , Facial cleanliness , and Environmental improvements ) strategy . [6] Mass antibiotic distributions have been shown to be effective at reducing the prevalence of trachoma . [10 , 11] However , while antibiotics may lead to local control of trachoma , alone they may not be sufficient for trachoma elimination in places with hyperendemic infection . [11] Multiple observational studies have demonstrated an association between poor facial hygiene , including the presence of flies on a child’s face , and trachoma . [12–16] It is possible that improvements in hygiene , and especially facial hygiene , may alter the transmission dynamics of trachoma and create more favorable conditions for trachoma elimination . The use of soap for face washing has been shown to be associated with decreased risk of trachoma in some[16–19] but not all[20] studies . Soap may decrease the bacterial load on children’s faces , which could decrease the probability of transmission of trachoma . A recent meta-analysis of observational studies demonstrated that use of soap was associated with a lower prevalence of trachoma . [16] However , soap specifically for face washing is rarely included or advocated for in trachoma elimination campaigns . The association between poor facial hygiene and trachoma suggests that interventions to promote facial cleanliness may be helpful in reducing trachoma prevalence and ultimately achieving trachoma elimination . These interventions will benefit from understanding current knowledge , practices and beliefs related to face washing . Here , we analyze knowledge , beliefs , and practices related to face washing , and their relation to trachoma , in a mixed methods study in a trachoma-hyperendemic region of rural Ethiopia .
This study took place in a rural agrarian region in the Goncha Siso Enese woreda of East Gojjam , Amhara , Ethiopia . The communities in this study were participating in a series of cluster-randomized trials testing different mass drug administration strategies for trachoma elimination beginning in 2006 . Each community has approximately 275 residents . These communities received mass azithromycin distributions annually or biannually between 2006 and 2013 . [11] Methods for these trials are described in detail elsewhere . [11] At baseline , the prevalence of trachoma in children 1–10 years old was 48 . 5% and 15 . 5% in children 11 years and older . [11] For the present study , we selected five communities that were within a one-hour walk from the farthest place a four-wheel drive vehicle could reach . All households in each community included in this study were eligible to participate in the quantitative survey . Before and during the study , all communities continued to receive the prescribed government package of hygiene promotion activities . In this report , the five communities are labeled Community A , B , C , D , and E to protect anonymity of the communities . The quantitative and qualitative surveys were designed to gain an understanding of existing knowledge and behaviors in relation to face washing , and to identify gaps between knowledge and behaviors .
Universally , participants endorsed daily face washing of their children , typically in the morning , although many participants also indicated that more frequent washing would be beneficial for their children . Whereas there was universal endorsement of face washing among participants in quantitative and qualitative data , the reported usage of soap for face washing varied widely by community , from 4% in Community C to 82% in Community E ( Table 2 ) . To gain a deeper understanding of face washing behaviors , focus groups explicitly probed for reasons behind face washing . Participants in all communities mentioned fly control as a reason for face washing , especially for children . Participants noted both that face washing was necessary when flies were seen on children’s faces , but also that regular face washing prevented flies from landing on faces . Many participants cited trachoma prevention as a reason for face washing for fly control . The benefits of fly control were seen as beneficial for the general health and wellbeing of children , as well .
In this mixed methods study , we document high prevalence of reported daily face washing among a rural population in a hyperendemic area for trachoma in Ethiopia . However , despite face washing being a common practice , soap was less commonly used as part of face washing routines . Face washing is a key component of the SAFE strategy for trachoma prevention , and use of soap may improve the ability of face washing to prevent trachoma transmission . [17 , 18] The use of soap for face washing in this study varied widely by community . Previous studies have demonstrated clustering of active trachoma and trachoma infection at the household and village level . [21 , 22] Geographic clustering of trachoma is likely due to both increased probability of transmission in areas with higher trachoma prevalence as well as shared characteristics such as environmental or climate factors . [23 , 24] The results of the present study suggest that there may be shared behavioral characteristics within villages that may also contribute to geographic clustering . Although use of soap varied widely by community , focus group participants from all communities reported high levels of knowledge of the importance of soap . The focus groups suggested that economic barriers are important in limiting the regular use of soap for face washing , indicating economic interventions may be important for improving face washing with soap behaviors . In Community C , which had the lowest reported use of soap for face washing , focus group participants explained that while they knew the benefits of using soap for face washing , they were simply not in the habit of doing so . These results suggest that , in this community , individuals are further along the knowledge-attitude-behavior continuum in terms of behavior change . [25] As such , interventions promoting face washing with soap may be more effective if they focus on habit formation and practice rather than improving knowledge , as the community members already have knowledge of the benefits of using soap for face washing . There is conflicting evidence of the relationship between distance to water source and trachoma . [16 , 26] Theoretically , increased distance to water source may reduce face washing behaviors because of water security in households . It is possible , however , that the inclusion of other factors in multivariable models ( such as hygiene practices ) obscures the relationship in some studies . In this study , we found an association between shorter distance to water source and face washing of children in the household . In addition , we noted a dose-response relationship , with households that reported longer times for water collecting less frequently reporting face-washing children . Similarly , households in which the survey respondent reported that the household had an adequate supply of water for their needs more often reported face washing of all children . The results indicate that face-washing behaviors may be facilitated by access to adequate water supply . Future work should consider the role of distance to water source on hygiene behaviors , as it is plausible that households with greater access to water have differential hygiene behaviors . Overall in this sample , participants had high health literacy related to trachoma . Focus group participants generally believed that face washing would help prevent trachoma . The communities in which this study was conducted were in a region that is hyperendemic for trachoma , and as such received mass drug administrations for trachoma and participated in trachoma trials for the eight years prior to the present study . [27–29] High levels of trachoma knowledge may be related to participants’ involvement in these studies . This knowledge of trachoma and the fact that it could cause blindness in their children was likely a motivator for face washing behavior , explaining high coverage of daily face washing in this population . There is also the possibility that participants noted trachoma prevention as a motivation for face-washing because they felt it was the ‘correct’ answer and not because it was a true motivation . These results may not be generalizable to areas that are hyperendemic for trachoma that have not experienced this intensity of trachoma programming . Future work may be needed in trachoma study or program-naive populations to determine if face-washing predictors and behaviors differ . The results of this study must be considered in the context of several limitations . Face washing behaviors in both the focus group discussions and the quantitative survey were collected via self-report . Although face washing is not necessarily a stigmatized behavior , it is possible that individuals’ responses may have been influenced by social desirability bias , as participants may have responded in ways which they perceived to be “correct” . There may have been alternative explanations , for example soap getting into children’s eyes , that were not discussed because participants perceived it was an incorrect answer . We anticipate that any outcome misclassification arising due to social desirability bias would be non-differential with respect to various predictors , and as such would , on average , bias towards the null in our regression models . In addition , while participants discussed their knowledge and current behaviors related to face washing and using soap , responses may not necessarily reflect motivations , and may rather reflect rationalization or normative reasons . Importantly , as qualitative data were collected as focus group discussions , individual responses may have been influenced by the responses of other members in the focus group . These results therefore should not be interpreted on the individual level , but instead represent community-level knowledge and behaviors . It is possible that some individual behaviors were masked in group discussions if some individuals did not want to discuss behaviors that were outlying from the rest of the group . Future work with individual interviews may yield additional insights into face washing and other hygiene behaviors in this region . This study provides important insights into face washing knowledge , attitudes , and behaviors for intervention development in a trachoma hyperendemic region of rural Ethiopia . Overall knowledge of the benefits of face washing was high , and the use of soap for face washing varied widely by community . Water access was associated with reduced odds of all children in the household washing their faces , but was not discussed during focus group discussions . Barriers to face washing with soap included cost and forgetting to use soap . Interventions for face washing that include habit formation , which may help to address forgetfulness , and address structural barriers to accessing soap , like cost , may be important for increasing facial cleanliness and ultimately trachoma control in hyperendemic regions .
|
Facial cleanliness is a core component of the SAFE ( Surgery , Antibiotics , Facial cleanliness , and Environmental improvements ) strategy for trachoma control . We conducted a mixed methods study in a trachoma hyperendemic region of rural Ethiopia to better understand knowledge , attitudes , and behaviors related to face washing . Overall , knowledge of the benefits of face washing was high , and participants reported regularly engaging in face washing practices . However , the use of soap for face washing varied more between communities . Participants cited cost and forgetting to use soap as the primary barriers to using soap for face washing . Trachoma prevention , including keeping flies from landing on children’s faces , was a commonly-cited motivator for face washing discussed in focus groups . Given the near-universal knowledge of the benefits of face washing , interventions focused on changing face washing behavior for trachoma control should focus on habit formation and removal of barriers to the use of soap rather than simply educational interventions .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"medicine",
"and",
"health",
"sciences",
"face",
"chemical",
"compounds",
"engineering",
"and",
"technology",
"tropical",
"diseases",
"salts",
"geographical",
"locations",
"cell",
"phones",
"bacterial",
"diseases",
"eye",
"diseases",
"neglected",
"tropical",
"diseases",
"ethiopia",
"eyes",
"africa",
"public",
"and",
"occupational",
"health",
"infectious",
"diseases",
"soaps",
"behavior",
"chemistry",
"hygiene",
"head",
"communication",
"equipment",
"people",
"and",
"places",
"anatomy",
"equipment",
"ophthalmology",
"biology",
"and",
"life",
"sciences",
"ocular",
"system",
"physical",
"sciences",
"trachoma"
] |
2016
|
‘If an Eye Is Washed Properly, It Means It Would See Clearly’: A Mixed Methods Study of Face Washing Knowledge, Attitudes, and Behaviors in Rural Ethiopia
|
Somatic copy number variations ( CNVs ) play a crucial role in development of many human cancers . The broad availability of next-generation sequencing data has enabled the development of algorithms to computationally infer CNV profiles from a variety of data types including exome and targeted sequence data; currently the most prevalent types of cancer genomics data . However , systemic evaluation and comparison of these tools remains challenging due to a lack of ground truth reference sets . To address this need , we have developed Bamgineer , a tool written in Python to introduce user-defined haplotype-phased allele-specific copy number events into an existing Binary Alignment Mapping ( BAM ) file , with a focus on targeted and exome sequencing experiments . As input , this tool requires a read alignment file ( BAM format ) , lists of non-overlapping genome coordinates for introduction of gains and losses ( bed file ) , and an optional file defining known haplotypes ( vcf format ) . To improve runtime performance , Bamgineer introduces the desired CNVs in parallel using queuing and parallel processing on a local machine or on a high-performance computing cluster . As proof-of-principle , we applied Bamgineer to a single high-coverage ( mean: 220X ) exome sequence file from a blood sample to simulate copy number profiles of 3 exemplar tumors from each of 10 tumor types at 5 tumor cellularity levels ( 20–100% , 150 BAM files in total ) . To demonstrate feasibility beyond exome data , we introduced read alignments to a targeted 5-gene cell-free DNA sequencing library to simulate EGFR amplifications at frequencies consistent with circulating tumor DNA ( 10 , 1 , 0 . 1 and 0 . 01% ) while retaining the multimodal insert size distribution of the original data . We expect Bamgineer to be of use for development and systematic benchmarking of CNV calling algorithms by users using locally-generated data for a variety of applications . The source code is freely available at http://github . com/pughlab/bamgineer .
The emergence and maturation of next-generation sequencing technologies , including whole genome sequencing , whole exome sequencing , and targeted sequencing approaches , has enabled researchers to perform increasingly more complex analysis of copy number variants ( CNVs ) [1] . While genome sequencing-based methods have long been used for CNV detection , these methods can be confounded when applied to exome and targeted sequencing data due to non-contiguous and highly-variable nature of coverage and other biases introduced during enrichment of target regions[1–5] . In cancer , this analysis is further challenged by bulk tumor samples that often yield nucleic acids of variable quality and are composed of a mixture of cell-types , including normal stromal cells , infiltrating immune cells , and subclonal cancer cell populations . Circulating tumor DNA presents further challenges due to a multimodal DNA fragment size distribution and low amounts of tumor-derived DNA in blood plasma . Therefore , development of CNV calling methods on arbitrary sets of tumor-derived data from public repositories may not reflect the type of tumor specimens encountered at an individual centre , particularly formalin-fixed-paraffin embedded tissues routinely profiled for diagnostic testing . Due to lack of a ground truth for validating CNV callers , many studies have used simulation to model tumor data[6] . Most often , simulation studies are used in an ad-hoc manner using customized formats to validate specific tools and settings with limited adaptability to other tools . More generalizable approaches aim at the de novo generation of sequencing reads according to a reference genome ( e . g . wessim[3] , Art-illumina[7] , and dwgsim[8] . However , de novo simulated reads do not necessarily capture subtle features of empirical data , such as read coverage distribution , DNA fragment insert size , quality scores , error rates , strand bias and GC content[6]; factors that can be more variable for exome and targeted sequencing data particularly when derived from clinical specimens . Recently , Ewing et al . developed a tool , BAMSurgeon , to introduce synthetic mutations into existing reads in a Binary alignment Mapping ( BAM ) file[9] . BAMSurgeon provides support for adjusting variant allele fractions ( VAF ) of engineered mutations based on prior knowledge of overlapping CNVs but does not currently support direct simulation of CNVs themselves . Here we introduce Bamgineer , a tool to modify existing BAM files to precisely model allele-specific and haplotype-phased CNVs ( Fig 1 ) . This is done by introducing new read pairs sampled from existing reads , thereby retaining biases of the original data such as local coverage , strand bias , and insert size . As input , Bamgineer requires a BAM file and a list of non-overlapping genomic coordinates to introduce allele-specific gains and losses . The user may explicitly provide known haplotypes or chose to use the BEAGLE[10] phasing module that we have incorporated within Bamgineer . We implemented parallelization of the Bamgineer algorithm for both standalone and high performance computing cluster environments , significantly improving the scalability of the algorithm . Overall , Bamgineer gives investigators complete control to introduce CNVs of arbitrary size , magnitude , and haplotype into an existing reference BAM file . We have uploaded all software code to a public repository ( http://github . com/pughlab/bamgineer ) ) .
For all proof-of-principle experiments , we used exome sequencing data from a single normal ( peripheral blood lymphocyte ) DNA sample . DNA was captured using the Agilent SureSelect Exome v5+UTR kit and sequenced to 220X median coverage as part of a study of neuroendocrine tumors . Reads were aligned to the hg19 build of the human genome reference sequence and processed using the Genome Analysis Toolkit ( GATK ) Best Practices pipeline . Following the validation of our tool for readily-detected chromosome- and arm-level events , we next used Bamgineer to simulate CNV profiles mimicking 3 exemplar tumors from each of 10 different cancer types profiled by The Cancer Genome Atlas using the Affymetrix SNP6 microarray platform: lung adenocarcinoma ( LUAD ) ; lung squamous cell ( LUSC ) ; head and neck squamous cell carcinoma ( HNSC ) ; glioblastoma multiformae ( GBM ) ; kidney renal cell carcinoma ( KIRC ) ; bladder ( BLCA ) ; colorectal ( CRC ) ; uterine cervix ( UCEC ) ; ovarian ( OV ) , and breast ( BRCA ) cancers ( Table 1 ) . To select 3 exemplar tumors for each cancer type , we chose profiles that best represented the copy number landscape for each cancer type . First , we addressed over-segmentation of the CNV calls from the microarray data by merging segments of <500 kb in size with the closest adjacent segment and removing the smaller event from the overlapping gain and loss regions . We then assigned a score to each tumor that reflects its similarity to other tumor of the same cancer type ( S7 Fig ) . This score integrates total number of CNV gain and losses ( Methods , Eq 6 ) , median size of each gain and loss , and the overlap of CNV regions with GISTIC peaks for each cancer type as reported by The Cancer Genome Atlas ( Table 1 ) . We selected three high ranking tumors for each cancer type such that , together , all significant GISTIC[15] peaks for that tumor type were represented . A representative profile from a single tumor is shown in Fig 2C . Subsequently , for each of the 30 selected tumor profiles ( 3 for each of 10 cancer types ) , we introduced the corresponding CNVs at 5 levels of tumor cellularity ( 20 , 40 , 60 , 80 , and 100% ) resulting in 150 BAM files in total . For each BAM file , we used Sequenza to generate allele-specific copy number calls as done previously . Tumor/normal log2 ratios are shown in Fig 3 for one representative from each cancer type . From this large set of tumors , we next set out to compare Picard metrics and CNV calls as we did for the arm- and chromosome-level pilot . We evaluated Bamgineer using several metrics: tumor allelic ratio , SNP phasing consistency , and tumor to normal log2 ratios ( Fig 4 ) . As expected , across all regions of a single copy gain , tumor allelic ratio was at ~0 . 66 ( interquartile range: 0 . 62–0 . 7 ) for the targeted haplotype and 0 . 33 ( interquartile range: 0 . 3–0 . 36 ) for the other haplotype . As purity was decreased , we observed a corresponding decrease in allelic ratios , from 0 . 66 down to 0 . 54 ( interquartile range: 0 . 5–0 . 57 ) for targeted and an increase ( from 0 . 33 ) to 0 . 47 ( interquartile range: 0 . 43–0 . 5 ) for the other haplotype for 20% purity ( Fig 4A and 4B ) . These changes correlated directly with decreasing purity ( R2 > 0 . 99 ) for both haplotypes . Similarly , for single copy loss regions , as purity was decreased from 100% to 20% the allelic ratio linearly decreased ( R2 > 0 . 99 ) from ~0 . 99 ( interquartile range: 0 . 98–1 . 0 ) for targeted haplotype to ~0 . 55 ( interquartile range: 0 . 51–0 . 58 ) for targeted haplotype and increases from 0 to ~0 . 43 ( interquartile range: 0 . 4–0 . 46 ) for the other haplotype ( Fig 4B ) . The results for log2 tumor to normal depth ratios of segments normalized for average ploidy were also consistent with the expected values ( Methods , Eq 2 ) . For CNV gain regions , log2 ratio decreased from ~0 . 58 ( log2 of 3/2 ) to ~0 . 13 as purity was decreased from 100% to 20% . For CNV loss regions , as purity was decreased from 100% to 20% , the log2 ratio increased from -1 ( log2 of 1/2 ) to -0 . 15 , consistent with Eq 2 ( Fig 4C; S1-S4 for individual cancers ) . Ultimately , we wanted to assess whether Bamgineer was introducing callable CNVs consistent with segments corresponding to the exemplar tumor set . To assess this , we calculated an accuracy metric ( Fig 4D ) as: accuracy=TP+TFTP+TF+FP+FN where TP , TF , FP and FN represent number of calls from Sequenza corresponding to true positives ( perfect matches to desired CNVs ) , true negatives ( regions without CNVs introduced ) , false positives ( CNV calls outside of target regions ) and false negatives ( target regions without CNVs called ) . TP , TF , TN , FN were calculated by comparing Sequenza absolute copy number ( predicted ) to the target regions for introduction of 1 Mb CNV bins across the genome . As tumor content decreased , accuracy for both gains and losses decreased as false negatives became increasingly prevalent due to small shifts in log2 ratios . We note that ( as expected ) , decreasing cancer purity from 100% to 20% generally decreases the segmentation accuracy . Additionally , we observe that segmentation accuracy is on average , significantly higher for gain regions compared to the loss regions for tumor purity levels below 40% ( Fig 4D ) . This is consistent with previous studies that show the sensitivity of CNV detection from sequencing data is slightly higher for CNV gains compared to CNV losses[16] . We also note that with decreasing cancer purity , the decline in segmentation accuracy follows a linear pattern of decline for gain regions and an abrupt stepwise decline for loss regions ( Fig 4D; segmentation accuracies are approximately similar for 40% and 20% tumor purities ) . Finally , we observed a degree of variation in terms of segmentation accuracy across individual cancer types ( S1–S4 Figs ) . Segmentation accuracy was lower for LUAD , OV and UCEC compared to other simulated cancer types for this study . The relative decline in performance is seen in cancer types where CNV gains and losses cover a sizeable portion of the genome; and hence , the original loss and gain events sampled from TCGA had significant overlaps . As a result , after resolving overlapping gain and loss regions ( S7 Fig ) , on average , the final target regions constitute a larger number of small ( < 200 kb ) loss regions immediately followed by gain regions and vice versa; making the accurate segmentation challenging for the CBS ( circular binary segmentation ) algorithm implemented by Sequenza relying on presence of heterozygous SNPs . This can cause uncertainties in assignments of segment boundaries . In summary , application of an allele-specific caller to BAMs generated by Bamgineer recapitulated CNV segments consistent with >95% ( medians: 95 . 1 for losses and 97 . 2 for gains ) of those input to the algorithm . However , we note some discrepancies between the expected and called events , primarily due to small CVNs as well as large segments of unprobed genome between exonic sequences . To evaluate the use of Bamgineer for circulating tumor DNA analysis , we simulated the presence of an EGFR gene amplification in read alignments from a targeted 5-gene panel ( 18 kb ) applied to a cell-free DNA from a healthy donor and sequenced to >50 , 000X coverage . To mirror concentrations of tumor-derived fragments commonly encountered in cell-free DNA[17 , 18] , we introduced gain of an EGFR haplotype at frequencies of 100 , 10 , 1 , 0 . 1 , and 0 . 01% . This haplotype included 3 SNPs covered by our panel , which were phased and subject to allele-specific gain accordingly . As with the exome data , we observed shifts in coverage of specific allelic variants , and haplotype representation consistent with the targeted allele frequencies ( Fig 5A , Supplemental S1 Table ) . Furthermore , read pairs introduced to simulate gene amplification retain the bimodal insert size distribution characteristic of cell-free DNA fragments ( Fig 5B and 5C ) . While this experiment showcases the ability of Bamgineer to faithfully represent features of original sequencing data while controlling allelic amplification at the level of the individual reads , these subtle shifts are currently beyond the sensitivity of conventional CNV callers when applied to small , deeply covered gene panels . Therefore , it is our hope that Bamgineer may be of value to aid develop of new methods capable of detecting copy number variants supported by a small minority of DNA fragments in a specimen . Bamgineer is computationally intensive and the runtime of the program is dictated by the number of reads that must be processed , a function of the coverage of the genomic footprint of target regions . To ameliorate the computational intensiveness of the algorithm , we employed a parallelized computing framework to maximize use of a high-performance compute cluster environment when available . We took advantage of two features in designing the parallelization module . First , we required that added CNVs are independent for each chromosome ( although nested events can likely be engineered through serial application of Bamgineer ) . Second , since we did not model interchromosomal CNV events , each chromosome can be processed independently . As such , CNV regions for each chromosome can be processed in parallel and aggregated as a final step . S8 Fig shows the runtimes for The Cancer Genome Atlas simulation experiments . Using a single node with 12 cores and 128 GB of RAM , each synthetic BAM took less than 3 . 5 hours to generate . We also developed a version of Bamgineer that can be launched from sun grid engine cluster environments . It uses python pipeline management package ruffus to parallelize tasks automatically and log runtime events . It is highly modular and easily updatable . If disrupted during a run , the pipeline can continue to completion without re-running previously completed intermediate steps .
Here , we introduced Bamgineer , to introduce user-defined haplotype-phased allele-specific copy number events into an existing Binary Alignment Mapping ( BAM ) file , obtained from exome and targeted sequencing experiments . As proof of principle , we generated , from a single high coverage ( mean: 220X ) BAM file derived from a human blood sample , a series of 30 new BAM files containing a total of 1 , 693 simulated copy number variants ( on average , 56 CNVs comprising 1800Mb i . e . ~55% of the genome per tumor ) corresponding to profiles from exemplar tumors for each of 10 cancer types . To demonstrate quantitative introduction of CNVs , we further simulated 4 levels of tumor cellularity ( 20 , 40 , 60 , 80% purity ) resulting in an additional 120 new tumor BAM files . We validated our approach by comparing CNV calls and inferred purity values generated by an allele-specific CNV-caller ( Sequenza[14] ) as well as a focused comparison of allelic variant ratios , haplotype-phasing consistency , and tumor/normal log2 ratios for inferred CNV segments ( S1–S4 Figs ) . In every case , inferred purity values were within ±5% of the targeted purity; and majority of engineered CNV regions were correctly called by Sequenza ( accuracy > 94%; S1–S4 Figs ) . Allele variant ratios were also consistent with the expected values both for targeted and the other haplotypes ( Median within ±3% of expected value ) . Median tumor/normal log2 ratios were within ±5% of the expected values . To demonstrate feasibility beyond exome data , we next evaluated these same metrics in a targeted 5-gene panel applied to a cell-free DNA sequencing library generated from a healthy blood donor and sequenced to >10 , 000X coverage[17] To simulate concentrations of tumor-derived fragments typically encountered in cancer patients , we introduced EGFR amplifications at frequencies of 100 , 10 , 1 , 0 . 1 , and 0 . 01% . As with the exome data , we observed highly specific shifts in allele variant ratios , log2 coverage ratios , and haplotype representation consistent with the targeted allele frequencies . Our method also retained the bimodal DNA insert size distribution observed in the original read alignment . However , it is worthwhile noting that , these minute shifts are currently beyond the sensitivity of existing CNV callers when applied to small , deeply covered gene panels . Consequently , we anticipate that Bamgineer may be of value to aid develop of new methods capable of detecting copy number variants supported by a small minority of DNA fragments . In the experiments conducted in this study , we limited ourselves to autosomes and to maximum total copy number to 4 . Naturally , Bamgineer can readily simulate higher-level copy number states and alter sex chromosomes as well ( S10 Fig ) . While chromosome X in diploid state ( e . g . XX in normal female ) is treated identically to autosomes , for both X and Y chromosomes beginning in haplotype state ( e . g . XY in normal male ) , the haplotype phasing step is skipped and Bamgineer samples all reads on these chromosomes independently . For high-level amplifications , the ability of Bamgineer to faithfully retain the features of the input Bam file ( e . g . DNA fragment insert size , quality scores and so on ) , depends on the intrinsic factors such as the length of the desired CNV , mean depth of coverage and fragment length distribution of the original input BAM file ( see Materials and Methods ) . The significance of this work in the context of CNV inference in cancer is twofold: 1 ) users can simulate CNVs using their own locally-generated alignments so as to reflect lab- , biospecimen- , or pipeline-specific features; 2 ) bioinformatic methods development can be better supported by ground-truth sequencing data reflecting CNVs without reliance on generated test data from suboptimal tissue or plasma specimens . Bamgineer addresses both problems by creating standardized sequencing alignment format ( BAM files ) harbouring user-defined CNVs that can readily be used for algorithm optimization , benchmarking and other purposes . We expect our approach to be applicable to tune algorithms for detection of subtle CNV signals such as somatic mosaicism or circulating tumor DNA . As these subtle shifts are beyond the sensitivity of many CNV callers , we expect our tool to be of value for the development of new methods for detecting such events trained on conventional DNA sequencing data . By providing the ability to create customized user-generated reference data , Bamgineer will prove valuable inn development and benchmarking of CNV calling and other sequence data analysis tools and pipelines . The work presented herein can be extended in several directions . First , Bamgineer is not able to reliably perform interchromosomal operations such as chromosomal translocation , as our focus has been on discrete regions probed by exome and targeted panels . Second , while Bamgineer is readily applicable to whole genome sequence data , sufficient numbers of reads are required for re-pairing when introducing high-level amplifications . As such , shallow ( 0 . 1-1X ) or conventional ( ~30X ) whole genome sequence data may only be amenable to introduction of arm-level alterations as smaller , focal targets may not contain sufficient numbers of reads to draw from to simulate high-level amplifications . Additionally , in our current implementation , we limited the simulated copy numbers to non-overlapping regions . Certainly , such overlapping CNV regions occur in cancer and iterative application of Bamgineer may enable introduction of complex , nested events . Finally , introduction of compound , serially acquired CNVs may be of interest to model subclonal phylogeny developed over time in bulk tumor tissue samples .
The user provides 2 mandatory inputs to Bamgineer as command-line arguments: 1 ) a BAM file containing aligned paired-end sequencing reads ( “Normal . bam” ) , 2 ) a BED file containing the genome coordinates and type of CNVs ( e . g . allele-specific gain ) to introduce ( “CNV regions . bed” ) . Bamgineer can be used to add four broad categories of CNVs: Balanced Copy Number Gain ( BCNG ) , Allele-specific Copy Number Gain ( ASCNG ) , Allele-specific Copy Number Loss ( ACNL ) , and Homozygous Deletion ( HD ) . For example , consider a genotype AB at a genomic locus where A represents the major and B represents the minor allele . Bamgineer can be applied to convert that genomic locus to any of the following copy number states: {A , B , ABB , AAB , ABB , AABB , AAAB , ABBB , …} An optional VCF file containing phased germline calls can be provided ( phased_het . vcf ) . If this file is not provided , Bamgineer will call germline heterozygous single nucleotide polymorphisms ( SNPs ) using the GATK HaplotypeCaller and then categorize alleles likely to be co-located on the same haplotypes using BEAGLE and population reference data from the HapMap project . To obtain paired-reads in CNV regions of interest , we first intersect Normal . bam with the targeted regions overlapping user-defined CNV regions ( roi . bed ) . This operation generates a new BAM file ( roi . bam ) . Subsequently , depending on whether the CNV event is a gain or loss , the algorithms performs two separate steps as follows . To introduce copy number gains , Bamgineer creates new read-pairs constructed from existing reads within each region of interest . This approach thereby avoids introducing pairs that many tools would flag as molecular duplicates due to read 1 and read 2 having start and end positions identical to an existing pair . If desired , these read pairs can be restricted to reads meeting a specific SAM flag . For our exome experiments , we used read pairs with a SAM flag equal to 99 , 147 , 83 , or 163 , i . e . read paired , read mapped in proper pair , mate reverse ( forward ) strand , and first ( second ) in pair . To enable support for the bimodal distribution of DNA fragment sizes in ctDNA , we removed the requirement for “read mapped in proper pair” and used read pairs with a SAM flag equal to 97 , 145 , 81 , or 161 . Users considering engineering of reads supporting large inserts or intrachromosomal read pairs may also want to remove the requirement for “read mapped in proper pair” . Additionally , we required that the selection of the newly paired read is within ±50% ( ±20% for ctDNA ) of the original read size . The newly created read- pairs are provided unique read names to avoid confusion with the original input BAM file . To enable inspection of these reads , these newly created read pairs are stored in a new BAM file , gain_re_paired_renamed . bam , prior to merging into the final engineered BAM . Since we only consider high quality reads ( i . e . properly paired reads , primary alignments and mapping quality > 30 ) , the newly created BAM file contains fewer reads compared to the input file ( ~90–95% in our proof-of-principle experiments ) . As such , at every transition we log the ratio between number of reads between the input and output files . High-level copy number amplification ( ASCN > = 4 ) . To achieve higher than 4 copy number amplifications , during the read/mate pairing step , we pair each read with more than one mate read ( Fig 1 ) to generate more new reads ( to accommodate the desired copy number state ) . Though , since as stated a small portion of newly created paired reads do not meet the inclusion criteria , we aim to create more reads than necessary in the initial phase and use the sampling to adjust them in a later phase . For instance , to simulate copy number of 6 , in theory we need create two new read pairs for every input read . Hence , in the initial “re-pairing” step we aim to create four paired reads per read ( instead of 3 ) , so that the newly created Bam file includes enough number of reads ( as a rule of thumb , we use read-paring window size of ~20% higher than theoretical value ) . It should be noted that the maximum copy number amplification that can faithfully retain the features of the input BAM file ( e . g . DNA fragment insert size , quality scores and so on ) , depends on the intrinsic factors such as the length of the desired CNV , mean depth of coverage and fragment length distribution of the original input BAM file . Introduction of mutations according to haplotype state . To ensure newly constructed read-pairs match the desired haplotype , we alter the base at heterozygous SNP locations ( phased_het . vcf ) within each read according to haplotype provided by the user or inferred using the BEAGLE algorithm . To achieve this , we iterate through the set of re-paired reads used to increase coverage ( gain_re_paired_renamed . bam ) and modify bases overlapping SNPs corresponding to the target haplotype ( phased_het . vcf ) . We then write these reads to a new BAM file ( gain_re_paired_renamed_mutated . bam ) prior to merging into the final engineered BAM ( S9 Fig ) . As an illustrative example consider two heterozygous SNPs , AB and CD both with allele frequencies of ~0 . 5 in the original BAM file ( i . e . approximately half of the reads supporting reference bases and the other half supporting alternate bases . To introduce a 2-copy gain of a single haplotype , reads to be introduced must match the desired haplotype rather than the two haplotypes found in the original data . If heterozygous AB and CD are both located on a haplotype comprised of alternative alleles , at the end of this step , 100% of the newly re-paired reads will support alternate base-pairs ( e . g . BB and DD ) . Based on the haplotype structure provided , other haplotype combinations are possible including AA/DD , BB/CC , etc . Sampling of reads to reflect desired allele fraction . Depending on the absolute copy number desired for the for CNV gain regions , we sample the BAM files according to the desired copy number state . We define conversion coefficient as the ratio of total reads in the created BAM from previous step ( gain_repaired_mutated . bam ) to the total reads extracted from original input file ( roi . bam ) : ρ=no . ofreadsingain_re_paired_mutated . bamno . ofreadsinroi . bam ( 1 ) According to the maximum number of absolute copy number ( ACN ) for simulated CNV gain regions ( defined by the user ) , two scenarios are conceivable as follows . Copy number gain example . For instance , to achieve the single copy gain ( ACN = 3 , e . g . ABB copy state ) , the file in the previous step ( gain_re_paired_renamed_mutated . bam ) , should be sub-sampled such that on average depth of coverage is half that of extracted reads from the target regions from the original input normal file ( roi . bam ) . Thus , the final sampling rate is calculated by dividing ½ ( 0 . 5 ) by ρ ( subsample gain_re_paired_renamed_mutated . bam such that we have half of the roi . bam depth of coverage for the region; in practice adjusted sampling rate is in the range of 0 . 51–0 . 59 i . e . 0 . 85 < ρ < 1 for CN = 3 ) and the new reads are written to a new BAM file ( gain_re_paired_renamed_mutated_sampled . bam ) that we then merge with the original reads ( roi . bam ) to obtain gain_final . bam . Similarly to obtain three copy number gain ( ACN = 5 ) and the desired genotype ABBBB , the gain_re_paired_renamed_mutated . bam is subsampled such that depth of coverage is 3/2 ( 1 . 5 ) that of extracted reads from the target regions from the original input normal file ( note that as explained during the new paired-read generation step , we have already created more reads than needed ) . To introduce CNV losses , Bamgineer removes reads from the original BAM corresponding to a specific haplotype and does not create new read pairs from existing ones . To diminish coverage in regions of simulated copy number loss , we sub-sample the BAM files according to the desired copy number state and write these to a new file . The conversion coefficient is defined similarly as the number of reads in loss_mutated . bam divided by number of reads in roi_loss . bam ( > ~0 . 98 ) . Similar to CNV gains , the sampling rate is adjusted such that after the sampling , the average depth of coverage is half that of extracted reads from the target regions ( calculated by dividing 0 . 5 by conversion ratio , as the absolute copy number is 1 for loss regions ) . Finally , we subtract the reads in CNV loss BAMs from the input . bam ( or input_sampled . bam ) and merge the results with CNV gain BAM ( gain_final . bam ) to obtain , the final output BAM file harbouring the desired copy number events . To validate that the new paired-reads generated from the original BAM files show similar probability distribution , we used two-sided Kolmogorov–Smirnov ( KS ) test . The critical D-values where calculated for α = 0 . 01 as follows: Dα=c ( α ) n1+n2n1n2 ( 2 ) where coefficient c ( α ) is obtained from Table of critical values for KS test ( https://www . webdepot . umontreal . ca/Usagers/angers/MonDepotPublic/STT3500H10/Critical_KS . pdf; 1 . 63 for α = 0 . 01 ) and n1 and n2 are the number of samples in each dataset . To assess tumor allelic ratio consistency , for each SNP the theoretical allele frequency parameter was used as a reference point ( Eq 3 ) . Median , interquartile range and mean were drawn from the observed values for each haplotype-event pair for all the SNPs . The boxplot distribution of the allele frequencies were plotted and compared against the theoretical reference point . To assess the segmentation accuracy , we used log2 tumor to normal depth ratios of segments normalized for mean ploidy as the metric; where the mean ploidy is ( Eqs 4 and 5 ) . To benchmark the performance of segmentation accuracy , we used accuracy as the metrics . Statistical analysis was performed with the functions in the R statistical computing package using RStudio . Theoretical expected values . The expected value for tumor allelic frequencies at heterozygous SNP loci for tumor purity level of p ( 1-p: normal contamination ) is calculated as follows: AF ( snp ) =pAFtcnt+ ( 1−p ) AFncnnpcnt+ ( 1−p ) cnn ( 3 ) where AFt and AFn represent the expected allele frequencies for tumor and normal and cnt and cnn the expected copy number for tumor and normal at specific SNP loci . For CNV events used in this experiment AFt are ( 1/3 or 2/3 ) for gain and ( 1 or 0 ) for loss CNVs according to the haplotype information ( whether or not they are located on the haplotype that is affected by each CNV ) . The expected value for the average ploidy ( ∅^ ) is calculated as follows ∅^=1W ( ∑i=1ncngwgi+∑j=1mcnlwlj+∑i=1ncnn ( W−G−L ) ( 4 ) , where cng , cnl , cnn , wg and wl represent the expected ploidy for gain , loss and normal regions , and the length of individual gain and loss events respectively . W , G , and L represent total length ( in base pairs ) for gain regions , loss regions , and the entire genome ( ~ 3e9 ) . The expected log2ratio for each segment is calculated as follows log2ratio ( seg ) =log2 ( p×cnseg+ ( 1−p ) ×cnn∅^ ) ( 5 ) , where cnseg is the segment mean from Sequenza output , p tumor purity and ∅^ is the average ploidy calculated above . cnn is the copy number of copy neutral region ( i . e . 2 ) Similarity score to rank TCGA tumors . The similarity score for specific cancer type ( c ) and sampled tumor ( t ) is calculated as follows: S ( c , t ) =1/ ( |2gt−Gc−Go|+|2lt−Lc−Lo|+ϵ ) ( 6 ) , where gt , Gc , Go represent the total number of gains for specific tumor sampled from Cancer Genome Atlas ( after merging adjacent regions and removing overlapping regions ) , median number of gains for specific tumor type , and number of gain events overlapping with GISTIC peaks respectively; lt , Lc , Lo represent the above quantities for CNV loss regions ( ϵ is an arbitrary small positive value to avoid zero denominator ) . The higher score the closer is the sampled tumor to an exemplar tumor from specific cancer type .
|
We present Bamgineer , a software program to introduce user-defined , haplotype-specific copy number variants ( CNVs ) at any frequency into standard Binary Alignment Mapping ( BAM ) files . Copy number gains are simulated by introducing new DNA sequencing read pairs sampled from existing reads and modified to contain SNPs of the haplotype of interest . This approach retains biases of the original data such as local coverage , strand bias , and insert size . Deletions are simulated by removing reads corresponding to one or both haplotypes . In our proof-of-principle study , we simulated copy number profiles from 10 cancer types at varying cellularity levels typically encountered in clinical samples . We also demonstrated introduction of low frequency CNVs into cell-free DNA sequencing data that retained the bimodal fragment size distribution characteristic of these data . Bamgineer is flexible and enables users to simulate CNVs that reflect characteristics of locally-generated sequence files and can be used for many applications including development and benchmarking of CNV inference tools for a variety of data types .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] |
[
"sequencing",
"techniques",
"alleles",
"genetic",
"mapping",
"genome",
"analysis",
"copy",
"number",
"variation",
"molecular",
"genetics",
"molecular",
"biology",
"techniques",
"research",
"and",
"analysis",
"methods",
"sequence",
"analysis",
"genome",
"complexity",
"sequence",
"alignment",
"bioinformatics",
"molecular",
"biology",
"genetic",
"loci",
"haplotypes",
"dna",
"sequence",
"analysis",
"heredity",
"database",
"and",
"informatics",
"methods",
"genetics",
"biology",
"and",
"life",
"sciences",
"genomics",
"dna",
"sequencing",
"computational",
"biology"
] |
2018
|
Bamgineer: Introduction of simulated allele-specific copy number variants into exome and targeted sequence data sets
|
Andes virus ( ANDV ) is a human-pathogenic hantavirus . Hantaviruses presumably initiate their mRNA synthesis by using cap structures derived from host cell mRNAs , a mechanism called cap-snatching . A signature for a cap-snatching endonuclease is present in the N terminus of hantavirus L proteins . In this study , we aimed to solve the atomic structure of the ANDV endonuclease and characterize its biochemical features . However , the wild-type protein was refractory to expression in Escherichia coli , presumably due to toxic enzyme activity . To circumvent this problem , we introduced attenuating mutations in the domain that were previously shown to enhance L protein expression in mammalian cells . Using this approach , 13 mutant proteins encompassing ANDV L protein residues 1–200 were successfully expressed and purified . Protein stability and nuclease activity of the mutants was analyzed and the crystal structure of one mutant was solved to a resolution of 2 . 4 Å . Shape in solution was determined by small angle X-ray scattering . The ANDV endonuclease showed structural similarities to related enzymes of orthobunya- , arena- , and orthomyxoviruses , but also differences such as elongated shape and positively charged patches surrounding the active site . The enzyme was dependent on manganese , which is bound to the active site , most efficiently cleaved single-stranded RNA substrates , did not cleave DNA , and could be inhibited by known endonuclease inhibitors . The atomic structure in conjunction with stability and activity data for the 13 mutant enzymes facilitated inference of structure–function relationships in the protein . In conclusion , we solved the structure of a hantavirus cap-snatching endonuclease , elucidated its catalytic properties , and present a highly active mutant form , which allows for inhibitor screening .
Andes virus ( ANDV ) belongs to the genus Hantavirus within the family Bunyaviridae . Hantaviruses can be pathogenic to humans and are distributed worldwide [1–3] . Their natural hosts are various rodent species [4] . ANDV is endemic in Argentina and Chile and its main reservoir host is the long-tailed pygmy rice rat ( Oligoryzomys longicaudatus ) [5] . The virus causes hantavirus cardiopulmonary syndrome ( HCPS ) that is associated with a case fatality of up to 40% . Related HCPS-causing hantaviruses are found throughout the Americas . However , in contrast to other hantaviruses , ANDV can be transmitted from human to human [6 , 7] . The hantavirus genome consists of three RNA segments: L ( large ) , M ( medium ) , and S ( small ) [8] . The L segment encodes the 250-kDa L protein , which contains an RNA-dependent RNA polymerase ( RdRp ) domain in the center as indicated by sequence homology [9] . The L protein is likely to possess additional enzymatic functions involved in transcription and replication of the RNA genome . The M segment contains the gene for the glycoprotein precursor , which is cleaved co-translationally into the envelope proteins Gn and Gc . The S segment encodes the nucleoprotein , which encapsidates the viral genome . Several hantaviruses encode a second gene product in the S segment , the non-structural protein NSs . This small protein seems to play a role in modulating the immune response of the host cell [10 , 11] . The 5’ ends of hantavirus mRNAs contain a stretch of heterologous nucleotides suggesting that the virus uses a mechanism called ‘cap-snatching’ to initiate transcription of its mRNAs [12 , 13] . Cap-snatching involves two steps . First , the 5’ end of a cellular mRNA is bound by a viral cap-binding protein and secondly , the cap is cleaved off several nucleotides downstream by a viral endonuclease . The capped RNA fragment is used as a primer to initiate transcription of the viral mRNA by the RdRp . The viral strategy of stealing the 5’ cap from host cell mRNA was first described for influenza A virus ( IAV; family Orthomyxoviridae ) [14] . Here , the cap-binding and endonuclease domains reside in the PB2 and PA proteins , respectively [15–19] . In other segmented negative strand RNA viruses , the endonuclease resides in the very N terminus of the L protein , as shown for La Crosse virus ( LACV; genus Orthobunyavirus , family Bunyaviridae ) , Lassa virus , and lymphocytic choriomeningitis virus ( both family Arenaviridae ) [20–23] . An endonuclease motif ( H-PD-D/E-K ) is also found in the N terminus of hantavirus L proteins , although there is no experimental evidence for a functional enzyme . The endonuclease is a potential target for antiviral therapy , as cap-snatching is an essential step in virus replication [24–35] . Investigation of L protein of ANDV and other hantaviruses has been challenging , because the protein is difficult to express in heterologous systems [36 , 37] . We have previously shown that this phenotype is determined at post-transcriptional level and that mutations in the putative endonuclease can rescue L protein expression in mammalian cells [37] . Both suggests that a strong catalytic activity of the ANDV endonuclease prevents high-level expression by down regulating the level of its own and host cell mRNAs . The aim of the current study was to provide proof for the existence of a cap-snatching endonuclease in ANDV L protein using a structural and biochemical approach . We took advantage of the previously characterized mutations enhancing the L protein level in mammalian cells to facilitate expression of the endonuclease domain in bacteria .
For structural and biochemical characterization of the putative endonuclease of ANDV , we aimed to express the N terminus of L protein in E . coli . To this end , plasmids for expression of the N-terminal 163 , 179 , 191 , 194 , 197 , 200 , 211 , 214 , and 228 amino acid residues of wild-type ANDV L protein were generated using E . coli cloning strain DH5α . To facilitate purification , the expression constructs contained an N-terminal His-tag or His-tag plus maltose-binding protein ( MBP ) , respectively . However , we encountered difficulties in expressing the L protein fragments in bacteria , as observed previously in mammalian cells [36 , 37] . Proteins with less than 200 residues were found exclusively in inclusion bodies . Transformation of expression strains of E . coli with constructs containing 200 or more residues failed repeatedly , suggesting that background expression of soluble wild-type endonuclease is already toxic to the cells . To circumvent this problem , we took advantage of 15 amino acid exchanges that enhanced L protein expression in mammalian cells , presumably because they attenuate the enzymatic activity of the endonuclease [37] . All 15 mutations were individually introduced into the plasmids for expression of L protein amino acid residues 1–200 fused to an N-terminal His-tag ( ANDV L1–200 ) and the mutated constructs were tested for expression in E . coli BL21 cells . The results are summarized in Table 1 . Two mutants ( L1–200 Y32V and D37A ) showed a phenotype like the wild-type protein , i . e . no bacterial growth was observed . The remaining 13 mutants ( L1–200 R35H , H36R , D40E , I43A , K44A , N50A , P96A , D97E , N98A , E110A , K124A , K127A , and N167A ) were successfully expressed with varying yield and purified via nickel affinity and size exclusion chromatography ( S1 Fig ) . Noticeable slower growth of expressing E . coli was observed for L1–200 N167A . A thermofluor assay [38] was used to evaluate the stability of the mutant proteins under various conditions . The ANDV L1–200 mutants were most stable at high salt concentrations and low pH ( S2 Fig ) , as expected from their calculated isoelectric point of approximately 7 . 5 . Therefore , we used pH 5 . 5 and 1 M NaCl for protein purification . All mutants eluted from the size exclusion column according to their molecular mass , indicating they form monomers in solution . ANDV L1–200 K127A was chosen for structural studies , as this mutant was the most stable mutant , which retained the capacity to bind to manganese ions—a co-factor of nucleases—along with a residual enzymatic activity ( for details , see section on thermal stability below ) . The mutant was expressed as His-tag fusion of variable lengths ( 200 , 211 , 214 , and 228 L protein residues ) and subjected to crystallization trials . We succeeded to crystallize ANDV L1–200 K127A in the presence of Mn2+ after cleavage of the His-tag . The crystal structure was solved in space group P4212 by molecular replacement using coordinates for residues 32–162 from the endonuclease of Hantaan virus ( [39] , co-submission ) , allowing for the building of a contiguous polypeptide chain for one molecule in the asymmetric unit and refinement to a resolution of 2 . 4 Å ( see S1 Table , 5HSB . pdb ) . Clear electron density was visible for the complete structure with the exception of the loop between the first two α-helices ( residues 13–20 ) , where the signal was weak . The structure clearly shows a Mn2+ ion bound to the active site ( Fig 1A ) , as verified by an anomalous signal from data collected at 1 . 77 Å . The crystal structure of ANDV L1–200 K127A and the comparison to the cap-snatching endonucleases of LACV , LCMV , and IAV [15 , 19 , 21–23] provides clear evidence for the N terminus of ANDV L protein being an endonuclease ( Fig 1B ) . ANDV L1–200 has an overall structure similar to the other viral endonucleases with the highest similarity to its closest relative LACV . Considering the almost lacking sequence homology between the four proteins—except the catalytic H-PD-D/E-K motif—the structural homology is surprisingly high: RMSD values between the structural core ( orange ) of ANDV endonuclease and LACV , LCMV and IAV endonucleases are 1 . 7 Å , 1 . 9 Å and 2 . 2 Å , respectively ( as given by DALI [40] . The central β-sheet with the long α-helix αd that runs parallel to the β-sheet forms the core of the protein and provides the crucial active site residues in all four endonucleases ( Fig 1A and 1B , shown in orange ) . Further structural elements are helical , although number , length , as well as position of the α-helices vary significantly between the structures . In all endonucleases , the active site is located in a groove between two lobes . One lobe is formed by a helix bundle that consists of two to three α-helices from the N terminus and at least one long α-helix ( starting around residue 160 ) from the C-terminal end of the domain ( Fig 1A and 1B , shown in green ) . The interaction between the last residues of the C-terminal helix and the first residues of the second helix stabilizes the 3-helix bundle ( αa , αb , αf ) and provides a plausible explanation why constructs with less than 200 residues were misfolded and thus insoluble . The other lobe contains less conserved structural elements consisting of two to three small helices ( Fig 1A and 1B , shown in yellow ) . ANDV features a short additional α-helix αe between strand βc and the C-terminal helix αf , which is not present in LACV ( Fig 1C ) . One of the obvious differences between the endonuclease of ANDV and the other viruses are the overall dimensions of the domain: ANDV L1–200 is long and flat with approximate outer dimensions of 70 Å × 45 Å × 30 Å , whereas the other structures are more compact , in particular that of the distantly related IAV . The outer dimensions of the protein in the crystal fit with the dimensions in solution as determined by small angle X-ray scattering , indicating that the overall shape was not affected by crystallization artifacts ( Figs 1D and S3 ) . Most of the surface of ANDV L1–200 is formed by the helix bundle composed of the N- and C-terminal helices αa+b and αe+f , and the three small helices αc , αc’ and αc” ( Fig 1A and 1B ) . These helices differ between the endonucleases not only in their conformation and orientation , but also in their amino acid composition . Fig 2A shows how this leads to a significant alteration in surface charge distribution compared to the other enzymes . The ANDV endonuclease has large positively charged patches surrounding the active site groove , which are not or only to a lesser extent present in LACV and LCMV or even replaced by negatively charged areas . The more positively charged surface may increase the binding affinity of the enzyme to its substrate , i . e . negatively charged mRNA . A closer look at the active site of the ANDV endonuclease ( Fig 2C ) in comparison to LACV and IAV demonstrates the structural similarity of the essential catalytic residues . The side chains of the conserved H-PD-D/E-K motif superimpose well ( Fig 2B ) . The manganese bound to the active site also superimposes with one of the two metal ions present in the other structures . Sequence comparisons suggested that the active site of the ANDV endonuclease is more closely related to IAV than LACV [20 , 37] . Indeed , Lys124 of ANDV superimposes with Lys134 of IAV , whereas LACV has a threonine at this position and the equivalent lysine ( Lys94 ) is provided by a different part of the protein . Close to the active site , electron density for a sulfate and a glycerol molecule is visible in the structure ( Fig 2D ) . The glycerol ligand is coordinated by the conserved residues Tyr32 and Arg35 ( Fig 2E ) , both of which have previously been proposed to regulate the endonuclease activity of the L protein [37] . With the crystal structure of the ANDV cap-snatching endonuclease at hand , we may infer a more precise role for the 15 amino acid residues that have been implicated in the activity of the endonuclease in mammalian cells [37] ( Fig 3 ) . The residues can be allocated to five different groups according to their presumed role: ( 1 ) catalytic residues ( His36 , Asp97 , and Glu110 ) , ( 2 ) active site residues not essential for activity ( Asp37 , Pro96 , Lys124 , and Lys127 ) , ( 3 ) residues involved in RNA binding in the vicinity of the active site ( Tyr32 and Arg35 ) , ( 4 ) residues forming stabilizing contacts close to the active site ( Asp40 , Ile43 , Lys44 , Asn50 , and Asn98 ) , and ( 5 ) residues stabilizing helix αe ( Asn167 ) . The hypothetical role also considers biochemical and thermal stability data ( Table 1 and sections below ) . His36 , Asp97 , and Glu110 ( group 1 ) are the central catalytic residues coordinating the manganese ion ( Figs 2C and 3 ) . Asp37 and Pro96 ( group 2 ) are relevant for positioning the catalytic residues His36 and Asp97 , respectively . Lys124 and Lys127 ( group 2 ) , which emerge from helix αd , may position the RNA substrate in the active site . Tyr32 and Arg35 ( group 3 ) are located in helix αb and bind a glycerol solvent molecule in the structure . Their position at the entrance of the catalytic site and capacity to bind hydroxyl groups let us to speculate that these residues may also be involved in binding the negatively charged RNA substrate . Asp40 , Lys44 , Asn50 , and Asn98 ( group 4 ) form hydrogen bonds with each other and main chain atoms to stabilize and position the loop that contains the catalytic Asp97 . Ile43 forms hydrophobic interactions between the end of helix αb and several hydrophobic side chains both from the β-sheet and from helix αe . Group 5 contains only Asn167 . It is located in the small helix αe , which is not present in the LACV endonuclease ( Fig 1D ) and stabilizes this helix via hydrogen bonds with main chain atoms of strand βc . The thermal stability of the purified proteins without and with manganese was measured by thermofluor assay [38] in analogy to previously described experiments with the IAV and LACV cap-snatching endonucleases [16 , 20] . Melting temperature ( Tm ) as a measure of stability varies from 34 . 2°C for the least stable L1–200 I43A mutant to 49 . 2°C for the most stable L1–200 K124A mutant ( Fig 4A and 4B ) . For most mutants , the Tm increased by approximately 7°C in the presence of 4 mM manganese , presumably due to binding of the metal in the catalytic site . Consistent with this explanation , mutation of catalytic residues directly involved in metal coordination ( His36 , Asp97 , and Glu110 ) abolishes this stabilizing effect . Interestingly , increasing the manganese concentration to 16 mM led to a substantial additional Tm increase of up to 8°C for some mutants , suggesting binding of another metal ion by the active site ( Fig 4A and 4B ) . In contrast to manganese , magnesium ions showed a much weaker effect ( Fig 4A and 4B ) . Enzymatic activity was measured in a ribonuclease assay . A 5’-radioactively labeled 27mer single stranded ( ss ) RNA molecule was used as substrate , and the level of activity was measured by substrate degradation . None of the mutants was active in the absence of divalent ions . Addition of manganese rescued activity for some mutants ( Fig 5A ) , while magnesium had no enhancing effect at all . Therefore , all further assays were performed in the presence of manganese . As expected , active site mutants L1–200 H36R , D97E , and E110A were inactive despite addition of manganese ( Fig 5A and 5B ) . Mutation of Arg35 , Lys124 , and Lys127 , still allowed for very low residual activity . Mutation of Asn50 and Pro96 stabilizing the active site loop also strongly reduced endonuclease activity . Mutations affecting stability of the tertiary structure ( D40A , I43A , K44A , N98A , and N167A ) allowed for intermediate to high activity . We could not detect an intermediate cleavage product of specific length ( S4 Fig ) . The most active mutant was L1–200 N167A , which was therefore chosen for further biochemical experimentation . Thermal stability and enzymatic activity data are summarized in Table 1 . To investigate substrate specificity of the ANDV endonuclease , we used three mutants with a range of enzymatic activity: L1–200 N167A with high activity , L1–200 K44A with intermediate activity , and L1–200 D97E as a negative control . First , four different substrates were tested: a 27mer ssRNA that can form secondary structures ( predicted hairpin Tm = 77°C ) , the same sequence as double stranded ( ds ) RNA , and the corresponding ssDNA and dsDNA molecules ( Fig 5C ) . ANDV L1–200 N167A and K44A were most active on ssRNA , though they can also less efficiently cleave dsRNA . None of the mutants was active on ssDNA or dsDNA , indicating that the enzyme is a ribonuclease . Furthermore , we compared the ribonuclease activity on ssRNA with different sequence , structure , and length ( Fig 5D ) . Comparison of the 27mer RNA with hairpin-like structure vs . a 27mer unstructured poly ( A ) RNA reveals preference for the unstructured substrate . In addition , longer RNAs are degraded faster than shorter RNAs . The negative control ( L1–200 D97E ) did not show background degradation , demonstrating that our purification procedure removes potentially contaminating bacterial nucleases . The metal-dependent cap-snatching endonuclease is an attractive target for antiviral drugs against segmented negative strand RNA viruses and promising compounds have been described for influenza [24–35] . To proof suitability of the ANDV endonuclease expressed in this study for screening of potential inhibitors , we first measured the effect of the classical influenza virus endonuclease inhibitor 2 , 4-dioxo-4-phenylbutanoic acid ( DPBA ) [33] on thermal stability of ANDV L1–200 mutants H36R , K44A , N50A , D97E , E110A , K125A , K127A , and N167A in the presence of 16 mM MnCl2 . All mutants that showed a stabilization of > 8°C in the presence of manganese , also showed a further increase in the melting temperature upon addition of DPBA ( Fig 6A ) , indicating that DPBA binds to and stabilizes the manganese–active site complex . In agreement with this conclusion , mutation of residues His36 , Asp97 and Glu110 , which are involved in coordination of at least one manganese ion , abolishes the stabilizing effect of DPBA . To prove that DPBA not just binds to the active site of ANDV L1–200 , but also inhibits the ribonuclease activity , we incubated the most active mutant ANDV L1–200 N167A with Mn2+ and DPBA for 15 min prior to the addition of the RNA substrate . Fig 6B and 6C show a clear dose-dependent inhibition of the enzymatic activity comparable to IAV and LACV [16 , 20] .
This study provides structural and biochemical evidence for the existence of an endonuclease in the N terminus of ANDV L protein . In agreement with previous studies on expression of L protein in mammalian cells [36 , 37] , we have not been able to express the wild-type endonuclease domain in bacterial cells . Therefore , we took advantage of a range of mutations to facilitate expression of the domain . By using this approach , we expressed large amounts of various mutants and eventually solved the crystal structure of the ANDV L1–200 K127A mutant . The first structure of a hantavirus cap-snatching endonuclease reveals similarities to the related enzymes from LACV , LCMV , and IAV [16 , 18 , 20–23] . Besides the active site residues the sequence is hardly conserved . However there are obvious common structural features present: all cap-snatching endonucleases consist of two lobes with the conserved active site buried in a cavity in between . However , compared to the other enzymes , ANDV has positively charged patches surrounding the active site . The positively charged surface may increase the affinity of the enzyme for negatively charged RNA substrates . Whether or not this feature plays a role in the apparently strong endonuclease activity upon recombinant expression remains to be determined . The structure in conjunction with the thermal stability and enzymatic activity data of the attenuated mutants allow us to speculate on the contribution of individual residues for enzymatic activity , substrate binding , and overall stability of the tertiary structure ( Table 1 ) . The data demonstrate that residues His36 , Asp97 , and Glu110 coordinate the manganese ion and are essential for the catalytic process . All three residues act in a cooperative manner , as removal of one residue is sufficient to prevent binding of both metal ions and inactivate the enzyme . The two lysine residues at positions 124 and 127 in the active site may stabilize the attacking hydroxide nucleophile during catalysis and/or bind to the RNA substrate , as speculated previously for IAV and LACV [16 , 20] . Mutation of both lysines does not affect the stabilizing effect of DPBA , suggesting they play no essential role in metal binding . The two residues Tyr32 and Arg35 , which coordinate a glycerol ligand in the crystal structure , may also be involved in substrate binding . Arg35 seems to be of particular relevance , as it is conserved in all bunyavirus L proteins and the L1–200 R35H mutant is largely defective , although it clearly binds manganese in the thermal stability experiments . Tyr32 seems to be less important , as the L1–200 Y32V mutant behaves like the wild-type protein and is toxic to E . coli cells . Interestingly , several attenuating mutations ( D40E , I43A , K44A , N50A , N98A , and N167A ) seem to slightly destabilize the protein , which might affect the enzymatic activity of the enzyme . The corresponding wild-type residues are involved in hydrogen bonds with other side or main chain atoms or in hydrophobic interactions with neighboring secondary structure elements . Consistent with the structural data , most of these mutants are characterized by low temperature stability and high residual enzymatic activity . Residue Asn167 is exceptional as it is located distant to the active site . Via hydrogen bonds with β-sheet βc it positions a small helix ( αe ) that is not present in the otherwise similar LACV protein . Consistent with the peripheral location of Asn167 , the corresponding mutant is the least attenuated . Its wild-type like features include the highest enzymatic activity and a reduced growth rate of the expressing E . coli cells . Given the strong activity of an attenuated version of the enzyme on various RNA templates , it is plausible that the wild-type enzyme activity is too toxic for high-level expression in pro- and eukaryotic cells . Hantaviruses must have an intricate mechanism to control the activity of the endonuclease in infected cells or to keep the concentration of the L protein at a sub-toxic level . The biochemical data also revealed a clear co-factor and substrate specificity of the ANDV endonuclease . Consistent with its presumed function as a cap-snatching enzyme , it preferred ssRNA over dsRNA substrates and was inactive on ssDNA or dsDNA . RNA length and secondary structure seem to influence substrate binding . Dependence of enzymatic activity from manganese rather than magnesium has also been observed for the endonucleases of the other segmented negative strand viruses [16 , 20 , 21] . However , in contrast to LACV and IAV with two metal ions in the active site [16 , 20] , the ANDV crystal structure contains only a single manganese ion ( equivalent to Mn1 ) [21] , while the LCMV structure does not contain any ions in the active site . The reason for this discrepancy may be explained by our crystallization conditions and thermal stability data . At 2 mM manganese , the active site in the crystals contained a single ion . However , we observed an additional increase in thermal stability when increasing the manganese concentration from 4 mM to 16 mM , suggesting the binding of an additional ion ( equivalent to Mn2 ) at higher manganese concentration . This is in agreement with findings by Reguera et al . 2016 [39] for related endonucleases showing a high binding affinity for the first manganese ion ( Mn1 ) , but low affinity for the second manganese ion ( Mn2 ) . It is reasonable to propose that ANDV and other hantaviruses share the same two-metal dependent catalytic mechanism with LACV and IAV [16] . This study also has implications for development of antiviral drugs to treat hantavirus infections . The influenza virus endonuclease has been used as a target in antiviral drug discovery programs and several specific inhibitors have been found [24–35] . Likewise , the ANDV endonuclease is an attractive target for antiviral drug development . The experiments with the known endonuclease inhibitor DPBA [33] provide proof-of-concept that the ribonuclease activity of the enzyme is amenable to compound screening . The highly active ANDV L1–200 N167A mutant is the most promising enzyme candidate for this purpose .
The cDNA of fragments encoding residues 1–163 , 1–179 , 1–191 , 1–194 , 1–197 , 1–200 , 1–211 , 1–214 and 1–228 from the L protein of ANDV ( strain Chile 9717869 , GenBank accession no . AF291704 ) were amplified by PCR using the pCITE-ANDV-L plasmid [37] as a template . Virus sequences were cloned into pOPIN-F ( N-terminal His-tag–3C protease cleavage site ) and pOPIN-M ( N-terminal His-tag–maltose binding protein–3C protease cleavage site ) [41] . L protein mutants ( Y32V , R35H , H36R , D37A , D40E , I43A , K44A , N50A , P96A , D97E , N98A , E110A , K124A , K127A and N167A ) were generated via a classical two-step PCR mutagenesis approach as described previously [42] and the resulting fragments cloned into pOPIN-F . For cloning , E . coli strain DH5α was used . The sequence of the inserts was confirmed by sequencing . Proteins were expressed in E . coli strain BL21 Gold ( DE3 ) ( Novagen ) at 17°C overnight using LB medium and 0 . 5 mM isopropyl-β-D-thiogalactopyranosid for induction . Cells were pelleted , resuspended in lysis buffer containing 50 mM Tris-HCl pH 7 . 3 , 300 mM NaCl , 10 mM imidazole , 10% glycerol , 10 mM MnCl2 , and 1 mM phenylmethylsulfonyl fluorid and disrupted by sonication . The soluble material was loaded onto a Ni-NTA affinity column , washed with 10 column volumes of lysis buffer containing 50 mM imidazole , and eluted with 5 volumes of lysis buffer containing 500 mM imidazole . The His-tag of eluted protein was cleaved off using glutathione S-transferase-tagged 3C protease at 4°C overnight during dialysis against lysis buffer . The untagged proteins were further purified by size exclusion chromatography using a Superdex 200 column in 50 mM sodium citrate pH 5 . 5 , 1 M NaCl , 5% glycerol . Purified proteins were concentrated using centrifugal devices , flash frozen in liquid nitrogen , and stored in aliquots at –20°C . ANDV L1–200 K127A ( 10 mg/ml ) was crystallized in the presence of 2 mM MnCl2 , 100 mM sodium acetate pH 5 . 0 , 27 . 5% polyethylene glycol 3350 , and 500 mM ( NH4 ) 2SO4 by sitting drop vapor diffusion . Crystals were flash frozen in liquid nitrogen with 8% butanediol as cryo protectant and diffraction data were collected to 2 . 4 Å at beamline ID23-1 at the European Synchrotron Radiation Facility , Grenoble , with a wavelength of 0 . 98 Å and 1 . 77 Å for respectively native and anomalous data . Datasets were processed with iMosflm [43] . The crystal structure was solved in space group P4212 by molecular replacement using residues 32–162 from the endonuclease from Hantaan virus [39] and PHASER [44] , and refined by iterative cycles of manual model building in Coot [45] and computational optimization with PHENIX [46] . TLS refinement was performed during the final stages of refinement [47] with the entire molecule as rigid group . Data collection and refinement statistics are shown in the crystallographic table in the supplemental information . Structural data were visualized with Pymol and Chimera [48] . Electrostatic surfaces were calculated using PDB2PQR and APBS [49 , 50] . The coordinates have been deposited to the PDB ( 5HSB . pdb ) . Small angle X-ray scattering ( SAXS ) data for 1 , 2 , and 5 mg/ml of ANDV L1–200 K127A in 2 mM MnCl2 , 50 mM Tris-HCl pH 7 . 3 , 250 mM NaCl , 5% glycerol were collected at the SAXS beamline P12 at the PETRA III storage ring of the Deutsches Elektronen-Synchrotron , Hamburg [51] . Using a PILATUS 2M pixel detector at 3 . 1 m sample distance and 10 keV energy ( λ = 1 . 24 Å ) , a momentum transfer range of 0 . 01 Å–1 < s < 0 . 45 Å–1 was covered ( s = 4π sinθ/λ , where 2θ is the scattering angle ) . Data were analyzed using the ATSAS 2 . 6 package [52] . The forward scattering I ( 0 ) and the radius of gyration Rg were extracted from the Guinier approximation calculated with the AutoRG function within PRIMUS [53] . GNOM [54] provided the pair distribution function P ( r ) of the particle and the maximum size Dmax . Ab initio reconstructions were generated with the program DAMMIF [55] . Ten independent DAMMIF runs were superimposed by SUPCOMB [56] and averaged using the program DAMAVER [55] . Nuclease activity was measured by incubating 1 μM ANDV L1–200 protein with 0 . 1 μM 32P-labeled single stranded ( ss ) or double stranded ( ds ) 27mer RNA or DNA substrate ( 5’-GAU/TGAU/TGCU/TAU/TCACCGCGCU/TCGU/TCGU/TC-3’ ) or 40mer ssRNA substrate ( 5´-GAUGAUGCUAUCACCGCGCUCGUCGUCGAUGAUGCUAUCA-3’ ) in 50 mM Tris-HCl pH 7 . 3 , 250 mM NaCl , 5% glycerol , 0 . 25 U/μl RNasin ( Promega ) in the absence or presence of 2 mM MnCl2 at 37°C for 1 or 2 h . RNA was annealed by heating to 98°C for 2 minutes , followed by a slow cooling over 2 hours to room temperature . A 27 or 40mer poly ( A ) ssRNA was used as unstructured substrate . In selected experiments , 2 , 4-dioxo-4-phenylbutanoic acid ( DPBA ) was added to the assay . The reaction was stopped by adding 2 × loading buffer ( 95% formamide , 18 mM ethylenediaminetetraacetic acid ( EDTA ) , 0 . 025% sodium dodecyl sulfate , xylene cyanol , and bromophenol blue ) and heating the samples to 98°C for 5 min . The reaction products were separated by 8 M urea , 20% polyacrylamide , Tris-borate-EDTA gel electrophoresis and visualized by phosphor screen autoradiography using a Typhoon scanner ( GE Healthcare ) . Intensity of the signals was quantified by ImageJ software [57] . The stability of ANDV L1–200 protein was measured by thermofluor assay at a protein concentration of 10 μM in 100 mM Tris pH 7 . 0 , 250 mM NaCl , 5% glycerol complemented with either 10 mM EDTA , 4 mM MnCl2 , 16 mM MnCl2 , 16 mM MgCl2 or 16 mM MnCl2 plus 100 μM DPBA as described [38] .
|
Hantaviruses may cause severe disease in humans , either hemorrhagic fever or cardiopulmonary syndrome . Both conditions are associated with high lethality . Vaccines and effective treatments approved for application in humans are not available . The L protein plays a central role in hantavirus replication and represents a key target for antiviral drug development . However , the protein is difficult to study , as it can hardly be expressed in cells , and therefore structure and function are largely unknown . We succeeded to express and purify the N-terminal domain of ANDV , a South American hantavirus , in mutant forms . The atomic structure of the domain was solved by X-ray crystallography . The structural data in conjunction with our enzymatic activity data demonstrate that the N-terminal part of ANDV L protein harbors an endonuclease that cleaves RNA , presumably to “steal” cap structures from cellular mRNA for virus transcription . The investigation of a range of mutant enzymes revealed the detailed role of amino acid residues within the endonuclease . Importantly , using a known inhibitor of viral endonucleases we demonstrate the potential of the ANDV enzyme as an antiviral target . The discovery of a highly active mutant form of the endonuclease , which can be expressed at high level , will facilitate drug discovery programs for hantaviruses .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"medicine",
"and",
"health",
"sciences",
"crystal",
"structure",
"pathology",
"and",
"laboratory",
"medicine",
"manganese",
"pathogens",
"condensed",
"matter",
"physics",
"microbiology",
"enzymology",
"orthomyxoviruses",
"viruses",
"organisms",
"hantavirus",
"monomers",
"(chemistry)",
"rna",
"viruses",
"andes",
"virus",
"crystallography",
"thermodynamics",
"bunyaviruses",
"polymer",
"chemistry",
"influenza",
"a",
"virus",
"solid",
"state",
"physics",
"medical",
"microbiology",
"microbial",
"pathogens",
"chemistry",
"thermal",
"stability",
"physics",
"biochemistry",
"enzyme",
"structure",
"influenza",
"viruses",
"viral",
"pathogens",
"glycerol",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"chemical",
"elements"
] |
2016
|
Atomic Structure and Biochemical Characterization of an RNA Endonuclease in the N Terminus of Andes Virus L Protein
|
Elucidating virus-host interactions responsible for HIV-1 transmission is important for advancing HIV-1 prevention strategies . To this end , single genome amplification ( SGA ) and sequencing of HIV-1 within the context of a model of random virus evolution has made possible for the first time an unambiguous identification of transmitted/founder viruses and a precise estimation of their numbers . Here , we applied this approach to HIV-1 env analyses in a cohort of acutely infected men who have sex with men ( MSM ) and found that a high proportion ( 10 of 28; 36% ) had been productively infected by more than one virus . In subjects with multivariant transmission , the minimum number of transmitted viruses ranged from 2 to 10 with viral recombination leading to rapid and extensive genetic shuffling among virus lineages . A combined analysis of these results , together with recently published findings based on identical SGA methods in largely heterosexual ( HSX ) cohorts , revealed a significantly higher frequency of multivariant transmission in MSM than in HSX [19 of 50 subjects ( 38% ) versus 34 of 175 subjects ( 19% ) ; Fisher's exact p = 0 . 008] . To further evaluate the SGA strategy for identifying transmitted/founder viruses , we analyzed 239 overlapping 5′ and 3′ half genome or env-only sequences from plasma viral RNA ( vRNA ) and blood mononuclear cell DNA in an MSM subject who had a particularly well-documented virus exposure history 3–6 days before symptom onset and 14–17 days before peak plasma viremia ( 47 , 600 , 000 vRNA molecules/ml ) . All 239 sequences coalesced to a single transmitted/founder virus genome in a time frame consistent with the clinical history , and a molecular clone of this genome encoded replication competent virus in accord with model predictions . Higher multiplicity of HIV-1 infection in MSM compared with HSX is consistent with the demonstrably higher epidemiological risk of virus acquisition in MSM and could indicate a greater challenge for HIV-1 vaccines than previously recognized .
An effective sterilizing HIV-1 vaccine ideally should target virus in the earliest stages of transmission , prior to dissemination and establishment of persistent infection [1] , [2] , [3] , [4] . To be broadly protective , such a vaccine must defend against a genetically diverse set of viruses transmitted by different sexual practices and risk behaviors . Results from the recently reported ‘Thai Trial’ RV144 of an experimental HIV-1 vaccine showed a decrease in virus acquisition of 31 . 2% ( p = 0 . 04 ) based on a modified intention-to-treat analysis and a trend for greater vaccine effectiveness in those subjects identified as practicing lower risk behaviors [5] . These findings suggest that an HIV-1 vaccine might be more efficacious in preventing infection by some exposure routes than others [5] , [6] , [7] . Recently , we and others employed SGA , direct sequencing , and a model of random virus evolution to identify those viruses responsible for transmission and productive clinical infection in several largely heterosexual cohorts with acute HIV-1 subtype A , B or C infection [8] , [9] , [10] , [11] , [12] and in Indian rhesus macaques inoculated intra-rectally with SIVmac251 or SIVsmmE660 [13] . This experimental approach allows for the distinction of transmitted/founder viruses that differ by as little as a single nucleotide [10] , [13] . SGA-direct sequencing also makes possible the identification of transmitted viral sequences in linked transmissions , thereby enabling the unambiguous tracking of viruses from donor to recipient across mucosal surfaces [9] , [13] , and the molecular cloning and analysis of those viruses actually responsible for productive clinical infection [12] . Previous studies based on different experimental approaches have been informative with respect to determining the overall extent of viral diversity present in acute and early infection as a surrogate for identifying and quantifying transmitted viruses [14] , [15] , [16] , [17] , [18] , [19] , [20] , [21] , [22] , [23] , [24] . Such studies generally described new infections as being either “homogeneous , ” presumably reflecting infection by one or few viruses , or “heterogeneous , ” suggesting infection by more viruses . Based on these studies , a substantial bottleneck in virus transmission was recognized to exist , since the genetic complexity of viral quasispecies in the blood of chronically infected individuals was generally much greater than that in acutely infected subjects . Evidence for a bottleneck in virus transmission , although not necessarily at the mucosal interface , was further suggested by the longstanding observation that most new infections are caused by R5 tropic viruses and not by X4 tropic viruses , which are common in chronic infection [25] , [26] , [27] . These studies and others in the related Indian rhesus macaque-SIV infection model [13] , [28] , [29] , [30] , [31] thus focused attention on the mucosa and submucosa as a potentially important barrier to HIV-1 transmission and a site where critical early virus-host cell interactions leading to transmission and productive clinical infection likely take place [1] , [2] , [3] , [4] , [25] , [26] , [32] . However , it was not until the application of SGA , direct amplicon sequencing , and a model of random virus evolution to the analysis of viral genomes in the acute infection period that actual transmitted/founder viruses could be identified and their numbers precisely estimated [10] , [13] . In the present study , we have applied this strategy to a systematic analysis and comparison of multiplicity of HIV-1 infection in men who have sex with men ( MSM ) versus heterosexuals ( HSX ) .
SGA-direct sequencing was used to identify and enumerate transmitted/founder env sequences in 28 acutely infected MSM subjects who reported homosexual exposure as their primary HIV-1 risk behavior and who denied injection drug use ( IDU ) ( Table 1 ) . At the time of study , 14 subjects were HIV-1 ELISA negative/western immunoblot ( WB ) negative ( Fiebig stage II ) , 2 were ELISA+/WB− ( Fiebig stage III ) , 6 were ELISA+/WB indeterminate ( Fiebig stage IV ) and 6 were ELISA+/WB+/p31− ( Fiebig stage V ) [10] , [33] . Subjects were identified based on clinical symptoms of an acute retroviral syndrome , routine HIV testing in a health care setting , or contact tracing of an HIV-1 infected index case . Clinical histories of sexually transmitted diseases were not available . Envelope sequences from chronically infected sexual partners of two acutely infected subjects were also evaluated . A total of 1307 full-length env genes encoding gp160 were sequenced from plasma vRNA ( median of 40 sequences per subject; range 23–89 ) . In a composite neighbor-joining ( NJ ) phylogenetic tree ( Fig . 1 ) , viral sequences formed distinct patient-specific monophyletic lineages , each with high statistical support . Sequences from known sexual partners , including two acute-to-acute ( AD77 to AD75 and AD83 to 04013240 ) and two chronic-to-acute ( LACU9000 to HOBR0961 and AD18 to AD17 ) transmission pairs , also clustered significantly together ( Fig . 1 ) . All sequences were HIV-1 subtype B . Among the 28 acutely infected subjects , maximum within-patient env diversities ranged from 0 . 12% to 6 . 82% ( Table 2 ) . Sequences from 22 of these individuals had distinctly lower env diversities ( <0 . 75% ) compared with env diversities from six others ( >1 . 25% ) . The latter diversity is inconsistent with single virus transmission within the time frame of acute and early infection ( Fiebig stage I–V ) [8] , [9] , [10] , [34] , while env diversity <0 . 75% is consistent either with single variant transmission or with transmission of two or more closely related viruses . Phylogenetic and Highlighter analyses of env sequences distinguished between these possibilities for each subject ( Fig . 2; see also www . hiv . lanl . gov/content/sequence/HIV/USER_ALIGNMENTS/Li ) . Fig . 2A shows sequences from a subject ( 04013440 ) who was infected by a single virus , Fig . 2B a subject ( 04013211 ) infected by two viruses differing by only 4 nucleotides out of 2619 ( 0 . 15% ) , Fig . 2C a subject ( 04013383 ) infected by two viruses differing by 65 of 2547 nucleotides ( 2 . 55% ) , and Fig . 2D a subject ( 04013448 ) infected by four viruses differing by as many as 47 of 2655 nucleotides ( 1 . 79% ) with additional sequences showing recombination between the transmitted/founder lineages . Altogether , we determined that 10 of 28 subjects ( 36% ) had been productively infected by more than one virus ( Table 2 ) . We next analyzed the env sequences using a mathematical model of random virus evolution that we previously described [10] , [13] , [34] . Sequences resulting from multivariant transmission , APOBEC hypermutation , early stochastic mutation , selection by cytotoxic T-cells , or recombination violate model predictions ( Table 2 ) [10] , [34] . Once these confounders were accounted for , lineage-specific env sequences from each subject conformed to model predictions and coalesced to most recent common ancestor sequences at or near the time of virus transmission estimated from clinical histories and laboratory staging . These results thus corroborated a large body of evidence supporting the SGA-direct sequencing strategy for identifying transmitted/founder viruses [8] , [9] , [10] , [11] , [12] , [13] , [35] . As an additional test of the model's validity , we asked in subject AD17 whose history of virus exposure was particularly well-documented ( Table 1 and Fig . 3 ) , if plasma vRNA and PBMC viral DNA ( vDNA ) sequences spanning the complete ( 9 . 2 Kb ) viral genome coalesced to the same viral sequence as did env-only sequences and if a molecular clone of this viral genome encoded replication competent virus , as would be expected for an authentic transmitted/founder virus . For this analysis , we used SGA-direct sequencing to determine env-only sequences ( n = 51 ) and overlapping 5′ ( n = 92 ) and 3′ ( n = 96 ) half genome sequences ( Fig . 4 ) . All 239 vRNA and vDNA sequences coalesced to a single transmitted/founder genome in a time frame consistent with the clinical history of virus exposure as recently as 11 days earlier . The estimated time to a most recent common ancestor ( MRCA ) sequence for env-only sequences was 8 days ( 95% CI 5–11 ) and for all sequences 6–11 days ( CI 3–14 ) . MRCA estimates are frequently lower than clinical estimates [10] or experimentally determined intervals between transmission and virus sampling in the rhesus macaque-SIV infection model [13] because of purifying selection or variance in estimated parameters of virus replication [34] . The inferred transmitted/founder viral genome in subject AD17 contained intact LTR-gag-pol-vif-vpr-tat-rev-vpu-env-nef-LTR elements , a finding we have replicated for transmitted/founder viruses from 38 other subjects infected by HIV-1 subtypes A , B , C or D ( [12] and H . L . and G . M . S . , unpublished ) . A proviral clone ( pAD17 . 1 ) of the transmitted/founder viral genome from subject AD17 ( Fig . 5A ) , when transfected into 293T cells , produced virions that were infectious and highly replicative in human CD4+ T-cells , but interestingly , not in monocyte-derived macrophages from the same normal donors ( Fig . 5B ) . pAD17 . 1 virus was CCR5 tropic in JC53BL-13 cells ( Fig . 5C ) and in GHOST ( 3 ) cells [36] , where it infected cells bearing CD4 and CCR5 but not CD4 and CXCR4 ( G . M . S . , unpublished ) . Extremes in HIV-1 diversity in acute infection could be informative regarding biological events underlying virus transmission . Subject 04013171 had the greatest env diversity ( 6 . 82% ) ( Table 2 ) . This subject admitted to unprotected receptive anal intercourse with multiple partners over a single eight hour period four weeks before the onset of flu-like symptoms , consistent with his Fiebig IV staging . Fig . 6 shows a NJ tree and Highlighter plot of 86 plasma derived env sequences , which revealed 10 unique transmitted/founder virus lineages . In addition , 20 inter-lineage recombinants were identified based on shared polymorphisms in the Highlighter plot with corroboration by Recco analysis [37] . Among these recombinant sequences , the Hudson-Kaplan test [38] indicated a minimum of 44 recombination breakpoints . Interestingly , sequences corresponding to 4 of the 10 virus lineages in subject 04013171 were sampled only once . We could be confident that these represented unique transmitted/founder viruses and not recombinants between two or more predominant virus lineages because of the large number of unique nucleotide changes in each sequence that far exceeded the diversity observed empirically [8] , [9] , [10] , [13] or estimated to occur based on mathematical modeling of the first 35 days of infection ( eclipse phase to the end of Fiebig stage IV ) [10] . Power calculations further indicated that with a sample size of 86 sequences , there is a >95% probability of detecting minor sequences representing at least 4% of the population ( see Methods ) . These findings suggest that more extensive sampling might result in the detection of an even greater number of transmitted/founder viruses in this individual . Subject 701010068 had the second highest env diversity ( 4 . 43% ) among the study subjects ( Table 2 ) . He reported a single high risk exposure event involving unprotected receptive anal intercourse with two individuals , one HIV negative and the other HIV positive . He developed flu-like symptoms approximately two weeks later and was studied three weeks after that , again at Fiebig stage IV . Based on the earlier analysis of subject 04013171 ( Fig . 6 ) , we were concerned that viral recombination [39] , [40] , [41] , [42] , which is sequence length and time ( from infection ) dependent [10] , [13] , [40] , could confound the identification of discrete transmitted/founder virus lineages . This could be especially problematic in subjects infected with many different transmitted/founder viruses as opposed to two , since in the former case it is far more likely that doubly or multiply infected cells will spawn heterozygous virus progeny that lead in the next virus generation to recombinant viral genomes [39] , [40] , [41] . To test this hypothesis , we amplified and sequenced seventy-two 3′ half genome segments of plasma vRNA from subject 701010068 and then analyzed env gp41 ( 1035 bp ) , env gp160 ( 2630 bp ) , and 3′ half genome regions ( 4734 bp ) separately . The gp41 sequences ( Fig . 7A ) revealed discrete low diversity lineages comprised of identical or nearly identical sequences . We interpreted 7 of these sequence clusters as likely to have arisen from distinct transmitted viruses and the remaining sequences to represent inter-lineage recombinants . Clusters of identical or nearly identical sequences were also evident in gp160 sequences ( Fig . 7B ) , but with less clarity due to additional inter-lineage recombination events in the longer sequences . For example , sequences corresponding to lineage 4 in the gp41 sequences ( depicted in light blue in Fig . 7A ) were dispersed into five widely separated branches in the gp160 tree due entirely to recombination ( Fig . 3B ) . Similarly , sequences comprising lineage 6 in the gp41 sequences ( depicted in red in Fig . 7A ) were dispersed into three widely separated branches in the gp160 tree , again due entirely to recombination ( Fig . 7B ) . These findings were supported by Hudson-Kaplan analysis [38] , which indicated a minimum of 27 recombination breakpoints among the gp160 env sequences . Interspersion of sequences was even more dramatic in the 3′ half genome tree ( Fig . 7C ) . Remarkably , of the 72 3′ half genome sequences depicted in Fig . 7C , 63 ( 88% ) represented overt recombinants between two or more transmitted/founder lineages demonstrable by visual inspection and by computer-assisted algorithms . Only two ( dark blue ) sequences labeled L7 and L9 at the very top of the tree ( Fig . 7C ) , three ( green ) sequences labeled B1 , L1 and P7 in the middle of the tree , and four ( gray ) sequences labeled B27 , E1 , A2 and J1 at the very bottom of the tree showed no evidence of recombination . These findings , together with corroborating data from env-only sequences [8] , [10] , [11] , lead to the surprising conclusion that by the time of first antibody detection in acute HIV-1 infection ( Fiebig stages III/IV ) , a majority of circulating viruses may be recombinants . This finding is testament to the large number of doubly ( or multiply ) infected cells in acute and early infection and further evidence of the rapidity with which virus diversifies [43] , [44] , making clear that in order to identify non-recombinant transmitted/founder HIV-1 ( or SIV ) genomes [10] , [12] , [13] , it is necessary to characterize viral sequences as close to the transmission event as possible . Four studies , including the present one , have estimated the numbers of viruses responsible for transmission and productive HIV-1 infection after heterosexual or homosexual exposure using identical SGA-direct env amplicon sequencing methods [8] , [9] , [10] . One of these evaluated the frequency of multivariant transmission in a cohort of cohabitating HIV-1 discordant ( antiretroviral naïve ) heterosexual couples in Zambia and Rwanda followed prospectively for HIV-1 transmission [9] . Remarkably , only 2 of 20 [10%; 95% confidence interval ( CI ) 1–32%] of the epidemiologically linked infections resulted from multivariant transmission , a finding attributed to the chronicity of infection in the virus positive partner , lower prevalence of comorbid conditions such as untreated tuberculosis or sexually transmitted infections , and the heterosexual route of transmission . Since the frequency of multivariant transmission in HSX in that study was substantially less than what we observed for MSM [2 of 20 HSX ( 10% , CI 1–32% ) versus 10 of 28 MSM ( 36% , CI 19–56% ) ; Fisher's exact p = 0 . 042 , odds ratio 4 . 85 , 95% CI 1 . 1 - inf] , we performed a combined analysis of data from all four studies , which included 225 patients infected by HIV-1 subtypes A , B or C ( Table 3 ) . Again we found that the proportion of MSM subjects infected by more than one virus was substantially higher than for HSX [19 of 50 ( 38% ) versus 34 of 175 ( 19% ) ; Fisher's exact p = 0 . 008 , odds ratio 2 . 5 , 95% CI 1 . 2–5 . 3] . The MSM subjects were all infected with HIV-1 subtype B; a comparison to only the subset of HSX infections that were subtype B was still significant ( Fisher's exact p = 0 . 01 , odds ratio 2 . 9 , 95% CI 1 . 2–7 . 1 ) . The frequency of multiple infections in HSX was not statistically different among subtypes A , B and C nor was it different between males and females . Differences in the frequency of multivariant HIV-1 transmission in MSM versus HSX could not be accounted for by the numbers of sequences analyzed per subject nor by the clinical stage of subjects at the time of study . In the study by Haaland [9] , the median number of sequences analyzed was 40 , in Keele [10] it was 25 , and in Abraham [8] it was 22 . In the present study , the median number of sequences that we determined as part of our initial survey was 33 ( Table 2 ) . We used this lower number of sequences for statistical comparisons of single and multivariant transmissions in MSM versus HSX subjects in order to allow for comparability among the four studies . When , in this initial sequence set , we identified samples containing more than one transmitted/founder virus lineage , we went on to obtain additional sequences ( as many as 89 ) in order to estimate more precisely the numbers of transmitted/founder viruses ( Table 2 ) . Increasing the numbers of sequences analyzed allowed for greater accuracy and precision in estimating the numbers of viruses transmitted in those individuals with many transmitted viruses ( e . g . , subjects 04013448 , 04013171 and 701010068 in Figs . 2D , 6 and 7 ) , but it did not affect the discrimination between those subjects infected by one virus versus those infected by more than one virus . Finally , we found no significant correlation between the clinical stage of subjects at the time of plasma sampling and the numbers of transmitted/founder viruses identified in those samples: among the four studies , there was a total of 95 antibody negative subjects ( Fiebig stages I–II ) and 130 antibody positive subjects ( Fiebig stages III–VI ) . Twenty subjects ( 21% ) in the former group and 33 subjects ( 25% ) in the latter group had evidence of productive infection by more than one virus , which was not significantly different ( odds ratio 1 . 27; 95% CI 0 . 65–2 . 54; Fisher's exact p = 0 . 53 ) .
Previous studies used less precise methods for estimating multiplicity of HIV-1 infection in HSX and MSM subjects and reported widely varying results with a trend for higher multiplicities in MSM [15] , [16] , [17] , [18] , [19] , [20] , [21] , [22] , [24] . We report here new SGA-based determinations that show significant differences in the multiplicity of virus infection between the two risk groups: MSM were twice as likely as HSX subjects to become infected by more than one virus , with some MSM acquiring as many as 7 to 10 or more viruses . These findings are consistent with the higher epidemiological risk of HIV-1 acquisition in MSM compared with HSX and may be explained in part by the anatomical and immunohistological differences between the male and female genitourinary tracts and the lower intestine . A limitation of the current study is that it represents a retrospective comparison of multivariant HIV-1 transmission among patient cohorts having different enrollment criteria and different behavioral risk assessments . It must be noted , however , that all study subjects from all cohorts were queried extensively with regard to potential HIV-1 infection risk behaviors . This included acutely infected subjects identified by cross-sectional screening methods [45] , subjects enrolled prospectively into HIV-1 discordant couple [9] or Acute Infection Early Disease Research Program cohorts [46] , and source plasma donors who became HIV-1 infected during a period of serial plasma collections [10] . The latter subjects , whom we studied anonymously , underwent exhaustive pre-enrollment interrogation for HIV and IDU risk behaviors according to a standardized FDA-approved protocol ( http://www . fda . gov/BiologicsBloodVaccines/GuidanceComplianceRegulatoryInformation/Guidances/Blood/ucm073445 . htm ) that included a written questionnaire and interview inquiring about MSM and IDU activities , sex-for-money , sex with a partner who had sex-for-money , or sex with an individual known to be HIV positive . Source plasma donors also underwent serial laboratory testing for surrogate laboratory markers that could indicate injection drug use ( e . g . , liver transaminase elevations and hepatitis B or C nucleic acids or antibodies ) , and these markers were negative among qualified donors . Nonetheless , self-reporting of risk behaviors among paid plasma donors is imperfect [47] , and it is possible that some subjects whom we categorized as belonging to the HSX risk group actually had additional risks such as IDU or MSM . However , even if this were the case , it would likely bias the findings in the HSX group toward a greater ( not lesser ) number of transmitted viruses [48] . Not surprisingly , when we excluded all source plasma donor subjects from our comparative analysis of multivariant HIV-1 transmission , the difference between MSM and HSX groups was still statistically significant [19 of 50 ( 38% ) versus 25 of 119 ( 21% ) , respectively; Fisher's exact p = 0 . 03 , odds ratio 2 . 3 , 95% CI 1 . 04–5 . 02] . We conclude that within the limitations of self-reporting and surrogate marker testing , study subjects in the cohorts we examined were correctly assigned to HSX and MSM risk groups , differences in multiplicity of virus transmission between the two groups were significant , and overall study findings were unlikely to have been confounded by injection drug use . Future studies can benefit from a prospective trial design and a common behavioral and medical questionnaire [49] , [50] . It is noteworthy that while multivariant HIV-1 transmission was twice as common in MSM than in HSX , still more than half of MSM subjects showed evidence of productive infection by just one virus . Moreover , the adjusted median ( calculated from subjects with multivariant transmissions only ) was 3 in MSM compared with 2 in HSX ( Table 3 ) . Even in the Fiebig II subject AD17 , where we analyzed a total of 239 sequences ( giving us a 95% probability of detecting a second transmitted/founder virus lineage at 1 . 25% prevalence ) , all of the sequences coalesced phylogenetically to a single virus , thus providing no evidence for transmission of more than one virus . Elsewhere , we have used 454 deep sequencing to analyze tens of thousands of sequences from three additional Fiebig stage II MSM subjects in whom SGA-direct sequencing suggested transmission and productive clinical infection by a single virus ( Will Fischer , B . F . K . , G . M . S . B . T . K . , unpublished ) . Even with this greatly enhanced sensitivity of detection of minor sequences , we found no evidence of transmission by more than one virus in these subjects . Considered together , the findings of the present study , previously published studies [8] , [9] , [10] , and work in progress ( Will Fischer , B . F . K . , G . M . S . , B . T . K ) , all suggest that a substantial proportion of HSX and MSM patients acquire HIV-1 infection as a consequence of transmission and productive infection by literally one virion or one infected cell . The implication of this finding is that in order for a vaccine , microbicide or other prevention modality to be protective in this fraction of individuals , it need only prevent infection by a single virus or infected cell . Conversely , there is another subset of HSX and MSM subjects in whom the multiplicity of infection is higher . Since the proportion of such multiply infected individuals is far higher than would be expected from a Poisson distribution of independent , low frequency events ( see Abrahams [8] for discussion ) , we suspect that biological events underlying virus transmission in these subjects compared with those infected by a single virus are different and that challenges faced by vaccines and microbicides in the higher multiplicity infection group may be higher . Another interesting observation from the present study relates to viral recombination . Although recombination was not a primary study objective , the identification of two or more transmitted/founder genomes in acutely infected subjects gave us a unique opportunity to examine the dynamics and extent of recombination in primary HIV-1 infection . Five features of our study distinguish it from previous reports of HIV-1 recombination [39] , [40] , [41] , [42] . First , we studied subjects at very early clinical stages following virus transmission ( Fiebig stages II–V ) . Second , we used SGA-direct amplicon sequencing , which provides for a proportional representation of virus present in the plasma , including those that are recombinant [11] . Third , SGA eliminates in vitro recombination artifacts resulting from Taq polymerase-mediated template switching [11] . Fourth , SGA allowed us to identify the exact nucleotide sequences of full-length transmitted/founder virus env genes unambiguously and to distinguish these viruses and their progeny from viruses that contained even short regions of recombinant sequence . Fifth , SGA-direct sequencing of 3′ half genomes allowed us to examine recombination across the boundaries of vif-vpr-tat-rev-vpu-env-nef-LTR . Figs 2D , 6 and 7 illustrate examples of recombination and Table 2 summarizes the findings in all multiply infected subjects . Seven of 9 subjects had evidence of recombination within gp160 env ( one subject , AD77 , could not be analyzed because of excessive virus diversity at a late Fiebig stage ) . The proportion of recombinants ranged from 0 of 30 sequences in subject 04013211 to 30 of 72 sequences ( 42% ) in subject 701010068 . In the latter subject , we amplified a longer fragment of the viral genome so as to include the 3′ half; this allowed us to compare recombination frequencies within gp41 ( only ) , gp160 ( only ) or the full-length 3′ half genome . The proportion of recombinants in these three regions was 13/72 ( 18% ) , 30/72 ( 42% ) and 63/72 ( 88% ) , respectively . Recombination breakpoints were more common in sequences flanking gp160 env than within env ( Fig . 7C ) , a finding similar to that reported by Simon-Loriere and colleagues for HIV-1 inter-subtype recombination [42] . In subject 701010068 , where 88% of sequences corresponding to only half the viral genome were recombinant , it is likely that nearly all of the full-genome sequences at this time point are recombinant . Since recombination requires an earlier infection event in which a cell is infected by two or more viruses , our findings suggest that in acutely infected humans at or near antibody seroconversion ( Fiebig stages III/IV ) , a substantial fraction of productively infected cells are infected by more than one virus , a circumstance undoubtedly facilitated by initially high virus loads at a time when target cell availability is rapidly declining [51] . A final unique aspect to our study was its in-depth analysis of early virus replication kinetics ( Fig . 3 ) and diversification ( Fig . 4 ) in subject AD17 who was exposed to HIV-1 by receptive anal intercourse approximately 6 days before developing symptoms of the acute retroviral syndrome and 14–17 days before peak plasma viremia of 47 , 600 , 000 RNA molecules/ml . This exposure to HIV-1 was through a new sexual partner ( AD18 ) whom we could prove by phylogenetic analysis was the source of subject AD17's acute HIV-1 infection ( Fig . 1 ) . Assuming a plasma viral load ( vL ) of 10 RNA copies/ml at the time of symptom onset 6 days after virus infection , then during the period between days 6 and 14 , vL increased by a factor of ∼106 . This implies virus grew exponentially with growth rate r = 1 . 73/day , i . e . exp ( 1 . 73*8 ) ∼106 . This expansion rate is slower than the expansion rate calculated by Little [52] of 2 . 0/day but similar to that reported by Stafford [53] of 1 . 67/day . Subject AD17 began HAART on day 17 , and between days 17 and 25 , vL fell approximately 200-fold . Assuming HAART is nearly 100% effective [54] , then the productively infected cell death rate , δ , can be calculated from the rate of vL decline as ln ( 200 ) /8 = 0 . 66/day . These values can then be used to estimate R0 , the basic reproductive number , as ( 1+ r/δ ) exp ( rτ ) , where τ is the intracellular delay phase . If we ignore the delay phase , then R0 = ( 1+ r/δ ) and the estimate of R0 is 3 . 6 . However , if we include the delay phase and assume τ is one day , then R0 = 20 . 4 . This is larger than the estimates in Stafford [53] . These data support the basic assumptions used in the development of our model of early HIV-1 evolution [10] , [34] , and the genomic integrity and replication competence of the full-length proviral clone pAD17 . 1 provide further corroboration of the model . Only four other transmitted/founder virus molecular clones have been described ( [12] and J . S . G . and G . M . S . , unpublished ) , and all of these correspond to HIV-1 subtype C viruses resulting from heterosexual transmissions . With the addition of the pAD17 . 1 clone , we now have molecular proviral clones representing male-to-male rectal transmission ( pAD17 . 1 ) , male-to-female vaginal transmission ( pZM246F-10; pZM247Fv1; pZM247Fv2 ) , and female-to-male penile transmission ( pZM249M-1 ) . All of these viruses are R5 tropic , replicate efficiently in activated human CD4+ T cells , but fail to replicate efficiently in monocyte-derived macrophages . Such molecular clones of transmitted/founder viruses should represent a rich resource for studying the biology of HIV-1 transmission and its prevention . In summary , the findings presented here provide for the first time a comparative , quantitative view of the HIV-1 transmission event in two patient risk groups that dominate the HIV-1 pandemic . In doing so , they highlight both challenges and opportunities confronting candidate vaccines , microbicides , and other prevention modalities . Elucidation of the biological basis of single versus multivariant transmission in MSM and HSX could help advance prevention strategies [55] , [56] , [57] , with quantitative analyses of transmitted/founder viruses representing a potentially valuable new endpoint in vaccine and microbicide trial design and assessment [5] , [6] , [49] , [50] .
This study was conducted according to the principles expressed in the Declaration of Helsinki . It was approved by the Institutional Review Boards of the University of Alabama at Birmingham , Rockefeller University , Duke University , and the University of North Carolina . All patients provided written informed consent for the collection of samples and subsequent analysis . Blood samples were obtained from 28 subjects with acute HIV-1 infection and from chronically infected sexual partners of two of them . Blood specimens were generally collected in acid citrate dextrose and plasma separated and stored at −70°C . PBMCs were stored in vapor phase liquid nitrogen . Plasma samples were tested for HIV-1 RNA , p24 antigen , and viral specific antibodies by a battery of commercial tests . These included quantitative Chiron bDNA 3 . 0 or Roche Amplicor vRNA assays; Coulter or Roche p24 Ag assays; Genetic Systems Anti-HIV-1/2 Plus O; and Genetic Systems HIV-1 Western Blot Kit . Based on these test results , subjects were staged according to the Fiebig classification system for acute and early HIV-1 infection [33] . For each sample , approximately 20 , 000 viral RNA copies were extracted using the Qiagen BioRobot EZ1 Workstation with EZ1 Virus Mini Kit v2 . 0 ( Qiagen , Valencia , CA ) . RNA was eluted in 60 ul of elution buffer and subjected to first strand cDNA synthesis immediately by using the SuperScript III Reverse Transcriptase according to manufacturer's instructions ( Invitrogen Life Technologies ) . Each first strand synthesis reaction included ∼10 , 000 or fewer vRNA molecules , 1X reverse transcription buffer , 0 . 5 mM of each dNTP , 5 mM DTT , 2 units/ul of RnaseOUT , 10 units/ul of SuperScript III reverse transcriptase and 0 . 25 uM of antisense primer . The cDNA syntheses were performed using antisense primers located at different genomic regions . The primers for synthesizing the cDNA of env , 5′ half genome ( U5 , gag and pol ) and 3′ half genome ( vif , vpr , tat , rev , vpu , env , nef , U3 and R ) were env3out 5′-TTGCTACTTGTGATTGCTCCATGT-3′ , 1 . int . R1 5′-CTTGCCACACAATCATCACCTGCCAT-3′ and 1 . R3 . B3R 5′-ACTACTTGAAGCACTCAAGGCAAGCTTTATTG-3′ , respectively . The reactions were incubated at 50°C for 60 min , followed by 55°C for an additional 60 min incubation . The reaction was heat-inactivated at 70°C for 15 min , and then treated with RNaseH at 37°C for 20 min . The synthesized cDNA was subjected to 1st round PCR immediately or stored frozen at −80°C . Blood was collected from subject AD17 14–17 days following exposure to HIV-1 at Fiebig stage II . Genomic DNA was extracted from 1 . 3 million PBMCs using Qiagen Tissue DNA Extraction kit according to manufacturer's instructions . cDNA or genomic DNA was serially diluted and distributed in replicates of 8 PCR reactions in MicroAmp 96-well plates ( Applied Biosystems , Foster City , CA ) so as to identify a dilution where PCR positive wells constituted less than 30% of total number of the reactions . At this dilution , most wells contain amplicons derived from a single cDNA molecule . Additional PCR amplifications were performed using this dilution in 96-well reaction plates . PCR amplification was carried out in presence of 1x High Fidelity Platinum Taq PCR buffer , 2 mM MgSO4 , 0 . 2 mM each deoxynucleoside triphosphate , 0 . 2 uM each primer , and 0 . 025 units/ul of Platinum Taq High Fidelity polymerase in a 20-ul reaction ( Invitrogen , Carlsbad , CA ) . The nested primers for generating different genomic segments included: ( 1 ) full length env: 1st round sense primer env5out 5′-TAGAGCCCTGGAAGCATCCAGGAAG-3′ , 1st round antisense primer env3out 5′- TTGCTACTTGTGATTGCTCCATGT-3′ , 2nd round sense primer env5in 5′-TTAGGCATCTCCTATGGCAGGAAGAAG-3′ and 2nd round antisense primer env3in 5′-GTCTCGAGATACTGCTCCCACCC-3′; ( 2 ) 5′ half genome: 1st round sense primer 1 . U5 . F1 5′- CCTTGAGTGCTTCAAGTAGTGTGTGCCCGTCTGT-3′ , 1st round antisense primer 1 . int . R1 5′-CTTGCCACACAATCATCACCTGCCAT-3′ , 2nd round sense primer 2 . U5 . F2 5′-GTAGTGTGTGCCCGTCTGTTGTGTGACTC-3′ and 2nd round antisense primer 2 . int . R2 5′-CAATCATCACCTGCCATCTGTTTTCCATA-3′; ( 3 ) 3′ half genome: 1st round sense primer 1 . int . F1 5′- ACAGCAGTACAAATGGCAGTATT-3′ , 1st round antisense primer 1 . R3 . B3R 5′- ACTACTTGAAGCACTCAAGGCAAGCTTTATTG-3′ , 2nd round sense primer 2 . int . F2 5′-TGGAAAGGTGAAGGGGCAGTAGTAATAC-3′ and 2nd round antisense primer 2 . R3 . B6R 5′- TGAAGCACTCAAGGCAAGCTTTATTGAGGC-3′ . PCR parameters were as follows: 94°C for 2 min , followed by 35 cycles of 94°C for 15 s , 58°C for 30 s , and 68°C for 4 min ( env ) or 5 min ( 5′ or 3′ half genomes ) , followed by a final extension of 68°C for 10 min . The product of the first-round PCR was subsequently used as a template in the second-round PCR under same conditions but with a total of 45 cycles . The amplicons were inspected on precast 1% agarose E-gel 96 ( Invitrogen Life Technologies , Carlsbad , CA ) . All PCR procedures were carried out under PCR clean room conditions using procedural safeguards against sample contamination , including pre-aliquoting of all reagents , use of dedicated equipment , and physical separation of sample processing from pre- and post-PCR amplification steps . Amplicons were directly sequenced by cycle-sequencing using BigDye Terminator chemistry and protocols recommended by the manufacturer ( Applied Biosystems , Foster City , CA ) . Sequencing reaction products were analyzed with an ABI 3730xl genetic analyzer ( Applied Biosystems; Foster City , CA ) . Both DNA strands were sequenced using partially overlapping fragments . Individual sequence fragments for each amplicon were assembled and edited using the Sequencher program 4 . 8 ( Gene Codes; Ann Arbor , MI ) . All chromatograms were inspected for sites of mixed bases ( double peaks ) , which would be evidence of priming from more than one template or the introduction of PCR error in early cycles . Any sequence with evidence of double peaks was excluded from further analysis . All the sequence alignments were initially made with ClustalW and then hand-checked using MacClade 4 . 08 to improve the alignments according to the codon translation . Consensus sequences were generated for each individual . The full sequence alignment is available as a supplemental data file ( www . hiv . lanl . gov/content/sequence/HIV/USER_ALIGNMENTS/Li ) and sequences are deposited in GenBank ( accession numbers: GU330247–GU331770 ) . Complete env sequences ( n = 1307 ) were derived from 30 individuals , and 5′ ( U5 , gag and pol ) and 3′ ( vif , vpu , tat , rev , env , nef , U3 , and R ) half genome sequences ( n = 188 ) were derived from PBMC and plasma at two different time points from subject AD17 . We analyzed sequences for maximum sequence diversity and then visually inspected each set of sequences using neighbor-joining ( NJ ) and Highlighter tools ( www . hiv . lanl . gov ) . Phylogenetic trees were generated by the neighbor-joining method using ClustalW or PAUP . Enrichment for APOBEC3G/F mutations violates the assumption of constant mutation rate across positions as the editing performed by these enzymes are base and context sensitive . Enrichment for mutations with APOBEC3G/F signatures was assessed using Hypermut 2 . 0 ( www . hiv . lanl . gov ) . Sequences that yielded a p-value of 0 . 05 or lower were considered significantly hypermutated and excluded from subsequent analyses . To obtain an infectious molecular clone of the transmitted/founder virus of subject AD17 , we amplified overlapping 5′ and 3′ half genomes from proviral DNA of earliest sample ( day 14; 6/11/99 ) by single round PCR using Phusion Hot Start High-Fidelity DNA polymerase ( Biolabs ) . Both fragments contained a complete LTR element and an overlap of 170 base pairs encompassing a unique SalI restriction site . For the 5′ half genome , U3-R-U5 , gag , pol , vif , vpr and tat1 was amplified . For the 3′ half genome , tat1 , rev1 , vpu , env , nef , tat2 , rev2 and U3-R-U5 was amplified . The primers were designed to complement exactly the confirmed transmitted/founder sequence as determined by SGA-direct amplicon sequencing . The recognition sequences of MluI and NotI restriction enzymes were appended to the 5′ ends of the sense and antisense primers , respectively . Single round bulk PCR amplifications were carried out in the presence of 1X Phusion Hot Start HiFi buffer , 0 . 2 mM of each deoxynucleoside triphosphate , 0 . 5 uM of each primer , 3% final concentration of DMSO , and 0 . 02 units/ul of Phusion Hot Start High Fidelity polymerase in 50 ul reactions . The PCR product of each half genome was subjected to MluI and NotI digestion and gel purification and then independently cloned into the MluI-NotI site of TOPO XL vector ( Invitrogen ) . The ligation mixture was transformed into XL2 Blue MRF competent cells and plated onto LB agar plates supplemented with 50 ug/ml of kanamycin and grown overnight at 30°C . Single colonies were selected and grown overnight in LB medium with same concentration of kanamycin at 30°C with constant shaking . Plasmid DNA was isolated and sequenced to confirm the identity of transmitted/founder sequences . The 5′ genome half was excised and cloned into 3′ TOPO XL vector by utilizing the MluI and SalI restriction sites thereby generating the full length clone of the transmitted/founder provirus . Replication competency of the full length molecular proviral clone AD17 . 1 was assessed using 293T cells , JC53BL-13 cells ( NIH AIDS Research and Reference Reagent Program catalogue #8129 , TZM-bl ) , activated primary human CD4+ lymphocytes , and monocyte-derived macrophages . Infectious virus stock generation , Env-pseudotyped virus stocks , titrations , cell infections and virus neutralization assays were performed according to methods previously described [12] . Virus controls ( replication competent or Env-pseudotyped ) included the HIV-1 macrophage-tropic strains YU2 and BaL , the non-macrophage tropic T-cell line-adapted strain NL4 . 3 , the dual R5/X4 tropic strain WEAU1 . 60 , and the xenotropic MuLV env [12] . The coreceptor inhibitors TAK779 and AMD3100 were obtained from the NIH AIDS Research and Reference Reagent Program ( 4983 and 8128 ) . R5 and X4 tropism was assessed in both JC53BL-13 cells and in GHOST ( 3 ) cells that stably express CD4 along with CCR5 or CXCR4 or both or neither coreceptor . Recombination was evaluated using GARD [58] and Recco [37] and by visual inspection of Highlighter plots . The minimum number of recombination events required to explain sequence datasets was estimated using the four-gamete method of Hudson and Kaplan [38] as implemented in DNASP v5 . 00 . 07 [59] . Recombinant sequences reported in Table 2 were identified by Highlighter analysis and confirmed by Hudson-Kaplan , GARD and Recco analyses . The model employed in the present study has been described [10] , [12] , [34] as have measured parameters of early virus expansion [52] , [53] , [54] . Under this model , with no selection pressure and fast expansion , one can expect small samples from homogeneous virus populations to have evolved from a founder strain in a star-like phylogeny with all sequences coalescing at the founder [10] , [13] . Occasional deviations from a star phylogeny are , however , expected . The sampling of 10 sequences , for example , from a later generation of an exponentially growing population with six-fold growth per generation ( R0 = 6 ) has about 3% chance of including at least one pair sharing the first four generations , a 19% chance of including sequences that share three , and a 75% chance of sharing two . Using a point mutation rate of about 1 per 5 generations for the full-length 9 kb HIV-1 genome [10] , [34] , there is about 75% chance of finding among ten sequences two that share one mutation , about 20% chance of finding two sequences that share a pair of mutations , and <2% chance of sharing more than that . These probabilities are slightly enhanced by early stochastic events that can lead to the virus producing less than six descendants in some generations but are diminished by the chances that mutations cause a fitness disadvantage that results in early purifying selection , as previously observed [10] , [60] . We found examples of such early stochastic mutations leading to deviations from star phylogeny in several subjects ( Table 2 ) . We calculated the statistical significance of differences in rates of single versus multivariant HIV-1 transmission using Fisher's exact test . Differences were considered statistically significant at a value of p≤0 . 05 . To estimate the likelihood of missing infrequently represented transmitted variants , we described previously a power study that estimated the probability of sampling low frequency plasma viral sequences [10] . In a sample set of at least n = 20 , there is a 95% probability that a given missed variant comprises less than 14% of the virus population . For a sample set of 30 , there is a 95% probability not to miss a variant that comprises at least 10% of the total viral population . And for a sample set of 80 , there is a 95% probability not to miss a variant that comprises at least 4% of the total viral population . We also considered the possibility that the number of transmitted/founder viruses detected could be influenced by the clinical stage ( Fiebig stage ) of the subjects at the time of virus sampling , because differences in virus replication rates could lead to increasing differences in virus frequencies with time . If this were the case and some viruses were outcompeted , the prediction would be that at later Fiebig stages the numbers of transmitted/founder virus lineages would be less than at earlier Fiebig stages . Our model ( which is based on previously estimated parameters of an HIV-1 generation time of 2 days , a reproductive ratio [R0] of 6 , and a reverse transcriptase error rate of 2 . 16×10−5 and assumes that the initial virus replicates exponentially infecting R0 new cells at each generation and diversifies under a model of evolution that assumes no selection ) predicts that descendants of a transmitted virus at 45% replicative disadvantage compared to another transmitted virus , still have more than a 5% chance of occurring in a sample size of 20 , ten generations ( ∼20 days ) later . In humans , the eclipse period , defined as the time between HIV-1 transmission and first detection of virus in the plasma , has been estimated to be approximately 10–14 days , and the eclipse period plus Fiebig stages I and II , approximately 22–26 days [10] , [33] . In the present study , the numbers of subjects in Fiebig stages I/II , III , IV and V were 14 , 2 , 6 and 6 , respectively , and this relative distribution was similar in the three other studies included in the combined analysis [8] , [9] , [10] . As described in the main text , there was no significant correlation between clinical stage and multiplicity of infection ( Fisher's exact p = 0 . 53 ) .
|
Understanding the biology of sexual transmission of HIV-1 could contribute importantly to the development of effective prevention measures . However , different routes of virus transmission ( vaginal , rectal , penile or oral ) and inaccessibility of tissues at or near the time of virus transmission make this goal elusive . Here , we apply single genome amplification and sequencing of plasma HIV-1 and a model of random virus evolution to a cohort of acutely infected men who have sex with men ( MSM ) and find that MSM are twice as likely as heterosexuals to become infected by multiple viruses as opposed to a single virus . Some MSM subjects were infected by as many as 7 to 10 or more genetically distinct viruses as a consequence of a single exposure event . We go on to molecularly clone the first full-length transmitted/founder subtype B HIV-1 virus and show that it is highly replicative in human CD4+ T-cells but not macrophages . Our study provides the first comparative , quantitative analysis of the multiplicity of HIV-1 infection in the two primary risk groups—MSM and heterosexuals—driving the global pandemic , and we discuss the implications of the findings to HIV-1 vaccine development and prevention research .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"virology/virus",
"evolution",
"and",
"symbiosis",
"virology/vaccines",
"virology/immunodeficiency",
"viruses",
"virology/animal",
"models",
"of",
"infection",
"virology",
"infectious",
"diseases/hiv",
"infection",
"and",
"aids",
"virology/immune",
"evasion",
"infectious",
"diseases/sexually",
"transmitted",
"diseases",
"virology/host",
"invasion",
"and",
"cell",
"entry"
] |
2010
|
High Multiplicity Infection by HIV-1 in Men Who Have Sex with Men
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.